Energy storage charging and discharging control method and system for multi-operation scenario management of distribution network
By using BiLSTM networks and model predictive control theory, a multi-scenario day-ahead governance model for distribution networks was constructed, realizing intelligent regulation of distributed energy storage across multiple time scales. This solved the problem of insufficient stability and flexibility caused by the application of a single strategy in the distribution network, and improved the stability of power grid operation and power quality.
Patent Information
- Application Number
- CN202410985269.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-23
AI Technical Summary
Existing energy storage strategies and systems have limited application scenarios in distribution networks and fail to effectively integrate multiple strategies, resulting in insufficient stability and flexibility in distribution network operation and an inability to effectively solve problems such as insufficient transmission and transformation capacity and difficulties in peak shaving and absorption during holidays.
BiLSTM network is used to predict electrical quantity data, and a day-ahead governance model for multiple scenarios of distribution network operation is constructed. Combined with model predictive control theory, energy storage charging and discharging is optimized, and intelligent regulation is carried out on multiple time scales based on distributed energy storage. The energy storage control power is adjusted in real time to cope with multiple types of operation scenarios.
It has improved the flexibility and operational stability of the distribution network, enhanced power quality, and solved problems such as short-term heavy overload, high and low voltage for users, and reverse transmission of photovoltaic power, achieving comprehensive governance of multi-scenario strategies.
Smart Images

Figure CN118539496B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of energy storage charging and discharging, and in particular, relates to an energy storage charging and discharging control method and system for managing multiple operating scenarios of a distribution network. Background Art
[0002] Due to the rapid development of new energy sources such as photovoltaics, the transformation of the power grid has not kept pace with it, resulting in a huge increase in both transformer capacity and power grid absorption pressure. In some areas, problems such as insufficient transmission and transformation capacity and difficulty in peak load regulation during holidays have emerged, which have affected the stable operation of the power system. Therefore, for power grid companies, how to solve this challenge and promote the absorption of new energy is an important task currently facing them, and it is urgent to take multiple measures to resolve the problem of new energy absorption.
[0003] In this context, energy storage technology and equipment are developing rapidly. It can realize efficient and friendly use of green energy, improve the randomness and volatility of distributed new energy, and reduce energy costs through energy storage to achieve peak shaving and valley filling, demand response, and grid peak and frequency regulation.
[0004] However, in terms of the participation of distributed energy storage in the operation and management of distribution networks, most of the current energy storage strategies and systems still have a single usage mode and application scenario, do not integrate multiple strategies, and rarely participate in distribution network management. Summary of the invention
[0005] The present invention provides a method and system for regulating energy storage charging and discharging for managing multiple operating scenarios of a distribution network, which is used to solve the technical problem that most current energy storage strategies and systems still have a single usage mode and application scenario, resulting in a need for improved comprehensiveness.
[0006] In a first aspect, the present invention provides a method for regulating energy storage charging and discharging for managing multiple operation scenarios of a distribution network, comprising:
[0007] Acquire historical electrical quantity data of the distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data;
[0008] Iteratively training a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network;
[0009] A multi-scenario day-ahead governance model for distribution network operation is constructed based on the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the multi-scenario day-ahead governance model for distribution network operation under the preset constraint conditions is:
[0010] ,
[0011] In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period;
[0012] Considering the real-time electricity consumption and weather conditions during the day, the charging and discharging optimization power output by the multi-scenario day-ahead governance model of the distribution network operation is adjusted in real time based on the model predictive control theory, and the intra-day corrected energy storage control power is output according to the preset multi-scenario governance strategy of the distribution network operation. The second objective function of the multi-scenario governance strategy of the distribution network operation under the preset constraint conditions is:
[0013] ,
[0014] In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day;
[0015] The energy storage device is controlled to charge and discharge according to the intra-day corrected energy storage control power.
[0016] In a second aspect, the present invention provides an energy storage charging and discharging control system for managing multiple operation scenarios of a distribution network, comprising:
[0017] An acquisition module is configured to acquire historical electrical quantity data of a distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data;
[0018] A training module, configured to iteratively train a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network;
[0019] A construction module is configured to construct a multi-scenario day-ahead governance model for distribution network operation according to the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the multi-scenario day-ahead governance model for distribution network operation under the condition of satisfying a preset first constraint is:
[0020] ,
[0021] In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period;
[0022] The output module is configured to consider the real-time power consumption and weather conditions during the day, adjust the charging and discharging optimization power output by the multi-scenario day-ahead governance model of the distribution network operation in real time based on the model predictive control theory, and output the intra-day corrected energy storage control power according to the preset multi-scenario governance strategy of the distribution network operation, wherein the second objective function of the multi-scenario governance strategy of the distribution network operation under the preset second constraint condition is:
[0023] ,
[0024] In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day;
[0025] The control module is configured to control the charging and discharging of the energy storage device according to the intra-day corrected energy storage control power.
[0026] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the energy storage charging and discharging control method for managing multiple operating scenarios of a distribution network according to any embodiment of the present invention.
[0027] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the energy storage charging and discharging control method for managing multiple operating scenarios of a distribution network according to any embodiment of the present invention.
[0028] The energy storage charging and discharging control method and system for the management of multiple operating scenarios of distribution networks in the present application, for multiple operating scenarios such as short-term heavy overload, high / low voltage of users, photovoltaic power reverse transmission, etc. in the operation of distribution networks, intelligently adjust the charging and discharging power at multiple time scales based on distributed energy storage, and build a distribution network operation multi-scenario day-ahead management model according to the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model. Considering the real-time electricity consumption and weather conditions during the day, the charging and discharging optimization power output by the distribution network operation multi-scenario day-ahead management model is adjusted in real time based on the model predictive control theory, and the daily corrected energy storage control power is output according to the preset distribution network operation multi-scenario management strategy. It can solve multiple types of operating problems such as short-term heavy overload, high / low voltage of users, photovoltaic power reverse transmission, etc. in a targeted manner, enhance the flexibility of the distribution network, improve the power quality, and enhance the operation stability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0030] Figure 1 A flow chart of a method for regulating energy storage charging and discharging for managing multiple operation scenarios of a distribution network provided by an embodiment of the present invention;
[0031] Figure 2 A structural block diagram of an energy storage charging and discharging control system for managing multiple operation scenarios of a distribution network provided by an embodiment of the present invention;
[0032] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0034] See also Figure 1 , which shows a flow chart of a method for energy storage charging and discharging control for managing multiple operating scenarios of distribution networks in the present application.
[0035] like Figure 1 As shown in FIG. 1 , the energy storage charging and discharging control method for managing multiple operation scenarios of the distribution network specifically includes the following steps:
[0036] Step S101, acquiring historical electrical quantity data of a distribution station area, and performing standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data.
[0037] In this step, the power, voltage, current, photovoltaic reverse power and other electrical information of the distribution station area are collected through the terminal equipment data collection module. One day is divided into 96 time periods, and the data is collected every 15 minutes. m is the length of the data set.
[0038] Station power ,Voltage , current , photovoltaic reverse power .
[0039] in, is the mth historical power data of the substation, is the mth historical area voltage data, is the mth historical current data of the substation, The photovoltaic reverse power data of the mth historical substation;
[0040] The historical electrical quantity data is normalized and the data range is linearly scaled to the interval [0,1] to make the neural network training process more stable, accurate and fast. Represents the original value of the data, Represents the value after the original value is normalized. , Indicates the maximum and minimum values of the data. Specifically:
[0041] .
[0042] Step S102, iteratively training a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network.
[0043] In this step, the BiLSTM network (bidirectional long short-term memory network) consists of a forward LSTM network and a backward LSTM network. During the BiLSTM network training process, due to the addition of the backward LSTM network, while learning from the time series forward, the BiLSTM network can well handle the bidirectional long-term correlation between sequence time steps and mine the abstract features inside the time series data.
[0044] Neural network model building
[0045] (1) Forget gate: selectively removes weakly related information in data such as power, voltage, and current.
[0046] ,
[0047] In the formula, Represents the output of the hidden layer at time t-1 (storing power, voltage, current and other information); Indicates the input power, voltage, current and other information values at time t; express Activation function; represents the output of the forget gate; represents the forget gate weight matrix; Represents the forget gate offset value matrix;
[0048] (2) Input gate: Updates the state of the memory unit together with the sigmoid layer and hidden layer:
[0049] ,
[0050] ,
[0051] ,
[0052] Where: represents the input gate output; represents the input gate weight matrix; Represents the input gate offset value matrix; Indicates the updated value of the unit status at time t regarding power, voltage, current, etc.; , Represent the neuron weight matrix and bias value matrix respectively; It represents important information such as power, voltage, current, etc. retained in the unit state at time t.
[0053] (3) Output gate: The bidirectional LSTM matrix is superimposed and outputted together:
[0054] ,
[0055] ,
[0056] ,
[0057] Where: represents the output gate output, , Represent the output gate weight matrix and offset value matrix respectively. , Respectively represent the output of the forward and reverse hidden layers at time t; , Respectively represent the forward and reverse output layer weight matrices; represents the output layer bias matrix; Represents the output layer output.
[0058] 2. Neural network model parameter selection
[0059] In terms of training hyperparameter settings, the number of hidden units in the BiLSTM layer is set to 200, the batch size is set to 128, the state activation function and gate activation function of the network are selected as tanh and sigmoid respectively, the Adam solver is selected, and the training iteration is 300 rounds. At the same time, in order to reduce overfitting, a Dropout layer is added after the Bi-LSTM layer to improve the generalization ability of the model. The initial learning rate is set to 0.006, and the learning rate is reduced by multiplying it by a factor of 0.2 after 150 rounds of training.
[0060] During training, the input weights are initialized using Xavier to keep the activation values of each layer and the variance of the state gradient as consistent as possible during the transmission process; during the recursive process, the weight values are initialized using Orthogonal to solve the problems of gradient disappearance and gradient explosion in deep networks. 70% of the data is set as the training set, 15% as the validation set, and 15% as the test set.
[0061] Step S103: construct a multi-scenario day-ahead management model for distribution network operation based on the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model.
[0062] In this step, the rated power of the substation is , set the rated voltage of the station area , energy storage charging and discharging power ( When it is greater than 0, it is charged, and when it is less than 0, it is discharged), the operating voltage In terms of governance strategies, for heavy overload management in the substation, when the load rate exceeds 80%, the energy storage discharge is started to keep the load rate below 80% as much as possible; for high and low voltage management, when the substation voltage is 10% higher than the rated voltage, it is high voltage, and when the substation voltage is 10% lower than the rated voltage, it is low voltage; for photovoltaic reverse power, the strategy of "eliminating as much as possible" is adopted to absorb all photovoltaic power as much as possible.
[0063] The comprehensive balance of energy storage charging and discharging power is used to manage the multi-scenario problems of distribution network operation, and a multi-scenario day-ahead management model for distribution network operation is established. The first objective function of the multi-scenario day-ahead management model for distribution network operation under the preset constraints is:
[0064] ,
[0065] In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period;
[0066] It is the overall objective function, and its purpose is to comprehensively manage the overall distribution network operation scenario.
[0067] If a heavy overload occurs, the energy storage will discharge to supply the load to solve the problem of heavy overload in the area, and try to keep the load rate of the area below 80%. ,when hour, , otherwise 0;
[0068] This means that if photovoltaic power reverse transmission occurs, the energy storage will be charged to absorb the photovoltaic reverse transmission power as much as possible and reduce the situation of wind and solar power abandonment. ,when hour, =1, otherwise 0;
[0069] Indicates that if high / low voltage occurs in the substation, the high / low voltage situation can be alleviated by discharging / charging energy storage, and the high / low voltage control coefficient , when high voltage appears, , ; When low voltage occurs, , ; The rest of the time , ; Finally multiply by the coefficient Convert voltage value to power value to , Optimize calculations for uniform units.
[0070] It should be noted that the constraints include energy storage charging and discharging power constraints and energy storage real-time capacity constraints;
[0071] The expression of the energy storage charging and discharging power constraint is:
[0072] ,
[0073] In the formula, is the maximum discharge power, The maximum charging power;
[0074] The expression of the energy storage real-time capacity constraint is:
[0075] ,
[0076] In the formula, is the real-time capacity of energy storage, is the rated capacity of energy storage.
[0077] When the energy storage capacity reaches the lower limit When the energy storage capacity reaches the upper limit, the discharge of the energy storage is stopped and only charging of the energy storage is allowed; when the energy storage capacity reaches the upper limit When , the energy storage is stopped from being discharged and only the energy storage is allowed to be charged. The energy storage charging and discharging power of the day before is finally obtained through linear programming model calculation .
[0078] Step S104, taking into account the real-time electricity consumption and weather conditions during the day, the charging and discharging optimization power output by the distribution network operation multi-scenario day-ahead management model is adjusted in real time based on model predictive control theory, and the daily corrected energy storage control power is output according to the preset distribution network operation multi-scenario management strategy.
[0079] In this step, within the day, based on the day-ahead energy storage charging and discharging power , update the collected electrical quantity data in real time, and make rolling optimization adjustments to the charging and discharging power.
[0080] Taking into account the inaccurate forecast information (PV fluctuations, load fluctuations, etc.) of the day, the real-time energy storage charging and discharging power is set to optimize the distribution network operation management in the next hour (one control point every 15 minutes, 4 control points). is the difference variable between the energy storage charging and discharging power within the day and the previous day’s prediction, The amount of energy stored during the day, is the energy storage charging and discharging power output on the previous day, keeping the other constraints unchanged. The second objective function of the distribution network operation multi-scenario governance strategy under the preset constraints is:
[0081] ,
[0082] In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day;
[0083] ,
[0084] It means that the distribution network scenario of heavy overload in the substation is managed through power optimization within the day;
[0085] It means that the photovoltaic reverse transmission distribution network scenario is managed through power optimization during the day;
[0086] It means that the photovoltaic reverse transmission distribution network scenario is managed through power optimization during the day;
[0087] After optimizing and calculating the time range [i,i+3] at the i-th moment, only the output power at the i-th moment is executed for the energy storage, and the energy storage is regulated in real time. At the i+1th moment, the time range [i+1,i+4] is optimized and calculated, and only the output power at the i+1th moment is executed, and so on. At the same time, the data values predicted by the BiLSTM network are updated in real time to achieve the effect of model predictive control.
[0088] Step S105 , controlling the charging and discharging of the energy storage device according to the intra-day corrected energy storage control power.
[0089] In this step, the actual charging and discharging power of the energy storage is finally obtained as , which is the sum of the day-ahead predicted output power value and the real-time corrected power value.
[0090] In a specific embodiment, after the energy storage charging and discharging power is controlled, the difference variable between the energy storage charging and discharging power and the day-ahead prediction is obtained. In order to effectively quantify and evaluate the accuracy of the day-ahead optimization model, the accuracy index of the energy storage day-ahead optimization model is proposed to reasonably evaluate the model optimization capability. The accuracy index includes the charging and discharging power accuracy index and the energy storage battery capacity accuracy index;
[0091] The expression for calculating the charge and discharge power accuracy index is:
[0092] ,
[0093] In the formula, It is the indicator of charge and discharge power accuracy;
[0094] The expression for calculating the energy storage battery capacity accuracy index is:
[0095] ,
[0096] In the formula, It is an indicator of the accuracy of the energy storage battery capacity. To optimize the energy storage capacity during the day, Optimize energy storage capacity for the day ahead.
[0097] The precision evaluation quantitative index is used as the input feature together with the electrical quantity data as the training set of the BiLSTM network for the next day, namely, the charging and discharging power accuracy index, the energy storage battery capacity accuracy index, and the power ,Voltage , current , photovoltaic reverse power , improve the BiLSTM network training level and further improve the accuracy of the prediction model.
[0098] In summary, the method of the present application is applied to scenarios such as photovoltaic consumption, load fluctuation smoothing, low voltage control, peak shaving and valley filling, effectively alleviating distribution network operation problems, improving distribution network flexibility, improving power quality, and improving power grid stability. It also integrates multi-scenario strategies, and intelligently adjusts charging and discharging power at multiple time scales based on distributed energy storage. It can comprehensively manage multiple types of operating problems in the same distribution station area, and intelligently and automatically adjust the charging and discharging power of energy storage. It adopts a model-data dual-driven method to comprehensively solve the problem of inaccurate prediction of traditional neural network models. Model-based energy storage charging and discharging control cannot avoid considering long-term operation problems, and uses the intraday optimized corrected power as the input feature set of the neural network to further improve the model prediction accuracy.
[0099] See also Figure 2 , which shows a structural block diagram of an energy storage charging and discharging control system for managing multiple operating scenarios of distribution networks in the present application.
[0100] like Figure 2 As shown, the energy storage charging and discharging control system 200 includes an acquisition module 210, a training module 220, a construction module 230, an output module 240 and a control module 250.
[0101] The acquisition module 210 is configured to acquire historical electrical quantity data of the distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data;
[0102] A training module 220 is configured to iteratively train a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network;
[0103] The construction module 230 is configured to construct a distribution network operation multi-scenario day-ahead governance model according to the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the distribution network operation multi-scenario day-ahead governance model under the condition of satisfying the preset first constraint is:
[0104] ,
[0105] In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period;
[0106] The output module 240 is configured to consider the real-time power consumption and weather conditions during the day, adjust the charging and discharging optimization power output by the distribution network operation multi-scenario day-ahead governance model in real time based on the model predictive control theory, and output the intra-day corrected energy storage control power according to the preset distribution network operation multi-scenario governance strategy, wherein the second objective function of the distribution network operation multi-scenario governance strategy under the preset second constraint condition is:
[0107] ,
[0108] In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day;
[0109] The control module 250 is configured to control the charging and discharging of the energy storage device according to the intra-day corrected energy storage control power.
[0110] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.
[0111] In other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, and when the program instructions are executed by a processor, the processor executes the energy storage charging and discharging control method for distribution network multi-operation scenario management in any of the above method embodiments;
[0112] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:
[0113] Acquire historical electrical quantity data of the distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data;
[0114] Iteratively training a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network;
[0115] A multi-scenario day-ahead governance model for distribution network operation is constructed based on the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the multi-scenario day-ahead governance model for distribution network operation under the preset constraint conditions is:
[0116] ,
[0117] In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period;
[0118] Considering the real-time electricity consumption and weather conditions during the day, the charging and discharging optimization power output by the multi-scenario day-ahead governance model of the distribution network operation is adjusted in real time based on the model predictive control theory, and the intra-day corrected energy storage control power is output according to the preset multi-scenario governance strategy of the distribution network operation. The second objective function of the multi-scenario governance strategy of the distribution network operation under the preset constraint conditions is:
[0119] ,
[0120] In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day;
[0121] The energy storage device is controlled to charge and discharge according to the intra-day corrected energy storage control power.
[0122] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the energy storage charging and discharging control system for the management of multiple operating scenarios of the distribution network, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the energy storage charging and discharging control system for the management of multiple operating scenarios of the distribution network via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0123] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3In the example, the connection through the bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the energy storage charging and discharging control method for the management of multiple operating scenarios of the distribution network in the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function controls of the energy storage charging and discharging control system for the management of multiple operating scenarios of the distribution network. The output device 340 may include display devices such as display screens.
[0124] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.
[0125] As an implementation mode, the electronic device is applied to an energy storage charging and discharging control system for managing multiple operation scenarios of a distribution network, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:
[0126] Acquire historical electrical quantity data of the distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data;
[0127] Iteratively training a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network;
[0128] A multi-scenario day-ahead governance model for distribution network operation is constructed based on the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the multi-scenario day-ahead governance model for distribution network operation under the preset constraint conditions is:
[0129] ,
[0130] In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period;
[0131] Considering the real-time electricity consumption and weather conditions during the day, the charging and discharging optimization power output by the multi-scenario day-ahead governance model of the distribution network operation is adjusted in real time based on the model predictive control theory, and the intra-day corrected energy storage control power is output according to the preset multi-scenario governance strategy of the distribution network operation. The second objective function of the multi-scenario governance strategy of the distribution network operation under the preset constraint conditions is:
[0132] ,
[0133] In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day;
[0134] The energy storage device is controlled to charge and discharge according to the intra-day corrected energy storage control power.
[0135] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.
[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for regulating energy storage charging and discharging for multiple operation scenarios of distribution networks. It is characterized in that include: Acquire historical electrical quantity data of the distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data; Iteratively training a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network; A multi-scenario day-ahead governance model for distribution network operation is constructed based on the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the multi-scenario day-ahead governance model for distribution network operation under the preset constraint conditions is: , In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period; Considering the real-time electricity consumption and weather conditions during the day, the charging and discharging optimization power output by the multi-scenario day-ahead governance model of the distribution network operation is adjusted in real time based on the model predictive control theory, and the intra-day corrected energy storage control power is output according to the preset multi-scenario governance strategy of the distribution network operation. The second objective function of the multi-scenario governance strategy of the distribution network operation under the preset constraint conditions is: , In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day; Controlling the charging and discharging of the energy storage device according to the intra-day corrected energy storage control power; Wherein, after outputting the intraday corrected energy storage control power according to the preset distribution network operation multi-scenario governance strategy, the method further includes: The accuracy index is calculated based on the difference between the intra-day corrected energy storage control power and the charge-discharge optimization power output by the multi-scenario day-ahead governance model of the distribution network operation; The accuracy index is used as an input feature and the second target historical electrical quantity data is input into a preset BiLSTM network for iterative training to obtain a second electrical quantity prediction model for the next day, wherein the second target historical electrical quantity data is the electrical quantity data within the day after standard normalization processing; Constructing a multi-scenario day-ahead governance model for the next day's distribution network operation based on the day-ahead predicted electrical quantity data output by the second electrical quantity prediction model; Taking into account the real-time electricity consumption and weather conditions of the next day, the charging and discharging optimization power output by the multi-scenario day-ahead management model of the distribution network operation for the next day is adjusted in real time based on the model predictive control theory, and the corrected energy storage control power for the next day is output according to the preset multi-scenario management strategy of the distribution network operation.
2. According to claim 1, a method for regulating energy storage charging and discharging for managing multiple operation scenarios of distribution networks, It is characterized in that The accuracy index includes the charge and discharge power accuracy index and the energy storage battery capacity accuracy index; The expression for calculating the charge and discharge power accuracy index is: , In the formula, It is the indicator of charge and discharge power accuracy; The expression for calculating the energy storage battery capacity accuracy index is: , In the formula, It is an indicator of the accuracy of the energy storage battery capacity. To optimize the energy storage capacity during the day, Optimize energy storage capacity for the day ahead.
3. According to claim 1, a method for regulating energy storage charging and discharging for managing multiple operation scenarios of distribution networks, It is characterized in that The constraints include energy storage charging and discharging power constraints and energy storage real-time capacity constraints; The expression of the energy storage charging and discharging power constraint is: , In the formula, is the maximum discharge power, The maximum charging power; The expression of the energy storage real-time capacity constraint is: , In the formula, To optimize the energy storage capacity on the day before, is the rated capacity of energy storage.
4. A storage energy charging and discharging control system for managing multiple operation scenarios of distribution networks according to any of the methods described in claims 1-3, It is characterized in that include: An acquisition module is configured to acquire historical electrical quantity data of a distribution station area, and perform standard normalization processing on the historical electrical quantity data to obtain first target historical electrical quantity data; A training module, configured to iteratively train a preset BiLSTM network according to the first target historical electrical quantity data to obtain a first electrical quantity prediction model, wherein the BiLSTM network is a bidirectional long short-term memory network including a forward LSTM network and a backward LSTM network; A construction module is configured to construct a multi-scenario day-ahead governance model for distribution network operation according to the day-ahead predicted electrical quantity data output by the first electrical quantity prediction model, wherein the first objective function of the multi-scenario day-ahead governance model for distribution network operation under the condition of satisfying a preset first constraint is: , In the formula, is the heavy overload existence coefficient, is the rated power of the station area, For the day-ahead forecast of the area power, To optimize the energy storage charging and discharging power variables on the day before, is the photovoltaic reverse transmission existence coefficient, is the photovoltaic reverse power, For the day-ahead current forecast, is the high voltage coefficient of the station area, is the day-ahead voltage forecast, is the rated voltage of the station area, is the low voltage coefficient of the substation, and n is the time period; The output module is configured to consider the real-time power consumption and weather conditions during the day, adjust the charging and discharging optimization power output by the multi-scenario day-ahead governance model of the distribution network operation in real time based on the model predictive control theory, and output the intra-day corrected energy storage control power according to the preset multi-scenario governance strategy of the distribution network operation, wherein the second objective function of the multi-scenario governance strategy of the distribution network operation under the preset second constraint condition is: , In the formula, To optimize the energy storage charging and discharging power variables within the day, To predict the power of the substation area within the day, To predict the photovoltaic reverse power within the day, For intraday current forecast, Forecast voltage for the day; The control module is configured to control the charging and discharging of the energy storage device according to the intra-day corrected energy storage control power.
5. An electronic device, It is characterized in that include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any one of claims 1 to 3.
6. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.
