CN119341127A - An intelligent power distribution cabinet for renewable energy access - Google Patents
An intelligent power distribution cabinet for renewable energy access Download PDFInfo
- Publication number
- CN119341127A CN119341127A CN202411750939.XA CN202411750939A CN119341127A CN 119341127 A CN119341127 A CN 119341127A CN 202411750939 A CN202411750939 A CN 202411750939A CN 119341127 A CN119341127 A CN 119341127A
- Authority
- CN
- China
- Prior art keywords
- distribution cabinet
- power
- power generation
- power distribution
- renewable energy
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
- H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- H02J2103/30—
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Environmental & Geological Engineering (AREA)
- Power Engineering (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
本发明涉及一种可再生能源接入的智能配电柜，以特定时间为步长，监测并收集当地天气预报未来一段时间序列的实时天气环境数据；采用主成分分析方法提取实时天气环境数据的主要特征；通过历史数据建立天气环境数据主要特征为父节点，配电柜发电情况为子节点的贝叶斯网络模型，将实时天气环境数据输入贝叶斯网络模型，预测当前配电柜发电情况；根据当前配电柜发电情况，对配电柜的供电模式进行控制。本发明解决了现有技术难以根据天气环境因素的变化优化配电柜的控制策略，导致能源利用效率降低的问题。本发明结合天气预报对未来发电情况进行预测，实现提前优化配电柜的供电模式，提高配电柜控制效果。
The present invention relates to an intelligent power distribution cabinet connected to renewable energy, which monitors and collects real-time weather environment data of a sequence of a future period of time in the local weather forecast with a specific time as the step length; uses the principal component analysis method to extract the main features of the real-time weather environment data; establishes a Bayesian network model with the main features of the weather environment data as the parent node and the power generation situation of the power distribution cabinet as the child node through historical data, inputs the real-time weather environment data into the Bayesian network model, predicts the current power generation situation of the power distribution cabinet; controls the power supply mode of the power distribution cabinet according to the current power generation situation of the power distribution cabinet. The present invention solves the problem that it is difficult to optimize the control strategy of the power distribution cabinet according to the changes in weather and environmental factors in the prior art, resulting in reduced energy utilization efficiency. The present invention predicts future power generation conditions in combination with weather forecasts, optimizes the power supply mode of the power distribution cabinet in advance, and improves the control effect of the power distribution cabinet.
Description
Technical Field
The invention belongs to the technical field of power distribution cabinets, and relates to an intelligent power distribution cabinet with renewable energy access.
Background
Under the dual pressures of energy crisis and environmental protection, traditional energy faces dilemma, and renewable energy technologies such as solar energy, wind energy, water energy, geothermal energy, biomass energy and the like are continuously developed. Meanwhile, smart grid technology is emerging, and technical support is provided for renewable energy access. Under the background, the intelligent power grid can be efficiently connected into renewable energy sources, and intelligent monitoring and management, fault diagnosis and protection and energy optimization and management of electric power are realized. The method has wide application prospect in the fields of distributed energy systems, intelligent buildings, industrial parks, remote areas for power supply and the like, and provides an effective solution for realizing sustainable development of energy in the future.
The intelligent power distribution cabinet is used as a key component in the intelligent power grid, and plays an important role in distributing, controlling and protecting power in a power distribution link. However, the power generation condition of the renewable energy source can fluctuate along with the change of external factors such as environment, weather and the like, so that the power generation of the power distribution cabinet connected with the renewable energy source shows instability, and further the utilization efficiency of the power energy source is adversely affected. Therefore, in the prior art, the energy inflow needs to be adjusted according to the change of external factors, and the control strategy of the power distribution cabinet is optimized, so that the energy utilization efficiency is improved.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an intelligent power distribution cabinet with renewable energy access, which aims to establish an artificial intelligent model between weather environment factors and renewable energy power generation capacity through historical period big data, acquire weather environment data of a future time sequence through local weather forecast, input the weather environment data into the artificial intelligent model so as to predict the power generation capacity of the power distribution cabinet at the future time, and formulate a control strategy of the power distribution cabinet according to the power generation capacity, so that the energy utilization efficiency is improved.
The aim of the invention can be achieved by the following technical scheme:
the application provides an intelligent power distribution cabinet for renewable energy access, which comprises a monitoring and collecting module, an intelligent prediction module and an optimization control module, wherein the monitoring and collecting module, the intelligent prediction module and the optimization control module are in communication connection, and the intelligent power distribution cabinet comprises the following components:
the monitoring and collecting module is used for monitoring and collecting real-time weather environment data of a time sequence in the future of local weather forecast by taking specific time as step length;
The intelligent prediction module is used for constructing an artificial intelligent model by collecting weather environment data in a historical period and power generation condition data of the renewable energy power distribution cabinet, inputting the real-time weather environment data into the artificial intelligent model and predicting the current power generation condition of the power distribution cabinet;
And the optimizing control module is used for controlling the power supply mode of the power distribution cabinet according to the current power generation condition of the power distribution cabinet.
Further, the weather environment data includes air temperature, humidity, wind speed, illumination intensity, and rainfall.
Further, the weather forecast is obtained through a China weather data network, a business weather data API, a weather station, an official weather forecast website or a weather forecast application program.
Further, the power generation condition data comprises power generation capacity, power generation power, power generation stability index and power generation efficiency, wherein:
The generated energy represents the total electric quantity generated by the intelligent power distribution cabinet accessed by renewable energy sources in a specific time period;
The generated power represents generated energy in unit time;
the power generation stability index is used for measuring the stability of the power generation condition and is determined by standard deviation or variation coefficient of the generated energy or the generated power in a period of time;
The power generation efficiency represents the ratio of the renewable energy source to the electric energy, and is calculated by the ratio of the actual power generation amount to the theoretical maximum power generation amount.
Further, the artificial intelligence model is configured as a bayesian network model, and comprises the following construction steps:
s1, determining variable parameters in a Bayesian network according to the weather environment data and the numerical range of the data variables of the power generation condition of the renewable energy power distribution cabinet;
s2, constructing a structure of the Bayesian network according to the dependency relationship among the variables;
s3, determining conditional probability distribution among variable parameters, and constructing a conditional probability table;
And S4, after the Bayesian network model is constructed, evaluating and optimizing the Bayesian network by using cross verification, accuracy or recall.
Further, in step S1, the variable parameters are also determined by using a cluster analysis method configured as a K-means method, a Ward clustering method, or an SVM method.
Further, in step S2, a structure of the bayesian network is constructed according to the dependency relationship between the variables, and specifically, the dependency relationship between the variables is determined through expert knowledge, a data analysis model or a machine learning algorithm.
Further, the power supply modes include a pure renewable energy power supply mode, a hybrid power supply mode, an energy storage auxiliary power supply mode and a standby power supply mode, wherein:
The intelligent power distribution cabinet only depends on the power generated by renewable energy sources to supply power for loads in the pure renewable energy source power supply mode, and is not connected with the traditional energy sources;
The hybrid power supply mode is characterized in that renewable energy and traditional energy supply power to a load at the same time, and the proportion of the renewable energy and the traditional energy is dynamically adjusted according to the real-time power generation condition and the load demand;
The energy storage auxiliary power supply mode is used for storing redundant electric energy into the energy storage equipment when the power generation amount of the renewable energy source is sufficient, and releasing the electric energy by the energy storage equipment to supply power to the load when the power generation condition of the renewable energy source is unstable or the load requirement cannot be met within a specific time period;
and in the standby power supply mode, when the renewable energy source and the traditional energy source cannot normally supply power, the standby power supply is started to supply power to the load.
Further, the optimization control module comprises a receiving module, a matching module and an executing module, wherein:
the receiving module is used for receiving the power generation condition of the current power distribution cabinet and generating a control signal;
The matching module is used for matching corresponding power supply modes according to the control signals;
And the execution module is used for supplying power to the load according to the matched power supply mode.
Further, the optimization control module also utilizes the visualization tool to display the power supply state of the power distribution cabinet, wherein the power supply state comprises weather environment data change, future power generation condition prediction, power supply mode suggestion and current load demand condition.
The invention has the beneficial effects that:
(1) The method comprises the steps of taking specific time as step length, monitoring and collecting real-time weather environment data of a time sequence in the future of local weather forecast, extracting main characteristics of the real-time weather environment data by adopting a principal component analysis method, constructing an artificial intelligent model by collecting the main characteristics of the weather environment data in a historical period and power generation condition data of a renewable energy power distribution cabinet, inputting the real-time weather environment data into the artificial intelligent model, predicting the current power generation condition of the power distribution cabinet, and controlling the power supply mode of the power distribution cabinet according to the current power generation condition of the power distribution cabinet. The invention solves the problem that the control strategy of the power distribution cabinet is difficult to optimize according to the change of weather environmental factors in the prior art, so that the energy utilization efficiency is reduced.
(2) By establishing an artificial intelligent model between weather environment data and power generation conditions of the power distribution cabinet, the future power generation conditions are predicted by combining weather forecast, so that a power supply mode of the power distribution cabinet is optimized in advance, and the control effect of the power distribution cabinet is improved.
Drawings
The present invention is further described below with reference to the accompanying drawings for the convenience of understanding by those skilled in the art.
Fig. 1 is a block diagram of an intelligent power distribution cabinet with renewable energy access in the invention.
FIG. 2 is a flow chart of Bayesian network model construction in one embodiment of the present invention.
FIG. 3 is a block diagram of an optimization control module in one embodiment of the invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention for achieving the intended purpose, the following detailed description will refer to the specific implementation, structure, characteristics and effects according to the present invention with reference to the accompanying drawings and preferred embodiments.
Referring to fig. 1-3, the application provides an intelligent power distribution cabinet for renewable energy access, which comprises a monitoring and collecting module, an intelligent prediction module and an optimization control module, wherein the monitoring and collecting module, the intelligent prediction module and the optimization control module are in communication connection, and the intelligent power distribution cabinet comprises:
the monitoring and collecting module is used for monitoring and collecting real-time weather environment data of a time sequence in the future of local weather forecast by taking specific time as step length;
in this embodiment, the monitoring and collecting module works in specific time steps, which means that it can systematically develop monitoring for the weather environmental data of a time series in the future of the local weather forecast at preset time intervals. The specific time step can be adjusted according to actual requirements, so that the dynamic change of the weather environment can be captured as accurately as possible without wasting excessive resources. The weather environment data encompasses a number of important aspects. Among other things, temperature data is critical, and different temperature ranges can have an impact on the performance of renewable energy devices. For example, at higher temperatures, the efficiency of certain solar panels may be affected to some extent, whereas for some devices that utilize thermoelectric generation, the change in air temperature is more directly related to the amount of generated electricity.
The humidity data cannot be ignored, the higher humidity can possibly cause the insulation performance of the equipment to be reduced, the risk of equipment faults is increased, meanwhile, the humidity can also influence the heat conduction performance of air, further the heat dissipation effect is influenced, and the operation efficiency of the renewable energy equipment is indirectly influenced.
The wind speed data has decisive significance for the wind power generation system, namely the output power of the wind power generator is directly determined by the wind speed and the change trend. Through accurate monitoring wind speed, the potential of wind energy power generation can be predicted better, and important basis is provided for the control strategy of the intelligent power distribution cabinet.
The illumination intensity data is a core parameter for a solar power generation system, wherein sufficient illumination intensity can enable a solar panel to generate more electric energy, and the fluctuation of the illumination intensity can directly influence the fluctuation of the solar energy generating capacity. Therefore, accurate monitoring of illumination intensity is critical to optimizing operation of solar power generation systems and improving energy utilization efficiency.
Rainfall data also has the effect that although rainfall itself is not greatly related to direct energy power generation, excessive rainfall may affect the power generation efficiency of the solar panel, and may damage some outdoor renewable energy devices. In addition, the rainfall can influence the water level of the river, thereby indirectly influencing renewable energy forms such as hydroelectric generation and the like.
It should be noted that the acquired weather environment data exhibits a long time series of characteristics. This means that various weather environmental factors such as air temperature, humidity, wind speed, illumination intensity, rainfall and the like are continuously monitored and collected in a long time range, so that a large number of data samples are accumulated. However, a large number of data samples is not entirely beneficial. Although the rich data provides more comprehensive information to some extent, there may be many duplicate or highly relevant data portions, so-called data redundancy. For example, over a continuous period of time, the air temperature may fluctuate over a small range with a relatively smooth trend, so that continuously recorded air temperature data may have a large number of similar values that do not provide much additional valuable information for subsequent analytical modeling, but rather may be a factor in interference.
When modeling is performed using such data that includes a large amount of redundancy, such as building an artificial intelligence model for predicting the power generation condition of a power distribution cabinet, the excessive redundancy data can make it difficult for the model to accurately grasp critical information during learning, and can be easily misled by repeated and non-substantial-difference data. The model can not accurately capture real and effective association between the weather environmental factors and the power generation conditions, so that the modeling accuracy is reduced, and the power generation conditions of the power distribution cabinet which are finally predicted have larger deviation from the actual conditions.
In order to solve the above problem of the decrease in modeling accuracy that may be caused by data redundancy, the present embodiment introduces a Principal Component Analysis (PCA) method. Principal component analysis is a commonly used multivariate statistical analysis technique, and the core purpose of the principal component analysis is to transform a data set originally having a plurality of variables (in this case, various weather environment data variables) into a new set of independent (or extremely low-correlation) variable combinations, that is, principal components, by performing linear transformation on original data.
In a specific operation process, the principal component analysis method finds out principal features capable of representing original data information to the greatest extent according to statistical properties such as variance of data. For weather environment data, the main characteristics which can reflect the whole condition of the weather environment most and are independent from each other can be extracted from a plurality of data samples such as air temperature, humidity, wind speed and the like which possibly have redundancy.
For example, it may be found through PCA analysis that a new variable (principal component) that integrates factors such as air temperature, illumination intensity, and wind speed can represent the influence of the weather environment on the power generation condition to a great extent, while the importance of other original data variables that have higher correlation with the new variable is relatively reduced. By extracting the main characteristics, the data redundancy can be effectively reduced while the key information of the weather environment data is maintained.
Therefore, when the data subjected to principal component analysis processing is used for subsequent modeling work, such as construction of an artificial intelligent model for predicting the power generation condition of the power distribution cabinet, the model can be focused on key features which really have important influence on the power generation condition, so that the modeling accuracy is improved, and the prediction result is closer to the actual power generation condition of the power distribution cabinet.
Thus, further, the weather environmental data includes air temperature, humidity, wind speed, illumination intensity, and rainfall.
Further, the weather forecast is obtained by the following method:
(1) Using a professional weather data platform or API:
The Chinese weather data network is an official weather data platform and provides rich weather data resources. According to the basic weather data open shared catalogue issued by the China weather bureau, various shared weather data and data service products are available, but special needs or data acquisition outside the shared catalogue may need to be applied to the provincial weather bureau.
Commercial weather data API some commercial companies provide weather data interface services, such as the Goldmap weather API. Through applying for the API key, the related interface can be called to acquire weather forecast information of a specific region. Taking the Goldmap API as an example, when in use, a request URL needs to be constructed according to a specified format, corresponding parameters such as city codes and the like are transmitted, then returned JSON format data are analyzed, and the required weather environment data including air temperature, humidity, wind speed, illumination intensity (partial areas may not be provided directly but can be inferred indirectly through cloud amount and other information), rainfall and the like are extracted from the JSON format data.
(2) Using a weather monitoring device:
Weather stations are provided in some areas with professional weather stations equipped with various sensors and monitoring instruments, and can collect local weather data in real time, including basic weather environment data such as temperature, humidity, wind speed, rainfall and the like. The relevant departments or institutions can acquire real-time monitoring data by establishing data connection with the weather stations, and collect and sort the monitoring data according to specific time steps.
Personal weather monitoring devices-for some individuals or small organizations with specific needs for weather data, small weather monitoring devices, such as home weather stations, may be used. These devices can typically measure basic parameters such as temperature, humidity, air pressure, etc., and some advanced devices can also measure information such as wind speed. By connecting the device to a computer or mobile device, local weather environment data can be acquired and recorded in real time, but the range of data acquired in this way is relatively small and may not be as accurate as a professional weather station.
(3) Acquisition from weather forecast websites or applications:
Official weather forecast websites the weather departments of each place usually have own official weather forecast websites to provide weather forecast information of the local area. The web sites can be accessed according to specific time intervals through web crawler technology, and the required weather environment data is captured. But this approach requires attention to comply with relevant legal regulations and website usage regulations.
Weather forecast applications there are many weather forecast applications on the market that acquire real-time weather forecast data by cooperating with weather data providers. Some applications provide an open interface or data export functionality that can be used to obtain the desired weather environment data.
The intelligent prediction module is used for constructing an artificial intelligent model by collecting weather environment data in a historical period and power generation condition data of the renewable energy power distribution cabinet, inputting the real-time weather environment data into the artificial intelligent model and predicting the current power generation condition of the power distribution cabinet;
In this embodiment, the intelligent prediction module first aims at collecting weather environment data during a historical period and power generation condition data of the renewable energy power distribution cabinet. The historical weather environment data comprises various information such as air temperature, humidity, wind speed, illumination intensity, rainfall and the like in different seasons and different time periods. The power generation condition data of the renewable energy power distribution cabinet cover the details of power generation capacity, power generation power fluctuation and the like of each time node. By sorting and analyzing a large amount of historical data, a precise artificial intelligent model is constructed.
The artificial intelligence model can learn the complex relation between the weather environment factors and the power generation condition of the power distribution cabinet. For example, when the light intensity is high, the power generation amount of the solar power generation section is generally increased, and when the wind speed is high, the contribution of wind power generation may be increased. Meanwhile, the model also considers the comprehensive influence of a plurality of factors, such as the influence on the equipment performance under the combined action of high humidity and high temperature and the indirect influence on the renewable energy power generation under the condition of different rainfall.
When real-time weather environment data is input, the intelligent prediction module can rapidly transmit the data to the constructed artificial intelligent model. The model can calculate and analyze rapidly according to the input data, and predicts the power generation condition of the current power distribution cabinet. Such predictions may include estimates of power generation over a period of time in the future, trends in power generation, etc. Through the mode, the intelligent power distribution cabinet can know the power generation capacity of the intelligent power distribution cabinet in advance, and provides important basis for formulating a reasonable control strategy, so that the utilization efficiency of energy sources is effectively improved, and the stable operation of a power system is ensured.
Further, the power generation condition data is characterized by the following variables:
1. generating capacity:
This is the variable that most intuitively reflects the power generation situation. It represents the total amount of electricity produced by an intelligent power distribution cabinet that is accessed by renewable energy, typically in kilowatt-hours (kWh), over a specified period of time. By monitoring and recording the power generation at different points in time, the overall scale and trend of the power generation can be known.
2. Generating power:
The generated power represents the generated energy per unit time, typically in kilowatts (kW). The intelligent power distribution cabinet reflects the power generation intensity at the current moment, and has important significance for monitoring and controlling the running state of the intelligent power distribution cabinet in real time. For example, when the generated power suddenly drops, it may mean that a certain renewable energy power generation apparatus malfunctions or that a weather condition changes adversely.
3. Power generation stability index:
The index is used for measuring the stability of the power generation condition. The method can be used for determining statistics such as standard deviation, variation coefficient and the like of the generated energy or the generated power in a period of time. If the power generation stability index is low, the power generation condition fluctuation is large, and corresponding measures may need to be taken to stabilize the power generation, such as adjusting the charge and discharge strategy of the energy storage device or optimizing the control parameters of the intelligent power distribution cabinet.
4. Generating efficiency:
the power generation efficiency represents the ratio of renewable energy to electrical energy. It can be calculated by the ratio of the actual power generation amount to the theoretical maximum power generation amount. For example, for solar power generation, the theoretical maximum power generation can be estimated from the solar radiation intensity and the conversion efficiency of the photovoltaic panel. The improvement of the power generation efficiency is one of important targets for optimizing the performance of the intelligent power distribution cabinet, and can be realized by selecting high-efficiency renewable energy power generation equipment, optimizing system configuration and control strategies and the like.
Further, the artificial intelligence model is configured as a bayesian network model, and comprises the following construction steps:
S1, determining variable parameters:
First, the variable parameters in the bayesian network are clarified. These variables include, but are not limited to, weather environmental factors such as air temperature, humidity, wind speed, light intensity, rainfall, and the like, and the power generation conditions of the renewable energy power distribution cabinet. For each variable, it is necessary to determine its possible range of values determining parameters. For example, the air temperature can be divided into different sections such as low temperature, medium temperature, high temperature and the like according to actual conditions, and the generated energy can be divided into different power ranges. The partitioning may also be performed using a cluster analysis method.
Further, in step S1, the variable parameters are also determined by using a cluster analysis method configured as a K-means method, a Ward clustering method, or an SVM method.
In the embodiment, in the step of constructing the bayesian network model of the intelligent power distribution cabinet for renewable energy access, step S1 determines that the variable parameters have a key effect by using a cluster analysis method. The K-means method, the Ward clustering method and the SVM method are respectively characterized.
The K-means method firstly randomly selects K data points as initial clustering centers, distributes the data points by calculating the distance from the data points to each center, recalculates the centers, and repeats until the data points are stable. When the variable parameters are determined, historical weather and power generation data are taken as input, the data are divided into different clusters, each cluster represents a combination of specific weather and power generation conditions, and weather factors such as average air temperature and the like and average power generation capacity and the like are extracted to serve as the variable parameters of the Bayesian network model. The method has the advantages of simple and easy realization of algorithm, high calculation speed and capability of processing a large-scale data set, but the number of clusters needs to be predetermined and is sensitive to an initial center.
Ward clustering method is based on analysis of variance, and aims to minimize the sum of internal variances of all clusters after clustering, and two clusters with the smallest increase of total variances after each selection and combination are combined. When the variable parameters are determined, the historical data are taken as input, the clustering characteristics are analyzed, and parameters related to weather and power generation are extracted. The method can generate compact and balanced clusters, is suitable for different data distribution, but has higher calculation complexity and is sensitive to abnormal values.
The SVM (support vector machine) method can be used for clustering by improving the detection outliers of the One-Class SVM, for example, to perform single-Class clustering. When the variable parameters are determined, firstly, abnormal value detection is carried out on weather and power generation data, normal data are used as clusters, the characteristics of the clusters are analyzed, and parameters such as a power generation capacity interval and the like under normal weather are extracted. The method has good processing capability on high-dimensional complex data distribution, can effectively detect abnormal values, but has large influence on parameter selection and high calculation complexity. The clustering analysis method provides important variable parameter basis for constructing the Bayesian network model, is beneficial to more accurately predicting the power generation condition of the renewable energy power distribution cabinet, and improves the energy utilization efficiency.
S2, constructing a network structure:
and constructing the structure of the Bayesian network according to the relation among the variables. The dependency between variables may be determined by expert knowledge, data analysis, or machine learning algorithms. For example, the intensity of illumination may directly affect the amount of solar power generation and wind speed may be related to the amount of wind power generation. In building a network structure, a graphical tool may be used to visually represent the relationships between variables.
S3, determining conditional probability distribution:
after determining the network structure, a conditional probability distribution for each variable parameter needs to be determined. This may be achieved by a statistical analysis of the training data. For example, for a given air temperature and illumination intensity, a probability distribution of solar power generation is calculated. The parameters may be determined using maximum likelihood estimation, bayesian estimation, and the like.
S4, model evaluation and optimization:
After the bayesian network model is constructed, the bayesian network model needs to be evaluated and optimized. The performance of the model may be evaluated using cross-validation, accuracy, precision, recall, and F1 values, among other metrics. If the performance of the model is not ideal, the model can be improved by adjusting the network structure, adding variables, optimizing parameters and the like.
The accuracy rate is the proportion of the number of samples which are correctly predicted by the model to the total number of samples, is one of the most commonly used evaluation indexes, the accuracy rate measures how many of the positive examples are actually positive examples, which are important indexes for evaluating the prediction accuracy of the model, the recall rate measures the proportion of the number of positive examples which can be correctly predicted by the model to the number of actual positive examples, which is also called recall rate, the F1 value is the harmonic average of the accuracy rate and the recall rate, the accuracy and the recall of the model are comprehensively considered, and the higher the F1 value is, the better balance between the accuracy rate and the recall rate is achieved by the model.
Further, in step S5, the accuracy rate is calculated according to the following formula:
,
In the formula, ACC represents accuracy, TP represents a real example, TN represents a real negative example, FP represents a false positive example, and FN represents a false negative example.
Further, in step S5, the accuracy rate is calculated according to the following formula:
,
In the formula, PPV represents the precision, TP represents the real example, and FP represents the false positive example.
Further, in step S5, the recall ratio is calculated according to the following formula:
,
where TRP represents the recall rate, TP represents the true case, and FN represents the false negative case.
Further, in step S5, the calculation formula of the F1 value is:
,
Wherein F1 represents the F1 value, PPV represents the precision rate, and TRP represents the recall rate.
And the optimizing control module is used for controlling the power supply mode of the power distribution cabinet according to the current power generation condition of the power distribution cabinet.
In this embodiment, the optimization control module closely implements accurate control on the power supply mode of the power distribution cabinet according to the power generation condition of the current power distribution cabinet. And after the intelligent prediction module predicts the power generation condition of the current power distribution cabinet, the optimization control module can immediately receive the information and conduct deep analysis. The power supply modes include the following:
1. pure renewable energy power mode:
When the generated energy of the renewable energy sources is sufficient and stable, and the load requirements connected with the power distribution cabinet can be completely met, a pure renewable energy source power supply mode is adopted. In the mode, the intelligent power distribution cabinet only depends on electric power generated by renewable energy sources such as solar energy, wind energy and the like to supply power to the load, and the intelligent power distribution cabinet is not connected with the traditional energy sources. The mode has the advantages of zero emission and environmental protection, can utilize renewable energy sources to the greatest extent, and reduces the dependence on traditional energy sources.
2. Hybrid power mode:
when the power generation of the renewable energy source is insufficient to meet the whole load demand, the hybrid power supply mode is started. In this mode, the renewable energy source and the conventional energy source (e.g., mains) simultaneously supply the load. The optimization control module can dynamically adjust the proportion of renewable energy sources and traditional energy sources according to the real-time power generation condition and load requirements. For example, when the renewable energy generation amount is high, the use proportion of the conventional energy is reduced, and when the renewable energy generation amount is insufficient, the supply of the conventional energy is increased to ensure the stable operation of the load.
3. Energy storage auxiliary power supply mode:
the energy storage auxiliary power supply mode functions when the power generation amount of the renewable energy source is unstable or cannot meet the load demand within a specific period of time. In this mode, the intelligent power distribution cabinet is connected with energy storage equipment, such as a battery pack, a super capacitor and the like. And when the renewable energy generating capacity is insufficient, the energy storage equipment releases the electric energy to supply power for the load. The mode can effectively smooth the fluctuation of renewable energy power generation, and improve the stability and reliability of the system.
4. Standby power supply mode:
As an emergency power supply mode, a standby power supply mode is started when both renewable energy and conventional energy cannot normally supply electric power. The backup power source may be a diesel generator, UPS (uninterruptible power supply), or the like. This mode ensures that in extreme cases, critical loads can still be supplied with power, improving the safety and reliability of the system.
Further, the optimization control module comprises a receiving module, a matching module and an executing module, wherein:
the receiving module is used for receiving the power generation condition of the current power distribution cabinet and generating a control signal;
The matching module is used for matching corresponding power supply modes according to the control signals;
And the execution module is used for supplying power to the load according to the matched power supply mode.
In this embodiment, the receiving module is used as an initial link of the whole optimization control flow and is responsible for receiving the power generation condition of the current power distribution cabinet. Through the tight connection with the intelligent prediction module, key data about the generated energy, the generated power, the generated stability and the like are obtained in real time. The receiving module can perform rapid analysis processing on the data, and generates corresponding control signals according to preset judgment standards. This control signal, like the same instruction, provides an explicit direction for the operation of the following module.
The matching module starts working after receiving the control signal generated by the receiving module. Its main task is to match the corresponding power supply mode according to the control signal. The matching module stores the corresponding relation between different power generation conditions and power supply modes. For example, when the control signal indicates that the current renewable energy power generation is sufficient, the matching module can quickly identify and select a pure renewable energy power supply mode, and when the power generation is insufficient to meet the load demand, the matching module can match a hybrid power supply mode or an energy storage auxiliary power supply mode and the like. Through accurate matching, the selected power supply mode is ensured to be capable of adapting to the current power generation condition to the greatest extent, and efficient utilization of energy and stable operation of the system are realized.
The execution module is the last link of the optimization control module and is also a key part for putting the power supply mode into practice. Once a matching power mode is determined, the execution module will perform power operations in accordance with that mode. The power supply system comprises a power distribution cabinet, an execution module, a hybrid power supply mode, an energy storage auxiliary power supply mode and an energy storage auxiliary power supply mode, wherein the execution module can adjust circuit connection of the power distribution cabinet to ensure that a load is only supplied by renewable energy, the execution module can coordinate input proportion of the renewable energy and traditional energy to meet load requirements in the hybrid power supply mode, and the execution module can reasonably control inflow and outflow of electric energy according to the state of energy storage equipment in the energy storage auxiliary power supply mode. The efficient operation of the execution module ensures an accurate implementation of the power mode, providing a stable and reliable power supply to the load.
Further, the optimization control module also utilizes the visualization tool to display the power supply state of the power distribution cabinet, wherein the power supply state comprises weather environment data change, future power generation condition prediction, power supply mode suggestion and current load demand condition.
In the embodiment, the comprehensive visual display is performed, so that a user can intuitively know the power supply state of the power distribution cabinet, energy management and decision making are better performed, and the energy utilization efficiency and the stability of the system are improved. For presentation of weather environment data changes, the visualization tools are presented in the form of dynamic charts or graphs. For example, the air temperature may be shown as a line graph over time, with different colored curves representing different temperature ranges. Humidity can be represented by a bar graph, which intuitively reflects the change in the humidity level of the air. The wind speed can clearly show the distribution of wind direction and the fluctuation of the wind speed by combining the wind direction rose diagram with the wind speed change curve. The illumination intensity can be represented by a graph similar to the intensity of solar radiation, and the shade of the color represents the intensity change of illumination. The rainfall may show the total amount of rainfall change over a period of time using an accumulated histogram.
In future power generation prediction, the visualization tool is presented in a combination of prediction curves and data tables. The prediction curve shows the trend of the power generation amount in a future period of time, and the curves with different colors can represent different types of renewable energy power generation predictions. The data table lists specific predicted values, including the predicted power generation amount, the generated power and other information in different time periods, so that the user can clearly know the future power generation condition.
For power mode suggestions, the visualization tools are presented in a highlighting and text description manner. According to the current weather environment data, future power generation condition prediction and load demand condition, the system can give corresponding power supply mode suggestions. For example, if renewable energy generation is sufficient and stable, the visualization tool will highlight "pure renewable energy power mode" and attach a brief description explaining why such a mode is recommended. The reasons for this are also clearly shown and described if hybrid or energy storage auxiliary power is required.
The current load demand situation is shown by real-time data and graphs. The current power demand of different types of loads can be represented by a bar graph, with colors distinguishing different load categories. Meanwhile, the total load demand and the comparison situation with the current power supply capacity can be displayed digitally, so that a user can clearly know whether the load demand is met or not and whether the power supply strategy needs to be adjusted or not.
The present invention is not limited in any way by the above-described preferred embodiments, but is not limited to the above-described preferred embodiments, and any person skilled in the art will appreciate that the present invention can be embodied in the form of a program for carrying out the method of the present invention, while the above disclosure is directed to equivalent embodiments capable of being modified or altered in some ways, it is apparent that any modifications, equivalent variations and alterations made to the above embodiments according to the technical principles of the present invention fall within the scope of the present invention.
Claims (10)
1. The intelligent power distribution cabinet with renewable energy access is characterized by comprising a monitoring and collecting module, an intelligent prediction module and an optimization control module, wherein the monitoring and collecting module, the intelligent prediction module and the optimization control module are in communication connection, and the intelligent power distribution cabinet is characterized in that:
the monitoring and collecting module is used for monitoring and collecting real-time weather environment data of a time sequence in the future of local weather forecast by taking specific time as step length;
The intelligent prediction module is used for constructing an artificial intelligent model by collecting weather environment data in a historical period and power generation condition data of the renewable energy power distribution cabinet, inputting the real-time weather environment data into the artificial intelligent model and predicting the current power generation condition of the power distribution cabinet;
And the optimizing control module is used for controlling the power supply mode of the power distribution cabinet according to the current power generation condition of the power distribution cabinet.
2. The intelligent power distribution cabinet with renewable energy access according to claim 1, wherein the power generation condition data comprises power generation amount, power generation power, power generation stability index and power generation efficiency, and wherein:
The generated energy represents the total electric quantity generated by the intelligent power distribution cabinet accessed by renewable energy sources in a specific time period;
The generated power represents generated energy in unit time;
the power generation stability index is used for measuring the stability of the power generation condition and is determined by standard deviation or variation coefficient of the generated energy or the generated power in a period of time;
The power generation efficiency represents the ratio of the renewable energy source to the electric energy, and is calculated by the ratio of the actual power generation amount to the theoretical maximum power generation amount.
3. The intelligent power distribution cabinet for renewable energy access according to claim 1, wherein the Bayesian network model comprises the following construction steps:
s1, determining variable parameters in a Bayesian network according to the weather environment data and the numerical range of the data variables of the power generation condition of the renewable energy power distribution cabinet;
s2, constructing a structure of the Bayesian network according to the dependency relationship among the variables;
s3, determining conditional probability distribution among variable parameters, and constructing a conditional probability table;
And S4, after the Bayesian network model is constructed, evaluating and optimizing the Bayesian network by using cross verification, accuracy or recall.
4. The intelligent power distribution cabinet for renewable energy access according to claim 3, wherein in step S1, the variable parameters are further determined by using a cluster analysis method, and the cluster analysis method is configured as a K-means method, a Ward clustering method or an SVM method.
5. The intelligent power distribution cabinet for renewable energy access according to claim 3, wherein in step S2, a structure of a Bayesian network is constructed according to the dependency relationship between variables, and the dependency relationship between variables is determined specifically through expert knowledge, a data analysis model or a machine learning algorithm.
6. The intelligent power distribution cabinet with renewable energy access according to claim 1, wherein the power supply modes comprise a pure renewable energy power supply mode, a hybrid power supply mode, an energy storage auxiliary power supply mode and a standby power supply mode, wherein:
The intelligent power distribution cabinet only depends on the power generated by renewable energy sources to supply power for loads in the pure renewable energy source power supply mode, and is not connected with the traditional energy sources;
The hybrid power supply mode is characterized in that renewable energy and traditional energy supply power to a load at the same time, and the proportion of the renewable energy and the traditional energy is dynamically adjusted according to the real-time power generation condition and the load demand;
The energy storage auxiliary power supply mode is used for storing redundant electric energy into the energy storage equipment when the power generation amount of the renewable energy source is sufficient, and releasing the electric energy by the energy storage equipment to supply power to the load when the power generation condition of the renewable energy source is unstable or the load requirement cannot be met within a specific time period;
and in the standby power supply mode, when the renewable energy source and the traditional energy source cannot normally supply power, the standby power supply is started to supply power to the load.
7. The intelligent power distribution cabinet with renewable energy access according to claim 1, wherein the optimizing control module comprises a receiving module, a matching module and an executing module, wherein:
the receiving module is used for receiving the power generation condition of the current power distribution cabinet and generating a control signal;
The matching module is used for matching corresponding power supply modes according to the control signals;
And the execution module is used for supplying power to the load according to the matched power supply mode.
8. The intelligent power distribution cabinet with renewable energy access according to claim 1, wherein the optimization control module further displays a power supply state of the power distribution cabinet by using the visualization tool, wherein the power supply state comprises weather environment data change, future power generation condition prediction, power supply mode suggestion and current load demand condition.
9. The intelligent power distribution cabinet with renewable energy access according to claim 1, wherein the weather environment data comprises air temperature, humidity, wind speed, illumination intensity and rainfall.
10. The intelligent power distribution cabinet with renewable energy access according to claim 1, wherein the weather forecast is obtained through a China weather data network, a business weather data API, a weather station, an official weather forecast website or a weather forecast application program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411750939.XA CN119341127A (en) | 2024-12-02 | 2024-12-02 | An intelligent power distribution cabinet for renewable energy access |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411750939.XA CN119341127A (en) | 2024-12-02 | 2024-12-02 | An intelligent power distribution cabinet for renewable energy access |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN119341127A true CN119341127A (en) | 2025-01-21 |
Family
ID=94266721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411750939.XA Pending CN119341127A (en) | 2024-12-02 | 2024-12-02 | An intelligent power distribution cabinet for renewable energy access |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119341127A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110307109A1 (en) * | 2010-05-27 | 2011-12-15 | Sri-Jayantha Sri M | Smarter-Grid: Method to Forecast Electric Energy Production and Utilization Subject to Uncertain Environmental Variables |
| CN109242143A (en) * | 2018-07-31 | 2019-01-18 | 中国电力科学研究院有限公司 | A kind of neural network wind power forecasting method and system |
| CN112952815A (en) * | 2021-03-05 | 2021-06-11 | 广东电网有限责任公司佛山供电局 | Microgrid optimization configuration method, device, equipment and storage medium |
| CN118605244A (en) * | 2024-04-24 | 2024-09-06 | 华能桐乡燃机热电有限责任公司 | A method and device for intelligently controlling photovoltaic power generation system |
| CN118737488A (en) * | 2024-06-18 | 2024-10-01 | 成都市疾病预防控制中心(成都市预防医学研究院) | An integrated management system for HIV prevention, intervention and testing |
-
2024
- 2024-12-02 CN CN202411750939.XA patent/CN119341127A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110307109A1 (en) * | 2010-05-27 | 2011-12-15 | Sri-Jayantha Sri M | Smarter-Grid: Method to Forecast Electric Energy Production and Utilization Subject to Uncertain Environmental Variables |
| CN109242143A (en) * | 2018-07-31 | 2019-01-18 | 中国电力科学研究院有限公司 | A kind of neural network wind power forecasting method and system |
| CN112952815A (en) * | 2021-03-05 | 2021-06-11 | 广东电网有限责任公司佛山供电局 | Microgrid optimization configuration method, device, equipment and storage medium |
| CN118605244A (en) * | 2024-04-24 | 2024-09-06 | 华能桐乡燃机热电有限责任公司 | A method and device for intelligently controlling photovoltaic power generation system |
| CN118737488A (en) * | 2024-06-18 | 2024-10-01 | 成都市疾病预防控制中心(成都市预防医学研究院) | An integrated management system for HIV prevention, intervention and testing |
Non-Patent Citations (1)
| Title |
|---|
| 刘宇熙 著: "《Python机器学习案例教程 原书第2版》", 31 May 2021, pages: 82 - 86 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guarino et al. | Load match optimisation of a residential building case study: A cross-entropy based electricity storage sizing algorithm | |
| CN118214078B (en) | A photovoltaic inverter intelligent control method and system | |
| Panda | Intelligent data analysis for sustainable smart grids using hybrid classification by genetic algorithm based discretization | |
| WO2020250648A1 (en) | Power management system, power management method, and power management program | |
| CN118825965A (en) | A load dynamic response modeling method and system integrating multiple working conditions, processes and techniques | |
| Elkholy et al. | Impact of time resolution averaging analysis on integrated photovoltaic with office buildings and grid interaction metrics: Case study | |
| CN116468302A (en) | Building park energy structure evaluation method, system, electronic equipment and storage medium | |
| Yang et al. | Household power consumption pattern modeling through a single power sensor | |
| CN112904749A (en) | Comprehensive energy information acquisition and monitoring system and method based on ubiquitous Internet of things | |
| CN118313957B (en) | Comprehensive energy system safety improvement method and device based on scenario attribution | |
| CN119341127A (en) | An intelligent power distribution cabinet for renewable energy access | |
| Sun et al. | Power load disaggregation of households with solar panels based on an improved long short-term memory network | |
| Harit et al. | Quantification of power system flexibility: A review | |
| CN118966847A (en) | Intelligent regulation method and device based on solar power generation prediction | |
| KR20220026387A (en) | Energy management system using environment and status information of renewable power generation | |
| CN117114335A (en) | A building energy consumption intelligent management system and method | |
| Yang et al. | Comprehensive evaluation of distribution network reliability for power consumer based on AHP and entropy combination method | |
| Zaidi et al. | Automated demand side management in microgrids using load recognition | |
| Liasi et al. | Smart grid vs. intelligent grid; artificial intelligence takes the power system to the next level | |
| CN115619159A (en) | Intelligent park energy consumption analysis scheduling method and system | |
| Feng et al. | Design of Comprehensive Energy System Considering Anomaly Detection of Multi-Energy Data | |
| Zhang et al. | Configuration Optimization of Hybrid Energy Storage System Based on Dynamic Weighted Power Prediction Error Distribution | |
| Chen et al. | Season-independent pv disaggregation using multi-scale net load temporal feature extraction and weather factor fusion | |
| CN113077175A (en) | Comprehensive energy system operation state evaluation method | |
| Chen et al. | Correlation analysis between electrification and energy efficiency of park-level integrated energy system considering operational optimization |
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 |