Intelligent electric power dispatching system based on artificial intelligence

Abstract

The invention provides an intelligent power dispatching system based on artificial intelligence. The intelligent electric power dispatching system based on artificial intelligence comprises a data acquisition module used for collecting power grid state, load demand, power generation condition and external environment information in real time through various sensors and intelligent equipment, the sensors comprise a current sensor, a voltage sensor and a meteorological sensor, and the intelligent equipment comprises a power supply module and a power supply module; the intelligent device comprises an intelligent electric meter and a load prediction device, and the data processing module is used for cleaning, analyzing and fusing the collected data. According to the intelligent electric power scheduling system based on artificial intelligence, through real-time data acquisition of various sensors and intelligent equipment and data processing of adaptive filtering and a machine learning algorithm, the system can significantly improve the accuracy of scheduling decisions, reduce the risk of decision errors caused by data acquisition errors or delay, and improve the scheduling accuracy. The innovation not only improves the reliability of the system, but also enhances the response capability to the dynamic change of the power market.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power intelligent dispatching, and in particular to an electric power intelligent dispatching system based on artificial intelligence. Background Art

[0002] The power intelligent dispatching system based on artificial intelligence usually consists of four parts: data acquisition module, data processing module, decision support module and execution control module. The data acquisition module is responsible for obtaining real-time information on the power grid status, load demand and power generation situation, and transmitting the data to the system through sensors and smart devices. The data processing module uses machine learning and data analysis technology to clean, analyze and predict the collected data, identify load patterns and power generation potential. Based on the processing results, the decision support module uses optimization algorithms to generate dispatch plans and propose the optimal power distribution strategy. Finally, the execution control module transmits the dispatch plan to each power generation and distribution equipment to ensure that the system is executed as planned.

[0003] Although the AI-based power intelligent dispatching system has many advantages, it also has some drawbacks. The system relies heavily on data quality. If there are errors or delays in data collection, it may lead to inaccurate dispatching decisions. The black box nature of the system makes the dispatching process lack transparency, making it difficult to explain the reasons for certain decisions, which may reduce the trust of users and operators. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides an artificial intelligence-based electric power intelligent dispatching system, which solves the problem that the system has a high dependence on data quality. If there are errors or delays in data collection, the dispatching decision will be inaccurate; the black box characteristics of the system make the dispatching process lack transparency, making it difficult to explain the reasons for certain decisions, thereby reducing the trust of users and operators.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence-based power intelligent dispatching system, comprising:

[0006] A data acquisition module is used to collect grid status, load demand, power generation and external environment information in real time through a variety of sensors and smart devices. The sensors include current sensors, voltage sensors, and meteorological sensors. The smart devices include smart meters and load forecasting devices.

[0007] A data processing module is used to clean, analyze and fuse the collected data, combine adaptive filtering and machine learning algorithms to improve the quality and accuracy of the data, and reduce the impact of data errors on scheduling decisions;

[0008] The decision support module is used to generate a scheduling plan based on the processed data. It adopts a multi-layer neural network model based on time series, and its output formula is:

[0009] Y t =σ(W 1 ·[X t-1 ,X t-2 ,…,X t-n ]+b 1 )·W 2 +b 2 ;

[0010] Among them, Y t is the predicted load at time t, X is the input historical load data, W and b are weights and biases;

[0011] The execution control module is used to quickly transmit the scheduling plan to each power generation and distribution equipment. It adopts a distributed control architecture and combines it with a reinforcement learning algorithm. Its decision formula is:

[0012]

[0013] Where Q(s,a) represents the expected return of taking action a in state s, R(s,a) is the immediate reward, and γ is the discount factor;

[0014] The user interaction interface is used to display the power grid status, dispatch process and key decision-making basis in real time, enhancing the user's understanding and trust in the dispatch process.

[0015] Preferably, the data acquisition module further includes multiple data source interfaces, which can access electricity market data, weather forecast data and user demand forecast data to enhance the comprehensiveness of scheduling decisions. The data processing module adopts an anomaly detection model based on deep learning, and uses an automatic encoder to identify outliers in the data in real time to improve the quality of the data.

[0016] Preferably, the output results of the decision support module are transparently processed through explanatory artificial intelligence technology to quantify the contribution of each input variable to the final decision result to ensure improved user trust. The execution control module supports adaptive scheduling strategies, can automatically adjust the operating status of power generation and distribution equipment according to real-time load changes, and update strategies based on reinforcement learning.

[0017] Preferably, the system has a real-time feedback mechanism, which dynamically adjusts model parameters by monitoring feedback data to improve the accuracy of scheduling decisions. The decision support module introduces a model based on a graph neural network, which can handle node relationships in complex power grids and achieve more efficient load forecasting and scheduling optimization through an adjacency matrix.

[0018] Preferably, the system includes multiple scheduling modes, including real-time scheduling, short-term scheduling and long-term planning to adapt to different power markets and operational needs. The data acquisition module can monitor the health status of power equipment in real time, generate fault warnings and automatically adjust the scheduling plan.

[0019] Preferably, the execution control module supports cross-regional scheduling and can be dynamically adjusted according to the real-time demand and supply conditions of each region.

[0020] Preferably, the system is able to adjust the scheduling strategy based on user feedback to achieve a personalized response to user needs.

[0021] Preferably, the scheduling scheme generated by the optimization algorithm introduced into the decision support module has real-time update capability and can quickly respond to changes in power demand.

[0022] The present invention provides an artificial intelligence-based power intelligent dispatching system, which has the following beneficial effects:

[0023] Through real-time data collection from multiple sensors and smart devices, as well as data processing using adaptive filtering and machine learning algorithms, the system can significantly improve the accuracy of dispatch decisions and reduce the risk of decision errors caused by data collection errors or delays. This innovation not only improves the reliability of the system, but also enhances its ability to respond to dynamic changes in the power market.

[0024] The system's transparent design and explanatory artificial intelligence technology enhance users' understanding and trust in the dispatching process. By quantifying the contribution of each input variable to the decision-making result, users can clearly understand the basis behind the decision. In addition, the combination of real-time feedback mechanism and adaptive dispatching strategy enables the system to automatically adjust the operating status according to real-time load changes, improving the stability and flexibility of power supply, so as to better meet the needs of different users. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the structure of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] like Figure 1As shown, an embodiment of the present invention provides an artificial intelligence-based electric power intelligent dispatching system, including a data acquisition module for collecting power grid status, load demand, power generation status and external environment information in real time through a variety of sensors and intelligent devices. In the description of the data acquisition module, specific sensor parameters and acquisition frequency information are supplemented, as well as how to integrate data from different sensors through data fusion technology. Specifically, the acquisition frequency of the sensor can reach once per second, the current sensor has an accuracy of ±0.5%, and the error of the voltage sensor does not exceed ±1%. Through data fusion technology, the data collected by different sensors can be synchronized in time to ensure data consistency and timeliness. The sensors include current sensors, voltage sensors, and meteorological sensors. The intelligent devices include smart meters and load forecasting equipment. The data acquisition module further includes multiple data source interfaces, which can access power market data, weather forecast data, and user demand forecast data to enhance the comprehensiveness of dispatching decisions.

[0028] The data processing module is used to clean, analyze and fuse the collected data. During the data cleaning process, the mean interpolation method is used to process missing values, and the Z-score-based outlier detection algorithm is used to identify and eliminate abnormal data that exceeds the standard deviation range. The data fusion adopts the weighted average method, combined with the reliability and importance of each sensor for fusion, combined with adaptive filtering and machine learning algorithms to improve the quality and accuracy of the data and reduce the impact of data errors on scheduling decisions. The data processing module adopts an anomaly detection model based on deep learning, and uses an automatic encoder to identify outliers in the data in real time to improve the quality of the data.

[0029] The decision support module is used to generate a dispatching plan based on the processed data. The multi-layer neural network model contains three hidden layers, each with 64 nodes. The ReLU activation function is used to improve the nonlinear fitting ability of the model. The training data uses the load data of the past 24 hours, and combines the seasonal and trend decomposition methods to improve the accuracy of the model prediction. The multi-layer neural network model based on time series is used, and its output formula is:

[0030] Y t =σ(W 1 ·[X t-1 ,X t-2 ,…,X t-n ]+b 1 )·W 2 +b 2 ;

[0031] Among them, Y tis the predicted load at time t, X is the input historical load data, W and b are weights and biases, and the output results of the decision support module are transparently processed through explanatory artificial intelligence technology to quantify the contribution of each input variable to the final decision result and ensure the improvement of user trust.

[0032] The execution control module is used to quickly transmit the scheduling plan to each power generation and distribution equipment. The execution control module adopts the deep Q network algorithm to update the strategy in combination with the current environmental status and historical experience. The Q value of each action is updated through the Bellman equation to ensure the selection of the best scheduling plan in a complex power environment. It adopts a distributed control architecture and combines the reinforcement learning algorithm. The decision formula is:

[0033]

[0034] Among them, Q(s,a) represents the expected return of taking action a in state s, R(s,a) is the immediate reward, γ is the discount factor, and the execution control module supports adaptive scheduling strategies, which can automatically adjust the operating status of power generation and distribution equipment according to real-time load changes. Based on the reinforcement learning update strategy, the execution control module supports cross-regional scheduling and can make dynamic adjustments based on the real-time demand and supply conditions of each region.

[0035] The user interaction interface is used to display the power grid status, dispatching process and key decision-making basis in real time. The user interaction interface supports the user feedback collection function, allowing users to evaluate the dispatching decision and automatically adjust the algorithm parameters according to the feedback information. The system provides a visual data display function, using charts and dynamic dashboards to present the power grid status, load changes and dispatching process in real time, enhancing the user's understanding and trust in the dispatching process. The system has a real-time feedback mechanism, which dynamically adjusts the model parameters by monitoring the feedback data to improve the accuracy of the dispatching decision. The decision support module introduces a model based on graph neural network, which can handle the node relationship in complex power grids and achieve more efficient load forecasting and dispatching optimization through the adjacency matrix. The system includes multiple dispatching modes, including real-time dispatching, short-term dispatching and long-term planning to adapt to different power markets and operational needs. The data acquisition module can monitor the health status of power equipment in real time, generate fault warnings and automatically adjust the dispatching plan. The system can adjust the dispatching strategy according to user feedback to achieve personalized response to user needs. The dispatching plan generated by the optimization algorithm introduced in the decision support module has real-time update capabilities and can quickly respond to changes in power demand.

[0036] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence-based power intelligent dispatching system, characterized in that: include: A data acquisition module is used to collect grid status, load demand, power generation and external environment information in real time through a variety of sensors and smart devices. The sensors include current sensors, voltage sensors, and meteorological sensors. The smart devices include smart meters and load forecasting devices. A data processing module is used to clean, analyze and fuse the collected data, combine adaptive filtering and machine learning algorithms to improve the quality and accuracy of the data, and reduce the impact of data errors on scheduling decisions; The decision support module is used to generate a scheduling plan based on the processed data. It adopts a multi-layer neural network model based on time series, and its output formula is: Y t =σ(W1·[X t-1 ,X t-2 ,…,X t-n ]+b1)·W2+b2; Among them, Y t is the predicted load at time t, X is the input historical load data, W and b are weights and biases; The execution control module is used to quickly transmit the scheduling plan to each power generation and distribution equipment. It adopts a distributed control architecture and combines it with a reinforcement learning algorithm. Its decision formula is: Where Q(s,a) represents the expected return of taking action a in state s, R(s,a) is the immediate reward, and γ is the discount factor; The user interaction interface is used to display the power grid status, dispatch process and key decision-making basis in real time, enhancing the user's understanding and trust in the dispatch process.

2. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The data acquisition module further includes multiple data source interfaces, which can access power market data, weather forecast data and user demand forecast data to enhance the comprehensiveness of scheduling decisions. The data processing module adopts an anomaly detection model based on deep learning, and uses an automatic encoder to identify outliers in the data in real time to improve the quality of the data.

3. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The output results of the decision support module are processed transparently through explanatory artificial intelligence technology to quantify the contribution of each input variable to the final decision result, ensuring the improvement of user trust. The execution control module supports adaptive scheduling strategies, which can automatically adjust the operating status of power generation and distribution equipment according to real-time load changes and update strategies based on reinforcement learning.

4. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The system has a real-time feedback mechanism, which dynamically adjusts model parameters by monitoring feedback data to improve the accuracy of scheduling decisions. The decision support module introduces a model based on graph neural network, which can handle node relationships in complex power grids and achieve more efficient load forecasting and scheduling optimization through the adjacency matrix.

5. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The system includes multiple dispatching modes, including real-time dispatching, short-term dispatching and long-term planning to adapt to different power markets and operational needs. The data acquisition module can monitor the health status of power equipment in real time, generate fault warnings and automatically adjust the dispatching plan.

6. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The execution control module supports cross-regional scheduling and can be dynamically adjusted according to the real-time demand and supply conditions of each region.

7. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The system can adjust the scheduling strategy according to user feedback to achieve a personalized response to user needs.

8. The power intelligent dispatching system based on artificial intelligence according to claim 1 is characterized in that: The scheduling scheme generated by the optimization algorithm introduced in the decision support module has real-time updating capability and can quickly respond to changes in power demand.

Images