Community power dispatching optimization method, system and device and medium

Abstract

The invention relates to a community power dispatching optimization method, system and device and a medium, and the method comprises the steps: monitoring and collecting the power utilization data in a designated region in real time, and obtaining the external environment data of the designated region; storing the collected data according to a time sequence form; capturing a long dependency relationship in the stored time sequence data by adopting a long-short term memory (LSTM) network, and generating a prediction model by learning historical power consumption data and combining external environment data to predict a power consumption demand at a specified time in the future; acquiring real-time electricity price, actual electricity demand of a user and power grid load information, and calculating and generating an optimal scheduling strategy by using a multi-target optimization algorithm on the basis of the predicted electricity demand; and monitoring the operation state of user power equipment in real time, and adjusting load distribution and equipment start and stop based on the generated optimal scheduling strategy. According to the invention, the intelligence and flexibility of power dispatching can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power dispatching, in particular to a community power dispatching optimization method and system, an electronic device and a computer readable storage medium. BACKGROUND

[0002] With the popularity of smart home devices, power management of households and communities has become an important part of grid optimization. Traditional power dispatching systems mainly rely on fixed algorithms and rules, and cannot make real-time adjustments according to user individual needs and environmental changes. Due to the diversity of user behavior and the complexity of power demand, traditional dispatching systems are difficult to adapt to the individual needs of households and communities, resulting in power waste and low resource utilization efficiency.

[0003] Existing smart home systems usually use preset dispatching rules to control power consumption to some extent, but lack the ability to dynamically adjust based on real-time data and intelligent analysis. This makes the power management system unable to make accurate load prediction and dispatching in peak periods or when the grid load is too heavy, causing great pressure on the grid and even possibly causing power shortages. In addition, existing systems are difficult to adjust power consumption plans according to dynamic factors such as seasonal changes, weather changes, user habits, etc., and cannot achieve truly intelligent and personalized power management. SUMMARY

[0004] The purpose of the present application is to provide a community power dispatching optimization method and system, an electronic device and a computer readable storage medium, aiming to improve the intelligence and flexibility of power dispatching.

[0005] To achieve the above purpose, in a first aspect, the present application provides a community power dispatching optimization method, which comprises: Real-time monitoring and collecting power consumption data in a specified area, and obtaining external environmental data of the specified area; storing the collected data in time series form; Using a long short-term memory network (LSTM) to capture long dependencies in the stored time series data, and generating a prediction model by learning historical power consumption data and combining external environmental data to predict power consumption demand at a future specified time; Obtain real-time electricity price, user actual power demand and grid load information, and use a multi-objective optimization algorithm to calculate and generate an optimal dispatching strategy based on the predicted power demand; wherein the multi-objective optimization algorithm comprises: Calculate load balancing parameters , ; Calculate power cost minimization parameters : , Computing user comfort maximization parameters : , wherein, is total load at time, is electricity price, is actual electricity demand of user , is allocated power; computing comprehensive objective function: , wherein, is a load balancing weight coefficient for controlling peak reduction; is a cost minimization weight coefficient for controlling power cost reduction; is a user comfort weight coefficient for controlling user demand matching accuracy; Real-time monitoring of user power equipment operation state, and adjusting load distribution and equipment start-stop based on generated optimal scheduling strategy.

[0006] The real-time monitoring collects electricity data in the specified area and obtains external environment data of the specified area; the step of storing the collected data in time sequence form includes: Real-time collection of electricity data and external environment data from each user equipment through sensors and smart meters deployed in the smart grid, wherein the electricity data includes electricity consumption, voltage and / or current information, and the external environment data includes temperature, humidity and / or weather information; The collected data is stored in time sequence form and supports high frequency update; the collected data is represented by time sequence as: , wherein, represents the electricity consumption of the equipment at time , is the total number of equipment; The data sampling frequency is defined as: .

[0007] The step of capturing long dependency in stored time series data by using long short-term memory network LSTM and generating prediction model by learning historical electricity data and combining external environment data includes: The collected data is preprocessed, denoised, normalized and / or imputed, and power features and / or environmental features related to the prediction task are extracted from the data, and the extracted features are input into an embedding layer to be converted into vectors suitable for processing by a long short-term memory network (LSTM); In the LSTM, the historical data to be discarded is determined by the forget gate component; the latest data added to the state is determined by the input gate component; the current cell state is updated by the state update component by integrating the outputs of the forget gate and the input gate; the hidden state of each time step is controlled by the output gate component to capture long dependencies in time series data; The hidden state of the LSTM is passed to a fully connected layer and mapped to the predicted output; after post-processing, the predicted results are de-normalized and / or smoothed to obtain the final predicted power demand.

[0008] The state update formula of the LSTM is: , wherein, , , are the forget gate, input gate and output gate, is the cell state, is the hidden state, , are weight and bias matrices; The predicted power demand at a specified time in the future by the prediction model specifically includes: , wherein, is the model parameter.

[0009] The optimal scheduling strategy generated based on the optimal scheduling strategy adjusts the load distribution and device start-stop, including: The device state is defined as: , Based on the actual power demand of the user in the optimal scheduling strategy and the power allocated by the system, the start-stop of the device is dynamically adjusted, wherein the scheduling strategy includes: , When the actual power demand of the user is greater than the power allocated by the system, the device is controlled to start, When the actual power demand of the user is less than the power allocated by the system, the device is controlled to stop.

[0010] The optimal scheduling strategy generated based on the optimal scheduling strategy adjusts the load distribution and device start-stop, including: Controlling peak load wherein, is the average value of the system load to avoid grid overload.

[0011] Preferably, the step of monitoring the running state of the user's power equipment in real time and adjusting the load distribution and equipment start-stop based on the generated optimal scheduling strategy further comprises: monitoring the actual power consumption of the user's equipment in real time and calculating the deviation value between the actual power consumption and the predicted power consumption, if the deviation value is greater than the set threshold, feedback is performed and the prediction model is triggered to calculate the future power consumption demand of the user's equipment again.

[0012] In a second aspect, the present application provides a community power scheduling optimization system, the system comprising: a data acquisition module for real-time monitoring and collecting power consumption data in a specified area and obtaining external environmental data of the specified area; storing the collected data in time sequence form; a prediction module for capturing long dependencies in the stored time series data using a long short-term memory network (LSTM) and generating a prediction model by learning historical power consumption data and combining external environmental data to predict power consumption demand at a future specified time; an optimization module for obtaining real-time electricity prices, actual power consumption demand of users, and power grid load information, and generating an optimal scheduling strategy based on the predicted power consumption demand using a multi-objective optimization algorithm; wherein the multi-objective optimization algorithm comprises: calculating load balancing parameters , ; calculating power cost minimization parameters : , calculating user comfort maximization parameters : , wherein, is the total load at time t, is the electricity price, is the actual power consumption demand of the user, is the allocated power; calculating the comprehensive objective function: , , wherein, is the weight coefficient of load balancing for controlling peak reduction; is the weight coefficient of cost minimization for controlling the reduction of power cost; ​is a weight coefficient of user comfort, used to control the accuracy of user demand matching; The scheduling module is configured to monitor the running state of the user power equipment in real time, and adjust the load distribution and equipment start-stop based on the generated optimal scheduling strategy.

[0013] In a third aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, and the computer program is executed by the processor to implement the steps of the community power scheduling optimization method as described above.

[0014] In a fourth aspect, the present application provides a computer readable storage medium, which stores a community power scheduling optimization system, and the community power scheduling optimization system can be executed by at least one processor to make the at least one processor execute the steps of the community power scheduling optimization method as described above.

[0015] The beneficial effects of the above embodiments are as follows: By monitoring the power terminal and external environment data in real time, accurate power consumption data is provided, which provides a basis for subsequent analysis and decision-making. Using historical data and external environment characteristics, the LSTM model is used to predict future power demand, so as to make reasonable scheduling based on accurate demand estimation. According to the prediction result, a balance is made between load balancing, cost minimization and user comfort through a multi-objective optimization algorithm, and an optimal scheduling strategy is generated. Finally, the optimal scheduling strategy is applied to equipment control, and the load distribution is dynamically adjusted to ensure stable operation of the system and meet user demand. Thus, the power resources can be efficiently managed, and intelligent scheduling and optimization can be realized to improve the intelligence and flexibility of power scheduling. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The method flowchart of the community power scheduling optimization method provided by the present application is provided. Figure 2 The principle block diagram of the community power scheduling optimization system provided by the present application is provided. Figure 3 The flowchart of the LSTM network in the community power scheduling optimization method provided by the present application is provided. Figure 4 The principle block diagram of the electronic device provided by the present application is provided. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present application clearer, further detailed description will be given below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and not to limit the present application. Based on the examples in the present application, all other examples obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0018] It should be noted that the descriptions involving "first", "second" and the like in the present application are only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first" and "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions of various embodiments can be combined with each other, but it must be based on the fact that the technical solutions can be realized by those of ordinary skill in the art. When the combination of technical solutions contradicts each other or cannot be realized, it should be considered that the combination of technical solutions does not exist and is not within the scope of protection required by the present application.

[0019] The inventors of the present application have found that, with the rapid development of deep learning technology, especially in the application of time series prediction, user behavior modeling and resource optimization, the smart home system is expected to optimize power dispatch through deep learning algorithm, reduce energy consumption, and improve the utilization efficiency of power resources. That is, a deep learning driven power dispatch optimization system is proposed, which combines deep learning with smart home devices to improve the intelligence and flexibility of power dispatch.

[0020] Further, the present application provides a community power dispatch optimization method, system, electronic device and storage medium, which are described in detail below.

[0021] Referring to Figure 1 A method flowchart of an embodiment of the community power dispatch optimization method provided by the present application is provided. The method comprises the following steps: S1, real-time monitoring and collecting power consumption data (such as power, voltage, current, etc.) in a specified area, and acquiring external environmental data (such as temperature, humidity and weather conditions, etc.) of the specified area; storing the collected data in time sequence form; S2, using a long short-term memory network (LSTM) to capture long dependency relationships in the stored time series data, and generating a prediction model by learning historical power consumption data and combining external environmental data to predict power consumption demand at a future specified time; S3, acquiring real-time electricity price, actual power consumption demand of users and power grid load information, and calculating and generating an optimal dispatch strategy based on the predicted power consumption demand using a multi-objective optimization algorithm; wherein the multi-objective optimization algorithm comprises: Calculate load balancing parameters , ; computing power cost minimization parameter : , computing user comfort maximization parameter : , wherein, is the total load at the moment, is the electricity price, is the actual electricity demand of the user, is the allocated power; computing the comprehensive objective function: , , wherein, is the weight coefficient of load balancing, used to control peak reduction; is the weight coefficient of cost minimization, used to control the reduction of power cost; is the weight coefficient of user comfort, used to control the accuracy of user demand matching; S4, real-time monitoring the running state of the user power equipment, and adjusting the load distribution and equipment start-stop based on the generated optimal scheduling strategy.

[0022] The embodiment provides accurate electricity consumption data by real-time monitoring of power terminal and external environment data, providing a basis for subsequent analysis and decision-making. Using historical data and external environment characteristics, the future electricity demand is predicted using an LSTM model to make reasonable scheduling based on accurate demand estimation. According to the prediction result, a balance is made between load balancing, cost minimization and user comfort through a multi-objective optimization algorithm to generate an optimal scheduling strategy. Finally, the optimal scheduling strategy is applied to equipment control to dynamically adjust load distribution, ensuring stable operation of the system and meeting user demand. Thus, it can efficiently manage power resources, realize intelligent scheduling and optimization, and improve the intelligence and flexibility of power scheduling.

[0023] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0024] The embodiment also provides a community power dispatching optimization system, which corresponds one-to-one to the community power dispatching optimization method in the above embodiment. As Figure 2 ​As shown, the community power dispatch optimization system includes a data acquisition module, a prediction module, an optimization module, and a dispatch module. The data acquisition module is configured to monitor and collect real-time power consumption data in a specified area, and to obtain external environmental data of the specified area; and store the collected data in a time sequence form. The prediction module is configured to use a long short-term memory network (LSTM) to capture long dependencies in the stored time sequence data, and to generate a prediction model by learning historical power consumption data and combining external environmental data, so as to predict power consumption demand at a future specified time. The optimization module is configured to obtain real-time electricity prices, actual power consumption demand of users, and power grid load information, and to calculate and generate an optimal dispatch strategy by using a multi-objective optimization algorithm based on the predicted power consumption demand. The dispatch module is configured to monitor the running state of user power equipment in real time, and to adjust load distribution and equipment start-stop based on the generated optimal dispatch strategy.

[0025] Specifically, in an alternative embodiment, the present embodiment proposes an efficient power dispatch optimization system integrating data acquisition, load prediction, and optimization dispatch. The system architecture includes four modules: The data acquisition module provides real-time and accurate power consumption data to support the calculation of other modules.

[0026] The prediction module predicts future power consumption demand based on historical data and external factors, and transmits the prediction results to the optimization module.

[0027] The optimization module performs multi-objective optimization based on the predicted demand data, generates an optimal power dispatch strategy, and transmits it to the dispatch module.

[0028] The dispatch module controls the switching of equipment according to the dispatch plan of the optimization module, dynamically adjusts load distribution, ensures stable system operation, and adjusts feedback according to actual demand, finally realizing closed-loop control.

[0029] These four modules each undertake different but closely coordinated tasks in the system. The data acquisition module provides accurate power consumption data by monitoring power terminals and environmental data in real time, providing a basis for subsequent analysis and decision-making. The prediction module uses historical data and external characteristics to predict future power demand using an LSTM model, ensuring that the optimization module can make reasonable dispatch based on accurate demand estimates. The optimization module balances load balancing, cost minimization, and user comfort based on prediction results, using a multi-objective optimization method to generate an optimal dispatch strategy. Finally, the dispatch module applies the optimization strategy to equipment control, dynamically adjusts load distribution, ensures stable system operation, and meets user demand. Through the coordinated work of these four modules, the system can efficiently manage power resources, achieve intelligent dispatch and optimization.

[0030] The data acquisition module is the foundation of the entire system, responsible for real-time monitoring of electricity consumption in households and regions, ensuring high precision and reliability of the collected data. This module records information such as power consumption, voltage, current, and related environmental data such as temperature, humidity, and weather conditions through smart meters installed on various appliances and power terminals. These data not only reflect the current power consumption status but also provide historical trend analysis, providing solid data support for subsequent prediction and optimization.

[0031] The collected data can be stored in time series form and supports high-frequency updates to meet real-time requirements. To reduce communication burden, the module has a data filtering and compression mechanism that only transmits key information for upper module processing. The collected data is represented in time series as:

[0032] where, represents the power consumption of the th device at time , is the total number of devices.

[0033] The data sampling frequency is defined as:

[0034] The accuracy and real-time performance of the data acquisition module directly affect the overall performance of the system and are critical for subsequent prediction and optimization.

[0035] The prediction module is one of the cores of the system, responsible for accurate prediction of future power demand based on historical data and external features. This module uses Long Short-Term Memory Network (LSTM) as the prediction model, which can effectively capture long dependencies in time series and handle complex nonlinear patterns.

[0036] The prediction module not only considers the user's electricity consumption history but also integrates external information such as weather changes and time characteristics to achieve multi-dimensional accurate prediction. This module provides the predicted value of electricity demand at future time for the optimization module, providing data basis for load distribution and scheduling decisions.

[0037] In actual operation, the prediction module uses a sliding window mechanism, using a certain length of historical data as input to predict the electricity demand at future time steps. The prediction results are dynamically fed back to the optimization module to ensure the real-time performance and robustness of the system. The state update formula of the LSTM unit is: , where, , , forget gate, input gate, output gate, respectively, is the unit state, is the hidden state, , is the weight and bias matrix.

[0038] predicts the electricity demand at the future time :

[0039] where, is the model parameter.

[0040] The optimization module is the decision center of the power dispatch system, responsible for generating the optimal power dispatch strategy based on the electricity demand provided by the prediction module through mathematical optimization methods. The optimization module takes meeting user demand as a prerequisite, while considering various factors such as grid load, energy cost, and equipment operating state to ensure that the dispatch strategy can maximize overall efficiency. This module uses a multi-objective optimization framework to reduce grid peak load, reduce power cost, and improve user satisfaction as optimization objectives. By introducing weight coefficients, users can flexibly adjust the relative importance of each optimization objective to customize different scenario requirements. Multi-objective optimization includes: Load balancing: reduce peak load :

[0041] Cost minimization:

[0042] User comfort maximization:

[0043] where, is the total load at is the electricity price, is the actual electricity demand of user i, is the allocated power. represents the peak load, represents the power cost, represents the user comfort. Comprehensive objective function:

[0044]

[0045] where, is the weight coefficient of load balancing, controlling the importance of peak reduction; is the weight coefficient of cost minimization, controlling the reduction of power cost; ​is the weight coefficient of user comfort, controlling the accuracy of user demand matching. The smaller the objective function value means that the model performs better in load balancing, cost minimization, and user comfort. Therefore, in this optimization problem, the smaller the objective function value means the better the optimization effect.

[0046] The scheduling module is the execution unit of the system, responsible for dynamically applying the strategies generated by the optimization module to the operation of the devices. This module monitors the running state of the device in real time, adjusts the load distribution and device start-stop, and ensures that the system can continue to run efficiently under changing environmental conditions.

[0047] During execution, the scheduling module also needs to feedback the actual power demand of the user, evaluate the execution effect, and adjust the strategy in a timely manner, so as to realize closed-loop control. The module is also equipped with an emergency mechanism to deal with sudden load fluctuations or device failures, ensuring the safe operation of the power grid and user devices.

[0048] Device state is defined as:

[0049] During the scheduling process, the start and stop of the device need to be dynamically adjusted according to the actual power demand of the user and the power allocated by the system.

[0050] Adjustment strategy:

[0051] When the actual demand of the user is greater than the power allocated by the system, the device needs to start, when the actual demand of the user is less than the power allocated by the system, the device needs to be turned off.

[0052] The scheduling module also needs to monitor the running state of the device in real time and dynamically adjust the load according to the optimization results. For this purpose, the scheduling module considers the following factors: Power demand satisfaction:

[0053] By maximizing satisfaction, the scheduling module ensures that the user's demand is met as much as possible, and controls the demand difference. Goal: Keep the satisfaction , which means that the scheduling module needs to make the system allocated power and the actual demand of the user as close as possible to improve the user's satisfaction.

[0054] In order to avoid power grid overload or load fluctuation, the scheduling module needs to ensure that the load does not exceed the set peak load, and avoid excessive burden on the power grid. Peak load needs to meet the following conditions: ​Peak load:

[0055] where, is the average value of system load, is the total load at the highest moment. If the load approaches or exceeds this limit at a certain period, the dispatching module will take measures to avoid overload operation, possibly by delaying the start of some loads or reducing the power demand of some devices.

[0056] The flow of the entire power dispatching optimization system is formed by the cooperation of four modules: data acquisition, demand prediction, optimization decision, and dynamic dispatching, forming a closed-loop system to ensure the accuracy and real-time performance of power dispatching. First, the data acquisition module collects real-time power consumption information from each user device through sensors and smart meters deployed in the smart grid , and combines external features such as weather data and time information to deliver multi-dimensional data in a unified format to the demand prediction module. The completeness and real-time performance of these data provide a reliable foundation for subsequent modules.

[0057] The demand prediction module uses the Long Short-Term Memory network (LSTM) to model the input time series data. This module learns historical power consumption patterns and combines external features to generate power demand prediction values for the next time steps . The prediction result not only considers the periodicity and short-term fluctuations of user behavior, but also adjusts the weight of the prediction value through external factors, improving the prediction accuracy.

[0058] The optimization decision module takes the demand prediction results as input, combines real-time electricity prices and power resource constraints, and solves the optimization objective function: to determine the optimal dispatching strategy. During the optimization process, this module considers the differences between peak and valley periods, user power demand priorities, and the stability of power supply to ensure that the dispatching scheme is both economical and efficient and meets the power supply safety requirements.

[0059] The dispatching module is responsible for converting the dispatching plan output by the optimization decision module into specific dispatching instructions. This module monitors the actual power consumption of user devices in real time and feeds back the deviation value . If there is a large deviation, the dynamic dispatching module will adjust the allocation scheme and trigger the optimization module to recalculate, in order to achieve a quick response to unexpected events.

[0060] Throughout the whole process, each module cooperates closely through data flow and feedback mechanism: data collection provides accurate input for prediction, prediction results drive optimization decisions, optimization schemes guide dynamic scheduling, and scheduling feedback acts on data collection and prediction modules. Through this closed-loop operation mechanism, the system realizes a complete closed loop from data to decision-making, and then to actual scheduling, ensuring the intelligence and real-time of power resource allocation.

[0061] In some embodiments, as shown in FIG. 8, a flowchart of the LSTM network is provided. Figure 3

[0062] First, the power data is input into the model, and after data preprocessing, denoising, normalization, and missing value filling, etc., the data quality is ensured. Next, through feature extraction, useful features for the prediction task are extracted from the original data, such as power load, weather factors, and time information, etc. These features are passed into the embedding layer and converted into vector representations suitable for LSTM processing.

[0063] In the LSTM unit, data passes through multiple key components: the forget gate decides which historical information needs to be discarded, the input gate decides which new information can be added to the state, the state update integrates the outputs of the forget and input gates to update the current cell state, and finally, the output gate controls the hidden state at each time step, thereby capturing long-term dependencies in time series.

[0064] Next, the hidden state of the LSTM is passed to the fully connected layer, which is further mapped to the prediction output, i.e., power demand prediction. Finally, after post-processing, the prediction results are de-normalized, smoothed, etc., and the accurate predicted power demand is obtained. This process helps the LSTM network effectively learn the patterns of time series data in power prediction and provide accurate prediction results The specific limitations of the community power dispatch optimization method can be referred to the limitations of each module of the community power dispatch optimization system in the above, which will not be repeated here. Each module in the above community power dispatch optimization system can be realized by software, hardware, and their combinations in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0065] Referring to Figure 4 is a schematic diagram of the running environment of the preferred embodiment of the community power dispatch optimization system 10 of the present application.

[0066] ​In the embodiment, the community power dispatching optimization system 10 is installed and operated in the electronic device 1. The electronic device 1 is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions. The electronic device 1 can be a computer, a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of hosts or network servers based on cloud computing, wherein cloud computing is a kind of distributed computing, and the cloud is a super virtual computer composed of a group of loosely coupled computer clusters. (The electronic device 1 can be a server, a smart phone, a tablet computer, a portable computer, a desktop computer, and the like, which are terminal devices having storage and operation functions. In an embodiment, when the electronic device 1 is a server, the server can be one or more of a rack-mounted server, a blade server, a tower server, or a cabinet server.) In the embodiment, the electronic device 1 can include, but is not limited to, a memory 11, a processor 12, and a network interface 13 which are communicatively connected to each other through a system bus, and the memory 11 stores the community power dispatching optimization system 10 which can be run on the processor 12. It should be noted that, Figure 1 Only the electronic device 1 with components 11-13 is shown, but it should be understood that all the shown components are not required to be implemented, and more or fewer components can be alternatively implemented.

[0067] The memory 11 includes a memory and at least one type of readable storage medium. The memory provides a cache for the operation of the electronic device 1; the readable storage medium can be a non-volatile storage medium such as a flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the readable storage medium can be an internal storage unit of the electronic device 1, for example, a hard disk of the electronic device 1; in other embodiments, the non-volatile storage medium can also be an external storage device of the electronic device 1, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, and the like, which are equipped on the electronic device 1. In the embodiment, the readable storage medium of the memory 11 is usually used to store an operating system and various application software installed in the electronic device 1, for example, the community power dispatching optimization system 10 in the embodiment of the present application, and the like. In addition, the memory 11 can also be used to temporarily store various data which have been output or will be output.

[0068] The processor 12 may, in some embodiments, be a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 12 is generally used to control the overall operation of the electronic device 1, such as performing control and processing related to data interaction or communication with other devices, etc. In the present embodiment, the processor 12 is used to run program codes or process data stored in the memory 11, such as the community power dispatch optimization system 10, etc.

[0069] The network interface 13 may include a wireless network interface or a wired network interface, and is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0070] The community power dispatch optimization system 10 includes at least one computer readable instruction stored in the memory 11, which can be executed by the processor 12 to implement the embodiments of the present application.

[0071] In addition, the present application also provides a computer readable storage medium storing a community power dispatch optimization system, which can be executed by at least one processor to enable the at least one processor to perform the steps of the above embodiments.

[0072] The computer readable storage medium of the present application has substantially the same implementation as the above electronic device 1 and method embodiments, and will not be described here again.

[0073] It should be noted that, in this document, the term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article or device. Without more limitations, the element defined by the statement "comprises a" does not exclude the presence of another identical element in the process, method, article or device including the element.

[0074] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and necessary general hardware platform, of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of software product, and the computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a plurality of instructions to make a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) execute the method described in various embodiments of the present application.

[0075] The preferred embodiments of the present application are described above with reference to the accompanying drawings, and the scope of the present application is not limited thereto. The above-mentioned embodiment serial numbers are only for description, not representing the advantages and disadvantages of the embodiments. In addition, although the logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in an order different from that shown here.

[0076] Those skilled in the art can have various modification schemes to realize the present application without departing from the scope and essence of the present application, such as the features of one embodiment can be used in another embodiment to obtain another embodiment. Any modification, equivalent replacement and improvement within the technical concept of the present application shall be within the scope of the present application.

Claims

1. A community power dispatch optimization method, characterized in that, The community power dispatch optimization method comprises: Real-time monitoring and collecting power consumption data in a specified area, and obtaining external environment data of the specified area; storing the collected data in a time sequence form; Using a long short-term memory network (LSTM) to capture long dependencies in the stored time sequence data, and generating a prediction model by learning historical power consumption data and combining external environment data to predict power consumption demand at a future specified time; Obtaining real-time electricity prices, actual power consumption demand of users, and power grid load information, and calculating an optimal dispatch strategy based on the predicted power consumption demand using a multi-objective optimization algorithm; wherein the multi-objective optimization algorithm comprises: Computing load balancing parameters , ; Computing a power cost minimization parameter : ， Computing user comfort maximization parameters : ， in, for Total load at any given time For electricity price, For users Actual electricity demand For the distribution of electricity; Calculating a comprehensive objective function: ， wherein, is a weight coefficient for load balancing, for controlling peak reduction; is a weight coefficient for cost minimization, for controlling reduction of electricity cost; is a weight coefficient for user comfort, for controlling accuracy of user demand matching; Real-time monitoring of the operating state of user power equipment, and adjusting load distribution and equipment start-stop based on the generated optimal dispatch strategy.

2. The community power dispatch optimization method of claim 1, wherein, The real-time monitoring and collecting power consumption data in a specified area, and obtaining external environment data of the specified area; The step of storing the collected data in a time sequence form comprises: Real-time collection of power consumption data and external environment data from each user equipment through sensors and smart meters deployed in the smart grid, wherein the power consumption data comprises power consumption, voltage, and / or current information, and the external environment data comprises temperature, humidity, and / or weather information; The collected data is stored in a time sequence form and supports high-frequency updates; the collected data is represented in a time sequence table as follows: ， in, Indicates the first Each device in time Electricity consumption Total number of devices; Data sampling frequency is defined as: 。 3. The community power dispatch optimization method of claim 1, wherein, The step of using a long short-term memory network (LSTM) to capture long dependencies in the stored time sequence data, and generating a prediction model by learning historical power consumption data and combining external environment data comprises: Data preprocessing, denoising, normalization, and / or missing value filling operations are performed on the collected data, and power features and / or environmental features related to the prediction task are extracted from the data; the extracted features are input into an embedding layer to be converted into vectors suitable for processing by the long short-term memory network (LSTM); In the LSTM, the historical data to be discarded is determined through a forget gate component; the latest data added to the state is determined through an input gate component; the current cell state is updated through a state update component by integrating the outputs of the forget gate and the input gate; the hidden state of each time step is controlled through an output gate component to capture long dependencies in the time sequence data; The hidden state of the LSTM is passed to a fully connected layer and mapped to a prediction output; after post-processing, the prediction result is de-normalized and / or smoothed to obtain the final predicted power demand.

4. The community power dispatch optimization method of claim 3, wherein, The state update formula of the LSTM is: ， wherein, , , are respectively a forget gate, an input gate, an output gate, is a cell state, is a hidden state, , are weight and bias matrices; predicting future electricity demand at a specified time by a predictive model specifically includes: ， wherein are model parameters.

5. The community power dispatch optimization method of any one of claims 1-4, wherein, The step of adjusting load distribution and equipment start-stop based on the generated optimal dispatch strategy comprises: To set the device state is defined as: ， based on actual power consumption needs of users in an optimal dispatch strategy and power distributed by the system to dynamically adjust start and stop of devices, wherein the dispatch strategy comprises: ， When the actual power consumption demand of a user is greater than the power allocated by the system, the equipment is started, When the actual power consumption demand of a user is less than the power allocated by the system, the equipment is turned off.

6. The community power dispatch optimization method of any one of claims 1-4, wherein, The step of adjusting load distribution and equipment start-stop based on the generated optimal dispatch strategy comprises: Controlling peak loads wherein, is the average value of the system load to avoid grid overload.

7. The community power dispatch optimization method of any one of claims 1-4, wherein, After the step of real-time monitoring of the operating state of user power equipment, and adjusting load distribution and equipment start-stop based on the generated optimal dispatch strategy, the method further comprises: Real-time monitoring of actual power consumption of user equipment, and calculating the deviation value between actual power consumption and predicted power consumption, if the deviation value is greater than the set threshold, feedback is performed, and the prediction model is triggered to calculate the future power consumption demand of the user equipment again.

8. A community power dispatch optimization system, characterized by, The system comprises: A data acquisition module for real-time monitoring and collecting power consumption data in a specified area, and obtaining external environmental data of the specified area; the collected data is stored in time sequence form; A prediction module for capturing long dependencies in stored time series data using a long short-term memory network (LSTM), and generating a prediction model by learning historical power consumption data and combining external environmental data to predict power consumption demand at a future specified time; An optimization module for obtaining real-time electricity prices, user actual power consumption demand and power grid load information, and generating an optimal scheduling strategy using a multi-objective optimization algorithm based on the predicted power consumption demand; A scheduling module for real-time monitoring of the operating state of user power equipment, and adjusting load distribution and equipment start-stop based on the generated optimal scheduling strategy.

9. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory stores a computer program executable on the processor, and the computer program is executed by the processor to implement the steps of the community power scheduling optimization method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the community power scheduling optimization method according to any one of claims 1 to 7.

Images