CN120235461B - Coal gangue storage yard environment risk real-time monitoring method and system based on edge calculation - Google Patents
Coal gangue storage yard environment risk real-time monitoring method and system based on edge calculationInfo
- Publication number
- CN120235461B CN120235461B CN202510730406.3A CN202510730406A CN120235461B CN 120235461 B CN120235461 B CN 120235461B CN 202510730406 A CN202510730406 A CN 202510730406A CN 120235461 B CN120235461 B CN 120235461B
- Authority
- CN
- China
- Prior art keywords
- risk
- data
- response
- environmental
- monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/0895—Weakly supervised learning, e.g. semi-supervised or self-supervised learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/096—Transfer learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Computational Biology (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Primary Health Care (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Game Theory and Decision Science (AREA)
Abstract
The application relates to the technical field of data processing, and discloses a coal gangue storage yard environment risk real-time monitoring method and system based on edge calculation. The method comprises the steps of obtaining environment risk original data collected by a multi-source sensor, preprocessing the data through edge computing nodes, constructing a multi-mode identification model to obtain pollution and geological disaster risk results, generating hierarchical early warning information and a dynamic risk map, inputting an intelligent decision system to generate an emergency response plan, updating model parameters based on execution data, and optimizing a risk monitoring system. According to the application, by introducing an edge computing technology, the localization processing of data is realized, the data transmission quantity is reduced, the response time is shortened, and meanwhile, the unified monitoring, prevention and control of pollution risks and geological disaster risks are realized by integrating the multi-mode identification model, so that the timeliness, the accuracy and the response efficiency of the real-time monitoring of the environmental risks of the coal gangue storage yard are improved.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a method and a system for monitoring environmental risk of a coal gangue storage yard in real time based on edge calculation.
Background
The gangue is used as solid waste generated in the coal exploitation and washing process, and the yard management faces double risks of environmental pollution and geological disasters. The traditional gangue storage yard monitoring method mainly relies on manual inspection and single parameter sampling analysis, and has the problems of low monitoring frequency, limited coverage range, data processing lag and the like. With the development of sensor technology, some automatic monitoring systems are applied to coal gangue storage yard management, but the systems generally adopt a centralized architecture, and collected data are transmitted to a remote server for processing, so that the data transmission quantity is large and the response time is long. Meanwhile, most of existing monitoring systems are designed for separating pollution risks and geological disaster risks, and composite risks are difficult to identify and cope with. In addition, the traditional monitoring system lacks intelligent analysis and decision support functions, cannot automatically generate a coping strategy according to risk evolution trend, and reduces timeliness and effectiveness of risk prevention and control.
The method mainly comprises the following steps that firstly, data transmission delay and network bandwidth pressure are caused by a centralized data processing architecture, the requirement of real-time monitoring of environmental risks of a coal gangue storage yard cannot be met, secondly, a single risk monitoring mode cannot effectively cope with complex situations of coexistence of pollution risks and geological disaster risks in the coal gangue storage yard, thirdly, the data analysis process lacks multi-mode data fusion capability, information provided by different types of sensors is difficult to fully utilize, fourth, a pre-warning mechanism lacks space visual expression, risk distribution and evolution trend are difficult to visually show, and fifth, a decision support system is single in function and cannot generate a customized emergency response scheme according to multi-dimensional risk information, and finally, the system lacks an adaptive optimization mechanism, and cannot continuously improve risk identification and prevention and control capability based on history coping experience.
Disclosure of Invention
The application provides a coal gangue storage yard environmental risk real-time monitoring method and system based on edge calculation, which are used for realizing localized processing of data, reducing data transmission quantity, shortening response time, integrating a multi-mode identification model to realize unified monitoring, prevention and control of pollution risks and geological disaster risks and improving timeliness, accuracy and response efficiency of the coal gangue storage yard environmental risk real-time monitoring by introducing an edge calculation technology.
The application provides a coal gangue storage yard environment risk real-time monitoring method based on edge calculation, which comprises the steps of carrying out data acquisition on a coal gangue storage yard and surrounding environment layout multi-source sensor networks to obtain environment risk original monitoring data, inputting the environment risk original monitoring data into edge calculation nodes to carry out data verification, compensation correction, standardization and feature extraction processing to obtain preprocessing data, constructing an environment risk multi-mode recognition model based on the preprocessing data and carrying out training to obtain a pollution risk recognition result and a geological disaster risk recognition result, comprising the steps of inputting soil heavy metal detection data, water quality parameter data and gas concentration data into a convolutional neural network based on the preprocessing data to carry out feature extraction and classification to obtain a pollution risk preliminary recognition result, inputting surface deformation monitoring data, soil humidity data and weather parameter data into a long-term and short-term memory network based on the preprocessing data to carry out time sequence mode analysis to obtain a geological risk preliminary recognition result, constructing an environment risk multi-mode recognition model based on the preprocessing data and carrying out training to obtain a pollution risk recognition result and a geological disaster risk recognition result, constructing a pollution risk recognition result and a pollution risk analysis result, inputting a semi-source risk assessment algorithm to obtain a semi-source risk assessment result and a semi-source analysis result, carrying out a fusion risk assessment result and a fuzzy risk assessment result, carrying out a fuzzy risk assessment result and a fuzzy result matching result, carrying out a fuzzy risk assessment algorithm, the method comprises the steps of obtaining an optimized risk identification model, inputting the preprocessing data monitored in real time into the optimized risk identification model, calculating a risk state to obtain a pollution risk identification result and a geological disaster risk identification result, generating grading early warning information according to the pollution risk identification result and the geological disaster risk identification result, constructing a dynamic risk map to obtain a risk distribution visualization expression, inputting the grading early warning information and the dynamic risk map into an intelligent decision support system to generate an emergency response suggestion to obtain a customized emergency response plan, carrying out system evaluation and knowledge accumulation based on execution data of the emergency response plan, and carrying out parameter update on the environment risk multi-mode identification model to obtain the optimized risk monitoring system.
In a second aspect, the application provides a real-time monitoring system for environmental risk of a coal gangue storage yard based on edge calculation, which comprises:
The acquisition module is used for carrying out data acquisition on the coal gangue storage yard and the surrounding environment by arranging a multi-source sensor network so as to obtain environment risk original monitoring data;
The verification module is used for inputting the environment risk original monitoring data into an edge computing node for data verification, compensation correction, standardization and feature extraction processing to obtain preprocessing data;
The training module is used for constructing an environment risk multi-mode recognition model based on the preprocessing data and executing training to obtain a pollution risk recognition result and a geological disaster risk recognition result, and comprises the steps of inputting soil heavy metal detection data, water quality parameter data and gas concentration data into a convolutional neural network based on the preprocessing data, carrying out feature extraction and classification to obtain a pollution risk preliminary recognition result, inputting surface deformation monitoring data, soil humidity data and meteorological parameter data into a long-term and short-term memory network based on the preprocessing data, carrying out time sequence mode analysis to obtain a geological disaster risk preliminary recognition result, inputting the pollution risk preliminary recognition result and the geological disaster risk preliminary recognition result into a comprehensive risk evaluation layer, carrying out multi-source data fusion to obtain a fusion risk evaluation result, carrying out matching analysis on the fusion risk evaluation result and a historical risk event database, carrying out risk type discrimination and grade assessment to obtain a risk matching result, executing a semi-supervised learning algorithm and an incremental learning algorithm on the risk matching result, carrying out model parameter updating to obtain an optimized risk recognition model, inputting the preprocessing data into the optimized risk recognition model, and carrying out calculation of the optimized risk recognition result to obtain a pollution risk recognition result and a geological disaster risk recognition result;
The generation module is used for generating grading early warning information according to the pollution risk identification result and the geological disaster risk identification result and constructing a dynamic risk map to obtain a risk distribution visual expression;
the input module is used for inputting the grading early warning information and the dynamic risk map into an intelligent decision support system to generate an emergency response suggestion so as to obtain a customized emergency response plan;
And the updating module is used for carrying out system evaluation and knowledge accumulation based on the execution data of the emergency response plan, and carrying out parameter updating on the environment risk multi-mode identification model to obtain an optimized risk monitoring system.
The third aspect provides a coal gangue storage yard environmental risk real-time monitoring device based on edge calculation, which comprises a memory and at least one processor, wherein instructions are stored in the memory, and the at least one processor calls the instructions in the memory so that the coal gangue storage yard environmental risk real-time monitoring device based on edge calculation can execute the coal gangue storage yard environmental risk real-time monitoring method based on edge calculation.
In a fourth aspect, a computer readable storage medium is provided, where instructions are stored, when the computer readable storage medium runs on a computer, to cause the computer to execute the method for monitoring environmental risk of a coal gangue storage yard based on edge calculation in real time.
According to the technical scheme, comprehensive, continuous and real-time monitoring of environmental risks is achieved through arrangement of multi-source sensor networks in a gangue storage yard and the periphery, compared with a traditional single-parameter and low-frequency monitoring method, monitoring range and depth are remarkably expanded, the problem of data transmission delay caused by a traditional centralized architecture is solved through local processing of environment risk original monitoring data input edge computing nodes, response time is shortened by 85%, response speed of a monitoring system to sudden events is greatly improved, environmental risk multi-mode identification models constructed based on pretreatment data integrate identification functions of pollution risks and geological disaster risks, limitation of stroke risk monitoring separation design in the traditional system is overcome, recognition capability of the composite risks is improved, characteristic extraction and classification of the pollution risk data are carried out by an applied convolutional neural network, time sequence mode analysis of the geological disaster data is carried out by a long-period memory network, the two algorithms are respectively optimized for spatial characteristics and time sequence characteristics, multi-dimensional characteristics of the environment risk data of the gangue storage yard are fully adapted, visual early warning information and dynamic early warning information generated according to the identification results are provided, the visual early warning information and dynamic state identification model are integrated, the accuracy and the security risk control system is controlled based on the map, the visual response is controlled by a map, the visual performance is convenient to take account of the map, the visual performance of the map and the performance of the map is improved, the performance of the system is optimized, and the performance of the performance is controlled by using a map-oriented response, and the system is convenient to take the performance of a map-oriented response system, by analyzing and accumulating knowledge of the execution data of the emergency response, the system realizes self-adaptive optimization and continuously improves risk identification and prevention and control capability.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an embodiment of a method for monitoring environmental risk of a coal gangue storage yard in real time based on edge calculation in an embodiment of the present application;
FIG. 2 is a schematic diagram of an embodiment of a system for monitoring environmental risk of a coal gangue storage yard in real time based on edge calculation in an embodiment of the present application;
fig. 3 is a schematic block diagram of a coal gangue storage yard environmental risk real-time monitoring device based on edge calculation in an embodiment of the invention.
Detailed Description
The embodiment of the application provides a method and a system for monitoring environmental risk of a gangue storage yard in real time based on edge calculation. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present application is described below, referring to fig. 1, and an embodiment of a method for monitoring environmental risk of a coal gangue storage yard based on edge calculation in the embodiment of the present application includes:
Step S101, a multisource sensor network is distributed for data acquisition on a gangue storage yard and the surrounding environment, and environment risk original monitoring data are obtained;
step S102, inputting the environment risk original monitoring data into an edge computing node for data verification, compensation correction, standardization and feature extraction processing to obtain preprocessing data;
Step S103, constructing an environmental risk multi-modal identification model based on the preprocessing data and performing training to obtain a pollution risk identification result and a geological disaster risk identification result;
step S104, generating grading early warning information according to the pollution risk identification result and the geological disaster risk identification result, and constructing a dynamic risk map to obtain a risk distribution visual expression;
step S105, inputting the grading early warning information and the dynamic risk map into an intelligent decision support system to generate an emergency response suggestion, and obtaining a customized emergency response plan;
and step S106, performing system evaluation and knowledge accumulation based on the execution data of the emergency response plan, and performing parameter updating on the environment risk multi-mode identification model to obtain the optimized risk monitoring system.
It can be understood that the execution subject of the application can be a coal gangue storage yard environment risk real-time monitoring system based on edge calculation, and can also be a terminal or a server, and the execution subject is not limited in the specific description. The embodiment of the application is described by taking a server as an execution main body as an example.
Specifically, a multisource sensor network is distributed for the gangue storage yard and the surrounding environment thereof. These sensors include soil temperature sensors, soil humidity sensors, gas concentration sensors, heavy metal detection sensors, surface deformation sensors, etc., each of which is used to collect different environmental data. For example, a soil temperature sensor monitors the soil temperature change of different depths of a storage yard, a soil humidity sensor monitors the moisture change inside the storage yard, a gas concentration sensor is used for collecting harmful gas concentrations such as methane, hydrogen sulfide and the like, and a heavy metal sensor monitors harmful substances such as lead, cadmium, arsenic and the like in the soil on the surface layer of the storage yard. The collection of such data is critical to a comprehensive understanding of the environmental risk of the yard.
And preprocessing the original monitoring data. When the preprocessing process starts, data verification is first performed. In this step, the monitoring data with obvious abnormality is removed by setting a threshold range and a change rate checking rule. For example, if the readings of a certain gas concentration sensor deviate significantly from the expected range, the data may be marked as abnormal and rejected. During the verification process, compensation correction is then performed taking into account the effects of drift effects of the sensor and temperature variations on the measurement results. And (3) carrying out temperature and drift correction on the sensor data by establishing a compensation model, so as to ensure the accuracy of the data. And the standardization step is to perform unified processing on the data of different sensors, and convert the data into the same unit and standard format, so that the subsequent data analysis is facilitated. The data noise reduction processing is to remove random noise in the data through wavelet transformation or a moving average algorithm, so that the accuracy of the data is further improved. The feature extraction extracts important statistical features (such as mean, variance and change rate) and frequency domain features from the original data to generate final preprocessing data.
And constructing an environment risk multi-modal identification model based on the preprocessing data, and training. This model contains a pollution risk identification module and a geological disaster risk identification module. The pollution risk identification module utilizes soil heavy metal data, water quality data and gas concentration data to perform feature extraction and classification through a deep learning method (such as a convolutional neural network), so that potential pollution risks are identified. For example, if the lead concentration in the soil exceeds a set safety threshold, the system may identify a risk of contamination and give an early warning. And the geological disaster risk identification module utilizes the earth surface deformation data, the soil humidity data and the meteorological data to carry out time sequence mode analysis through a long-term and short-term memory network (LSTM) so as to identify the risks of geological disasters such as landslide, debris flow and the like. In the training process, the model parameters are continuously updated by utilizing the historical data and the newly acquired data, and the incremental learning and semi-supervised learning methods are adopted to improve the recognition capability of the model on rare risk events.
And generating four-level risk early warning information according to the pollution risk and the geological disaster risk identification result, and constructing a dynamic risk map. The generation of the early warning information depends on the risk type and the intensity identified by the model, and the probability, the influence range and the development trend of the risk are comprehensively considered. For example, if an anomaly in soil heavy metal concentration is detected and combined with landslide predictive data, the system will evaluate the area for contamination and geological disaster risk and determine its risk level. When a dynamic risk map is constructed, the system combines the risk data with a Geographic Information System (GIS), generates a risk thermodynamic diagram through space mapping, and dynamically updates the risk distribution according to time variation. By color coding and time series superposition, the system can intuitively demonstrate the spatial distribution and temporal evolution of risk.
Based on the risk early warning information and the dynamic risk map, the system inputs the data into an intelligent decision support system to generate a customized emergency response plan. And the system screens out the optimal countermeasure according to the risk level and the regional position by combining the environmental risk countermeasure knowledge base of the gangue storage yard. For example, in areas with high risk of contamination, the system may recommend activation of the contamination interception dams and delivery of the adsorbent material, while for areas with high risk of geological disasters, the system may recommend yard reinforcement and evacuation of surrounding residents.
Based on the execution data of the emergency response program, the system evaluates and performs knowledge accumulation. By tracking the execution process, the system calculates indexes such as response timeliness, resource utilization efficiency and the like, and evaluates the effect of emergency response. If the response effect of a link is not good (for example, the processing efficiency is low due to improper resource allocation), the system updates model parameters and optimizes the emergency response strategy according to the evaluation result. The system can extract successful coping strategies and risk modes by storing the execution data and the effect evaluation result into the historical case database, so that the efficiency of risk identification and emergency response is continuously improved.
In the embodiment of the application, the comprehensive, continuous and real-time monitoring of environmental risks is realized by arranging the multi-source sensor network at the coal gangue storage yard and the periphery, compared with the traditional single-parameter and low-frequency monitoring method, the monitoring range and depth are remarkably expanded, the problem of data transmission delay caused by the traditional centralized architecture is solved by locally processing the environmental risk original monitoring data input edge computing nodes, the response time is shortened by 85%, the response speed of a monitoring system to sudden events is greatly improved, the environmental risk multi-mode recognition model constructed based on the preprocessing data integrates the recognition functions of pollution risks and geological disaster risks, the limitation of stroke risk monitoring separation design of the traditional system is overcome, the recognition capability of the composite risks is improved, the characteristic extraction and classification of the pollution risk data are carried out by the applied convolutional neural network, the time sequence mode analysis of the geological disaster risk data is carried out by the long-period memory network, the two algorithms are respectively optimized for spatial characteristics and time sequence characteristics, the multi-dimensional characteristics of the environmental risk data of the coal gangue storage yard are fully adapted, the visual response speed of the monitoring system is improved, the environmental risk recognition system is integrated based on the recognition information generated by the recognition result, the environmental risk multi-mode recognition model is used for the visual early warning system, the map is convenient to respond to the map-oriented response and the intelligent system is customized by the map-oriented and the map-oriented system, the comprehensive response is conveniently-oriented by the map-oriented and the map-oriented system, the comprehensive response system has the requirements of the visual performance has been improved, and the performance of the comprehensive performance of the comprehensive performance has the performance of the control and the system has the requirements of the requirements on the performance of the system, the system realizes self-adaptive optimization and continuously improves risk identification and prevention and control capability.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
Burying a soil temperature sensor, a soil humidity sensor and a gas concentration sensor at different depths in a coal gangue storage yard, and collecting internal parameters of the storage yard to obtain temperature distribution data, water content change data and harmful gas concentration data;
Arranging a rapid heavy metal content detection sensor on the surface of the gangue storage yard, and detecting the element content of surface soil to obtain heavy metal element content data of lead, cadmium and arsenic;
Arranging an earth surface deformation monitoring device around the coal gangue storage yard, and monitoring earth surface changes in real time to obtain earth surface subsidence data, landslide precursor data and debris flow precursor data;
a water quality monitoring sensor is arranged on a water body at the downstream of the coal gangue storage yard, and physical and chemical properties of the water body are detected to obtain pH value data, conductivity data and dissolved oxygen data;
Weather parameter acquisition devices are arranged around the coal gangue storage yard, and environmental weather conditions are recorded to obtain rainfall data, wind speed data and air pressure data;
and summarizing and integrating temperature distribution data, water content change data, harmful gas concentration data, heavy metal element content data, ground surface subsidence data, landslide precursor data, debris flow precursor data, pH value data, conductivity data, dissolved oxygen data, rainfall data, wind speed data and air pressure data to obtain environment risk original monitoring data.
Specifically, the soil temperature sensor, the soil humidity sensor and the gas concentration sensor are buried at different depths inside the gangue storage yard, so as to comprehensively monitor key environmental parameters such as temperature distribution, water content change, harmful gas concentration and the like in the storage yard. The soil temperature sensor is used for helping to judge whether overheat phenomenon or other risk events possibly causing temperature abnormality exist in the storage yard by recording the temperature change of the soil in real time, the soil humidity sensor is used for monitoring the moisture change in the soil of the storage yard, the risk of the storage yard stability change possibly caused by the too high or too low humidity of the storage yard is important for analysis, the gas concentration sensor is mainly used for monitoring the concentration change of harmful gases such as methane, hydrogen sulfide and the like, and the leakage of the gases often indicates the possibility of the accumulation or leakage of the harmful gases in the storage yard. The data collection of all these sensors will generate raw monitoring data including temperature distribution data, water content variation data and harmful gas concentration data, which provides a basis for subsequent data analysis.
The rapid heavy metal content detection sensor is arranged on the surface of the gangue storage yard and is specially used for detecting the heavy metal element content in the surface soil. The sensors can monitor common heavy metal elements such as lead, cadmium, arsenic and the like in real time, and the elements have obvious harm to the environment and human health. Therefore, the method for accurately monitoring the concentration of the heavy metal in the surface soil of the storage yard is important for timely finding a pollution source and taking corresponding prevention and control measures. The sensor records and generates heavy metal element content data in real time, and provides key information for subsequent pollution risk analysis.
The surface deformation monitoring device is arranged in the peripheral area of the storage yard and mainly comprises an inclination angle sensor, a displacement sensor and a vibration sensor. These devices can monitor the ground surface changes of the yard and its surrounding areas in real time. By monitoring the ground surface subsidence data, landslide precursor data and debris flow precursor data, the system can grasp the state of geological change around the storage yard in real time, discover early symptoms of geological disasters in time, and further provide decision support for preventing natural disasters such as landslide, subsidence or debris flow.
For the water body at the downstream of the heap, the water quality monitoring sensor is arranged for detecting the physicochemical properties of the water body, and mainly comprises indexes such as pH value, conductivity, dissolved oxygen and the like. The pH value sensor is used for monitoring the pH value change of the water body, the conductivity sensor is used for monitoring the dissolved salt in the water body, and the dissolved oxygen sensor is used for monitoring the oxygen content of the water body. The change of the parameters can reflect the environmental problems of acidification of the water body, dissolution of pollutants and the like, and has important significance for effective management of water quality and pollution prevention and control. In addition, weather parameter acquisition devices installed around the storage yard comprise a rainfall gauge, an anemometer and an barometer, and the change condition of environmental weather conditions can be recorded. Rainfall data, wind speed data and air pressure data are key indexes for knowing climate condition changes, particularly weather early warning can be provided in time under extreme weather conditions, and adverse effects of weather disasters on the storage yard environment are prevented.
All the collected original data (including temperature distribution data, water content change data, harmful gas concentration data, heavy metal element content data, ground surface subsidence data, landslide sign data, mud-rock flow sign data, pH value data, conductivity data, dissolved oxygen data, rainfall data, wind speed data and air pressure data) are summarized and integrated to form the environment risk original monitoring data. The data not only cover the environmental parameters in the storage yard, but also include the ecological environment data around the storage yard, and provide abundant basic data support for subsequent environmental risk assessment and early warning. Through comprehensive analysis of the raw data, potential risks of a storage yard can be found in time, environmental safety of the storage yard can be evaluated, and a data basis is provided for making an emergency response scheme.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
setting a threshold range and a change rate checking rule for the original monitoring data of the environmental risk, and performing data verification processing to obtain verification data with abnormal values removed;
performing compensation correction calculation on the verification data according to a sensor drift compensation formula and a temperature influence correction model to obtain corrected monitoring data;
performing dimension conversion and unit unification operation on the corrected monitoring data, and performing standardization processing to obtain standard data with uniform format;
Performing a time stamp calibration algorithm on the standard data, and performing time synchronization processing to obtain synchronous data with consistent time sequence;
inputting the synchronous data into a wavelet transformation filtering algorithm and a moving average filtering algorithm, and carrying out data noise reduction treatment to obtain denoised smooth data;
and calculating the statistical characteristics of the mean value, variance and change rate of the smooth data, extracting the frequency domain characteristics through fast Fourier transformation, and carrying out characteristic extraction processing to obtain the preprocessed data.
Specifically, the original environmental risk monitoring data is subjected to a data verification process. In this step, a threshold range and a change rate check rule are set. By means of these rules, the system is able to identify data that is significantly outside the expected range or that has too great a range of variation. Such data may be caused by sensor failures, environmental disturbances, or other anomalies. For example, if the temperature reading of a certain sensor suddenly jumps from 20 ℃ to 50 ℃, the change is obviously not physical, and the system automatically marks and rejects the outliers. After the outliers are removed, the remaining data is considered valid verification data.
And compensating and correcting the verification data. The sensor may drift during long-term use or may be affected by temperature changes under different environmental conditions, resulting in deviation of the measurement results. Thus, at this stage, the system will calculate and make compensation corrections based on the sensor's drift compensation formula and temperature impact correction model, in combination with the actual performance of the sensor. For example, if a humidity sensor has systematic deviation in a high-temperature environment, the system adjusts the measured value according to a preset compensation formula to eliminate the influence of temperature on the data. The compensated data is more similar to the monitoring data of the real environment. And carrying out dimension conversion and unified standardization processing. The main purpose of this step is to unify the different types of sensor data into one standard format and unit for subsequent analysis and processing. For example, the soil moisture sensor may be in units of "% RH", while the gas concentration sensor may be in units of "ppm". During the normalization process, the data is converted into a unified unit (such as a unified unit system SI unit), and all data is ensured to be stored in the same format, so that the data integration and comparison are convenient.
Then, a time stamp calibration algorithm is performed for time synchronization. In a multi-source sensor system, there may be time bias or data acquisition time differences for the different sensors. To ensure that all the sensor data can be aligned and analyzed at the same time line, the system will time-stamp each piece of data. For example, if the sampling time of one sensor lags behind the other sensors, the system will correct according to the actual time, so that the data of all the sensors are kept consistent in time, and the time sequence consistency of the multi-source data is ensured.
After the time synchronization is completed, the system performs noise reduction processing on the data, and removes random noise in the data by using a wavelet transform filtering algorithm and a moving average filtering algorithm. Wavelet transform filtering can effectively separate noise components in a signal, preserve useful signal characteristics, and is particularly excellent when processing environmental data with sudden fluctuations. While moving average filtering smoothes out rapidly fluctuating portions by computing the neighborhood average of data points, which is very effective in removing transient, non-periodic noise. The data after the noise reduction processing is smoother and more real.
And then, the system performs feature extraction on the denoised smooth data. The feature extraction process includes calculating statistical features such as mean, variance, and rate of change of the data, and extracting frequency domain features by Fast Fourier Transform (FFT). The mean and variance are common statistical features used to describe the central tendency and degree of dispersion of the data, and the rate of change reveals the rate of change of the data over time. For example, when monitoring changes in soil humidity, the rate of change can reflect sharp fluctuations in humidity, indicating a possible environmental risk. The fast fourier transform converts the time domain signal into a frequency domain signal, which can extract the frequency characteristics in the signal, and is very useful for identifying periodic variations and long-term trends. Through the feature extraction, the system can effectively capture key features related to environmental risks and provide rich information for subsequent risk identification.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
Constructing a pollution risk identification layer based on the pretreatment data, inputting soil heavy metal detection data, water quality parameter data and gas concentration data into a convolutional neural network, and performing feature extraction and classification to obtain a preliminary pollution risk identification result;
Constructing a geological disaster risk identification layer based on the preprocessing data, inputting surface deformation monitoring data, soil humidity data and meteorological parameter data into a long-term and short-term memory network, and analyzing a time sequence mode to obtain a geological disaster risk preliminary identification result;
Inputting the pollution risk preliminary identification result and the geological disaster risk preliminary identification result into a comprehensive risk assessment layer, carrying out relevance analysis and weight distribution on the two types of risk results through a weighted fusion algorithm, and carrying out multi-source data fusion to obtain a fusion risk assessment result;
Carrying out matching analysis on the fusion risk assessment result and a historical risk event database, and carrying out risk type discrimination and grade assessment to obtain a risk matching result;
Executing a semi-supervised learning algorithm and an incremental learning algorithm on the risk matching result, and updating model parameters to obtain an optimized risk identification model;
and inputting the preprocessing data monitored in real time into the optimized risk identification model, and calculating a risk state to obtain the pollution risk identification result and the geological disaster risk identification result.
Specifically, a pollution risk identification layer is constructed based on the preprocessed data. The method comprises the steps of carrying out data recombination on preprocessed soil heavy metal detection data, water quality parameter data and gas concentration data according to a time sequence and a spatial position to form a three-dimensional tensor input format, wherein a first dimension represents a data type (heavy metal concentration value, pH value, conductivity, dissolved oxygen, gas concentration and the like), a second dimension represents a spatial position coordinate, and a third dimension represents a time sequence; then, designing a convolutional neural network architecture comprising an input layer, three convolutional layers, two pooling layers and a full-connection layer, wherein the input layer receives four-dimensional tensors with the dimension [ batch size x data type number x space position number x time window ], the first convolutional layer uses 32 3x3 convolutional kernels to perform feature extraction and adopts a ReLU activation function, the first pooling layer adopts 2 x 2 maximum pooling to reduce the data dimension, the second convolutional layer uses 64 3x3 convolutional kernels to further extract high-level features, the second pooling layer performs 2 x 2 maximum pooling again, the third convolutional layer uses 128 3x3 convolutional kernels to extract more complex pollution pattern features, the feature map is converted into one-dimensional vectors through global average pooling, and finally the full-connection layer comprising 256 neurons and the output layer adopting a softmax activation function are connected; in the training process, historical pollution event data is used as a labeling sample, pollution risk levels are divided into four categories (no risk, slight risk, medium risk and serious risk), a cross entropy loss function and an Adam optimizer are adopted to update network parameters, the learning rate is set to be 0.001, the batch size is set to be 32, the training period is 100 epochs, after the network training is completed, real-time preprocessing data is input into a trained CNN model, probability distribution of the four risk levels is obtained through forward propagation calculation, and selecting the category with the highest probability as a pollution risk preliminary identification result, and simultaneously outputting a confidence coefficient score for subsequent risk assessment, thereby realizing the construction of a pollution risk identification layer based on the preprocessed data. The layer is mainly used for processing soil heavy metal detection data, water quality parameter data and gas concentration data. These data can reflect the pollution conditions inside the yard and its surrounding environment. The contamination risk identification layer employs Convolutional Neural Networks (CNNs) for feature extraction and classification. CNNs are excellent in image and multidimensional data processing, and key features such as a change pattern of heavy metal concentration, abnormal fluctuation of water quality, and abrupt change of harmful gas concentration can be effectively extracted from these environmental data. By inputting these data into the CNN network, the network is able to automatically learn the spatial and temporal dependencies that exist in the data, and progressively extract more abstract and advanced features through multiple convolution layers. After training, the CNN can output a preliminary recognition result of pollution risk on the basis of real-time monitoring data. For example, if the lead concentration in the soil exceeds a set safety threshold, a sudden rise in gas concentration, etc., the network will determine a potential risk of contamination and output a preliminary risk assessment result.
And constructing a geological disaster risk identification layer. the method comprises the steps of arranging and organizing pretreated ground surface deformation monitoring data, soil humidity data and meteorological parameter data according to a time sequence, wherein the ground surface deformation data comprise displacement amounts and deformation rates in X, Y, Z directions, the soil humidity data comprise water content percentages of different depths, the meteorological parameter data comprise indexes such as rainfall, wind speed, air pressure and temperature, the multidimensional data are sliced according to a fixed time window (such as 24 hours), and three-dimensional input tensors with dimensions of [ sample number multiplied by characteristic dimension multiplied by time step ] are formed; then, a long-term and short-term memory network architecture is designed, which comprises an input layer, two LSTM hidden layers, Dropout layer and output layer, wherein the input layer receives sequence data with dimension of [ batch size x time step x characteristic number ], the first LSTM layer comprises 64 memory units, adopts tanh activating function and sigmoid gating mechanism, can learn short-term time sequence dependency relationship, the second LSTM layer comprises 32 memory units for capturing long-term time sequence mode and trend change, dropout layer (discarding rate of 0.2) is added between two LSTM layers to prevent overfitting, and finally fully connected layer comprising 16 neurons and output layer adopting sigmoid activating function are connected, in the time sequence mode analysis process, LSTM network uses forgetting gate to determine which historical information is discarded, uses input gate and candidate value to update current state, uses output gate to control output content so as to identify progressive change of surface deformation, The method comprises the steps of carrying out a network training by using time sequence data of historical geological disaster events as positive samples and data during normal monitoring as negative samples, carrying out parameter updating by using RMSprop optimizers by adopting a binary cross entropy loss function, setting a learning rate to 0.0001, setting a training batch size to 16 and a maximum training round number to 200, inputting real-time preprocessing data into a trained LSTM model according to the same time window after the training is finished, outputting risk probability values between 0 and 1 by a network, judging that geological disaster risks exist when the probability values exceed 0.5, dividing risk grades (0-0.3 are low risks, 0.3-0.7 are medium risks and 0.7-1.0 are high risks) according to a probability interval, and obtaining a geological disaster risk preliminary identification result, thereby realizing a geological disaster risk identification layer construction based on time sequence pattern analysis. The layer processes earth's surface deformation monitoring data, soil moisture data and meteorological parameter data. Geological disasters, such as landslide, debris flow, etc., are often accompanied by significant time series changes, and are therefore amenable to time series pattern analysis using long and short term memory networks (LSTM). LSTM is a recurrent neural network suitable for processing sequence data that is capable of remembering critical information in long time sequences and effectively capturing long-term dependencies in time. In this layer, the LSTM inputs time series data of surface subsidence, soil moisture and meteorological parameters into the network, trains and identifies potential geological disaster risks. For example, as soil humidity gradually increases, combined with increased rainfall, LSTM may identify disaster risks such as landslide or debris flow that these patterns may cause. After training, the LSTM can output a preliminary identification result of geological disaster risk according to real-time data. After the two primary identification results are obtained, the system inputs the information into the comprehensive risk assessment layer for multi-source data fusion. The layer combines a pollution risk identification result and a geological disaster risk identification result, firstly calculates a correlation coefficient between two types of risks through a correlation analysis algorithm, then carries out weight distribution on each risk type based on the risk severity and the influence range, and comprehensively processes multi-source risk data by adopting a weighted fusion algorithm, thereby comprehensively evaluating the overall risk condition of the environment. For example, when the pollution risk and the geological disaster risk exist at the same time, the system analyzes whether the heavy metal pollution of the soil can be accelerated and diffused due to the deformation of the earth surface or whether the landslide caused by rainfall can cause the migration of the pollutant to the downstream, quantitatively calculates the interaction strength of the two risks through a risk coupling model, and outputs the comprehensive risk assessment result by combining the respective severity, the development trend and the interaction influence factor.
And the system performs matching analysis on the fusion risk assessment result and the historical risk event database. The purpose of this step is to conduct risk type discrimination and ranking by comparison with historical risk events. The system can judge which type the current risk event belongs to according to the past case data and evaluate the risk level of the current risk event. Through such matching analysis, the system can effectively evaluate the urgency of the current environmental risk event. For example, if historical data indicates that similar contaminating events have caused serious environmental damage, the system will rate the current event as a high risk and take corresponding precautions.
After the risk matching result is obtained, the system executes a semi-supervised learning algorithm and an incremental learning algorithm to optimize the model. Semi-supervised learning allows the system to help improve the recognition capability of the model through unlabeled data under the condition that only part of the data is labeled, and incremental learning can continuously update model parameters when new data arrives, so that the accuracy and adaptability of the model are continuously improved. By the support of the algorithms, the model can be gradually optimized and adapt to different environment risk modes, so that more efficient risk identification is realized.
And inputting the preprocessed data monitored in real time into an optimized risk identification model, and calculating a risk state. The model can dynamically adjust the risk state through the environment monitoring data updated in real time, and re-evaluate the pollution risk and the grade of the geological disaster risk. For example, at a certain time point, if the gas concentration is increased or the soil humidity is changed sharply, the model recalculates pollution risk and geological disaster risk according to the optimized parameters, and gives a new risk assessment result.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
performing risk probability calculation and influence range evaluation algorithm on the pollution risk recognition result and the geological disaster risk recognition result, and performing risk classification to obtain fourth-level risk early warning data;
Generating a data structure containing risk type description, affected area, early warning level, development trend and response advice based on the four-level risk early warning data, and packaging early warning information to obtain structured early warning information;
Carrying out spatial registration on the structured early warning information and a geographic information system base map, and carrying out geographic coordinate mapping to obtain geographic coordinated risk information;
Performing color coding algorithm and thermodynamic diagram generation algorithm on the geographical coordinated risk information, and performing risk visualization rendering to obtain a risk distribution heat diagram;
Performing time sequence superposition processing on the risk distribution heat map, and performing space-time evolution analysis to obtain a space-time dynamic risk evolution sequence;
And (3) importing the risk evolution sequence into a multi-scale display engine, and performing view generation and interactive interface construction to obtain the risk distribution visual expression.
Specifically, after the pollution risk recognition result and the geological disaster risk recognition result are obtained, the system executes risk probability calculation and an influence range evaluation algorithm. The main purpose of this process is to calculate the probability of risk occurrence and the possible impact range on the environment, personnel, facilities, etc. based on the recognition result. For example, the probability calculation of pollution risk can be based on historical data of the change of the heavy metal concentration and the gas concentration of soil, and the probability of occurrence of pollution event can be obtained by combining meteorological conditions and topographic information, and the influence range of geological disaster risk can be used for determining the area possibly affected by landslide or debris flow through comprehensive evaluation of factors such as ground deformation monitoring data, soil humidity and rainfall. These calculations will provide basis for subsequent risk classification. By risk classification, the system classifies the risk into four classes, green (no risk), yellow (slight risk), orange (medium risk) and red (serious risk), and according to the difference of each class, the system adopts different early warning measures and response schemes.
Next, based on the four-level risk early warning data, the system generates an early warning information structure containing detailed information. The warning information will include a description of the type of risk (e.g., pollution risk or geological disaster risk), the extent of the affected area (e.g., inside the yard, surrounding the yard, downstream waters, etc.), the warning level (e.g., green, yellow, orange, or red), the trend of the risk development (e.g., exacerbating, slowing down, or remaining stable), and countermeasures (e.g., immediate taking protective measures, monitoring reinforcement, personnel evacuation, etc.). After the information is packaged, the structured early warning information is formed, so that the subsequent processing, transmission and display of the system are facilitated.
On the basis, the system performs spatial registration on the structured early warning information and a Geographic Information System (GIS) base map. The GIS base map comprises geographic information of a storage yard area and surrounding environments, the system matches early warning information with actual geographic positions through geographic coordinate mapping, and risk data is mapped to specific spatial positions. The process ensures that the early warning information can accurately reflect the geographical distribution condition of the occurrence of the risk event. Through the registered geographic coordinated risk information, the system can provide accurate spatial positioning and real-time environment monitoring for the user.
The system then color codes and thermodynamic diagrams the geoco-ordinated risk information. The color coding algorithm maps different risk levels to different color ranges, e.g. green for no risk, yellow for light risk, orange for medium risk, and red for severe risk. By color coding, the spatial distribution of risk becomes intuitive and understandable. The system then renders the risk information into a thermodynamic diagram using a thermodynamic diagram generation algorithm that can clearly demonstrate the risk intensity of different regions. For example, areas of higher risk may be highlighted in red, while areas of lower risk are displayed in green. The thermodynamic diagram is generated so that a user can clearly see the risk distribution of different areas at a glance, and a decision maker is helped to take countermeasures in time.
In order to further dynamically show the risk change process, the system executes time sequence superposition processing on the risk distribution heat map to perform space-time evolution analysis. The time sequence superposition processing is to integrate the data of multiple time points, display the change of the spatial distribution of risks along with time and display the dynamic evolution process of risks. This analysis can reveal expansion or contraction of the risk areas, the trend of the change in risk level, and the impact of environmental conditions on risk. For example, if the risk level of a certain area continuously rises in a period of time, the system can timely reflect the risk level through a thermodynamic diagram, and the manager can be helped to early warn of potential danger.
The system imports the time-space dynamic risk evolution sequence into a multi-scale display engine to perform view generation and interactive interface construction. The multi-scale display engine allows a user to view risk profiles from a macroscopic to microscopic view. For example, a user may view a risk overview of the entire gangue dump and its surroundings, or a detailed risk analysis focused on a small area. The interactive interface is constructed so that the user can freely zoom and drag the map, view risk information of different layers and conduct deep analysis. The interactive view display enables a decision maker to flexibly view and analyze data according to specific requirements, so that accurate basis is provided for subsequent decisions.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
Matching and searching the grading early warning information, the dynamic risk map and the coal gangue storage yard environmental risk coping knowledge base, and carrying out coping scheme primary screening to obtain a candidate coping scheme set;
Performing multi-objective optimization calculation on the candidate coping scheme set in combination with the current storage yard condition, the available resource condition and the risk development trend, and performing scheme evaluation sequencing to obtain a sequenced coping scheme;
Constructing a multi-scenario decision tree based on the ordered countermeasure, calculating an expected utility value of each decision node according to the expected risk reduction degree, implementation success probability and time benefit of countermeasure by adopting a weighted summation mode, and simultaneously carrying out accumulation calculation based on labor input cost, equipment use cost, material consumption cost and time cost to obtain a resource consumption value, evaluating the goodness of each decision path according to the ratio of the expected utility value to the resource consumption value, and carrying out decision path generation to obtain a decision branch network;
Integrating and analyzing the decision branch network with weather forecast data, engineering activity plans and sensitive target distribution information, and carrying out future scene deduction to obtain a risk development prediction model;
Executing a Monte Carlo simulation algorithm on each decision branch based on the risk development prediction model, and evaluating the risk coping effect to obtain a coping scheme optimization result;
and converting the response scheme optimization result into a structured document containing risk description, response targets, technical measures, resource allocation, personnel division and time nodes, and generating a planning document to obtain a customized emergency response plan.
Specifically, matching and searching are carried out on the grading early warning information, the dynamic risk map and the gangue storage yard environment risk coping knowledge base. The system matches the real-time monitored risk level (such as pollution risk and geological disaster risk) and risk spatial distribution (namely dynamic risk map) with the preset environment risk to cope with the emergency scheme in the knowledge base. An emergency knowledge base is a database containing standard coping processes, resource requirements and technical measures for different environmental risk situations. By matching and searching with the early warning information and the risk map, the system can rapidly acquire the candidate coping scheme set. These sets of schemes are pre-designed standard emergency response policies for the currently identified risk type and risk class.
The system performs multi-objective optimization calculation on the candidate coping scheme set, and performs coping scheme evaluation and sequencing by combining the actual condition, the available resource condition and the risk development trend of the current storage yard. In practical applications, each of the solutions may consider different objectives, such as the effect of risk control, consumption of required resources, and implementation difficulty. The multi-objective optimization calculation evaluates the merits of each scheme through a mathematical model, balances between different objectives, and finally outputs a sequenced response scheme. For example, if a yard is at a significant risk of contamination, the system may prioritize the quick taking of contamination interception measures and select the most cost effective solution in situations where resources are limited. After multi-objective optimization, the system will get a set of evaluated and ranked treatment schemes.
Then, based on the ordered coping schemes, the system builds a multi-scenario decision tree. At this stage, the decision process of each coping scheme is broken down into a plurality of decision nodes. Each node represents a possible decision path under certain conditions, e.g. decisions to take different countermeasures at different risk levels. And the system calculates an expected utility value of each decision node according to the expected risk reduction degree, the implementation success probability and the time benefit of the counter measures by adopting a weighted summation mode, and simultaneously calculates a resource consumption value by accumulation based on the labor input cost, the equipment use cost, the material consumption cost and the time cost, and evaluates the superiority and inferiority of each decision path according to the ratio of the expected utility value to the resource consumption value. These indicators will help the decision maker to measure the effect and cost of different decisions and ultimately generate decision paths, thus obtaining a complete decision branch network. Through this decision tree, the decision maker can clearly see the possible consequences of each decision, as well as the merits of the various choices.
After the decision branch network is constructed, the system integrates and analyzes weather forecast data, a project activity plan of a storage yard and sensitive target distribution information. Weather forecast data can provide early warning of weather change for decision making, for example, landslide or debris flow risk is increased due to increased rainfall, engineering activity plans can reflect influences of storage yard construction or other artificial activities on risks, and sensitive target distribution information can provide key area information (such as residential areas and important infrastructure) of surrounding areas of the storage yard. By integrating this information into the analysis of the decision tree, the system is able to make future scenario deductions, predicting the trend of risk in different situations. For example, if the weather forecast indicates that a storm is imminent, the system may adjust the countermeasures to strengthen the reinforcement of the yard side slope and avoid landslide hazards caused by rainfall.
Based on the predicted risk development trend, the system uses a Monte Carlo simulation algorithm to evaluate the risk coping effect of each decision branch. Monte Carlo simulation is an algorithm that simulates multiple scenarios that may occur through random sampling and calculates the results thereof. Through Monte Carlo simulation, the system can simulate the effects of different coping strategies under various uncertain factors (such as weather changes and resource limitations), and evaluate whether each decision path can effectively reduce risks under the possible future change conditions. Monte Carlo simulation can provide a more comprehensive and reliable coping strategy evaluation for a decision maker, ensuring that a selected coping scheme can cope with a variety of potential uncertain risks.
The system converts the optimized response scheme into an emergency response scheme document containing detailed information. The document includes the contents of risk descriptions, goals of care, specific technical measures, resource allocation, personnel division, and time nodes. The emergency response program will generate specific embodiments based on the optimization results, ensuring that the response can be handled quickly and effectively when at risk. For example, for a high risk area, the system may recommend immediate deployment personnel for environmental remediation, while deploying equipment for contaminant interception, and ensuring that relevant personnel complete the task on time. The emergency response plan also includes resource allocation plans, such as allocation policies for funds, manpower, and equipment, to ensure that the solution is performed successfully.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
Comparing the environment monitoring data before and after the emergency response plan is executed, calculating response timeliness, risk control effectiveness, resource utilization efficiency and secondary risk prevention and control indexes, and performing effect evaluation to obtain risk coping performance evaluation results;
Based on the performance deficiency points of each module of the risk coping performance evaluation result identification system, carrying out improvement strategy generation to obtain optimization suggestions comprising sensor layout adjustment, sampling frequency optimization, algorithm improvement and threshold calibration;
storing the complete data of the risk event, the coping process and the effect evaluation result into a historical case database, and carrying out case induction to obtain structured risk coping experience data;
Executing an association rule mining algorithm and a sequence pattern mining algorithm on the structured risk coping experience data, and extracting knowledge to obtain risk pattern characteristics and an optimal coping strategy;
performing parameter adjustment and structure optimization on the environmental risk multi-mode recognition model based on the risk mode characteristics and the optimal coping strategy, and performing model update to obtain an optimized risk recognition model;
Integrating the optimized risk identification model with the optimized sensor network layout, the data processing algorithm and the decision support system, and performing system integration to obtain the optimized risk monitoring system.
Specifically, the system compares the environmental monitoring data before and after execution of the emergency response plan to calculate a plurality of performance indicators. These include response timeliness, risk control effectiveness, resource utilization efficiency, and secondary risk prevention and control indicators. The response timeliness refers to the time difference from the occurrence of risks to the implementation of response measures, the shorter response time means the stronger emergency response capability of the system, the risk control effectiveness measures whether the change condition of the risk level effectively reduces the risks by taking the countermeasures, the resource utilization efficiency pays attention to the use condition of resources (such as manpower, equipment and funds) in the emergency response process, the rationality of resource allocation is ensured, and the secondary risk prevention and control index evaluates whether new risks or unforeseen risk events are avoided in the emergency response process. For example, in response to certain environmental risks, if enhanced monitoring and real-time data analysis can avoid secondary pollution or geological disasters, it is indicated that secondary risk prevention and control measures are effectively performed.
By calculating these performance metrics, the risk response performance evaluation results obtained by the system will reveal the overall effect of the emergency response program. If not performing well on certain criteria, the system may identify possible deficiencies and provide basis for subsequent optimization.
Based on the risk handling performance evaluation results, the system can identify performance shortages for each module and generate an improvement strategy. For example, if the accuracy of the monitoring sensors is insufficient, resulting in inaccurate data, the system may recommend adjusting the layout of the sensors, optimizing the layout of the sensors to ensure more accurate capture of critical data, if the data acquisition frequency does not meet the real-time requirements of the emergency response, the system may recommend increasing the sampling frequency, or dynamically adjusting for different risk types and variations. Furthermore, if the data processing algorithm is inefficient or the existing threshold settings do not adapt to the actual risk changes, the system may suggest algorithm improvements and threshold calibration. These optimization suggestions will help to improve overall system performance, ensuring that future exposure to risks can provide more efficient countermeasures.
The system then stores the complete data of the risk event, the coping process and the effect evaluation results in a historical case database. This database will contain various data accumulated during multiple emergency responses, including environmental monitoring data, risk assessment results, implemented emergency measures, resource allocation conditions, and effect assessment results, etc. By storing this information in a structured manner, the system can provide valuable historical data support for future emergency responses, helping to analyze and summarize coping experience in different situations.
For structured data stored in the historical database, the system will use association rule mining algorithms and sequence pattern mining algorithms to perform knowledge extraction. Association rule mining algorithms are able to identify patterns and associations that often occur in different emergency responses, e.g., which emergency measures tend to be most effective when a particular type of pollution risk occurs. The sequence pattern mining algorithm can extract risk pattern features from time sequence data, such as how a certain geological disaster risk evolves over time, and which factors play a decisive role in the risk development in different time periods. Through the algorithms, the system can extract the optimal coping strategy from the historical data, and help optimize future emergency response measures.
Based on the risk pattern features and the optimal coping strategies extracted from the historical data, the system can perform parameter adjustment and structure optimization on the multi-pattern recognition model of the environmental risk. The purpose of this optimization process is to promote the accuracy and adaptability of the model in terms of recognition in the face of emerging risks. For example, if the system identifies certain new sources of contamination or risk types, the model may automatically adjust parameters based on historical data to improve the ability to identify these new risks. Through the optimization, the risk identification model can continuously improve the accuracy and efficiency of the risk identification model in long-term operation.
The risk identification system after model optimization is integrated with the optimized sensor network layout, the data processing algorithm and the decision support system. The optimization of the sensor network layout ensures that the monitoring system can cover all key areas and avoid information blind areas, the optimization of the data processing algorithm improves the processing speed and accuracy of data, ensures that effective data support can be obtained in time in the emergency response process, and the optimization of the decision support system provides more scientific and reasonable decision basis for managers and helps the managers to make decisions quickly in emergency situations. By this integration, the end result will be an optimized risk monitoring system that can more efficiently and accurately cope with various environmental risks that may occur in the future.
The method for monitoring the environmental risk of the coal gangue storage yard based on the edge calculation in the embodiment of the present application is described above, and the system for monitoring the environmental risk of the coal gangue storage yard based on the edge calculation in the embodiment of the present application is described below, referring to fig. 2, one embodiment of the system for monitoring the environmental risk of the coal gangue storage yard based on the edge calculation in the embodiment of the present application includes:
The acquisition module 201 is used for carrying out data acquisition on a multi-source sensor network arranged in a gangue storage yard and the surrounding environment to obtain environment risk original monitoring data;
the verification module 202 is configured to input the environmental risk raw monitoring data into an edge computing node for data verification, compensation correction, standardization and feature extraction processing, so as to obtain preprocessed data;
the training module 203 is configured to construct an environmental risk multi-modal identification model based on the preprocessing data and perform training to obtain a pollution risk identification result and a geological disaster risk identification result;
The generating module 204 is configured to generate hierarchical early warning information according to the pollution risk identification result and the geological disaster risk identification result, and construct a dynamic risk map to obtain a risk distribution visual expression;
The input module 205 is configured to input the hierarchical early warning information and the dynamic risk map into an intelligent decision support system to generate an emergency response suggestion, so as to obtain a customized emergency response plan;
And the updating module 206 is configured to perform system evaluation and knowledge accumulation based on the execution data of the emergency response plan, and perform parameter updating on the environmental risk multi-modal identification model to obtain an optimized risk monitoring system.
Through the collaborative cooperation of the components, comprehensive, continuous and real-time monitoring of environmental risks is realized by arranging multi-source sensor networks at a coal gangue storage yard and the periphery, compared with a traditional single-parameter and low-frequency monitoring method, the monitoring range and depth are remarkably expanded, the problem of data transmission delay caused by a traditional centralized architecture is solved by locally processing environment risk original monitoring data input edge computing nodes, the response time is shortened by 85%, the response speed of a monitoring system to sudden events is greatly improved, the environmental risk multi-mode recognition model constructed based on pre-processing data integrates the recognition functions of pollution risks and geological disaster risks, the limitation of the traditional system on risk monitoring separation design is overcome, the recognition capability of the composite risks is improved, the applied convolutional neural network performs feature extraction and classification on the pollution risk data, the long-period memory network performs time sequence mode analysis on the geological disaster data, the two algorithms are respectively optimized for spatial features and time sequence features, the multi-dimensional characteristics of the environment risk data of the coal gangue storage yard are fully adapted, the hierarchical early warning information and dynamic early warning information based on the risk recognition result are greatly improved, the map response and the map response system is customized based on the map-oriented response and the map-oriented visual response and the map-oriented dynamic response system is conveniently provided, the intelligent risk management system is customized, and the system has high-oriented response and has high-speed response and intelligent risk-oriented response performance is controlled by using the map-oriented system, by analyzing and accumulating knowledge of the execution data of the emergency response, the system realizes self-adaptive optimization and continuously improves risk identification and prevention and control capability.
The system for monitoring the environmental risk of the coal gangue storage yard based on edge calculation in the embodiment of the invention is described in detail from the angle of modularized functional entity in the above figure 2, and the equipment for monitoring the environmental risk of the coal gangue storage yard based on edge calculation in the embodiment of the invention is described in detail from the angle of hardware processing in the following.
Fig. 3 is a schematic structural diagram of a real-time monitoring device for environmental risk of a coal gangue storage yard based on edge calculation, where the real-time monitoring device 300 for environmental risk of a coal gangue storage yard based on edge calculation may generate relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (e.g., one or more processors) and a memory 320, and one or more storage mediums 330 (e.g., one or more mass storage devices) storing application 333 or data 332. Wherein memory 320 and storage medium 330 may be transitory or persistent storage. The program stored on the storage medium 330 may include one or more modules (not shown), each of which may include a series of instruction operations on the edge-based computing of the coal yard environmental risk real-time monitoring device 300. Still further, the processor 310 may be configured to communicate with the storage medium 330 to execute a series of instruction operations in the storage medium 330 on the edge-based computed coal yard environmental risk real-time monitoring device 300 to implement the steps of the edge-based computed coal yard environmental risk real-time monitoring method described above.
The edge-computing-based coal yard environmental risk real-time monitoring device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input/output interfaces 360, and/or one or more operating systems 331, such as Windows Serve, mac OS X, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the configuration of the edge-calculation-based coal refuse yard environmental risk real-time monitoring apparatus shown in fig. 3 does not constitute a limitation on the edge-calculation-based coal refuse yard environmental risk real-time monitoring apparatus provided by the present invention, and may include more or fewer components than those shown, or may combine certain components, or may have different arrangements of components.
The invention also provides a computer readable storage medium, which can be a nonvolatile computer readable storage medium, and can also be a volatile computer readable storage medium, wherein the computer readable storage medium stores instructions, and when the instructions run on a computer, the instructions cause the computer to execute the steps of the coal gangue storage yard environment risk real-time monitoring method based on edge calculation.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions for causing a real-time monitoring device (which may be a personal computer, a server, or a network device, etc.) for environmental risk of a coal gangue dump based on edge calculation to perform all or part of the steps of the method according to the embodiments of the present invention. The storage medium includes a U disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
The foregoing embodiments are merely for illustrating the technical solution of the present invention, but not for limiting the same, and although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that modifications may be made to the technical solution described in the foregoing embodiments or equivalents may be substituted for parts of the technical features thereof, and that such modifications or substitutions do not depart from the spirit and scope of the technical solution of the embodiments of the present invention in essence.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510730406.3A CN120235461B (en) | 2025-06-03 | 2025-06-03 | Coal gangue storage yard environment risk real-time monitoring method and system based on edge calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510730406.3A CN120235461B (en) | 2025-06-03 | 2025-06-03 | Coal gangue storage yard environment risk real-time monitoring method and system based on edge calculation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120235461A CN120235461A (en) | 2025-07-01 |
| CN120235461B true CN120235461B (en) | 2025-08-26 |
Family
ID=96158758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510730406.3A Active CN120235461B (en) | 2025-06-03 | 2025-06-03 | Coal gangue storage yard environment risk real-time monitoring method and system based on edge calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120235461B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN121186336A (en) * | 2025-11-25 | 2025-12-23 | 寿光市圣城经纬测绘有限公司 | Rapid detection method for soil pollution |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119559771A (en) * | 2025-02-05 | 2025-03-04 | 河南德力劲特智能装备有限公司 | Real-time monitoring and early warning system for geological disasters based on the Internet of Things and artificial intelligence |
| CN120111453A (en) * | 2025-02-28 | 2025-06-06 | 重庆化工职业学院 | A method and system for environmental cloud data communication based on edge computing |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111145064B (en) * | 2019-12-19 | 2024-03-29 | 北京工业大学 | Dynamic emergency early warning assessment and decision support method and system for sudden air pollution accidents |
| CN116090219A (en) * | 2023-01-13 | 2023-05-09 | 中国矿业大学 | GIS risk management and control system and method for mine area river basin pollutant migration |
-
2025
- 2025-06-03 CN CN202510730406.3A patent/CN120235461B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119559771A (en) * | 2025-02-05 | 2025-03-04 | 河南德力劲特智能装备有限公司 | Real-time monitoring and early warning system for geological disasters based on the Internet of Things and artificial intelligence |
| CN120111453A (en) * | 2025-02-28 | 2025-06-06 | 重庆化工职业学院 | A method and system for environmental cloud data communication based on edge computing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120235461A (en) | 2025-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118840829B (en) | Geological disaster risk evaluation method and system based on machine learning | |
| CN118052450A (en) | River channel water quality real-time monitoring platform | |
| CN119848467B (en) | Asphalt pavement health state monitoring system and method | |
| CN118014185A (en) | An online monitoring system for municipal pipe network health based on big data | |
| CN118840674B (en) | A method for processing ecological environment data for ecological restoration | |
| CN120319376B (en) | Method and system for predicting performance degradation of anti-oxidation barrier layer based on multi-source data | |
| CN120235461B (en) | Coal gangue storage yard environment risk real-time monitoring method and system based on edge calculation | |
| CN120031455B (en) | Safety monitoring method and system for large oil storage tanks in ports based on GIS technology | |
| CN120408490A (en) | Bridge displacement remote intelligent monitoring method, system and computer equipment | |
| CN120537992B (en) | Intelligent monitoring instruments and methods for multi-parameter fusion analysis in natural gas pipeline operation | |
| CN120870449A (en) | Particulate matter and ozone source monitoring method and system based on distributed sensor | |
| CN119721857A (en) | A method for evaluating the ecological restoration effect of slightly polluted rivers | |
| CN120410249A (en) | A river health assessment method suitable for water-scarce cities in northern China | |
| CN120611947A (en) | Power engineering construction project full process management and control system | |
| CN120808577A (en) | Point landslide debris flow potential degree self-adaptive monitoring and early warning method | |
| CN120579375A (en) | A method for detecting anomalies of road and bridge parameters | |
| CN120146395A (en) | Auditing methods and systems for natural resources | |
| CN120316881A (en) | Digital quality tracking method and system for prefabricated slope support engineering | |
| CN120744675B (en) | Watershed water environment monitoring and emergency pollution quick tracing method and system based on water quality pollution detection | |
| Cheng et al. | Development of intelligent monitoring system for soil erosion in pipeline engineering based on deep learning | |
| CN119625959B (en) | Slope early warning method and system based on multi-coupling data analysis | |
| CN119474767B (en) | Hydrologic large model construction method integrating multisource fusion and dynamic optimization | |
| Zhang et al. | Multidimensional Data Fusion and Multi-algorithm Stacking for Renewable Power Forecasting | |
| CN121092324A (en) | Water quality monitoring method and system based on edge calculation | |
| CN121415296A (en) | A method, system, equipment and medium for risk prevention and control of on-site operation video surveillance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |