Intelligent regulation and control energy storage optimization method and system for distributed energy system

Abstract

The invention discloses an intelligent regulation and control energy storage optimization method and system for a distributed energy system, and the method comprises the steps: monitoring the dust distribution data and temperature data of the surface of a photovoltaic module in real time, constructing a power generation efficiency correction model, calculating the actual power generation attenuation, and outputting the corrected theoretical power generation; establishing a digital twinborn model in combination with energy storage operation data, and synchronizing to an energy storage system state and power grid parameters; constructing a multi-objective optimization model based on a digital twin model, and cooperatively distributing output strategies of each stored energy in peak regulation and standby power scenes; and when it is detected that the battery attenuation exceeds a first preset threshold value or the local power grid load is unbalanced, triggering a replanning algorithm to recalculate the energy storage capacity distribution and optimize the charge and discharge priority, and generating a final energy storage configuration scheme. The actual power generation attenuation can be accurately calculated by considering the influence of dust, temperature and the like, a reliable basis is provided for energy storage planning, the output strategies of each energy storage in different scenes can be cooperatively distributed, and the operation efficiency and stability of the energy storage system are improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy technologies, and particularly to an intelligent control energy storage optimization method and system for distributed energy systems. Background Art

[0002] In distributed energy systems, the effective application of energy storage technologies is crucial for improving energy utilization efficiency and ensuring the stable operation of the power grid. There are many deficiencies in the existing technologies for energy storage regulation in distributed energy systems. On the one hand, for common distributed energy generation devices such as photovoltaic modules, there is a lack of accurate monitoring and consideration of their operating environmental factors. The dust distribution on the surface of photovoltaic modules will block light irradiation and reduce the photoelectric conversion efficiency, and temperature changes will also significantly affect their power generation performance. However, existing technologies often ignore these effects, resulting in an inability to accurately evaluate the actual power generation capacity of photovoltaic modules, large deviations in power generation calculations, and difficulties in efficient energy storage planning and regulation.

[0003] On the other hand, in the operation management of energy storage systems, existing technologies cannot achieve accurate synchronization of the energy storage system status and grid parameters. When constructing an energy storage optimization model, it is difficult to comprehensively and accurately consider various complex factors, resulting in a lack of coordination and scientific nature in the output strategies of energy storage in different scenarios such as peak shaving and power backup, and the inability to fully utilize the effectiveness of energy storage devices. At the same time, when abnormal situations such as battery degradation or local grid load imbalance occur, existing technologies cannot timely and intelligently re-plan the energy storage capacity allocation and charge-discharge priorities, resulting in poor emergency response capabilities of the energy storage system and affecting the stability and reliability of the entire distributed energy system. Summary of the Invention

[0004] To solve at least one of the above-mentioned technical problems, the present invention provides an intelligent control energy storage optimization method and system for distributed energy systems.

[0005] In the first aspect, the present invention provides an intelligent control energy storage optimization method for distributed energy systems, and the method includes:

[0006] Real-time monitoring of the dust distribution data and temperature data on the surface of photovoltaic modules;

[0007] Based on the dust distribution data and temperature data, construct a power generation efficiency correction model, calculate the actual power generation attenuation according to the power generation efficiency correction model, and output the corrected theoretical power generation;

[0008] According to the corrected theoretical power generation and energy storage operation data, establish a digital twin model, synchronize it to the energy storage system status and grid parameters; based on the digital twin model, construct a multi-objective optimization model, and collaboratively allocate the output strategies of each energy storage in peak shaving and power backup scenarios;

[0009] When it is detected that the battery degradation exceeds the first preset threshold or the local grid load is unbalanced, the replanning algorithm is triggered to recalculate the energy storage capacity allocation and optimize the charge and discharge priorities, generating the final energy storage configuration plan.

[0010] Preferably, a power generation efficiency correction model is constructed based on the dust distribution data and temperature data, and the actual power generation degradation is calculated according to the power generation efficiency correction model, and the corrected theoretical power generation is output, including:

[0011] Collect the dust coverage rate, particle size distribution, and plate surface temperature gradient;

[0012] Based on the dust coverage rate, particle size distribution, and plate surface temperature gradient, a dust deposition-temperature field-power generation efficiency coupling model is constructed;

[0013] Based on the coupling model, the actual efficiency degradation rate of the photovoltaic module is dynamically calculated through the heat conduction equation and the light attenuation coefficient, and the corrected theoretical power generation is output.

[0014] Preferably, the construction method of the coupling model includes

[0015] ;

[0016] Among them, is the corrected efficiency, is the dust deposition thickness, is the temperature deviation, is the dust influence coefficient, and the value range is ; is the temperature influence coefficient, and the value range is .

[0017] Preferably, a digital twin model is established according to the corrected theoretical power generation and energy storage operation data, and synchronized to the energy storage system state and grid parameters; a multi-objective optimization model is constructed based on the digital twin model to collaboratively allocate the output strategies of each energy storage in peak shaving and power backup scenarios, including:

[0018] Based on the corrected theoretical power generation and energy storage system operation data, a digital twin body covering photovoltaic modules, energy storage batteries, and grid topologies is established, and the state of charge of the battery, line load, and environmental parameters are synchronized in real time;

[0019] Using the real-time data of the digital twin body, taking the distributed energy storage system as a game participant, designing a multi-objective optimization model based on the Shapley value, dynamically allocating the output strategies of each energy storage in peak shaving and power backup scenarios, and generating the initial energy storage configuration plan.

[0020] Preferably, the design of the multi-objective optimization model based on the Shapley value includes:

[0021] According to the real-time line load data provided by the digital twin, the output range of each energy storage system is constrained so that the voltage deviation of the power grid node does not exceed ±5%, and the line load rate is lower than 90%.

[0022] The peak shaving benefits of each energy storage system are allocated through the Shapley value to ensure that the individual benefits are not lower than the benefits when operating independently, and the total benefits satisfy superadditivity.

[0023] Preferably, the replanning algorithm is as follows:

[0024] According to the battery health state prediction model, calculate the capacity attenuation rate:

[0025] ;

[0026] Wherein, is the battery health state, is the number of charge and discharge cycles, is the attenuation coefficient, and the value range is ;

[0027] When is lower than the second preset threshold, trigger the capacity reallocation strategy, and transfer the load of the high-attenuation battery to the low-attenuation battery proportionally.

[0028] Preferably, the data synchronization frequency of the digital twin is not lower than 10 Hz, and local data processing is realized through the edge computing node to reduce the cloud transmission delay.

[0029] In a second aspect, the present invention also provides an intelligent regulation energy storage optimization system for a distributed energy system, and the system includes:

[0030] A data acquisition unit for real-time monitoring of the dust distribution data and temperature data on the surface of the photovoltaic module;

[0031] A power generation amount correction unit for constructing a power generation efficiency correction model based on the dust distribution data and temperature data, calculating the actual power generation amount attenuation according to the power generation efficiency correction model, and outputting the corrected theoretical power generation amount;

[0032] A multi-objective optimization unit for establishing a digital twin model according to the corrected theoretical power generation amount and the energy storage operation data, and synchronizing to the energy storage system state and the power grid parameters; constructing a multi-objective optimization model based on the digital twin model, and coordinating the output strategies of each energy storage in the peak shaving and power backup scenarios;

[0033] An energy storage configuration update unit for triggering the replanning algorithm to recalculate the energy storage capacity allocation and optimize the charge and discharge priority when detecting that the battery attenuation exceeds the first preset threshold or the local power grid load is unbalanced, and generating a final energy storage configuration plan.

[0034] In a third aspect, the present invention further provides an electronic device, including a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes the method according to the first aspect and any possible implementation manner thereof as described above.

[0035] In a fourth aspect, the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor is caused to execute the method according to the first aspect and any possible implementation manner thereof as described above.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] The intelligent control energy storage optimization method for a distributed energy system provided by the present invention effectively solves these technical problems. By real-time monitoring the dust distribution data and temperature data on the surface of photovoltaic modules and constructing a power generation efficiency correction model, it can accurately calculate the actual power generation attenuation and output a more practical corrected theoretical power generation amount, providing a reliable basis for energy storage planning. Based on this, a digital twin model is established to synchronize the energy storage system state and grid parameters, and a multi-objective optimization model is constructed to collaboratively allocate the output strategies of each energy storage in different scenarios, greatly improving the operation efficiency of the energy storage system. When it is detected that the battery attenuation exceeds the first preset threshold or the local grid load is unbalanced, a replanning algorithm is triggered to recalculate the energy storage capacity allocation and optimize the charge and discharge priorities, generating a final energy storage configuration plan to ensure that the energy storage system can still operate stably and efficiently under complex working conditions, enhancing the ability of the distributed energy system to cope with various changes and improving the stability, reliability and energy utilization efficiency of the entire system.

[0038] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the following will describe the drawings required to be used in the embodiments of the present invention or the background art.

[0040] The drawings here are incorporated into the specification and form a part of this specification. These drawings show embodiments that conform to the present disclosure and are used together with the specification to explain the technical solutions of the present disclosure.

[0041] Figure 1 It is a schematic flowchart of an intelligent control energy storage optimization method for a distributed energy system provided by an embodiment of the present invention;

[0042] Figure 2 is Figure 1 a schematic flow chart of the sub-steps of step S20 in

[0043] Figure 3 is Figure 1 a schematic flow chart of the sub-steps of step S30 in

[0044] Figure 4 a schematic structural diagram of an intelligent control energy storage optimization system for a distributed energy system provided by an embodiment of the present invention. Specific embodiments

[0045] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0046] Referring to "embodiments" herein means that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0047] Please refer to Figure 1 , Figure 1 is a schematic flow chart of an intelligent control energy storage optimization method for a distributed energy system provided by an embodiment of the present invention. As Figure 1 shown, the method includes:

[0048] S10. Real-time monitor the dust distribution data and temperature data on the surface of the photovoltaic module;

[0049] Dust on the surface of the photovoltaic module will block light, reducing the photoelectric conversion efficiency of the photovoltaic module. Moreover, different dust distribution conditions have different degrees of influence on the power generation efficiency. Temperature also significantly affects the power generation performance of the photovoltaic module. Too high or too low temperature will lead to a decrease in power generation efficiency. Only by obtaining the dust distribution data and temperature data in real time can basic data be provided for accurately evaluating the power generation efficiency of the photovoltaic module subsequently.

[0050] Therefore, in this embodiment, a high-resolution image sensor and a high-precision temperature sensor are installed on the surface of the photovoltaic module. The high-resolution image sensor periodically collects images of the surface of the photovoltaic module, and analyzes distribution data such as the coverage area and thickness of dust in the images through an image recognition algorithm; the high-precision temperature sensor monitors the surface temperature of the photovoltaic module in real time, and transmits the collected temperature data to the data processing center through a wireless communication module. Real-time and accurate monitoring of the key environmental factors affecting the power generation efficiency of the photovoltaic module is realized, providing reliable data support for the subsequent construction of a power generation efficiency correction model, making the calculation of power generation more in line with the actual situation, avoiding the deviation of power generation estimation caused by ignoring dust and temperature factors, and thus laying a foundation for the reasonable planning of the energy storage system.

[0051] S20. Construct a power generation efficiency correction model based on the dust distribution data and temperature data, calculate the actual power generation attenuation according to the power generation efficiency correction model, and output the corrected theoretical power generation;

[0052] Since the influence of dust and temperature on the power generation efficiency of the photovoltaic module is relatively complex, it is difficult to obtain accurate results through simple formulas. In this embodiment, historical operation data can be used, combined with the relationship between dust distribution data, temperature data and actual power generation, to train a power generation efficiency correction model through a machine learning algorithm (such as a neural network algorithm). Input the real-time monitored dust distribution data and temperature data into the model, and the model calculates the power generation efficiency of the photovoltaic module under the current working condition according to the learned relationship, compares it with the power generation efficiency under the standard working condition, obtains the actual power generation attenuation ratio, and then calculates and outputs the corrected theoretical power generation.

[0053] The data-driven power generation efficiency correction model can accurately quantify the influence of dust and temperature on the power generation efficiency of the photovoltaic module, and effectively improve the accuracy of theoretical power generation calculation. Compared with the traditional power generation calculation method that does not consider these factors, the corrected theoretical power generation can more truly reflect the actual power generation capacity of the photovoltaic module, which helps the energy storage system to more reasonably plan the charge and discharge strategies and improve the energy utilization efficiency.

[0054] S30. Establish a digital twin model according to the corrected theoretical power generation and energy storage operation data, and synchronize it to the energy storage system state and grid parameters; construct a multi-objective optimization model based on the digital twin model, and collaboratively allocate the output strategies of each energy storage in peak shaving and power backup scenarios;

[0055] The digital twin model can real-time and accurately simulate the operation status of the energy storage system and the power grid, providing an intuitive and visual decision-making basis for energy storage optimization control. Constructing a multi-objective optimization model can comprehensively consider multiple important objectives, solve through intelligent optimization algorithms, and can find the best output strategies of each energy storage in different scenarios in a complex operation environment, realizing the efficient utilization of energy storage resources and the stable operation of the power grid.

[0056] In this embodiment, the corrected theoretical power generation, the charge and discharge status, the remaining capacity, the charge and discharge power and other energy storage operation data of the energy storage device, as well as the parameters of the power grid such as voltage, frequency, and load are transmitted to the digital twin platform in real time through the data interface. Using three-dimensional modeling technology and simulation algorithms, a digital twin model highly similar to the actual energy storage system and power grid is constructed on the digital twin platform to achieve real-time synchronization of the energy storage system status and power grid parameters. Based on the digital twin model, with the goal of improving energy utilization efficiency, reducing operating costs, and ensuring power grid stability, a multi-objective optimization model is constructed, and intelligent optimization algorithms (such as genetic algorithms, particle swarm algorithms) are used to solve the model to obtain the optimal output strategies of each energy storage in different scenarios such as peak shaving and power backup.

[0057] The establishment of the digital twin model realizes the real-time and accurate simulation of the operation status of the energy storage system and the power grid, enabling operators to comprehensively understand the system operation conditions. Based on the constructed multi-objective optimization model, the output strategies obtained through intelligent algorithms can collaboratively allocate the work tasks of each energy storage in different scenarios, avoiding the disorderly operation of the energy storage device. In the peak shaving scenario, it can effectively smooth the output fluctuations of photovoltaic power and improve the power grid's ability to accommodate distributed energy; in the power backup scenario, it can reasonably allocate the energy storage capacity, ensure the power supply of critical loads, and significantly improve the overall operation efficiency of the energy storage system and the power grid stability.

[0058] S40. When it is detected that the battery degradation exceeds the first preset threshold or the local power grid load is unbalanced, trigger the replanning algorithm to recalculate the energy storage capacity allocation and optimize the charge and discharge priority to generate the final energy storage configuration plan.

[0059] Battery degradation will lead to a decline in the performance of the energy storage device, affecting the overall efficiency of the energy storage system; local power grid load imbalance may cause problems such as voltage fluctuations and frequency abnormalities, threatening the safe and stable operation of the power grid. When these situations occur, the original energy storage configuration plan may no longer be applicable, and it is necessary to promptly replan the energy storage capacity allocation and charge and discharge priority to ensure that the energy storage system can continue to operate stably and efficiently and guarantee the power grid safety.

[0060] During the operation of the energy storage system, the charge and discharge cycle times, capacity retention rate and other parameters of the battery are monitored in real time, and the degree of battery degradation is judged through data analysis. At the same time, the load conditions of each node of the power grid are monitored, and the load balance degree is calculated. When the battery degradation exceeds the first preset threshold or the degree of local power grid load imbalance exceeds the set standard, trigger the replanning algorithm. The replanning algorithm takes the corrected theoretical power generation, the current energy storage system status, the power grid load conditions, etc. as inputs, and uses dynamic programming algorithms or heuristic algorithms to recalculate the capacity allocation of each energy storage device, optimize the charge and discharge priority, and finally generate the final energy storage configuration plan adapted to the new working conditions and send the plan to the energy storage control system for execution.

[0061] By means of real-time monitoring and timely triggering of the replanning algorithm, abnormal conditions such as battery degradation and local grid load imbalance can be quickly responded to. The recalculated energy storage capacity allocation and optimized charge and discharge priorities enable the energy storage system to still operate efficiently under new working conditions, effectively reducing the adverse effects of battery degradation on the energy storage system, improving the ability of the energy storage system to cope with grid load changes, enhancing the stability and reliability of the distributed energy system, and ensuring the continuity and quality of power supply.

[0062] See Figure 2 , in one embodiment, a power generation efficiency correction model is constructed based on the dust distribution data and temperature data, and the actual power generation attenuation is calculated according to the power generation efficiency correction model, and the corrected theoretical power generation is output, including:

[0063] S201. Collect the dust coverage rate, particle size distribution and panel surface temperature gradient;

[0064] The dust coverage rate directly affects the area of the photovoltaic module receiving light. The higher the coverage rate, the less available light energy; the particle size distribution determines the scattering and absorption characteristics of dust to light, and dust with different particle sizes has different degrees of light attenuation; the panel surface temperature gradient reflects the non-uniformity of the component surface temperature. Temperature changes will change the carrier concentration and mobility of photovoltaic materials, affecting the power generation efficiency. These three parameters comprehensively reflect the influence of environmental factors on power generation performance from the optical and thermodynamic perspectives.

[0065] Traditional single sensors cannot obtain multi-dimensional information of dust and temperature, and it is difficult to accurately evaluate the environmental impact. By using a sensor array and advanced data processing technology, refined and real-time monitoring of dust and temperature can be achieved, and high spatio-temporal resolution data can be obtained. If the data is missing or the accuracy is insufficient, the subsequent constructed model will not be able to accurately reflect the actual power generation process, resulting in calculation deviation of the power generation amount and affecting the regulation effect of the energy storage system.

[0066] Therefore, in this embodiment, a micro-optical sensor array and a thin-film temperature sensor are evenly deployed on the surface of the photovoltaic module. The optical sensor uses the principle of laser scattering. By emitting a laser beam, the scattering angle and intensity of the laser by dust particles are analyzed to calculate the dust coverage rate and particle size distribution; the thin-film temperature sensor is closely attached to the component surface, forming a grid layout at a 5-cm interval to collect temperature data at each point in real time, and the discrete temperature data is processed by means of an interpolation algorithm to construct a temperature gradient distribution map of the photovoltaic module panel surface. The collected data is preliminarily processed by an edge computing device and then transmitted to the central data processing server through 5G or industrial Ethernet.

[0067] Through high-precision sensors and data processing technologies, the measurement error of the dust coverage rate can be controlled within 3%, the measurement error of the particle size distribution is less than 5%, and the measurement accuracy of the temperature gradient reaches ±0.5°C. These high-quality data provide a solid foundation for subsequent model construction, making the corrected calculation of power generation efficiency more accurate, significantly reducing the prediction error of power generation, helping the energy storage system to plan charge and discharge strategies more reasonably, and improving energy utilization efficiency.

[0068] S202. Construct a dust deposition-temperature field-power generation efficiency coupling model based on the dust coverage rate, particle size distribution, and plate surface temperature gradient;

[0069] Dust deposition causes light attenuation and reduces the effective light intensity received by the photovoltaic module; the change in the temperature field affects the bandgap width and carrier mobility of semiconductor materials, thereby changing the power generation efficiency. These two factors are interrelated. Dust deposition affects the heat dissipation characteristics of the module surface, thereby changing the temperature field distribution, and temperature changes also affect the physical and chemical properties and deposition state of dust. By integrating these interaction relationships, the coupling model can more realistically describe the variation law of power generation efficiency. If only the influence of dust or temperature on power generation efficiency is considered separately, the synergistic effect between the two cannot be reflected, resulting in a large deviation between the model calculation results and the actual situation. Constructing a coupling model and adopting a data-driven parameter optimization method can comprehensively consider the interactive effects of multiple factors, improving the accuracy and generality of the model. Traditional empirical formulas are difficult to adapt to complex and variable actual environments, while the coupling model can meet the power generation efficiency calculation requirements under different scenarios through parameter adjustment.

[0070] Therefore, in this embodiment, based on the theories of heat transfer, optics, and semiconductor physics, combined with a large amount of historical experimental data, a physical coupling model between dust deposition, temperature field, and power generation efficiency is established using the finite element analysis method. By performing multiple nonlinear regression on the power generation experimental data of photovoltaic modules under different dust conditions and temperature environments, the model parameters are determined.

[0071] Preferably, the construction method of the coupling model includes

[0072] ;

[0073] wherein, is the corrected efficiency, is the dust deposition thickness, is the temperature deviation, is the dust influence coefficient, and its value range is ; is the temperature influence coefficient, and its value range is .

[0074] The dust influence coefficient represents the proportion of photovoltaic efficiency loss caused by each millimeter of dust deposition thickness. For example, if indicates that the efficiency loss is 30% when 1 mm of dust is deposited; the temperature influence coefficient represents the proportion of photovoltaic efficiency loss caused by each degree Celsius of temperature deviation. For example, if , then when the temperature rises ( ), the efficiency loss is 5%. When calculating, it is necessary to ensure that is the difference between the actual temperature and the standard test temperature ( ). Preferably, a genetic algorithm can be used to perform global optimization on and , and dynamically adjust the parameter values according to different regions, seasons and component types, so that the model can adapt to complex and changeable actual working conditions.

[0075] The average error between the power generation efficiency calculated by the coupling model and the actual measured value is within 2.5%. Compared with the single-factor model, the error is greatly reduced. The accurate power generation efficiency model provides a reliable power generation prediction for the energy storage system, makes the formulation of the energy storage charge and discharge strategy more scientific, reduces the waste or shortage of energy storage resources caused by inaccurate prediction, and improves the overall economy and stability of the distributed energy system.

[0076] S203. Based on the coupling model, dynamically calculate the actual efficiency decay rate of the photovoltaic module through the heat conduction equation and the light attenuation coefficient, and output the corrected theoretical power generation.

[0077] The heat conduction equation describes the law of heat transfer inside the photovoltaic module. By solving this equation, the temperature of each part of the module can be obtained, and then the influence of temperature on the power generation efficiency can be calculated; the light attenuation coefficient reflects the absorption and scattering degree of dust on light, and different dust characteristics correspond to different light attenuation coefficients. Based on the coupling model, comprehensively calculate the influence of heat conduction and light attenuation on the power generation efficiency, which can accurately reflect the efficiency decay of the photovoltaic module under actual operating conditions, so as to correct the theoretical power generation.

[0078] The power generation efficiency of the photovoltaic module changes dynamically with the environmental conditions, and the traditional static calculation method cannot adapt to this change. By collecting data in real time and dynamically calculating the efficiency decay rate in combination with physical equations, it can timely reflect the influence of environmental changes on the power generation performance, and make the corrected theoretical power generation closer to the actual situation. The calculation method based on physical equations and coupling models has a solid theoretical basis. Compared with empirical formulas, the calculation results are more reliable and can provide an accurate basis for the regulation of the energy storage system.

[0079] In this embodiment, the real-time dust coverage rate, particle size distribution, and panel temperature gradient data collected in step S201 are input into the coupling model. Using the heat conduction equation and combining the thermophysical parameters of the component materials, the temperature distribution inside the component is calculated to determine the impact of temperature on the power generation efficiency; according to the dust coverage rate and particle size distribution, the light attenuation coefficient is calculated through the Mie scattering theory to evaluate the light loss caused by dust. The efficiency attenuation caused by temperature and dust is superimposed to obtain the actual efficiency attenuation rate of the photovoltaic module. Finally, according to the corrected efficiency formula the corrected power generation efficiency is calculated, and combined with the nominal power and operating time of the component, the corrected theoretical power generation is output.

[0080] The dynamic calculation method can achieve real-time correction of the power generation efficiency of photovoltaic modules, and the average error between the corrected theoretical power generation and the actual power generation is controlled within 2%. Accurate power generation prediction enables the energy storage system to adjust the charge and discharge strategies more timely and accurately, efficiently store excess power when the light is sufficient, and reasonably release power during peak electricity consumption or insufficient light, improving the response speed and regulation accuracy of the energy storage system, enhancing the stability and energy utilization efficiency of the distributed energy system, and reducing the system operation cost.

[0081] See Figure 3 , in one embodiment, a digital twin model is established based on the corrected theoretical power generation and energy storage operation data, and synchronized to the energy storage system state and grid parameters; a multi-objective optimization model is constructed based on the digital twin model to collaboratively allocate the output strategies of each energy storage in peak shaving and backup power scenarios, including:

[0082] S301. Based on the corrected theoretical power generation and the operation data of the energy storage system, a digital twin covering the photovoltaic module, energy storage battery, and grid topology is established, and the state of charge of the battery, line load, and environmental parameters are synchronized in real time;

[0083] Traditional centralized monitoring systems cannot reflect the dynamic characteristics of distributed energy systems and the coupling relationships between devices in real time. Digital twin realizes real-time mapping and prediction of the state of physical systems by establishing a virtual model highly similar to the physical system and using a method that combines data-driven and physical modeling. There are complex energy flows and interactions between photovoltaic modules, energy storage batteries, and the grid. Only by establishing a unified model covering all three can the operating state of the entire system be accurately reflected.

[0084] Specifically, in step S301, a photovoltaic-storage-grid joint simulation model is established based on the electrical characteristic equation of the photovoltaic module, the equivalent circuit model of the energy storage battery, and the power grid power flow equation. The thermal distribution of the photovoltaic module and the electrochemical process inside the battery are modeled through the finite element analysis method to accurately reflect the physical characteristics of the equipment. The temperature, light intensity, output power of the photovoltaic module, the voltage, current, temperature, and SOC (state of charge) of the energy storage battery, as well as the node voltage and line current of the power grid are collected in real time through the IoT sensor network. The on-site data is transmitted to the digital twin platform using a general protocol, and the data is denoised and state-estimated through a Kalman filter to ensure the accuracy and real-time nature of the data. Finally, the actual operation data is compared with the simulation results of the digital twin model, and the model parameters are dynamically adjusted through the particle swarm optimization algorithm to control the error between the model output and the actual system within 5%. In this way, panoramic monitoring and accurate prediction of the distributed energy system are achieved, the system state estimation error is reduced to within 3%, and the fault warning lead time is correspondingly increased. Through the pre-simulation function of the digital twin, the effects of different control strategies can be evaluated in advance, the trial-and-error cost can be reduced, and the operational stability of the system can be improved.

[0085] S302. Using the real-time data of the digital twin, taking the distributed energy storage system as a game participant, design a multi-objective optimization model based on the Shapley value to dynamically allocate the output strategies of each energy storage in the peak shaving and power backup scenarios, and generate an initial energy storage configuration plan.

[0086] Traditional centralized optimization methods are difficult to adapt to the decentralization and uncertainty of distributed energy systems. The Shapley value theory solves the balance problem between individual interests and collective interests in distributed systems through a fair distribution mechanism that quantifies the contribution of each participant to the coalition. In multi-objective optimization, different output strategies will have different impacts on the peak shaving effect, power backup reliability, and the lifespan of energy storage devices, and an optimal balance point needs to be found through mathematical methods.

[0087] In step S302, each energy storage unit in the distributed energy storage system is regarded as an independent game participant, and each participant aims to maximize its own interests and optimize the overall system. The goal of the peak shaving scenario is defined as minimizing the peak-valley difference of the power grid, and the goal of the backup power scenario is defined as maximizing the power supply reliability of critical loads. The marginal contribution of each energy storage unit to the system goal under different coalition combinations is calculated through the Monte Carlo simulation method. The revenue weights of each energy storage unit are allocated according to the Shapley value formula, a multi-objective function including economic benefits, system reliability, and equipment life is established, and the NSGA-II algorithm is used to solve the Pareto optimal solution set. The optimal compromise solution is determined through the fuzzy decision theory to generate the initial energy storage configuration plan. Compared with the traditional optimization method, the multi-objective optimization strategy based on the Shapley value can reduce the peak-valley difference of the power grid by 15%-20%, improve the power supply reliability of critical loads to more than 99.9%, and extend the life of energy storage equipment. This method fairly distributes the revenue of each energy storage unit on the premise of ensuring the overall interests of the system, improves the enthusiasm of participants, and promotes the healthy development of the distributed energy system. Therefore, in this embodiment, through the combination of digital twin and game theory, intelligent collaborative control of the energy storage system is realized.

[0088] In one embodiment, the design of the multi-objective optimization model based on the Shapley value includes:

[0089] 1) According to the real-time line load data provided by the digital twin, the output range of each energy storage system is restricted so that the voltage deviation of the power grid nodes does not exceed ±5%, and the line load rate is lower than 90%;

[0090] The safe operation of the power system has strict requirements for voltage and line load. Excessive voltage deviation at nodes will affect the life and performance of electrical equipment, and line overload may cause equipment damage or even power outages. By using the digital twin model to monitor the system status in real time and adding constraint conditions during the optimization process, it can ensure that the energy storage control strategy is carried out on the premise of meeting system safety.

[0091] State estimation and power flow calculation: Using the real-time line load, node voltage, and equipment parameters provided by the digital twin platform, the power flow equation of the power grid is solved by the Newton-Raphson method to calculate the voltage of each node and the line load rate.

[0092] Constraint condition construction:

[0093] Voltage constraint: ;

[0094] Load rate constraint: ;

[0095] Energy storage output constraint: ;

[0096] Among them, is the voltage of the node ; is the reference voltage ; is the current of the line ; is the maximum current of the line ;

[0097] Embed the above constraints into the multi-objective optimization model and solve it using the interior point method or sequential quadratic programming algorithm. Through this constraint mechanism, the voltage deviation of the power grid nodes can be controlled within ±3%, and the line load rate can be maintained below 80%, significantly improving the operational safety of the power grid. At the same time, the regulation strategy of the energy storage system can fully consider the actual operating state of the power grid and avoid system instability problems caused by blind regulation.

[0098] 2) Allocate the peak shaving benefits of each energy storage system through the Shapley value to ensure that the individual benefits are not lower than the benefits when operating independently, and the total benefits satisfy superadditivity.

[0099] As a fair allocation mechanism, the Shapley value can quantify the marginal contribution of each participant to the coalition. In the peak shaving scenario, the positions, capacities, and response characteristics of different energy storage systems are different, and their contributions to the system are also different. By allocating benefits through the Shapley value, each energy storage system can be motivated to actively participate in peak shaving, while ensuring the fairness and stability of the allocation results. This method can make the matching degree between the peak shaving benefits of each energy storage system and its actual contribution reach more than 90%, effectively solving the fairness problem caused by the traditional capacity-based benefit allocation. By satisfying the individual rationality and collective rationality constraints, the enthusiasm of the energy storage system to participate in peak shaving is improved, and the overall peak shaving efficiency of the system is greatly enhanced.

[0100] Therefore, through the power grid constraints in the above embodiments, the risks of voltage over-limit and line overload caused by energy storage regulation are reduced, and the system failure rate is decreased. The Shapley value allocation mechanism makes the benefits of each energy storage system positively correlated with its contribution, and the participation degree is improved. On the premise of meeting the system safety constraints, the peak shaving cost is reduced, achieving the balance between the overall system benefits and individual interests.

[0101] Preferably, the replanning algorithm is as follows:

[0102] Calculate the capacity attenuation rate according to the battery health state prediction model:

[0103] ;

[0104] where is the battery health state is the number of charge and discharge cycles is the attenuation coefficient, and the value range is ;

[0105] When is lower than the second preset threshold, a capacity reallocation strategy is triggered to transfer the load of high-decay batteries to low-decay batteries proportionally.

[0106] Battery capacity decay is a gradual process and has a non-linear relationship with the number of charge-discharge cycles. The square root model can better fit the physical mechanism of battery decay and reflect the process of membrane growth and active material loss inside the battery. By real-time monitoring and prediction , the downward trend of battery performance can be detected in advance, providing a decision-making basis for capacity reallocation.

[0107] The charge-discharge current, voltage, and temperature data of the battery are collected in real time through the battery management system. The sliding window filtering algorithm is used to remove noise and extract the number of charge-discharge cycles and depth. For different types of batteries, such as lithium iron phosphate batteries, experimental data are obtained through accelerated life tests, and the decay coefficient is fitted by the least squares method , a battery health state prediction model is established, and the Kalman filter is used to fuse real-time data with the prediction model to achieve dynamic update, with the estimation error controlled within ±3%.

[0108] When it is monitored that the of any battery is lower than the second preset threshold, usually 80%, the capacity reallocation algorithm is triggered. According to the value of each battery, the load distribution weight is calculated, that is, the of the current battery and the sum of all batteries ratio. The PID controller is used to achieve smooth transfer of the load, avoiding the impact on the power grid caused by power mutation. The transfer rate should be limited to ensure the safe operation of the battery.

[0109] By transferring the load of high-decay batteries to low-decay batteries, the charge-discharge depth of each battery can be balanced, the aging speed of high-decay batteries can be slowed down, and the service life of the entire energy storage system can be extended. Based on weight allocation mechanism ensures the fairness and rationality of load transfer, maximizing the overall performance of the system.

[0110] Therefore, in this embodiment, through predictive maintenance and capacity reallocation, the overall life of the energy storage system is extended, the battery replacement frequency is reduced, the operation and maintenance cost is reduced, the available capacity retention rate of the system is increased, and the charge-discharge efficiency is improved.

[0111] Preferably, the data synchronization frequency of the digital twin is not lower than 10 Hz, and local data processing is achieved through edge computing nodes to reduce the cloud transmission delay.

[0112] Deploy micro multi-parameter sensors on the surface of photovoltaic modules, integrating functions of detecting light intensity, temperature, and dust coverage, with a sampling frequency ≥ 100 Hz. The energy storage battery pack is configured with a BMS (Battery Management System) to collect voltage, current, and temperature data in real time, with a sampling frequency ≥ 50 Hz; install PMUs (Phasor Measurement Units) at key grid nodes to synchronously collect data such as voltage phasors and frequencies, with a sampling frequency ≥ 60 Hz; use an industrial-grade edge computing gateway equipped with a multi-core processor and an FPGA co-processor, deploy a lightweight containerized platform to achieve distributed deployment of algorithm modules; the edge node has a built-in data processing pipeline, including functional modules such as data cleaning, feature extraction, and state estimation. Adopt a hybrid mechanism of time-triggered and event-triggered. Under normal operating conditions, synchronize key state data at a frequency of 10 Hz to achieve two-way data synchronization between the edge and the cloud, supporting breakpoint resumption and data verification mechanisms. In this way, the end-to-end data synchronization delay is reduced from 200 ms in the traditional solution to within 8 ms, the state estimation update cycle is improved from the second level to the 100 ms level, significantly improving the system response speed, and the fault detection time is shortened to within 50 ms, realizing rapid perception and response to power grid disturbances. Through the deep integration of hardware acceleration and algorithm optimization, a low-latency and highly reliable digital twin data channel is constructed.

[0113] In summary, the intelligent regulation energy storage optimization method for a distributed energy system provided by the embodiment of the present invention can accurately calculate the actual power generation attenuation by real-time monitoring the dust distribution data and temperature data on the surface of photovoltaic modules and constructing a power generation efficiency correction model, and output a more realistic corrected theoretical power generation amount, providing a reliable basis for energy storage planning. Based on this, a digital twin model is established, the state of the energy storage system and grid parameters are synchronized, and a multi-objective optimization model is constructed to collaboratively allocate the output strategies of each energy storage in different scenarios, greatly improving the operating efficiency of the energy storage system. When it is detected that the battery attenuation exceeds the first preset threshold or there is a local power grid load imbalance, the replanning algorithm is triggered to recalculate the energy storage capacity allocation and optimize the charge and discharge priorities to generate the final energy storage configuration plan, ensuring that the energy storage system can still operate stably and efficiently under complex operating conditions, enhancing the ability of the distributed energy system to cope with various changes, and improving the stability, reliability, and energy utilization efficiency of the entire system.

[0114] See Figure 4 , in one embodiment, the present invention also provides an intelligent regulation energy storage optimization system for a distributed energy system, and the system includes:

[0115] A data acquisition unit 100 for real-time monitoring of the dust distribution data and temperature data on the surface of photovoltaic modules;

[0116] The power generation correction unit 200 is configured to construct a power generation efficiency correction model based on dust distribution data and temperature data, calculate the actual power generation attenuation according to the power generation efficiency correction model, and output the corrected theoretical power generation amount;

[0117] The multi-objective optimization unit 300 is configured to establish a digital twin model based on the corrected theoretical power generation amount and energy storage operation data, and synchronize it to the energy storage system state and grid parameters; construct a multi-objective optimization model based on the digital twin model, and collaboratively allocate the output strategies of each energy storage in peak shaving and power backup scenarios;

[0118] The energy storage configuration update unit 400 is configured to trigger a replanning algorithm to recalculate the energy storage capacity allocation and optimize the charge and discharge priorities when it detects that the battery attenuation exceeds the first preset threshold or the local grid load is unbalanced, and generate a final energy storage configuration plan.

[0119] It can be understood that the functions or modules included in the system provided in this embodiment can be used to execute the methods described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0120] The present invention also provides an electronic device, including a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes the method in any of the above possible implementation manners.

[0121] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by the processor of the electronic device, the processor is caused to execute the method in any of the above possible implementation manners.

[0122] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

Claims

1. An intelligent regulation energy storage optimization method for a distributed energy system, characterized in that, The method includes: Real-time monitoring of the dust distribution data and temperature data on the surface of photovoltaic modules; Based on the dust distribution data and temperature data, constructing a power generation efficiency correction model, calculating the actual power generation attenuation according to the power generation efficiency correction model, and outputting the corrected theoretical power generation; Establishing a digital twin model based on the corrected theoretical power generation and energy storage operation data, and synchronizing it to the energy storage system status and grid parameters; constructing a multi-objective optimization model based on the digital twin model, and collaboratively allocating the output strategies of each energy storage in peak shaving and backup power scenarios; When it is detected that the battery attenuation exceeds the first preset threshold or the local grid load is unbalanced, triggering a replanning algorithm to recalculate the energy storage capacity allocation and optimize the charge and discharge priorities, and generating a final energy storage configuration plan.

2. The intelligent regulation energy storage optimization method for a distributed energy system according to claim 1, wherein Based on the dust distribution data and temperature data, constructing a power generation efficiency correction model, calculating the actual power generation attenuation according to the power generation efficiency correction model, and outputting the corrected theoretical power generation, including: Collecting the dust coverage rate, particle size distribution, and board surface temperature gradient; Based on the dust coverage rate, particle size distribution, and board surface temperature gradient, constructing a dust deposition-temperature field-power generation efficiency coupling model; Based on the coupling model, dynamically calculating the actual efficiency attenuation rate of the photovoltaic module through the heat conduction equation and the light attenuation coefficient, and outputting the corrected theoretical power generation.

3. The intelligent regulation energy storage optimization method for a distributed energy system according to claim 2, characterized in that, The construction method of the coupling model includes ； Among them, is the corrected efficiency, is the dust deposition thickness, is the temperature deviation, is the dust influence coefficient, and its value range is ; is the temperature influence coefficient, and its value range is .

4. The intelligent control energy storage optimization method for a distributed energy system according to claim 1, characterized in that Establishing a digital twin model according to the corrected theoretical power generation and energy storage operation data, and synchronizing it to the energy storage system status and grid parameters; Constructing a multi-objective optimization model based on the digital twin model, and collaboratively allocating the output strategies of each energy storage in peak shaving and backup power scenarios, including: Based on the corrected theoretical power generation and energy storage system operation data, establishing a digital twin body covering photovoltaic modules, energy storage batteries, and grid topology, and real-time synchronizing the state of charge of the battery, line load, and environmental parameters; Using the real-time data of the digital twin body, taking the distributed energy storage system as a game participant, designing a multi-objective optimization model based on the Shapley value, dynamically allocating the output strategies of each energy storage in peak shaving and backup power scenarios, and generating an initial energy storage configuration plan.

5. The intelligent control energy storage optimization method for a distributed energy system according to claim 4, characterized in that The design of the multi-objective optimization model based on the Shapley value includes: According to the real-time line load data provided by the digital twin body, restricting the output range of each energy storage system, so that the grid node voltage deviation does not exceed ±5%, and the line load rate is lower than 90%; Allocating the peak shaving benefits of each energy storage system through the Shapley value, ensuring that the individual benefits are not lower than the benefits when operating independently, and the total benefits satisfy superadditivity.

6. The intelligent regulation energy storage optimization method for a distributed energy system according to claim 1, wherein The replanning algorithm is: According to the battery health state prediction model, calculating the capacity attenuation rate: ； Among them, is the battery health status, is the number of charge and discharge cycles, is the attenuation coefficient, and the value range is ; When below the second preset threshold, a capacity reallocation strategy is triggered to transfer the load of the high-decay battery to the low-decay battery in proportion.

7. The intelligent control energy storage optimization method for a distributed energy system according to claim 1, wherein The data synchronization frequency of the digital twin body is not less than 10Hz, and local data processing is realized through edge computing nodes to reduce the cloud transmission delay.

8. An intelligent regulation energy storage optimization system for a distributed energy system, characterized in that, The system includes: A data acquisition unit for real-time monitoring of the dust distribution data and temperature data on the surface of photovoltaic modules; A power generation amount correction unit for constructing a power generation efficiency correction model based on the dust distribution data and temperature data, calculating the actual power generation attenuation according to the power generation efficiency correction model, and outputting the corrected theoretical power generation; The multi-objective optimization unit is used to establish a digital twin model based on the corrected theoretical power generation and energy storage operation data, and synchronize it to the energy storage system state and grid parameters; construct a multi-objective optimization model based on the digital twin model, and collaboratively allocate the output strategies of each energy storage in peak shaving and backup power scenarios; The energy storage configuration update unit is used to trigger the replanning algorithm to recalculate the energy storage capacity allocation and optimize the charge and discharge priorities when it detects that the battery decay exceeds the first preset threshold or the local grid load is unbalanced, and generate the final energy storage configuration plan.

9. An electronic device, characterized in that, Comprising: A processor and a memory, the memory is used to store computer program codes, the computer program codes include computer instructions, and when the processor executes the computer instructions, the electronic device executes the intelligent regulation energy storage optimization method for a distributed energy system according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by the processor of the electronic device, the processor is caused to execute the intelligent regulation energy storage optimization method for a distributed energy system according to any one of claims 1 to 7.

Images