Power system energy storage control method and device

Abstract

The invention discloses an electric power system energy storage control method and device, and the method comprises the steps: obtaining the operation data of a power grid, and predicting the load demand of the electric power system based on the operation data of the power grid; evaluating a health state of an energy storage unit in the power system based on the power grid operation data, and obtaining a health score of the energy storage unit; and matching a charging and discharging strategy based on the load demand, the health score and a preset learning algorithm, and performing energy storage control on the power system based on the charging and discharging strategy. The load demand of the power system is predicted through the power grid operation data, and the health score of the power system is obtained by evaluating the health state of the power system, so that the charging and discharging strategy is more matched with the actual power demand based on the load demand and the health score in combination with the preset learning algorithm; the most suitable charging and discharging strategy is effectively selected, flexible scheduling of the power system is achieved, and the charging and discharging efficiency and stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage control, and in particular to a method and device for controlling energy storage in a power system. Background Art

[0002] Energy storage devices refer to devices or systems that can store electrical energy and release it when needed. These devices can convert various forms of energy into electrical energy for use in the power system. The main functions of energy storage include balancing the volatility of renewable energy, providing backup power, improving grid stability, regulating loads, and improving the overall flexibility of the power system. As the proportion of new energy in the power system gradually increases, the traditional power system operation mode has been unable to adapt to the challenges brought about by the high proportion of new energy access.

[0003] Existing energy storage control methods lack intelligent management and cannot be flexibly dispatched and optimized according to the actual needs of the power system. They have low charging and discharging efficiency and low stability of energy storage control. Summary of the invention

[0004] The present invention provides a method and device for controlling energy storage in an electric power system, so as to improve charging and discharging efficiency and stability.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a power system energy storage control method, including:

[0006] Acquire power grid operation data, and predict the load demand of the power system based on the power grid operation data;

[0007] Evaluate the health status of the power system based on the power grid operation status to obtain a health score of the power system;

[0008] A charging and discharging strategy is matched based on the load demand and the health score, and energy storage control is performed on the power system based on the charging and discharging strategy.

[0009] The present invention predicts the load demand of the power system through the operation data of the power grid to determine the power demand in the future. At the same time, by evaluating the health status of the power system, the health score of the power system is obtained, so as to match the charging and discharging strategy based on the load demand and the health score. Through real-time demand load monitoring and health assessment combined with Q learning algorithm, the charging and discharging strategy is more closely matched with the actual power demand, and the most suitable charging and discharging strategy is effectively selected, potential problems of the power system are discovered in time, flexible scheduling of the power system is realized, and charging and discharging efficiency and stability are improved.

[0010] Furthermore, the power grid operation data includes power production data, power grid state data, operation data, and time data; and predicting the load demand of the power system based on the power grid operation data includes:

[0011] Obtain historical power grid operation data;

[0012] Building an initial load forecasting model based on a first machine learning algorithm, and training the initial load forecasting model based on the historical operation data of the power grid to obtain a load forecasting model;

[0013] The power production data, the grid status data, the operation data and the time data are input into the load forecasting model to obtain the load demand of the power system.

[0014] The load forecasting model constructed by the present invention through machine learning algorithm can learn and capture complex load change patterns from historical data to monitor the power demand of the power grid in real time, so as to provide adjustment basis for the energy storage control of the power system.

[0015] Furthermore, the grid operation data also includes equipment status data; and the step of evaluating the health status of the energy storage unit in the power system based on the grid operation data to obtain the health score of the energy storage unit includes:

[0016] Get historical device status data;

[0017] Building an initial health status assessment model based on a second machine learning algorithm, and training the initial health status assessment model based on the historical device status data to obtain a health status assessment model;

[0018] The device status data is input into the health status assessment model, and the health score of each energy storage unit in the power system is obtained to obtain the health score of the power system.

[0019] The present invention evaluates the health status of the power system based on the operating status of the power grid and obtains a health score, thereby evaluating the health status of each energy storage unit in the power system in real time, ensuring that the system is in a safe operating state at all times, promptly discovering and handling potential risks, reducing the possibility of accidents, and improving the safety and stability of the system.

[0020] Furthermore, the preset learning algorithm is a Q learning algorithm; matching the charging and discharging strategy based on the load demand, the health score and the preset learning algorithm, and performing energy storage control on the power system based on the charging and discharging strategy, includes:

[0021] Setting a state space and a reward function of a Q-learning algorithm based on the load demand and the health score;

[0022] Setting the action space of the Q learning algorithm based on a preset multi-level energy storage strategy;

[0023] The charging and discharging strategy is matched based on the state space, the reward function and the action space, and energy storage control is performed on the power system based on the charging and discharging strategy.

[0024] The present invention can dynamically adjust the charging and discharging strategy of the energy storage system according to the current load demand and equipment health status through the Q learning algorithm to achieve more efficient energy management. Through the intelligent scheduling of the Q learning algorithm, it can discharge during peak hours and charge during off-peak hours, smooth the load curve and reduce the pressure on the power grid.

[0025] Furthermore, the multi-level energy storage strategy is a control strategy for a multi-level energy storage layer, and the multi-level energy storage layer includes a first energy storage layer, a second energy storage layer, and a third energy storage layer; the energy storage control of the power system based on the charging and discharging strategy includes:

[0026] When the load demand is a short-term load fluctuation demand and the health score of the first energy storage layer is higher than a preset score threshold, performing energy storage control based on the first energy storage layer;

[0027] When the load demand is a short-term energy storage demand and the health score of the second energy storage layer is higher than a preset score threshold, performing energy storage control based on the second energy storage layer;

[0028] When the load demand is a long-term energy storage demand and the health score of the third energy storage layer is higher than a preset score threshold, energy storage control is performed based on the third energy storage layer.

[0029] In a second aspect, the present invention provides a power system energy storage control device, including: a load prediction module, a health assessment module and a strategy matching module;

[0030] The load forecasting module is used to obtain power grid operation data and forecast the load demand of the power system based on the power grid operation data;

[0031] The health assessment module is used to assess the health status of the power system based on the power grid operation status and obtain a health score of the power system;

[0032] The strategy matching module is used to match the charging and discharging strategy based on the load demand, the health score and a preset learning algorithm, and perform energy storage control on the power system based on the charging and discharging strategy.

[0033] The present invention predicts the load demand of the power system through the operation data of the power grid to determine the power demand in the future. At the same time, by evaluating the health status of the power system, the health score of the power system is obtained, so as to match the charging and discharging strategy based on the load demand and the health score. Through real-time demand load monitoring and health assessment combined with a preset learning algorithm, the charging and discharging strategy is more closely matched with the actual power demand, and the most suitable charging and discharging strategy is effectively selected to achieve flexible scheduling of the power system and improve charging and discharging efficiency and stability.

[0034] Furthermore, the power grid operation data includes power production data, power grid status data, operation data, and time data; the load forecasting module is used to:

[0035] Obtain historical power grid operation data;

[0036] Building an initial load forecasting model based on a first machine learning algorithm, and training the initial load forecasting model based on the historical operation data of the power grid to obtain a load forecasting model;

[0037] The power production data, the grid status data, the operation data and the time data are input into the load forecasting model to obtain the load demand of the power system.

[0038] Furthermore, the power grid operation data also includes equipment status data; the health assessment module is used to:

[0039] Get historical device status data;

[0040] Building an initial health status assessment model based on a second machine learning algorithm, and training the initial health status assessment model based on the historical device status data to obtain a health status assessment model;

[0041] The device status data is input into the health status assessment model, and the health score of each energy storage unit in the power system is obtained to obtain the health score of the power system.

[0042] Furthermore, the preset learning algorithm is a Q learning algorithm; the strategy matching module is used to:

[0043] Setting a state space and a reward function of a Q-learning algorithm based on the load demand and the health score;

[0044] Setting the action space of the Q learning algorithm based on a preset multi-level energy storage strategy;

[0045] The charging and discharging strategy is matched based on the state space, the reward function and the action space, and energy storage control is performed on the power system based on the charging and discharging strategy.

[0046] Furthermore, it also includes a multi-level energy storage layer, the multi-level energy storage layer includes a first energy storage layer, a second energy storage layer and a third energy storage layer; the multi-level energy storage strategy is a control strategy for the multi-level energy storage layer, and the strategy matching module performs energy storage control on the power system based on the charging and discharging strategy, including:

[0047] When the load demand is a short-term load fluctuation demand and the health score of the first energy storage layer is higher than a preset score threshold, performing energy storage control based on the first energy storage layer;

[0048] When the load demand is a short-term energy storage demand and the health score of the second energy storage layer is higher than a preset score threshold, performing energy storage control based on the second energy storage layer;

[0049] When the load demand is a long-term energy storage demand and the health score of the third energy storage layer is higher than a preset score threshold, energy storage control is performed based on the third energy storage layer. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of a flow chart of a power system energy storage control method provided in an embodiment of the present invention;

[0051] Figure 2 A schematic structural diagram of an energy storage control device for an electric power system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0053] The terms "first" and "second" and the like in the specification, claims and drawings of this application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0054] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] Example 1

[0056] See also Figure 1 , Figure 1 The present invention provides a method for controlling energy storage in a power system, which is as follows:

[0057] Step 101: Acquire power grid operation data, and predict the load demand of the power system based on the power grid operation data;

[0058] In this embodiment, the obtaining of power grid operation data includes:

[0059] The power grid operation data is acquired, wherein the power grid operation data includes power production data, power grid status data, operation data, time data and equipment status data.

[0060] In this embodiment, the power production data includes power generation data, environmental data and energy management data.

[0061] In this embodiment, by collecting power production data, the power production situation is monitored in real time to ensure the normal operation of the generator set. And by combining historical data and weather forecasts, the future power demand is predicted so as to reasonably dispatch power generation resources; at the same time, according to the power grid operation data, the abnormal situation of the generator set is discovered in time, and repair and maintenance are carried out to ensure the continuity and safety of power production.

[0062] In this embodiment, the grid state data includes grid frequency, grid voltage, grid current and grid kilometer factor.

[0063] In this embodiment, the state of the power grid is monitored in real time through the power grid state data, including voltage, current, frequency, etc., to ensure the stable operation of the power grid, timely discover the fault points in the power grid, quickly locate and handle them, and reduce power outage time.

[0064] In this embodiment, the operation data includes power generation cost, power generation revenue and maintenance data.

[0065] In this embodiment, the changing trend of electricity prices is analyzed through operation data, and power dispatch is optimized according to the operation data to reduce electricity costs and improve economic benefits.

[0066] In this embodiment, time series data is analyzed to discover the periodicity and trend of the data and predict future power demand and grid status.

[0067] In this embodiment, the device status data includes device operating status, device temperature, device pressure, device vibration, and device wear degree.

[0068] In this embodiment, the running status of the equipment is monitored in real time to ensure the normal operation of the equipment, and abnormal conditions of the equipment are discovered in time to perform fault diagnosis and repair to reduce equipment downtime.

[0069] In this embodiment, the power grid operation data includes power production data, power grid state data, operation data, and time data; and predicting the load demand of the power system based on the power grid operation data includes:

[0070] Obtain historical power grid operation data;

[0071] Building an initial load forecasting model based on a first machine learning algorithm, and training the initial load forecasting model based on the historical operation data of the power grid to obtain a load forecasting model;

[0072] The power production data, the grid status data, the operation data and the time data are input into the load forecasting model to obtain the load demand of the power system.

[0073] In this embodiment, the load forecasting model is trained based on the historical operation data of the power system, and its input is the operation data of the power system, and feature extraction is performed based on the operation data, so as to predict the real-time load demand of the power system based on the extracted features. The historical operation data includes historical power production data, historical power grid status data, historical operation data and historical time data.

[0074] In this embodiment, the first machine learning algorithm includes linear regression, decision tree, random forest, support vector machine (SVM), neural network, etc.

[0075] In this embodiment, the load forecasting model constructed by the first machine learning algorithm can learn and capture complex load change patterns from historical data to monitor the power demand of the power grid in real time, so as to provide an adjustment basis for the energy storage control of the power system.

[0076] Step 102: Evaluate the health status of the power system based on the power grid operation status to obtain a health score of the power system;

[0077] In this embodiment, the grid operation data also includes equipment status data; and evaluating the health status of the energy storage unit in the power system based on the grid operation data to obtain the health score of the energy storage unit includes:

[0078] Get historical device status data;

[0079] Building an initial health status assessment model based on a second machine learning algorithm, and training the initial health status assessment model based on the historical device status data to obtain a health status assessment model;

[0080] The device status data is input into the health status assessment model, and the health score of each energy storage unit in the power system is obtained to obtain the health score of the power system.

[0081] In this embodiment, historical device status data is collected, based on which the historical device status data includes operation data, maintenance records, and fault records of each energy storage unit in the power system.

[0082] In this embodiment, the second machine learning algorithm is a convolutional neural network algorithm. An initial health status assessment model is constructed based on the convolutional neural network, and the health status assessment model is trained based on historical equipment status data to obtain a trained health status assessment model.

[0083] In this embodiment, feature extraction is performed on the device status data based on the health status assessment model to obtain key features of each energy storage unit, such as temperature changes, pressure fluctuations, vibration frequency, etc., and a health assessment is performed on each energy storage unit based on the key features. The assessment results are discretized to obtain a health score.

[0084] In this embodiment, the health status of the power system is evaluated based on the operating status of the power grid to obtain a health score, thereby evaluating the health status of each energy storage unit in the power system in real time, ensuring that the system is in a safe operating state at all times, promptly discovering and handling potential risks, reducing the possibility of accidents, and improving the safety and stability of the system.

[0085] Step 103: Matching a charging and discharging strategy based on the load demand, the health score and a preset learning algorithm, and performing energy storage control on the power system based on the charging and discharging strategy.

[0086] In this embodiment, the preset learning algorithm is a Q learning algorithm; matching the charging and discharging strategy based on the load demand, the health score and the preset learning algorithm, and performing energy storage control on the power system based on the charging and discharging strategy include:

[0087] Setting a state space and a reward function of a Q-learning algorithm based on the load demand and the health score;

[0088] Setting the action space of the Q learning algorithm based on a preset multi-level energy storage strategy;

[0089] The charging and discharging strategy is matched based on the state space, the reward function and the action space, and energy storage control is performed on the power system based on the charging and discharging strategy.

[0090] In this embodiment, the multi-level energy storage strategy is a control strategy for a multi-level energy storage layer, and the multi-level energy storage layer includes a first energy storage layer, a second energy storage layer, and a third energy storage layer; the energy storage control of the power system based on the charging and discharging strategy includes:

[0091] When the load demand is a short-term load fluctuation demand and the health score of the first energy storage layer is higher than a preset score threshold, performing energy storage control based on the first energy storage layer;

[0092] When the load demand is a short-term energy storage demand and the health score of the second energy storage layer is higher than a preset score threshold, performing energy storage control based on the second energy storage layer;

[0093] When the load demand is a long-term energy storage demand and the health score of the third energy storage layer is higher than a preset score threshold, energy storage control is performed based on the third energy storage layer.

[0094] In this embodiment, the energy storage strategy includes a multi-level energy storage layer energy storage strategy.

[0095] In this embodiment, the energy storage strategies include a first energy storage layer energy storage strategy, a second energy storage layer energy storage strategy and a third energy storage layer energy storage strategy.

[0096] In this embodiment, the first energy storage layer is a fast response layer, which includes supercapacitors and flywheel energy storage systems with high power density energy storage devices; supercapacitors can provide instantaneous power support within milliseconds through their fast charging and discharging characteristics to cope with short-term fluctuations in new energy power generation.

[0097] In this embodiment, the second energy storage layer is a conventional response layer, which includes lithium-ion batteries and sodium-sulfur batteries, which are energy storage devices with high energy density and high efficiency; these devices are used to handle load changes from minutes to hours and balance electricity demand and supply in a short period of time.

[0098] In this embodiment, the third energy storage layer is a long-term energy storage layer, which includes energy storage devices such as compressed air energy storage and liquid flow batteries, which are used to handle daytime load differences and long-term electricity storage needs.

[0099] In this embodiment, the first energy storage layer, the second energy storage layer and the third energy storage layer are interconnected, and the energy storage devices of the three corresponding modules are interconnected through the power electronic interface and the control system to form a complete energy storage network. The first energy storage layer, the second energy storage layer and the third energy storage layer are coordinated and linked through data transmission and control signals. Data and control signals are transmitted through the Internet of Things platform, and each energy storage device maintains information exchange and collaborative work to optimize the operation of the overall system.

[0100] In this embodiment, the action space of the Q learning algorithm is set by the three-level energy storage layer, taking into account the independence and coordination of each layer of energy storage device. The action space includes the first energy storage layer, the second energy storage layer and the third energy storage layer, which can be expressed as: action = (first energy storage layer action, second energy storage layer action and third energy storage layer action).

[0101] In this embodiment, the state space and reward function of the Q learning algorithm are set based on the load demand and the health score.

[0102] In this embodiment, the state space includes load demand, health score, current time and energy storage state. The load demand is the load demand of the current power system, the health score is the health score of the power system, and the current time can be a segment of a day, such as peak hours, valley hours, etc. The energy storage state is the charge and discharge state of the current energy storage system, such as the remaining power, charge and discharge power, etc.

[0103] In this embodiment, the reward function includes load satisfaction, health status, energy efficiency and stability. Among them, load satisfaction means that if the load demand is met, a positive reward is given; otherwise, a negative reward is given. Health status means that if the action taken has a positive impact on the health status of the system, a positive reward is given; otherwise, a negative reward is given. Energy efficiency means the efficiency of charging and discharging, high efficiency gives positive rewards, and low efficiency gives negative rewards. Stability means that if the action taken can improve the stability of the system, a positive reward is given; otherwise, a negative reward is given.

[0104] In this embodiment, when the health score of a certain energy storage layer is lower than or equal to the score threshold, it indicates that the energy storage layer has serious fault loss, and therefore the energy storage layer needs to be isolated.

[0105] In this embodiment, the low health score of a certain energy storage layer is calculated by taking a weighted average of the health scores of the energy storage units in the energy storage layer.

[0106] In this embodiment, the Q learning algorithm can dynamically adjust the charging and discharging strategy of the energy storage system according to the current load demand and equipment health status to achieve more efficient energy management. Through the intelligent scheduling of the Q learning algorithm, discharge can be performed during peak hours and charge during off-peak hours, smoothing the load curve and reducing the pressure on the power grid. The charging and discharging strategy stipulates that different energy storage layers are selected based on different load demands for charging and discharging control, thereby ensuring sufficient power supply during high demand and effectively storing excess power during low demand.

[0107] In this embodiment, after the charging and discharging strategy is obtained, the charging and discharging strategy is adjusted based on a multi-objective optimization algorithm and influencing factors, and resource scheduling is performed on the energy storage device based on the adjusted charging and discharging strategy.

[0108] In this embodiment, an electricity price prediction model is constructed based on deep learning, and future electricity prices are predicted based on real-time electricity market data. Based on the prediction results, load forecasts, and the status of energy storage devices, a heuristic algorithm or a mixed integer linear programming (MILP) algorithm is used to optimize the charging and discharging strategy to achieve the best economic benefit.

[0109] In this embodiment, the multi-objective optimization algorithm is specifically:

[0110] min x J=α·C cost (x)+β·C emission (x)-γ·U user (x) (2)

[0111] Among them, Ccost(x) is the cost function, Cemi ssion(x) is the carbon emission function, Uuser(x) is the user utility function, α, β, γ are weight coefficients;

[0112] In this embodiment, the charging and discharging strategy is adjusted in real time according to the multi-objective algorithm to determine the optimal operating parameters of the energy storage system, thereby adjusting the scheduling strategy of the energy storage system in real time to avoid excessive use of energy storage equipment during peak electricity price periods and reduce electricity procurement costs.

[0113] In this embodiment, a distributed energy storage network is constructed based on the Internet of Things, and resource scheduling is performed on energy storage devices of each network node based on the distributed energy storage network.

[0114] Please refer to Figure 2 , Figure 2 A schematic diagram of a structure of a power system energy storage control device provided by an embodiment of the present invention includes: a load prediction module 201, a health assessment module 202 and a strategy matching module 203;

[0115] The load prediction module 201 is used to obtain power grid operation data and predict the load demand of the power system based on the power grid operation data;

[0116] The health assessment module 202 is used to assess the health status of the power system based on the power grid operation status to obtain a health score of the power system;

[0117] The strategy matching module 203 is used to match the charging and discharging strategy based on the load demand, the health score and a preset learning algorithm, and perform energy storage control on the power system based on the charging and discharging strategy.

[0118] In this embodiment, the power grid operation data includes power production data, power grid status data, operation data, and time data; the load forecasting module is used to:

[0119] Obtain historical power grid operation data;

[0120] Building an initial load forecasting model based on a first machine learning algorithm, and training the initial load forecasting model based on the historical operation data of the power grid to obtain a load forecasting model;

[0121] The power production data, the grid status data, the operation data and the time data are input into the load forecasting model to obtain the load demand of the power system.

[0122] In this embodiment, the power grid operation data also includes equipment status data; the health assessment module is used to:

[0123] Get historical device status data;

[0124] Building an initial health status assessment model based on a second machine learning algorithm, and training the initial health status assessment model based on the historical device status data to obtain a health status assessment model;

[0125] The device status data is input into the health status assessment model, and the health score of each energy storage unit in the power system is obtained to obtain the health score of the power system.

[0126] In this embodiment, the preset learning algorithm is a Q learning algorithm; the strategy matching module is used to:

[0127] Setting a state space and a reward function of a Q-learning algorithm based on the load demand and the health score;

[0128] Setting the action space of the Q learning algorithm based on a preset multi-level energy storage strategy;

[0129] The charging and discharging strategy is matched based on the state space, the reward function and the action space, and energy storage control is performed on the power system based on the charging and discharging strategy.

[0130] In this embodiment, a multi-level energy storage layer is further included, and the multi-level energy storage layer includes a first energy storage layer, a second energy storage layer, and a third energy storage layer; the multi-level energy storage strategy is a control strategy for the multi-level energy storage layer, and the strategy matching module performs energy storage control on the power system based on the charging and discharging strategy, including:

[0131] When the load demand is a short-term load fluctuation demand and the health score of the first energy storage layer is higher than a preset score threshold, performing energy storage control based on the first energy storage layer;

[0132] When the load demand is a short-term energy storage demand and the health score of the second energy storage layer is higher than a preset score threshold, performing energy storage control based on the second energy storage layer;

[0133] When the load demand is a long-term energy storage demand and the health score of the third energy storage layer is higher than a preset score threshold, energy storage control is performed based on the third energy storage layer.

[0134] In an embodiment of the present invention, a terminal device is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, the above-mentioned power system energy storage control method is implemented.

[0135] In an embodiment of the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored computer program. When the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned power system energy storage control method.

[0136] Exemplarily, the computer program may be divided into one or more modules, one or more modules are stored in a memory and executed by a processor to implement the present invention. One or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in a terminal device.

[0137] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The terminal device may include, but is not limited to, a processor, a memory, and a display. Those skilled in the art will appreciate that the above components are merely examples of the terminal device and do not constitute a limitation on the terminal device. The terminal device may include more or fewer components than the components, or may combine certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0138] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0139] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0140] Among them, if the module based on the processing of the multi-device access platform is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0141] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A power system energy storage control method, characterized in that: include: Acquire power grid operation data, and predict the load demand of the power system based on the power grid operation data; Evaluate the health status of the energy storage unit in the power system based on the power grid operation data to obtain a health score of the energy storage unit; A charging and discharging strategy is matched based on the load demand, the health score and a preset learning algorithm, and energy storage control is performed on the power system based on the charging and discharging strategy.

2. A power system energy storage control method according to claim 1, characterized in that: The power grid operation data includes power production data, power grid state data, operation data, and time data; and the load demand of the power system predicted based on the power grid operation data includes: Obtain historical power grid operation data; Building an initial load forecasting model based on a first machine learning algorithm, and training the initial load forecasting model based on the historical operation data of the power grid to obtain a load forecasting model; The power production data, the grid status data, the operation data and the time data are input into the load forecasting model to obtain the load demand of the power system.

3. A power system energy storage control method as claimed in claim 2, characterized in that: The grid operation data also includes equipment status data; the health status of the energy storage unit in the power system is evaluated based on the grid operation data to obtain the health score of the energy storage unit, including: Get historical device status data; Building an initial health status assessment model based on a second machine learning algorithm, and training the initial health status assessment model based on the historical device status data to obtain a health status assessment model; The device status data is input into the health status assessment model, and the health score of each energy storage unit in the power system is obtained to obtain the health score of the power system.

4. A power system energy storage control method as claimed in claim 3, characterized in that: The preset learning algorithm is a Q learning algorithm; The matching of the charging and discharging strategy based on the load demand, the health score and the preset learning algorithm, and performing energy storage control on the power system based on the charging and discharging strategy, includes: Setting a state space and a reward function of a Q-learning algorithm based on the load demand and the health score; Setting the action space of the Q learning algorithm based on a preset multi-level energy storage strategy; The charging and discharging strategy is matched based on the state space, the reward function and the action space, and energy storage control is performed on the power system based on the charging and discharging strategy.

5. A power system energy storage control method as claimed in claim 4, characterized in that: The multi-level energy storage strategy is a control strategy for a multi-level energy storage layer, wherein the multi-level energy storage layer includes a first energy storage layer, a second energy storage layer, and a third energy storage layer; the energy storage control of the power system based on the charging and discharging strategy includes: When the load demand is a short-term load fluctuation demand and the health score of the first energy storage layer is higher than a preset score threshold, performing energy storage control based on the first energy storage layer; When the load demand is a short-term energy storage demand and the health score of the second energy storage layer is higher than a preset score threshold, performing energy storage control based on the second energy storage layer; When the load demand is a long-term energy storage demand and the health score of the third energy storage layer is higher than a preset score threshold, energy storage control is performed based on the third energy storage layer.

6. A power system energy storage control device, characterized in that: include: Load forecasting module, health assessment module and strategy matching module; The load forecasting module is used to obtain power grid operation data and forecast the load demand of the power system based on the power grid operation data; The health assessment module is used to assess the health status of the power system based on the power grid operation status and obtain a health score of the power system; The strategy matching module is used to match the charging and discharging strategy based on the load demand, the health score and the Q learning algorithm, and perform energy storage control on the power system based on the charging and discharging strategy.

7. The power system energy storage control device according to claim 6, characterized in that: The power grid operation data includes power production data, power grid status data, operation data, and time data; the load forecasting module is used to: Obtain historical power grid operation data; Building an initial load forecasting model based on a first machine learning algorithm, and training the initial load forecasting model based on the historical operation data of the power grid to obtain a load forecasting model; The power production data, the grid status data, the operation data and the time data are input into the load forecasting model to obtain the load demand of the power system.

8. The power system energy storage control device according to claim 7, characterized in that: The power grid operation data also includes equipment status data; the health assessment module is used to: Get historical device status data; Building an initial health status assessment model based on a second machine learning algorithm, and training the initial health status assessment model based on the historical device status data to obtain a health status assessment model; The device status data is input into the health status assessment model, and the health score of each energy storage unit in the power system is obtained to obtain the health score of the power system.

9. The power system energy storage control device according to claim 8, characterized in that: The preset learning algorithm is a Q learning algorithm; the strategy matching module is used to: Setting a state space and a reward function of a Q-learning algorithm based on the load demand and the health score; Setting the action space of the Q learning algorithm based on a preset multi-level energy storage strategy; The charging and discharging strategy is matched based on the state space, the reward function and the action space, and energy storage control is performed on the power system based on the charging and discharging strategy.

10. The power system energy storage control device according to claim 9, characterized in that: It also includes a multi-level energy storage layer, the multi-level energy storage layer includes a first energy storage layer, a second energy storage layer and a third energy storage layer; the multi-level energy storage strategy is a control strategy for the multi-level energy storage layer, and the strategy matching module performs energy storage control on the power system based on the charging and discharging strategy, including: When the load demand is a short-term load fluctuation demand and the health score of the first energy storage layer is higher than a preset score threshold, performing energy storage control based on the first energy storage layer; When the load demand is a short-term energy storage demand and the health score of the second energy storage layer is higher than a preset score threshold, performing energy storage control based on the second energy storage layer; When the load demand is a long-term energy storage demand and the health score of the third energy storage layer is higher than a preset score threshold, energy storage control is performed based on the third energy storage layer.

Images