Expressway micro-grid dispatching method and system, terminal and storage medium

Abstract

The invention relates to the technical field of micro-grids, and particularly provides a highway micro-grid dispatching method and system, a terminal and a storage medium, and the method comprises the steps: monitoring the power generation data and power utilization data of each micro-grid, and generating a power generation data sequence and a power utilization data sequence of the micro-grids; inputting the power generation data sequence and the power utilization data sequence of the micro-grid into a pre-trained LSTM model to obtain a predicted power generation amount and a predicted power utilization amount of the micro-grid; establishing a scheduling decision model based on a difference value between the predicted generating capacity and the predicted power consumption of the micro-grid; and optimizing the scheduling scheme of the scheduling decision model by using a reinforcement learning algorithm to obtain an optimal scheduling scheme. According to the invention, the prediction capability of the LSTM and the optimization capability of the reinforcement learning algorithm are combined, the precise scheduling of the highway microgrid is realized, the safety and stability of the plastic-coated highway are ensured, and the energy consumption is reduced to the minimum.

Description

Technical Field

[0001] The present invention belongs to the technical field of microgrids, and in particular relates to a highway microgrid dispatching method, system, terminal and storage medium. Background Art

[0002] As an important part of the distributed power system, microgrids can effectively integrate a variety of renewable energy sources, realize local power generation and local consumption, and are an important technical means to achieve zero-carbon highways. However, highway microgrids have the following characteristics and challenges: Coexistence of AC and DC modes: Highway microgrids have both DC power supply systems (such as solar photovoltaic power generation systems and energy storage systems) and AC power supply systems (such as wind power generation systems and traditional power generation equipment). How to achieve efficient power dispatch and energy management in the case of coexistence of AC and DC modes is a complex issue.

[0003] Multiple voltage levels: There are different voltage levels in highway microgrids, including high voltage, medium voltage and low voltage. These power equipment and systems of different voltage levels need to be coordinated and dispatched to ensure the safe and stable operation of the entire microgrid.

[0004] There are many sources of electricity with different characteristics: The sources of electricity in highway microgrids are diverse, including solar energy, wind energy, energy storage equipment, and electric vehicles. These energy sources have different characteristics. For example, solar energy and wind energy are intermittent and volatile, while energy storage equipment can provide smooth power output. How to dynamically configure and optimize the scheduling between these energy sources with different characteristics is the key to achieving efficient energy management.

[0005] Large fluctuations in load demand: Highway load demand is highly volatile and uncertain, and traffic flow and energy demand vary significantly in different time periods. How to dynamically configure and dispatch energy based on real-time load demand to ensure a balance between supply and demand is an important part of microgrid dispatch optimization.

[0006] Existing microgrid dispatching optimization technologies mainly rely on traditional power system dispatching methods, which are difficult to fully cope with the diverse demands and rapidly changing operating environment in the above-mentioned complex scenarios, resulting in low efficiency in power transmission, conversion and use, and difficulty in making full use of renewable energy. Summary of the invention

[0007] In view of the above-mentioned deficiencies in the prior art, the present invention provides a highway microgrid scheduling method, system, terminal and storage medium to solve the above-mentioned technical problems.

[0008] In a first aspect, the present invention provides a highway microgrid dispatching method, comprising: Monitor the power generation data and power consumption data of each microgrid, and generate a power generation data sequence and a power consumption data sequence of the microgrid; Input the power generation data sequence and power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid; Based on the difference between the predicted power generation and the predicted power consumption of the microgrid, a scheduling decision model is established; The reinforcement learning algorithm is used to optimize the scheduling plan of the scheduling decision model to obtain the best scheduling plan.

[0009] In an optional implementation, monitoring the power generation data and power consumption data of each microgrid to generate a power generation data sequence and a power consumption data sequence of the microgrid includes: Regularly collect power generation data and power consumption data of the microgrid, as well as energy storage data, wherein the energy storage data includes rated capacity and used capacity; The collected power generation data and power consumption data are sorted according to the collection time to obtain a power generation data queue and a power consumption data queue; A time window of a specified length is used to extract a power generation data sequence and a power consumption data sequence from the power generation data queue and the power consumption data queue respectively.

[0010] In an optional implementation, a scheduling decision model is established based on the difference between the predicted power generation and the predicted power consumption of the microgrid, including: If the difference is positive, the difference is recorded as the remaining power of the microgrid; if the difference is negative, the absolute value of the difference is recorded as the required power of the microgrid; Calculate the total remaining power and total required power of the microgrid on the target road section; If the total power demand exceeds the total remaining power, the microgrid is granted the authority to change the equipment working mode; Establish a scheduling decision model: ; Among them, Q(s,a) represents the value of taking action a in state s, and action a includes adjusting energy distribution and changing equipment working mode; r is the immediate reward, γ is the discount factor, s′ and a′ represent the next state and action respectively; the non-clean electricity and dispatch loss electricity used by the microgrid when taking action a in state s are calculated, and the immediate reward is generated based on the non-clean electricity and dispatch loss electricity.

[0011] In an optional implementation, a reinforcement learning algorithm is used to optimize the scheduling scheme of the scheduling decision model to obtain an optimal scheduling scheme, including: Generate actions using a greedy strategy; Substituting the action into the scheduling decision model to obtain the value of the action; The action is updated using an update function, the update function comprising: ; Among them, α is the learning rate. If Qnew(s,a)>Q(s,a), s′ and a′ are used; if Qnew(s,a)<Q(s,a), s and a are used and s′ and a′ are discarded.

[0012] In an optional embodiment, the method further comprises: The optimal scheduling scheme is sent to the scheduling agent of each microgrid, and the scheduling agent generates internal scheduling instructions of the microgrid based on the optimal scheduling scheme, and sends the internal scheduling instructions to the power generation agent, energy storage agent and load agent of the same microgrid.

[0013] In a second aspect, the present invention provides a highway microgrid dispatching system, comprising: A data monitoring module is used to monitor the power generation data and power consumption data of each microgrid and generate a power generation data sequence and a power consumption data sequence of the microgrid; The data prediction module is used to input the power generation data sequence and power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid; The dispatch modeling module is used to establish a dispatch decision model based on the difference between the predicted power generation and the predicted power consumption of the microgrid; The scheduling solution module is used to optimize the scheduling plan of the scheduling decision model using the reinforcement learning algorithm to obtain the best scheduling plan.

[0014] In an optional embodiment, the data monitoring includes: A data acquisition unit, used to regularly collect power generation data and power consumption data of the microgrid, as well as energy storage data, wherein the energy storage data includes rated capacity and used capacity; A data sorting unit, used to sort the collected power generation data and power consumption data according to the collection time, to obtain a power generation data queue and a power consumption data queue; The sequence interception unit is used to intercept the power generation data sequence and the power consumption data sequence from the power generation data queue and the power consumption data queue respectively using a time window of a specified length.

[0015] In an optional embodiment, the scheduling modeling module includes: A state judgment unit, configured to record the difference as the remaining power of the microgrid if the difference is positive; and record the absolute value of the difference as the required power of the microgrid if the difference is negative; A total calculation unit, used to calculate the total remaining power and total required power of the microgrid of the target section; An authority setting unit, used to grant the microgrid authority to change the device working mode if the total power demand exceeds the total remaining power; Model building unit, used to build scheduling decision model: ; Among them, Q(s,a) represents the value of taking action a in state s, and action a includes adjusting energy distribution and changing equipment working mode; r is the immediate reward, γ is the discount factor, s′ and a′ represent the next state and action respectively; the non-clean electricity and dispatch loss electricity used by the microgrid when taking action a in state s are calculated, and the immediate reward is generated based on the non-clean electricity and dispatch loss electricity.

[0016] In a third aspect, a terminal is provided, including: processor, memory, wherein: The memory is used to store computer programs. The processor is used to call and run the computer program from the memory, so that the terminal executes the above-mentioned terminal method.

[0017] According to a fourth aspect, a computer storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the methods described in the above aspects.

[0018] The beneficial effect of the present invention is that the highway microgrid dispatching method, system, terminal and storage medium provided by the present invention, combined with the prediction ability of LSTM and the optimization ability of reinforcement learning algorithm, realize the precise dispatching of highway microgrids, ensure the safety and stability of Baosu Highway and reduce energy consumption to a minimum.

[0019] In addition, the invention has a reliable design principle, a simple structure and a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention.

[0022] Figure 2 is another schematic flow chart of a method according to an embodiment of the present invention.

[0023] Figure 3 is a schematic block diagram of a system according to an embodiment of the present invention.

[0024] Figure 4 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 should fall within the scope of protection of the present invention.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0027] The highway microgrid dispatching method provided in the embodiment of the present invention is executed by a computer device, and accordingly, the highway microgrid dispatching system runs in the computer device.

[0028] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention. Figure 1 The execution subject may be a highway microgrid dispatching system. According to different requirements, the order of the steps in the flow chart may be changed, and some may be omitted.

[0029] like Figure 1 As shown, the method includes: Step 110, monitoring the power generation data and power consumption data of each microgrid, and generating a power generation data sequence and a power consumption data sequence of the microgrid; Step 120, inputting the power generation data sequence and the power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid; Step 130, establishing a scheduling decision model based on the difference between the predicted power generation and the predicted power consumption of the microgrid; Step 140, using a reinforcement learning algorithm to optimize the scheduling plan of the scheduling decision model to obtain the best scheduling plan.

[0030] To facilitate understanding of the present invention, the highway microgrid dispatching method provided by the present invention is further described below based on the principle of the highway microgrid dispatching method of the present invention and in combination with the process of dispatching the highway microgrid in the embodiment.

[0031] For details, please refer to Figure 2 ,The highway microgrid dispatching methods include: S1. Monitor the power generation data and power consumption data of each microgrid, and generate the power generation data sequence and power consumption data sequence of the microgrid.

[0032] Specifically, the power generation data and power consumption data of the microgrid, as well as the energy storage data, are collected regularly, and the energy storage data includes the rated capacity and the used capacity; the collected power generation data and power consumption data are sorted in chronological order according to the collection time to obtain the power generation data queue and the power consumption data queue; and the power generation data sequence and the power consumption data sequence are respectively intercepted from the power generation data queue and the power consumption data queue using a time window of a specified length.

[0033] Specifically, historical and real-time data are collected from each node of the highway microgrid, including but not limited to: voltage, current, power, energy type (such as solar energy, wind energy, energy storage equipment, etc.), and load conditions. The frequency of data collection should be high enough to capture the dynamic changes of the system, usually in seconds or minutes.

[0034] The collected data may contain noise and outliers, and data cleaning and normalization are required.

[0035] Data cleaning: remove obviously erroneous data points (such as values ​​outside a reasonable range) and fill in missing data (such as using the average of the data before and after).

[0036] Data normalization: In order to eliminate the dimensional differences of different data dimensions, the data are normalized to the range of [0, 1].

[0037] S2. Input the power generation data sequence and power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid.

[0038] The Long Short-Term Memory (LSTM) model is used to build the forecasting model. LSTM can effectively process time series data and capture the time dependency of electricity demand and renewable energy generation.

[0039] Temporal dependency refers to the regularity and trend of electricity demand and renewable energy generation over time. For example, electricity demand is usually higher during certain times of the day, such as morning and evening, and may be lower late at night. Similarly, solar power generation will show a clear daily periodicity as the sunshine changes, and wind power generation may be affected by the change of wind speed over time. In the process of forecasting electricity demand and renewable energy generation, LSTM will input time series data (such as historical electricity demand, solar and wind power generation records) into the network in segments. By processing this data step by step, LSTM will learn the time dependency pattern in the data.

[0040] LSTM model training: Use historical data to train the LSTM model.

[0041] x t : The time series data matrix input to the LSTM model contains variables such as historical electricity demand and renewable energy generation.

[0042] x t =[x 1,t ,x 2,t ,…,x n,t ] represents the observed value of the i-th variable at time t.

[0043] In this embodiment, the historical power generation data is used to train the first LSTM model, the historical power consumption data is used to train the second LSTM model, the trained first LSTM model is used to predict the power generation, and the trained second LSTM model is used to predict the power consumption.

[0044] S3. Based on the difference between the predicted power generation and the predicted power consumption of the microgrid, a scheduling decision model is established; the scheduling plan of the scheduling decision model is optimized using the reinforcement learning algorithm to obtain the optimal scheduling plan.

[0045] If the difference is positive, the difference is recorded as the remaining power of the microgrid; if the difference is negative, the absolute value of the difference is recorded as the required power of the microgrid; Calculate the total remaining power and total required power of the microgrid on the target road section; If the total power demand exceeds the total remaining power, the microgrid is granted the authority to change the equipment working mode; Establish a scheduling decision model: ; Among them, Q(s,a) represents the value of taking action a in state s, and action a includes adjusting energy distribution and changing equipment working mode; r is the immediate reward, γ is the discount factor, s′ and a′ represent the next state and action respectively; the non-clean electricity and dispatch loss electricity used by the microgrid when taking action a in state s are calculated, and the immediate reward is generated based on the non-clean electricity and dispatch loss electricity.

[0046] Among them, state s: The state s of the microgrid can include the total remaining power, the total required power, the current status of various energy sources (such as solar energy, wind energy, energy storage equipment, etc.), the working mode of the equipment (such as standby, low power, high power, etc.), time information (such as season, time period), etc.

[0047] Action a: In addition to adjusting energy distribution (such as increasing or decreasing the supply of a certain energy source) and changing the working mode of the equipment, action a can also include starting or stopping certain equipment, adjusting the charging and discharging strategies of energy storage equipment, etc.

[0048] Summarize the set of actions that can be executed in different states s of the microgrid, and update the actions by selecting new actions from the action set. For example, if the power demand of the microgrid is X, it can release the energy storage power, reduce the power of some equipment or shut down some equipment, and obtain some power from the outside.

[0049] The calculation method of the instant reward r includes: Non-clean electricity: The non-clean electricity (such as fossil fuel power generation) used by the microgrid after taking action a in the calculation state s. The more non-clean electricity, the lower the immediate reward r should be.

[0050] Dispatch loss: Dispatch loss is caused by supply-demand mismatch, equipment switching costs, energy conversion efficiency loss, etc. The greater the dispatch loss, the lower the immediate reward r should be.

[0051] Comprehensive reward: The instant reward r can be a function that comprehensively considers non-clean electricity and scheduling loss electricity, for example, calculated by weighted summation or nonlinear function.

[0052] The scheduling decision model refinement methods include: Q-learning or other reinforcement learning algorithms: Explicitly use Q-learning or other suitable reinforcement learning algorithms to train the scheduling decision model. Q(s,a) represents the expected long-term reward of taking action a in state s.

[0053] State transition: Explicitly describe the state transition function, that is, how to transfer from the current state s and action a to the next state s′.

[0054] Action Space and Selection: Detailed description of the size of the action space and the selection mechanism, such as using a greedy strategy (ε-greedy) to balance exploration and exploitation.

[0055] The optimization process includes: Generate an action using a greedy strategy; substitute the action into the scheduling decision model to obtain the value of the action; and update the action using an update function, wherein the update function includes: ; Among them, α is the learning rate. If Qnew(s,a)>Q(s,a), s′ and a′ are used; if Qnew(s,a)<Q(s,a), s and a are used and s′ and a′ are discarded.

[0056] Continue to iterate until the best scheduling solution is selected.

[0057] S4. Send the optimal scheduling plan to the scheduling agent of each microgrid. The scheduling agent generates internal scheduling instructions of the microgrid based on the optimal scheduling plan, and sends the internal scheduling instructions to the power generation agent, energy storage agent and load agent of the same microgrid.

[0058] A microgrid system consists of N agents A1, A2, ..., A N , where each agent is responsible for a specific function: Power Generation Agent A gen : Responsible for controlling and optimizing the power generation process of renewable energy such as solar energy and wind energy.

[0059] Energy Storage Agent A storage :Manage the charging and discharging process of energy storage equipment and optimize the storage and release of energy.

[0060] Load Agent A load : Responsible for managing the load of the microgrid to ensure the balance between supply and demand.

[0061] Scheduling Agent A schedule :Coordinate the scheduling between power generation, energy storage and load to optimize the distribution of energy.

[0062] Collaboration mechanism between agents Intelligent agents collaborate through information exchange and negotiation mechanisms to ensure the stable and efficient operation of the entire microgrid.

[0063] Assume that agent A i and A j The collaboration between them is achieved through the shared state variable S ij (t) and the execution control input U ij (t) to achieve, then its collaboration mechanism can be expressed by the following formula: ; in: U ij (t) is agent A i A j The control input applied depends on the state S shared by both ij (t).

[0064] f(S ij (t)) is a direct control strategy based on the current state.

[0065] g(S ik (t), S kj (t)) is the indirect influence of other related agents.

[0066] Design a reasonable reward function to encourage agents to work together and achieve global optimal scheduling: ; Among them, r i is the reward of the ith agent at time step t, P i (t) is the revenue, C i (t) is the cost.

[0067] Through the collaboration and competition of multi-agent systems, the global optimal scheduling strategy is achieved: ; in, is the optimal measurement of the ith agent in state s, is the expected value, is the discount factor, The reward of the ith agent at time step t.

[0068] In some embodiments, the highway microgrid dispatching system may include a plurality of functional modules composed of computer program segments. The computer programs of the various program segments in the highway microgrid dispatching system may be stored in a memory of a computer device and executed by at least one processor to execute (see Figure 1 Description) Functions of highway microgrid dispatch.

[0069] In this embodiment, the highway microgrid dispatching system can be divided into multiple functional modules according to the functions it performs, such as Figure 3As shown. The functional modules of the system 300 may include: a data monitoring module 310, a data prediction module 320, a scheduling modeling module 330 and a scheduling solution module 340. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0070] A data monitoring module is used to monitor the power generation data and power consumption data of each microgrid and generate a power generation data sequence and a power consumption data sequence of the microgrid; The data prediction module is used to input the power generation data sequence and power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid; The dispatch modeling module is used to establish a dispatch decision model based on the difference between the predicted power generation and the predicted power consumption of the microgrid; The scheduling solution module is used to optimize the scheduling plan of the scheduling decision model using the reinforcement learning algorithm to obtain the best scheduling plan.

[0071] Optionally, as an embodiment of the present invention, the data monitoring includes: A data acquisition unit, used to regularly collect power generation data and power consumption data of the microgrid, as well as energy storage data, wherein the energy storage data includes rated capacity and used capacity; A data sorting unit, used to sort the collected power generation data and power consumption data according to the collection time, to obtain a power generation data queue and a power consumption data queue; The sequence interception unit is used to intercept the power generation data sequence and the power consumption data sequence from the power generation data queue and the power consumption data queue respectively using a time window of a specified length.

[0072] Optionally, as an embodiment of the present invention, the scheduling modeling module includes: A state judgment unit, configured to record the difference as the remaining power of the microgrid if the difference is positive; and record the absolute value of the difference as the required power of the microgrid if the difference is negative; A total calculation unit, used to calculate the total remaining power and total required power of the microgrid of the target section; An authority setting unit, used to grant the microgrid authority to change the device working mode if the total power demand exceeds the total remaining power; Model building unit, used to build scheduling decision model: ; Among them, Q(s,a) represents the value of taking action a in state s, and action a includes adjusting energy distribution and changing equipment working mode; r is the immediate reward, γ is the discount factor, s′ and a′ represent the next state and action respectively; the non-clean electricity and dispatch loss electricity used by the microgrid when taking action a in state s are calculated, and the immediate reward is generated based on the non-clean electricity and dispatch loss electricity.

[0073] Figure 4 The present invention provides a schematic diagram of the structure of a terminal 400 provided in an embodiment of the present invention. The terminal 400 can be used to execute the highway microgrid scheduling method provided in an embodiment of the present invention.

[0074] The terminal 400 may include: a processor 410, a memory 420 and a communication unit 430. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention, and it may be a bus structure or a star structure, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0075] The memory 420 can be used to store the execution instructions of the processor 410, and the memory 420 can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory 420 are executed by the processor 410, the terminal 400 can perform some or all of the steps in the following method embodiments.

[0076] The processor 410 is the control center of the storage terminal, and uses various interfaces and lines to connect various parts of the entire electronic terminal. It runs or executes software programs and / or modules stored in the memory 420, and calls data stored in the memory to perform various functions of the electronic terminal and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor 410 can include only a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.

[0077] The communication unit 430 is used to establish a communication channel so that the storage terminal can communicate with other terminals, receive user data sent by other terminals or send user data to other terminals.

[0078] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, and when the program is executed, the program may include some or all of the steps in each embodiment provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0079] Therefore, the present invention combines the prediction ability of LSTM and the optimization ability of reinforcement learning algorithm to achieve precise scheduling of highway microgrids, ensure the safety and stability of Baosu Highway and minimize energy consumption. The technical effects that can be achieved by this embodiment can be found in the description above and will not be repeated here.

[0080] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes, including several instructions for enabling a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention.

[0081] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

[0082] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or modules, which can be electrical, mechanical or other forms.

[0083] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0085] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.

Claims

1. A highway microgrid dispatching method, characterized in that: include: Monitor the power generation data and power consumption data of each microgrid, and generate a power generation data sequence and a power consumption data sequence of the microgrid; Input the power generation data sequence and power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid; Based on the difference between the predicted power generation and the predicted power consumption of the microgrid, a scheduling decision model is established; The reinforcement learning algorithm is used to optimize the scheduling plan of the scheduling decision model to obtain the best scheduling plan.

2. The method according to claim 1, characterized in that Monitor the power generation data and power consumption data of each microgrid, and generate the power generation data sequence and power consumption data sequence of the microgrid, including: Regularly collect power generation data and power consumption data of the microgrid, as well as energy storage data, wherein the energy storage data includes rated capacity and used capacity; The collected power generation data and power consumption data are sorted according to the collection time to obtain a power generation data queue and a power consumption data queue; A time window of a specified length is used to extract a power generation data sequence and a power consumption data sequence from the power generation data queue and the power consumption data queue respectively.

3. The method according to claim 1, characterized in that Based on the difference between the predicted power generation and the predicted power consumption of the microgrid, a scheduling decision model is established, including: If the difference is positive, the difference is recorded as the remaining power of the microgrid; if the difference is negative, the absolute value of the difference is recorded as the required power of the microgrid; Calculate the total remaining power and total required power of the microgrid on the target road section; If the total power demand exceeds the total remaining power, the microgrid is granted the authority to change the equipment working mode; Establish a scheduling decision model: ； Among them, Q(s,a) represents the value of taking action a in state s, and action a includes adjusting energy distribution and changing equipment working mode; r is the immediate reward, γ is the discount factor, s′ and a′ represent the next state and action respectively; the non-clean electricity and dispatch loss electricity used by the microgrid when taking action a in state s are calculated, and the immediate reward is generated based on the non-clean electricity and dispatch loss electricity.

4. The method according to claim 3, characterized in that The scheduling scheme of the scheduling decision model is optimized using the reinforcement learning algorithm to obtain the best scheduling scheme, including: Generate actions using a greedy strategy; Substituting the action into the scheduling decision model to obtain the value of the action; The action is updated using an update function, the update function comprising: ； Among them, α is the learning rate. If Qnew(s,a)>Q(s,a), s′ and a′ are used; if Qnew(s,a)<Q(s,a), s and a are used and s′ and a′ are discarded.

5. The method according to claim 1, characterized in that The method further comprises: The optimal scheduling scheme is sent to the scheduling agent of each microgrid, and the scheduling agent generates internal scheduling instructions of the microgrid based on the optimal scheduling scheme, and sends the internal scheduling instructions to the power generation agent, energy storage agent and load agent of the same microgrid.

6. A highway microgrid dispatching system, characterized in that: include: A data monitoring module is used to monitor the power generation data and power consumption data of each microgrid and generate a power generation data sequence and a power consumption data sequence of the microgrid; The data prediction module is used to input the power generation data sequence and power consumption data sequence of the microgrid into the pre-trained LSTM model to obtain the predicted power generation and predicted power consumption of the microgrid; The scheduling modeling module is used to establish a scheduling decision model based on the difference between the predicted power generation and the predicted power consumption of the microgrid; The scheduling solution module is used to optimize the scheduling plan of the scheduling decision model using the reinforcement learning algorithm to obtain the best scheduling plan.

7. The system according to claim 6, characterized in that The data monitoring includes: A data acquisition unit, used to regularly collect power generation data and power consumption data of the microgrid, as well as energy storage data, wherein the energy storage data includes rated capacity and used capacity; A data sorting unit, used to sort the collected power generation data and power consumption data according to the collection time, to obtain a power generation data queue and a power consumption data queue; The sequence interception unit is used to intercept the power generation data sequence and the power consumption data sequence from the power generation data queue and the power consumption data queue respectively using a time window of a specified length.

8. The system according to claim 6, characterized in that The scheduling modeling module includes: A state judgment unit, configured to record the difference as the remaining power of the microgrid if the difference is positive; and record the absolute value of the difference as the required power of the microgrid if the difference is negative; A total calculation unit, used to calculate the total remaining power and total required power of the microgrid of the target section; An authority setting unit, used to grant the microgrid authority to change the device working mode if the total power demand exceeds the total remaining power; Model building unit, used to build scheduling decision model: ； Among them, Q(s,a) represents the value of taking action a in state s, and action a includes adjusting energy distribution and changing equipment working mode; r is the immediate reward, γ is the discount factor, s′ and a′ represent the next state and action respectively; the non-clean electricity and dispatch loss electricity used by the microgrid when taking action a in state s are calculated, and the immediate reward is generated based on the non-clean electricity and dispatch loss electricity.

9. A terminal, characterized in that: include: A memory, used for storing a highway microgrid dispatching program; A processor is used to implement the steps of the highway microgrid scheduling method as described in any one of claims 1 to 5 when executing the highway microgrid scheduling program.

10. A computer-readable storage medium storing a computer program, characterized in that: The readable storage medium stores a highway microgrid scheduling program, and when the highway microgrid scheduling program is executed by a processor, the steps of the highway microgrid scheduling method according to any one of claims 1 to 5 are implemented.

Images