Micro-grid source-grid-load-storage integrated optimal scheduling method and system

Abstract

The invention discloses a source-grid-load-storage integrated optimal scheduling method and system for a micro-grid, and relates to the technical field of smart grids and micro-grids, and the method comprises the steps: firstly collecting multi-source data in a specified micro-grid, and carrying out the optimization storage of the multi-source data in a time sequence database; constructing a time sequence prediction model, predicting generating capacity and load demand according to historical data, generating a preliminary scheduling strategy according to the generating capacity and load demand, realizing comprehensive cost minimization, analyzing the strategy, evaluating deviation and abnormity, and finally correcting the strategy according to an evaluation result; comprising an intelligent data aggregation module, a time sequence data prediction module, a preliminary scheduling generation module, a preliminary scheduling strategy analysis module and a real-time optimization feedback module, influences of various data sources are considered, the model can be adjusted in real time, variable power demands and power generation conditions in a micro-grid environment can be better coped with, and the real-time optimization feedback is realized. The accuracy of a prediction result is greatly improved, so that a powerful basis is provided for scheduling decision making, and risks possibly caused by prediction deviation are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grids and microgrids, and in particular to a method and system for optimizing the scheduling of integrated source-grid-load-storage microgrids. Background Art

[0002] With the massive access of new energy sources to microgrids, the real-time carbon emission factor of electricity at the energy supply end fluctuates frequently, and there is strong uncertainty in each link of source, grid, load and storage. Traditional scheduling methods are difficult to meet the demand, and it is necessary to optimize the management of power resources. The present invention realizes the integrated optimization scheduling of source, grid, load and storage of microgrids through data aggregation, prediction and other links, thereby improving stability and economy.

[0003] Prior art, such as the invention patent application with publication number CN113991648A, discloses a microgrid monitoring system based on source, grid, load and storage, including a data collection module, a microgrid management module and an energy storage management module; the data collection module is used to build a microgrid database, collect weather information, the power load of the equipment cluster, collect the power parameters of each virtual subnet in the microgrid in real time, and enter the comprehensive data of the microgrid into the system; the microgrid management module is used to build a virtual subnet, analyze the impact of weather information on the power generation of new energy power generation, further analyze the comprehensive data of the microgrid, use the comprehensive data of the microgrid to train the model, and predict the power generation of each virtual subnet of the microgrid, and match the virtual subnet with the power load of the equipment cluster according to the power generation of the virtual subnet; the microgrid management module is also used to monitor the power load of each virtual subnet in real time, and dispatch power for power compensation according to the power load situation of each virtual subnet.

[0004] There are at least the following technical problems with the above scheme: 1. The above scheme lacks a detailed description of the time series data prediction model, which will lead to the lack of scientificity and accuracy of the microgrid dispatching system in predicting future power generation and load demand. The time series data prediction model is the core component of the microgrid dispatching system, which is conducive to predicting future power generation and load demand through historical data, thereby providing a basis for the generation of dispatching strategies. The above scheme only mentions the use of microgrid comprehensive data to train the model and predict power generation, but does not explain in detail how to build a time series prediction model, how to process historical data, how to consider the impact of environmental factors on power generation and other key issues. This may cause the system to have large deviations when predicting future power generation, thereby affecting the accuracy of the dispatching strategy and the stability of the microgrid.

[0005] 2. The above scheme lacks a detailed description of the load demand balance model, which will make it difficult for the microgrid to achieve supply and demand balance during the scheduling process. Although the above scheme mentions matching the virtual subnet with the power load of the equipment cluster according to the power generation of the virtual subnet, it does not explain in detail how to achieve load demand balance through a mathematical model, nor does it mention how to consider economic factors such as power generation costs and energy storage discharge costs. This will make it difficult for the system to achieve the optimal supply and demand balance during the scheduling process, resulting in power waste or power shortages.

[0006] 3. The above scheme lacks a detailed description of the abnormal evaluation and real-time optimization feedback mechanism, which will make it difficult for the microgrid to respond to emergencies and adjust the dispatch strategy in real time during operation. Although the above scheme mentions real-time monitoring of the power load of each virtual subnet and dispatching power for compensation according to the load situation, it does not explain in detail how to evaluate the deviation between the load and the predicted value through a mathematical model, nor does it mention how to correct the dispatch strategy based on the deviation, which will make it difficult for the system to respond to emergencies during operation, such as sudden load changes, equipment failures, etc. Summary of the invention

[0007] The purpose of the present invention is to provide a microgrid source-grid-load-storage integrated optimization scheduling method and system, which solves the problems existing in the background technology.

[0008] To solve the above technical problems, the present invention adopts the following technical solutions: The first aspect of the present invention provides a microgrid source-grid-load-storage integrated optimization scheduling method, including: Step 1, intelligent data aggregation: In a specified microgrid, various types of data sources are collected and stored in a time series database in an optimized manner.

[0009] Step 2: Time series data prediction: Based on the data sources stored in the time series database, a time series prediction model for renewable energy power generation and energy storage status is constructed to predict power generation and load demand within a set time period in the future through historical power generation and load demand data.

[0010] Step 3: Generate preliminary dispatch: Generate a preliminary dispatch strategy based on the predicted power generation and load demand within a set time period in the future to minimize the overall cost.

[0011] Step 4: Preliminary dispatch strategy analysis: Based on the generated preliminary dispatch strategy, the deviation between the microgrid load and the predicted value is evaluated. When there is a deviation between the microgrid load and the predicted value, an abnormality assessment is performed by collecting the real-time operation data of the microgrid.

[0012] Step 5: Real-time optimization feedback: According to the abnormal evaluation results of the preliminary scheduling strategy, make abnormal corrections to the preliminary scheduling strategy.

[0013] In a second aspect, the present invention provides a microgrid source-grid-load-storage integrated optimization dispatching system, including: an intelligent data aggregation module, which is used to optimize and store various types of data sources collected in a specified microgrid into a time series database.

[0014] The time series data prediction module is used to build a time series prediction model for renewable energy power generation and energy storage status based on the data sources stored in the time series database, so as to predict the power generation and load demand in the future set time period through historical power generation and load demand data.

[0015] The preliminary scheduling generation module is used to generate a preliminary scheduling strategy based on the predicted power generation and load demand within a set time period in the future, so as to minimize the overall cost.

[0016] The preliminary dispatch strategy analysis module is used to evaluate the deviation between the microgrid load and the predicted value based on the generated preliminary dispatch strategy. When there is a deviation between the microgrid load and the predicted value, an abnormality assessment is performed by collecting the real-time operation data of the microgrid.

[0017] The real-time optimization feedback module is used to make abnormal corrections to the preliminary scheduling strategy based on the abnormal evaluation results of the preliminary scheduling strategy.

[0018] The beneficial effects of the present invention are: 1. A microgrid source-grid-load-storage integrated optimization scheduling method and system provided by an embodiment of the present invention, in the process of intelligent data aggregation, optimizes the storage of various types of data sources in a time series database, which is conducive to improving the efficiency and accuracy of data processing. By storing multiple data sources such as new energy power generation data, energy storage system data, load demand data and environmental data in a time series and uniformly formatting them into data sorted by timestamps, the integrity and consistency of the data are ensured, which helps to provide high-quality input data for subsequent prediction models, reduce redundancy and errors in data storage and processing, and improve the accuracy of subsequent decision-making.

[0019] 2. In the process of time series data prediction, the embodiment of the present invention constructs a time series prediction model of renewable energy power generation and energy storage status, which is beneficial to improving the accuracy of future power generation and load demand prediction. It uses historical power generation and load demand data, as well as a set power generation contribution threshold, to screen valid data from the time series database, and optimizes the modeling through adaptive weights. It not only takes into account the impact of multiple data sources, but also can adjust the model in real time to better cope with the variable power demand and power generation conditions in the microgrid environment, greatly improving the accuracy of the prediction results, thereby providing a strong basis for scheduling decisions and avoiding the risks that may be caused by prediction deviations.

[0020] 3. In the process of generating the preliminary dispatching strategy, the embodiment of the present invention optimizes the power dispatching through the load demand balance model, which is conducive to maximizing the economic benefits of the microgrid. According to the predicted power generation and load demand, combined with the power generation cost and energy storage discharge cost, a preliminary dispatching strategy that minimizes the overall cost is generated through optimization calculation, which realizes the reasonable allocation of resources, ensures the satisfaction of power demand, optimizes energy costs, and ensures the efficient operation of the microgrid.

[0021] 4. In the preliminary dispatch strategy analysis process, the embodiment of the present invention evaluates the deviation between the load and the predicted value, which is conducive to discovering potential problems in the dispatch and making timely corrections. By comparing with the set deviation threshold, the deviation between the microgrid load and the predicted value can be identified in real time. If the deviation exceeds the set threshold, the system will issue an alarm in time and collect real-time operating data for abnormal evaluation, which provides a basis for adjusting and optimizing the dispatch strategy and effectively avoids the negative impact of dispatch deviation on the stability and economic benefits of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 It is a schematic flow chart of the implementation steps of the present invention.

[0024] Figure 2 It is a schematic diagram of the system structure connection of the present invention.

[0025] Figure 3 It is a schematic diagram of the flow chart of the optimization scheduling method of the present invention. DETAILED DESCRIPTION

[0026] 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.

[0027] See also Figure 1-2 As shown, the present invention provides a microgrid source-grid-load-storage integrated optimization scheduling method, the method comprising: step one, intelligent data aggregation: in a specified microgrid, various types of data sources are collected and stored in a time series database in an optimized manner.

[0028] In a specific embodiment, the various types of data sources collected are optimized and stored in a time series database. The specific process is as follows: various types of data sources include new energy power generation data, energy storage system data, load demand data and environmental data. Various types of data sources corresponding to the specified microgrid are collected through sensor equipment, system equipment itself and user-side interface, and the collected various types of data sources are stored in a time series manner, and then the various types of data sources are uniformly converted into a format sorted by timestamp, so that the various types of data sources containing each timestamp are stored in the time series database, and the various types of data sources stored in the time series database are preprocessed. When the data preprocessing is completed, it indicates that the collected various types of data sources have completed the optimized storage.

[0029] It should be noted that the new energy power generation data is monitored in real time by smart sensors installed on the power generation equipment. The sensors measure parameters such as wind speed, light intensity and power generation, and transmit the data to the central server through wireless communication technology. The energy storage system data comes from the energy storage equipment itself, such as the battery management system. The battery management system records the battery's charging and discharging status, battery power, voltage and current in real time, and uploads it to the central server through wired or wireless networks. The load demand data comes from the smart meter at the user end of the microgrid. The smart meter measures the power consumption of the user end according to the set monitoring time point and uploads it to the central server through the data acquisition terminal. Environmental factors such as weather and temperature data are collected through weather stations, temperature and humidity sensors or meteorological service APIs, and connected to the central server through an interface.

[0030] It should also be noted that the process of data preprocessing mainly includes steps such as data cleaning, data format conversion, outlier detection and correction, missing value processing, and data normalization. For example, the timestamps of renewable energy power generation data, energy storage system data, load demand data, and environmental data need to be uniformly sorted so that they can be stored in time series. In terms of outlier detection, statistical methods or machine learning models are used to identify and correct abnormal fluctuations or mutation values ​​in the data to ensure the authenticity of the data. For missing values, interpolation or filling methods based on historical data are usually used to reduce the impact of incomplete data. Finally, data normalization is performed to convert data of different magnitudes to a unified scale for subsequent analysis and modeling.

[0031] Step 2: Time series data prediction: Based on the data sources stored in the time series database, a time series prediction model for renewable energy power generation and energy storage status is constructed to predict power generation and load demand within a set time period in the future through historical power generation and load demand data.

[0032] In a specific embodiment, the time series prediction model for renewable energy power generation and energy storage status is constructed, and the specific process is as follows: the historical power generation contribution corresponding to each power generation mode in the specified microgrid is obtained from the time series database, and compared with the set power generation contribution threshold. If the historical power generation contribution of a power generation mode is greater than or equal to the set power generation contribution threshold, the power generation mode is included in the corresponding input variable of the time series prediction model, and the corresponding input variables included in the time series prediction model are The power generation at the time is recorded as , It is represented by the number corresponding to each power generation mode. , is a positive integer, and the power load corresponding to each user end is collected from the specified microgrid, which is recorded as the load demand , Represents the time index, , is a positive integer, when the prediction When the total power generation forecast value at the moment is obtained, the forecast model is used: ,get The total power generation forecast corresponding to the time ,in Expressed as The adaptive weights corresponding to the power generation modes are: Expressed as The model is based on The time corresponding to The corresponding forecast value is calculated based on the power generation and load demand of each power generation mode. Indicated as set The environmental correction factor corresponding to the moment is used to predict the power generation and load demand in the future set time period through historical power generation and load demand data.

[0033] It should be noted that the power generation contribution refers to the proportion of power generation contributed by each power generation method in a specified time period to the total power generation. In a specified microgrid, different power generation methods will generate different amounts of electricity in different time periods, such as solar energy, wind energy and energy storage systems. By analyzing historical data, the relative contribution of power generation of various power generation methods within a certain time range is calculated. For example, solar power generation accounts for 40% of the total power generation, wind power generation accounts for 30%, energy storage systems account for 20%, and other power generation methods account for 10%. Then the contribution of each power generation method is 40%, 30%, 20% and 10% respectively.

[0034] It should also be noted that the process of setting the power generation contribution threshold is based on the performance of different power generation methods in the microgrid in historical data. First, by analyzing the power generation data of each power generation method in the set historical time period, the contribution ratio to the total power generation in each time period is calculated. The results reflect the relative importance of various power generation methods. Then, according to the operation requirements, system stability and optimization goals of the microgrid, a minimum contribution threshold is determined. If the contribution of a certain power generation method in the historical data exceeds or is equal to the set power generation contribution threshold, then the power generation method is considered to be an important power generation source and is worthy of being included in the time series prediction model. Assuming that in the past period of time, the average historical contribution of wind energy at different times is 60%, the contribution of photovoltaics is 30%, and the contribution of load demand is 10%, then the power generation contribution threshold is set to 30%, which means that wind energy and photovoltaics will be taken into account in the model, but load demand will not.

[0035] It should also be noted that, for example, when there are two power generation modes, wind power generation and solar power generation, in the specified microgrid At this moment, the wind power generation is 150 kW, the solar power generation is 200 kW, and the load demand is 120 kW. The adaptive weight values ​​corresponding to wind power generation, solar power generation, and load demand are 0.4, 0.3, and 0.3, respectively. The model analysis shows that the predicted value of wind power generation is 160 kW, the predicted value of solar power generation is 210 kW, and the predicted value of load demand is 130 kW. If it is set to 50 kilowatts, the contribution of wind power generation is 64 kilowatts, the contribution of solar power generation is 63 kilowatts, and the contribution of load demand is 39 kilowatts, then the total power generation forecast is 216 kilowatts. The RNN model is a recurrent neural network. The calculation process involves multiple levels of neural network training and data processing. It is an existing model calculation process and will not be elaborated here.

[0036] It should also be noted that the adaptive weight is set according to the importance of the data source or its impact on the power generation of the microgrid. The setting of the adaptive weight is based on the contribution of each power generation mode and load demand to the total power generation. The weight values ​​of wind power, photovoltaic power and load demand are adjusted according to the actual effects of historical data and prediction models. Set to 50 kW, which means At this moment, the external environment or other factors that cannot be directly captured by the RNN prediction model have an impact of 50 kilowatts on the power generation. The setting process of the environmental correction factor is determined according to the value of the historical environmental correction factor, such as collecting environmental correction factor data in the historical time period, including influencing factors such as meteorological changes, equipment status, and the health status of the energy storage system, identifying the influencing factors that have an impact on the power generation of the microgrid under different time periods and conditions, thereby determining the setting value of the environmental correction factor.

[0037] In the process of time series data prediction, the embodiment of the present invention constructs a time series prediction model of renewable energy power generation and energy storage status, which is beneficial to improving the accuracy of future power generation and load demand prediction. It uses historical power generation and load demand data, as well as a set power generation contribution threshold, to screen valid data from the time series database, and optimizes the modeling through adaptive weights. It not only takes into account the influence of multiple data sources, but also can adjust the model in real time to better cope with the variable power demand and power generation conditions in the microgrid environment, greatly improving the accuracy of the prediction results, thereby providing a strong basis for scheduling decisions and avoiding the risks that may be caused by prediction deviations.

[0038] Step 3: Generate preliminary dispatch: Generate a preliminary dispatch strategy based on the predicted power generation and load demand within a set time period in the future to minimize the overall cost.

[0039] In a specific embodiment, the specific process of generating a preliminary scheduling strategy is as follows: based on the predicted power generation and load demand within a set time period in the future, the load demand balance constraint corresponding to the specified microgrid within the set time period is performed through a load demand balance model, so as to obtain the power demand that the specified microgrid needs to dispatch from the energy storage system, and optimize the power scheduling by obtaining the power generation cost and the energy storage discharge cost, so as to obtain a preliminary scheduling strategy that meets the requirements of minimizing the overall cost.

[0040] In a specific embodiment, the load demand balance model is used to specify the load demand balance constraint of the microgrid within the set time period. The specific process is as follows: The total power generation forecast corresponding to the time , obtained from the management platform of the specified microgrid The grid output power corresponding to the time and energy storage systems in The charging and discharging power corresponding to the moment ,pass The specified microgrid is calculated The load demand forecast value corresponding to the time , and then through the load demand balancing model: , the specified microgrid needs to be It corresponds to the power demand dispatched from the energy storage system at all times.

[0041] It should be noted that The specific application process is as follows: is 10 kilowatts, is 8 kWh, the equation does not hold, indicating that the specified microgrid needs to be The corresponding power demand dispatched from the energy storage system at that moment is 10 kW minus 8 kW, which is 2 kW.

[0042] In a specific embodiment, the optimization of power dispatch by obtaining power generation cost and energy storage discharge cost is as follows: obtaining The unit electricity cost of the grid corresponding to the time Charging and discharging costs corresponding to the energy storage system , and then through the calculation formula: , get the minimum comprehensive cost corresponding to the specified microgrid ,in and Respectively expressed as The specified microgrid power generation and energy storage system dispatching power corresponding to the time are The power demand dispatched from the energy storage system at the time is recorded as ,but: , Indicates that it is used to solve The minimum value of .

[0043] It should be noted that The specific meaning of the formula is: each part corresponds to different cost items and scheduling quantities. Specifically, the formula is The total comprehensive cost is calculated from the time To time The cost at each moment is composed of two parts: one is the cost of using the power grid , which is the cost when the microgrid purchases electricity from the grid, and the other is the discharge or charging cost of the energy storage system , that is, the cost of the microgrid obtaining electricity from the energy storage system or storing electricity in the energy storage system when needed, and the grid cost and energy storage cost at each moment and the corresponding amount of electricity and Multiply them together to get the cost at that moment, and sum up the costs at all moments to get the total comprehensive cost for the entire period.

[0044] In the process of generating a preliminary dispatching strategy, the embodiment of the present invention optimizes power dispatching through a load demand balance model, which is beneficial to maximizing the economic benefits of the microgrid. According to the predicted power generation and load demand, combined with the power generation cost and energy storage discharge cost, a preliminary dispatching strategy that meets the minimum comprehensive cost is generated through optimization calculation, thereby achieving a reasonable allocation of resources, ensuring that power demand is met, optimizing energy costs, and ensuring the efficient operation of the microgrid.

[0045] Step 4: Preliminary dispatch strategy analysis: Based on the generated preliminary dispatch strategy, the deviation between the microgrid load and the predicted value is evaluated. When there is a deviation between the microgrid load and the predicted value, an abnormality assessment is performed by collecting the real-time operation data of the microgrid.

[0046] In a specific embodiment, the deviation evaluation of the microgrid load and the predicted value is performed as follows: obtaining the microgrid load from the management platform of the specified microgrid. The actual load demand of the power grid at any time is expressed as ,Will and The result of the difference calculation is recorded as ,Will Compared with the set deviation threshold, if If it is greater than the set deviation threshold, it indicates that there is a deviation between the microgrid load and the predicted value. Otherwise, it indicates that there is no deviation between the microgrid load and the predicted value.

[0047] In a specific embodiment, the abnormality assessment is performed by collecting real-time operation data of the microgrid. The specific process is as follows: obtaining the real-time operation data of the microgrid from the management platform of the specified microgrid. Total power corresponding to load demand at any time and The actual total power supply corresponding to the time , the actual total power supply Including the power of the power grid, the power of the energy storage system and the power provided by renewable energy, that is It is equal to the sum of the power of the power grid, the power of the energy storage system and the power provided by renewable energy, and then calculated by the formula: ,get The corresponding power loss rate when there is a deviation at any time .

[0048] Through the discriminant formula: ,get The corresponding fault frequency when there is a deviation at any time ,in It represents the number of each power device in the specified microgrid. , is a positive integer, is the indicator function, indicating the Power equipment in Power of the moment Is the power lower than the set threshold? ,when < At this time The value is ,on the contrary, The value is .

[0049] Will The corresponding power loss rate when there is a deviation at any time and The corresponding fault frequency when there is a deviation at any time The power loss rate threshold is set respectively and the set fault frequency threshold For comparison, Less than the set power loss rate threshold and When it is less than the set fault frequency threshold, it indicates that there is no need to make abnormal corrections to the preliminary scheduling strategy. Otherwise, it indicates that there is a need to make abnormal corrections to the preliminary scheduling strategy.

[0050] It should be noted that renewable energy includes wind power and solar energy, and the set power threshold is based on the corresponding power threshold used in the historical judgment process and the corresponding power threshold used in the model training process, as well as the set value obtained in combination with the engineer's opinions.

[0051] In the preliminary dispatching strategy analysis process, the embodiment of the present invention evaluates the deviation between the load and the predicted value, which is conducive to discovering potential problems in the dispatching and making corrections in time. By comparing with the set deviation threshold, the deviation between the microgrid load and the predicted value is identified in real time. If the deviation exceeds the set threshold, the system will issue an alarm in time and collect real-time operation data for abnormal evaluation, which provides a basis for adjusting and optimizing the dispatching strategy and effectively avoids the negative impact of the dispatching deviation on the stability and economic benefits of the microgrid.

[0052] Step 5: Real-time optimization feedback: According to the abnormal evaluation results of the preliminary scheduling strategy, make abnormal corrections to the preliminary scheduling strategy.

[0053] In a specific embodiment, the abnormal correction of the preliminary scheduling strategy is carried out, and the specific process is as follows: when there is a deviation between the microgrid load and the predicted value, the difference between the energy storage capacity and the power generation is compared, and the required supplementary power is calculated based on the actual power demand of the specified microgrid, and the scheduling optimization is performed through the load demand balance model and the minimum comprehensive cost calculation formula until there is no deviation between the microgrid load and the predicted value.

[0054] It should be noted that, for example, when there is a deviation between the microgrid load and the predicted value, the gap between the current energy storage capacity and the power generation capacity is first determined. If the energy storage capacity is not enough to make up for the gap between the load and the predicted value, the power is supplemented from the power grid or other power generation methods. Assuming that at time , the predicted total power generation is 150 kilowatts, but the actual load demand is 180 kilowatts, resulting in a load deviation of 30 kilowatts. At this time, the microgrid dispatches 30 kilowatts of electricity from the energy storage system or the power grid to make up for this gap. Through the load demand balancing model, first calculate the current remaining power of the energy storage system and the expected charging and discharging capacity. For example, the energy storage system currently has 20 kilowatts of remaining power, but can only discharge 15 kilowatts. The remaining 15 kilowatts of gap needs to be obtained from the power grid. At this time, it is calculated that the power supplemented by the power grid is 15 kilowatts, and the power supplemented by the energy storage system is 15 kilowatts. The capacity is 15 kilowatts. When optimizing the dispatch, it is necessary to consider minimizing the comprehensive cost, which involves the power grid's electricity purchase cost and the energy storage discharge cost. Assuming that the unit cost of the power grid's electricity purchase is 0.2 yuan per kilowatt-hour, and the discharge cost of the energy storage system is 0.15 yuan per kilowatt-hour, the goal of dispatch optimization is to calculate the minimum comprehensive cost based on these two costs, and to ensure that the load demand is balanced and the total cost reaches the minimum by adjusting the dispatch of the energy storage system and the power grid. The dispatch result is continuously adjusted until the deviation between the actual load of the microgrid and the predicted value reaches within the set range.

[0055] See also Figure 2 As shown, a microgrid source-grid-load-storage integrated optimization scheduling system includes the following modules: an intelligent data aggregation module, a time series data prediction module, a preliminary scheduling generation module, a preliminary scheduling strategy analysis module, a real-time optimization feedback module and a time series database.

[0056] The intelligent data aggregation module is connected to the time series database, the time series database is connected to the time series data prediction module, the time series data prediction module is connected to the preliminary scheduling generation module, the preliminary scheduling generation module is connected to the preliminary scheduling strategy analysis module, and the preliminary scheduling strategy analysis module is connected to the real-time optimization feedback module.

[0057] The intelligent data aggregation module is used to optimize and store various types of data sources collected in a specified microgrid into a time series database.

[0058] The time series data prediction module is used to build a time series prediction model for renewable energy power generation and energy storage status based on the data sources stored in the time series database, so as to predict the power generation and load demand in the future set time period through historical power generation and load demand data.

[0059] The preliminary scheduling generation module is used to generate a preliminary scheduling strategy based on the predicted power generation and load demand within a set time period in the future, so as to minimize the overall cost.

[0060] The preliminary dispatch strategy analysis module is used to evaluate the deviation between the microgrid load and the predicted value based on the generated preliminary dispatch strategy. When there is a deviation between the microgrid load and the predicted value, an abnormality assessment is performed by collecting the real-time operation data of the microgrid.

[0061] The real-time optimization feedback module is used to make abnormal corrections to the preliminary scheduling strategy based on the abnormal evaluation results of the preliminary scheduling strategy.

[0062] The time series database is used to store various types of data sources including timestamps, and also stores the historical power generation contribution of each power generation mode in the specified microgrid.

[0063] An embodiment of the present invention provides a microgrid source-grid-load-storage integrated optimization scheduling method and system. In the process of intelligent data aggregation, by optimizing the storage of various types of data sources in a time series database, it is beneficial to improve the efficiency and accuracy of data processing. By storing multiple data sources such as new energy power generation data, energy storage system data, load demand data and environmental data in a time series and uniformly formatting them into data sorted by timestamps, the integrity and consistency of the data are ensured, which helps to provide high-quality input data for subsequent prediction models, reduce redundancy and errors in data storage and processing, and improve the accuracy of subsequent decision-making.

[0064] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they shall all fall within the protection scope of the present invention.

Claims

1. A microgrid source-grid-load-storage integrated optimization scheduling method, characterized in that: include: Step 1: Intelligent data aggregation: In a specified microgrid, various types of data sources are collected and stored in a time series database in an optimized manner; Step 2: Time series data prediction: Based on the data sources stored in the time series database, a time series prediction model for renewable energy power generation and energy storage status is constructed to predict power generation and load demand in a future set time period based on historical power generation and load demand data; Step 3: Preliminary dispatch generation: Based on the predicted power generation and load demand in the future set time period, a preliminary dispatch strategy is generated to minimize the overall cost; Step 4: Preliminary dispatch strategy analysis: Based on the generated preliminary dispatch strategy, the deviation between the microgrid load and the predicted value is evaluated. When there is a deviation between the microgrid load and the predicted value, an abnormality assessment is performed by collecting the real-time operation data of the microgrid; Step 5: Real-time optimization feedback: According to the abnormal evaluation results of the preliminary scheduling strategy, make abnormal corrections to the preliminary scheduling strategy.

2. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 1, characterized in that: The process of optimizing and storing various types of data sources in a time series database is as follows: Various types of data sources include renewable energy power generation data, energy storage system data, load demand data and environmental data. Various types of data sources corresponding to the specified microgrid are collected through sensor equipment, system equipment itself and user-side interface, and the collected various types of data sources are stored in a time series manner, and then the various types of data sources are uniformly converted into a format sorted by timestamp, so that the various types of data sources containing each timestamp are stored in a time series database, and the various types of data sources stored in the time series database are preprocessed. When the data preprocessing is completed, it indicates that the collected various types of data sources have completed optimized storage.

3. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 2, characterized in that: The specific process of constructing the time series prediction model of new energy power generation and energy storage status is as follows: The historical power generation contribution of each power generation mode in the specified microgrid is obtained from the time series database and compared with the set power generation contribution threshold. If the historical power generation contribution of a power generation mode is greater than or equal to the set power generation contribution threshold, the power generation mode is included in the corresponding input variable of the time series prediction model, and the corresponding input variables included in the time series prediction model are The power generation at the time is recorded as , It is represented by the number corresponding to each power generation mode. , is a positive integer, and the power load corresponding to each user end is collected from the specified microgrid, which is recorded as the load demand , Represents the time index, , is a positive integer, when the prediction When the total power generation forecast value at the moment is obtained, the forecast model is used: ,get The total power generation forecast corresponding to the time ,in Expressed as The adaptive weights corresponding to the power generation modes are: Expressed as The model is based on The time corresponding to The corresponding forecast value is calculated based on the power generation and load demand of each power generation mode. Indicated as set The environmental correction factor corresponding to the moment is used to predict the power generation and load demand in the future set time period through historical power generation and load demand data.

4. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 3, characterized in that: The specific process of generating the preliminary scheduling strategy is as follows: According to the predicted power generation and load demand within a set time period in the future, the load demand balance constraint corresponding to the specified microgrid within the set time period is performed through the load demand balance model, so as to obtain the power demand that the specified microgrid needs to dispatch from the energy storage system, and optimize the power dispatch by obtaining the power generation cost and energy storage discharge cost, so as to obtain a preliminary dispatch strategy that meets the minimum comprehensive cost.

5. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 4, characterized in that: The load demand balance model is used to specify the load demand balance constraint of the microgrid within the set time period. The specific process is as follows: According to the specified microgrid The total power generation forecast corresponding to the time , obtained from the management platform of the specified microgrid The grid output power corresponding to the time and energy storage systems in The charging and discharging power corresponding to the moment ,pass The specified microgrid is calculated The load demand forecast value corresponding to the time , and then through the load demand balancing model: , the specified microgrid needs to be It corresponds to the power demand dispatched from the energy storage system at all times.

6. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 5, characterized in that: The specific process of optimizing power dispatch by obtaining power generation cost and energy storage discharge cost is as follows: Obtained from the management platform of the specified microgrid The unit electricity cost of the grid corresponding to the time Charging and discharging costs corresponding to the energy storage system , and then through the calculation formula: , get the minimum comprehensive cost corresponding to the specified microgrid ,in and Respectively expressed as The specified microgrid power generation and energy storage system dispatching power corresponding to the time are The power demand dispatched from the energy storage system at the time is recorded as ,but: , Indicates that it is used to solve The minimum value of .

7. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 6, characterized in that: The specific process of evaluating the deviation between the microgrid load and the predicted value is as follows: Obtained from the management platform of the specified microgrid The actual load demand of the power grid at any time is expressed as ,Will and The result of the difference calculation is recorded as ,Will Compared with the set deviation threshold, if If it is greater than the set deviation threshold, it indicates that there is a deviation between the microgrid load and the predicted value. Otherwise, it indicates that there is no deviation between the microgrid load and the predicted value.

8. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 7, characterized in that: The abnormality assessment is performed by collecting the real-time operation data of the microgrid. The specific process is as follows: Obtained from the management platform of the specified microgrid The total power corresponding to the load demand at any time and The actual total power supply corresponding to the time , the actual total power supply Including the power of the power grid, the power of the energy storage system and the power provided by renewable energy, that is It is equal to the sum of the power of the power grid, the power of the energy storage system and the power provided by renewable energy, and then calculated by the formula: ,get The corresponding power loss rate when there is a deviation at any time ; Through the discriminant formula: ,get The corresponding fault frequency when there is a deviation at any time ,in It represents the number of each power device in the specified microgrid. , is a positive integer, is the indicator function, indicating the Power equipment in Power of the moment Is the power lower than the set threshold? ,when < At this time The value is ,on the contrary, The value is ; Will The corresponding power loss rate when there is a deviation at any time and The corresponding fault frequency when there is a deviation at any time The power loss rate threshold is set respectively and the set fault frequency threshold For comparison, Less than the set power loss rate threshold and When it is less than the set fault frequency threshold, it indicates that there is no need to make abnormal corrections to the preliminary scheduling strategy. Otherwise, it indicates that there is a need to make abnormal corrections to the preliminary scheduling strategy.

9. A microgrid source-grid-load-storage integrated optimization scheduling method according to claim 8, characterized in that: The specific process of performing abnormal correction of the preliminary scheduling strategy is as follows: When there is a deviation between the microgrid load and the predicted value, the difference between the energy storage capacity and the power generation is compared, and the required amount of supplementary electricity is calculated based on the actual power demand of the specified microgrid. The scheduling optimization is performed through the load demand balance model and the minimum comprehensive cost calculation formula until there is no deviation between the microgrid load and the predicted value.

10. A microgrid source-grid-load-storage integrated optimization dispatching system that implements the microgrid source-grid-load-storage integrated optimization dispatching method according to any one of claims 1 to 9, characterized in that: Includes the following modules: Intelligent data aggregation module, used to optimize and store various types of data sources collected in a specified microgrid into a time series database; The time series data prediction module is used to build a time series prediction model for renewable energy power generation and energy storage status based on the data sources stored in the time series database, so as to predict the power generation and load demand in the future set time period through the historical power generation and load demand data; The preliminary dispatch generation module is used to generate a preliminary dispatch strategy based on the predicted power generation and load demand within a set time period in the future, so as to minimize the overall cost; The preliminary dispatch strategy analysis module is used to evaluate the deviation between the microgrid load and the predicted value according to the generated preliminary dispatch strategy. When there is a deviation between the microgrid load and the predicted value, an abnormality assessment is performed by collecting the real-time operation data of the microgrid; The real-time optimization feedback module is used to make abnormal corrections to the preliminary scheduling strategy based on the abnormal evaluation results of the preliminary scheduling strategy.

Images