Community microgrid scheduling method considering photovoltaic energy storage

Abstract

The invention provides a community microgrid scheduling method and system considering photovoltaic energy storage, and relates to the technical field of data processing, and the method comprises the steps: collecting photovoltaic energy storage data in a community; according to the photovoltaic energy storage data, trend and periodicity in the data are identified through time sequence analysis to obtain a time sequence analysis result; according to a time sequence analysis result, constructing a neural network architecture, and according to the neural network architecture, predicting future photovoltaic energy storage data to obtain a prediction result; and determining a scheduling strategy of the power equipment according to the prediction result and the running state of the current power distribution network. According to the method, the technical means of time sequence analysis, neural network prediction, linear programming and the cutting plane method are comprehensively utilized, the utilization efficiency of the photovoltaic energy storage system is improved, the electric power cost is reduced, and the stability of the electric power system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a community microgrid dispatching method considering photovoltaic energy storage. Background Art

[0002] Photovoltaic power generation is a clean, renewable form of energy that is environmentally friendly, sustainable and distributed. The output of photovoltaic power generation is affected by factors such as weather, season and sunshine intensity, and is unstable and intermittent. This leads to problems such as power supply fluctuation and power fluctuation in the photovoltaic power generation system of the community microgrid, which brings certain challenges to the operation of the power grid. The community microgrid is a small-scale power system composed of renewable energy power generation devices (such as photovoltaics), energy storage systems, traditional power generation devices (such as diesel generators) and load equipment. It can operate independently or be connected to the main power grid, with a high degree of autonomy and controllability. In order to ensure the stable operation of the community microgrid and improve energy utilization efficiency, an effective scheduling method needs to be adopted.

[0003] Energy storage technology is an effective means to solve the instability and intermittency of photovoltaic power generation. By storing excess photovoltaic power in energy storage devices and releasing it when needed, the difference between grid load and supply can be balanced, and the reliability and economy of the grid can be improved. In community microgrids, the scheduling method considering photovoltaic energy storage can optimize power generation efficiency and improve the reliability and economy of the grid. By predicting the output of photovoltaic power generation and combining the charging and discharging strategy of the energy storage system, a reasonable power generation scheduling plan can be formulated, and the microgrid can operate stably under different operating modes. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a community microgrid dispatching method taking photovoltaic energy storage into consideration, which improves the utilization efficiency of the photovoltaic energy storage system, reduces electricity costs and improves the stability of the power system by comprehensively using time series analysis, neural network prediction, linear programming and cutting plane method technical means.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a community microgrid dispatching method considering photovoltaic energy storage is provided, the method comprising:

[0007] Collect PV energy storage data within the community;

[0008] Based on the photovoltaic energy storage data, the trend and periodicity in the data are identified through time series analysis to obtain the results of time series analysis;

[0009] According to the results of time series analysis, a neural network architecture is constructed, and based on the neural network architecture, future photovoltaic energy storage data is predicted to obtain prediction results;

[0010] Determine the dispatching strategy of power equipment based on the prediction results and the current operating status of the distribution network;

[0011] According to the dispatching strategy of power equipment, linear programming is performed on the power distribution to obtain the distribution result;

[0012] According to the allocation results, the dispatch plan is adjusted in real time through the cutting plane method to realize the dispatch of community microgrid.

[0013] Furthermore, by deploying corresponding sensors and data acquisition equipment at key nodes of the community microgrid and performing calibration and initialization settings;

[0014] After the equipment is initialized, real-time data collection is started, and the collected raw data is transmitted to the central data processing unit in real time via wired or wireless means;

[0015] After the central data processing unit receives the data, it is stored in the high-speed data storage system. Before data storage, the system will verify the data and obtain the photovoltaic energy storage data in the community.

[0016] Furthermore, by collecting and organizing PV energy storage data, we can observe how the data changes over time;

[0017] By observing the changes in data over time, arranging them in chronological order, analyzing the speed of change of trends and the length of cycles;

[0018] Combine the trend and periodicity analysis results to comprehensively analyze the changing characteristics of the data;

[0019] Based on the trend and periodic change characteristics, the results of time series analysis are obtained.

[0020] Furthermore, based on the results of the time series analysis, a recurrent neural network architecture is constructed with input layer, hidden layer, output layer and activation function;

[0021] Train the neural network architecture to obtain a trained neural network architecture;

[0022] According to the trained neural network architecture, future photovoltaic energy storage data is predicted to obtain prediction results.

[0023] Further, an objective function and constraint conditions are set, wherein the objective function includes minimizing the power generation cost and minimizing the load power shortage rate or voltage deviation;

[0024] Through real number coding or binary coding, each gene represents the output power of a device, and the initial chromosomes are randomly generated to form the initial population;

[0025] Calculate the objective function value of each scheduling scheme;

[0026] According to the roulette wheel selection, the parent chromosome is selected according to the fitness value, the chromosome with high fitness is selected first, the excellent genes are retained, the selected parent is subjected to single-point crossover or multi-point crossover to generate the offspring chromosome, and the partial gene value of the offspring chromosome is randomly changed with a certain probability to increase the population diversity;

[0027] Merge the newly generated offspring population with the parent population, select the M chromosomes with the highest fitness as the next generation population, reach the maximum number of iterations, and get the final solution;

[0028] The global final solution is decoded to determine the dispatch strategy of the power equipment.

[0029] Furthermore, the calculation formula of the objective function is: ;

[0030] in, is the minimum total optimization objective value, , is the weight coefficient, is the total number of generators, It is The linear generation cost coefficient of the generator, It is The output power of the generator, It is The secondary power generation cost coefficient of the generator, is the scheduling interval, is the total number of power purchasing nodes, It is The electricity price of each power purchasing node, It is The power purchase power of each power purchase node, is the total number of load nodes, is the number of load subsets, It is The power shortage of load nodes, It is The total required power of the load nodes is is the weight coefficient of the load power failure rate, is the total number of distribution network nodes, It is The actual voltage of the node, It is The rated voltage of each node, is the weight coefficient of voltage deviation, It is The expected voltage at each node is and is the index.

[0031] Furthermore, according to the dispatching strategy and historical data of the power equipment, the power load demand is predicted to obtain the prediction result;

[0032] Based on the prediction results, a linear programming framework is constructed;

[0033] Convert the scheduling problem into a linear programming problem to determine the decision variables, objective function, and constraints of the linear programming framework;

[0034] Calculate the objective function and get the final solution;

[0035] The final solution is converted into the electricity distribution plan to obtain the distribution result.

[0036] In the second aspect, a community microgrid dispatching system considering photovoltaic energy storage includes:

[0037] The acquisition module is used to collect photovoltaic energy storage data in the community; based on the photovoltaic energy storage data, the trend and periodicity in the data are identified through time series analysis to obtain the results of the time series analysis;

[0038] The prediction module is used to construct a neural network architecture based on the results of time series analysis, and predict future photovoltaic energy storage data based on the neural network architecture to obtain prediction results; based on the prediction results and the current operating status of the distribution network, the dispatching strategy of the power equipment is determined;

[0039] The allocation module is used to formulate a dispatch plan according to the dispatch strategy of the power equipment, perform linear programming on the power distribution, and obtain the distribution result; according to the distribution result, the dispatch plan is adjusted in real time through the cutting plane method to realize the dispatch of the community microgrid.

[0040] According to a third aspect, a computing device includes:

[0041] one or more processors;

[0042] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0043] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0044] The above solution of the present invention includes at least the following beneficial effects:

[0045] By collecting and analyzing photovoltaic energy storage data in real time, it is possible to predict future photovoltaic energy storage conditions, thereby optimizing the dispatching strategy of power equipment. This helps to make full use of renewable energy, reduce dependence on traditional energy, and improve energy efficiency. Through time series analysis and neural network prediction, it is possible to predict power load demand and power generation costs, and then formulate more economical dispatching strategies, which helps to reduce power generation costs, reduce electricity purchase costs, and minimize load power shortage rates, thereby reducing overall operating costs. Through the application of linear programming and cutting plane method, the system can meet power demand while stabilizing voltage and reducing voltage deviation. Enhance the stability of the distribution network, improve power supply reliability, and be able to collect microgrid operation data in real time, and adjust the dispatching plan according to real-time deviations. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of a community microgrid dispatching method considering photovoltaic energy storage provided by an embodiment of the present invention.

[0047] Figure 2 It is a schematic diagram of a community microgrid dispatching system taking photovoltaic energy storage into consideration provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0049] like Figure 1 As shown, an embodiment of the present invention proposes a community microgrid scheduling method considering photovoltaic energy storage, and the method includes the following steps:

[0050] Step 12, collecting photovoltaic energy storage data within the community;

[0051] Step 13, based on the photovoltaic energy storage data, identifying the trend and periodicity in the data through time series analysis to obtain the result of the time series analysis;

[0052] Step 14, constructing a neural network architecture according to the results of the time series analysis, and predicting future photovoltaic energy storage data according to the neural network architecture to obtain a prediction result;

[0053] Step 15, determining a dispatching strategy for power equipment according to the prediction results and the current operating status of the distribution network;

[0054] Step 16, according to the dispatching strategy of the power equipment, linear programming is performed on the power distribution to obtain the distribution result;

[0055] Step 17, according to the allocation result, the scheduling plan is adjusted in real time through the cutting plane method to realize the scheduling of the community microgrid.

[0056] In the embodiment of the present invention, the above step 12 may include:

[0057] Step 121, by arranging corresponding sensors and data acquisition equipment at key nodes of the community microgrid and performing calibration and initialization settings;

[0058] Step 122, the device is initialized, real-time data collection is started, and the collected raw data is transmitted to the central data processing unit in real time via wired or wireless means;

[0059] Step 123, after receiving the data, the central data processing unit stores it in a high-speed data storage system. Before data storage, the system will verify the data to obtain the photovoltaic energy storage data in the community.

[0060] In the embodiment of the present invention, by carefully arranging and calibrating sensors and data acquisition equipment at key nodes of the community microgrid, the collected raw data is reliable. After the equipment is initialized, the real-time data acquisition mechanism is immediately started, and the data is transmitted to the central data processing unit in real time by wired or wireless means. This enables the system to monitor the operating status of the microgrid in real time, detect abnormalities or potential problems in a timely manner, and respond quickly, ensuring the stable operation of the microgrid and the continuity of power supply. After receiving the data, the central data processing unit stores it in a high-speed data storage system.

[0061] In a specific embodiment of the present invention, the specific steps include:

[0062] Step 121, in the community microgrid, according to the key paths and nodes of power flow and information flow, select the specific locations where sensors and data acquisition equipment need to be arranged. These locations may include the output end of the photovoltaic power station, the input and output ends of the energy storage system, the nodes of the key distribution lines, etc. According to the selected nodes, install the corresponding sensors and data acquisition equipment, use a standard signal source or a device with a known accurate value to calibrate the sensor, and the measurement error of the sensor is within an acceptable range.

[0063] Step 122, after completing the initialization settings, start the data acquisition device and start real-time acquisition of relevant data of the photovoltaic energy storage system. The data acquisition device transmits the collected raw data to the central data processing unit in real time via wired mode. During the transmission process, the integrity and security of the data are ensured to prevent data loss or tampering.

[0064] Step 123, after receiving the data, the central data processing unit stores it in a high-speed data storage system. Before the data is stored, the central data processing unit will verify the data. After the verified data, the central data processing unit will further process and convert it. The data after the above processing and conversion is the valid data of the photovoltaic energy storage system in the community.

[0065] In the embodiment of the present invention, the above step 13 may include:

[0066] Step 131, collecting and organizing photovoltaic energy storage data, and observing changes in the data over time;

[0067] Step 132, by observing the change of data over time, arranging them in chronological order, analyzing the speed of change of the trend and the length of the cycle;

[0068] Step 133, combining the trend and periodicity analysis results to comprehensively analyze the change characteristics of the data;

[0069] Step 134, obtaining the result of time series analysis according to the trend and periodic change characteristics.

[0070] In an embodiment of the present invention, by collecting and organizing photovoltaic energy storage data and observing the changes of these data over time, a deeper understanding of the operating status and performance of the photovoltaic energy storage system can be achieved. This helps to identify abnormal behavior, potential problems or room for improvement in the system. By observing the changes of data over time and arranging them in chronological order, the trends and periodic characteristics in the data can be accurately captured. Trend analysis can reveal the long-term development direction and speed of the data, while periodic analysis can identify recurring patterns or cycles in the data. The trend and periodic analysis results are combined for comprehensive analysis to fully reveal the changing characteristics of the data. This comprehensive analysis discovers hidden laws, associations or anomalies in the data, providing strong support for subsequent decision-making and optimization. The results of time series analysis can provide an important basis for the prediction and planning of photovoltaic energy storage systems. By understanding the trends and periodic characteristics of the data, future system performance, demand changes or potential problems can be predicted, thereby formulating effective plans and strategies.

[0071] In a specific embodiment of the present invention, the specific steps include:

[0072] Step 131 , obtaining historical data of the photovoltaic energy storage system from a central data processing unit or a data storage system, and using statistical tools to observe the trend and pattern of the data over time.

[0073] Step 132, observe the changing trend of the data over time, determine whether it is rising, falling or remaining stable, identify periodic changes in the data, such as daily cycles, weekly cycles, monthly cycles, etc., select a starting time point as the starting point of the cycle, select an ending time point as the end point of the cycle, and calculate the length of the cycle.

[0074] Step 133 combines the results of trend analysis and periodicity analysis, comprehensively considers the long-term trend and short-term fluctuations in the data, and conducts a comprehensive analysis of the overall change characteristics of the data. Identify abnormal points, turning points or trend changes in the data to provide a basis for subsequent decision-making and optimization.

[0075] Step 134, present the results of the time series analysis in the form of charts, reports or data tables, explain and illustrate the analysis results, explain the trends, periodic characteristics and possible causes or effects in the data, and formulate corresponding strategies or measures based on the results of the time series analysis.

[0076] In the embodiment of the present invention, the above step 14 may include:

[0077] Step 141, constructing a recurrent neural network architecture, input layer, hidden layer, output layer and activation function according to the results of the time series analysis;

[0078] Step 142, training the neural network architecture to obtain a trained neural network architecture;

[0079] Step 143, predicting future photovoltaic energy storage data based on the trained neural network architecture to obtain a prediction result.

[0080] In the embodiment of the present invention, the recurrent neural network is good at processing time series data and can capture the time dependency and long-term trend in the data, thereby improving the prediction accuracy of photovoltaic energy storage data. The neural network model has a strong adaptive ability and can learn and adjust according to new data to adapt to the ever-changing photovoltaic energy storage environment. By automating the prediction process, the dependence on human experience and judgment is reduced, and the efficiency of the prediction is improved. The prediction results can provide strong support for the optimal scheduling and decision-making of the photovoltaic energy storage system and improve the overall operating efficiency of the system.

[0081] In a specific embodiment of the present invention, the specific steps include:

[0082] Step 141, perform time series analysis on historical photovoltaic energy storage data, identify trends, seasonality, periodicity and other characteristics in the data, input data dimensions and formats, input layer: design an input layer for receiving time series data, and the number of input layer neurons matches the characteristic dimensions of the input data. Hidden layer: plan the number of hidden layers and the number of neurons in each layer. The hidden layer is the core of RNN and is responsible for capturing the temporal dependencies in the time series. Output layer: design the output layer to generate predicted values ​​for future photovoltaic energy storage data. The number of output layer neurons is determined according to the prediction target. Activation function: select a suitable activation function to enhance the nonlinear expression ability of the model.

[0083] Step 142, historical photovoltaic energy storage data is divided into training set, validation set and test set. Preprocess the data, such as normalization or standardization, to improve training efficiency. The input data enters the RNN through the input layer and is passed to the hidden layer and output layer layer by layer. At each time step, the hidden layer calculates the new hidden state based on the current input and the hidden state of the previous time step. The output layer calculates the prediction result based on the current hidden state. The loss function value during the training process is monitored, and hyperparameters such as learning rate and batch size are adjusted to optimize the training effect. The validation set is used to evaluate the performance of the model to prevent overfitting. The prediction performance of the trained RNN model is evaluated on the test set, using the indicator mean square error. , n is the total number of samples, is the true value of the qth sample, is the predicted value of the qth sample, where q is the index.

[0084] Step 143, collect the latest photovoltaic energy storage data, perform the same preprocessing operation on the new data as the training data, input the preprocessed new data into the trained RNN model, and the model generates the predicted value of future photovoltaic energy storage data based on the input data and the learned time dependency, and post-processes the predicted results, such as denormalization or destandardization, to obtain practical and interpretable prediction results. The prediction results are used for the scheduling and management of the photovoltaic energy storage system, optimizing the energy storage strategy, improving the system efficiency, monitoring the difference between the actual data and the predicted data, and timely adjusting the prediction model to adapt to system changes.

[0085] In the embodiment of the present invention, the above step 15 may include:

[0086] Step 151, setting an objective function and constraints, wherein the objective function includes minimizing the power generation cost and minimizing the load power shortage rate or voltage deviation;

[0087] Step 152, by real number coding or binary coding, each gene represents the output power of a device, and the initial chromosomes are randomly generated to form an initial population;

[0088] Step 153, calculating the objective function value of each scheduling scheme;

[0089] Step 154, selecting the parent chromosome according to the fitness value based on the roulette wheel selection, giving priority to the chromosome with high fitness, retaining the excellent genes, performing single-point crossover or multi-point crossover on the selected parent to generate the offspring chromosome, and randomly changing the value of some genes of the offspring chromosome with a certain probability to increase the population diversity;

[0090] Step 155, merging the newly generated offspring population with the parent population, selecting the M chromosomes with the highest fitness as the next generation population, reaching the maximum number of iterations, and obtaining the final solution;

[0091] The global final solution is decoded to determine the dispatch strategy of the power equipment.

[0092] In an embodiment of the present invention, clear objective functions and constraints are set to provide the genetic algorithm with an optimization direction and scope. The output power of the equipment is represented by real number coding or binary coding, and the initial population is randomly generated, which provides a starting point for searching for the optimal solution. The objective function value of each scheduling scheme is calculated to evaluate the pros and cons of the scheduling scheme. New offspring populations are generated through selection, crossover and mutation operations, which increases the diversity of the population and the search range. At the same time, the constraints of the mutated chromosomes are checked to ensure that the generated scheduling scheme meets the actual requirements. By merging the parent and offspring populations and selecting the individuals with the highest fitness as the next generation population, the optimal solution is gradually approached. The optimal scheduling strategy finally decoded can guide the actual operation of the power equipment.

[0093] In a specific embodiment of the present invention, the specific steps include:

[0094] Step 151, objective function, which is an important economic indicator in power dispatching, aims to reduce the cost of power generation by optimizing equipment dispatching. The load power shortage rate reflects the extent to which the system cannot meet the load demand, and the voltage deviation affects the normal operation of the power equipment and the stability of the power grid. Equipment operation constraints, such as the maximum and minimum output power limits of the equipment, the start and stop time limits of the equipment, etc., power grid safety constraints, such as the limits of electrical parameters such as voltage, current, and power factor, are used to ensure the safe operation of the power grid. Load demand constraints: meet the load demand within a specific time period to ensure the adequacy of power supply.

[0095] Step 152, real number encoding, each gene is represented by a real number, such as the output power of the device can be directly represented by a real number, binary encoding, the output power of the device and other continuous variables are discretized into binary strings. A set of chromosomes is randomly generated, each chromosome represents a possible scheduling scheme. The length of the chromosome is determined by the number of devices and the encoding method.

[0096] Step 153, each chromosome is evaluated by Calculate the corresponding objective function value and get the final solution, where: is the minimum total optimization objective value, , is the weight coefficient, is the total number of generators, It is The linear generation cost coefficient of the generator, It is The output power of the generator, It is The secondary power generation cost coefficient of the generator, is the scheduling interval, is the total number of power purchasing nodes, It is The electricity price of each power purchasing node, It is The power purchase power of each power purchase node, is the total number of load nodes, is the number of load subsets, It is The power shortage of load nodes, It is The total required power of the load nodes is is the weight coefficient of the load power failure rate, is the total number of distribution network nodes, It is The actual voltage of the node, It is The rated voltage of each node, is the weight coefficient of voltage deviation, It is The expected voltage at each node is and is the index.

[0097] Step 154, selection operation, according to the roulette wheel selection method, selects the parent chromosome according to the fitness value. The chromosome with a high fitness value has a greater probability of being selected to retain excellent genes. The crossover operation performs a single-point crossover or a multi-point crossover on the selected parent chromosome to generate a child chromosome. The crossover operation helps to combine the excellent genes of the parent chromosome to generate a new scheduling scheme. The mutation operation randomly changes the value of some genes of the child chromosome with a certain probability to increase the population diversity. The mutation operation helps to jump out of the local optimal solution and explore a broader solution space. The constraints of the mutated chromosome are checked to ensure that it meets all constraints. If the constraints are not met, the necessary repair operations are performed.

[0098] Step 155, merge the population, merge the newly generated offspring population with the parent population to form a new population. Select the optimal solution, and select the M chromosomes with the highest fitness as the next generation population according to the fitness value. Repeat the above process until the maximum number of iterations is reached or other stop conditions are met to obtain the final solution. After the maximum number of iterations is reached, the chromosome with the highest fitness is selected as the final solution. The global final solution is decoded to determine the dispatching strategy of the power equipment.

[0099] In the embodiment of the present invention, the above step 16 may include:

[0100] Step 161, predicting the power load demand according to the dispatching strategy and historical data of the power equipment to obtain a prediction result;

[0101] Step 162, constructing a linear programming framework according to the prediction results;

[0102] Step 163, converting the scheduling problem into a linear programming problem to determine the decision variables, objective function and constraints of the linear programming framework;

[0103] Step 164, calculating the objective function to obtain a final solution;

[0104] Step 165, convert the final solution into the power distribution plan to obtain the distribution result.

[0105] In the embodiment of the present invention, by minimizing the power generation cost and the power purchase cost, the operating cost of the power company is reduced and the economic benefit is improved. By optimizing the power dispatching plan, the load power shortage rate and voltage deviation are reduced, and the power supply quality and reliability are improved. By constructing a linear programming framework, the dispatching problem is converted into a mathematical optimization problem, which is convenient for solving using algorithms and improving the flexibility and adaptability of the system. Provide scientific power dispatching plans for power companies, support decision-making, and improve management levels. By optimizing power dispatching, resource waste and environmental pollution are reduced.

[0106] In a specific embodiment of the present invention, the specific steps include:

[0107] Step 161, using historical power load data, weather conditions, holiday information, economic activity indicators and other multivariate data, combined with the current power equipment dispatch strategy, using time series analysis, machine learning algorithms or neural network models and other methods to predict future power load demand. The prediction result should give a power load curve or load volume in the future, which will serve as the basis for subsequent linear programming.

[0108] Step 162, based on the load forecast results, determine the basic framework of linear programming, including the selection of decision variables, the setting of objective functions, and the construction of constraints. Decision variables usually include the output power of the generator set, the power purchase power of the power purchase node, etc. The objective function and constraints will be set according to the specific optimization goals.

[0109] Step 163, clarify the decision variables in the linear programming framework, such as setting For the The output power of the generator, is the power purchase power of the jth power purchase node, etc. Set the objective function, such as the total optimization target value Z, which needs to be minimized. Its expression may include the weighted sum of multiple aspects such as power generation cost, power purchase cost, load power shortage rate, voltage deviation, etc. Construct constraints, such as supply and demand balance constraints (that is, power generation and power purchase should meet load demand), equipment operation constraints (such as output power limit of generator sets, start and stop time limit, etc.), voltage and current limits, etc.

[0110] Step 164, solving the objective function to obtain an optimal solution that satisfies the constraints.

[0111] Step 165, formulate a specific power distribution plan based on the final solution obtained. The power distribution plan should clearly define the power generation plan of each generator set (including power generation time, power generation, etc.), the power purchase plan of each power purchase node (including power purchase time, power purchase amount, etc.), and the power transmission plan of the distribution network.

[0112] In the embodiment of the present invention, the above step 17 may include:

[0113] Step 171, based on the allocation result, collect the operation data of the microgrid in real time and identify the deviation from the initial allocation result;

[0114] Step 172, constructing a cutting plane according to the real-time deviation to exclude currently infeasible scheduling solutions;

[0115] Step 173, adjusting the scheduling plan according to the cutting plane to obtain an adjusted scheduling plan;

[0116] Step 174, transmitting the adjusted dispatching plan to each power generation unit in the microgrid to implement the dispatching of the community microgrid.

[0117] In the embodiment of the present invention, the microgrid operation data is collected in real time and compared with the initial allocation results, so that deviations can be found in time to ensure that the scheduling plan is consistent with the actual operating conditions. By constructing a cutting plane to exclude infeasible solutions, the system can more flexibly respond to changes in the microgrid, improve the adaptability and reliability of scheduling, and adjust the scheduling plan according to the cutting plane, so that the resources in the microgrid can be more reasonably allocated, and the resource utilization efficiency can be improved. The adjusted scheduling plan can be delivered to each power generation unit in time, and the scheduling instructions can be quickly executed to improve the scheduling efficiency of the entire microgrid.

[0118] In a specific embodiment of the present invention, the specific steps include:

[0119] Step 171, using sensors, smart meters and other equipment, collect data such as output power, status information and load demand of each power generation unit (such as solar photovoltaic panels, wind turbines, energy storage batteries, etc.) in the microgrid in real time. The accuracy and timeliness of data collection can reflect the actual operating status of the microgrid in a timely manner.

[0120] Compare the real-time collected operating data with the initial allocation results to identify the differences or deviations between the actual operating status and the plan. Deviations may include fluctuations in the output power of the power generation unit, sudden changes in load demand, etc.

[0121] Step 172, analyze the identified deviations to evaluate the degree and scope of their impact on the scheduling plan. Whether the deviation will cause the scheduling plan to become infeasible or inefficient. Based on the results of the deviation analysis, construct a cutting plane to exclude those scheduling plans that are infeasible or inefficient under the current conditions.

[0122] Step 173, based on the constructed cutting plane, the initial scheduling plan is adjusted to adapt to the actual operating state of the microgrid. The adjustment may include changing the output plan of the power generation unit, adjusting the charging and discharging strategy of the energy storage device, and redistributing the load. The adjusted scheduling plan is verified to meet the operating constraints and conditions of the microgrid, and the scheduling plan is further optimized to improve its efficiency and reliability.

[0123] Step 174, convert the adjusted dispatch plan into specific dispatch instructions and transmit them to each power generation unit in the microgrid through the communication network. The dispatch instructions are accurately transmitted and executed in a timely manner so that each power generation unit can operate according to the plan. Each power generation unit adjusts and operates according to the received dispatch instructions to achieve the overall dispatch goal of the microgrid.

[0124] like Figure 2 As shown, an embodiment of the present invention further provides a community microgrid dispatching system considering photovoltaic energy storage, including:

[0125] The acquisition module is used to collect photovoltaic energy storage data in the community; based on the photovoltaic energy storage data, the trend and periodicity in the data are identified through time series analysis to obtain the results of the time series analysis;

[0126] The prediction module is used to construct a neural network architecture based on the results of time series analysis, and predict future photovoltaic energy storage data based on the neural network architecture to obtain prediction results; based on the prediction results and the current operating status of the distribution network, the dispatching strategy of the power equipment is determined;

[0127] The allocation module is used to formulate a dispatching plan according to the dispatching strategy of the power equipment, perform linear programming on the power distribution, and obtain the distribution result; according to the distribution result, the dispatching plan is adjusted in real time through the cutting plane method to realize the dispatching of the community microgrid.

[0128] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A community microgrid dispatching method considering photovoltaic energy storage, characterized in that: The method comprises: Collect data on photovoltaic energy storage within the community; Based on the photovoltaic energy storage data, the trend and periodicity in the data are identified through time series analysis to obtain the results of time series analysis; According to the results of time series analysis, a neural network architecture is constructed, and based on the neural network architecture, future photovoltaic energy storage data is predicted to obtain prediction results; Determine the dispatching strategy of power equipment based on the prediction results and the current operating status of the distribution network; According to the dispatching strategy of power equipment, linear programming is performed on the power distribution to obtain the distribution result; According to the allocation results, the dispatch plan is adjusted in real time through the cutting plane method to realize the dispatch of community microgrid.

2. A community microgrid dispatching method considering photovoltaic energy storage according to claim 1, characterized in that: Collect PV energy storage data within the community, including: By deploying corresponding sensors and data acquisition equipment at key nodes of the community microgrid and performing calibration and initialization settings; After the equipment is initialized, real-time data collection is started, and the collected raw data is transmitted to the central data processing unit in real time via wired or wireless means; After the central data processing unit receives the data, it is stored in the high-speed data storage system. Before data storage, the system will verify the data and obtain the photovoltaic energy storage data in the community.

3. A community microgrid dispatching method considering photovoltaic energy storage according to claim 2, characterized in that: Based on the photovoltaic energy storage data, the trend and periodicity in the data are identified through time series analysis to obtain the results of time series analysis, including: By collecting and organizing photovoltaic energy storage data, we can observe the changes of data over time; By observing the changes in data over time, arranging them in chronological order, analyzing the speed of change of trends and the length of cycles; Combine the trend and periodicity analysis results to comprehensively analyze the changing characteristics of the data; Based on the trend and periodic change characteristics, the results of time series analysis are obtained.

4. A community microgrid dispatching method considering photovoltaic energy storage according to claim 3, characterized in that: According to the results of time series analysis, a neural network architecture is constructed, and based on the neural network architecture, future photovoltaic energy storage data is predicted to obtain prediction results, including: Based on the results of time series analysis, construct the recurrent neural network architecture, input layer, hidden layer, output layer and activation function; Train the neural network architecture to obtain a trained neural network architecture; According to the trained neural network architecture, future photovoltaic energy storage data is predicted to obtain prediction results.

5. A community microgrid dispatching method considering photovoltaic energy storage according to claim 4, characterized in that: According to the prediction results and the current operating status of the distribution network, the dispatching strategy of the power equipment is determined, including: Setting objective functions and constraints, wherein the objective function includes minimizing power generation costs and minimizing load power shortage rate or voltage deviation; Through real number coding or binary coding, each gene represents the output power of a device, and the initial chromosomes are randomly generated to form the initial population; Calculate the objective function value of each scheduling scheme; According to the roulette wheel selection, the parent chromosome is selected according to the fitness value, the chromosome with high fitness is selected first, the excellent genes are retained, the selected parent is subjected to single-point crossover or multi-point crossover to generate the offspring chromosome, and the partial gene value of the offspring chromosome is randomly changed with a certain probability to increase the population diversity; Merge the newly generated offspring population with the parent population, select the M chromosomes with the highest fitness as the next generation population, reach the maximum number of iterations, and get the final solution; The global final solution is decoded to determine the dispatch strategy of the power equipment.

6. A community microgrid dispatching method considering photovoltaic energy storage according to claim 5, characterized in that: The calculation formula of the objective function is: ； in, is the minimum total optimization objective value, , is the weight coefficient, is the total number of generators, It is The linear generation cost coefficient of the generator, It is The output power of the generator, It is The secondary power generation cost coefficient of the generator, is the scheduling interval, is the total number of power purchasing nodes, It is The electricity price of each power purchasing node, It is The power purchase power of each power purchase node, is the total number of load nodes, is the number of load subsets, It is The power shortage of load nodes, It is The total required power of the load nodes is is the weight coefficient of the load power failure rate, is the total number of distribution network nodes, It is The actual voltage of the node, It is The rated voltage of each node, is the weight coefficient of voltage deviation, It is The expected voltage at each node is and is the index.

7. A community microgrid dispatching method considering photovoltaic energy storage according to claim 6, characterized in that: According to the dispatching strategy of power equipment, linear programming is performed on power distribution to obtain distribution results, including: According to the dispatching strategy and historical data of power equipment, the power load demand is predicted and the prediction results are obtained; Based on the prediction results, a linear programming framework is constructed; Convert the scheduling problem into a linear programming problem to determine the decision variables, objective function, and constraints of the linear programming framework; Calculate the objective function and get the final solution; The final solution is converted into the electricity distribution plan to obtain the distribution result.

8. A community microgrid dispatching system considering photovoltaic energy storage, the system implementing the method as described in any one of claims 1 to 7, characterized in that: include: Acquisition module, used to collect photovoltaic energy storage data within the community; Based on the photovoltaic energy storage data, the trend and periodicity in the data are identified through time series analysis to obtain the results of time series analysis; The prediction module is used to construct a neural network architecture based on the results of time series analysis, and predict future photovoltaic energy storage data based on the neural network architecture to obtain prediction results; based on the prediction results and the current operating status of the distribution network, the dispatching strategy of the power equipment is determined; The allocation module is used to formulate a dispatch plan according to the dispatch strategy of the power equipment, perform linear programming on the power distribution, and obtain the distribution result; According to the allocation results, the dispatch plan is adjusted in real time through the cutting plane method to realize the dispatch of community microgrid.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images