Energy storage system power smoothing method and system based on model predictive control

Abstract

The invention relates to the technical field of energy storage systems, in particular to an energy storage system power smoothing method and system based on model predictive control. By introducing the adaptive prediction model and combining with the dynamic frequency decomposition module, the power signal can automatically adjust the frequency decomposition threshold according to the dynamic change of the signal, so that the flexibility and precision of frequency decomposition are remarkably improved, and the accuracy of frequency decomposition is improved by automatically adjusting the frequency decomposition threshold on the basis of the dynamic change information. Therefore, the power difference value sequence is decomposed into a plurality of dynamic frequency components, the change trend of each frequency component is predicted, the future power requirement can be estimated more accurately, the charging and discharging strategy of the energy storage equipment is adjusted in real time, it is guaranteed that the energy storage system can provide efficient and accurate power compensation in practical application, and the energy storage efficiency is improved. According to the invention, the parameters of the adaptive prediction model can be corrected in real time, and the continuous smoothing of the power of the energy storage system is realized by continuously optimizing the frequency decomposition precision and the power prediction accuracy.

Description

Technical Field

[0001] This invention relates to the field of energy storage system technology, and in particular to a power smoothing method and system for energy storage systems based on model predictive control. Background Technology

[0002] Power smoothing in energy storage systems refers to reducing the fluctuations in the output power of a power grid or system by using energy storage devices (such as batteries and supercapacitors), making the system output more stable and avoiding the impact on the power grid or the power quality caused by excessive fluctuations. In the application of renewable energy equipment such as wind power and photovoltaic power generation, the instability and fluctuation of their output power often put pressure on the power grid. The goal of power smoothing in energy storage systems is to reduce such fluctuations by absorbing and releasing energy through energy storage devices to balance the fluctuations in power supply and make the power output of the power grid tend to be stable.

[0003] Current technologies for achieving power balancing in energy storage systems primarily employ filtering techniques. The principle involves processing the output power or load power signal of renewable energy devices using filters, decomposing it into different frequency components, and then allocating these components according to the characteristics of the energy storage devices. This allows the energy storage system to compensate for power differences, thereby achieving power balance. However, this method is based on fixed-frequency decomposition. Since these frequency responses are static, they cannot be adjusted according to dynamic changes in the signal, thus affecting the smoothing effect and resulting in poor power balancing performance. Summary of the Invention

[0004] The main objective of this invention is to provide a power smoothing method for energy storage systems based on model predictive control, aiming to solve the technical problems in the prior art.

[0005] This invention proposes a power smoothing method for energy storage systems based on model predictive control, comprising:

[0006] The system acquires real-time output power data of multiple renewable energy devices connected to the energy storage system, dynamic change information of the energy storage system, real-time load power data of multiple devices, and operating parameters of the energy storage devices, and obtains a power difference sequence based on the multiple real-time output power data and real-time load power data.

[0007] An adaptive prediction model is constructed based on the power difference sequence and operating parameters, wherein the adaptive prediction model includes a dynamic frequency decomposition module and a power prediction module;

[0008] The dynamic change information is input into the dynamic frequency decomposition module of the adaptive prediction model, and the frequency decomposition threshold is automatically adjusted to obtain the real-time adjustment threshold. The power difference sequence is then decomposed into multiple dynamic frequency components based on the real-time adjustment threshold.

[0009] The multiple dynamic frequency components are input into the power prediction module and combined with the operating parameters to predict the changing trend of each dynamic frequency component within a future preset time period, thereby obtaining the power prediction curve;

[0010] Obtain the charging and discharging constraints of the energy storage device, and obtain the power compensation command of the energy storage system based on the charging and discharging constraints and the power prediction curve;

[0011] The power compensation command is sent to the energy storage device to obtain the corresponding actual compensation power, and the power smoothing error is obtained based on the actual compensation power;

[0012] The parameters of the adaptive prediction model are iteratively corrected based on the power smoothing error to optimize the dynamic frequency decomposition accuracy and achieve continuous power smoothing of the energy storage system.

[0013] This application also provides a power smoothing system for an energy storage system based on model predictive control, comprising:

[0014] The first acquisition module is used to acquire multiple real-time output power data of renewable energy devices connected to the energy storage system, dynamic change information of the energy storage system, multiple real-time load power data and operating parameters of the energy storage devices, and to acquire a power difference sequence based on the multiple real-time output power data and real-time load power data.

[0015] A construction module is used to construct an adaptive prediction model based on the power difference sequence and operating parameters, wherein the adaptive prediction model includes a dynamic frequency decomposition module and a power prediction module;

[0016] The adjustment decomposition module is used to input the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model, and to automatically adjust the frequency decomposition threshold to obtain the real-time adjustment threshold, and decompose the power difference sequence into multiple dynamic frequency components according to the real-time adjustment threshold.

[0017] The prediction module is used to input multiple dynamic frequency components into the power prediction module and combine them with operating parameters to predict the changing trend of each dynamic frequency component within a future preset time period, thereby obtaining a power prediction curve.

[0018] The second acquisition module is used to acquire the charging and discharging constraints of the energy storage device, and to acquire the power compensation command of the energy storage system based on the charging and discharging constraints and the power prediction curve.

[0019] The third acquisition module is used to send the power compensation command to the energy storage device to obtain the corresponding actual compensation power, and to obtain the power smoothing error based on the actual compensation power.

[0020] The correction module is used to iteratively correct the parameters of the adaptive prediction model based on the power smoothing error, so as to optimize the dynamic frequency decomposition accuracy and achieve continuous power smoothing of the energy storage system.

[0021] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described model predictive control-based power smoothing method for energy storage systems.

[0022] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described model predictive control-based power smoothing method for energy storage systems.

[0023] The beneficial effects of this invention are as follows: By introducing an adaptive prediction model and combining it with a dynamic frequency decomposition module, this invention enables the power signal to automatically adjust the frequency decomposition threshold according to the dynamic changes of the signal, thereby improving the flexibility and accuracy of frequency decomposition. By automatically adjusting the frequency decomposition threshold based on dynamic change information, the power difference sequence is decomposed into multiple dynamic frequency components. By predicting the changing trend of each frequency component, future power demand can be estimated more accurately, and the charging and discharging strategies of energy storage devices can be adjusted in real time, ensuring that the energy storage system can provide efficient and accurate power compensation in practical applications. This invention can correct the parameters of the adaptive prediction model in real time, and by continuously optimizing the frequency decomposition accuracy and power prediction accuracy, it achieves continuous smoothing of the power of the energy storage system, improving the response speed and smoothing performance of the energy storage system. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of a method flow according to an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the system structure according to an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of the internal structure of a computer device according to an embodiment of this application.

[0027] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0028] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0029] like Figure 1 As shown, this application provides a power smoothing method for energy storage systems based on model predictive control, including:

[0030] S1. Acquire real-time output power data of multiple renewable energy devices connected to the energy storage system, dynamic change information of the energy storage system, real-time load power data of multiple devices, and operating parameters of the energy storage devices, and obtain a power difference sequence based on the real-time output power data and the real-time load power data.

[0031] S2. Construct an adaptive prediction model based on the power difference sequence and operating parameters. The adaptive prediction model includes a dynamic frequency decomposition module and a power prediction module.

[0032] S3. Input the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model to automatically adjust the frequency decomposition threshold, obtain the real-time adjustment threshold, and decompose the power difference sequence into multiple dynamic frequency components according to the real-time adjustment threshold.

[0033] S4. Input multiple dynamic frequency components into the power prediction module and combine them with the operating parameters to predict the changing trend of each dynamic frequency component within a preset time period in the future, and obtain the power prediction curve.

[0034] S5. Obtain the charging and discharging constraints of the energy storage device, and obtain the power compensation command of the energy storage system based on the charging and discharging constraints and the power prediction curve.

[0035] S6. Send the power compensation command to the energy storage device to obtain the corresponding actual compensation power, and obtain the power smoothing error based on the actual compensation power;

[0036] S7. Iteratively correct the parameters of the adaptive prediction model based on the power smoothing error to optimize the dynamic frequency decomposition accuracy and achieve continuous power smoothing of the energy storage system.

[0037] As described in steps S1-S2 above, the step of obtaining a power difference sequence based on multiple real-time output power data and real-time load power data includes obtaining corresponding data acquisition timestamps based on multiple real-time output power data and real-time load power data; calculating the difference between real-time output power data and real-time load power data at the same timestamp to obtain the power difference at a single time point; and sorting the power difference at each time point according to the chronological order of multiple timestamps to form a power difference sequence arranged in time series.

[0038] The difference between the actual compensation power and the theoretical compensation power is calculated by combining the actual compensation power with the actual value of the power difference sequence at the corresponding time. The single-time compensation deviation is obtained by combining the actual compensation power with the actual value of the power difference sequence at the corresponding time. The deviation sequence is then arranged in time order based on multiple single-time compensation deviations. The power smoothing error can be obtained by calculating the root mean square value of the deviation sequence.

[0039] This invention, by acquiring multi-source data such as the output and load power of renewable energy devices, helps to comprehensively analyze various factors affecting the performance of energy storage systems. The power difference sequence serves as the basis for power regulation of the energy storage system, ensuring real-time response to power fluctuations. By constructing an adaptive prediction model, it helps to accurately predict future power changes of the energy storage system through intelligent algorithms and make real-time adjustments based on dynamic output. Through the combination of the dynamic frequency decomposition module and the power prediction module, it can not only analyze past power data, but also predict future power demand and fluctuation trends in real time based on the current operating status. This ensures that the power regulation of the energy storage system can fully consider changes in historical data and the operating environment, thereby effectively improving the smoothing effect of the energy storage system on power fluctuations. The dynamic frequency decomposition module automatically adjusts the frequency decomposition threshold according to real-time dynamic change information, thereby realizing dynamic processing of power signals. The frequency division threshold is adjusted according to the actual changes in the signal, flexibly adapting to the characteristics of different power fluctuations. This avoids the defects of unsuitable frequency decomposition and poor smoothing effect caused by fixed frequency division thresholds in traditional methods. Dynamically adjusting the frequency division threshold allows each frequency component to more accurately match the actual power fluctuation situation, improving the accuracy and response speed of power smoothing.

[0040] By inputting dynamic frequency components into the power prediction module and combining them with operating parameters for trend prediction, a more accurate power compensation scheme can be provided. By independently predicting the changing trend of each frequency component, the impact of each frequency component on the energy storage system can be more accurately grasped, and the precise formulation of compensation measures can be ensured. This improves the power smoothing capability of the energy storage system. The charging and discharging constraints of energy storage devices are usually determined by multiple factors such as battery performance, environmental factors, and usage specifications. Obtaining the charging and discharging constraints and calculating them in conjunction with the power prediction curve can effectively ensure that the operation of the energy storage system does not exceed its capacity range. Intelligent scheduling based on real-time prediction data and constraints ensures the energy storage system operates within its capacity. This system ensures long-term stability and power smoothing of the energy storage system, while maximizing the utilization efficiency of energy storage devices. It effectively reduces unnecessary losses in power compensation. By comparing the actual compensation power with the predicted results, the power compensation strategy can be adjusted in real time to ensure that the gap between the actual compensation power and the ideal state is minimized, so that the system is always in the best operating state. By adjusting the parameters of the adaptive prediction model according to the power smoothing error, the accuracy of frequency decomposition can be continuously optimized, and the power smoothing effect of the energy storage system can be continuously optimized. This allows the prediction model to continuously self-optimize over time and with changes in the operating environment, thereby improving the system's adaptability to complex and variable loads and energy inputs.

[0041] In one embodiment, step S2, which constructs an adaptive prediction model based on the power difference sequence and operating parameters, includes:

[0042] S21. Obtain the key feature set of the power difference sequence, and determine the core input dimension of the adaptive prediction model based on the key feature set and operating parameters to obtain the model input parameter system;

[0043] S22. Based on the model input parameter system, select the dynamic frequency decomposition algorithm and the power prediction algorithm as the basic framework of the dynamic frequency decomposition module and the power prediction module of the adaptive prediction model, respectively, to obtain the initial framework of the module.

[0044] S23. Based on the fluctuation frequency range in the key feature set, the initial frequency threshold of the dynamic frequency decomposition module is preset to obtain the module pre-configuration parameters.

[0045] S24. Initialize the weight parameters of the power prediction module based on the equipment charging and discharging efficiency and attenuation coefficient in the operating parameters to obtain the initial weight of the module;

[0046] S25. Based on the initial module framework, pre-configured module parameters, and initial module weights, the dynamic frequency decomposition module and the power prediction module are integrated to construct an adaptive prediction model.

[0047] As described in steps S21-S25 above, the key feature set includes key features such as power fluctuation period, peak deviation, and rate of change. Based on the energy storage device response speed and capacity limitation parameters in the operating parameters, the device performance constraints are determined to obtain a constraint list. Based on the key feature set and the constraint list, features and constraints that have a significant impact on prediction accuracy are selected to determine the core input dimension of the adaptive prediction model. Based on the core input dimension, the feature items and constraints are classified and organized to obtain the model input parameter system.

[0048] Based on the model input parameter system, the parameter dimensions, data volatility, and real-time requirements are extracted to obtain the algorithm selection criteria. Based on the algorithm selection criteria, dynamic frequency decomposition algorithms that are suitable for parameter dimensions and volatility are selected to determine the basic algorithm of the dynamic frequency decomposition module. Based on the algorithm selection criteria, power prediction algorithms that meet real-time requirements are selected to determine the basic algorithm of the power prediction module. Based on the dynamic frequency decomposition algorithm and the power prediction algorithm, the corresponding module structures are built to obtain the initial framework of the module.

[0049] By extracting key features from the power difference sequence, important changes in the power balance of the energy storage system can be accurately captured. By combining operating parameters to determine the core input dimension, the model can be adjusted according to different actual application scenarios, helping the model focus on the core factors that truly affect the power difference, thereby improving prediction accuracy. By selecting dynamic frequency decomposition algorithm and power prediction algorithm as the basic framework of the dynamic frequency decomposition module and power prediction module respectively, the model structure can be made clearer and more modular, facilitating later optimization and replacement. The selection of dynamic frequency decomposition and power prediction algorithms can adjust the prediction parameters in real time according to different features of the model input, making the model highly adaptive. By selectively choosing suitable algorithms, the response speed and accuracy of the frequency decomposition module can be significantly improved, thereby improving the accuracy of power balance.

[0050] By pre-setting frequency thresholds based on the fluctuation frequency range of key feature sets, the dynamic frequency decomposition module can focus on the key frequency range of power fluctuations, improving the efficiency and accuracy of the decomposition process. After pre-setting the initial frequency thresholds, the dynamic frequency decomposition module can monitor frequency changes in real time and dynamically adjust the frequency range according to actual operating conditions, better adapting to actual system fluctuations and enhancing the model's ability to cope with complex changes. By initializing the weight parameters of the power prediction module in conjunction with the device's charging and discharging efficiency and attenuation coefficient, it can be ensured that the initial prediction results of the model match the actual device performance. The device's charging and discharging efficiency and attenuation coefficient are key factors affecting power balance. The introduction of these parameters enables the power prediction module to effectively adjust according to the equipment characteristics in the early stages of system operation. This not only improves prediction accuracy but also provides a reasonable starting point for subsequent adaptive adjustments. The dynamic frequency decomposition module provides accurate frequency decomposition, while the power prediction module performs power prediction based on the actual equipment performance. The combination of these two modules greatly optimizes the overall performance of the system. The integrated model can make real-time adaptive adjustments based on factors such as changes in the actual power difference and adjustments to equipment operating parameters, ensuring the stability and continuity of the power balance effect. By dynamically adjusting the decomposition frequency through actual signal changes, the power balance effect and system response speed are improved.

[0051] In one embodiment, step S3, which involves inputting dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model to automatically adjust the frequency decomposition threshold to obtain a real-time adjusted threshold, and decomposing the power difference sequence into multiple dynamic frequency components based on the real-time adjusted threshold, includes:

[0052] S31. Obtain the real-time load fluctuation amplitude of the energy storage system, the frequency of output power mutation of renewable energy equipment, and the grid voltage stability coefficient based on dynamic change information, and obtain the load fluctuation ratio and load fluctuation deviation value based on the real-time load fluctuation amplitude and the preset benchmark fluctuation amplitude.

[0053] S32. Obtain the historical average mutation frequency of renewable energy equipment, and obtain the mutation frequency deviation rate and the mutation frequency absolute deviation value based on the output power mutation frequency and the historical average mutation frequency.

[0054] S33. Obtain the dynamic influence factor by weighted summation of the mutation deviation rate and load fluctuation ratio, and determine the frequency decomposition threshold adjustment direction of the dynamic frequency decomposition module based on the dynamic influence factor to obtain the threshold adjustment trend.

[0055] S34. Obtain the basic adjustment amount of the frequency decomposition threshold based on the load fluctuation deviation value and the absolute deviation value of the sudden frequency, and obtain the real-time adjustment threshold of the frequency decomposition threshold based on the threshold adjustment trend, the basic adjustment amount and the grid voltage stability coefficient.

[0056] S35. Determine multiple frequency band intervals of the dynamic frequency decomposition module based on the real-time adjustment threshold, input the power difference sequence into the dynamic frequency decomposition module, and perform signal separation on the power difference sequence according to each frequency band interval to obtain multiple dynamic frequency components.

[0057] As described in steps S31-S35 above, the load fluctuation ratio is calculated by the ratio of the real-time load fluctuation amplitude to the preset reference fluctuation amplitude, the load fluctuation deviation value is calculated by the difference between the real-time load fluctuation amplitude and the preset reference fluctuation amplitude, the sudden frequency deviation rate is calculated by the difference between the output power sudden frequency and the historical average sudden frequency, and the sudden frequency absolute deviation value is calculated by the ratio of the sudden frequency deviation rate to the historical average sudden frequency.

[0058] By adjusting the threshold in real time, the upper and lower limits of the frequency included in the threshold are extracted to obtain the basic frequency boundary. Based on the basic frequency boundary and the frequency band division rules preset by the dynamic frequency decomposition module, the start and end ranges of the corresponding low frequency band, mid frequency band, and high frequency band are calculated. The start and end ranges of each frequency band are checked to see if they meet the bandwidth requirements of the signal processing within the module. If they do, they are determined as multiple frequency band intervals of the dynamic frequency decomposition module. The frequency band division rule can be to split according to the high and low frequency ratio.

[0059] This invention accurately captures the dynamic changes in load, renewable energy equipment output, and grid status in the current energy storage system by acquiring real-time load fluctuation amplitude, output power mutation frequency of renewable energy equipment, and grid voltage stability coefficient. By simultaneously considering multiple factors such as load fluctuation amplitude, power mutation frequency, and grid voltage stability coefficient, it helps to comprehensively evaluate the dynamic changes of the system, thereby more comprehensively reflecting the actual working status of the system and providing more accurate input for subsequent frequency decomposition threshold adjustment. By calculating the mutation frequency deviation rate and the absolute deviation value of the mutation frequency, the system can evaluate the difference between the mutation frequency and historical data in real time and make dynamic adjustments according to the degree of difference. This not only provides a refined identification of the mutation frequency but also helps the system identify abnormal fluctuations in the mutation frequency. The mutation frequency deviation rate and the absolute deviation value can quantify the degree of mutation, thereby helping the dynamic frequency decomposition module to make more flexible frequency adjustments when the load fluctuation is large. By weighted summation of dynamic influence factors, multiple factors such as mutation frequency deviation rate and load fluctuation ratio can be comprehensively considered to ensure that the system has a balanced response to different types of fluctuations.

[0060] The basic adjustment amount is obtained by calculating the load fluctuation deviation and the absolute deviation of the sudden frequency. This accurately reflects the actual changes in the current load and the output of renewable energy equipment. By quantifying the actual fluctuation amplitude, it ensures that the frequency decomposition adjustment amount matches the actual needs of the system, significantly improving the system's power balance capability. Through real-time threshold adjustment, the system can continuously track load changes and energy fluctuations in the energy storage system, ensuring that the frequency decomposition threshold remains consistent with the system state. Real-time adjustment significantly enhances the system's ability to cope with different fluctuation modes, avoiding power compensation failure due to inaccurate frequency decomposition. By combining real-time adjustments with the grid voltage stability coefficient, it ensures that the power supply... When grid fluctuations are significant, the system can automatically adjust the frequency accordingly. Based on the real-time adjusted threshold, the dynamic frequency decomposition module can accurately determine multiple frequency band intervals, ensuring refined signal separation. The decomposition of different frequency bands not only helps identify power fluctuations from different sources but also accurately maps different frequency components to specific energy storage system adjustment intervals, thereby improving the accuracy of power balance. By dynamically adjusting frequency band intervals, precise signal separation can be performed on power difference sequences at different frequencies. Through fine decomposition of power fluctuations, each frequency band can be adjusted independently, thus avoiding the transmission and amplification of overall power fluctuations and improving the power balance effect.

[0061] In one embodiment, step S4, which involves inputting multiple dynamic frequency components into the power prediction module and combining them with operating parameters to predict the changing trend of each dynamic frequency component within a preset time period to obtain the power prediction curve, includes:

[0062] S41. Obtain the historical feature set of each dynamic frequency component, and determine the prediction weight of the corresponding dynamic frequency component in the power prediction module based on each historical feature set and operating parameters.

[0063] S42. Input the corresponding dynamic frequency component into the power prediction module according to each prediction weight, and use the time-series prediction algorithm to make a preliminary prediction of the amplitude and phase changes of the corresponding dynamic frequency component within a preset time period in the future, so as to obtain the preliminary change trend of the corresponding dynamic frequency component.

[0064] S43. Obtain the capacity limit of the energy storage device in the operating parameters, and determine the upper and lower limits of the adjustable power of the energy storage device per unit time based on the capacity limit of the energy storage device to obtain the power adjustment range;

[0065] S44. Based on the power adjustment range, determine whether the predicted power value at each moment in the initial trend exceeds the power adjustment range.

[0066] If the predicted power value at a certain moment exceeds the power adjustment range, the predicted power value at that moment is corrected to the closest limit within the power adjustment range, and the corresponding correction trend is obtained.

[0067] S45. Based on each correction trend, the predicted values ​​of each dynamic frequency component at the same time are superimposed along the time axis to obtain the total power prediction value at each time.

[0068] S46. Plot a time-power axis with time as the horizontal axis and total power prediction value as the vertical axis. Plot the total power prediction value corresponding to each moment as a connection point on the time-power axis, and connect multiple connection points sequentially with curves to obtain the power prediction curve.

[0069] As described in steps S41-S46 above, the historical feature set includes historical fluctuation cycles, amplitude variation ranges, and differences between adjacent time points. The operating parameters include, for example, the response speed of the energy storage device and the power adjustment range. By extracting the fluctuation stability and prediction deviation rate of the dynamic frequency component based on the historical feature set, the predictive influence parameters of the component itself are obtained. The parameter matching degree is calculated based on the predictive influence parameters of the component itself and the operating parameters. Components with stable fluctuations and adapted to the device parameters are given higher weights based on the parameter matching degree, while components with violent fluctuations or low adaptability are given lower weights, thus obtaining the prediction weights of the corresponding dynamic frequency components. Fluctuation stability can be reflected by the fluctuation amplitude variance, and parameter matching degree can be reflected by the degree of adaptation between fluctuation stability and device response speed.

[0070] This invention determines the prediction weight of each frequency component by acquiring the historical feature set of dynamic frequency components and combining it with operating parameters. This enables personalized prediction adjustments and more accurately reflects the system's performance under different operating conditions. The historical feature set contains the variation patterns of the frequency components, while the operating parameters provide information on the current state of the system. By combining these two for weight setting, it ensures that the prediction weight of each frequency component reflects the actual operating conditions, thus providing more accurate power prediction. By using a time-series prediction algorithm, the historical variation trend of each dynamic frequency component can be taken into account, and predictions can be made in the context of time series. The time-series prediction algorithm can capture the amplitude and phase variation trends of each frequency component in the future time period, reducing the impact of inaccurate frequency response and thus improving the accuracy of power prediction. By considering the actual capacity limitations of the energy storage device, the consistency between the power prediction and the actual adjustable range is ensured. By calculating the upper and lower limits of the adjustable power, it is possible to effectively prevent the predicted value from exceeding the actual adjustment capability of the device, thereby enhancing the safety and reliability of the system.

[0071] By correcting power predictions in real time, the system ensures that the energy storage system does not exceed its adjustable range, thus preventing overload or unadjustable situations. Correcting over-predicted values ​​ensures that the entire system does not exceed its adjustment capabilities during dynamic adjustments, improving power balance stability. Real-time judgment and correction of predicted power values ​​ensures the entire system remains within a safe and operable range during predictive control, contributing to improved operational efficiency and stability. The total power prediction value is obtained by superimposing each corrected dynamic frequency component, considering not only the independence of each component but also effectively integrating them into a comprehensive power prediction value. This facilitates optimization of the overall power regulation scheme. The superimposed total power prediction value provides the energy storage system with a comprehensive scheduling plan, ensuring consistency and coordination between different frequency components, thereby achieving optimal power balance. Plotting time-power curves visually displays power change trends within a preset time period, helping energy storage system dispatchers and monitors understand power fluctuations and providing a basis for subsequent power regulation.

[0072] In one embodiment, step S5, which obtains the power compensation command of the energy storage system based on the charge / discharge constraints and the power prediction curve, includes:

[0073] S51. Obtain the constraint parameter set of the energy storage device according to the charging and discharging constraint conditions, and determine the feasible range of power compensation according to the constraint parameter set to obtain the compensation boundary conditions.

[0074] S52. Obtain the power prediction values ​​for multiple moments within a preset time period based on the power prediction curve, and obtain the corresponding power deviation value based on the power prediction value at each moment and the preset system allowable power range.

[0075] S53. Obtain the corresponding theoretical compensation power based on each power deviation value and compensation boundary condition, and sort the multiple theoretical compensation powers according to the corresponding time sequence to obtain the theoretical compensation power sequence.

[0076] S54. Correct the theoretical compensation power sequence according to the constraint parameter set to obtain the corrected compensation power sequence, and generate power compensation instructions in time order according to the corrected compensation power sequence. The power compensation instructions contain instruction information on compensation power and compensation duration at multiple times.

[0077] As described in steps S51-S54 above, the constraint parameter set includes the maximum charging / discharging power, charging / discharging duration limit, and remaining capacity threshold. The upper and lower limits of the single compensation power are determined based on the maximum charging / discharging power to obtain the power constraint interval. The longest duration of a single compensation is calculated based on the charging / discharging duration limit and the remaining capacity threshold to obtain the duration constraint interval. The power constraint interval and the duration constraint interval are integrated to form the feasible range of power compensation, thereby obtaining the compensation boundary conditions. The preset system allowable power range is a pre-set power value that can fluctuate to ensure the stable operation of the energy storage system and connected renewable energy equipment and loads. The power prediction curve has upper and lower limit ranges. When the predicted power value at a certain moment exceeds the upper limit of the range, power compensation is required by discharging the energy storage device. When the predicted value is lower than the lower limit of the range, power compensation is required by charging the energy storage device, thereby achieving power smoothing. The power prediction value at each moment within a preset time period is extracted from the power prediction curve. It is then determined whether the predicted power value at each moment exceeds the preset system allowable power range. If it does, the corresponding compensation time and power deviation value are obtained. The feasible range of power compensation means that the compensation power does not exceed the maximum charging and discharging power, and the compensation time does not exceed the charging and discharging time limit.

[0078] The deviation type is determined by identifying the positive or negative attribute of each power deviation value. The compensation direction and upper limit of compensation power are determined based on the deviation type and the maximum charging and discharging power in the compensation boundary conditions. The compensation amount that matches the absolute value of the deviation value and does not exceed the upper limit is calculated based on the magnitude of the deviation value and the upper limit of compensation power, thus obtaining the corresponding theoretical compensation power. Positive deviation indicates power excess, corresponding to discharge compensation, while negative deviation indicates power deficiency, corresponding to charging compensation. The theoretical compensation power that exceeds the remaining capacity carrying capacity is corrected by using the theoretical compensation power sequence and the remaining capacity threshold in the constraint parameter set, thus obtaining the corrected compensation power sequence.

[0079] This invention clarifies the maximum and minimum power range that energy storage devices can provide during power compensation by using a set of constraint parameters, thereby determining the feasible range of power compensation and ensuring the feasibility of power compensation commands. Defining compensation boundary conditions provides constraints for the implementation of subsequent compensation strategies, preventing unrealistic or inoperable compensation power commands in the system. By obtaining power prediction values ​​at multiple moments within a preset time period through power prediction curves, the power demand of the system in the future period can be understood in advance, and the trend of power change within that period can be accurately reflected. The theoretical compensation power sequence ensures that the calculation of theoretical compensation power is based on the power deviation at each moment, taking into account actual needs and the physical constraints of the equipment.

[0080] By arranging the theoretical compensation power in a time sequence, compensation commands can be generated step by step according to actual needs, enabling the energy storage system to make accurate power compensation at different points in time. This significantly improves the power balance effect and system response efficiency. The theoretical compensation power sequence is corrected based on the constraint parameter set of the energy storage device to ensure that the corrected compensation power sequence conforms to the actual capacity and operational limitations of the energy storage device. Through correction, the compensation power can be further optimized to ensure that the energy storage device does not exceed its operating range due to excessive load, while ensuring the accuracy and timeliness of power compensation. The corrected compensation power sequence will be converted into specific power compensation commands, including compensation power and compensation duration. The power compensation commands are generated in time sequence to ensure that the energy storage system can accurately execute the required compensation task at each moment, enabling the energy storage system to adapt more flexibly to power fluctuations and achieve accurate power balance.

[0081] In one embodiment, step S7, which iteratively corrects the parameters of the adaptive prediction model based on the power smoothing error, includes:

[0082] S71. Obtain the actual power curve based on multiple actual compensation power, real-time output power data and real-time load power data, and obtain the deviation segment between the power prediction curve and the actual power curve based on the power smoothing error.

[0083] S72. Obtain the deviation frequency characteristics based on the frequency characteristics of power fluctuations within the deviation range, and determine the type of model parameters that need to be corrected based on the deviation frequency characteristics.

[0084] S73. Obtain parameter correction adjustment values ​​based on power smoothing error, and perform preliminary correction on the model parameter types that need to be corrected in the adaptive prediction model based on the parameter correction adjustment values ​​and update the model parameter configuration to obtain the first corrected adaptive prediction model.

[0085] S74. Send the power compensation command to the initial correction adaptive prediction model to obtain the error value after iteration, and obtain the error difference based on the error value after iteration and the original power smoothing error.

[0086] S75. Determine whether the error difference is within the preset acceptable range;

[0087] If the error difference is greater than the upper limit of the preset acceptable range, the parameter correction adjustment value needs to be increased to obtain a new correction adjustment value;

[0088] If the error difference is within the preset acceptable range, there is no need to adjust the parameters to correct the adjustment value;

[0089] If the error difference is less than the lower limit of the preset acceptable range, the parameter correction adjustment value needs to be reduced to obtain a new correction adjustment value;

[0090] S76. Based on the newly corrected adjustment value, the types of model parameters that need to be corrected in the initial corrected adaptive prediction model are corrected again, and the model parameter configuration is updated until the error difference after iteration is within the preset acceptable range.

[0091] As described in steps S71-S76 above, the actual power curve is obtained by adding the actual compensation power and the real-time output power data at each moment and then subtracting the real-time load power data. Then, by connecting the actual total power data points at each moment with time as the horizontal axis and actual total power as the vertical axis, the actual power curve can be obtained.

[0092] This invention constructs an actual power curve by acquiring multiple real-time data on actual compensation power, output power, and load power. This curve reflects the power output characteristics of the system under different load conditions in real time. By comparing the actual power curve with the power prediction curve, the deviation segments between the predicted and actual power can be clearly identified, allowing the prediction model to optimize and correct these deviation segments, thus improving power balance accuracy. By analyzing the frequency characteristics of power fluctuations within the deviation segments, the frequency distribution characteristics of power fluctuations can be accurately identified. Since power fluctuations typically exhibit different frequency characteristics, fixed-frequency decomposition methods cannot adapt to dynamically changing power characteristics. By identifying the deviation frequency characteristics, the parameter types in the model can be modified in a targeted manner, ensuring dynamic adaptation to different power fluctuation characteristics. By calculating parameter correction adjustment values, the parameters of the prediction model can be adjusted according to the actual error. This not only allows for initial correction of model parameters but also enables dynamic updates based on the current power smoothing error. Adjusting the model based on the actual power smoothing effect allows the model to better adapt to the current power fluctuation situation, significantly improving the power balance effect and ensuring that the system can respond in real time and correct errors in power prediction.

[0093] By sending power compensation commands to the initially corrected adaptive prediction model, real-time power adjustment can be achieved. This allows for continuous updating and optimization of model parameters through a feedback mechanism, and correction of the prediction model based on the error values ​​after iteration. This ensures that the model can respond to system power changes in real time during continuous operation. By determining whether the error difference is within a preset acceptable range, the accuracy of model correction can be effectively controlled, helping to ensure that the error in the entire power balance process does not exceed the predetermined tolerance range. When the error difference is outside the preset range, adjusting the correction value and re-correcting the model parameters can further improve the model's adaptability, ensuring that each model correction is adjusted towards a more accurate target, thereby gradually approaching the optimal power balance state.

[0094] like Figure 2 As shown, this application also provides a power smoothing system for an energy storage system based on model predictive control, comprising:

[0095] The first acquisition module is used to acquire multiple real-time output power data of renewable energy devices connected to the energy storage system, dynamic change information of the energy storage system, multiple real-time load power data and operating parameters of the energy storage devices, and to acquire a power difference sequence based on the multiple real-time output power data and real-time load power data.

[0096] A construction module is used to construct an adaptive prediction model based on the power difference sequence and operating parameters, wherein the adaptive prediction model includes a dynamic frequency decomposition module and a power prediction module;

[0097] The adjustment decomposition module is used to input the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model, and to automatically adjust the frequency decomposition threshold to obtain the real-time adjustment threshold, and decompose the power difference sequence into multiple dynamic frequency components according to the real-time adjustment threshold.

[0098] The prediction module is used to input multiple dynamic frequency components into the power prediction module and combine them with operating parameters to predict the changing trend of each dynamic frequency component within a future preset time period, thereby obtaining a power prediction curve.

[0099] The second acquisition module is used to acquire the charging and discharging constraints of the energy storage device, and to acquire the power compensation command of the energy storage system based on the charging and discharging constraints and the power prediction curve.

[0100] The third acquisition module is used to send the power compensation command to the energy storage device to obtain the corresponding actual compensation power, and to obtain the power smoothing error based on the actual compensation power.

[0101] The correction module is used to iteratively correct the parameters of the adaptive prediction model based on the power smoothing error, so as to optimize the dynamic frequency decomposition accuracy and achieve continuous power smoothing of the energy storage system.

[0102] It should be noted that each module and unit in the model predictive control-based energy storage system power smoothing system corresponds one-to-one with the steps in the model predictive control-based energy storage system power smoothing method.

[0103] like Figure 3 As shown, this application also provides a computer device, which can be a server, and its internal structure can be as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores all data required for the process of a model predictive control-based power smoothing method for energy storage systems. The network interface is used for communication with external terminals via a network connection. When the processor executes the computer program, it implements the model predictive control-based power smoothing method for energy storage systems.

[0104] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer equipment on which the present application is applied.

[0105] An embodiment of this application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements any of the above-described model predictive control-based power smoothing methods for energy storage systems.

[0106] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0107] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0108] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A model predictive control based power smoothing method for energy storage system, characterized in that, The method comprises the following steps: acquiring a plurality of output power real-time data of renewable energy equipment connected to an energy storage system, dynamic change information of the energy storage system, a plurality of load power real-time data, and operating parameters of the energy storage equipment, and acquiring a power difference sequence according to the plurality of output power real-time data and load power real-time data; constructing an adaptive prediction model according to the power difference sequence and the operating parameters, wherein the adaptive prediction model comprises a dynamic frequency decomposition module and a power prediction module; inputting the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model to automatically adjust a frequency decomposition threshold, obtaining a real-time adjustment threshold, and decomposing the power difference sequence into a plurality of dynamic frequency components according to the real-time adjustment threshold; inputting the plurality of dynamic frequency components into the power prediction module and combining the operating parameters to predict the change trend of each dynamic frequency component in a future preset time period, and obtaining a power prediction curve; acquiring charge and discharge constraints of the energy storage equipment, and acquiring a power compensation instruction of the energy storage system according to the charge and discharge constraints and the power prediction curve; sending the power compensation instruction to the energy storage equipment to output a corresponding actual compensation power, and acquiring a power smoothing error according to the actual compensation power; iteratively correcting parameters of the adaptive prediction model according to the power smoothing error to optimize the dynamic frequency decomposition accuracy and realize continuous smoothing of the power of the energy storage system.

2. The model predictive control based power smoothing method for energy storage system according to claim 1, wherein, The step of constructing the adaptive prediction model according to the power difference sequence and the operating parameters comprises: acquiring a key feature set of the power difference sequence, and determining a core input dimension of the adaptive prediction model according to the key feature set and the operating parameters to obtain a model input parameter system; selecting a dynamic frequency decomposition algorithm and a power prediction algorithm as a basic framework of the dynamic frequency decomposition module and the power prediction module of the adaptive prediction model according to the model input parameter system to obtain a module initial framework; pre-setting an initial frequency threshold of the dynamic frequency decomposition module according to a fluctuation frequency range in the key feature set to obtain a module pre-configuration parameter; initializing a weight parameter of the power prediction module according to an equipment charge and discharge efficiency and a decay coefficient in the operating parameters to obtain a module initial weight; integrating the dynamic frequency decomposition module and the power prediction module according to the module initial framework, the module pre-configuration parameter, and the module initial weight to construct the adaptive prediction model.

3. The model predictive control based power smoothing method for energy storage system according to claim 1, wherein, The step of inputting the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model to automatically adjust the frequency decomposition threshold, obtaining a real-time adjustment threshold, and decomposing the power difference sequence into a plurality of dynamic frequency components according to the real-time adjustment threshold comprises: acquiring a real-time load fluctuation amplitude of the energy storage system, an output power mutation frequency of the renewable energy equipment, and a power grid voltage stability coefficient according to the dynamic change information, and acquiring a load fluctuation ratio and a load fluctuation deviation value according to the real-time load fluctuation amplitude and a preset reference fluctuation amplitude; acquiring a historical average mutation frequency of the renewable energy equipment, and acquiring a mutation frequency deviation rate and a mutation frequency absolute deviation value according to the output power mutation frequency and the historical average mutation frequency; According to the mutation deviation rate and the load fluctuation ratio, a dynamic influence factor is obtained, and a frequency decomposition threshold adjustment direction of a dynamic frequency decomposition module is determined according to the dynamic influence factor, so as to obtain a threshold adjustment trend; According to the load fluctuation deviation value and the mutation frequency absolute deviation value, a basic adjustment amount of the frequency decomposition threshold is obtained, and a real-time adjustment threshold of the frequency decomposition threshold is obtained according to the threshold adjustment trend, the basic adjustment amount and a power grid voltage stability coefficient; According to the real-time adjustment threshold, a plurality of frequency band intervals of the dynamic frequency decomposition module are determined, the power difference value sequence is input into the dynamic frequency decomposition module, and signal separation is performed on the power difference value sequence according to each frequency band interval, so as to obtain a plurality of dynamic frequency components.

4. The model predictive control based power smoothing method for energy storage system according to claim 1, wherein, The step of inputting the plurality of dynamic frequency components into the power prediction module and combining the operating parameters to predict the change trend of each dynamic frequency component in a future preset time period to obtain a power prediction curve comprises: A historical feature set of each dynamic frequency component is obtained, and a prediction weight of a corresponding dynamic frequency component in the power prediction module is determined according to each historical feature set and the operating parameters; According to each prediction weight, the corresponding dynamic frequency component is input into the power prediction module, and a time sequence prediction algorithm is used to preliminarily predict the amplitude and phase change of the corresponding dynamic frequency component in the future preset time period, so as to obtain a preliminary change trend of the corresponding dynamic frequency component; The capacity limit of the energy storage device in the operating parameters is obtained, and the upper limit and the lower limit of the adjustable power of the energy storage device in a unit time are determined according to the capacity limit of the energy storage device, so as to obtain a power adjustment range; According to the power adjustment range, it is judged whether the predicted power value at each moment in the compared preliminary change trend exceeds the power adjustment range; If the predicted power value at a certain moment exceeds the power adjustment range, the predicted power value at the moment is corrected to the nearest limit value within the power adjustment range to obtain a corresponding corrected change trend; According to each corrected change trend, the predicted values of each dynamic frequency component at the same moment are superimposed according to the time axis to obtain a total power prediction value at each moment; A time-power axis is drawn with time as the horizontal axis and total power prediction value as the vertical axis, the total power prediction value corresponding to each moment is taken as a connection point and drawn on the time-power axis, and a plurality of connection points are sequentially connected by a curve to obtain a power prediction curve.

5. The model predictive control based power smoothing method for energy storage system according to claim 1, wherein, The step of obtaining a power compensation instruction of the energy storage system according to the charge-discharge constraint condition and the power prediction curve comprises: According to the charge-discharge constraint condition, a constraint parameter set of the energy storage device is obtained, and a feasible range of power compensation is determined according to the constraint parameter set to obtain a compensation boundary condition; According to the power prediction curve, power prediction values at a plurality of moments in a future preset time period are obtained, and a corresponding power deviation value is obtained according to the power prediction value at each moment and a preset system allowable power range; According to each power deviation value and the compensation boundary condition, a corresponding theoretical compensation power is obtained, and a theoretical compensation power sequence is obtained by sequentially sorting a plurality of theoretical compensation powers according to corresponding moments. According to the constraint parameter set, the theoretical compensation power sequence is modified to obtain a modified compensation power sequence, and power compensation instructions are generated in chronological order according to the modified compensation power sequence, wherein the power compensation instructions include instruction information of compensation power at multiple time points and compensation time length.

6. The model predictive control-based energy storage system power smoothing method of claim 1, wherein, The step of iteratively correcting the parameters of the adaptive prediction model according to the power smoothing error comprises: An actual power curve is obtained according to the multiple actual compensation powers, output power real-time data and load power real-time data, and a deviation section of the power prediction curve and the actual power curve is obtained according to the power smoothing error; A deviation frequency feature is obtained according to the frequency characteristics of power fluctuations in the deviation section, and the type of model parameters that need to be corrected is determined according to the deviation frequency feature; A parameter correction adjustment value is obtained according to the power smoothing error, and the type of model parameters that need to be corrected in the adaptive prediction model is preliminarily corrected and the model parameter configuration is updated according to the parameter correction adjustment value, to obtain a first-corrected adaptive prediction model; The power compensation instructions are sent to the first-corrected adaptive prediction model, an error value after iteration is obtained, and an error difference value is obtained according to the error value after iteration and the original power smoothing error; It is judged whether the error difference value is within a preset acceptable range; If the error difference value is greater than the upper limit of the preset acceptable range, the parameter correction adjustment value needs to be increased to obtain a new correction adjustment value; If the error difference value is within the preset acceptable range, the parameter correction adjustment value does not need to be adjusted; If the error difference value is less than the lower limit of the preset acceptable range, the parameter correction adjustment value needs to be reduced to obtain a new correction adjustment value; The type of model parameters that need to be corrected in the first-corrected adaptive prediction model is corrected again and the model parameter configuration is updated according to the new correction adjustment value, until the error difference value after iteration is within the preset acceptable range.

7. A model predictive control based power smoothing system for an energy storage system, characterized in that, Comprise: The first acquisition module is used for acquiring multiple output power real-time data of renewable energy equipment connected with the energy storage system, dynamic change information of the energy storage system, multiple load power real-time data and operating parameters of the energy storage equipment, and obtaining a power difference value sequence according to the multiple output power real-time data and load power real-time data; The construction module is used for constructing an adaptive prediction model according to the power difference value sequence and operating parameters, wherein the adaptive prediction model comprises a dynamic frequency decomposition module and a power prediction module; The adjustment decomposition module is used for inputting the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model, and inputting the dynamic change information into the dynamic frequency decomposition module of the adaptive prediction model to automatically adjust the frequency decomposition threshold to obtain a real-time adjustment threshold, and decomposing the power difference value sequence into multiple dynamic frequency components according to the real-time adjustment threshold; The prediction module is used for inputting the multiple dynamic frequency components into the power prediction module and predicting the change trend of each dynamic frequency component in a future preset time period in combination with the operating parameters to obtain a power prediction curve; The second acquisition module is configured to acquire a charge-discharge constraint condition of the energy storage device, and acquire a power compensation instruction of the energy storage system according to the charge-discharge constraint condition and the power prediction curve; The third acquisition module is configured to send the power compensation instruction to the energy storage device to obtain a corresponding actual compensation power, and acquire a power smoothing error according to the actual compensation power; The correction module is configured to iteratively correct parameters of the adaptive prediction model according to the power smoothing error, so as to optimize the dynamic frequency decomposition accuracy and realize continuous smoothing of the power of the energy storage system.

8. The model predictive control-based energy storage system power smoothing system of claim 7, wherein, The second acquisition module comprises: The first acquisition unit is configured to acquire a constraint parameter set of the energy storage device according to the charge-discharge constraint condition, and determine a feasible range of power compensation according to the constraint parameter set to obtain a compensation boundary condition; The second acquisition unit is configured to acquire power prediction values at multiple time points in a future preset time period according to the power prediction curve, and acquire a corresponding power deviation value according to the power prediction value at each time point and a preset system allowed power range; The third acquisition unit is configured to acquire a corresponding theoretical compensation power according to each power deviation value and the compensation boundary condition, and sort the multiple theoretical compensation powers according to the corresponding time points in sequence to obtain a theoretical compensation power sequence; The correction generation unit is configured to correct the theoretical compensation power sequence according to the constraint parameter set to obtain a corrected compensation power sequence, and generate a power compensation instruction in time sequence according to the corrected compensation power sequence, wherein the power compensation instruction comprises instruction information of compensation power and compensation time length at multiple time points. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the method in any one of claims 1 to 6.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 6.

Images