CN115729186B - A safe state multi-modal real-time intelligent management and control master machine, method and system - Google Patents

Abstract

The invention discloses a safety state multi-mode real-time intelligent control master machine, a method and a system, wherein the safety state multi-mode real-time intelligent control master machine comprises a parameter selection module, a data transceiver module and a control module, wherein the parameter selection module is used for a user to select monitoring parameters, the data transceiver module is used for receiving real-time monitoring values of the monitoring parameters selected by the user and sending the real-time monitoring values to the control module, the control module is embedded with a plurality of multi-level multi-mode fusion evaluation early warning models corresponding to different monitoring scenes, the multi-level multi-mode fusion evaluation early warning models corresponding to the monitoring scenes are used for determining the monitoring scenes according to the monitoring parameters selected by the user, and the safety evaluation early warning is carried out by combining the real-time monitoring values of the monitoring parameters. The invention develops the real-time intelligent management and control equipment with the safety state of changing parameters and changing modes across scenes based on the technology of the Internet of things, can expand the parameter types when scene monitoring parameters change, is suitable for the mode expansion of multi-level safety multi-mode management and control, and provides multiple safety guarantees by multi-mode fusion evaluation and early warning.

Description

A safe state multi-mode real-time intelligent management and control master machine, method and system

technical field

The invention belongs to the field of safety monitoring, and more specifically relates to a safe state multi-mode real-time intelligent control master machine, method and system.

Background technique

Safety monitoring and monitoring is a technical means of safety production in all walks of life, and the intelligence level of safety monitoring equipment determines the effect of safety production control. At present, the intelligence of safety production monitoring equipment in various fields needs to be further improved. The main technical problems existing in the existing production safety monitoring and monitoring equipment include: (1) The contradiction between relatively perfect monitoring of production safety status and insufficient real-time analysis and comprehensive evaluation of production safety status. (2) The intelligence of monitoring and the decision-making of safety production control mainly rely on manpower. (3) The contradiction between the multi-parameters of safety production monitoring and the multi-level and multi-modal control of safety production status in real time.

Therefore, there is an urgent need for a real-time intelligent management and control master machine with selectable parameter types, multi-level security status and multi-modality that adapts to changing scenarios of production and living activities and equipment and facility characteristics.

Contents of the invention

Aiming at the above defects or improvement needs of the prior art, the present invention provides a safe state multi-modal real-time intelligent management and control master machine, method and system, the purpose of which is to use the technology of the Internet of Things to develop The real-time intelligent management and control equipment of the security status can expand the parameter types when the scene monitoring parameters change, and at the same time adapt to the modal expansion of multi-level security and multi-modal control, multi-modal fusion evaluation and early warning. As a result, the intelligence and versatility of safety monitoring instruments will be promoted, the threshold of safety supervision will be lowered, the cost of safety supervision will be saved, the breadth of scientific monitoring and monitoring technology services for safety production will be improved, and safety production will be promoted.

In order to achieve the above object, according to the first aspect of the present invention, a safe state multi-modal real-time intelligent control master machine is provided, including: a parameter selection module, a data transceiver module, and a control module;

The parameter selection module is used for users to select monitoring parameters;

The data transceiver module is used to receive the real-time monitoring value of the monitoring parameter selected by the user and send it to the control module;

The control module embeds multi-level and multi-modal fusion evaluation and early warning models corresponding to multiple different monitoring scenarios, and is used to determine the monitoring scene according to the monitoring parameters selected by the user, and call the multi-level and multi-modal fusion evaluation and early warning models corresponding to the monitoring scenarios. The model, combined with the real-time monitoring value of the monitoring parameters, performs safety assessment and early warning;

Wherein, the multi-level multi-modal fusion assessment early warning model includes a bottom-up node-level mode fusion layer, a process-level mode fusion layer and a process-level mode fusion layer;

The node-level modal fusion layer is used to determine whether each node has a security risk according to the real-time monitoring value and the preset threshold of the monitoring parameters of each node, and if so, an alarm is issued, and the security assessment score of each node is calculated;

The process-level modal fusion layer is used to calculate the safety assessment score of each process according to the security assessment score of each node, so as to determine whether there is a safety risk in each process, and to report to the police if it exists;

The process-level modal fusion layer is used to judge whether there is a safety risk in each process according to the safety assessment score of each process, and to issue an alarm if there is.

Preferably, the node-level mode fusion layer adopts a series fusion model: when any monitoring parameter of a node exceeds a preset threshold, it is determined that the node has a security risk, and the security assessment score of the node is 100, or, the node The security assessment score for , is determined based on the percentage that exceeds a preset threshold.

Preferably, the process-level modal fusion layer adopts a weighted fusion model: calculate the weighted score of the process according to the weights and scores of the nodes included in the process, and if it exceeds a preset risk threshold, it is determined that the process has a safety risk.

Preferably, the process-level modal fusion layer adopts a series fusion model: when any process of the process has a safety risk, it is determined that the process has a safety risk.

Preferably, a modality extension module is also included, which is used for users to expand or select monitoring parameters, nodes, procedures or processes.

According to the second aspect of the present invention, a multi-modal real-time intelligent management and control method for a safe state is provided, which is applied to the control module of the parent machine as described in the first aspect, including:

Determine the monitoring scene according to the monitoring parameters selected by the user, call the multi-level and multi-modal fusion evaluation and early warning model corresponding to the monitoring scene, and combine the real-time monitoring values of the monitoring parameters to perform safety assessment and early warning;

Wherein, the multi-level multi-modal fusion assessment early warning model includes a bottom-up node-level mode fusion layer, a process-level mode fusion layer and a process-level mode fusion layer;

The node-level modal fusion layer is used to determine whether each node has a security risk according to the real-time monitoring value and the preset threshold of the monitoring parameters of each node, and if so, an alarm is issued, and the security assessment score of each node is calculated;

The process-level modal fusion layer is used to calculate the safety assessment score of each process according to the security assessment score of each node, so as to determine whether there is a safety risk in each process, and to report to the police if it exists;

The process-level modal fusion layer is used to judge whether there is a safety risk in each process according to the safety assessment score of each process, and to issue an alarm if there is.

According to the third aspect of the present invention, a security state multi-modal real-time intelligent management and control system is provided, including: a computer-readable storage medium and a processor;

The computer-readable storage medium is used to store executable instructions;

The processor is configured to read executable instructions stored in the computer-readable storage medium, and execute the method as described in the second aspect.

Generally speaking, compared with the prior art, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

The security state multi-modal real-time intelligent management and control base machine provided by the present invention uses Internet of Things technology to develop real-time intelligent management and control equipment with cross-scenario variable parameters and variable modes. When the scene monitoring parameters change, the parameter types can be expanded, At the same time, it adapts to the mode extension of multi-level security and multi-modal management and control, multi-modal fusion assessment and early warning, and provides multiple security guarantees. As a result, the intelligence and versatility of safety monitoring instruments will be promoted, the threshold of safety supervision will be lowered, the cost of safety supervision will be saved, the breadth of scientific monitoring and monitoring technology services for safety production will be improved, and safety production will be promoted.

Description of drawings

Fig. 1 is one of the structural schematic diagrams of the multi-modal real-time intelligent management and control master machine in a safe state provided by an embodiment of the present invention;

Fig. 2 is the second structural diagram of the multi-modal real-time intelligent management and control master machine in a safe state provided by the embodiment of the present invention;

Fig. 3 is a schematic diagram of the multi-level multi-modal fusion data flow of the multi-level multi-modal fusion evaluation early warning model provided by the embodiment of the present invention;

Fig. 4 is a schematic diagram of the application process of a multi-modal real-time intelligent management and control master machine in a safe state according to the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not constitute a conflict with each other.

In various factories and engineering safety monitoring and testing, technicians usually select monitoring parameters based on the scene safety risk assessment results, deploy sensors at sensitive positions of safety risks, and use acquisition and analysis instruments to collect corresponding types of sensor data. Based on experience and related norms and standards, the monitoring single-parameter status and multi-parameter object security status assessment and early warning standards based on real-time data are formulated, and monitoring personnel analyze the monitoring data based on this to carry out security risk management and control. Among them, it integrates parameter status, multi-modal status assessment and early warning of various security risks in scenarios. High cost and insufficient real-time performance. There is an urgent need to establish intelligent multi-parameter and multi-modal security state real-time monitoring and control equipment for the real-time nature of security risk control, multiple security parameters, and multiple security modes.

Based on this, the present invention provides a safe state multi-modal real-time intelligent control master machine, method and system.

For ease of understanding, the relevant definitions are first explained as follows:

Modality: The direct or indirect way in which the subject describes and observes the properties of the object in order to achieve a specific goal. A way for the subject to obtain object attribute target information or a means for the subject to achieve the object is a modality.

Multimodality: The multi-attribute characteristics of the object or the multiple means or ways that the subject describes the object are multimodal.

Safety: Refers to the state in which the production and life scenes of human beings will not endanger human life, health and safety.

Safety risks: Refers to the root causes or factors in human production and living scenarios that may endanger human life, health and safety.

Safety risk mode: refers to factors that may endanger human life, health and safety in human production and living scenarios. Each type of factor is a mode, such as the risk mode of explosion accidents in process factory production and the risk of collapse accidents in factory construction. modal etc.

Parameters of the safety risk mode: various parameters that may lead to the occurrence of the safety risk mode, such as factors affecting the risk mode of explosion accidents: flammable and explosive materials, leakage of facilities storing flammable and explosive materials, poor site ventilation, excessive temperature higher.

Scenario: A general term for production and living activities and their places.

An embodiment of the present invention provides a safe state multi-modal real-time intelligent management and control master machine, as shown in Figure 1, including:

Parameter selection module, data transceiver module, control module and alarm module;

The parameter selection module is used for users to select monitoring parameters;

The data transceiver module is used to receive the real-time monitoring value of the monitoring parameter selected by the user and send it to the control module;

The control module embeds multi-level and multi-modal fusion evaluation and early warning models corresponding to multiple different monitoring scenarios, and is used to determine the monitoring scene according to the monitoring parameters selected by the user, and call the multi-level and multi-modal fusion evaluation and early warning models corresponding to the monitoring scenarios. Model (parameter-node-process-process multi-level and multi-modal fusion assessment early warning model), combined with the real-time monitoring values of the monitoring parameters, to perform safety assessment and early warning;

Wherein, the multi-level multi-modal fusion assessment early warning model includes a bottom-up node-level mode fusion layer, a process-level mode fusion layer and a process-level mode fusion layer;

The node-level modal fusion layer is used to determine whether each node has a security risk according to the real-time monitoring value and the preset threshold of the monitoring parameters of each node, and if so, an alarm is issued, and the security assessment score of each node is calculated;

The process-level modal fusion layer is used to calculate the safety assessment score of each process according to the security assessment score of each node, so as to determine whether there is a safety risk in each process, and to report to the police if it exists;

The process-level modal fusion layer is used to judge whether there is a safety risk in each process according to the safety assessment score of each process, and to issue an alarm if there is.

Preferably, the node-level mode fusion layer adopts a series fusion model: when any monitoring parameter of a node exceeds a preset threshold, it is determined that the node has a security risk, and the security assessment score of the node is 100, or, the node The security assessment score for , is determined based on the percentage that exceeds a preset threshold.

For example, if the monitoring value of a certain monitoring parameter exceeds 10 percent of the preset threshold, the security assessment score of the node is 10, and if the monitoring value of a monitoring parameter exceeds 50 percent of the preset threshold, the security assessment score is 50. The more the preset threshold is exceeded, the higher the security assessment score, which means the higher the security risk.

Preferably, the process-level modal fusion layer adopts a weighted fusion model: calculate the weighted score of the process according to the weights and scores of the nodes included in the process, and use it as the safety assessment score of the process. If it exceeds the preset risk threshold, It is determined that the process has a safety risk.

Preferably, the process-level modal fusion layer adopts a series fusion model: when any process of the process has a safety risk, it is determined that the process has a safety risk.

Preferably, the intelligent management and control master machine also includes a modal expansion module, which is used for users to expand or select monitoring parameters, nodes, procedures or processes, that is, for users to perform parameter-like modals and nodes, procedures, The multi-level and multi-modal expansion or selection of the process is for users to personalize the multi-level and multi-modal fusion assessment early warning model.

Further, the alarm module is an acousto-optic alarm module, which can be installed at each monitoring parameter, each node, each process, and each process, and audible and visual alarms can be provided when there is a safety risk.

Specifically, the safe state multi-modal real-time intelligent control master includes: a control module, a parameter selection module, a modal expansion module, and a real-time data sending and receiving module; modules, parameter selection modules and modal extension modules. As shown in Figure 2, the functions of the real-time intelligent monitoring of the security state provided by the present invention include: monitoring parameter type selection function, multi-level security mode expansion function, data sending and receiving function, multi-level and multi-modal fusion analysis, evaluation, early warning functions, and functional modules for realizing the above functions, as shown in Figure 1.

The control module embeds multiple parameters, nodes, procedures, and multi-level safety status and multi-modal fusion assessment and early warning models corresponding to multiple different monitoring scenarios, which are used to determine the monitoring scenario according to the monitoring parameters selected by the user, and call the corresponding monitoring scenario. Monitoring parameters-node-process-process multi-level multi-modal fusion assessment early warning model, and adapt monitoring parameters to the model for safety assessment and early warning.

That is to say, the control module includes multi-level multi-modal fusion analysis, evaluation and early warning modules (that is, multi-modal fusion analysis modules at all levels of scene-parameter-node-process-process, security assessment at all levels of scene-parameter-node-process-process Early warning module) is used to fuse the security multi-modal state parameters of each level according to the multi-modal parameter state data of each level scene, according to the multi-modal fusion model embedded in the corresponding monitoring scene of the main machine, and then according to the scene embedded in the main machine The multi-level status evaluation level model judges the multi-mode real-time status level of each level of security, and compares it with the scene security warning threshold embedded in the host machine to give an early warning as needed. as shown in picture 2.

Among them, the corresponding relationship between monitoring parameters-monitoring scenarios-multi-level and multi-modal fusion assessment and early warning models has been written into the control module in advance. Therefore, the control module can determine the monitoring scene according to the monitoring parameters selected by the management and control subject, and call the corresponding multi-level multi-modal fusion evaluation early warning model, and the multi-level multi-modal fusion evaluation early warning model also includes the monitoring parameters. Safety thresholds (ie preset thresholds) and preset risk thresholds for each process. The above corresponding relationship can be written into the user manual.

The parameter selection module is used for the user to select the parameter type to be monitored, and can adapt to the change of security monitoring parameters in variable scenarios. By embedding the parameter type library of the parent machine, the user can select the scene security monitoring parameters that the parent machine needs to control; multi-level The security mode expansion module is used for users to expand the multi-level security status mode of parameters, nodes, procedures, and processes that adapt to the scene.

The real-time status data transceiving module is used to receive the real-time sensing parameter status data of the selected parameter type, and at the same time transmit relevant data to the upper device; that is, the data transceiver module is used to receive the intelligent monitoring parameter status of the selected type of parameters Data, send the related data of the master machine to the upper device.

As shown in Figure 3, the parameter, node, process, and process multi-level security state multi-modal fusion assessment and early warning model includes a bottom-up node-level modal fusion layer, a process-level modal fusion layer, and a process-level modal The fusion layer is used for multi-level security and multi-modal fusion and multi-level security status assessment and early warning.

The control module is connected to the parameter selection module, which is used to select the parameter type that needs to be controlled. After the parameter type is selected, the control module drives the receiving module to receive the real-time data of the parameter status sent by the scene intelligent perception sensor of the selected parameter, and the control module automatically sends the real-time data of the parameter status. Adapted to the corresponding parameter-level multi-modal, node, process, and process multi-modal fusion model, edge computing evaluates the multi-level security and multi-modal status and gives an early warning, and drives the data upload module to upload data. The control module uses a single-chip microcomputer as a device, and embeds a parameter type library, a scene mode model library, a scene multi-modal fusion model library, a scene security multi-modal evaluation model library, a scene security state evaluation level model library, and a data identification library.

The multi-level multi-modal fusion function shown in Figure 2 includes a multi-modal fusion model library embedded in the parameters, nodes, procedures, and process levels of the control module. Multi-level security status assessment and early warning functions, including embedded control module scenarios and parameters, nodes, procedures, and process-level security status assessment level model libraries. That is, the control module is embedded with a parameter type database, a multi-layer modal database of scenarios and parameters, nodes, procedures, and processes, a multi-level scenario security multi-modal fusion database, and a scenario process security status assessment and early warning model library.

As shown in Figure 4, the control subject selects the parameters according to the security monitoring parameters of the control scene, and determines the security mode at each level of the scene-parameter-node-process-process, the control module receives parameter selection and modal extension information, and the drive data receiving module receives The scene parameter status data sent by the scene smart sensor in real time adapts to the security mode model of each level of the scene, drives the multi-modal fusion analysis of the scene-parameter-node-process-process level, judges the security status of each level of the scene, and warns of exceeding the threshold. Then upload the master machine to receive data and decision data.

Furthermore, both the parameter selection module and the modal expansion module can use the keyboard connected to the port of the single-chip microcomputer as a device, and realize parameter selection and modal expansion through mapping between it and parameter types and modalities.

The data receiving module is equipped with a wireless transceiver device and is driven by the control module. When the parameters and modes of the scene are determined, the control module drives the data receiving module to receive the real-time data of the parameter status sent by the intelligent sensor for controlling the scene parameters. After the real-time data is processed and a multi-level and multi-modal security management and control decision is formed, the decision information and the received data are wirelessly uploaded to the upper device.

The scene-parameter-node-process-process multi-modal fusion analysis module at each level receives the scene real-time monitoring parameter status data transmitted by the data transceiver module, and integrates the multi-modality of each level according to the parameter multi-modal fusion model. Parameter-level multi-modal data, followed by node multi-modal and process multi-modal fusion analysis.

The security assessment and early warning module at each level of scene-parameter-node-process-process, according to the multi-modal fusion analysis results of scene-parameter-node-process-process at each level, according to the security of each level of scene, parameter, node, process and process The state assessment and early warning model performs security state assessment and early warning at each level in turn.

The intelligent management and control master machine provided by the present invention selects the middle-end PIC16F877 single-chip microcomputer as the chip of the expansion module of the multi-mode real-time intelligent management and control master machine in a safe state, and the WIFI communication module for sending and receiving data is connected to the B port, and the A port is connected to the project scene selection setting and monitoring parameter selection setting , C port is connected to the scene security multi-modal fusion model selection setting port, D port is connected to the mode parameter selection setting port of the management and control level, and E port is connected to the voice warning and light prompt. The control module chip is embedded with a parameter type database, a safety mode database, a safety multi-mode fusion model library and a safety state assessment model library.

The security state multi-modal real-time intelligent control base machine provided by the present invention has parameter type selection function, mode expansion function, multi-mode fusion and security state evaluation and early warning function, and data sending and receiving function; in order to realize the above-mentioned function device, embedded parameter type information library, modal information library, multi-modal fusion model library, security status assessment and early warning module library. The production method is to realize the expansion of parameter types and security modes by embedding the parameter type information library and modal information library of the parent machine, using the internal communication channel of the parent machine and the designed parameter mode selection key; providing a multi-modal fusion model library embedded in the parent machine, Use edge computing to integrate multi-modal security from parameter-node-process-process layer by layer; by embedding the supervision project, scene and parameter status evaluation model library of the parent machine, use edge computing to analyze and evaluate the status evaluation and warning of supervision objects, and realize safety supervision data information application. When the intelligent management and control master machine is in use, it can start normal use after selecting the monitoring parameters of the receiving supervision object according to the scene, and determining the various modes of the scene according to the safety supervision content of each supervision level. The supervisory master machine control module drives the data sending and receiving module to receive the monitoring parameter status data of the supervised object, the mode fusion module fuses the safety multi-modal data step by step according to the level, the safety assessment and early warning module evaluates the level of security status level by level and gives an early warning, and the data is sent and received The module uploads relevant information.

The application of the safe state multi-modal real-time intelligent management and control master machine provided by the embodiment of the present invention will be further described below by taking the ceiling lifting operation of the building construction tower as an example.

The goal of the tower crane lifting operation is to disconnect the tower top that was originally connected to the standard section of the tower body and jack up a standard section height, add a new standard section, connect the lower part with the tower body, and the upper part with the tower top, so as to realize the increase of the tower crane. high. Its operation process is complex and high-risk, and it has become the first high-risk activity in the construction process. The main monitoring parameters involved in its safe operation include: 1) Meteorological parameters: wind speed, direction, temperature, humidity and other indicators; 2) Quality parameters of operation activities: such as cylinder jacking displacement, speed indicators; 3) Working quality parameters of each process : such as the climbing claws in place and the safety status of the beam; 4) The safety condition parameters of each process operation: such as the deflection of the tower crane, the connection status of the standard section of the tower crane, etc. The main procedures included in the lifting operation process of the tower crane are: 1) Preparation (including inspection and confirmation of equipment and facilities status and safe operating conditions); 2) Trial lifting; 3) First lifting; 4) Lifting the beam of the jacking cylinder; 5 ) Repeat steps 3) and 4) until one standard section can be accommodated in the jacking space; 6) Connect the newly added standard section; 7) After the safety check is confirmed, the operation of adding a standard section is completed. Tower crane jacking process operation safety monitoring supervision level, including front-end parameter layer (station work quality and safety monitoring - front-end parameter single index status), parameter multi-modal fusion layer (station work quality and safety monitoring - station multi-indicator comprehensive status ), process multimodal fusion layer (quality and safety control at the operation team level - multi-station and multi-indicator comprehensive status), process multi-modal fusion layer (project-level quality and safety supervision - multi-process multi-indicator comprehensive quality and safety status).

Utilize the multi-mode real-time intelligent management and control master machine in the safe state provided by the present invention, select the parameters monitored by intelligent sensors on site: ambient temperature, humidity, wind speed, wind direction parameters, tower crane tilt parameters, jacking cylinder displacement and speed parameters and jacking operation related parameters Component state logic switch parameters, the control module determines the monitoring scene as the tower crane lifting operation according to the above parameters, and calls the corresponding multi-level and multi-modal fusion evaluation early warning model.

1) The front-end parameter layer, the first jacking process includes three stations: oil cylinder operation, jacking beam operation, and movable claw operation. Here, the oil cylinder operation station is taken as an example: including two parameters of oil cylinder piston displacement and displacement speed. Displacement parameter value range: step spacing h+50mm displacement speed 0-15mm/s. If the warning is exceeded, the cylinder operator should correct it immediately.

2) The parameter multi-modal fusion layer (that is, the node multi-modal fusion layer), still taking the oil cylinder operation station in the first lifting process as an example: two parameter states of displacement and speed need to be considered. Usually, any parameter exceeding the standard will be Therefore, a series model can be adopted here, that is, take the most unfavorable parameter state and perform parameter fusion, that is, as long as one index exceeds the limit, it will issue a parameter multi-modal state warning and output the data identification and value of the problem sensor. The station operator must correct it immediately.

3) Process multi-modal fusion layer, taking the first jacking process as an example, including oil cylinder operation station 1, jacking beam operation station 2, and movable claw operation station 3. Oil cylinder operation station Operation control The extension length of the oil cylinder piston is slightly greater than the step height of the standard section and returns slightly, the jacking beam operation station 2 is in charge of locking the left and right pins of the beam (logic state index value is 1), the movable crawling claw operation station 3 Operate the movable crawling claw, and after the movable crawling claw has crossed the step (logic state index value 0), operate it close to the standard pitch chord rod, and when the cylinder piston retracts, it will accurately fall into the step (logic state index value 1). Multi-parameter fusion layer: station 1 is two parameters of oil cylinder piston displacement and speed; station 2 is two logic switch parameters of left and right pin shafts of beam; station 3 is two logic switch parameters of left and right movable claws. In addition to the quality and safety of each station, the multi-modal safety of this process, the environmental wind speed and the tilting state of the tower body during the entire oil cylinder jacking period directly affect the safety of this process. Therefore, for these 3 stations, each station Both include 2 safety parameters of work quality plus 2 safety parameters of ambient wind speed and tower inclination, a total of 4 safety parameters. From the perspective of process safety, any failure of any one of the 4 parameters in each station will cause the station to work Failed, therefore, all 3-station multi-parameter fusions should adopt a 4-parameter series model. However, the multimodal fusion of the first lifting process, including the multimodal fusion of 3 stations, the failure of the work quality and safety of any one of the 3 stations will lead to the failure of this process, but the jacking The work quality and safety of the beam operation station 2 and the movable crawler operation station 3 are fatal, and usually lead to major production safety accidents. Therefore, the multimodal fusion including the first jacking process can adopt a weighted model, in which the work Bit 1 has a weight of 0.2, and bits 2 and 3 are equally important, each with a weight of 0.4.

Taking the jacking beam lifting process as an example, it includes oil cylinder operation station 1, jacking beam operation station 2, and movable claw station 3. The parameters of each station are the same as those of the first jacking process, but The parameter thresholds are different. In the lifting process of the jacking beam: the multi-parameters of station 1 are the cylinder piston displacement (with 1 step h reduced to -50-0mm), the speed parameter (0-15mm/s); the multi-parameters of station 2 are the left and right beams The logic state of the pin shaft (1→0→1); the multi-parameter of station 3 is the logic state of the left and right movable claws (always 1). And considering the environmental wind speed and tower tilt parameters that affect the safety of the process operation, the quality and safety of the work station adopts a 4-parameter series model for multi-modal fusion. It is still recommended to use a weighted model for multi-modal fusion in this process. The state of the active claws in this process is the basis and the most important. Therefore, the weights of the three stations are respectively 0.2, 0.2, and 0.6.

4) Process multi-modal fusion layer. The process here only considers the first two processes of jacking and lifting of the jacking beam as an example. Process quality and safety includes factors such as the multi-modal state of process quality and safety, as well as the connection status of tower crane structural components that affect the quality and safety of process operations, the slewing braking status of tower cranes, and the lifting operators with certificates. Therefore, the multi-modal integration of process safety Among them, it is necessary to increase the process operation safety condition index including three parameters including structural component connection, slewing brake, and operating personnel with certificates, which are much safer than the first jacking process and the jacking beam lifting process according to the series model. Modal fusion. The process multi-modal safety includes the safety of the first jacking process and the lifting process of the jacking beam, and adopts the series fusion model.

An embodiment of the present invention provides a multi-modal real-time intelligent management and control method for a security state, which is applied to the control module of the management and control master machine described in any of the above embodiments, including:

Determine the monitoring scene according to the monitoring parameters selected by the user, call the multi-level and multi-modal fusion evaluation and early warning model corresponding to the monitoring scene, and combine the real-time monitoring values of the monitoring parameters to perform safety assessment and early warning;

Wherein, the multi-level multi-modal fusion assessment early warning model includes a bottom-up node-level mode fusion layer, a process-level mode fusion layer and a process-level mode fusion layer;

The node-level modal fusion layer is used to determine whether each node has a security risk according to the real-time monitoring value and the preset threshold of the monitoring parameters of each node, and if so, an alarm is issued, and the security assessment score of each node is calculated;

The process-level modal fusion layer is used to calculate the safety assessment score of each process according to the security assessment score of each node, so as to determine whether there is a safety risk in each process, and to report to the police if it exists;

The process-level modal fusion layer is used to judge whether there is a safety risk in each process according to the safety assessment score of each process, and to issue an alarm if there is.

An embodiment of the present invention provides a security state multi-modal real-time intelligent management and control system, which is characterized in that it includes: a computer-readable storage medium and a processor;

The computer-readable storage medium is used to store executable instructions;

The processor is configured to read executable instructions stored in the computer-readable storage medium, and execute the methods described in the above-mentioned embodiments.

It is easy for those skilled in the art to understand that the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention, All should be included within the protection scope of the present invention.

Claims (6)

1. A safe state multi-mode real-time intelligent management and control master machine, characterized in that it includes: a parameter selection module, a data transceiver module, and a control module; The parameter selection module is used for users to select monitoring parameters, which can adapt to changes in safety monitoring parameters in variable scenarios, and is used for users to select scene safety monitoring parameters that need to be controlled by the parent machine by embedding the parameter type library of the parent machine; The data transceiver module is used to receive the real-time monitoring value of the monitoring parameter selected by the user and send it to the control module; The control module embeds multi-level and multi-modal fusion evaluation and early warning models corresponding to multiple different monitoring scenarios, and is used to determine the monitoring scene according to the monitoring parameters selected by the user, and call the multi-level and multi-modal fusion evaluation and early warning models corresponding to the monitoring scenarios. model, combining the real-time monitoring values of the monitoring parameters, to perform safety assessment and early warning; wherein, after the parameter type is selected, the control module drives the receiving module to receive the real-time data of the parameter status sent by the scene intelligent perception sensor of the selected parameter, and the control module automatically converts the parameter status The real-time data of the state is adapted to the corresponding multi-level and multi-modal fusion evaluation and early warning model, and the edge computing evaluates the multi-level security and multi-modal status and gives an early warning; Wherein, the multi-level multi-modal fusion assessment early warning model includes a bottom-up node-level mode fusion layer, a process-level mode fusion layer and a process-level mode fusion layer; The node-level modal fusion layer is used to determine whether each node has a security risk according to the real-time monitoring value and the preset threshold of the monitoring parameters of each node, and if so, an alarm is issued, and the security assessment score of each node is calculated; The process-level modal fusion layer is used to calculate the safety assessment score of each process according to the security assessment score of each node, so as to determine whether there is a safety risk in each process, and to report to the police if it exists; The process-level modal fusion layer is used to judge whether there is a safety risk in each process according to the safety assessment score of each process, and to report to the police if it exists; It also includes a modality expansion module for users to expand or select monitoring parameters, nodes, procedures or processes; Scene refers to the general term of production and living activities and their places. 2. The management and control master machine according to claim 1, wherein the node-level mode fusion layer adopts a series fusion model: when any monitoring parameter of a node exceeds a preset threshold, it is determined that the node has a security risk, and the The security assessment score of the node is 100, or, the security assessment score of the node is determined according to the percentage exceeding the preset threshold. 3. The management and control master machine according to claim 2, wherein the process-level modal fusion layer adopts a weighted fusion model: calculate the weighted score of the process according to the weights and scores of the nodes included in the process, if it exceeds the preset If the risk threshold is set, it is determined that there is a safety risk in the process. 4. The management and control master machine according to claim 3, wherein the process-level modal fusion layer adopts a serial fusion model: when any process of the process has a security risk, it is determined that the process has a security risk. 5. A safe state multi-modal real-time intelligent control method, applied to the control module of the control master machine according to any one of claims 1-4, characterized in that, comprising: Determine the monitoring scene according to the monitoring parameters selected by the user, call the multi-level and multi-modal fusion evaluation and early warning model corresponding to the monitoring scene, and combine the real-time monitoring values of the monitoring parameters to perform safety assessment and early warning; Wherein, the multi-level multi-modal fusion assessment early warning model includes a bottom-up node-level mode fusion layer, a process-level mode fusion layer and a process-level mode fusion layer; The node-level modal fusion layer is used to determine whether each node has a security risk according to the real-time monitoring value and the preset threshold of the monitoring parameters of each node, and if so, an alarm is issued, and the security assessment score of each node is calculated; The process-level modal fusion layer is used to calculate the safety assessment score of each process according to the security assessment score of each node, so as to determine whether there is a safety risk in each process, and to report to the police if it exists; The process-level modal fusion layer is used to judge whether there is a safety risk in each process according to the safety assessment score of each process, and to issue an alarm if there is. 6. A security state multi-modal real-time intelligent control system, characterized in that it comprises: a computer-readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is configured to read executable instructions stored in the computer-readable storage medium, and execute the method as claimed in claim 5 .