Load management method and system for distributed energy storage virtual power plant
By collecting and integrating power grid load data, optimizing load prediction model using prediction neural algorithms and feedback mechanisms, dynamically adjusting the charging and discharging strategies of energy storage equipment, solving the problems of inaccurate prediction and inflexible scheduling in the load management of distributed energy storage virtual power plants. , improves the operating efficiency and stability of the power system and promotes the intelligent management of the power system.
Patent Information
- Application Number
- CN202410904913.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-11-26
AI Technical Summary
In load management, distributed energy storage virtual power plants have problems such as inaccurate load prediction, inflexible scheduling and lack of feedback optimization mechanism, which affects the operating efficiency and stability of the power system.
By collecting and integrating real-time grid load data, using prediction neural algorithms to predict loads, establishing a feedback mechanism to optimize prediction models, dynamically adjust the charging and discharging strategies of energy storage equipment, and flexibly manage them according to the load range.
It improves the accuracy and real-time load prediction, enhances the utilization efficiency of energy storage equipment, balances the grid load, improves the stability and reliability of the power grid, and promotes the intelligent management of the power system and the energy utilization efficiency.
Smart Images

Figure CN119029899A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of load management systems, and specifically to a load management method and system for a distributed energy storage virtual power plant. Background Art
[0002] With the continuous growth of energy demand and the gradual popularization of renewable energy, the power system is facing more and more challenges. The traditional centralized power generation model has many problems in meeting power demand, such as energy waste, environmental pollution, unstable power supply, etc. In order to solve these problems, distributed energy storage virtual power plant technology came into being and became one of the important directions for the transformation and upgrading of the power system.
[0003] In terms of the development of related technologies, with the rapid development of intelligent technology and new energy technology, distributed energy storage virtual power plant technology has gradually become a research hotspot. This technology connects distributed energy storage devices (such as batteries, supercapacitors, etc.) with the power system to achieve the management and optimization of the power grid load. By collecting power grid load data, predicting load change trends, and adjusting the charging and discharging strategies of energy storage devices based on the prediction results, the operating efficiency of the power system can be effectively improved, energy waste can be reduced, power supply quality can be improved, energy costs can be reduced, and the use of clean energy can be promoted, thereby realizing the intelligent and sustainable development of the power system.
[0004] However, despite the many advantages of distributed energy storage virtual power plant technology, there are still some shortcomings in practical applications. First, the existing load management methods often lack the ability to accurately predict changes in grid load, which makes it difficult to accurately adjust the charging and discharging strategies of energy storage equipment, affecting the operating efficiency and stability of the power system. Secondly, traditional load management methods often lack feedback mechanisms, and are unable to promptly discover and correct errors in the prediction model, making it difficult to optimize and continuously improve the load prediction model. In addition, existing load management systems often have limited responsiveness to external environmental factors, making it difficult to effectively regulate external factors, affecting the accuracy and flexibility of load management. Summary of the invention
[0005] In view of the above-mentioned problems, the present invention is proposed.
[0006] Therefore, the technical problem solved by the present invention is: how to solve the technical problems of inaccurate load forecasting, inflexible scheduling and lack of feedback optimization mechanism in load management of distributed energy storage virtual power plants.
[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a load management method for a distributed energy storage virtual power plant, which comprises the following steps:
[0008] Collect and integrate grid load data; input the integrated grid load data into the load forecasting model to forecast load data; perform load management based on the forecasted grid load.
[0009] As a preferred solution of the load management method of the distributed energy storage virtual power plant described in the present invention, wherein: the real-time power grid load data includes power grid load data, weather data and economic indicators;
[0010] The weather data includes temperature and humidity.
[0011] The economic indicators include production unit output, distributed energy capacity, and distributed energy output.
[0012] As a preferred solution of the load management method of the distributed energy storage virtual power plant described in the present invention, the integration includes cleaning the collected data, removing erroneous, duplicate and incomplete records, unifying the time format and numerical units in different data sources, and filling in missing data.
[0013] As a preferred solution of the load management method of the distributed energy storage virtual power plant described in the present invention, the predicted load data includes a load prediction model that automatically learns from real-time power grid load data based on a predictive neural algorithm, identifies the power grid load, and establishes a feedback mechanism to optimize the load prediction model.
[0014] The prediction neural algorithm is expressed as,
[0015]
[0016] Where L(t) represents the predicted grid load at time t, T(s) represents the temperature data at time s, H(s) represents the humidity data at time s, α and β represent the parameters that adjust the impact of temperature and humidity on the load, and P i (s) represents the seasonal factor at time s, γ i It represents the parameter that adjusts the influence of various influencing factors on the load, D j (t) represents the economic index, N represents the number of seasonal factors, M represents the number of economic indicators, Δ is a normal number, |C j | represents a constant related to an economic indicator.
[0017] When L(t) is less than the low load threshold L low When L(t) is greater than or equal to the low load threshold L low Less than or equal to the high load threshold L high When L(t) is greater than the high load threshold L high, it indicates that the predicted grid load is in the high load range.
[0018] As a preferred solution of the load management method of the distributed energy storage virtual power plant described in the present invention, the load management is to adjust the charging and discharging strategy of the virtual power plant, including further comparing L(t) with L when in a low load range. low The size of L(t) is less than or equal to When the load is extremely low, the discharge activity of the energy storage device is completely stopped. If the power grid has excess energy, the charging speed of the energy storage device is increased, and the low electricity price during the low load period is used to actively charge the energy storage device. The low load period is used to repair and inspect the energy storage device. When the load is relatively low, the energy storage device is allowed to discharge, and the charging plan is adjusted according to the real-time load data. When the energy storage device detects a rapid increase in load, it quickly switches to high discharge mode.
[0019] When in the medium load range, the grid load changes are monitored in real time, the output of the energy storage equipment is automatically adjusted according to the increase or decrease in load, and the charging and discharging plans are adjusted using real-time or predicted electricity price information, increasing charging when the electricity price is low and increasing discharging when the electricity price is high.
[0020] When in the high load range, further compare L(t) with L high When L(t) is greater than L high Less than or equal to 105% L high When L(t) is greater than 105% L high When it is close to the high load range, the overall operation status of the power grid is monitored, power flow and potential bottlenecks are tracked in real time, and the advanced power management system is used to optimize the scheduling of renewable energy and traditional power generation. The discharge rate of energy storage equipment is gradually increased. When it is in the extremely high load range, the energy storage equipment is discharged at maximum capacity, and additional backup power generation resources are called upon to temporarily adjust the electricity price.
[0021] As a preferred scheme of the load management method of the distributed energy storage virtual power plant described in the present invention, wherein: the establishment of the feedback mechanism includes comparing the actually observed load data with the prediction results to obtain the accuracy, and based on the obtained accuracy, performing feedback loop adjustment on the load prediction model.
[0022] The accuracy is expressed as,
[0023]
[0024] Where n represents the number of samples, e i represents the forecast error of the i-th time period, wi Represents the weighted coefficient.
[0025] The feedback loop adjustment of the load forecasting model includes when the accuracy is greater than or equal to the accuracy θ, it indicates that the load forecasting model is in the expected effect and no adjustment is required; when the accuracy is less than the accuracy θ, it indicates that the load forecasting model is not ideal and a reverse loop adjustment is performed.
[0026] The reverse cycle adjustment includes optimizing the parameters in the load forecasting model, and the optimization includes determining the adjustment range of the parameters, and performing experiments using the selected parameter optimization algorithm within the adjustment range of the parameters, and determining the parameters α, β and γ in the load forecasting model based on different accuracy rates after the experiments. i .
[0027] Another object of the present invention is to provide a load management system for a distributed energy storage virtual power plant, which can collect and analyze power grid load data in real time, dynamically adjust the charging and discharging strategies of energy storage equipment, and introduce a feedback mechanism to optimize the load forecasting model, thereby solving the problems of insufficient load forecasting accuracy and weak ability to respond to external changes in the prior art.
[0028] In order to solve the above technical problems, the present invention provides the following technical solutions: a load management system of a distributed energy storage virtual power plant, including a data acquisition and integration module, a load prediction model module, and an energy management module.
[0029] The data acquisition and integration module is used to acquire and integrate real-time power grid load data.
[0030] The load prediction model module is used to input the integrated power grid load data into the load prediction model to predict the power grid load.
[0031] The energy management module performs load management based on the predicted grid load.
[0032] As a preferred solution for the load management system of the distributed energy storage virtual power plant described in the present invention, the data acquisition and integration module is used to collect and integrate real-time power grid load data.
[0033] The real-time power grid load data includes power grid load data, weather data and economic indicators.
[0034] The weather data includes temperature and humidity.
[0035] The economic indicators include production unit output, distributed energy capacity, and distributed energy output.
[0036] The integration includes cleaning the collected data, removing erroneous, duplicate and incomplete records, unifying the time format and numerical units in different data sources, and filling in missing data.
[0037] The load forecasting model module includes a load forecasting model that automatically learns from real-time power grid load data based on a predictive neural algorithm and identifies the power grid load, while establishing a feedback mechanism to optimize the load forecasting model.
[0038] The prediction neural algorithm is expressed as,
[0039]
[0040] Where L(t) represents the predicted grid load at time t, T(s) represents the temperature data at time s, H(s) represents the humidity data at time s, α and β represent the parameters that adjust the impact of temperature and humidity on the load, and P i (s) represents the seasonal factor at time s, γ i It represents the parameter that adjusts the influence of various influencing factors on the load, D j (t) represents the economic index, N represents the number of seasonal factors, M represents the number of economic indicators, Δ is a normal number, |C j | represents a constant related to an economic indicator.
[0041] When L(t) is less than the low load threshold L low When L(t) is greater than or equal to the low load threshold L low Less than or equal to the high load threshold L high When L(t) is greater than the high load threshold L high , it indicates that the predicted grid load is in the high load range.
[0042] The energy management module includes further comparing L(t) with L when in a low load range. low The size of L(t) is less than or equal to When the load is extremely low, the discharge activity of the energy storage device is completely stopped. If the power grid has excess energy, the charging speed of the energy storage device is increased, and the low electricity price during the low load period is used to actively charge the energy storage device. The low load period is used to repair and inspect the energy storage device. When the load is relatively low, the energy storage device is allowed to discharge, and the charging plan is adjusted according to the real-time load data. When the energy storage device detects a rapid increase in load, it quickly switches to high discharge mode.
[0043] When in the medium load range, the grid load changes are monitored in real time, the output of the energy storage equipment is automatically adjusted according to the increase or decrease in load, and the charging and discharging plans are adjusted using real-time or predicted electricity price information, increasing charging when the electricity price is low and increasing discharging when the electricity price is high.
[0044] When in the high load range, further compare L(t) with L high When L(t) is greater than L high Less than or equal to 105% L high When L(t) is greater than 105% L high When it is close to the high load range, the overall operation status of the power grid is monitored, power flow and potential bottlenecks are tracked in real time, and the advanced power management system is used to optimize the scheduling of renewable energy and traditional power generation. The discharge rate of energy storage equipment is gradually increased. When it is in the extremely high load range, the energy storage equipment is discharged at maximum capacity, and additional backup power generation resources are called upon to temporarily adjust the electricity price.
[0045] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the load management method of the distributed energy storage virtual power plant as described above when executing the computer program.
[0046] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the load management method of the distributed energy storage virtual power plant as described above.
[0047] Beneficial effects of the present invention: The present invention improves the accuracy and real-time performance of load forecasting by collecting and integrating real-time grid load data, thereby making the charging and discharging strategy of the energy storage device more accurate and efficient. Secondly, dynamically adjusting the charging and discharging strategy of the energy storage device not only improves the efficiency of energy use, but also helps to balance the grid load and enhance the stability and reliability of the grid. In addition, the introduced feedback mechanism can continuously optimize the prediction model, adapt to changes in the external environment and internal operating conditions, and ensure the long-term and efficient operation of the system. Overall, these effects jointly promote the intelligent management of the power system, improve energy utilization efficiency, and reduce operating costs, providing strong support for the sustainable development of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0049] Figure 1 An overall flow chart of the load management method for a distributed energy storage virtual power plant provided in the first embodiment of the present invention.
[0050] Figure 2 An overall framework diagram of a load management system for a distributed energy storage virtual power plant provided in accordance with the second embodiment of the present invention. DETAILED DESCRIPTION
[0051] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
[0052] Example 1
[0053] Reference Figure 1 , is an embodiment of the present invention, and provides a load management method for a distributed energy storage virtual power plant, characterized in that:
[0054] S1: Collect and integrate real-time grid load data.
[0055] Real-time grid load data includes grid load data, weather data and economic indicators.
[0056] Weather data includes temperature and humidity.
[0057] Economic indicators include production unit output, distributed energy capacity, and distributed energy output.
[0058] Integration includes cleaning the collected data, removing erroneous, duplicate and incomplete records, unifying the time format and numerical units in different data sources, and filling in missing data.
[0059] Furthermore, by cleaning, unifying the format and filling missing data of the collected power grid load data, the accuracy and completeness of the data are ensured, and the accuracy of subsequent load forecasting is improved. Increased data availability: The integrated data can be used in each step, avoiding information islands and improving data utilization efficiency.
[0060] S2: Input the integrated grid load data into the load forecasting model to forecast the load data.
[0061] The predicted load data includes a load forecasting model that automatically learns from real-time grid load data based on a predictive neural algorithm and identifies the grid load. At the same time, a feedback mechanism is established to optimize the load forecasting model.
[0062] The prediction neural algorithm is expressed as,
[0063]
[0064] Where L(t) represents the predicted grid load at time t, T(s) represents the temperature data at time s, H(s) represents the humidity data at time s, α and β represent the parameters that adjust the impact of temperature and humidity on the load, and P i (s) represents the seasonal factor at time s, γ i It represents the parameter that adjusts the influence of various influencing factors on the load, D j (t) represents the economic index, N represents the number of seasonal factors, M represents the number of economic indicators, Δ is a normal number, |C j | represents a constant related to an economic indicator.
[0065] When L(t) is less than the low load threshold L low When L(t) is greater than or equal to the low load threshold L low Less than or equal to the high load threshold L high When L(t) is greater than the high load threshold L high , it indicates that the predicted grid load is in the high load range.
[0066] Furthermore, by using historical load data, weather data, economic indicators and other information, and through predictive neural algorithms and other methods, accurate prediction of future grid loads can be achieved, providing an important basis for the operation and dispatch of power systems. The ability to capture load fluctuation trends in a timely manner helps to formulate reasonable dispatch plans and ensure the stable operation of the power grid.
[0067] S3: Load management based on predicted grid load.
[0068] Load management is to adjust the charging and discharging strategy of the energy storage device, including further comparing L(t) with L when in the low load range. low The size of L(t) is less than or equal to When the load is extremely low, the discharge activity of the energy storage device is completely stopped. If the power grid has excess energy, the charging speed of the energy storage device is increased, and the low electricity price during the low load period is used to actively charge the energy storage device. The low load period is used to repair and inspect the energy storage device. When the load is relatively low, the energy storage device is allowed to discharge, and the charging plan is adjusted according to the real-time load data. When the energy storage device detects a rapid increase in load, it quickly switches to high discharge mode.
[0069] When in the medium load range, the grid load changes are monitored in real time, the output of the energy storage equipment is automatically adjusted according to the increase or decrease in load, and the charging and discharging plans are adjusted using real-time or predicted electricity price information, increasing charging when the electricity price is low and increasing discharging when the electricity price is high.
[0070] When in the high load range, further compare L(t) with L high When L(t) is greater than L high Less than or equal to 105% L high When L(t) is greater than 105% L high When it is close to the high load range, the overall operation status of the power grid is monitored, power flow and potential bottlenecks are tracked in real time, and the advanced power management system is used to optimize the scheduling of renewable energy and traditional power generation. The discharge rate of the energy storage equipment is gradually increased. When it is in the extremely high load range, the energy storage equipment discharges at maximum capacity, and additional backup power generation resources are called upon at the same time. The electricity price is temporarily adjusted to reduce consumers' electricity use through economic incentives.
[0071] To further add, why 105% is chosen as the threshold for the high load range:
[0072] Safety buffer: set at L high The 5% high load threshold provides a safety buffer, ensuring that there is enough warning and preparation time to take action when the grid approaches its design limit. This helps prevent potential power system overload or failure.
[0073] Response time and flexibility: Grid operators need time to respond to load changes. When forecasted load approaches or slightly exceeds design capacity (e.g., 105%), this small excess allows operators time to mobilize reserve resources or adjust power flows without immediately risking system instability.
[0074] Practical operational considerations: Load management and planning of power grids usually take into account that a small overload may not have a serious impact on the system in a short period of time. Therefore, 105% can be regarded as a practical and practical operating limit to ensure that the load capacity of the power grid is maximized without affecting system stability.
[0075] Why 50% is chosen as the threshold for the low load interval:
[0076] Energy efficiency and cost management: When grid load is significantly lower than normal (e.g., below L low When the power grid is overloaded (over 50%), it indicates that there is excess generation capacity in the grid. In this case, increasing the charging activity of energy storage equipment can effectively utilize this excess power while reducing the operation of generators, reducing operating costs and fuel consumption.
[0077] Equipment maintenance and life extension: Reducing the operation of certain power generation facilities in the grid during low load periods can extend the life of these equipment and reduce maintenance requirements. At the same time, this is also an ideal time for scheduled maintenance and upgrades of the power system because the reliance on electricity is relatively low at this time.
[0078] The need for stable operation: Keeping the grid operating above a load level of at least 50% helps avoid power system stability issues such as voltage instability or frequency anomalies caused by underload.
[0079] Furthermore, according to the predicted grid load, the charging and discharging strategies of energy storage equipment can be flexibly adjusted to enable it to respond quickly and accurately to load changes and ensure the stability and security of the power system. By optimizing the charging and discharging strategies, the low electricity prices during low-load periods can be fully utilized for charging, reducing energy waste and lowering energy costs.
[0080] Establishing a feedback mechanism includes comparing the actual observed load data with the forecast results to obtain the accuracy, and based on the obtained accuracy, making feedback loop adjustments to the load forecasting model.
[0081] The accuracy is expressed as,
[0082]
[0083] Where n represents the number of samples, e i represents the forecast error of the i-th time period, w i Represents the weighted coefficient.
[0084] Feedback loop adjustment of the load forecasting model includes when the accuracy is greater than or equal to the accuracy θ, it indicates that the load forecasting model is in the expected effect and no adjustment is needed; when the accuracy is less than the accuracy θ, it indicates that the load forecasting model is not ideal and a reverse loop adjustment is performed.
[0085] The reverse cycle adjustment includes optimizing the parameters in the load forecasting model, wherein the optimization includes determining the adjustment range of the parameters, performing experiments using the selected parameter optimization algorithm within the adjustment range of the parameters, and determining the parameters α, β and γ in the load forecasting model based on different accuracy rates after the experiments. i .
[0086] Furthermore, by comparing actual data with forecast results in real time, the load forecasting model is continuously optimized, the accuracy and stability of the forecast are improved, and more reliable decision support is provided for power grid management and operation. The load forecasting model is adjusted through feedback loops to continuously improve system performance, adapt to the dynamic changes of the power system, and improve the intelligence level and response capabilities of the system.
[0087] We have proposed a load management method for a distributed energy storage virtual power plant. By collecting grid load data and integrating it into a database, we use a load forecasting model to accurately forecast the grid load, and dynamically adjust the charging and discharging strategy of the energy storage device based on the forecast results. At the same time, we establish a feedback mechanism to continuously optimize the load forecasting model and continuously improve the accuracy and stability of load forecasting. Compared with traditional methods, our invention has higher load forecasting accuracy, more flexible load management strategies, and stronger adaptability, which can effectively improve the operating efficiency and stability of the power system and promote the development and utilization of clean energy.
[0088] Example 2
[0089] Reference Figure 2 , which is an embodiment of the present invention, provides a system for a load management method of a distributed energy storage virtual power plant. The load management system of the distributed energy storage virtual power plant includes a data acquisition and integration module, a load prediction model module, and an energy management module.
[0090] The data acquisition and integration module is used to collect and integrate real-time power grid load data.
[0091] The load forecasting model module is used to input the integrated power grid load data into the load forecasting model to forecast the power grid load.
[0092] The energy management module performs load management based on the predicted grid load.
[0093] The data acquisition and integration module is used to collect and integrate real-time power grid load data.
[0094] Real-time grid load data includes grid load data, weather data and economic indicators.
[0095] Weather data includes temperature and humidity.
[0096] Economic indicators include production unit output, distributed energy capacity, and distributed energy output.
[0097] Integration includes cleaning the collected data, removing erroneous, duplicate and incomplete records, unifying the time format and numerical units in different data sources, and filling in missing data.
[0098] The load forecasting model module includes a load forecasting model that automatically learns from real-time power grid load data based on a predictive neural algorithm and identifies the power grid, while establishing a feedback mechanism to optimize the load forecasting model.
[0099] The prediction neural algorithm is expressed as,
[0100]
[0101] Where L(t) represents the predicted grid load at time t, T(s) represents the temperature data at time s, H(s) represents the humidity data at time s, α and β represent the parameters that adjust the impact of temperature and humidity on the load, and P i (s) represents the seasonal factor at time s, γ i It represents the parameter that adjusts the influence of various influencing factors on the load, D j (t) represents the economic index, N represents the number of seasonal factors, M represents the number of economic indicators, Δ is a normal number, |C j | represents a constant related to an economic indicator.
[0102] When L(t) is less than the low load threshold L low When L(t) is greater than or equal to the low load threshold L low Less than or equal to the high load threshold L high When L(t) is greater than the high load threshold L high , it indicates that the predicted grid load is in the high load range.
[0103] The energy management module includes further comparing L(t) with L when in the low load range. low The size of L(t) is less than or equal to When the load is extremely low, the discharge activity of the energy storage device is completely stopped. If the power grid has excess energy, the charging speed of the energy storage device is increased, and the low electricity price during the low load period is used to actively charge the energy storage device. The low load period is used to repair and inspect the energy storage device. When the load is relatively low, the energy storage device is allowed to discharge, and the charging plan is adjusted according to the real-time load data. When the energy storage device detects a rapid increase in load, it quickly switches to high discharge mode.
[0104] When in the medium load range, the grid load changes are monitored in real time, the output of the energy storage equipment is automatically adjusted according to the increase or decrease in load, and the charging and discharging plans are adjusted using real-time or predicted electricity price information, increasing charging when the electricity price is low and increasing discharging when the electricity price is high.
[0105] When in the high load range, further compare L(t) with L high When L(t) is greater than L high Less than or equal to 105% L high When L(t) is greater than 105% L highWhen it is close to the high load range, the overall operation status of the power grid is monitored, power flow and potential bottlenecks are tracked in real time, and the advanced power management system is used to optimize the scheduling of renewable energy and traditional power generation. The discharge rate of the energy storage equipment is gradually increased. When it is in the extremely high load range, the energy storage equipment discharges at maximum capacity, and additional backup power generation resources are called upon at the same time. The electricity price is temporarily adjusted to reduce consumers' electricity use through economic incentives.
[0106] The feedback mechanism module compares the actual observed load data with the forecast results to obtain the accuracy, and based on the obtained accuracy, performs feedback loop adjustments on the load forecasting model.
[0107] The accuracy is expressed as,
[0108]
[0109] Where n represents the number of samples, e i represents the forecast error of the i-th time period, w i Represents the weighted coefficient.
[0110] Feedback loop adjustment of the load forecasting model includes when the accuracy is greater than or equal to the accuracy θ, it indicates that the load forecasting model is in the expected effect and no adjustment is needed; when the accuracy is less than the accuracy θ, it indicates that the load forecasting model is not ideal and a reverse loop adjustment is performed.
[0111] The reverse cycle adjustment includes optimizing the parameters in the load forecasting model, wherein the optimization includes determining the adjustment range of the parameters, performing experiments using the selected parameter optimization algorithm within the adjustment range of the parameters, and determining the parameters α, β and γ in the load forecasting model based on different accuracy rates after the experiments. i .
[0112] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.
[0113] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.
[0114] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0115] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0116] Example 3
[0117] In this embodiment, in order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments. We designed two experimental groups: distributed energy storage virtual power plant load management method (experimental group A) and traditional method (experimental group B). Each experimental group will conduct simulation experiments on the same load data set. The experiment will be evaluated from the following aspects:
[0118] Load forecast accuracy: The root mean square error (RMSE) is used to evaluate the accuracy of load forecasting. The smaller the RMSE value, the better the forecasting effect.
[0119] Energy storage equipment utilization rate: The charging / discharging frequency and capacity utilization rate of the energy storage equipment are used as evaluation indicators. The higher the utilization rate, the higher the utilization efficiency of the energy storage equipment.
[0120] Economic analysis: Considering the costs and benefits of the two methods under different electricity prices, their economic feasibility is evaluated through comprehensive cost and benefit indicators.
[0121] Table 1 Experimental effect comparison chart
[0122] Experimental Group Load forecast accuracy Energy storage equipment utilization Economic Analysis A(this method) 0.05 80% Higher B (Traditional method) 0.1 60% generally
[0123] In summary, load forecasting accuracy: Experimental group A uses a load forecasting model based on a predictive neural algorithm, which has a smaller RMSE value than the traditional method, indicating a higher forecasting accuracy.
[0124] Energy storage equipment utilization: Experimental group A can use energy storage equipment more flexibly and achieve higher utilization rate by dynamically adjusting the charging and discharging strategies, while the fixed strategy of the traditional method limits the utilization efficiency of energy storage equipment.
[0125] Economic analysis: Considering the improvement of load forecasting accuracy and energy storage equipment utilization, experimental group A performs better in terms of economy and has higher cost-effectiveness.
[0126] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A load management method for a distributed energy storage virtual power plant, characterized in that: include: Collect and integrate real-time grid load data; Inputting the integrated power grid load data into the load forecasting model to forecast the power grid load; Load management based on predicted grid load.
2. The load management method of a distributed energy storage virtual power plant according to claim 1, characterized in that: The real-time power grid load data includes power grid load data, weather data and economic indicators; The weather data includes temperature and humidity; The economic indicators include production unit output, distributed energy capacity, and distributed energy output.
3. The load management method of a distributed energy storage virtual power plant according to claim 2, characterized in that: The integration includes cleaning the collected data, removing erroneous, duplicate and incomplete records, unifying the time format and numerical units in different data sources, and filling in missing data.
4. The load management method of a distributed energy storage virtual power plant according to claim 3, characterized in that: The predicted load data includes a load prediction model that automatically learns from real-time power grid load data based on a predictive neural algorithm and identifies the power grid load, while establishing a feedback mechanism to optimize the load prediction model; The prediction neural algorithm is expressed as, Where L(t) represents the predicted grid load at time t, T(s) represents the temperature data at time s, H(s) represents the humidity data at time s, α and β represent the parameters that adjust the impact of temperature and humidity on the load, and P i (s) represents the seasonal factor at time s, γ i It represents the parameter that adjusts the influence of various influencing factors on the load, D j (t) represents the economic index, N represents the number of seasonal factors, M represents the number of economic indicators, Δ is a normal number, |C j | represents a constant related to an economic indicator; When L(t) is less than the low load threshold L low When L(t) is greater than or equal to the low load threshold L low Less than or equal to the high load threshold L high When L(t) is greater than the high load threshold L high , it indicates that the predicted grid load is in the high load range.
5. The load management method of a distributed energy storage virtual power plant according to claim 4, characterized in that: The load management is to adjust the charging and discharging strategy of the virtual power plant, including further comparing L(t) with L when in a low load range. low The size of L(t) is less than or equal to When the load is too low, it is considered as an extremely low load situation, otherwise it is considered as a relatively low load situation. When the load is too low, the discharge activity of the energy storage device is completely stopped. If the power grid has excess energy, the charging speed of the energy storage device is increased, and the low electricity price during the low load period is used to actively charge. The energy storage device is repaired and inspected during the low load period. When the load is relatively low, the energy storage device is allowed to discharge, and the charging plan is adjusted according to the real-time data of the load. When the energy storage device detects a rapid increase in load, it quickly switches to a high discharge mode. When in the medium load range, the load changes of the power grid are monitored in real time, the output of the energy storage equipment is automatically adjusted according to the increase or decrease of the load, and the charging and discharging plans are adjusted using the real-time or predicted electricity price information, increasing charging when the electricity price is low and increasing discharging when the electricity price is high; When in the high load range, further compare L(t) with L high When L(t) is greater than L high Less than or equal to 105% L high When L(t) is greater than 105% L high When it is close to the high load range, the overall operation status of the power grid is monitored, power flow and potential bottlenecks are tracked in real time, and the advanced power management system is used to optimize the scheduling of renewable energy and traditional power generation. The discharge rate of energy storage equipment is gradually increased. When it is in the extremely high load range, the energy storage equipment is discharged at maximum capacity, and additional backup power generation resources are called upon to temporarily adjust the electricity price.
6. The load management method of a distributed energy storage virtual power plant according to claim 5, characterized in that: The establishment of the feedback mechanism includes comparing the actually observed load data with the forecast results to obtain the accuracy, and based on the obtained accuracy, performing feedback loop adjustment on the load forecasting model; The accuracy is expressed as, Where n represents the number of samples, e i represents the forecast error of the i-th time period, w i represents the weighted coefficient; The feedback loop adjustment of the load forecasting model includes when the accuracy rate is greater than or equal to the accuracy rate θ, it indicates that the load forecasting model is in the expected effect and no adjustment is required; when the accuracy rate is less than the accuracy rate θ, it indicates that the load forecasting model is not ideal and a reverse loop adjustment is performed; The reverse cycle adjustment includes optimizing the parameters in the load forecasting model, and the optimization includes determining the adjustment range of the parameters, and performing experiments using the selected parameter optimization algorithm within the adjustment range of the parameters, and determining the parameters α, β and γ in the load forecasting model based on different accuracy rates after the experiments. i .
7. A load management system for a distributed energy storage virtual power plant, characterized in that: It includes data collection and integration module, load forecasting model module and energy management module; The data acquisition and integration module is used to collect and integrate real-time power grid load data; The load prediction model module is used to input the integrated power grid load data into the load prediction model to predict the power grid load; The energy management module performs load management based on the predicted grid load.
8. A load management system using the distributed energy storage virtual power plant as claimed in claim 7, characterized in that: The data acquisition and integration module is used to collect and integrate real-time power grid load data; The real-time power grid load data includes power grid load data, weather data and economic indicators; The weather data includes temperature and humidity; The economic indicators include production unit output, distributed energy capacity, and distributed energy output; The integration includes cleaning the collected data, removing erroneous, duplicate and incomplete records, unifying the time format and numerical units in different data sources, and filling in missing data; The load forecasting model module includes a load forecasting model that automatically learns from real-time power grid load data based on a predictive neural algorithm and identifies the power grid load, while establishing a feedback mechanism to optimize the load forecasting model; The prediction neural algorithm is expressed as, Where L(t) represents the predicted grid load at time t, T(s) represents the temperature data at time s, H(s) represents the humidity data at time s, α and β represent the parameters that adjust the impact of temperature and humidity on the load, and P i (s) represents the seasonal factor at time s, γ i It represents the parameter that adjusts the influence of various influencing factors on the load, D j (t) represents the economic index, N represents the number of seasonal factors, M represents the number of economic indicators, Δ is a normal number, |C j | represents a constant related to an economic indicator; When L(t) is less than the low load threshold L low When L(t) is greater than or equal to the low load threshold L low Less than or equal to the high load threshold L high When L(t) is greater than the high load threshold L high When , it means that the predicted grid load is in the high load range; The energy management module includes further comparing L(t) with L when in a low load range. low The size of L(t) is less than or equal to When the load is too low, it is considered as an extremely low load situation, otherwise it is considered as a relatively low load situation. When the load is too low, the discharge activity of the energy storage device is completely stopped. If the power grid has excess energy, the charging speed of the energy storage device is increased, and the low electricity price during the low load period is used to actively charge. The energy storage device is repaired and inspected during the low load period. When the load is relatively low, the energy storage device is allowed to discharge, and the charging plan is adjusted according to the real-time data of the load. When the energy storage device detects a rapid increase in load, it quickly switches to a high discharge mode. When in the medium load range, the load changes of the power grid are monitored in real time, the output of the energy storage equipment is automatically adjusted according to the increase or decrease of the load, and the charging and discharging plans are adjusted using the real-time or predicted electricity price information, increasing charging when the electricity price is low and increasing discharging when the electricity price is high; When in the high load range, further compare L(t) with L high When L(t) is greater than L high Less than or equal to 105% L high When L(t) is greater than 105% L high When it is close to the high load range, the overall operation status of the power grid is monitored, power flow and potential bottlenecks are tracked in real time, and the advanced power management system is used to optimize the scheduling of renewable energy and traditional power generation. The discharge rate of energy storage equipment is gradually increased. When it is in the extremely high load range, the energy storage equipment is discharged at maximum capacity, and additional backup power generation resources are called upon to temporarily adjust the electricity price.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the load management method of the distributed energy storage virtual power plant described in any one of claims 1 to 6 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the load management method of the distributed energy storage virtual power plant described in any one of claims 1 to 6 are implemented.
Citation Information
Cited By
Distributed household energy storage all-in-one machine charging and discharging management control system
CN119382216A
Distributed energy storage converter optimization method and system based on big data analysis
CN119921357A
A Distributed Energy Storage Inverter Optimization Method and System Based on Big Data Analysis
CN119921357B
