A power conversion control method and system for a bidirectional charging pile

Abstract

The present invention relates to the field of power control technology, specifically to a power conversion control method and system for a bidirectional charging pile, comprising the following steps: collecting grid demand, electricity price changes, vehicle arrival time, expected stay time, and battery status data, integrating the data into an initial input set, calculating the information entropy value of the data in the set, analyzing the degree of dispersion of the data according to the information entropy value, and obtaining multiple types of entropy weighted data of the bidirectional charging pile. In the present invention, by analyzing the information entropy values ​​of operating parameters such as grid demand, electricity price changes, and vehicle arrival time, the system is allowed to identify the parameters with the largest changes at different time points, thereby effectively adjusting their weights and achieving more flexible and economical energy allocation. The control of power flow is enhanced, so that the charging station can effectively allocate energy during high-demand periods. At the same time, combined with the Gaussian process regression model, the accuracy of energy demand prediction is further improved.

Description

Technical Field

[0001] The present invention relates to the field of power control technology, and in particular to a power conversion control method and system for a bidirectional charging pile. Background Art

[0002] The field of power control technology involves the management and optimization of power systems, including the generation, transmission, distribution and use of electrical energy. The field focuses on how to effectively control the flow of electricity to improve energy efficiency, reduce energy losses, and ensure the stability and reliability of the power grid. This includes regulating voltage and current, as well as automatic management of power equipment through intelligent control. It also involves power conversion, such as frequency converters and inverters, which can convert the form and characteristics of electricity to adapt to different usage requirements and improve the efficiency of power use.

[0003] Among them, the power conversion control method of the bidirectional charging pile is used to manage the power flow in the electric vehicle charging station. In the bidirectional charging pile, power can not only be transmitted from the power grid to the electric vehicle for charging, but also fed back from the electric vehicle to the power grid, which can improve energy utilization and achieve more flexible energy management. For example, when the power grid load is low, the vehicle can send the stored power back to the power grid; and during peak hours, it can obtain energy from the power grid.

[0004] Although existing bidirectional charging pile technologies support the bidirectional flow of electric energy, they are often limited in terms of the flexibility and prediction accuracy of power dispatch. Existing methods mainly rely on static preset conditions, which makes it difficult to cope with rapidly changing market conditions and grid conditions in practical applications. For example, in the event of sudden peak demand or unpredictable grid failures, the system may not be able to adapt quickly, resulting in insufficient or excessive energy supply, which affects the economy and stability of the system. In addition, existing technologies have difficulty analyzing the importance of uncertainty and parameter changes when processing data, which limits the optimization potential of decision-making and the flexibility of response strategies, making power distribution less than ideal during high or low demand periods, which can easily lead to resource waste or energy shortages. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a power conversion control method and system for a bidirectional charging pile.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a power conversion control method for a bidirectional charging pile, comprising the following steps:

[0007] S1: Collect grid demand, electricity price changes, vehicle arrival time, expected stay time, and battery status data, integrate the data into the initial input set, calculate the information entropy value of the data in the set, analyze the dispersion of the data based on the information entropy value, and obtain multi-category entropy weight data of the bidirectional charging pile;

[0008] S2: extracting the parameters that have the greatest impact on power control from the multi-category entropy weight data of the bidirectional charging pile, adjusting the influence weights of the parameters according to the dispersion degree of the parameters, re-proportioning the parameters according to the adjusted weights, and performing linear weight superposition to construct a weight adjustment model;

[0009] S3: Collect weather conditions, vehicle flow and historical charging and discharging data of bidirectional charging piles, analyze the nonlinear relationship between the data and energy demand through Gaussian process regression method, integrate the data into the weight adjustment model, and form an energy flow prediction model;

[0010] S4: Apply the energy flow prediction model to perform simulation tests, adjust the charging and discharging process according to the simulation results, collect the adjusted operation data, optimize the power conversion parameters according to the operation data, and obtain the bidirectional charging pile power control optimization results.

[0011] As a further solution of the present invention, the step of obtaining the information entropy value of each type of data in the calculation set is specifically:

[0012] S111: Collect data, including grid demand, electricity price changes, vehicle arrival time, estimated stay time, and battery status, and integrate them into an initial input set. The set of data types is defined as: , according to each type of data, using the formula: , calculate the frequency of occurrence of each data type ;

[0013] in, Represents the i-th type of data, i is the data index, represent The number of occurrences of Represents the total number of data. represents the smoothing constant added to the frequency, Represents the total number of data types;

[0014] S112: According to the occurrence frequency of each data type , using the formula: , analyze the amount of data information and calculate the information entropy value of each type of data ;

[0015] in, represent The frequency of the jth value of represent The number of values ​​of Used to avoid zero parameters in the logarithm, represent The index of the value in , () is a logarithmic function.

[0016] As a further solution of the present invention, the steps for obtaining multiple types of entropy weighted data of the bidirectional charging pile are specifically as follows:

[0017] S121: Based on the information entropy value of each type of data , using the formula: , calculate the relative entropy weight of each type of data, and get the entropy weight of the i-th type of data ;

[0018] S122: According to the entropy weight of the i-th type of data , using the formula: , analyze the dispersion of the data and obtain the average entropy weight of all data types ;

[0019] S123: Calculate the average entropy weight of all data types , using the formula: , calculate the entropy weight of the adjusted i-th category data , integrating multiple types of entropy weight data of bidirectional charging piles;

[0020] in, is the influence coefficient of dispersion, is a constant.

[0021] As a further solution of the present invention, the step of obtaining the influence weight of the adjustment parameter is specifically as follows:

[0022] S211: according to the adjusted entropy weight of the i-th category data , using the formula: , extract the most influential parameters ;

[0023] in, is the adjustment parameter that affects the sensitivity of the i-th type of data, is the base of natural logarithms, is the data index, The index i used to capture the maximum data;

[0024] S212: Extracting the most influential parameter according to the method , using the formula: , analyze the parameters with the greatest impact The data characteristics of the data are used to obtain the coefficient of variation of the data type with the greatest impact on power control. ;

[0025] in, for The standard deviation indicates the fluctuation of the data. for The average value represents the center point of the data;

[0026] S213: Coefficient of variation according to the largest data type , using the formula: , calculate the adjusted weight ;

[0027] in, is the adjustment coefficient, which is used to nonlinearly adjust the influence of dispersion. is another adjustment coefficient used to increase the nonlinear adjustment characteristics.

[0028] As a further solution of the present invention, the step of obtaining the weight adjustment model is specifically as follows:

[0029] S221: Based on the adjusted weight , using the formula: , re-proportion the parameters, and obtain Rebalanced weights ;

[0030] S222: Using the basis Rebalanced weights , using the formula; , accumulate weighted parameters and build a weight adjustment model .

[0031] As a further solution of the present invention, the steps of acquiring the energy flow prediction model are specifically as follows:

[0032] S311: Integrated weather conditions , traffic volume , and historical charge and discharge data For a set of predictor variables, use the formula: , and obtain the set of predictor variables ;

[0033] in, To adjust the weight coefficient of weather data contribution, To adjust the weight coefficient of the contribution of traffic flow data, To adjust the weight coefficient of the contribution of historical charge and discharge data;

[0034] S312: Based on the prediction variable set and historical charging energy requirements Data, using the formula: , calculate the predicted energy demand value and build a regression model ;

[0035] in, is an exponential function used for kernel function calculation, for The mean of for The standard deviation of the data;

[0036] S313: According to the regression model and weight adjustment model , using the formula: , perform data integration and obtain the energy flow prediction model .

[0037] As a further solution of the present invention, the steps for obtaining the bidirectional charging pile power control optimization result are specifically as follows:

[0038] S411: According to the energy flow prediction model , using the formula: , calculate the simulated energy demand value ;

[0039] in, represents the adjustment factor for current environmental conditions, Represents the environmental stability index, Represents a small amount;

[0040] S412: Based on the simulated energy demand value , using the formula: , calculate the adjusted charge and discharge operation data ;

[0041] in, represents the target efficiency parameter, represents the adjustment intensity coefficient;

[0042] S413: Based on the adjusted charge and discharge operation data , using the formula: , and obtain the optimization result of bidirectional charging pile power control ;

[0043] in, represents the new conversion efficiency coefficient, Represents the performance adjustment factor.

[0044] A power conversion control system for a bidirectional charging pile, the power conversion control system for the bidirectional charging pile is used to execute the power conversion control method for the bidirectional charging pile, the system comprising:

[0045] The data collection and analysis module collects grid demand, electricity price change information, vehicle arrival time, expected stay time and battery status, integrates them into the initial data set, calculates the information entropy value of the data, evaluates the degree of data dispersion, and generates multi-category entropy weighted data of bidirectional charging piles;

[0046] The parameter weight adjustment module analyzes multiple types of entropy weight data of the bidirectional charging pile, identifies key parameter weights that affect power control, reconfigures parameter weights through a linear superposition method, and generates a weight adjustment model;

[0047] The energy demand prediction module integrates weather conditions, vehicle flow and historical charging and discharging data of charging piles, performs nonlinear relationship analysis on these data, optimizes energy demand prediction in combination with the weight adjustment model, and generates an energy flow prediction model;

[0048] The charge and discharge simulation test module uses the energy flow prediction model to perform simulation tests, adjusts the charge and discharge process according to the test results, collects adjusted operation data, and generates optimized operation data;

[0049] The power control optimization module uses the optimized operation data to iteratively adjust the power conversion parameters to obtain the power control optimization results of the bidirectional charging pile.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are:

[0051] In the present invention, by analyzing the information entropy values ​​of operating parameters such as grid demand, electricity price changes, and vehicle arrival time, the system is allowed to identify the parameters with the largest changes at different time points, thereby effectively adjusting their weights and achieving more flexible and economical energy allocation. The control of power flow is enhanced, so that charging stations can effectively allocate energy during high-demand periods. At the same time, combined with the Gaussian process regression model, the accuracy of energy demand prediction is further improved. The energy management of charging piles is not only more responsive, but also can predict future trends, optimize resource allocation, significantly reduce energy waste, and enhance the economic benefits and energy efficiency of the entire charging system through precise regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0053] Figure 2 A flowchart of calculating the information entropy value of data in a set according to the present invention;

[0054] Figure 3 It is a flow chart of multiple types of entropy weight data of the bidirectional charging pile of the present invention;

[0055] Figure 4 A flow chart of adjusting the parameter influence weights of the present invention;

[0056] Figure 5 A flow chart of the weight adjustment model of the present invention;

[0057] Figure 6 It is a flow chart of the energy flow prediction model of the present invention;

[0058] Figure 7 This is a flow chart of the power control optimization results of the bidirectional charging pile of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0061] See also Figure 1 The present invention provides a technical solution: a power conversion control method for a bidirectional charging pile, comprising the following steps:

[0062] S1: Collect grid demand, electricity price changes, vehicle arrival time, expected stay time, and battery status data, integrate the data into the initial input set, calculate the information entropy value of the data in the set, analyze the dispersion of the data based on the information entropy value, and obtain multi-category entropy weight data of the bidirectional charging pile;

[0063] S2: Extract the parameters that have the greatest impact on power control from the multi-category entropy weight data of the bidirectional charging pile, adjust their influence weights according to the dispersion of the parameters, re-proportion the parameters according to the adjusted weights, perform linear weight superposition, and construct a weight adjustment model;

[0064] S3: Collect weather conditions, vehicle flow, and historical charging and discharging data of bidirectional charging piles, analyze the nonlinear relationship between data and energy demand through Gaussian process regression method, integrate the data into the weight adjustment model, and form an energy flow prediction model;

[0065] S4: Apply the energy flow prediction model to perform simulation tests, adjust the charging and discharging process according to the simulation results, collect the adjusted operation data, optimize the power conversion parameters according to the operation data, and obtain the power control optimization results of the bidirectional charging pile;

[0066] The multiple types of entropy weight data of bidirectional charging piles include the entropy weight of grid load changes in multiple time periods, the entropy weight of peak-valley electricity price differences, the entropy weight of the vehicle's expected charging start time, the entropy weight of the estimated parking time, and the entropy weight of the battery's remaining capacity. The weight adjustment model includes adjustments to the grid demand weights during key operating periods, dynamic electricity price weights, vehicle arrival weights during peak charging periods, charging station parking lot occupancy weights, and battery health status weights. The energy flow prediction model includes the impact of seasonal weather conditions, differences in traffic flow between daily and holidays, and historical operating efficiency data of charging piles. The power control optimization results of the bidirectional charging piles include adjusted charging and discharging schedules, power conversion efficiency parameters, and charging power settings.

[0067] See also Figure 2 , the specific steps for obtaining the information entropy value of each type of data in the calculation set are:

[0068] S111: Collect data, including grid demand, electricity price changes, vehicle arrival time, estimated stay time, and battery status, and integrate them into an initial input set. The set of data types is defined as: , according to each type of data, using the formula: , calculate the frequency of occurrence of each data type ;

[0069] in, Represents the i-th type of data, i is the data index, represent The number of occurrences of Represents the total number of data. represents a smoothing constant added to the frequency to avoid inaccurate calculations due to sparse data. Represents the total number of data types;

[0070] S112: Based on the frequency of occurrence of each data type , using the formula: , analyze the amount of data information and calculate the information entropy value of each type of data ;

[0071] in, represent The frequency of the jth value of represent The number of values ​​of Used to prevent the parameters in the logarithm from being zero, thus avoiding calculation errors. represent The index of a value in , () is a logarithmic function. The smaller the probability, the larger the logarithmic value.

[0072] The calculation process is as follows:

[0073] For the formula: ;

[0074] Assume there are 5 types of data ( ), the total number of data , smoothing parameter , the number of occurrences of the first type of data ;

[0075] Calculate the denominator:

[0076] ;

[0077] Calculate the numerator:

[0078] ;

[0079] Calculate frequency: ;

[0080] result It means that the first data type accounts for 20% of the total data, showing its proportion in the data set.

[0081] For the formula: , assuming that the first type of data has only two values ​​( ), the first frequency is known to be , the second frequency is , ;

[0082] Calculate the information content of the first value:

[0083] ;

[0084] Calculate the information content of the second value:

[0085] ;

[0086] Calculate the total information entropy:

[0087] , assuming The two values ​​of have equal probability, that is, .

[0088] Assuming maximum information entropy: ;

[0089] in, Approx. ;

[0090] therefore: ;

[0091] Actual information entropy and compared to, Lower, indicating data The distribution of is not completely uniform, and there is a certain degree of skewness. A value lower than the maximum entropy indicates that the uncertainty of the data is smaller than the ideal uniform distribution.

[0092] The information entropy value shows that although the data distribution has a certain randomness, it has more structure and predictability than the uniform distribution. It is possible to optimize the charging time or adjust the charging strategy to adapt to the predicted grid demand and vehicle usage pattern.

[0093] See also Figure 3 ,The specific steps for obtaining multi-category entropy weight data of bidirectional charging piles are:

[0094] S121: According to the information entropy value of each type of data , using the formula: , calculate the relative entropy weight of each type of data, and get the entropy weight of the i-th type of data ;

[0095] in, is the total number of data types, Index for data;

[0096] S122: According to the entropy weight of the i-th type of data , using the formula: , analyze the dispersion of the data and obtain the average entropy weight of all data types ;

[0097] S123: Based on the average entropy weight of all data types , using the formula: , calculate the entropy weight of the adjusted i-th category data , integrating multiple types of entropy weight data of bidirectional charging piles;

[0098] in, is the influence coefficient of the degree of dispersion, which is used to adjust the influence of the degree of dispersion on the weight. is a constant used to ensure that the original entropy weight is maintained when there is no dispersion effect;

[0099] The calculation process is as follows:

[0100] For the formula: , assuming there are three data types, so ;

[0101] The information entropy values ​​of the data types are ;

[0102] Calculate the sum of information entropy values ​​for all data types: ;

[0103] Calculate the entropy weight for each data type:

[0104] ;

[0105] ;

[0106] ;

[0107] The results show that the data type It has the greatest importance in the entire system because it occupies the largest entropy weight ratio.

[0108] For the formula: ;

[0109] Compute the average of the entropy weights: , the average entropy weight Used to compare the deviation of each data type's entropy weight from the average level.

[0110] For the formula: , assuming As the influence coefficient of dispersion degree.

[0111] Modified entropy weight calculation:

[0112] ;

[0113] ;

[0114] ;

[0115] The corrected entropy weight is:

[0116] and After comparison, The value from Down to , although slightly decreased, is still much higher than the average entropy weight, indicating Class data is still the most important. Its higher entropy weight reflects its importance in the data set and its significant impact on decision making. When formulating power conversion strategies, it should be given priority consideration. Class data. The value from Slightly increased to , close to the average entropy weight, indicating The importance of this type of data is on par with the average level, indicating that its impact on the power conversion control strategy is relatively balanced and does not need to be particularly emphasized, nor should it be ignored. The value from Increase to , still lower than the average entropy weight, but slightly increased, indicating that However, the impact is still small, indicating that in the overall power conversion control, Class data is not the main decision factor.

[0117] See also Figure 4 , the specific steps for obtaining the weights affected by adjusting the parameters are:

[0118] S211: According to the adjusted entropy weight of the i-th category data , using the formula: , extract the most influential parameters ;

[0119] in, To adjust the parameters that affect the sensitivity of the i-th type of data, the sigmoid function is used to increase the nonlinear characteristics of the selection. is the base of the natural logarithm, used for the sigmoid function, is the data index, Used to find the maximized data index i;

[0120] S212: Extract the most influential parameters , using the formula: , analyze the parameters with the greatest impact The data characteristics of the data are used to obtain the coefficient of variation of the data type with the greatest impact on power control. ;

[0121] in, for The standard deviation indicates the fluctuation of the data. for The average value represents the center point of the data;

[0122] S213: Coefficient of variation according to the largest data type , using the formula: , calculate the adjusted weight ;

[0123] in, is the adjustment coefficient, which is used to nonlinearly adjust the influence of dispersion. is another adjustment coefficient, adding nonlinear adjustment characteristics;

[0124] The calculation process is as follows:

[0125] Assume a modified entropy weight with three parameters:

[0126] ;

[0127] ;

[0128] ;

[0129] For the formula: ;

[0130] Assumptions , which makes , so the formula is: ;

[0131] Here, because is a constant and does not affect the determination of the maximum value, so: , the parameter with the largest modified entropy weight is selected, indicating that it has the greatest impact on power control. refers to The parameter corresponding to the maximum value.

[0132] In this example, for ,because maximum.

[0133] Assumptions: ;

[0134] For the formula: , calculate the coefficient of variation , which is used to measure the dispersion of the data relative to its mean.

[0135] for : ,express The data is relatively concentrated and does not fluctuate much.

[0136] Assuming the weight adjustment factor used and They are: ;

[0137] For the formula: ;

[0138] for Adjustment weights: ;

[0139] Assume the base value is , the final adjusted weight show is more important than the baseline value. It is critical in power control and its higher adjusted weight reflects its strong influence.

[0140] See also Figure 5 , the specific steps for obtaining the weight adjustment model are:

[0141] S221: Based on the adjusted weights , using the formula: , re-proportion the parameters, and obtain Rebalanced weights ;

[0142] S222: Use according to Rebalanced weights , using the formula; , accumulate weighted parameters and build a weight adjustment model ;

[0143] The calculation process is as follows:

[0144] For the formula: ;

[0145] Assumptions: , , total number of parameters

[0146] Compute the sum of all adjusted weights:

[0147] ;

[0148] Normalize each weight using this sum:

[0149] ;

[0150] ;

[0151] ;

[0152] Results directly reflects the relative importance of each parameter after adjustment, where Keep the highest influence weight.

[0153] For the formula: ;

[0154] Assuming the actual control parameters They represent different power control variables, such as current regulation, voltage regulation, and frequency regulation;

[0155] Assumptions The actual values ​​are 100, 200, and 150 respectively;

[0156] ;

[0157] ;

[0158] The baseline value is assumed to be , which is used to evaluate the average effectiveness of the control strategy. It is slightly lower than the baseline value, indicating that the current control weight configuration has a slightly lower than expected effect on power adjustment.

[0159] See also Figure 6 , the specific steps of obtaining the energy flow prediction model are:

[0160] S311: Integrated weather conditions , traffic volume , and historical charge and discharge data For a set of predictor variables, use the formula: , and obtain the set of predictor variables ;

[0161] Among them, a is the weight coefficient for adjusting the contribution of weather data. To adjust the weight coefficient of the contribution of traffic flow data, To adjust the weight coefficient of the contribution of historical charge and discharge data;

[0162] S312: Based on the set of predictor variables and historical charging energy requirements Data, using the formula: , calculate the predicted energy demand value and build a regression model ;

[0163] in, is an exponential function, used for kernel function calculation in Gaussian process. for The mean of for The standard deviation of the data;

[0164] S313: Based on the regression model and weight adjustment model , using the formula: , perform data integration and obtain the energy flow prediction model ;

[0165] The calculation process is as follows:

[0166] For the formula: ;

[0167] Assumptions: Weather conditions (e.g. temperature is 20°C), traffic volume (e.g. daily traffic flow), historical charging and discharging data (e.g. total number of charges and discharges in a day), the weight coefficient assumes , , ;

[0168] ,

[0169] ;

[0170] The value combines the weighted effects of weather, traffic flow, and charging and discharging data, resulting in 620.

[0171] For the formula: ;

[0172] Assumption: Historical charging energy demand (e.g. average electricity demand per hour), mean Standard Deviation ;

[0173] Compute the Gaussian kernel:

[0174] ,

[0175] ,

[0176] ;

[0177] The result is 48.6, which indicates the predicted electricity demand after taking into account the current input variables.

[0178] For the formula:

[0179] ,

[0180] ;

[0181] The result is about 6818.58, which is obtained according to the prediction model and reflects the predicted total charging energy demand of the bidirectional charging pile under given input variables (such as weather, traffic flow and historical charging and discharging data). Compared with the assumed baseline value of 5000, the predicted value of 6818.58 shows a higher charging energy demand under the considered conditions, which may be due to external factors, such as hotter weather increases the use of air conditioning, or increased traffic volume leads to more vehicles needing to charge.

[0182] See also Figure 7 , the specific steps for obtaining the optimization results of the power control of the bidirectional charging pile are:

[0183] S411: Based on the energy flow prediction model , using the formula: , calculate the simulated energy demand value ;

[0184] in, represents the adjustment factor for current environmental conditions, Represents the environmental stability index, represents a small amount, used to ensure stability;

[0185] S412: Based on the simulated energy demand value , using the formula: , calculate the adjusted charge and discharge operation data ;

[0186] in, represents the target efficiency parameter, represents the adjustment intensity coefficient;

[0187] S413: Based on the adjusted charge and discharge operation data , using the formula: , and obtain the optimization result of bidirectional charging pile power control ;

[0188] in, represents the new conversion efficiency coefficient, represents the performance adjustment factor, represents a small amount, used to ensure stability;

[0189] The calculation process is as follows:

[0190] For the formula: ;

[0191] Assumptions: (Adjustment factor for current environmental conditions, assuming slightly adverse environmental effects), (an indicator of environmental stability, indicating that the impact of environmental changes is small), (a small amount, used to prevent the denominator from being zero);

[0192] Calculate the denominator:

[0193] ;

[0194] Evaluate the entire fraction:

[0195] ;

[0196] calculate :

[0197] ;

[0198] The result is about 119325.15, which is the predicted simulation test result under the given environmental conditions, indicating that the energy demand may increase significantly under the current settings.

[0199] For the formula: ;

[0200] Assumptions: , (target efficiency parameter, set to a higher target to promote efficiency), (Adjust the intensity coefficient, less than 1 to reduce the impact of the adjustment);

[0201] Calculate the fraction:

[0202] ;

[0203] Apply an adjustment strength factor:

[0204] ;

[0205] The result is about 104.258, which indicates the new parameters after adjusting the charge and discharge operations to optimize the efficiency of power conversion.

[0206] For the formula: ;

[0207] Assumptions: , (New conversion efficiency coefficient, assuming a slight improvement), (Performance adjustment factor, indicating the magnitude of performance improvement), (small amount, remain unchanged);

[0208] Calculate the fraction:

[0209] ;

[0210] calculate :

[0211] ;

[0212] The result is approximately 1042.58, which is the power control optimization result obtained after adjusting the operating data, indicating that after optimization, the performance and efficiency of the charging pile have been improved.

[0213] Assume that before optimization, the power conversion efficiency of the bidirectional charging pile is low, and the benchmark efficiency value is set to Reflects the average performance of the system under standard operating conditions, taking into account factors such as equipment aging, inadequate maintenance, or less than ideal environmental conditions. Significantly higher than the baseline . It shows that after adjusting the charging and discharging process and optimizing the power conversion parameters, the performance of the bidirectional charging pile has been significantly improved. This may come from more accurate energy demand prediction, more effective energy management strategy and optimization of system response. It not only means that the energy utilization efficiency of the charging pile is higher, but also may bring better user experience (such as faster charging time and lower waiting time) and reduce energy waste.

[0214] A power conversion control system for a bidirectional charging pile, the power conversion control system for the bidirectional charging pile is used to execute the power conversion control method for the bidirectional charging pile, the system comprises:

[0215] The data collection and analysis module collects grid demand, electricity price change information, vehicle arrival time, expected stay time and battery status, integrates them into the initial data set, calculates the information entropy value of the data, evaluates the degree of data dispersion, and generates multi-category entropy weighted data of bidirectional charging piles;

[0216] The parameter weight adjustment module analyzes multiple types of entropy weight data of bidirectional charging piles, identifies the key parameter weights that affect power control, reconfigures the parameter weights through the linear superposition method, and generates a weight adjustment model;

[0217] The energy demand forecasting module integrates weather conditions, vehicle flow, and historical charging and discharging data of charging piles, performs nonlinear relationship analysis on these data, and optimizes energy demand forecasting in combination with a weight adjustment model to generate an energy flow forecasting model;

[0218] The charge and discharge simulation test module uses the energy flow prediction model to perform simulation tests, adjusts the charge and discharge process according to the test results, collects the adjusted operation data, and generates optimized operation data;

[0219] The power control optimization module uses the optimized operation data to iteratively adjust the power conversion parameters to obtain the power control optimization results of the bidirectional charging pile.

[0220] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A power conversion control method for a bidirectional charging pile, characterized in that: The following steps are involved: Collect grid demand, electricity price changes, vehicle arrival time, expected stay time, and battery status data, integrate the data into the initial input set, calculate the information entropy value of the data in the set, analyze the degree of data dispersion based on the information entropy value, and obtain multi-category entropy weight data of bidirectional charging piles; The steps for obtaining the multi-category entropy weight data of the bidirectional charging pile are specifically as follows: According to the information entropy value of each type of data , using the formula: , calculate the relative entropy weight of each type of data, and get the entropy weight of the i-th type of data , according to the entropy weight of the i-th type of data , using the formula: , analyze the dispersion of the data and obtain the average entropy weight of all data types ; The average value of entropy weights of all data types , using the formula: , calculate the entropy weight of the adjusted i-th category data , integrating multiple types of entropy weight data of bidirectional charging piles; in, is the influence coefficient of dispersion degree, 1 is a constant; Extracting the parameters that have the greatest impact on power control from the multi-category entropy weight data of the bidirectional charging pile, adjusting the impact weights of the parameters according to the dispersion degree of the parameters, re-proportioning the parameters according to the adjusted weights, and performing linear weight superposition to construct a weight adjustment model; The step of obtaining the weight adjustment model is specifically as follows: according to the adjusted weight , using the formula: , re-proportion the parameters, and obtain Rebalanced weights ; Use the basis Rebalanced weights , using the formula: , accumulate weighted parameters and build a weight adjustment model ; Collect weather conditions, vehicle flow, and historical charging and discharging data of bidirectional charging piles, analyze the nonlinear relationship between the data and energy demand through Gaussian process regression method, integrate the data into the weight adjustment model, and form an energy flow prediction model; The steps for obtaining the energy flow prediction model are specifically as follows: Integrate weather conditions T, traffic flow V, and historical charging and discharging data H into a set of prediction variables, using the formula: , get the prediction variable set X; Among them, a is the weight coefficient for adjusting the contribution of weather data, b is the weight coefficient for adjusting the contribution of traffic flow data, and c is the weight coefficient for adjusting the contribution of historical charging and discharging data; According to the prediction variable set X and the historical charging energy demand E data, the formula is used: , calculate the predicted energy demand value and construct the regression model R; in, is an exponential function used for kernel function calculation, is the mean of X, is the standard deviation of the X data; Adjust the model according to the regression model R and weights , using the formula: , perform data integration and obtain the energy flow prediction model P; Apply the energy flow prediction model to perform simulation tests, adjust the charging and discharging process according to the simulation results, collect the adjusted operation data, optimize the power conversion parameters according to the operation data, and obtain the bidirectional charging pile power control optimization results; The steps for obtaining the bidirectional charging pile power control optimization result are specifically as follows: According to the energy flow prediction model P, the formula is adopted: , calculate the simulated energy demand value S; in, represents the adjustment factor for current environmental conditions, Represents the environmental stability index, Represents a small amount; According to the simulated energy demand value S, the formula is adopted: , calculate the adjusted charge and discharge operation data Z; in, represents the target efficiency parameter, represents the adjustment intensity coefficient; According to the adjusted charge and discharge operation data Z, the formula is adopted: , and obtain the optimization result of bidirectional charging pile power control ; in, represents the new conversion efficiency coefficient, represents the performance adjustment factor; The multiple types of entropy weight data of the bidirectional charging pile include the entropy weight of grid load changes in multiple time periods, the entropy weight of peak-valley electricity price differences, the entropy weight of the vehicle's expected charging start time, the entropy weight of the estimated parking time, and the entropy weight of the battery's remaining capacity. The weight adjustment model includes adjustments to the grid demand weights during key operating periods, dynamic electricity price weights, vehicle arrival weights during peak charging periods, charging station parking lot occupancy weights, and battery health status weights. The energy flow prediction model includes the impact of seasonal weather conditions, differences in traffic flow between daily and holidays, and historical operating efficiency data of charging piles. The power control optimization results of the bidirectional charging pile include adjusted charging and discharging schedules, power conversion efficiency parameters, and charging power settings.

2. The power conversion control method of the bidirectional charging pile according to claim 1, characterized in that: The specific steps for obtaining the information entropy value of each type of data in the calculation set are: Collect data, including grid demand, electricity price changes, vehicle arrival time, estimated stay time, and battery status, and integrate them into the initial input set. Define the set of data types as follows: , according to each type of data, using the formula: , calculate the frequency of occurrence of each data type ; in, Represents the i-th type of data, i is the data index, represent The number of occurrences of Represents the total number of data. represents the smoothing constant added to the frequency, Represents the total number of data types; According to the frequency of occurrence of each data type , using the formula: , analyze the amount of data information and calculate the information entropy value of each type of data ; in, represent The frequency of the jth value of represent The number of values ​​​​of Used to avoid zero parameters in the logarithm, represent The index of the value in , is a logarithmic function.

3. The power conversion control method of the bidirectional charging pile according to claim 1, characterized in that: The specific steps to obtain the weight of the adjustment parameters are as follows: According to the adjusted entropy weight of the i-th category data , using the formula: , extract the most influential parameters ; in, is the adjustment parameter that affects the sensitivity of the i-th type of data, is the base of natural logarithms, is the data index, The index i used to capture the maximum data; Extract the most influential parameters according to , using the formula: , analyze the parameters with the greatest impact The data characteristics of the data are used to obtain the coefficient of variation of the data type with the greatest impact on power control. ; in, for The standard deviation indicates the fluctuation of the data. for The average value represents the center point of the data; Coefficient of variation according to the largest data type , using the formula: , calculate the adjusted weight ; in, The adjustment factor is used to nonlinearly adjust the effect of dispersion. is another adjustment coefficient used to increase the nonlinear adjustment characteristics.

4. A power conversion control system for a bidirectional charging pile, characterized in that: According to the power conversion control method of the bidirectional charging pile according to any one of claims 1 to 3, the system comprises: The data collection and analysis module collects grid demand, electricity price change information, vehicle arrival time, expected stay time and battery status, integrates them into the initial data set, calculates the information entropy value of the data, evaluates the degree of data dispersion, and generates multi-category entropy weighted data of bidirectional charging piles; The parameter weight adjustment module analyzes multiple types of entropy weight data of the bidirectional charging pile, identifies key parameter weights that affect power control, reconfigures parameter weights through a linear superposition method, and generates a weight adjustment model; The energy demand prediction module integrates weather conditions, vehicle flow and historical charging and discharging data of charging piles, performs nonlinear relationship analysis on these data, optimizes energy demand prediction in combination with the weight adjustment model, and generates an energy flow prediction model; The charge and discharge simulation test module uses the energy flow prediction model to perform simulation tests, adjusts the charge and discharge process according to the test results, collects adjusted operation data, and generates optimized operation data; The power control optimization module uses the optimized operation data to iteratively adjust the power conversion parameters to obtain the power control optimization results of the bidirectional charging pile.

Images