Distributed energy grid-connected scheduling optimization management system

Abstract

The invention discloses a distributed energy grid-connected dispatching optimization management system, and particularly relates to the field of energy management and power dispatching. Comprising a real-time monitoring acquisition module, a distributed energy characteristic sensing module, a data processing module, a distributed energy performance prediction module, an intelligent scheduling optimization module, a fault self-diagnosis and self-recovery module and a distributed energy intelligent maintenance management module. According to the distributed energy grid-connected scheduling optimization management system, distributed energy characteristics are extracted through a distributed energy characteristic sensing module, and the scheduling efficiency of an energy storage system is improved; a scheduling result is predicted through the data processing module, so that the potential of distributed energy is utilized to the maximum extent, and automatic and intelligent power grid energy management is realized; the scheduling operation is optimized through the intelligent scheduling optimization module, the scheduling of the power system is optimized, and the energy utilization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management and electric power dispatching, and more specifically, to a distributed energy grid-connected dispatching optimization management system. Background Art

[0002] With the rapid development of renewable energy, especially the large-scale access of distributed energy such as wind power, solar energy, and energy storage systems, the management and dispatching of smart grids and microgrids using traditional power dispatching methods has led to low energy utilization efficiency, serious waste of resources, and unstable grid operation.

[0003] At present, although there are some scheduling optimization methods based on traditional algorithms, and to a certain extent, they have realized the connection between distributed energy and the power grid, and played a positive role in improving the reliability of energy supply and promoting the consumption of renewable energy, by adopting traditional power electronic converters to realize the power conversion of distributed energy, it can meet the basic requirements of grid access; at the same time, a simple monitoring system is used to monitor the operating status of distributed energy, which provides a certain decision-making basis for power grid dispatchers.

[0004] However, there are still some shortcomings in its actual use, such as the lack of an effective response mechanism for dynamic adjustment and intelligent management of distributed energy characteristics; the real-time changes in distributed energy generation power are difficult to accurately predict; and the synergy between distributed energy sources and between distributed energy sources and the entire power grid system cannot be fully considered. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a distributed energy grid-connected dispatch optimization management system, which solves the problems raised in the above-mentioned background technology through the following scheme.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a distributed energy grid-connected dispatch optimization management system, including a system operation database, a system central processing module and a user information terminal, and also including:

[0007] Real-time monitoring and acquisition module: used to collect the first distributed energy in real time by deploying sensors and smart meters in the target power grid, and transmit it to the distributed energy characteristic perception module and the intelligent scheduling optimization module;

[0008] Distributed energy characteristic perception module: used for performing characteristic perception operation on the first distributed energy transmitted in the real-time monitoring and acquisition module, the characteristic perception operation is used for obtaining the first energy characteristic corresponding to the first distributed energy, and transmitting it to the data processing module;

[0009] Data processing module: used to obtain a distributed energy scheduling analysis model, obtain a second distributed energy according to the first energy characteristic transmitted by the distributed energy characteristic perception module through the distributed energy scheduling analysis model, and transmit it to the distributed energy performance prediction module;

[0010] Distributed energy performance prediction module: used to predict the performance of distributed energy through the second distributed energy according to the distributed energy scheduling analysis model obtained by the data processing module, obtain the second energy characteristics corresponding to the second distributed energy, and transmit them to the intelligent scheduling optimization module;

[0011] Intelligent scheduling optimization module: used to perform scheduling optimization operations on the first energy characteristics transmitted by the real-time monitoring and acquisition module and the second energy characteristics transmitted by the distributed energy performance prediction module. The scheduling optimization operation is used to obtain the optimal scheduling strategy corresponding to the second energy characteristics and transmit it to the fault self-diagnosis and self-recovery module and the distributed energy intelligent maintenance management module;

[0012] Fault self-diagnosis and self-recovery module: used to monitor the operating status of the target power grid in real time and diagnose faults through fault monitoring operations. The fault monitoring operation is used to obtain the third distributed energy and transmit it to the data processing module;

[0013] Distributed energy intelligent maintenance management module: Based on the optimal scheduling strategy corresponding to the second energy characteristics transmitted by the intelligent scheduling optimization module, output the intelligent maintenance management report to the user interface;

[0014] The system operation database includes all data texts of the distributed energy grid-connected scheduling and optimization management system, and collects the information texts output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control system, and the user information terminal is an information output device for receiving the distributed energy grid-connected scheduling and optimization management system.

[0015] Preferably, in the real-time monitoring and acquisition module, the first distributed energy includes environmental data, distributed energy output data, and electrical equipment status, wherein the environmental data includes meteorological data, ambient humidity, and ambient temperature; the distributed energy output data includes output charge data of each distributed energy; the electrical equipment status includes voltage, current, and power of each electrical equipment.

[0016] Preferably, the steps of the distributed energy characteristic perception module characteristic perception operation are as follows:

[0017] Key characteristic information collation: preliminary collation of the discrete data sequence of the first distributed energy with time as the horizontal axis and power generation as the vertical axis;

[0018] Processing abnormal data: Check the integrity and accuracy of the data, use adaptive filtering algorithms to process data noise and errors, and use linear interpolation to supplement a small amount of missing data;

[0019] Characteristic analysis: Carry out targeted processing according to the characteristic analysis requirements of different types of distributed energy;

[0020] Feature analysis data transmission and visualization processing: Dynamic visualization technology is used to set the visualization layout and transmit it through the high-speed communication 5G network.

[0021] Preferably, the data processing module obtains the second distributed energy, specifically including:

[0022] Based on the first energy characteristic and the corresponding historical data in the system operation database, the data set of the input model is set, denoted as Ec=Ec1, Ec2, ..., Ec n , where n represents the total number of data information in the data set of the input model, and the Ec corresponding to the data set of the input model i It is represented as a d-dimensional vector, i represents the index of the corresponding data information in the data set of the input model, and d represents the feature dimension of the data set of the input model;

[0023] Set the target label corresponding to the data information in the data set of the input model, denoted as Tec = tec1, tec2, ..., tec n , where n represents the total number of data information in the data set of the input model, and each target label tec i It is a scalar, i represents the index of the corresponding data information in the data set of the input model;

[0024] Initialize the weight matrix W from the input layer to the hidden layer, specifically expressed as: W∈R L×d , where R represents a set of real numbers, L represents the number of elements in the bias vector, and d represents the feature dimension of the data set input to the model;

[0025] Initialize the bias vector b from the input layer to the hidden layer, specifically expressed as: b∈R L , where R represents the set of real numbers and L represents the number of elements in the bias vector.

[0026] Preferably, the data processing module obtains the second distributed energy, specifically including:

[0027] The output layer weight β corresponding to the distributed energy scheduling analysis model is calculated by minimizing the mean square error, which is specifically expressed as:

[0028] β=(H T H) -1 H T H,

[0029] Among them, β∈R L , that is, taking values ​​from the weights of the hidden layer output to the output layer, H is represented as the hidden layer output matrix, and the matrix rows represent the input data samples, and the matrix columns represent the outputs of the hidden layer neurons;

[0030] Based on the corresponding output layer weight β and hidden layer output matrix H of the distributed energy scheduling analysis model, the predicted value AEc of the distributed energy scheduling analysis model is calculated, which is specifically expressed as:

[0031] AEc=f(W×Ec+b)·H·β,

[0032] Among them, Ec represents the data set of the input model, W represents the weight matrix from the input layer to the hidden layer, b represents the bias vector from the input layer to the hidden layer, and f represents the activation function of the corresponding prediction value of the distributed energy scheduling analysis model.

[0033] Preferably, the data processing module obtains the second distributed energy, specifically including:

[0034] Determine the scheduling and optimization goals; initialize particles; update the speed and position of particles: update the speed of particles based on the individual optimal solution and the global optimal solution of particles. The speed update formula of particles is specifically expressed as:

[0035] v j (t+1)=iw·v j (t)+c1·r1·(p j -x j (t))+c2·r2·(g * -x j (t)),

[0036] Among them, v j (t+1) represents the velocity of the jth particle at the t+1th iteration, iw represents the inertia weight, and v j (t) represents the velocity of the jth particle at the tth iteration, c1, c2 represent learning factors, r1, r2 represent random numbers between [0, 1]; p j It is represented as the individual optimal position of the jth particle; g * Represents the global optimal position; x j (t) represents the position of the jth particle at the tth iteration;

[0037] Based on the particle update speed, the particle position update formula is specifically expressed as:

[0038] x j (t+1)=x j (t)+v j (t+1),

[0039] Among them, x i (t+1) represents the position of the jth particle at the t+1th iteration, x i (t) represents the position of the jth particle at the tth iteration, v j (t+1) represents the velocity of the jth particle at the t+1th iteration;

[0040] Set the position boundary processing strategy; determine the iteration termination condition; obtain the optimal solution.

[0041] Preferably, the distributed energy performance prediction module obtains the second energy characteristic corresponding to the second distributed energy, specifically including:

[0042] Based on the predicted value of each distributed energy source in the target power grid within a unit time period and the actual performance value of each distributed energy source in the target power grid within a unit time period, the energy characteristic prediction coefficient is calculated, which is specifically expressed as:

[0043] PC=e pc1 / log pc2 e,

[0044] Among them, pc represents the energy characteristic prediction coefficient, pc1 and pc2 represent the influencing factors of the energy characteristic prediction coefficient, and e represents a constant with a value of 2.718.

[0045] Preferably, the optimal scheduling strategy of the intelligent scheduling optimization module includes a distributed energy generation power allocation strategy, an energy storage device charging and discharging strategy, and a power grid load regulation strategy.

[0046] To achieve the above object, the present invention provides the following technical solution: a distributed energy grid-connected dispatch optimization management method, implementing the above distributed energy grid-connected dispatch optimization management system, comprising:

[0047] S1: Real-time monitoring and collection of distributed energy information: by deploying sensors and smart meters in the target power grid, the first distributed energy is collected in real time;

[0048] S2: sensing distributed energy characteristics: performing a characteristic sensing operation on the first distributed energy, where the characteristic sensing operation is used to obtain a first energy characteristic corresponding to the first distributed energy;

[0049] S3: Data processing and model building: obtaining a distributed energy scheduling analysis model, and obtaining a second distributed energy source through the first energy source characteristics according to the distributed energy scheduling analysis model;

[0050] S4: predicting distributed energy performance: according to the distributed energy scheduling analysis model, predicting the performance of the distributed energy through the second distributed energy, and obtaining the second energy characteristic corresponding to the second distributed energy;

[0051] S5: Optimizing the intelligent scheduling strategy: performing a scheduling optimization operation on the second energy characteristic, where the scheduling optimization operation is used to obtain an optimal scheduling strategy corresponding to the second energy characteristic;

[0052] S6: Fault self-diagnosis and self-recovery: Through fault monitoring operation, the operating status of the target power grid is monitored in real time and the fault is diagnosed. The fault monitoring operation is used to obtain the third distributed energy;

[0053] S7: Distributed energy intelligent maintenance management: Based on the optimal scheduling strategy corresponding to the second energy characteristics, an intelligent maintenance management report is output to the user interface.

[0054] Technical effects and advantages of the present invention:

[0055] 1. The present invention extracts distributed energy characteristics through a distributed energy characteristic perception module to improve the scheduling efficiency of the energy storage system;

[0056] 2. The present invention predicts the dispatching results through the data processing module to maximize the potential of distributed energy and realize automatic and intelligent power grid energy management;

[0057] 3. The present invention optimizes the dispatching operation through the intelligent dispatching optimization module, improves the dispatching of the power system and improves the energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a system flow chart of the present invention.

[0059] Figure 2 This is a distributed energy scheduling analysis flow chart corresponding to the data processing module of the present invention.

[0060] Figure 3 This is a particle swarm optimization flow chart corresponding to the distributed energy scheduling analysis of the data processing module in the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0063] In the following, the terms "first", "second", and "third" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", and "third" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0064] As attached Figure 1 The distributed energy grid-connected dispatching optimization management system shown includes a system operation database, a system central processing module and a user information terminal, and also includes a real-time monitoring and acquisition module, a distributed energy characteristic perception module, a data processing module, a distributed energy performance prediction module, an intelligent dispatching optimization module, a fault self-diagnosis and self-recovery module, and a distributed energy intelligent maintenance management module.

[0065] The system operation database includes all data texts of the distributed energy grid-connected scheduling and optimization management system, and collects the information texts output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control system, and the user information terminal is an information output device for receiving the distributed energy grid-connected scheduling and optimization management system.

[0066] The real-time monitoring and acquisition module is used to collect the first distributed energy in real time by deploying sensors and smart meters in the target power grid, and transmit it to the distributed energy characteristic perception module and the intelligent scheduling optimization module.

[0067] Specifically, the distributed energy grid-connected dispatching optimization management system deploys corresponding sensors and smart meters based on the key locations of the target power grid. The key locations of the target power grid include but are not limited to generators, energy storage equipment, substations, distribution networks, etc.; monitoring and collection operations are performed through corresponding sensors and smart meters. The types of sensors include but are not limited to anemometers, wind vanes, voltage sensors, current sensors, power sensors, etc., to obtain the first distributed energy. The first distributed energy includes environmental data, distributed energy output data, and electrical equipment status, wherein the distributed energy output data includes the output charge data of each distributed energy; the electrical equipment status includes the voltage, current, and power of each electrical equipment; the environmental data includes meteorological data, ambient humidity, and ambient temperature.

[0068] In this embodiment, an anemometer with an accuracy of ±0.5m / s and a wind vane with an accuracy of ±5° are installed in the wind turbine cabin to measure wind speed and direction in real time; a voltage sensor with a measurement accuracy of ±0.2%, a current sensor with a measurement accuracy of ±0.2%, and a power sensor with a measurement accuracy of ±0.5% are installed at the output end of the generator; and a power sensor with a measurement accuracy of ±5W / m 2 The instrument is equipped with a radiometer and a temperature sensor with an accuracy of ±0.5°C.

[0069] The distributed energy characteristic perception module is used to perform characteristic perception operations on the first distributed energy transmitted in the real-time monitoring and acquisition module. The characteristic perception operation is used to obtain the first energy characteristic corresponding to the first distributed energy and transmit it to the data processing module.

[0070] Specifically, after obtaining the first distributed energy transmitted in the real-time monitoring and acquisition module, the distributed energy grid-connected dispatching optimization management system can obtain multiple key characteristic information corresponding to the first distributed energy, namely, the first energy characteristics, through the characteristic perception operation; the steps of the characteristic perception operation are as follows: key characteristic information sorting: preliminary sorting of the discrete data sequence of the first distributed energy with time as the horizontal axis and power generation as the vertical axis; abnormal data processing: checking the integrity and accuracy of the data, using an adaptive filtering algorithm to process the noise and error of the data, and using linear interpolation to supplement a small amount of missing data; characteristic analysis: targeted processing according to the characteristic analysis requirements of different types of distributed energy; it should be noted that the moving average method is used to analyze the time series change trend of the power generation, the time window size is set to 6 sampling periods, and the increase or decrease of the power is determined by calculating the first-order difference of the power sequence, so as to smooth the power sequence to eliminate the impact of short-term fluctuations, and then determine the rising and falling trend characteristics of the distributed energy power generation, as well as the frequency and amplitude of the trend change; characteristic analysis data transmission and visualization processing: dynamic visualization technology is used to set the visualization layout of the multiple key characteristic information obtained by the analysis according to the data type and display requirements, and transmit it through the high-speed communication 5G network.

[0071] The data processing module is used to obtain a distributed energy scheduling analysis model, obtain a second distributed energy according to the first energy characteristics transmitted by the distributed energy characteristics perception module through the distributed energy scheduling analysis model, and transmit it to the distributed energy performance prediction module.

[0072] Specifically, the distributed energy scheduling and analysis model is a pre-constructed learning model. By inputting the first energy characteristics into the distributed energy scheduling and analysis model, the distributed energy scheduling and analysis model obtains the second distributed energy according to the first energy characteristics. The second distributed energy includes the storage data corresponding to the first energy characteristics, the load prediction results output by the distributed energy scheduling and analysis model, and the power generation prediction results output by the distributed energy scheduling and analysis model.

[0073] In a possible implementation, obtaining the second distributed energy includes: storing the first energy characteristic by using a big data processing technology, and analyzing the corresponding historical data based on the system operation database to obtain a distributed energy scheduling analysis model;

[0074] It should be noted that the steps for obtaining the distributed energy scheduling analysis model are as follows:

[0075] A1. Normalize the grid operation data, distributed energy generation data, and meteorological data corresponding to the first energy characteristic, and store them using the Hadoop distributed database;

[0076] A2. Based on the corresponding historical data in the system operation database, a distributed energy scheduling analysis model is constructed through machine learning and deep learning algorithms. The distributed energy scheduling analysis model is used to perform load forecasting and power generation forecasting on the target power grid;

[0077] In this embodiment, the distributed energy scheduling analysis model adopts the ELM deep learning model; based on historical load data, weather data, and time characteristics, the load demand of the target power grid in the future unit time period is predicted; based on historical power generation data and meteorological forecast data, the power generation demand of each distributed energy is predicted;

[0078] It should be noted that the distributed energy dispatch analysis model generated by ELM deep learning model training includes:

[0079] Based on the first energy characteristic and the corresponding historical data in the system operation database, the data set of the input model is set, denoted as Ec=Ec1, Ec2, ..., Ec n , where i represents the index of the corresponding data information in the data set of the input model, n represents the total number of data information in the data set of the input model, and the corresponding Ec i It is represented as a d-dimensional vector, where d represents the feature dimension of the data set of the input model; the target label corresponding to the corresponding data information in the data set of the input model is set, denoted as Tec = tec1, tec2, ..., tec n , where i represents the index of the corresponding data information in the data set of the input model, n represents the total number of data information in the data set of the input model, and each target label tec i is a scalar; randomly initialize the weight W and bias b corresponding to the input layer to the hidden layer in the distributed energy scheduling analysis model. The input layer and hidden layer are represented by the hierarchical structure names corresponding to the machine learning and deep learning algorithms in the distributed energy scheduling analysis model; initialize the weight matrix W from the input layer to the hidden layer, specifically expressed as: W∈R L×d , where R represents a set of real numbers, L represents the number of neurons in the hidden layer, and d is the feature dimension of the data set input to the model; initialize the bias vector b from the input layer to the hidden layer, specifically expressed as: b∈R L , where R represents a set of real numbers and L represents the number of elements in the bias vector;

[0080] The output layer weight β corresponding to the distributed energy scheduling analysis model is calculated by minimizing the mean square error, which is specifically expressed as:

[0081] β=(H T H) -1 H T H,

[0082] Among them, β∈R L, that is, taking values ​​from the weights of the hidden layer output to the output layer, the output layer is represented by the hierarchical structure name corresponding to the machine learning and deep learning algorithms in the distributed energy scheduling analysis model, H is represented by the hidden layer output matrix, and the matrix rows represent the input data samples, and the matrix columns represent the outputs of the hidden layer neurons;

[0083] Based on the corresponding output layer weight β and hidden layer output matrix H of the distributed energy scheduling analysis model, the predicted value AEc of the distributed energy scheduling analysis model is calculated, which is specifically expressed as:

[0084] AEc=f(W×Ec+b)·H·β,

[0085] Among them, Ec represents the data set of the input model, W represents the weight matrix from the input layer to the hidden layer, b represents the bias vector from the input layer to the hidden layer, and f represents the activation function of the corresponding prediction value of the distributed energy scheduling analysis model;

[0086] In this embodiment, the activation function f of the distributed energy scheduling analysis model corresponding to the predicted value uses the Leaky-ReLU function;

[0087] A3. Use multi-objective optimization algorithm particle swarm optimization distributed energy scheduling analysis model;

[0088] Determine dispatching and optimization goals: Set dispatching and optimization goals, including but not limited to grid load balancing, maximizing distributed energy utilization, efficient charging and discharging of energy storage systems, etc., and optimize the distributed energy dispatching analysis model based on historical prediction results;

[0089] Initialize the particle swarm: In the storage space corresponding to the distributed energy scheduling analysis model, randomly determine the initial position and speed of the particle swarm, denoted as x j (0)∈R lo and v j (0)∈R lo , where R represents a real number set, lo represents the latitude of the storage space corresponding to the distributed energy scheduling analysis model, and j represents the index of the particle in the particle swarm; it should be noted that each particle in the particle swarm represents a potential scheduling solution, and the particle with the best fitness evaluated by the current position of each particle is recorded as the global optimal solution g * , the best fitness position of each particle in history is recorded as the individual optimal solution p i ;The latitude of the storage space is related to the number of variables corresponding to the distributed energy scheduling analysis model;

[0090] Update the speed and position of particles: Update the speed of particles based on the individual optimal solution and the global optimal solution of particles. The speed update formula of particles is specifically expressed as:

[0091] v j(t+1)=iw·v j (t)+c1·r1·(p j -x j (t))+c2·r2·(g * -x j (t)),

[0092] Among them, v j (t+1) represents the velocity of the jth particle at the t+1th iteration, iw represents the inertia weight, and v j (t) represents the velocity of the jth particle at the tth iteration, c1, c2 represent learning factors, r1, r2 represent random numbers between [0, 1]; p j It is represented as the individual optimal position of the jth particle; g * Represents the global optimal position; x j (t) represents the position of the jth particle at the tth iteration;

[0093] It should be noted that the learning factor is the ability of the control particle to learn its own historical optimal solution and the global optimal solution;

[0094] Based on the particle update speed, the particle position update formula is specifically expressed as:

[0095] x j (t+1)=x j (t)+v j (t+1),

[0096] Among them, x i (t+1) represents the position of the jth particle at the t+1th iteration, x i (t) represents the position of the jth particle at the tth iteration, v j (t+1) represents the velocity of the jth particle at the t+1th iteration;

[0097] Set the position boundary processing strategy: When a particle crosses the boundary of the scheduling and optimization target during the position update process, that is, exceeds the allowed storage space range, the particle that crosses the boundary will be bounced back to the boundary according to specific rules so that it can continue to participate in the optimization process in the legal search space; or the particle value can be directly trimmed to the boundary value to ensure that the particle position is always within the allowed search space range, or the particle that crosses the boundary will be transferred to a relative position so that it can continue to move in the simulated periodic space;

[0098] Determine the iterative termination condition: The iterative process continues until any of the following termination conditions is met: when the number of iterations reaches a preset upper limit; when the improvement of the fitness value is less than a certain set threshold; when the duration of the model optimization process exceeds a predetermined time;

[0099] Obtain the optimal solution: By iterating the above steps, output the global optimal solution g * The corresponding particle position is the optimal solution.

[0100] The distributed energy performance prediction module is used to predict the performance of the distributed energy through the second distributed energy according to the distributed energy scheduling analysis model obtained by the data processing module, obtain the second energy characteristics corresponding to the second distributed energy, and transmit them to the intelligent scheduling optimization module.

[0101] Specifically, based on the corresponding rules of input data and historical data, a distributed energy scheduling analysis model is started to perform prediction calculations to obtain the predicted performance of each distributed energy in the target power grid within a unit time period, and the actual performance of each distributed energy in the target power grid within a unit time period is monitored by the distributed energy grid-connected scheduling optimization management system to analyze and obtain the second energy characteristic corresponding to the second distributed energy, that is, the energy characteristic prediction coefficient;

[0102] In a possible implementation, obtaining the second energy characteristic corresponding to the second distributed energy includes: calculating an energy characteristic prediction coefficient based on a predicted value of each distributed energy in the target power grid within a unit time period and an actual performance value of each distributed energy in the target power grid within a unit time period, which is specifically expressed as:

[0103] PC=e pc1 / log pc2 e,

[0104] Among them, pc represents the energy characteristic prediction coefficient, pc1 and pc2 represent the influencing factors of the energy characteristic prediction coefficient, and e represents a constant with a value of 2.718;

[0105] It should be noted that the energy characteristic prediction coefficient is the predicted performance value of the distributed energy scheduling analysis model; the calculation formula of the influencing factor pc1 of the energy characteristic prediction coefficient is specifically expressed as:

[0106]

[0107] Where N represents the total number of prediction results, i represents the index of the corresponding data information in the data set of the input model, Represented as the prediction result corresponding to the i-th data, Represented as the actual result corresponding to the i-th data;

[0108] The calculation formula of the influencing factor pc2 of the energy characteristic prediction coefficient is specifically expressed as:

[0109]

[0110] Where N represents the total number of prediction results, i represents the index of the corresponding data information in the data set of the input model, and AEc i Represented as the prediction result corresponding to the i-th data, Represents the actual result corresponding to the i-th data, Expressed as an average of actual results.

[0111] The intelligent scheduling optimization module is used to perform scheduling optimization operations on the first energy characteristics transmitted by the real-time monitoring and acquisition module and the second energy characteristics transmitted by the distributed energy performance prediction module. The scheduling optimization operation is used to obtain the optimal scheduling strategy corresponding to the second energy characteristics and transmit it to the fault self-diagnosis and self-recovery module and the distributed energy intelligent maintenance management module.

[0112] Specifically, based on load forecasts, distributed energy generation forecasts and energy storage status, the optimal scheduling strategy is dynamically formulated. The optimal scheduling strategy includes a distributed energy generation power allocation strategy, an energy storage device charging and discharging strategy, and a power grid load regulation strategy. The scheduling strategy is applied to the actual target power grid operation to control the generator sets, energy storage devices, and load regulation. The scheduling strategy is executed by adjusting the power grid load distribution in real time to control the working status of distributed energy and energy storage devices. The operating status of the target power grid is checked through real-time monitoring equipment, and feedback information is obtained. The feedback information is the actual operating data of the target power grid, including but not limited to voltage, current, power, etc.

[0113] In this embodiment, the distributed energy generation power allocation strategy determines the generation power allocation of each distributed energy according to the first energy characteristics and the second energy characteristics of the distributed energy. For distributed energy with higher stability, the allocated generation power is used to bear the basic load part; for wind energy equipment with strong intermittency, the generation power is dynamically allocated in the target power grid according to the current wind speed and predicted changes.

[0114] Specifically, the system reduces the volatility of the power grid through the charging and discharging strategy of energy storage equipment, avoids excessive load on the power grid during peak hours, and improves the absorption capacity of distributed energy;

[0115] In this embodiment, during the late night when the grid load is low, the system controls the charging of the energy storage device to increase the energy storage system's energy storage; during the day when the grid load is peak, the system releases the stored electricity to reduce the peak load and balance the grid load; and monitors the charging and discharging status of the energy storage system in real time, obtains the status data of the energy storage system in real time, and dynamically adjusts the charging and discharging strategy.

[0116] The fault self-diagnosis and self-recovery module is used to monitor the operating status of the target power grid in real time and diagnose faults through fault monitoring operations. The fault monitoring operation is used to obtain the third distributed energy and transmit it to the data processing module.

[0117] Specifically, the distributed energy grid-connected dispatching optimization management system monitors the status of target grid equipment and systems in real time, detects faults and issues alarms; uses intelligent sensors and machine learning algorithms to monitor the status of grid equipment in real time, identifies potential fault risks, uses machine learning algorithms to analyze the possibility of equipment failure, and performs accurate diagnosis when a fault occurs; when a fault area occurs, the system activates a self-recovery mechanism, transfers power to other parts by adjusting the grid dispatching and control strategies, and ensures the normal operation of the grid; and after the fault is restored and stabilized, collects information about the corresponding equipment in the fault area and transmits it to the data processing module for analysis.

[0118] It should be noted that the third distributed energy is obtained through fault monitoring operations, including the real-time operating status of the target power grid and information about related equipment in the fault area.

[0119] The distributed energy intelligent maintenance management module outputs an intelligent maintenance management report to the user interface based on the optimal scheduling strategy corresponding to the second energy characteristic transmitted by the intelligent scheduling optimization module.

[0120] Specifically, based on the optimal scheduling strategy corresponding to the second energy characteristic transmitted by the intelligent scheduling optimization module, and combined with the maintenance manual and expert experience of the distributed energy equipment, an intelligent maintenance management report is generated. The intelligent maintenance management report includes an overall operating status evaluation of the distributed energy system, an operating status analysis of each distributed energy equipment, potential fault risk warnings, maintenance recommendations, and maintenance plan arrangements; the generated intelligent maintenance management report is output to the user interface in a predetermined format and layout.

[0121] In this embodiment, through the dynamic visualization technology in the distributed energy grid-connected dispatching optimization management system, the overall operation status of the power grid, load fluctuations, distributed energy power generation, energy storage equipment status, and meteorological changes are displayed in different charts respectively; for the operation status of the power grid, real-time curves are used to display the changing trends of voltage, current, and power, and the time axis resolution is set to seconds.

[0122] As attached Figure 2 The distributed energy grid-connected dispatch optimization management method shown includes: S1: real-time monitoring and collection, S2: distributed energy characteristics perception, S3: data processing, S4: distributed energy performance prediction, S5: intelligent dispatch optimization, S6: fault self-diagnosis and self-recovery, and S7: distributed energy intelligent maintenance management.

[0123] S1: Real-time monitoring and collection of distributed energy information: by deploying sensors and smart meters in the target power grid, the first distributed energy is collected in real time;

[0124] S2: sensing distributed energy characteristics: performing a characteristic sensing operation on the first distributed energy, where the characteristic sensing operation is used to obtain a first energy characteristic corresponding to the first distributed energy;

[0125] S3: Data processing and model building: obtaining a distributed energy scheduling analysis model, and obtaining a second distributed energy source through the first energy source characteristics according to the distributed energy scheduling analysis model;

[0126] S4: predicting distributed energy performance: according to the distributed energy scheduling analysis model, predicting the performance of the distributed energy through the second distributed energy, and obtaining the second energy characteristic corresponding to the second distributed energy;

[0127] S5: Optimizing the intelligent scheduling strategy: performing a scheduling optimization operation on the second energy characteristic, where the scheduling optimization operation is used to obtain an optimal scheduling strategy corresponding to the second energy characteristic;

[0128] S6: Fault self-diagnosis and self-recovery: Through fault monitoring operation, the operating status of the target power grid is monitored in real time and the fault is diagnosed. The fault monitoring operation is used to obtain the third distributed energy;

[0129] S7: Distributed energy intelligent maintenance management: Based on the optimal scheduling strategy corresponding to the second energy characteristics, an intelligent maintenance management report is output to the user interface.

[0130] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0131] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Distributed energy grid-connected dispatch optimization management system, including system operation database, system central processing module and user information terminal, characterized in that: Also includes: Real-time monitoring and acquisition module: used to collect the first distributed energy in real time by deploying sensors and smart meters in the target power grid, and transmit it to the distributed energy characteristic perception module and the intelligent scheduling optimization module; Distributed energy characteristic perception module: used for performing characteristic perception operation on the first distributed energy transmitted in the real-time monitoring and acquisition module, the characteristic perception operation is used for obtaining the first energy characteristic corresponding to the first distributed energy, and transmitting it to the data processing module; Data processing module: used to obtain a distributed energy scheduling analysis model, obtain a second distributed energy according to the first energy characteristic transmitted by the distributed energy characteristic perception module through the distributed energy scheduling analysis model, and transmit it to the distributed energy performance prediction module; Distributed energy performance prediction module: used to predict the performance of distributed energy through the second distributed energy according to the distributed energy scheduling analysis model obtained by the data processing module, obtain the second energy characteristics corresponding to the second distributed energy, and transmit them to the intelligent scheduling optimization module; Intelligent scheduling optimization module: used to perform scheduling optimization operations on the first energy characteristics transmitted by the real-time monitoring and acquisition module and the second energy characteristics transmitted by the distributed energy performance prediction module. The scheduling optimization operation is used to obtain the optimal scheduling strategy corresponding to the second energy characteristics and transmit it to the fault self-diagnosis and self-recovery module and the distributed energy intelligent maintenance management module; Fault self-diagnosis and self-recovery module: used to monitor the operating status of the target power grid in real time and diagnose faults through fault monitoring operations. The fault monitoring operation is used to obtain the third distributed energy and transmit it to the data processing module; Distributed energy intelligent maintenance management module: Based on the optimal scheduling strategy corresponding to the second energy characteristics transmitted by the intelligent scheduling optimization module, output the intelligent maintenance management report to the user interface; The system operation database includes all data texts of the distributed energy grid-connected scheduling and optimization management system, and collects the information texts output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control system, and the user information terminal is an information output device for receiving the distributed energy grid-connected scheduling and optimization management system.

2. The distributed energy grid-connected dispatch optimization management system according to claim 1 is characterized by: The real-time monitoring and acquisition module, the first distributed energy includes environmental data, distributed energy output data, and electrical equipment status, wherein the environmental data includes meteorological data, ambient humidity, and ambient temperature; the distributed energy output data includes the output charge data of each distributed energy; the electrical equipment status includes the voltage, current, and power of each electrical equipment.

3. The distributed energy grid-connected dispatch optimization management system according to claim 1 is characterized by: The steps of the distributed energy characteristic perception module characteristic perception operation are as follows: Key characteristic information collation: preliminary collation of the discrete data sequence of the first distributed energy with time as the horizontal axis and power generation as the vertical axis; Processing abnormal data: Check the integrity and accuracy of the data, use adaptive filtering algorithms to process data noise and errors, and use linear interpolation to supplement a small amount of missing data; Characteristic analysis: Carry out targeted processing according to the characteristic analysis requirements of different types of distributed energy; Feature analysis data transmission and visualization processing: Dynamic visualization technology is used to set the visualization layout and transmit it through the high-speed communication 5G network.

4. The distributed energy grid-connected dispatch optimization management system according to claim 1 is characterized by: The data processing module obtains the second distributed energy, specifically including: Based on the first energy characteristic and the corresponding historical data in the system operation database, the data set of the input model is set, denoted as Ec=Ec1, Ec2, ..., Ec n , where n represents the total number of data information in the data set of the input model, and the Ec corresponding to the data set of the input model i It is represented as a d-dimensional vector, i represents the index of the corresponding data information in the data set of the input model, and d represents the feature dimension of the data set of the input model; Set the target label corresponding to the data information in the data set of the input model, denoted as Tec = tec1, tec2, ..., tec n , where n represents the total number of data information in the data set of the input model, and each target label tec i It is a scalar, i represents the index of the corresponding data information in the data set of the input model; Initialize the weight matrix W from the input layer to the hidden layer, specifically expressed as: W∈R L×d , where R represents a set of real numbers, L represents the number of elements in the bias vector, and d represents the feature dimension of the data set input to the model; Initialize the bias vector b from the input layer to the hidden layer, specifically expressed as: b∈R L , where R represents the set of real numbers and L represents the number of elements in the bias vector.

5. The distributed energy grid-connected dispatch optimization management system according to claim 4 is characterized by: The data processing module obtains the second distributed energy, specifically including: The output layer weight β corresponding to the distributed energy scheduling analysis model is calculated by minimizing the mean square error, which is specifically expressed as: β=(H T H) -1 H T H, Among them, β∈R L , that is, taking values ​​from the weights of the hidden layer output to the output layer, H is represented as the hidden layer output matrix, and the matrix rows represent the input data samples, and the matrix columns represent the outputs of the hidden layer neurons; Based on the corresponding output layer weight β and hidden layer output matrix H of the distributed energy scheduling analysis model, the predicted value AEc of the distributed energy scheduling analysis model is calculated, which is specifically expressed as: AEc=f(W×Ec+b)·H·β, Among them, Ec represents the data set of the input model, W represents the weight matrix from the input layer to the hidden layer, b represents the bias vector from the input layer to the hidden layer, and f represents the activation function of the corresponding prediction value of the distributed energy scheduling analysis model.

6. The distributed energy grid-connected dispatch optimization management system according to claim 4 is characterized by: The data processing module obtains the second distributed energy, specifically including: Determine the scheduling and optimization goals; initialize particles; update the speed and position of particles: update the speed of particles based on the individual optimal solution and the global optimal solution of particles. The speed update formula of particles is specifically expressed as: v j (t+1)=iw·v j (t)+c1·r1·(p j -x j (t))+c2·r2·(g * -x j (t)), Among them, v j (t+1) represents the velocity of the jth particle at the t+1th iteration, iw represents the inertia weight, and v j (t) represents the velocity of the jth particle at the tth iteration, c1, c2 represent learning factors, r1, r2 represent random numbers between [0, 1]; p j It is represented as the individual optimal position of the jth particle; g * Represents the global optimal position; x j (t) represents the position of the jth particle at the tth iteration; Based on the particle update speed, the particle position update formula is specifically expressed as: x j (t+1)=x j (t)+v j (t+1), Among them, x i (t+1) represents the position of the jth particle at the t+1th iteration, x i (t) represents the position of the jth particle at the tth iteration, v j (t+1) represents the velocity of the jth particle at the t+1th iteration; Set the position boundary processing strategy; determine the iteration termination condition; obtain the optimal solution.

7. The distributed energy grid-connected dispatch optimization management system according to claim 1 is characterized by: The distributed energy performance prediction module obtains the second energy characteristic corresponding to the second distributed energy, specifically including: Based on the predicted value of each distributed energy source in the target power grid within a unit time period and the actual performance value of each distributed energy source in the target power grid within a unit time period, the energy characteristic prediction coefficient is calculated, which is specifically expressed as: PC=and pc1 / log pc2 And, Among them, pc represents the energy characteristic prediction coefficient, pc1 and pc2 represent the influencing factors of the energy characteristic prediction coefficient, and e represents a constant with a value of 2.

718.

8. The distributed energy grid-connected dispatch optimization management system according to claim 1 is characterized by: The optimal scheduling strategy of the intelligent scheduling optimization module includes a distributed energy generation power allocation strategy, an energy storage device charging and discharging strategy, and a power grid load regulation strategy.

Images