Distribution network system operation optimization method and device based on virtual power plant cluster, medium and product

Abstract

The invention discloses a distribution network system operation optimization method and device based on a virtual power plant cluster, a medium and a product, and relates to the field of virtual power plants, and the method comprises the steps: constructing a distribution network system framework based on the virtual power plant cluster; respectively determining the correlation between the characteristic parameter data and the distributed power generation historical data and the correlation between the characteristic parameter data and the user load historical data, and determining key parameters respectively corresponding to the distributed power generation historical data and the user load historical data; constructing a power generation prediction model of the distributed power generation unit; constructing a user load prediction model; constructing a shared energy storage unit configuration optimization model according to the power generation prediction data of the distributed power generation unit and the user load prediction data; and performing distribution network system operation optimization of the virtual power plant cluster according to the power generation prediction model of the distributed power generation unit, the user load prediction model and the configuration optimization model of the shared energy storage unit. According to the invention, the stability of the distribution network system can be improved, and the stability and reliability of VPP cluster operation can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of virtual power plants, and in particular to a distribution network system operation optimization method, equipment, medium and product based on a virtual power plant cluster. Background Art

[0002] In order to achieve the requirements of energy conservation and emission reduction, distributed power generation technology has developed rapidly. However, renewable energy sources such as photovoltaic power generation and wind power generation have obvious intermittent and volatile characteristics due to the significant impact of environmental factors. After large-scale access to the power grid, this type of power generation may bring major challenges to the stability, power quality and operation planning of the distribution network system. In addition, in the process of power dispatching, dispatching decisions need to be based on accurate power generation forecasts and load forecasts, and the operating status of various electrical equipment must be reasonably determined to form a scientific dispatching plan. With the transformation of new power systems, traditional prediction models and methods can no longer meet current needs and urgently need to be optimized and improved. Especially in the current economic context, flexible loads such as air conditioners, water heaters and electric vehicles are increasingly receiving attention. Therefore, the construction of prediction models must further explore the characteristic relationships under different scenarios.

[0003] Virtual power plants (VPPs) can effectively optimize load distribution, improve the absorption capacity of renewable energy, and enhance the operational stability of power systems by integrating distributed generators, energy storage systems, and controllable loads into a whole to participate in the electricity market. At the same time, power storage technology can improve the flexibility and reliability of grid operation through peak load regulation and frequency modulation, and further promote the absorption of renewable energy. Through the construction and application of virtual power plants, the potential of distributed energy will be fully released, helping to cope with the challenges brought by large-scale access to new energy, and promoting the development of power systems in a more efficient, flexible, and intelligent direction. In addition, with the rapid development of modern intelligent communication technology, modern detection technology, and modern control technology, the operating status of the distribution network system can be efficiently monitored and analyzed, providing more convenient conditions for the operation optimization of the distribution network system.

[0004] However, the existing technologies and methods for the research on power transmission methods and information interaction technologies for distribution network systems and VPP clusters lack a relatively complete system architecture, and rarely consider the data collection, storage and transmission process of VPP clusters, and have low practical value. In addition, when predicting and analyzing power generation data and power load, the impact of various external factors is not fully considered. As a result, in the process of constructing power generation prediction models and user load prediction models, dimensional disaster problems often occur, reducing the prediction accuracy of the prediction model. Existing technologies have little research on self-built energy storage devices in distribution network systems. The high cost of user-built energy storage devices has limited the development and application of energy storage technology, which is not conducive to the efficient consumption of distributed resources.

[0005] Based on the problems of the existing technology, it is urgently necessary to provide a new VPP cluster distribution network system operation optimization method to improve the stability of the distribution network system, thereby effectively improving the stability and reliability of the VPP cluster operation. Summary of the invention

[0006] The purpose of this application is to provide a distribution network system operation optimization method, equipment, medium and product based on a virtual power plant cluster, which can improve the stability of the distribution network system and effectively improve the stability and reliability of the VPP cluster operation.

[0007] To achieve the above objectives, this application provides the following solutions:

[0008] In a first aspect, the present application provides a distribution network system operation optimization method based on a virtual power plant cluster, and the distribution network system operation optimization method based on a virtual power plant cluster includes:

[0009] A distribution network system framework based on a virtual power plant cluster is constructed; the distribution network system framework includes: a shared energy storage unit that communicates with the virtual power plant cluster, a power grid control center, an electricity trading market, and common user loads; the common user loads are loads that do not participate in macro-control and are related to user habits; the shared energy storage unit of the virtual power plant is used to provide energy storage services to users in the form of users sharing their own energy storage; the form of users sharing their own energy storage is to enable all users to share the energy storage cost and enjoy the energy storage service together; the virtual power plant cluster includes multiple interconnected virtual power plants; each virtual power plant includes: an aggregation unit and a virtual power plant system control center that transmits power and interacts with the aggregation unit The aggregation unit includes: photovoltaic power generation unit, wind power generation unit and user controllable load; the user controllable load includes: user common load controlled by intelligent switch and user intelligent power load; the user common load controlled by intelligent switch is that the user submits the power plan to the virtual power plant control center according to its own power demand, and the virtual power plant regulates the current user load according to the user's power plan and the distribution network system regulation demand; the user intelligent power load is the user's power equipment that adjusts the power through the intelligent controller, and when the virtual power plant has the regulation demand, the power of the current load is adjusted; each virtual power plant transmits power and exchanges information through the virtual power plant control center;

[0010] Obtaining distributed generation historical data and user load historical data and corresponding characteristic parameter data according to the distribution network system framework; the characteristic parameter data includes: historical wind power, light, electricity price, temperature and humidity;

[0011] Respectively determine the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determine the key parameters corresponding to the distributed generation historical data and the user load historical data;

[0012] According to the historical data of distributed power generation and the corresponding key parameters, a network of comprehensive convolutional neural network and bidirectional long short-term memory network is used to build a distributed power generation unit power generation prediction model; according to the historical data of user load and the corresponding key parameters, a network of comprehensive convolutional neural network and bidirectional long short-term memory network is used to build a user load prediction model; the distributed power generation unit power generation prediction model is used to output the distributed power generation unit power generation prediction data to the virtual power plant system control center; the user load prediction model is used to output the user load prediction data to the virtual power plant system control center;

[0013] Based on the power generation forecast data of distributed generation units and user load forecast data, a shared energy storage unit configuration optimization model is constructed;

[0014] The distribution network system operation of the virtual power plant cluster is optimized based on the distributed generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model.

[0015] Optionally, the determining of the correlation between the characteristic parameter data and the distributed generation history data and the user load history data, and determining the key parameters corresponding to the distributed generation history data and the user load history data, respectively, specifically includes:

[0016] Pearson correlation analysis is used to associate the characteristic parameter data with the historical data of distributed generation and the historical data of user load, respectively, and the Pearson correlation coefficient is obtained;

[0017] Correlations were determined based on Pearson’s correlation coefficient;

[0018] The key parameters corresponding to the distributed generation historical data and the user load historical data are determined according to the correlation.

[0019] Optionally, the distributed generation unit power generation prediction model and the user load prediction model are both based on the mean square error M SE Cooperate with Adam optimizer for training; and use mean square error M SE and the mean percentage error M APE As an evaluation indicator.

[0020] Optionally, the shared energy storage unit configuration optimization model specifically includes:

[0021] Using the formula Determine the energy rate constraint of the shared energy storage unit; where γ is the fixed proportional coefficient of the energy storage power upper limit and capacity of the shared energy storage unit, is the maximum charging and discharging power of the shared energy storage unit, is the maximum energy storage capacity of the shared energy storage unit;

[0022] Using the formula Determine the energy storage power and energy storage capacity constraints of the shared energy storage unit;

[0023] in, is the minimum discharge power of the shared energy storage unit, is the maximum discharge power of the shared energy storage unit, is the minimum energy storage capacity of the shared energy storage unit, is the maximum energy storage capacity of the shared energy storage unit, P ESS is the discharge power of the shared energy storage unit, E ESS is the energy storage battery capacity of the shared energy storage unit, It is the energy storage battery capacity of the shared energy storage unit just before the distribution network system reaches the worst operating condition. Predict power shortage for the distribution network system under the worst operating conditions; Forecast power for user loads in the distribution network system, is the predicted power of distributed generation in the distribution network system, η is the capacity margin index of the distribution network system, and the capacity margin index meets the requirements of compensating the prediction deviation of the distribution network system and the maximum predicted value capacity requirements of the user's controllable load when it is not regulated, with the goal of ensuring the reliability of the distribution network power supply, Δt worst It is the duration of the worst operating state of the distribution network system.

[0024] Optionally, the operation optimization of the distribution network system of the virtual power plant cluster is performed according to the distributed power generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model, specifically including:

[0025] Establish an optimization objective function with the goal of minimizing the power loss of the distribution network system;

[0026] According to the optimization objective function and the distribution network system operation mode based on the virtual power plant cluster, the distribution network system operation optimization of the virtual power plant cluster is carried out.

[0027] Optionally, the optimizing the operation of the distribution network system of the virtual power plant cluster according to the optimization objective function and the operation mode of the distribution network system based on the virtual power plant cluster specifically includes:

[0028] When the operation mode of the distribution network system based on the virtual power plant cluster is the worst operation mode, the distribution network system can only obtain power from the virtual power plant cluster and the shared energy storage unit, and the optimization objective function is:

[0029] Among them, P ESS is the discharge power of the shared energy storage unit, P loss is the power loss power of the distribution network system; P support is the total power supply of the distribution network system; P load.regularP is the power consumption of the user's normal load; load.control is the power consumption of the user controllable load, P system.support is the power generation, in the worst operation mode P system，support Reaching the maximum value ΔP load.control is the control power of the user's controllable load, ΔP in the worst operation mode load.control Adjust to maximum value

[0030] When the operation mode of the distribution network system based on the virtual power plant cluster is the normal operation mode, the electric energy generated by the distributed generation unit supplies power to the user load according to the distribution service demand, and is also used for auxiliary service demand and for peak and frequency regulation services of the distribution network system. The optimization objective function is to fully absorb the new energy power generation, and the optimization objective function is:

[0031]

[0032] Among them, ΔP is the power demand for peak and frequency regulation services of the distribution network system, which is determined by the day-ahead power trading market. system.support , ΔP ESS , ΔP load.control They respectively represent the auxiliary service power of distributed generation units, shared energy storage units, and user-controllable loads participating in the peak and frequency regulation of the distribution network system.

[0033] In a second aspect, the present application provides a distribution network system operation optimization device based on a virtual power plant cluster, and the distribution network system operation optimization device based on a virtual power plant cluster includes:

[0034] A framework construction module is used to construct a distribution network system framework based on a virtual power plant cluster; the distribution network system framework includes: a shared energy storage unit that communicates with the virtual power plant cluster, a power grid control center, an electricity trading market, and user common loads; the user common loads are loads that do not participate in macro-control and are related to user habits; the shared energy storage unit of the virtual power plant is used to provide energy storage services to users in the form of users sharing their own energy storage; the form of users sharing their own energy storage is to enable all users to share the energy storage cost and enjoy the energy storage service together; the virtual power plant cluster includes multiple interconnected virtual power plants; each virtual power plant includes: an aggregation unit and a virtual power plant system that transmits power and interacts with the aggregation unit System control center; the aggregation unit includes: photovoltaic power generation unit and wind power generation unit and user controllable load; user controllable load includes: user common load controlled by intelligent switch and user intelligent power load; the user common load controlled by intelligent switch is the user submitting power plan to the virtual power plant control center according to its own power demand, and the virtual power plant regulates the current user load according to the user's power plan and the distribution network system regulation demand; the user intelligent power load is the user's power equipment that adjusts the power through the intelligent controller, and when the virtual power plant has regulation demand, the power of the current load is adjusted; each virtual power plant transmits power and exchanges information through the virtual power plant control center;

[0035] A data acquisition module is used to acquire historical data of distributed generation and user load and corresponding characteristic parameter data according to the distribution network system framework; the characteristic parameter data includes: historical wind power, light, electricity price, temperature and humidity;

[0036] A key parameter determination module, used to determine the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determine the key parameters corresponding to the distributed generation historical data and the user load historical data respectively;

[0037] The prediction model determination module is used to construct a distributed generation unit power generation prediction model based on the distributed generation historical data and the corresponding key parameters, using a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network; the user load prediction model is constructed based on the user load historical data and the corresponding key parameters, using a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network; the distributed generation unit power generation prediction model is used to output the distributed generation unit power generation prediction data to the virtual power plant system control center; the user load prediction model is used to output the user load prediction data to the virtual power plant system control center;

[0038] An optimization model determination module is used to construct a shared energy storage unit configuration optimization model based on the distributed generation unit power generation forecast data and the user load forecast data;

[0039] The operation optimization module is used to optimize the operation of the distribution network system of the virtual power plant cluster based on the distributed generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model.

[0040] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the distribution network system operation optimization method based on a virtual power plant cluster.

[0041] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the distribution network system operation optimization method based on a virtual power plant cluster.

[0042] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the distribution network system operation optimization method based on a virtual power plant cluster.

[0043] According to the specific embodiments provided in this application, this application has the following technical effects:

[0044] The present application provides a distribution network system operation optimization method, equipment, medium and product based on a virtual power plant cluster, constructs a distribution network system framework based on a virtual power plant cluster, analyzes and studies the power data and information data of the distribution network system through the VPP system control center, and then realizes energy transmission and control signal transmission between VPP clusters, and fully considers the data confidentiality requirements during data transmission. The constructed distribution network system framework fully considers the shared energy storage unit and the power grid power control center, which can ensure the power supply stability and reliability of the distribution network system. When predicting and analyzing the power generation forecast data of distributed power generation units and user load forecast data, the influence of various external factors is fully considered, and through the correlation analysis of relevant feature parameters, the relevant feature parameters are screened, the features with strong correlation are selected, the number of relevant feature parameters is reduced, the overfitting phenomenon is alleviated, and the mining ability of the associated feature parameters is strengthened, thereby improving the prediction accuracy. The use of shared energy storage units enables all users to share the energy storage cost and enjoy the energy storage service together, which is conducive to the efficient consumption of distributed resources, and takes into account the worst operating conditions of the distribution network system, improves the stability of the distribution network system, and effectively improves the stability and reliability of the VPP cluster operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 This is a flow chart of a method for optimizing the operation of a distribution network system based on a virtual power plant cluster in one embodiment of the present application;

[0047] Figure 2 This is a schematic diagram of the distribution network system framework based on the virtual power plant cluster;

[0048] Figure 3 Schematic diagram of the network structure of the integrated convolutional neural network and bidirectional long short-term memory network. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0051] In an exemplary embodiment, Figure 1 As shown, a distribution network system operation optimization method based on a virtual power plant cluster is provided, and the method includes the following S101 to S106. Among them:

[0052] S101, build a distribution network system framework based on virtual power plant cluster; Figure 2As shown, the distribution network system framework includes: a shared energy storage unit that communicates with a virtual power plant cluster, a power grid control center, an electricity trading market, and common user loads; the common user loads are loads that do not participate in macro-control and are related to user habits; the shared energy storage unit of the virtual power plant is used to provide energy storage services to users in the form of users sharing their own energy storage; the form of users sharing their own energy storage is to enable all users to share the energy storage costs and enjoy the energy storage services together; the shared energy storage unit is invested, constructed, operated, and maintained by the shared energy storage unit operator; the virtual power plant cluster includes multiple interconnected virtual power plants; each virtual power plant includes: an aggregation unit and a virtual Power plant system control center; the aggregation unit includes: photovoltaic power generation unit, wind power generation unit and user controllable load; user controllable load includes: user common load controlled by intelligent switch and user intelligent power load; the user common load controlled by intelligent switch is the user submitting power plan to the virtual power plant control center according to its own power demand, and the virtual power plant regulates the current user load according to the user's power plan and the distribution network system regulation demand; the user intelligent power load is the user's power equipment that adjusts the power through the intelligent controller, and when the virtual power plant has a regulation demand, the power of the current load is adjusted; each virtual power plant transmits power and exchanges information through the virtual power plant control center;

[0053] Each VPP is interconnected and independent of each other, that is, N VPPs can realize mutual power supply, that is, when the power is insufficient, they can first take the excess power from the adjacent VPP, or when there is excess power, they can first supply the excess power to other VPPs. And when there is a large power shortage in the system, electricity can be purchased from the shared energy storage unit and the power grid system.

[0054] The power generation data and load data of user loads, power generation units and shared energy storage units are collected through intelligent terminal devices, and the collected data is transmitted to each VPP system control center, which analyzes and processes the power generation data and load data of the distribution network system. In order to realize remote data transmission and analysis of distribution network system data, all data transmission devices and storage devices included in this application collect power generation data and load data to the intranet cloud platform through an intelligent gateway. In this application, the intranet cloud platform terminal only allows intranet data access to ensure user data security and privacy. The intelligent gateway used in this application supports multiple network standards at the same time, such as GPRS / 4G / 5G / 1.8GHz power line wireless private network / 230MHz power wireless private network / optical fiber private network and other network standards. At the same time, this application supports multiple types of communication protocols, such as TCP, DL / T645, DL / T698, CDT, Modbus, UDP and other protocols.

[0055] S102, according to the distribution network system framework, obtain the historical data of distributed generation and the historical data of user load and the corresponding characteristic parameter data; the characteristic parameter data include: historical wind power, light, electricity price, temperature and humidity; the historical data of user load includes both the user's ordinary load data and the user's controllable load data. The historical data of user controllable load is the historical load data generated when the VPP does not regulate the user's controllable load, that is, to obtain the maximum predicted value of the user load, so as to assist the reliability margin analysis of the configuration optimization of the shared energy storage unit of the distribution network system.

[0056] S103, respectively determining the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determining the key parameters corresponding to the distributed generation historical data and the user load historical data;

[0057] In order to improve the accuracy and efficiency of prediction, before conducting distributed generation and user load forecasting, it is necessary to screen the input features, select features that are highly correlated with the target variables, and reduce the number of feature parameters to alleviate the overfitting problem. By deeply mining key features, the generalization ability of the model can be effectively improved, and the prediction accuracy of distributed generation data and user load data can be further improved.

[0058] S103 specifically includes:

[0059] S1, using Pearson correlation analysis, the characteristic parameter data is associated with the historical data of distributed generation and the historical data of user load, and the Pearson correlation coefficient is obtained;

[0060] Specifically, by calculating the covariance and standard deviation of historical data samples and related characteristic parameter samples, the Pearson correlation coefficient between the two can be obtained. In this way, the Pearson correlation coefficient r of distributed generation historical data and related characteristic parameters, user load historical data and related characteristic parameters can be obtained respectively:

[0061]

[0062] In the formula, X i is the i-th data of the data sequence X; is the average value of the data sequence X; n is the length of the data sequence; Y i is the i-th data of data sequence Y; is the average value of the data sequence Y. This formula is used to calculate the correlation coefficient between the distributed generation historical data sequence, the user load historical data sequence and their corresponding characteristic parameter data sequence.

[0063] S2, correlation determined by Pearson correlation coefficient;

[0064] S3, determining key parameters corresponding to the distributed generation historical data and the user load historical data respectively according to the correlation.

[0065] When the absolute value of the Pearson correlation coefficient r is in the intervals of (0, 0.2], (0.2, 0.4], (0.4, 0.6], (0.6, 0.8] and (0.8, 1.0], the correlation between the two variables is extremely weak, weak, medium, strong and extremely strong, respectively. This patent chooses to eliminate the features that show extremely weak, weak and medium correlation with the target variable to strengthen the mining of feature parameters with strong correlation, thereby improving the accuracy and efficiency of the prediction model.

[0066] S104, based on the historical data of distributed power generation and the corresponding key parameters, a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network is used to construct a distributed power generation unit power generation prediction model; based on the historical data of user load and the corresponding key parameters, a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network is used to construct a user load prediction model; the distributed power generation unit power generation prediction model is used to output the distributed power generation unit power generation prediction data to the virtual power plant system control center; the user load prediction model is used to output the user load prediction data to the virtual power plant system control center;

[0067] like Figure 3 As shown in the figure, the network of the integrated convolutional neural network and the bidirectional long short-term memory network uses CNN to mine the correlation between feature vectors, and uses BiLSTM for bidirectional loop training to learn the global information of historical data more fully, and finally obtains the prediction results of the distributed generation unit power generation data and user load data, and transmits them to the VPP system control center;

[0068] In the CNN network, convolutional layers and pooling layers are used alternately. In the convolutional layer, the model can more effectively extract feature association information from the input data through local neuron connections and convolution kernel weight sharing. In the pooling layer, feature association information is extracted locally to reduce feature dimensions and effectively reduce the number of training feature parameters. The BiLSTM model is a combination of forward and reverse propagation bidirectional loop structures. From the perspective of time flow, the BiLSTM network adds data flow from the future to the past on the basis of the unidirectional flow from the past to the future of the LSTM network, so it can more fully mine the temporal characteristics of the data.

[0069] Input key features and historical data, alternately extract features and refine local data through four layers of one-dimensional convolutional layers and pooling layers, and then flatten the data through one layer of flattening. Secondly, input the processed data into the BiLSTM network for training. Finally, the training or prediction results are passed through a fully connected layer (Dense) to enhance the output data features.

[0070] The distributed generation unit generation prediction model and the user load prediction model are both based on the mean square error M SE Cooperate with Adam optimizer for training; and use mean square error M SE and the mean percentage error M APE As an evaluation indicator.

[0071] S105, constructing a shared energy storage unit configuration optimization model based on the distributed generation unit power generation forecast data and the user load forecast data;

[0072] The shared energy storage unit configuration optimization model specifically includes:

[0073] Using the formula Determine the energy rate constraint of the shared energy storage unit; there is an energy rate constraint between the capacity upper limit of the shared energy storage unit and the rated charge and discharge power upper limit;

[0074] Among them, γ is the fixed proportional coefficient of the upper limit of the energy storage power and capacity of the shared energy storage unit, is the maximum charging and discharging power of the shared energy storage unit, is the maximum energy storage capacity of the shared energy storage unit;

[0075] Using the formula Determine the energy storage power and energy storage capacity constraints of the shared energy storage unit;

[0076] in, is the minimum discharge power of the shared energy storage unit, is the maximum discharge power of the shared energy storage unit, is the minimum energy storage capacity of the shared energy storage unit, is the maximum energy storage capacity of the shared energy storage unit, P ESS is the discharge power of the shared energy storage unit, E ESS is the energy storage battery capacity of the shared energy storage unit, It is the energy storage battery capacity of the shared energy storage unit just before the distribution network system reaches the worst operating condition. Predict power shortage for the distribution network system under the worst operating conditions; Forecast power for user loads in the distribution network system, is the predicted power of distributed generation in the distribution network system, η is the capacity margin index of the distribution network system, and the capacity margin index meets the requirements of compensating the prediction deviation of the distribution network system and the maximum predicted value capacity requirements of the user's controllable load when it is not regulated, with the goal of ensuring the reliability of the distribution network power supply, Δt worst It is the duration of the worst operating state of the distribution network system.

[0077] S106, optimizing the operation of the distribution network system of the virtual power plant cluster according to the distributed power generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model.

[0078] S106 specifically includes:

[0079] S61, establishing an optimization objective function with the goal of minimizing the power loss of the distribution network system;

[0080] minP loss =P support -P load.regular -P load.control ;

[0081] S62, optimizing the operation of the distribution network system of the virtual power plant cluster according to the optimization objective function and the operation mode of the distribution network system based on the virtual power plant cluster.

[0082] S62 specifically includes:

[0083] When the operation mode of the distribution network system based on the virtual power plant cluster is the worst operation mode, the upper power grid has stopped supplying power to the distribution network system, that is, the distribution network system can only obtain power from the virtual power plant cluster and shared energy storage units, mainly including wind power generation units, photovoltaic power generation units and energy storage power sources. In the worst operation mode, the distributed power generation units should be fully utilized, that is, the power generation power reaches the maximum value. The control power ΔP of the user-controllable load load.control Also adjusted to the maximum In order to maximize the power supply area of ​​the distribution network system and ensure the power supply reliability to the distribution network users, the optimization objective function is:

[0084] Among them, P ESS is the discharge power of the shared energy storage unit, P loss is the power loss power of the distribution network system; P support is the total power supply of the distribution network system; P load.regular P is the power consumption of the user's normal load; load.control is the power consumption of the user controllable load, P system，support is the power generation, in the worst operation mode P system.support Reaching the maximum value ΔP load.control is the control power of the user's controllable load, ΔP in the worst operation mode load.control Adjust to maximum value

[0085] When the operation mode of the distribution network system based on the virtual power plant cluster is the normal operation mode, the upper power grid participates in the power supply of the distribution network system. In order to prevent the phenomenon of wind and solar abandonment, it is also required that the wind power generation unit and the photovoltaic power generation unit should be fully generated. The electric energy generated by the distributed generation unit supplies power to the user load according to the distribution service demand, and is also used for auxiliary service needs and for the peak and frequency regulation services of the distribution network system. The optimization objective function is to fully absorb the new energy power generation. The optimization objective function is:

[0086]

[0087] Among them, ΔP is the power demand for peak and frequency regulation services of the distribution network system, which is determined by the day-ahead power trading market. system.support , ΔP ESS , ΔP load.control They respectively represent the auxiliary service power of distributed generation units, shared energy storage units, and user-controllable loads participating in the peak and frequency regulation of the distribution network system.

[0088] Based on the same inventive concept, the embodiment of the present application also provides a distribution network system operation optimization device based on a virtual power plant cluster for implementing the distribution network system operation optimization method based on a virtual power plant cluster involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the distribution network system operation optimization device based on a virtual power plant cluster provided below can refer to the limitations of the distribution network system operation optimization method based on a virtual power plant cluster above, and will not be repeated here.

[0089] In an exemplary embodiment, a distribution network system operation optimization device based on a virtual power plant cluster is provided, comprising:

[0090] A framework construction module is used to construct a distribution network system framework based on a virtual power plant cluster; the distribution network system framework includes: a shared energy storage unit that communicates with the virtual power plant cluster, a power grid control center, an electricity trading market, and user common loads; the user common loads are loads that do not participate in macro-control and are related to user habits; the shared energy storage unit of the virtual power plant is used to provide energy storage services to users in the form of users sharing their own energy storage; the form of users sharing their own energy storage is to enable all users to share the energy storage cost and enjoy the energy storage service together; the virtual power plant cluster includes multiple interconnected virtual power plants; each virtual power plant includes: an aggregation unit and a virtual power plant system that transmits power and interacts with the aggregation unit System control center; the aggregation unit includes: photovoltaic power generation unit and wind power generation unit and user controllable load; user controllable load includes: user common load controlled by intelligent switch and user intelligent power load; the user common load controlled by intelligent switch is the user submitting power plan to the virtual power plant control center according to its own power demand, and the virtual power plant regulates the current user load according to the user's power plan and the distribution network system regulation demand; the user intelligent power load is the user's power equipment that adjusts the power through the intelligent controller, and when the virtual power plant has regulation demand, the power of the current load is adjusted; each virtual power plant transmits power and exchanges information through the virtual power plant control center;

[0091] A data acquisition module is used to acquire historical data of distributed generation and user load and corresponding characteristic parameter data according to the distribution network system framework; the characteristic parameter data includes: historical wind power, light, electricity price, temperature and humidity;

[0092] A key parameter determination module, used to determine the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determine the key parameters corresponding to the distributed generation historical data and the user load historical data respectively;

[0093] The prediction model determination module is used to construct a distributed generation unit power generation prediction model based on the distributed generation historical data and the corresponding key parameters, using a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network; the user load prediction model is constructed based on the user load historical data and the corresponding key parameters, using a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network; the distributed generation unit power generation prediction model is used to output the distributed generation unit power generation prediction data to the virtual power plant system control center; the user load prediction model is used to output the user load prediction data to the virtual power plant system control center;

[0094] An optimization model determination module is used to construct a shared energy storage unit configuration optimization model based on the distributed generation unit power generation forecast data and the user load forecast data;

[0095] The operation optimization module is used to optimize the operation of the distribution network system of the virtual power plant cluster based on the distributed generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model.

[0096] In an exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a distribution network system operation optimization method based on a virtual power plant cluster is implemented.

[0097] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0098] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0099] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0100] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0101] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0102] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0103] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A distribution network system operation optimization method based on a virtual power plant cluster, characterized in that: The distribution network system operation optimization method based on the virtual power plant cluster includes: A distribution network system framework based on a virtual power plant cluster is constructed; the distribution network system framework includes: a shared energy storage unit that communicates with the virtual power plant cluster, a power grid control center, an electricity trading market, and common user loads; the common user loads are loads that do not participate in macro-control and are related to user habits; the shared energy storage unit of the virtual power plant is used to provide energy storage services to users in the form of users sharing their own energy storage; the form of users sharing their own energy storage is to enable all users to share the energy storage cost and enjoy the energy storage service together; the virtual power plant cluster includes multiple interconnected virtual power plants; each virtual power plant includes: an aggregation unit and a virtual power plant system control center that transmits power and interacts with the aggregation unit The aggregation unit includes: photovoltaic power generation unit, wind power generation unit and user controllable load; the user controllable load includes: user common load controlled by intelligent switch and user intelligent power load; the user common load controlled by intelligent switch is that the user submits the power plan to the virtual power plant control center according to its own power demand, and the virtual power plant regulates the current user load according to the user's power plan and the distribution network system regulation demand; the user intelligent power load is the user's power equipment that adjusts the power through the intelligent controller, and when the virtual power plant has the regulation demand, the power of the current load is adjusted; each virtual power plant transmits power and exchanges information through the virtual power plant control center; Obtaining distributed generation historical data and user load historical data and corresponding characteristic parameter data according to the distribution network system framework; the characteristic parameter data includes: historical wind power, light, electricity price, temperature and humidity; Respectively determine the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determine the key parameters corresponding to the distributed generation historical data and the user load historical data; According to the historical data of distributed power generation and the corresponding key parameters, a network of comprehensive convolutional neural network and bidirectional long short-term memory network is used to build a distributed power generation unit power generation prediction model; according to the historical data of user load and the corresponding key parameters, a network of comprehensive convolutional neural network and bidirectional long short-term memory network is used to build a user load prediction model; the distributed power generation unit power generation prediction model is used to output the distributed power generation unit power generation prediction data to the virtual power plant system control center; the user load prediction model is used to output the user load prediction data to the virtual power plant system control center; Based on the power generation forecast data of distributed generation units and user load forecast data, a shared energy storage unit configuration optimization model is constructed; The distribution network system operation of the virtual power plant cluster is optimized based on the distributed generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model.

2. The method for optimizing the operation of a distribution network system based on a virtual power plant cluster according to claim 1, characterized in that: The step of respectively determining the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determining the key parameters corresponding to the distributed generation historical data and the user load historical data, specifically includes: Pearson correlation analysis is used to associate the characteristic parameter data with the historical data of distributed generation and the historical data of user load, respectively, and the Pearson correlation coefficient is obtained; Correlations were determined based on Pearson’s correlation coefficient; The key parameters corresponding to the distributed generation historical data and the user load historical data are determined according to the correlation.

3. The method for optimizing the operation of a distribution network system based on a virtual power plant cluster according to claim 1, characterized in that: The distributed generation unit generation prediction model and the user load prediction model are both based on the mean square error M SE Cooperate with Adam optimizer for training; and use mean square error M SE and the mean percentage error M APE As an evaluation indicator.

4. The method for optimizing the operation of a distribution network system based on a virtual power plant cluster according to claim 1, characterized in that: The shared energy storage unit configuration optimization model specifically includes: Using the formula Determine the energy rate constraint of the shared energy storage unit; where γ is the fixed proportional coefficient of the energy storage power upper limit and capacity of the shared energy storage unit, is the maximum charging and discharging power of the shared energy storage unit, is the maximum energy storage capacity of the shared energy storage unit; Using the formula Determine the energy storage power and energy storage capacity constraints of the shared energy storage unit; in, is the minimum discharge power of the shared energy storage unit, is the maximum discharge power of the shared energy storage unit, is the minimum energy storage capacity of the shared energy storage unit, is the maximum energy storage capacity of the shared energy storage unit, P ESS is the discharge power of the shared energy storage unit, E ESS is the energy storage battery capacity of the shared energy storage unit, It is the energy storage battery capacity of the shared energy storage unit just before the distribution network system reaches the worst operating condition. Predict power shortage for the distribution network system under the worst operating conditions; Forecast power for user loads in the distribution network system, is the predicted power of distributed generation in the distribution network system, η is the capacity margin index of the distribution network system, and the capacity margin index meets the requirements of compensating the prediction deviation of the distribution network system and the maximum predicted value capacity requirements of the user's controllable load when it is not regulated, with the goal of ensuring the reliability of the distribution network power supply, Δt worst It is the duration of the worst operating state of the distribution network system.

5. The method for optimizing the operation of a distribution network system based on a virtual power plant cluster according to claim 1, characterized in that: The operation optimization of the distribution network system of the virtual power plant cluster is performed according to the distributed generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model, specifically including: Establish an optimization objective function with the goal of minimizing the power loss of the distribution network system; According to the optimization objective function and the distribution network system operation mode based on the virtual power plant cluster, the distribution network system operation optimization of the virtual power plant cluster is carried out.

6. The method for optimizing the operation of a distribution network system based on a virtual power plant cluster according to claim 5, characterized in that: The optimization of the distribution network operation of the virtual power plant cluster is performed according to the optimization objective function and the distribution network operation mode based on the virtual power plant cluster, specifically including: When the operation mode of the distribution network system based on the virtual power plant cluster is the worst operation mode, the distribution network system can only obtain power from the virtual power plant cluster and the shared energy storage unit, and the optimization objective function is: Among them, P ESS is the discharge power of the shared energy storage unit, P loss is the power loss power of the distribution network system; P support is the total power supply of the distribution network system; P load.regular P is the power consumption of the user's normal load; load.control is the power consumption of the user controllable load, P system.support is the power generation, in the worst operation mode P system.support Reaching the maximum value ΔP load.control is the control power of the user's controllable load, ΔP in the worst operation mode load.control Adjust to maximum value When the operation mode of the distribution network system based on the virtual power plant cluster is the normal operation mode, the electric energy generated by the distributed generation unit supplies power to the user load according to the distribution service demand, and is also used for auxiliary service demand and for peak and frequency regulation services of the distribution network system. The optimization objective function is to fully absorb the new energy power generation, and the optimization objective function is: Among them, ΔP is the power demand for peak and frequency regulation services of the distribution network system, which is determined by the day-ahead power trading market. system.support , ΔP ESS , ΔP load.control They respectively represent the auxiliary service power of distributed generation units, shared energy storage units, and user-controllable loads participating in the peak and frequency regulation of the distribution network system.

7. A distribution network system operation optimization device based on a virtual power plant cluster, characterized in that: The distribution network system operation optimization device based on the virtual power plant cluster includes: A framework construction module is used to construct a distribution network system framework based on a virtual power plant cluster; the distribution network system framework includes: a shared energy storage unit that communicates with the virtual power plant cluster, a power grid control center, an electricity trading market, and user common loads; the user common loads are loads that do not participate in macro-control and are related to user habits; the shared energy storage unit of the virtual power plant is used to provide energy storage services to users in the form of users sharing their own energy storage; the form of users sharing their own energy storage is to enable all users to share the energy storage cost and enjoy the energy storage service together; the virtual power plant cluster includes multiple interconnected virtual power plants; each virtual power plant includes: an aggregation unit and a virtual power plant system that transmits power and interacts with the aggregation unit System control center; the aggregation unit includes: photovoltaic power generation unit and wind power generation unit and user controllable load; user controllable load includes: user common load controlled by intelligent switch and user intelligent power load; the user common load controlled by intelligent switch is the user submitting power plan to the virtual power plant control center according to its own power demand, and the virtual power plant regulates the current user load according to the user's power plan and the distribution network system regulation demand; the user intelligent power load is the user's power equipment that adjusts the power through the intelligent controller, and when the virtual power plant has regulation demand, the power of the current load is adjusted; each virtual power plant transmits power and exchanges information through the virtual power plant control center; A data acquisition module is used to acquire historical data of distributed generation and user load and corresponding characteristic parameter data according to the distribution network system framework; the characteristic parameter data includes: historical wind power, light, electricity price, temperature and humidity; A key parameter determination module, used to determine the correlation between the characteristic parameter data and the distributed generation historical data and the user load historical data, and determine the key parameters corresponding to the distributed generation historical data and the user load historical data respectively; The prediction model determination module is used to construct a distributed generation unit power generation prediction model based on the distributed generation historical data and the corresponding key parameters, using a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network; the user load prediction model is constructed based on the user load historical data and the corresponding key parameters, using a network of a comprehensive convolutional neural network and a bidirectional long short-term memory network; the distributed generation unit power generation prediction model is used to output the distributed generation unit power generation prediction data to the virtual power plant system control center; the user load prediction model is used to output the user load prediction data to the virtual power plant system control center; An optimization model determination module is used to construct a shared energy storage unit configuration optimization model based on the distributed generation unit power generation forecast data and the user load forecast data; The operation optimization module is used to optimize the operation of the distribution network system of the virtual power plant cluster based on the distributed generation unit power generation prediction model, the user load prediction model and the shared energy storage unit configuration optimization model.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the distribution network system operation optimization method based on a virtual power plant cluster as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the distribution network system operation optimization method based on a virtual power plant cluster as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the distribution network system operation optimization method based on a virtual power plant cluster as described in any one of claims 1-6.

Images