A multi-energy complementary industrial park microgrid autonomous control method and system

Abstract

The present invention provides a multi-energy complementary industrial park microgrid autonomous control method and system, which relates to the field of autonomous control technology, including: respectively collecting and obtaining target distributed power sources and target distributed memories in the target industrial park; randomly extracting the first grid; obtaining the first power source set and the first memory set; calling the first historical load database, combining with the predetermined period analysis to obtain the first period load state; analyzing under the first load peak period in the first period load state, generating a first power source priority scheduling sequence; extracting the first memory, combining with the predetermined control fitness function to perform control optimization, obtaining the first optimal scheduling control decision; and microgrid control of the first grid. The present invention solves the technical problems that the traditional method cannot effectively integrate the diversified energy in the industrial park, and cannot meet the dynamically changing load demand in the industrial park, resulting in low energy utilization efficiency and poor system stability.

Description

Technical Field

[0001] The present invention relates to the field of autonomous control technology, and in particular to a multi-energy complementary industrial park microgrid autonomous control method and system. Background Art

[0002] Industrial parks usually involve different types of energy, including traditional grid energy, distributed energy such as wind power and solar energy, as well as energy storage equipment. How to effectively integrate these diverse energy sources and achieve complementary and coordinated operation is one of the technical challenges faced by industrial park microgrids. At the same time, the energy consumption patterns of industrial parks usually show a high degree of time-variability, including working days and weekends, peak periods and trough periods, etc. How to achieve real-time scheduling of energy under dynamically changing load demands and meet actual needs is a problem that needs to be solved under existing technologies.

[0003] Therefore, a new method is needed to effectively solve the above problems and improve the energy utilization efficiency, stability and economy of industrial park microgrids. Summary of the invention

[0004] This application provides a multi-energy complementary industrial park microgrid autonomous control method, aiming to solve the technical problems that traditional methods cannot effectively integrate the diverse energy sources in the industrial park and cannot meet the dynamically changing load demands in the industrial park, resulting in low energy utilization efficiency and poor system stability.

[0005] In view of the above problems, the present application provides a multi-energy complementary industrial park microgrid autonomous control method and system.

[0006] The first aspect disclosed in the present application provides a method for autonomous control of a multi-energy complementary industrial park microgrid, the method comprising: respectively acquiring a target distributed power source and a target distributed memory in a target industrial park, the target distributed power source comprising a plurality of power sources with position identifiers, and the target distributed memory comprising a plurality of memories with position identifiers; randomly extracting a first grid in a target grid set obtained by grid segmentation of the target industrial park, the first grid having a first grid position identifier; based on the first grid position identifier, sequentially traversing the plurality of power sources with position identifiers to obtain a first power source set and the plurality of memories with position identifiers to obtain a first memory set; calling the first grid of the first grid; A historical load database is obtained, and a first cycle load state is obtained in combination with a predetermined cycle analysis; during a first load peak period in the first cycle load state, the first memory set is analyzed by a first edge computing center to generate a first power priority scheduling sequence for a first power source, wherein the first power source is any one of the first power sources in the first power source set, and the first edge computing center is an inherent computing center of the first power source; the first memory in the first power priority scheduling sequence is extracted, and the first memory is controlled and optimized in combination with a predetermined control fitness function to obtain a first optimal scheduling control decision; and the first grid in the target industrial park is microgrid controlled according to the first optimal scheduling control decision.

[0007] Another aspect disclosed in the present application provides an autonomous control system for a multi-energy complementary industrial park microgrid, the system being used for the above method, the system comprising: a distributed memory acquisition module, the distributed memory acquisition module being used to respectively acquire a target distributed power source and a target distributed memory in a target industrial park, the target distributed power source comprising a plurality of power sources with position identifiers, the target distributed memory comprising a plurality of memories with position identifiers; a first grid acquisition module, the first grid acquisition module being used to randomly extract a first grid in a target grid set obtained by grid segmentation of the target industrial park, the first grid having a first grid position identifier; a first memory set acquisition module, the first memory acquisition module being used to sequentially traverse the plurality of power sources with position identifiers based on the first grid position identifier to obtain a first power source set and the plurality of memories with position identifiers to obtain a first memory set; a load state acquisition module, The load state acquisition module is used to retrieve the first historical load database of the first grid, and obtain the first cycle load state in combination with the predetermined cycle analysis; the first memory set analysis module is used to analyze the first memory set through the first edge computing center during the first load peak period in the first cycle load state to generate a first power supply priority scheduling sequence for the first power supply, the first power supply is any one of the first power supply sets, and the first edge computing center is the inherent computing center of the first power supply; the control optimization module is used to extract the first memory in the first power supply priority scheduling sequence, and perform control optimization on the first memory in combination with a predetermined control fitness function to obtain a first optimal scheduling control decision; the microgrid control module is used to perform microgrid control on the first grid in the target industrial park according to the first optimal scheduling control decision.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By collecting the distributed power supply and storage information in the target industrial park, as well as grid segmentation and location identification, the collection and coordinated scheduling of multiple energy sources are realized, and the energy utilization efficiency of the entire microgrid system is optimized; by randomly extracting grids, retrieving historical load databases and real-time monitoring, accurate energy scheduling decisions can be made based on the real-time data to meet the real-time needs of the industrial park; through edge computing center analysis and optimization, priority scheduling sequences are extracted, and predetermined control fitness functions are applied to optimize the scheduling decisions of the storage, ensuring that the demand can be met during peak hours while considering the comprehensive impact of historical performance to the greatest extent; according to the optimal scheduling control decision, the grid in the target industrial park is controlled by the microgrid, the operation of distributed power supply and storage is effectively coordinated, and the reliability and stability of the entire microgrid system are improved. In general, this method solves a series of problems existing in the existing technology by introducing key features such as multi-energy complementarity, real-time, and optimization, so that the industrial park microgrid can be more intelligently and efficiently controlled autonomously, which helps to improve energy utilization efficiency and reduce energy costs, while ensuring that the system can operate stably and reliably under dynamically changing needs.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic flow chart of a multi-energy complementary industrial park microgrid autonomous control method is provided for the embodiment of the present application;

[0012] Figure 2 A schematic diagram of the structure of a multi-energy complementary industrial park microgrid autonomous control system is provided for an embodiment of the present application.

[0013] Explanation of the reference numerals: distributed memory acquisition module 10 , first grid acquisition module 20 , first memory set acquisition module 30 , load state acquisition module 40 , first memory set analysis module 50 , control optimization module 60 , microgrid control module 70 . DETAILED DESCRIPTION

[0014] The embodiment of the present application provides an autonomous control method for a multi-energy complementary industrial park microgrid, thereby solving the technical problems that traditional methods cannot effectively integrate the diverse energy sources in the industrial park and cannot meet the dynamically changing load demands in the industrial park, resulting in low energy utilization efficiency and poor system stability.

[0015] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0016] Embodiment 1

[0017] like Figure 1 As shown, the embodiment of the present application provides a multi-energy complementary industrial park microgrid autonomous control method, the method comprising:

[0018] The target distributed power source and the target distributed storage in the target industrial park are respectively acquired, wherein the target distributed power source includes a plurality of power sources with location identifiers, and the target distributed storage includes a plurality of storages with location identifiers;

[0019] Get the target industrial park to be controlled, which is an area including one or more industrial facilities, power supplies and storage. For the target distributed power supply, each power supply has a location identifier, which can be a unique identifier such as geographic coordinates, area number, etc., which is used to accurately locate the location of each power supply, and install sensor equipment in the industrial park to collect relevant data of each target distributed power supply, including the power supply's production capacity, operating status, real-time power and other information; for the target distributed storage, similar to the power supply, each target distributed storage has a location identifier to accurately locate the storage location, and install sensor equipment to collect relevant data of each target distributed storage, including the storage capacity, available status, real-time storage capacity and other information.

[0020] The collected power supply and storage data are integrated, and in the integrated data set, an inventory of all distributed power supplies and storage in the target industrial park is established, including their location identification and related characteristics, providing the necessary data basis for subsequent microgrid control.

[0021] Randomly extracting a first grid in a target grid set obtained by grid segmentation of the target industrial park, wherein the first grid has a first grid position identifier;

[0022] The grid size used to segment the target industrial park is preset. The grid can be a square or rectangular area of ​​fixed size, or it can be flexibly adjusted according to actual conditions. The target industrial park is segmented according to the preset grid size. For example, based on the method of geographic information system (GIS), the area is divided according to geographic coordinates to ensure reasonable and accurate division of the grid. The area after grid segmentation is composed of a target grid set, which contains the grids after the entire target industrial park is divided.

[0023] A random selection algorithm is used to randomly extract a first grid from the target grid set, and a first grid position identifier is assigned to the selected first grid. The position identifier can be a unique identifier such as the coordinates and number of the grid, which is used to accurately identify the position of the first grid.

[0024] Based on the first grid position identifier, sequentially traverse the plurality of power supplies with position identifiers to obtain a first power supply set and the plurality of memories with position identifiers to obtain a first memory set;

[0025] Using the first grid position identifier of the first grid as the target, from the power source set of the target industrial park, traverse each power source in turn according to the position identifier. During the traversal process, the power source belonging to the first grid position identifier is added to the first power source set. The power sources in the first power source set are all power sources in the first grid.

[0026] Similarly, from the memory set of the target industrial park, each memory is traversed in turn according to the location identifier, and the memory belonging to the first grid location identifier is added to the first memory set. The memories in the first memory set are all memories in the first grid.

[0027] Retrieving a first historical load database of the first grid, and obtaining a first cycle load state in combination with a predetermined cycle analysis;

[0028] According to the location identifier of the first grid, the corresponding historical load database is found. This database records the power load information of the grid in the past period of time. The query operation is performed to retrieve the historical load data of the first grid, including the power load values ​​at different time points, to establish a historical record of the load status.

[0029] A cycle is predefined, such as a day or a quarter, for analyzing changes in load status. This cycle can be adjusted according to system requirements. The load status within the predetermined cycle is analyzed using the retrieved historical load data, including identifying time periods with higher power demand as peak periods and time periods with lower power demand as valley periods, and then determining the characteristics of the load status of the first cycle, which include peak periods and valley periods. According to the results of the load analysis, the first load peak period and the first load valley period are determined. For example, there are peak periods and valley periods for power consumption in a day or throughout the year.

[0030] During a first load peak period in the first cycle load state, the first memory set is analyzed by a first edge computing center to generate a first power priority scheduling sequence for a first power source, where the first power source is any one of the first power sources in the first power source set, and the first edge computing center is an inherent computing center of the first power source;

[0031] Any power supply in the first power supply set has an inherent computing center, wherein a power supply is randomly selected in the first power supply set as the first power supply, and the first power supply corresponds to the first edge computing center, which can be a local computing device associated with the power supply.

[0032] During the first load peak period, the supply side is less than the demand side, and the energy stored in the memory needs to be dispatched to meet the current demand. Specifically, the first edge computing center is used to analyze the first memory set, the amount of energy stored in each memory and the availability are evaluated, and the scheduling algorithm is used to combine the energy reserve situation in the memory, the efficiency of the power supply and other factors to sort the memories for scheduling priority, and generate the first power supply priority scheduling sequence of the first power supply. In the sequence, the memory with a higher ranking has a higher energy scheduling priority, which is used to guide the energy scheduling operation of the power supply during the current load peak period, to meet the current demand, and to ensure reasonable energy scheduling when the supply side is less than the demand side during the load peak period.

[0033] Extracting a first memory in the first power priority scheduling sequence, and performing control optimization on the first memory in combination with a predetermined control fitness function to obtain a first optimal scheduling control decision;

[0034] A first memory, that is, a memory with the highest priority in the sequence, is selected from the first power priority scheduling sequence as the first memory. A predefined predetermined control fitness function is obtained, which may include factors such as efficiency, cost, reliability, etc. to evaluate the quality of the control decision.

[0035] The relevant parameters of the first storage, such as the amount of reserved energy, efficiency, etc., are input into a predetermined control fitness function for evaluation, and optimization is performed based on the evaluation results to obtain a first optimal scheduling control decision, which includes control parameters for the first storage to meet current load requirements and optimize energy utilization while considering various factors.

[0036] By executing the above steps, the memory with the highest priority scheduling is successfully extracted from the first power priority scheduling sequence, and the memory is controlled and optimized based on the predetermined control fitness function, and the first optimal scheduling control decision is obtained. This decision is used to guide the actual operation of the memory in microgrid control.

[0037] The first grid in the target industrial park is microgrid controlled according to the first optimal scheduling control decision.

[0038] The obtained first optimal dispatch control decision is executed and applied to the first grid, including dispatching energy in the storage to meet the current load demand.

[0039] Before optimizing the scheduling, the power supply, storage and load status in the target industrial park are monitored in real time, the current system status is analyzed, and the objects that need to be further optimized are determined, including power scheduling, storage energy utilization efficiency, etc. According to the real-time monitored system status, an optimization strategy is formulated, for example, adjusting the scheduling order of the power supply, optimizing the energy utilization of the storage, etc., and the determined optimization strategy is applied to the system to further improve the performance and efficiency of the microgrid. During the operation of the microgrid, multiple iterations of optimization adjustments are carried out to adapt to the real-time changes in load demand and energy supply conditions.

[0040] Through the above steps, the first grid in the target industrial park is prioritized according to the first optimal scheduling control decision, and then optimized scheduling is performed based on real-time monitoring to ensure that the microgrid optimizes energy utilization as much as possible while efficiently meeting load demand. This process can be iterative to adapt to changes in actual operation.

[0041] Further, the first historical load database of the first grid is retrieved, and a first cycle load state is obtained in combination with a predetermined cycle analysis, and then the method further includes:

[0042] In a first load valley period in the first cycle load state, analyzing the first memory set by the first edge computing center to generate a first memory priority storage sequence of the first power supply;

[0043] The energy of the first power supply is stored and controlled according to the first memory priority storage sequence.

[0044] When in the first load valley period, the supply side is greater than the demand side, and the energy generated by the current power supply needs to be dispatched and stored in the surrounding storage. Specifically, the first edge computing center analyzes the first storage set, evaluates the amount of energy stored in each storage and its availability, uses the storage scheduling algorithm, and combines the energy reserve in the storage, the efficiency of the storage, and other factors to sort the storage storage scheduling priorities, and generates the first storage priority storage sequence of the first power source. In the sequence, the higher the ranking of the storage, the higher the energy storage scheduling priority. The sequence guides the storage operation of the storage during the current load valley period.

[0045] According to the priority storage sequence of the first memory, the first memory is selected, which is the memory with the highest priority scheduling, and the energy generated by the first power source is stored in the memory for use during future peak periods or when needed. In this way, it is ensured that energy is effectively stored during the load valley period for use when needed in the future.

[0046] Further, during a first load peak period in the first periodic load state, the first memory set is analyzed by a first edge computing center to generate a first power priority scheduling sequence of the first power supply, including:

[0047] The first edge computing center stores a predetermined priority constraint;

[0048] Randomly obtain any memory in the first memory set, where the any memory has an arbitrary memory location identifier;

[0049] Generate a shortest scheduling path for the arbitrary memory based on the first power supply location identifier of the first power supply and the arbitrary memory location identifier;

[0050] Analyzing the shortest scheduling path in combination with the predetermined priority constraint to obtain a scheduling index of the arbitrary memory;

[0051] The first memory sets are arranged in descending order based on the scheduling index to form the first power priority scheduling sequence.

[0052] The predetermined priority constraints are standards set in advance based on specific optimization goals and constraints. These standards may include considerations such as efficiency, cost, and reliability. The predetermined priority constraints serve as the basis for guiding storage scheduling decisions to ensure that storage operations meet the system's optimization goals and constraints.

[0053] The first memory set is a collection of multiple memories. A memory is arbitrarily selected from the first memory set by a random selection method. The selected memory has a location identifier for identifying information about the location of the memory in the target industrial park.

[0054] The location identifier of the first power supply is information about the location of the first power supply in the target industrial park. Based on the location identifier of the first power supply and the location identifier of the selected memory, a path calculation algorithm is used to generate the shortest scheduling path connecting the two. This path is calculated based on the network structure in the target industrial park. The generated shortest scheduling path information includes key information such as nodes, edges, and distances on the path.

[0055] Extract the relevant information of the calculated shortest scheduling path, including the nodes, edges, distances, etc. on the path, apply the predetermined priority constraints to the path information to analyze the degree of compliance of the path, weigh multiple factors such as efficiency constraints and cost constraints, and calculate the scheduling index of any selected memory.

[0056] Using the calculated scheduling index of each memory, the memories in the first memory set are arranged in descending order from high to low according to their scheduling index. A high scheduling index indicates a memory that better meets the predetermined priority constraint. Based on the result of the descending order, a first power supply priority scheduling sequence is formed. The order of the memories in the sequence determines their scheduling priority during the current load peak period.

[0057] Furthermore, analyzing the shortest scheduling path in combination with the predetermined priority constraint to obtain the scheduling index of the arbitrary memory includes:

[0058] The predetermined priority constraints include efficiency constraints and cost constraints;

[0059] respectively obtaining an efficiency adjustment feedback coefficient of the shortest scheduling path based on the efficiency constraint and a cost feedback adjustment coefficient of the shortest scheduling path based on the cost constraint;

[0060] Normalization processing is performed to obtain path normalization data of the shortest scheduling path;

[0061] Obtaining a dispatching efficiency index according to the efficiency adjustment feedback coefficient and the path normalization data, and obtaining a dispatching cost index according to the cost feedback adjustment coefficient and the path normalization data;

[0062] The dispatching efficiency index and the dispatching cost index are weighted to obtain the dispatching index.

[0063] The efficiency constraint specifies the efficiency criteria that need to be considered when selecting the shortest scheduling path, that is, when calculating the shortest scheduling path, the energy transmission efficiency on the path will be considered, including minimizing energy transmission losses or ensuring efficient utilization of energy transmission; the cost constraint specifies the cost criteria that need to be considered when selecting the shortest scheduling path, that is, when calculating the shortest scheduling path, the cost factors on the path will be considered, including the actual cost of energy transmission, equipment maintenance costs, etc.

[0064] Through such predetermined priority constraints, the scheduling of storage during peak hours can be considered more comprehensively, ensuring that the selected scheduling path is carried out while improving efficiency and controlling costs, thereby achieving more sustainable and efficient microgrid energy management.

[0065] The efficiency adjustment feedback coefficient is a coefficient set for different path ranges of the shortest scheduling path based on efficiency constraints. It reflects the energy transmission efficiency of the path. Different efficiency adjustment feedback coefficients can be set for different path ranges. For example, the longer the path range, the higher the efficiency adjustment feedback coefficient.

[0066] The cost feedback adjustment coefficient is a coefficient set for the shortest dispatch path based on cost constraints, reflecting the dispatch cost of the path. The cost feedback adjustment coefficient is set based on historical experience, previous experimental results, or according to specific needs. For example, within the range of 0-5 kilometers, the cost coefficient is set to 1.2; within the range of 5-10 kilometers, the cost coefficient is set to 1.1; within the range of 10-15 kilometers, the cost coefficient is set to 1.3. In other words, the shorter the path, the lower the dispatch cost, but it is set independently based on historical experience.

[0067] By setting different adjustment coefficients according to efficiency and cost constraints, the characteristics of each path can be considered more specifically, ensuring that the path selection meets both efficiency requirements and cost constraints.

[0068] The relevant data of the shortest scheduling path is calculated, including information such as nodes, edges, and distances on the path, to ensure that the scale is relatively consistent within different paths and obtain normalized data. Normalization helps to easily compare the characteristics of different paths in comprehensive analysis. It ensures that the data of different paths are on the same scale to better balance efficiency and cost.

[0069] Applying the efficiency adjustment feedback coefficient to the normalized data, such as performing a simple product operation, obtains the dispatch efficiency index, which reflects the energy transmission efficiency of the path; applying the cost feedback adjustment coefficient to the normalized data, such as performing a simple product operation, obtains the dispatch cost index, which reflects the dispatch cost of the path. By calculating these two indices, the performance of each path can be evaluated more comprehensively.

[0070] The weights of scheduling efficiency and cost in the final index are set, which can be set according to system requirements, optimization goals, etc. to reflect the relative importance of efficiency and cost. According to the set weights, the scheduling efficiency index and the scheduling cost index are weighted and summed to obtain the final scheduling index. The final scheduling index reflects the comprehensive consideration of efficiency and cost, and is used to form a priority scheduling sequence for the first power source to guide the energy scheduling of the storage during peak periods.

[0071] By taking into account scheduling efficiency and cost in a weighted manner, storage scheduling decisions can be made more flexibly, ensuring that the most suitable path is selected during peak hours while balancing energy transmission efficiency and scheduling cost.

[0072] Furthermore, the method comprises:

[0073] Reading a predetermined scheduling control parameter, wherein the predetermined scheduling control parameter includes a scheduling power, a scheduling frequency, and a scheduling voltage, wherein the scheduling power has a power threshold, the scheduling frequency has a frequency threshold, and the scheduling voltage has a voltage threshold;

[0074] forming a first scheduling control decision based on a first power extracted from the power threshold, a first frequency extracted from the frequency threshold, and a first voltage extracted from the voltage threshold;

[0075] Matching a first historical dispatch record of the first dispatch control decision in a historical dispatch control database, wherein the first historical dispatch includes first historical dispatch efficiency, first historical dispatch stability, and first historical dispatch environmental friendliness;

[0076] Calculating the first historical scheduling efficiency, the first historical scheduling stability and the first historical scheduling environmental friendliness in combination with the predetermined control fitness function to obtain a first scheduling fitness;

[0077] When the first scheduling fitness reaches a predetermined fitness threshold, the first scheduling control decision is used as the first optimal scheduling control decision.

[0078] The dispatching power refers to the energy transmission power between the storage and the power supply in the microgrid system. The dispatching power has a power threshold, which is used to determine the upper limit of the energy transmission power between the storage and the power supply; the dispatching frequency refers to the energy transmission frequency between the storage and the power supply in the microgrid system. The dispatching frequency has a frequency threshold, which is used to set the upper limit of the energy transmission frequency; the dispatching voltage refers to the energy transmission voltage between the storage and the power supply in the microgrid system. The dispatching voltage has a voltage threshold, which is used to set the upper limit of the energy transmission voltage.

[0079] By reading these predetermined dispatching control parameters, it is possible to ensure that the energy transmission power, frequency and voltage are within a safe and controllable range when performing microgrid control, thereby maintaining the stability and reliability of the microgrid system.

[0080] Using the power threshold in the predetermined scheduling control parameter, a first power is extracted to represent the energy transmission power between the storage and the power supply in the microgrid system; using the frequency threshold in the predetermined scheduling control parameter, a first frequency is extracted to represent the frequency of energy transmission in the microgrid system; using the voltage threshold in the predetermined scheduling control parameter, a first voltage is extracted to represent the voltage of energy transmission in the microgrid system.

[0081] A first scheduling control decision is formed using the extracted first power, first frequency, and first voltage. The first scheduling control decision includes adjusting the energy transfer power, frequency, and voltage between the memory and the power supply to ensure that they are within a safe and controllable range.

[0082] The historical dispatch control database stores the historical records of the microgrid system's past dispatch control decisions, including various parameters and performance indicators. The current first dispatch control decision is used as the query condition, and based on the principle of similarity coefficient, the historical records matching the current dispatch decision are searched in the historical dispatch control database. Once the match is successful, the first historical dispatch record is extracted from the database, including the first historical dispatch efficiency, the first historical dispatch stability, and the first historical dispatch environmental protection.

[0083] The predetermined control fitness function is a mathematical function that comprehensively considers the indicators of historical dispatch efficiency, dispatch stability, and dispatch environmental protection. The first historical dispatch efficiency, the first historical dispatch stability, and the first historical dispatch environmental protection extracted from the historical dispatch control database are used, and these historical parameters are input into the predetermined control fitness function to calculate and obtain the first dispatch fitness, which is a comprehensive fitness value used to evaluate the comprehensive performance of historical dispatch.

[0084] The predetermined fitness threshold is a preset threshold value, which is set according to actual needs and historical experience and is used to measure the fitness value. If the fitness reaches the predetermined threshold value, it means that the current decision is superior under the past historical performance and can be regarded as the optimal scheduling control decision. Compare the calculated first scheduling fitness with the predetermined fitness threshold value. If the first scheduling fitness reaches the predetermined fitness threshold value, the first scheduling control decision is determined as the first optimal scheduling control decision; if it does not reach the threshold value, it may be necessary to reconsider or optimize the scheduling strategy.

[0085] Furthermore, matching the first historical scheduling record of the first scheduling control decision in the historical scheduling control database includes:

[0086] Randomly extracting a first historical record from the historical dispatch control database, the first historical record comprising a first historical control decision;

[0087] Performing a similarity analysis on the first scheduling control decision and the first historical control decision based on a similarity coefficient principle to obtain a first similarity;

[0088] The first historical scheduling record is obtained by screening based on the first similarity.

[0089] A historical record is randomly selected from the historical dispatch control database as a first historical record, where the record includes a first historical control decision, specifically including parameter information such as a first historical power, a first historical frequency, and a first historical voltage.

[0090] The similarity coefficient is an indicator used to measure the similarity between two objects. Here, the object is the dispatch control decision. The higher the similarity, the more similar the two objects are. Compare the parameters of the current first dispatch control decision with the parameters of the first historical control decision, that is, compare the first power with the first historical power, the first frequency with the first historical frequency, and the first voltage with the first historical voltage, and perform weighted summation on the comparison results of each pair of parameters. The calculation result is used as the similarity between the two decisions to obtain the first similarity, so as to judge whether the current decision is similar to the historical situation, so as to better draw on past experience to guide the current microgrid dispatch decision.

[0091] A similarity calculation is performed on multiple historical records in the historical scheduling database, and according to the calculation result, the historical scheduling record with the highest similarity to the current decision is screened out as the first historical scheduling record.

[0092] Furthermore, the expression of the predetermined control fitness function is as follows:

[0093] f(x)=α*eff(x)+β*sta(x)+γ*envir(x);

[0094] Among them, f(x) represents the first scheduling fitness of the first scheduling control decision x, eff(x) represents the normalized result of the first historical scheduling efficiency, sta(x) represents the normalized result of the first historical scheduling stability, envir(x) represents the normalized result of the first historical scheduling environmental friendliness, α, β and γ represent the first coefficient, the second coefficient and the third coefficient respectively, and α+β+γ=1.

[0095] This fitness function obtains a comprehensive fitness value by weighing different coefficients based on the efficiency, stability and environmental protection performance of historical scheduling. By adjusting the weights of the coefficients, we can pay more attention to a certain aspect of performance while considering historical performance. Such a fitness function is used to evaluate and select different scheduling control decisions, so that the system can comprehensively consider historical performance in different aspects.

[0096] Further, during a first load peak period in the first cycle load state, the first memory set is analyzed by a first edge computing center to generate a first power priority scheduling sequence for a first power source, which also includes:

[0097] reading a first predetermined load demand threshold;

[0098] determining whether the first real-time load of the first grid during the first load peak period is within the first predetermined load demand threshold;

[0099] If not, issuing a monitoring instruction, and dynamically monitoring the first grid during the first load peak period based on the monitoring instruction to obtain a first real-time operating state;

[0100] A first abnormal device is identified based on the first real-time operating status, and the first abnormal device is isolated and repaired.

[0101] The value of the first predetermined load demand threshold is read from a predetermined configuration file or a specified location. This threshold is used to determine whether the current load has reached the expected demand level, thereby determining whether dynamic monitoring and scheduling are required.

[0102] The first real-time load data of the first grid during the first load peak period is obtained from the real-time monitoring or data source, and the obtained real-time load is compared with the first predetermined load demand threshold. If the real-time load is greater than or equal to the first predetermined load demand threshold, it indicates that the load is within the expected range and no additional dynamic monitoring and scheduling is required; otherwise, it is considered that the load has not reached the expected level and corresponding processing is required.

[0103] If the first predetermined load demand threshold is not reached, a monitoring instruction is issued to monitor the first grid during the first load peak period, and the monitoring instruction is sent to relevant monitoring equipment. The monitoring equipment executes the monitoring instruction to obtain the first real-time operating status of the first grid during the first load peak period, including real-time load, power, frequency and other parameters.

[0104] Use intelligent models, machine learning algorithms or other technologies to analyze the first real-time operating status data, identify abnormal equipment that may exist in the first grid during the first load peak period through analysis of the real-time data, and once the abnormal equipment is identified, generate instructions for isolation and maintenance, send the isolation and maintenance instructions to the relevant equipment control system, and the equipment control system executes the isolation and maintenance instructions to isolate the abnormal equipment from the system, and perform necessary inspection and maintenance work on the isolated abnormal equipment.

[0105] By identifying and isolating abnormal equipment, it helps to ensure the stable operation of the microgrid system during peak periods and improve the reliability and safety of the system.

[0106] In summary, the multi-energy complementary industrial park microgrid autonomous control method and system provided by the embodiment of the present application has the following technical effects:

[0107] 1. By collecting the distributed power supply and storage information in the target industrial park, as well as grid segmentation and location identification, it realizes the collection and coordinated scheduling of multiple energy sources and optimizes the energy utilization efficiency of the entire microgrid system;

[0108] 2. Through random grid extraction, historical load database retrieval and real-time monitoring, accurate energy dispatch decisions can be made based on real-time data to meet the real-time needs of industrial parks;

[0109] 3. Through edge computing center, analysis and optimization are performed to extract the priority scheduling sequence, and a predetermined control fitness function is applied to optimize the scheduling decision of the storage, ensuring that the demand can be met during peak hours while taking into account the comprehensive impact of historical performance to the greatest extent;

[0110] 4. According to the optimal scheduling and control decision, the microgrid control of the grid in the target industrial park is realized, the operation of distributed power sources and storage is effectively coordinated, and the reliability and stability of the entire microgrid system are improved.

[0111] In summary, this method enables the industrial park microgrid to be more intelligent and efficient in autonomous control by introducing key features such as multi-energy complementarity, real-time, and optimization, which helps to improve energy utilization efficiency and reduce energy costs, while ensuring that the system can operate stably and reliably under dynamically changing demands.

[0112] Embodiment 2

[0113] Based on the same inventive concept as the multi-energy complementary industrial park microgrid autonomous control method in the aforementioned embodiment, Figure 2 As shown, the present application provides a multi-energy complementary industrial park microgrid autonomous control system, the system comprising:

[0114] A distributed memory acquisition module 10, wherein the distributed memory acquisition module 10 is used to respectively acquire a target distributed power source and a target distributed memory in a target industrial park, wherein the target distributed power source includes a plurality of power sources with location identifiers, and the target distributed memory includes a plurality of memories with location identifiers;

[0115] A first grid acquisition module 20, the first grid acquisition module 20 is used to randomly extract a first grid in a target grid set obtained by grid segmentation of the target industrial park, the first grid having a first grid position identifier;

[0116] A first memory set acquisition module 30, the first memory acquisition module 30 is used to sequentially traverse the plurality of power supplies with position identifiers based on the first grid position identifier to obtain a first power supply set and the plurality of memories with position identifiers to obtain a first memory set;

[0117] A load state acquisition module 40, the load state acquisition module 40 is used to retrieve a first historical load database of the first grid, and obtain a first period load state in combination with a predetermined period analysis;

[0118] A first memory set analysis module 50, the first memory set analysis module 50 is used to analyze the first memory set through a first edge computing center during a first load peak period in the first periodic load state to generate a first power priority scheduling sequence for a first power source, the first power source being any one of the first power sources in the first power source set, and the first edge computing center being an inherent computing center of the first power source;

[0119] A control optimization module 60, the control optimization module 60 is used to extract the first memory in the first power priority scheduling sequence, and perform control optimization on the first memory in combination with a predetermined control fitness function to obtain a first optimal scheduling control decision;

[0120] A microgrid control module 70 is used to perform microgrid control on the first grid in the target industrial park according to the first optimal scheduling control decision.

[0121] Furthermore, the system further comprises:

[0122] a storage sequence generation module, the storage sequence generation module being used to analyze the first memory set through the first edge computing center in a first load valley period in the first cycle load state to generate a first memory priority storage sequence of the first power supply;

[0123] A storage control module is used to control the storage of energy of the first power supply according to the first memory priority storage sequence.

[0124] Furthermore, the system further includes a priority scheduling sequence generation module to perform the following operation steps:

[0125] The first edge computing center stores a predetermined priority constraint;

[0126] Randomly obtain any memory in the first memory set, where the any memory has an arbitrary memory location identifier;

[0127] Generate a shortest scheduling path for the arbitrary memory based on the first power supply location identifier of the first power supply and the arbitrary memory location identifier;

[0128] Analyzing the shortest scheduling path in combination with the predetermined priority constraint to obtain a scheduling index of the arbitrary memory;

[0129] The first memory sets are arranged in descending order based on the scheduling index to form the first power priority scheduling sequence.

[0130] Furthermore, the system further includes a scheduling index generation module to perform the following operation steps:

[0131] The predetermined priority constraints include efficiency constraints and cost constraints;

[0132] respectively obtaining an efficiency adjustment feedback coefficient of the shortest scheduling path based on the efficiency constraint and a cost feedback adjustment coefficient of the shortest scheduling path based on the cost constraint;

[0133] Normalization processing is performed to obtain path normalization data of the shortest scheduling path;

[0134] Obtaining a dispatching efficiency index according to the efficiency adjustment feedback coefficient and the path normalization data, and obtaining a dispatching cost index according to the cost feedback adjustment coefficient and the path normalization data;

[0135] The dispatching efficiency index and the dispatching cost index are weighted to obtain the dispatching index.

[0136] Furthermore, the system also includes an optimal scheduling control decision generation module to perform the following operation steps:

[0137] Reading a predetermined scheduling control parameter, wherein the predetermined scheduling control parameter includes a scheduling power, a scheduling frequency, and a scheduling voltage, wherein the scheduling power has a power threshold, the scheduling frequency has a frequency threshold, and the scheduling voltage has a voltage threshold;

[0138] forming a first scheduling control decision based on a first power extracted from the power threshold, a first frequency extracted from the frequency threshold, and a first voltage extracted from the voltage threshold;

[0139] Matching a first historical dispatch record of the first dispatch control decision in a historical dispatch control database, wherein the first historical dispatch includes first historical dispatch efficiency, first historical dispatch stability, and first historical dispatch environmental friendliness;

[0140] Calculating the first historical scheduling efficiency, the first historical scheduling stability, and the first historical scheduling environmental friendliness in combination with the predetermined control fitness function to obtain a first scheduling fitness;

[0141] When the first scheduling fitness reaches a predetermined fitness threshold, the first scheduling control decision is used as the first optimal scheduling control decision.

[0142] Furthermore, the system further includes a first historical scheduling record generating module to perform the following operation steps:

[0143] Randomly extracting a first historical record from the historical dispatch control database, the first historical record comprising a first historical control decision;

[0144] Performing a similarity analysis on the first scheduling control decision and the first historical control decision based on a similarity coefficient principle to obtain a first similarity;

[0145] The first historical scheduling record is obtained by screening based on the first similarity.

[0146] Furthermore, the expression of the predetermined control fitness function is as follows:

[0147] f(x)=α*eff(x)+β*sta(x)+γ*envir(x);

[0148] Among them, f(x) represents the first scheduling fitness of the first scheduling control decision x, eff(x) represents the normalized result of the first historical scheduling efficiency, sta(x) represents the normalized result of the first historical scheduling stability, envir(x) represents the normalized result of the first historical scheduling environmental friendliness, α, β and γ represent the first coefficient, the second coefficient and the third coefficient respectively, and α+β+γ=1.

[0149] Furthermore, the system also includes an isolation and maintenance module to perform the following operation steps:

[0150] reading a first predetermined load demand threshold;

[0151] determining whether the first real-time load of the first grid during the first load peak period is within the first predetermined load demand threshold;

[0152] If not, issuing a monitoring instruction, and dynamically monitoring the first grid during the first load peak period based on the monitoring instruction to obtain a first real-time operating state;

[0153] A first abnormal device is identified based on the first real-time operating status, and the first abnormal device is isolated and repaired.

[0154] Through the above-mentioned detailed description of a multi-energy complementary industrial park microgrid autonomous control method in this specification, those skilled in the art can clearly know a multi-energy complementary industrial park microgrid autonomous control system in this embodiment. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0155] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-energy complementary industrial park microgrid autonomous control method, characterized in that: The method comprises: The target distributed power source and the target distributed storage in the target industrial park are respectively acquired, wherein the target distributed power source includes a plurality of power sources with location identifiers, and the target distributed storage includes a plurality of storages with location identifiers; Randomly extracting a first grid in a target grid set obtained by grid segmentation of the target industrial park, wherein the first grid has a first grid position identifier; Based on the first grid position identifier, sequentially traverse the plurality of power supplies with position identifiers to obtain a first power supply set and the plurality of memories with position identifiers to obtain a first memory set; Retrieving a first historical load database of the first grid, and obtaining a first cycle load state in combination with a predetermined cycle analysis; During a first load peak period in the first cycle load state, the first memory set is analyzed by a first edge computing center to generate a first power priority scheduling sequence for a first power source, where the first power source is any one of the first power sources in the first power source set, and the first edge computing center is an inherent computing center of the first power source; Extracting a first memory in the first power priority scheduling sequence, and performing control optimization on the first memory in combination with a predetermined control fitness function to obtain a first optimal scheduling control decision; Performing microgrid control on the first grid in the target industrial park according to the first optimal scheduling control decision; Wherein, during the first load peak period in the first cycle load state, the first memory set is analyzed by the first edge computing center to generate a first power priority scheduling sequence of the first power supply, including: The first edge computing center stores a predetermined priority constraint; Randomly obtain any memory in the first memory set, where the any memory has an arbitrary memory location identifier; Generate a shortest scheduling path for the arbitrary memory based on the first power supply location identifier of the first power supply and the arbitrary memory location identifier; Analyzing the shortest scheduling path in combination with the predetermined priority constraint to obtain a scheduling index of the arbitrary memory; The first memory sets are arranged in descending order based on the scheduling index to form the first power priority scheduling sequence.

2. The method according to claim 1, characterized in that: Retrieving the first historical load database of the first grid, and obtaining the first cycle load state in combination with the predetermined cycle analysis, and then further comprising: In a first load valley period in the first cycle load state, analyzing the first memory set by the first edge computing center to generate a first memory priority storage sequence of the first power supply; The energy of the first power supply is stored and controlled according to the first memory priority storage sequence.

3. The method according to claim 1, characterized in that: Analyzing the shortest scheduling path in combination with the predetermined priority constraint to obtain a scheduling index of the arbitrary memory includes: The predetermined priority constraints include efficiency constraints and cost constraints; respectively obtaining an efficiency adjustment feedback coefficient of the shortest scheduling path based on the efficiency constraint and a cost feedback adjustment coefficient of the shortest scheduling path based on the cost constraint; Normalization processing is performed to obtain path normalization data of the shortest scheduling path; Obtaining a dispatching efficiency index according to the efficiency adjustment feedback coefficient and the path normalization data, and obtaining a dispatching cost index according to the cost feedback adjustment coefficient and the path normalization data; The dispatching efficiency index and the dispatching cost index are weighted to obtain the dispatching index.

4. The method according to claim 1, characterized in that: include: Reading a predetermined scheduling control parameter, wherein the predetermined scheduling control parameter includes a scheduling power, a scheduling frequency, and a scheduling voltage, wherein the scheduling power has a power threshold, the scheduling frequency has a frequency threshold, and the scheduling voltage has a voltage threshold; forming a first scheduling control decision based on a first power extracted from the power threshold, a first frequency extracted from the frequency threshold, and a first voltage extracted from the voltage threshold; Matching a first historical dispatch record of the first dispatch control decision in a historical dispatch control database, wherein the first historical dispatch includes first historical dispatch efficiency, first historical dispatch stability, and first historical dispatch environmental friendliness; Calculating the first historical scheduling efficiency, the first historical scheduling stability and the first historical scheduling environmental friendliness in combination with the predetermined control fitness function to obtain a first scheduling fitness; When the first scheduling fitness reaches a predetermined fitness threshold, the first scheduling control decision is used as the first optimal scheduling control decision.

5. The method according to claim 4, characterized in that: Matching the first historical scheduling record of the first scheduling control decision in a historical scheduling control database includes: Randomly extracting a first historical record from the historical dispatch control database, the first historical record comprising a first historical control decision; Performing a similarity analysis on the first scheduling control decision and the first historical control decision based on a similarity coefficient principle to obtain a first similarity; The first historical scheduling record is obtained by screening based on the first similarity.

6. The method according to claim 4, characterized in that: The expression of the predetermined control fitness function is as follows: ； in, Characterizing the First Scheduling Control Decision The first scheduling fitness, A normalized result characterizing the first historical scheduling efficiency, A normalized result characterizing the stability of the first historical scheduling, The normalized result characterizing the environmental friendliness of the first historical dispatch, and and represent the first coefficient, the second coefficient and the third coefficient respectively, and .

7. The method according to claim 1, characterized in that: During a first load peak period in the first cycle load state, the first memory set is analyzed by a first edge computing center to generate a first power priority scheduling sequence for a first power source, which also includes: reading a first predetermined load demand threshold; determining whether the first real-time load of the first grid during the first load peak period is within the first predetermined load demand threshold; If not, issuing a monitoring instruction, and dynamically monitoring the first grid during the first load peak period based on the monitoring instruction to obtain a first real-time operating state; A first abnormal device is identified based on the first real-time operating status, and the first abnormal device is isolated and repaired.

8. A multi-energy complementary industrial park microgrid autonomous control system, characterized in that: A method for autonomously controlling a multi-energy complementary industrial park microgrid according to any one of claims 1 to 7, comprising: A distributed memory acquisition module, the distributed memory acquisition module is used to respectively acquire a target distributed power source and a target distributed memory in a target industrial park, the target distributed power source includes a plurality of power sources with location identifiers, and the target distributed memory includes a plurality of memories with location identifiers; A first grid acquisition module, the first grid acquisition module is used to randomly extract a first grid in a target grid set obtained by grid segmentation of the target industrial park, the first grid having a first grid position identifier; A first memory set acquisition module, the first memory acquisition module is used to sequentially traverse the plurality of power supplies with position identifiers based on the first grid position identifier to obtain a first power supply set and the plurality of memories with position identifiers to obtain a first memory set; A load state acquisition module, the load state acquisition module is used to retrieve a first historical load database of the first grid, and obtain a first period load state in combination with a predetermined period analysis; a first memory set analysis module, the first memory set analysis module being used to analyze the first memory set through a first edge computing center during a first load peak period in the first periodic load state to generate a first power priority scheduling sequence for a first power source, the first power source being any one of the first power sources in the first power source set, and the first edge computing center being an inherent computing center of the first power source; A control optimization module, the control optimization module is used to extract the first memory in the first power priority scheduling sequence, and perform control optimization on the first memory in combination with a predetermined control fitness function to obtain a first optimal scheduling control decision; A microgrid control module, wherein the microgrid control module is used to perform microgrid control on the first grid in the target industrial park according to the first optimal scheduling control decision.

Images