Photovoltaic power generation system energy storage management method and computer equipment based on dynamic regulation

Abstract

The present invention discloses a method and computer equipment for managing energy storage of photovoltaic power generation systems based on dynamic regulation, wherein the method includes: S1, acquisition and preliminary analysis of grid load data; S2, construction of a dynamic dispatch model for grid costs; S3, calculation and optimization of the number of energy storage units; S4, construction of an optimal dynamic dispatch model; S5, matching of historical data and strategy optimization. The present invention aims to solve the problem of unstable power output of photovoltaic power generation in the prior art due to the influence of weather and day and night changes. By acquiring information such as grid load data, photovoltaic power generation historical data and the number of energy storage units, the present invention makes decisions using historical data and load forecasting models to ensure that the energy storage management strategy is more scientific and reasonable, aiming to optimize energy storage management, improve the utilization rate and economic benefits of photovoltaic power generation, and achieve significant improvements in the economy and stability of photovoltaic power generation systems with effective energy storage management methods.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and in particular to a photovoltaic power generation system energy storage management method and computer equipment based on dynamic regulation. Background Art

[0002] The power output of photovoltaic systems is affected by weather, sunshine hours, and seasonal changes, which can lead to unstable power generation. In order to improve the efficiency and economy of photovoltaic power generation, the application of energy storage management technology has become particularly important. Energy storage technology can effectively balance the supply and demand of electricity, improve the utilization rate of renewable energy, and promote the development of smart grids. Traditional photovoltaic power generation systems generally rely on simple energy storage methods, such as battery energy storage or pumped storage. However, these technologies often have problems such as inflexible management and slow response. For example, although battery energy storage systems can provide short-term power support, it is difficult to effectively allocate power resources during peak loads. In addition, traditional energy storage management methods often lack real-time data analysis and prediction, and are difficult to adapt to rapidly changing grid load demands, resulting in waste of resources.

[0003] Although traditional technologies have met the needs of photovoltaic power generation to a certain extent, they still have obvious defects in power supply stability and economic benefits: first, the fixed energy storage capacity limits the system's responsiveness under high load conditions, resulting in unstable power supply; second, the lack of dynamic scheduling of energy storage management cannot fully utilize the economic advantages brought by electricity price fluctuations, increasing the overall operating costs. Therefore, a more flexible and intelligent energy storage management method is urgently needed to achieve the optimal operation of photovoltaic power generation systems and improve their economic benefits and power supply stability. Summary of the invention

[0004] The present invention provides a photovoltaic power generation system energy storage management method and computer equipment based on dynamic regulation with more stable power supply and higher economic benefits, which can solve at least one of the above technical problems.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A photovoltaic power generation system energy storage management method based on dynamic regulation includes the following steps:

[0007] S1. Acquisition and preliminary analysis of power grid load data: Preliminary setting of the first energy storage ratio to obtain the fluctuation characteristics of historical power grid load data;

[0008] S2. Construction of a dynamic dispatch model for power grid cost: Based on the analysis of historical power grid load data and time characteristics obtained in S1, a power grid operation cost model under each first energy storage ratio is constructed;

[0009] S3. Calculation and optimization of the number of energy storage units: Based on the grid operation cost model of S2, combined with the grid operation cost and the comprehensive cost of the photovoltaic power generation system, the total cost is weighted and calculated to determine the optimal energy storage configuration;

[0010] S4, construction of an optimal dynamic scheduling model: setting an objective function representing the minimization of the total cost, and determining the optimal number of energy storage units and the corresponding dynamic scheduling strategy that minimize the objective function based on the optimal energy storage configuration of S3;

[0011] S5. Matching of historical data and strategy optimization: Based on the optimal number of energy storage units and the optimal dynamic scheduling strategy of S4, quantify and compare the model prediction ability and the actual operation effect, judge the model prediction accuracy, optimize and improve the applicability of the model.

[0012] Furthermore, the S1 further comprises:

[0013] S11, extract the peak and valley values ​​in the historical power grid load data, and set the peak value as P max , the valley value is P min , calculate the variation range R of the power grid load data, R=P max -P min ;

[0014] S12, determining whether the variation range R is within the preset range, if so, preliminarily setting the first energy storage ratio, if not, first adjusting the energy storage ratio coefficient k, and then preliminarily setting the first energy storage ratio;

[0015] S13. Use time series analysis methods to identify periodic and trend changes in electricity demand and analyze the fluctuation characteristics of historical power grid load data;

[0016] S14. Predict grid load demand based on fluctuation characteristics.

[0017] Furthermore, in S13, the historical power grid load data is modeled using an autoregressive moving average model, and the modeling formula is:

[0018]

[0019] in:

[0020] t is the time point, which represents the current moment in the grid load data sequence;

[0021] L t is the historical grid load data;

[0022] μ is the mean of the historical power grid load data series;

[0023] Ф and θ are the autoregressive and moving average coefficients, respectively;

[0024] ε is the white noise term;

[0025] i is the lag order of the autoregressive term, ranging from 1 to p, indicating the degree of influence of historical grid load data on current load;

[0026] j is the lag order of the moving average term, ranging from 1 to q, indicating the degree of influence of the historical white noise term on the current load;

[0027] p is the order of the autoregressive model, indicating the maximum number of load lag terms considered;

[0028] q is the order of the moving average model, which indicates the maximum number of white noise lags considered.

[0029] Furthermore, in S14, a linear regression method is used to construct a prediction model based on historical power grid load data, and the power grid load prediction value is initially set to L pred , the prediction formula is:

[0030]

[0031] Among them, a is the rate at which the power grid load data changes over time, b is the initial load value, which can be obtained by fitting the historical power grid load data, and t is the time point, which represents the current moment in the power grid load data sequence.

[0032] Furthermore, the S2 further comprises:

[0033] S21, according to the peak value P in the historical power grid load data max and valley value P min , determine the time characteristics of the power grid load, the time characteristics can be expressed as load peak period T peak and load valley period T valley express;

[0034] S22, combined with historical grid load data L t and the time characteristics of the power grid load, construct a power grid load prediction model, and further set the power grid load prediction value to be L pred (t), the prediction formula is:

[0035]

[0036] in, is the smoothing coefficient, ranging from 0 to 1;

[0037] S23, obtaining the electricity price curve C at each first energy storage ratio price (t), different electricity price curves reflect the electricity price fluctuations in different time periods, and the electricity price fluctuation range C is calculated range , C range =Cmax -C min ;

[0038] S24, combined with the grid load prediction value L pred (t) and electricity price curve C price (t), construct the grid operation cost model C under each first energy storage ratio cost , the modeling formula is:

[0039]

[0040] By adding up the grid operation costs in different time periods, we can get the total grid operation cost C total :

[0041]

[0042] Where n is the total number of time periods.

[0043] Furthermore, the S3 further includes:

[0044] S31, obtaining the historical power generation and peak power limit of the photovoltaic power generation system, and combining them with the first energy storage ratio that is initially set, to determine the initial number of energy storage units;

[0045] S32. Based on the historical operation data of the photovoltaic power generation system, a comprehensive cost prediction model is constructed to reflect the cost performance under different numbers of energy storage units;

[0046] S33. According to the analysis results of historical operation data, determine whether the power generation of the photovoltaic power generation system meets the power demand of the power grid. If so, then calculate the power grid operation cost. If not, adjust the power generation capacity of the energy storage unit and then calculate the power grid operation cost.

[0047] S34, when calculating the grid operation cost, based on the grid load prediction model and the grid operation cost model, the predicted grid load data is combined with the electricity prices in different time periods to calculate the grid operation cost under different numbers of energy storage units;

[0048] S35. When calculating the comprehensive cost of the photovoltaic power generation system, by comparing the historical power generation of the photovoltaic power generation system, evaluating the power generation efficiency and cost-effectiveness under different numbers of energy storage units, the number of energy storage units corresponding to the lowest comprehensive cost of the photovoltaic power generation system is the best choice;

[0049] S36. Perform weighted calculation on the predicted total cost under all energy storage unit quantities to find the predicted minimum comprehensive cost C under each first energy storage ratio condition. min (S ratio ) and its corresponding number of energy storage units N optimal , the current number of energy storage units is the optimal energy storage configuration.

[0050] Furthermore, the S4 further comprises:

[0051] S41. Taking the number of energy storage units as the decision variable, set an objective function Z, which represents the minimization of the total cost, to ensure that under each energy storage ratio condition, the photovoltaic power generation system can reduce the operating cost while meeting the grid demand. Z is expressed as:

[0052]

[0053] Among them, C min (S ratio ) is the lowest comprehensive cost prediction, C grid (N storage ) is the grid operation cost, C PV (N storage ) is the comprehensive cost of the photovoltaic power generation system;

[0054] S42, setting constraints for the objective function Z, and determining the optimal number of energy storage units N that minimizes the objective function Z by adjusting the constraints and using an algorithm for solving the optimization problem. optimal And the corresponding dynamic scheduling strategy to form the optimal dynamic scheduling strategy.

[0055] Furthermore, the S5 further includes:

[0056] S51. Based on the optimal number of energy storage units and the optimal dynamic scheduling strategy, regression analysis methods are used in the historical data matching process to quantify the model prediction ability and actual operation effect;

[0057] S52. Calculate the mean square error (MSE) to further evaluate the prediction effect of the model. The MSE calculation formula is:

[0058]

[0059] in, is the actual operating energy storage efficiency of the i-th group, The predicted energy storage efficiency of the model for group i, n is the number of data points;

[0060] S53, judging whether the model prediction is accurate according to the calculation results, if so, maintaining the current model, if not, adjusting the parameters of the current model until the model prediction capability matches the actual operation effect;

[0061] S54, using the cross-validation method, divide the historical data into multiple subsets, train and validate the model one by one, and more comprehensively evaluate the performance of the model under different conditions;

[0062] S55. Identify the key factors that affect energy storage efficiency, construct a multiple linear regression model, and introduce these key factors into the multiple linear regression model, which is expressed as:

[0063]

[0064] Among them, X 1 , X 2 ...X k Represent different key factors, β 0 is the intercept term, β 1 , β 2 ...β k They represent different slopes respectively, and ò is the error term.

[0065] Furthermore, in S51, a regression model is defined to represent the relationship between the actual operation data and the predicted data. Let Y represent the energy storage efficiency of the actual operation, and X represent the energy storage efficiency predicted by the model. The regression model is expressed as:

[0066]

[0067] Among them, β 0 is the intercept term, β 1 is the slope and ò is the error term.

[0068] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the photovoltaic power generation system energy storage management method based on dynamic regulation.

[0069] The beneficial effects of the present invention are embodied in:

[0070] 1. Improve the utilization rate of photovoltaic power generation: By optimizing the energy storage management strategy, sufficient power support can be provided during the peak load period of the power grid, thereby significantly improving the utilization rate of photovoltaic power generation, reducing the energy waste caused by unstable power output, and improving overall energy efficiency.

[0071] 2. Enhance economic efficiency and cost-effectiveness: Dynamically adjust the number of energy storage units and scheduling strategies to enable the photovoltaic power generation system to flexibly respond to fluctuations in electricity prices and changes in grid demand, reduce overall operating costs, thereby improving the economy of the system and enhancing return on investment.

[0072] 3. Improve system stability and reliability: By analyzing grid load and photovoltaic power generation data in real time, ensure that the system can respond quickly to changes in power demand, enhance the photovoltaic power generation system's ability to adapt to grid fluctuations, ensure power supply stability, and reduce the risk of power outages.

[0073] 4. Support the sustainable development of renewable energy: Effective energy storage management solutions promote the rational use of photovoltaic power generation, provide a solid foundation for the integration and development of renewable energy, comply with global sustainable development goals, help reduce dependence on traditional fossil energy, and promote green energy transformation.

[0074] 5. Data-driven decision support: Use historical data and load forecasting models to make decisions, ensure that energy storage management strategies are more scientific and reasonable, improve the accuracy and efficiency of system operation, and promote the development of intelligent management. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0076] Figure 1 It is a schematic diagram of the overall process of a photovoltaic power generation system energy storage management method based on dynamic regulation according to an embodiment of the present invention.

[0077] Figure 2 It is a flowchart of the power grid load data analysis process according to an embodiment of the present invention.

[0078] Figure 3 It is a flowchart of the construction process of the dynamic dispatch model of power grid cost according to an embodiment of the present invention.

[0079] Figure 4 It is a flowchart of energy storage unit calculation and optimization process according to an embodiment of the present invention.

[0080] Figure 5 It is a flowchart of the optimal dynamic scheduling strategy formation process of an embodiment of the present invention.

[0081] Figure 6 It is a flowchart of historical data and strategy optimization process according to an embodiment of the present invention.

[0082] Figure 7 It is a structural block diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0083] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. In the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0084] It should be noted that if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, "multiple" refers to more than two.

[0085] See also Figure 1 The embodiment of the present invention provides a method for managing energy storage of a photovoltaic power generation system based on dynamic regulation, comprising the following steps:

[0086] S1. Acquisition and preliminary analysis of grid load data: Preliminarily set the first energy storage ratio and obtain the fluctuation characteristics of historical grid load data.

[0087] Obtaining grid load data is the first step in implementing this method. This data is usually provided by power companies and covers electricity demand in different time periods. Through in-depth analysis of grid load data, the peak and valley values ​​of historical grid load data can be extracted, which are of great significance for subsequent energy storage management. With the peak value as a reference, a preliminary first energy storage ratio can be set to provide sufficient power support for the peak load period of the grid. At the same time, analyzing the fluctuation characteristics of historical grid load data can also help formulate more flexible energy storage strategies to cope with electricity demand in different time periods.

[0088] S2. Construction of a dynamic dispatch model for power grid cost: Based on the analysis of the historical power grid load data and time characteristics obtained in S1, a power grid operation cost model under each first energy storage ratio is constructed.

[0089] After setting the first energy storage ratio, constructing a dynamic dispatch model for grid costs is a key step in achieving optimized dispatch.

[0090] S3. Calculation and optimization of the number of energy storage units: Based on the grid operation cost model of S2, combined with the grid operation cost and the comprehensive cost of the photovoltaic power generation system, the total cost is weighted and calculated to determine the optimal energy storage configuration.

[0091] Through this series of calculations and optimization steps, the number of energy storage units under the optimal energy storage configuration can be determined, thereby ensuring that the photovoltaic power generation system can achieve the best economic performance under various operating conditions.

[0092] S4. Construction of an optimal dynamic scheduling model: setting an objective function representing the minimization of total cost, and determining the optimal number of energy storage units that minimize the objective function and the corresponding dynamic scheduling strategy based on the optimal energy storage configuration of S3.

[0093] In order to minimize the overall cost, building an optimal dynamic scheduling model is a key step. The core goal of this model is to reasonably configure the number of energy storage units and optimize the dynamic scheduling strategy based on the prediction of the lowest comprehensive cost.

[0094] S5. Matching of historical data and strategy optimization: Based on the optimal number of energy storage units and the optimal dynamic scheduling strategy of S4, quantify and compare the model prediction ability and the actual operation effect, judge the model prediction accuracy, optimize and improve the applicability of the model.

[0095] After the number of energy storage units and the dynamic dispatch strategy are determined, the last step is to match these optimal solutions with historical data. The key to this process is to evaluate the adaptability and accuracy of the current model by analyzing historical data. By comparing the model output with historical operating conditions, potential problems can be discovered and necessary adjustments can be made to the model to improve its effectiveness in practical applications.

[0096] See also Figure 1 and Figure 2 In this embodiment, S1 further includes:

[0097] S11, extract the peak and valley values ​​in the historical power grid load data, and set the peak value as P max , the valley value is P min , calculate the variation range R of the power grid load data, R=P max -P min ;

[0098] S12, determining whether the variation range R is within the preset range, if so, preliminarily setting the first energy storage ratio, if not, first adjusting the energy storage ratio coefficient k, and then preliminarily setting the first energy storage ratio;

[0099] Through this variation range R, we can understand the fluctuation characteristics of power demand and thus set the energy storage management strategy. Taking the peak value as a reference, we can set the initial first energy storage ratio S ratio , the calculation formula is:

[0100]

[0101] Where k is the energy storage ratio coefficient, which is usually between 0.1 and 0.5, depending on the specific application scenario. By setting an appropriate energy storage ratio, it can ensure that sufficient power support is provided during the peak load period of the power grid;

[0102] S13. Use time series analysis methods to identify periodic and trend changes in electricity demand and analyze the fluctuation characteristics of historical power grid load data;

[0103] In S13, the autoregressive moving average model (ARIMA) is used to model the historical power grid load data, and the modeling formula is:

[0104]

[0105] in:

[0106] t is the time point, which represents the current moment in the grid load data sequence;

[0107] L t is the historical grid load data;

[0108] μ is the mean of the historical power grid load data series;

[0109] Ф and θ are the autoregressive and moving average coefficients, respectively;

[0110] ε is the white noise term;

[0111] i is the lag order of the autoregressive term, ranging from 1 to p, indicating the degree of influence of historical grid load data on current load;

[0112] j is the lag order of the moving average term, ranging from 1 to q, indicating the degree of influence of the historical white noise term on the current load;

[0113] p is the order of the autoregressive model, indicating the maximum number of load lag terms considered;

[0114] q is the order of the moving average model, indicating the maximum number of white noise lags considered;

[0115] Through parameter estimation, the power demand in future periods can be predicted, providing a basis for subsequent energy storage management;

[0116] For example, assuming that the peak value of the historical power grid load data in a certain period of time is 200MW and the valley value is 80MW, then the variation range R is 120MW. If k is set to 0.3, then the first energy storage ratio S ratio 60MW, which means that during the peak load period of the power grid, the energy storage system needs to have an electricity supply capacity of at least 60MW;

[0117] Further analysis of the fluctuation characteristics of historical loads can provide more flexible strategies for the design of energy storage systems, such as giving priority to releasing energy storage when the grid load is close to its peak to ensure the stability of power supply;

[0118] S14. Predicting grid load demand based on fluctuation characteristics;

[0119] In order to achieve real-time energy storage scheduling, it is also necessary to consider the prediction of power demand. In S14, a linear regression method is used to build a prediction model based on historical power grid load data, and the power grid load prediction value is initially set to L pred , the prediction formula is:

[0120]

[0121] Among them, a is the rate at which the power grid load data changes over time, b is the initial load value, which can be obtained by fitting the historical power grid load data, and t is the time point, which represents the current moment in the power grid load data sequence.

[0122] Combining the above steps, it is possible to accurately grasp and flexibly dispatch the grid load during the dynamic regulation of energy storage management, thereby improving the overall efficiency and economy of the photovoltaic power generation system.

[0123] See also Figure 1 and Figure 3 In this embodiment, S2 further includes:

[0124] S21, according to the peak value P in the historical power grid load data max and valley value P min , determine the time characteristics of the power grid load, the time characteristics can be expressed as load peak period T peak and load valley period T valley express;

[0125] This feature will directly affect the operation strategy of the power grid and provide a basis for the optimization of the energy storage system;

[0126] S22, combined with historical grid load data L t The load prediction model of the power grid can be constructed by using the autoregressive model or exponential smoothing method to predict the load. The load prediction value of the power grid is further set to L pred (t), the prediction formula is:

[0127]

[0128] in, is the smoothing coefficient, ranging from 0 to 1;

[0129] Through multiple iterations of historical grid load data, the grid load conditions at different times in the future can be predicted, providing necessary data support;

[0130] S23, obtaining the electricity price curve C at each first energy storage ratio price (t), different electricity price curves reflect the electricity price fluctuations in different time periods, and the electricity price fluctuation range C is calculated range , C range =Cmax -C min ;

[0131] S24, combined with the grid load prediction value L pred (t) and electricity price curve C price (t), construct the grid operation cost model C under each first energy storage ratio cost , the modeling formula is:

[0132]

[0133] By adding up the grid operation costs in different time periods, we can get the total grid operation cost C total :

[0134]

[0135] Where n is the total number of time periods.

[0136] This model provides an accurate calculation method for the grid operation cost for each time period, which not only provides important data support for the subsequent calculation of the number of energy storage units, but also lays the foundation for optimizing the operation efficiency of the grid. For example, assuming that the grid load predicted at a certain moment is 150MW and the corresponding electricity price is 0.1 yuan / kWh, the operation cost is 15 yuan. If the electricity price rises to 0.2 yuan / kWh during peak hours and the grid load increases to 200MW, the operation cost at this time is 40 yuan. By constructing a dynamic dispatch model, it is possible to flexibly respond to electricity price fluctuations and load changes and optimize the overall operation efficiency of the grid.

[0137] See also Figure 1 and Figure 4 In this embodiment, S3 further includes:

[0138] S31, obtaining the historical power generation and peak power limit of the photovoltaic power generation system, and combining them with the first energy storage ratio that is initially set, to determine the initial number of energy storage units;

[0139] After obtaining the historical power generation and peak power limit of the photovoltaic power generation system, calculating the number of energy storage units is one of the key steps. The selection of the number of energy storage units directly affects the overall efficiency and economy of the photovoltaic power generation system.

[0140] S32. Based on the historical operation data of the photovoltaic power generation system, a comprehensive cost prediction model is constructed to reflect the cost performance under different numbers of energy storage units;

[0141] By analyzing historical operating data, it is possible to identify the characteristics of the photovoltaic power generation system in different operating periods, and then adjust the power generation capacity of the energy storage unit to better adapt to the needs of the power grid. For example, during peak periods, the photovoltaic power generation system needs to be able to provide additional power support, which requires the energy storage unit to have sufficient capacity;

[0142] S33. According to the analysis results of historical operation data, determine whether the power generation of the photovoltaic power generation system meets the power demand of the power grid. If so, then calculate the power grid operation cost. If not, adjust the power generation capacity of the energy storage unit and then calculate the power grid operation cost.

[0143] S34, when calculating the grid operation cost, based on the grid load prediction model and the grid operation cost model, the predicted grid load data is combined with the electricity prices in different time periods to calculate the grid operation cost under different numbers of energy storage units;

[0144] This calculation process not only provides the necessary data support, but also helps identify the optimal energy storage configuration;

[0145] S35. When calculating the comprehensive cost of the photovoltaic power generation system, by comparing the historical power generation of the photovoltaic power generation system, evaluating the power generation efficiency and cost-effectiveness under different numbers of energy storage units, the number of energy storage units corresponding to the lowest comprehensive cost of the photovoltaic power generation system is the best choice;

[0146] That is, if the comprehensive cost of the photovoltaic power generation system under a certain quantity condition is significantly lower than that under other quantity conditions, then this quantity may be the best choice;

[0147] S36. Perform weighted calculation on the predicted total cost under all energy storage unit quantities to find the predicted minimum comprehensive cost C under each first energy storage ratio condition. min (S ratio ) and its corresponding number of energy storage units N optimal , the current number of energy storage units is the optimal energy storage configuration;

[0148] This process can significantly reduce the overall operating cost of the system and improve economic benefits. For example, if it is calculated that in a certain situation, selecting 10 energy storage units can minimize the total cost, then this configuration can be given priority to optimize the economy and stability of the photovoltaic power generation system.

[0149] See also Figure 1 and Figure 5 In this embodiment, S4 further includes:

[0150] S41. Taking the number of energy storage units as the decision variable, set an objective function Z, which represents the minimization of the total cost, to ensure that under each energy storage ratio condition, the photovoltaic power generation system can reduce the operating cost while meeting the grid demand. Z is expressed as:

[0151]

[0152] Among them, C min (S ratio ) is the lowest comprehensive cost prediction, C grid (N storage ) is the grid operation cost, C PV (N storage ) is the comprehensive cost of the photovoltaic power generation system;

[0153] S42, setting constraints for the objective function Z, and determining the optimal number of energy storage units N that minimizes the objective function Z by adjusting the constraints and using an algorithm for solving the optimization problem, such as a simplex method or an interior point method. optimal And the corresponding dynamic scheduling strategy to form the optimal dynamic scheduling strategy.

[0154] To achieve the objective function Z, linear programming or nonlinear programming can be used. Setting constraints is the basis for ensuring the safe and reliable operation of the system. For example, the minimum and maximum number of energy storage units N can be specified. min ≤N storage ≤N max In addition, the load limit of the power grid must be considered to ensure that the energy storage system can meet the load requirements of the power grid in different time periods. This can be achieved by introducing the power grid load forecast value L pred (t) As a constraint:

[0155]

[0156] Among them, N storage is the number of energy storage units, G unit The power supply capacity of a single energy storage unit;

[0157] Through such constraints, it is ensured that the energy storage system can provide sufficient power support during peak load periods.

[0158] Taking linear programming as an example, the above objective function and constraints can be converted into a standard linear programming form for solution.

[0159] Assume that in a certain situation, after solving the model, it is known that the optimal number of energy storage units is 8. Combined with this result, a corresponding dynamic scheduling strategy is constructed. This strategy can flexibly adjust the charging and discharging plan of the energy storage system based on time period division and electricity price fluctuations to maximize economic benefits.

[0160] Finally, the optimal dynamic dispatch model constructed not only ensures the economic efficiency of the photovoltaic power generation system, but also enhances its adaptability to grid fluctuations. Through scientific decision-making and optimization strategies, the photovoltaic power generation system can effectively respond to changes in grid load and achieve efficient and stable operation. This process provides strong support for the sustainable development of the power market and promotes the rational use of new energy.

[0161] See also Figure 1 and Figure 6 In this embodiment, S5 further includes:

[0162] S51. Based on the optimal number of energy storage units and the optimal dynamic scheduling strategy, regression analysis methods are used in the historical data matching process to quantify the model prediction ability and actual operation effect;

[0163] In S51, a regression model is defined to represent the relationship between the actual operation data and the predicted data. Let Y represent the energy storage efficiency of the actual operation, and X represent the energy storage efficiency predicted by the model. The regression model is expressed as:

[0164]

[0165] Among them, β 0 is the intercept term, β 1 is the slope, ò is the error term;

[0166] By performing regression analysis on historical data, we can estimate β 0 and β 1 The significance of these parameters can be assessed by t-test to determine the explanatory power of the model;

[0167] In regression analysis, t refers to the t-test, which is a statistical test method used to evaluate whether a parameter in a model is significantly different from zero;

[0168] S52. Calculate the mean square error (MSE) to further evaluate the prediction effect of the model. The MSE calculation formula is:

[0169]

[0170] in, is the actual operating energy storage efficiency of the i-th group, The predicted energy storage efficiency of the model for group i, n is the number of data points;

[0171] By analyzing the MSE, we can intuitively judge the accuracy of the model prediction and make necessary adjustments to the model parameters;

[0172] S53, judging whether the model prediction is accurate according to the calculation results, if so, maintaining the current model, if not, adjusting the parameters of the current model until the model prediction capability matches the actual operation effect;

[0173] S54, using the cross-validation method, divide the historical data into multiple subsets, train and validate the model one by one, and more comprehensively evaluate the performance of the model under different conditions;

[0174] This method can effectively avoid overfitting, ensure the robustness of the model in practical applications, and improve the applicability of the model;

[0175] S55. Identify the key factors that affect energy storage efficiency, construct a multiple linear regression model, and introduce these key factors into the multiple linear regression model, which is expressed as:

[0176]

[0177] Among them, X 1 , X 2 ...X k Represent different key factors, β 0 is the intercept term, β 1 , β 2 ...β k Represent different slopes respectively, ò is the error term;

[0178] In the analysis process, the identification of key factors is also particularly important. For example, through in-depth mining of historical data, some key factors, such as weather changes, fluctuations in photovoltaic power generation and grid load fluctuations, will affect energy storage efficiency;

[0179] By using a multivariate linear regression model to quantify the impact of each key factor, we can have a more comprehensive understanding of the impact of each factor on the performance of the energy storage system and provide a reference for future system improvements.

[0180] Through the matching of the above historical data and strategy optimization, not only the applicability of the model is improved, but also the energy storage management strategy can be dynamically adjusted to ensure the stability and economy of the system in actual operation. This process provides a practical solution for subsequent energy storage management, which helps to promote the efficient operation and sustainable development of photovoltaic power generation systems.

[0181] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned photovoltaic power generation system energy storage management method based on dynamic regulation.

[0182] See also Figure 7An embodiment of the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned photovoltaic power generation system energy storage management method based on dynamic regulation.

[0183] The embodiment of the present invention further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the steps of the above-mentioned photovoltaic power generation system energy storage management method based on dynamic regulation.

[0184] It is understandable that the system, device and storage medium provided in the embodiments of the present invention correspond to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts in the above-mentioned photovoltaic power generation system energy storage management method based on dynamic control.

[0185] It should be noted that those skilled in the art can understand that all or part of the steps implemented in the embodiments of the present invention can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using hardware, it can be implemented in whole or in part in the form of purchased standard parts or modified parts. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0186] In summary, the present invention proposes a method for managing energy storage in a photovoltaic power generation system based on dynamic regulation to solve the problem of unstable power output caused by the influence of weather and day and night changes in the prior art. This solution aims to optimize energy storage management, improve the utilization rate and economic benefits of photovoltaic power generation, and significantly improve the economy and stability of photovoltaic power generation systems by acquiring information such as grid load data, photovoltaic power generation historical data, and the number of energy storage units.

[0187] It should be understood that the examples and implementation modes described herein are for illustrative purposes only and are not intended to limit the present invention. Those skilled in the art may make various modifications or changes based on the examples and implementation modes. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A photovoltaic power generation system energy storage management method based on dynamic regulation, characterized in that: The following steps are involved: S1. Acquisition and preliminary analysis of power grid load data: Preliminary setting of the first energy storage ratio to obtain the fluctuation characteristics of historical power grid load data; S2. Construction of a dynamic dispatch model for power grid cost: Based on the analysis of historical power grid load data and time characteristics obtained in S1, a power grid operation cost model under each first energy storage ratio is constructed; S3. Calculation and optimization of the number of energy storage units: Based on the grid operation cost model of S2, combined with the grid operation cost and the comprehensive cost of the photovoltaic power generation system, the total cost is weighted and calculated to determine the optimal energy storage configuration; S4, construction of an optimal dynamic scheduling model: setting an objective function representing the minimization of the total cost, and determining the optimal number of energy storage units and the corresponding dynamic scheduling strategy that minimize the objective function based on the optimal energy storage configuration of S3; S5. Matching of historical data and strategy optimization: Based on the optimal number of energy storage units and the optimal dynamic scheduling strategy of S4, quantify and compare the model prediction ability and the actual operation effect, judge the model prediction accuracy, optimize and improve the applicability of the model; The S3 further comprises: S31, obtaining the historical power generation and peak power limit of the photovoltaic power generation system, and combining them with the first energy storage ratio that is initially set, to determine the initial number of energy storage units; S32. Based on the historical operation data of the photovoltaic power generation system, a comprehensive cost prediction model is constructed to reflect the cost performance under different numbers of energy storage units; S33. According to the analysis results of historical operation data, determine whether the power generation of the photovoltaic power generation system meets the power demand of the power grid. If so, then calculate the power grid operation cost. If not, adjust the power generation capacity of the energy storage unit and then calculate the power grid operation cost. S34, when calculating the grid operation cost, based on the grid load prediction model and the grid operation cost model, the predicted grid load data is combined with the electricity prices in different time periods to calculate the grid operation cost under different numbers of energy storage units; S35. When calculating the comprehensive cost of the photovoltaic power generation system, by comparing the historical power generation of the photovoltaic power generation system, evaluating the power generation efficiency and cost-effectiveness under different numbers of energy storage units, the number of energy storage units corresponding to the lowest comprehensive cost of the photovoltaic power generation system is the best choice; S36. Perform weighted calculation on the predicted total cost under all energy storage unit quantities to find the predicted minimum comprehensive cost C under each first energy storage ratio condition. min (S ratio ) and its corresponding number of energy storage units N optimal , the current number of energy storage units is the optimal energy storage configuration.

2. The method for managing energy storage of a photovoltaic power generation system based on dynamic regulation according to claim 1, characterized in that: The S1 further comprises: S11, extract the peak and valley values ​​in the historical power grid load data, and set the peak value as P max , the valley value is P min , calculate the variation range R of the power grid load data, R=P max -P min ; S12, determining whether the variation range R is within the preset range, if so, preliminarily setting the first energy storage ratio, if not, first adjusting the energy storage ratio coefficient k, and then preliminarily setting the first energy storage ratio; S13. Use time series analysis methods to identify periodic and trend changes in electricity demand and analyze the fluctuation characteristics of historical power grid load data; S14. Predict grid load demand based on fluctuation characteristics.

3. The method for managing energy storage of a photovoltaic power generation system based on dynamic regulation according to claim 2, characterized in that: In S13, the autoregressive moving average model is used to model the historical power grid load data, and the modeling formula is: in: t is the time point, which represents the current moment in the grid load data sequence; L t is the historical grid load data; μ is the mean of the historical power grid load data series; Ф and θ are the autoregressive and moving average coefficients, respectively; ε is the white noise term; i is the lag order of the autoregressive term, ranging from 1 to p, indicating the degree of influence of historical grid load data on current load; j is the lag order of the moving average term, ranging from 1 to q, indicating the degree of influence of the historical white noise term on the current load; p is the order of the autoregressive model, indicating the maximum number of load lag terms considered; q is the order of the moving average model, which indicates the maximum number of white noise lags considered.

4. The method for managing energy storage of a photovoltaic power generation system based on dynamic regulation according to claim 2, characterized in that: In S14, a linear regression method is used to construct a prediction model based on historical power grid load data, and the power grid load prediction value is initially set to L pred , the prediction formula is: Among them, a is the rate at which the power grid load data changes over time, b is the initial load value, which can be obtained by fitting the historical power grid load data, and t is the time point, which represents the current moment in the power grid load data sequence.

5. The photovoltaic power generation system energy storage management method based on dynamic regulation according to claim 4, characterized in that: The S2 further comprises: S21, according to the peak value P in the historical power grid load data max and valley value P min , determine the time characteristics of the power grid load, the time characteristics can be expressed as load peak period T peak and load valley period T valley express; S22, combined with historical grid load data L t and the time characteristics of the power grid load, construct a power grid load forecasting model, and further set the power grid load forecasting value to be L pred (t), the prediction formula is: in, is the smoothing coefficient, ranging from 0 to 1; S23, obtaining the electricity price curve C at each first energy storage ratio price (t), different electricity price curves reflect the electricity price fluctuations in different time periods, and the electricity price fluctuation range C is calculated range , C range =C max -C min ; S24, combined with the grid load prediction value L pred (t) and electricity price curve C price (t), construct the grid operation cost model C under each first energy storage ratio cost , the modeling formula is: By adding up the grid operation costs in different time periods, we can get the total grid operation cost C total : Where n is the total number of time periods.

6. The photovoltaic power generation system energy storage management method based on dynamic regulation according to claim 1, characterized in that: The S4 further comprises: S41. Taking the number of energy storage units as the decision variable, set an objective function Z, which represents the minimization of the total cost, to ensure that under each energy storage ratio condition, the photovoltaic power generation system can reduce the operating cost while meeting the grid demand. Z is expressed as: Among them, C min (S ratio ) is the lowest comprehensive cost prediction, C grid (N storage ) is the grid operation cost, C PV (N storage ) is the comprehensive cost of the photovoltaic power generation system; S42, setting constraints for the objective function Z, and determining the optimal number of energy storage units N that minimizes the objective function Z by adjusting the constraints and using an algorithm for solving the optimization problem. optimal And the corresponding dynamic scheduling strategy to form the optimal dynamic scheduling strategy.

7. The method for managing energy storage of a photovoltaic power generation system based on dynamic regulation according to claim 6, characterized in that: The S5 further comprises: S51. Based on the optimal number of energy storage units and the optimal dynamic scheduling strategy, regression analysis methods are used in the historical data matching process to quantify the model prediction ability and actual operation effect; S52. Calculate the mean square error (MSE) to further evaluate the prediction effect of the model. The MSE calculation formula is: in, is the actual operating energy storage efficiency of the i-th group, The predicted energy storage efficiency of the model for group i, n is the number of data points; S53, judging whether the model prediction is accurate according to the calculation results, if so, maintaining the current model, if not, adjusting the parameters of the current model until the model prediction capability matches the actual operation effect; S54, using the cross-validation method, divide the historical data into multiple subsets, train and validate the model one by one, and more comprehensively evaluate the performance of the model under different conditions; S55. Identify the key factors that affect energy storage efficiency, construct a multiple linear regression model, and introduce these key factors into the multiple linear regression model, which is expressed as: Among them, X1, X2...X k They represent different key factors, β0 is the intercept term, β1, β2...β k They represent different slopes respectively, and ò is the error term.

8. The photovoltaic power generation system energy storage management method based on dynamic regulation according to claim 7, characterized in that: In S51, a regression model is defined to represent the relationship between the actual operation data and the predicted data. Let Y represent the energy storage efficiency of the actual operation, and X represent the energy storage efficiency predicted by the model. The regression model is expressed as: Among them, β0 is the intercept term, β1 is the slope, and ò is the error term.

9. A computer device, characterized in that: It includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the photovoltaic power generation system energy storage management method based on dynamic regulation as described in any one of claims 1 to 8.

Images