Intelligent control method and system of photovoltaic inverter

Abstract

The invention relates to the technical field of photovoltaic equipment control, in particular to an intelligent control method and system for a photovoltaic inverter, and the method comprises the following steps: in a calibration period, obtaining real-time load data recorded by each group of inverter circuits at each time point, the time intervals between any adjacent time points are the same, and the time intervals between any two adjacent time points are the same; the shorter the time interval is, the more accurate the data is, calculating the average calibration load and the fault rate of each inverter circuit, calculating a Pearson's correlation coefficient according to the average calibration load and the fault rate to determine a linear relation, establishing a linear regression model through two groups of data of the average calibration load and the fault rate, and calculating an expected fault rate; and predicting the residual life according to the expected failure rate.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic equipment control, and in particular to an intelligent control method and system for a photovoltaic inverter. Background Art

[0002] A photovoltaic inverter is a device that converts the direct current (DC) electricity generated by solar panels into alternating current (AC). AC is the standard type of electricity used in homes and on the grid, so the inverter allows the solar system to connect to the grid and power a home or business.

[0003] Photovoltaic inverters are key components in photovoltaic power generation systems. Their performance and lifespan directly affect the stability and efficiency of the entire power generation system. At present, the rated lifespan of photovoltaic inverters is mostly determined by testing and designing the lifespan of photovoltaic inverter electronic components. Manufacturers determine the rated lifespan of photovoltaic inverters by measuring the operating parameters of various aspects of photovoltaic inverter electronic components before leaving the factory.

[0004] In the prior art, the rated life of the inverter under the rated environment is determined by simulating its working parameters. However, in actual use, the use situation is often more complicated, especially when faults occur frequently, the actual life is often much different from the rated life. Therefore, continuing to operate according to the rated life will result in certain losses. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent control method for a photovoltaic inverter to solve the above technical problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An intelligent control method for a photovoltaic inverter comprises the following steps:

[0008] 1. An intelligent control method for a photovoltaic inverter, characterized in that it comprises the following steps:

[0009] S1: Obtain a calibration period T, and select a number of time points t from the calibration period T according to a preset time interval Δt 0 ,t 1 ,t 2 ,t 3 ,..., where t 0 The start time of the calibration cycle is used to obtain the load data W at the inverter. 0 ,W 1 ,W 2 , W 3 , ...;

[0010] S2: Calculate the average rated load R of the inverter pn , and its calculation formula is as follows:

[0011]

[0012] Among them, W i represents the load data at the i-th time point, W i-1 represents the load data at the i-1th time point, R j represents the average load between the i-th time point and the i-1-th time point, R pn represents the average calibrated load of the nth inverter in the calibration period T;

[0013] S3: Obtain the inverter fault count G within the calibration period sn , according to the fault number G of the nth inverter in the calibration period sn Calculate the failure rate G n =G sn / T, by averaging the load R pn and failure rate G n Set the Pearson correlation coefficient r;

[0014] S4: When r≠0, obtain the average calibration load R pn and failure rate G n Establish a linear regression model, calculate the regression coefficients a and b, and establish the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula of ;

[0015] S5: Through real-time load W n Substitute into the linear regression equation to calculate the expected inverter failure rate G yn , get the rated life T of the inverter 额 , expected failure rate G yn , failure rate G n and its calibration period T, calculate the expected remaining life M n , the calculation formula is as follows, where G yn ≤G n :

[0016]

[0017] Get the remaining lifespan M n And sort them from large to small, and generate a remaining life sorting table.

[0018] As a further solution of the present invention: the calibration period is the actual total operating time.

[0019] As a further solution of the present invention: the average calibration load R pn, can also be calculated by the following method: obtain the load data change curve F(t) over time within the calibration period T, and the average calibration load R pn The calculation formula is as follows:

[0020]

[0021] Among them, t d The end time of the calibration cycle.

[0022] As a further solution of the present invention: the calculation formula for calculating the Pearson correlation coefficient r is as follows:

[0023]

[0024] Where h is the total number of inverters.

[0025] As a further solution of the present invention: when r=0, according to the average calibration load R pn and failure rate G n Draw a scatter plot and get the position of each point (R pn ,G n ), filter out abnormal data by drawing a box plot, exclude the abnormal data, and repeat step S3.

[0026] As a further solution of the present invention: the value range of the Pearson correlation coefficient r is [-1, 1], when r≠0, the average calibration load R is obtained pn and failure rate G n , with an average rated load R pn is the independent variable, the failure rate G n Establish a linear regression model for the dependent variable and establish the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula.

[0027] As a further solution of the present invention: the real-time load W n To obtain the load data of the inverter at the current time node.

[0028] A photovoltaic inverter intelligent control system, characterized by comprising:

[0029] The data acquisition module acquires the calibration period T and selects several time points t from the calibration period T at a preset time interval Δt. 0 ,t 1 ,t 2 ,t 3 ,..., obtain the load data W at the inverter 0 ,W 1 ,W 2 , W3 , ..., obtain the inverter fault times G within the calibration period sn , rated life T 额 , expected failure rate G yn , failure rate G n and its calibration period T, to obtain the remaining life M n And sort them from large to small, and generate a remaining life ranking table;

[0030] Data processing module: calculate the average rated load R of each inverter circuit pn , and its calculation formula is as follows:

[0031]

[0032] Calculate the failure rate G n , by averaging the load R pn and failure rate G n Set the Pearson correlation coefficient r. When r≠0, the average calibration load R pn and failure rate G n Establish a linear regression model, calculate the regression coefficients a and b, and obtain the linear regression equation formula for the expected failure rate. n Calculate the expected inverter failure rate G yn , calculate the expected remaining life M n , the calculation formula is as follows, where G yn ≤G n :

[0033]

[0034] Beneficial effects of the present invention: In the present invention, a corresponding calibration process is set. During the calibration period, real-time load data recorded at each time point of each group of inverter circuits is obtained, and the time intervals between any adjacent time points are the same. The average calibrated load and failure rate of each inverter circuit are calculated, and the Pearson correlation coefficient is set according to the average calibrated load and failure rate to determine the linear relationship. A linear regression model is established through the two groups of data of the average calibrated load and failure rate, and the expected failure rate is calculated. The remaining life is predicted based on the expected failure rate, thereby avoiding late detection and replacement of the inverter and reducing cost losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention will be further described below in conjunction with the accompanying drawings.

[0036] Figure 1 It is a flow chart of an intelligent control method for a photovoltaic inverter of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] See also Figure 1 As shown, the present invention is an intelligent control method for a photovoltaic inverter, comprising the following steps:

[0039] The present invention relates to a calibration and fault prediction method for an inverter circuit, wherein a corresponding calibration process is provided. During the calibration period, by acquiring the real-time load data recorded by each group of inverter circuits at each time point, it is ensured that the time intervals between any adjacent time points are the same, and the shorter the time interval, the higher the accuracy of the acquired data. Based on the acquired real-time load data, the average calibrated load of each inverter circuit and its corresponding failure rate are calculated. By setting the Pearson correlation coefficient, the linear relationship between the average calibrated load and the failure rate is determined. In addition, a linear regression model is established based on the two sets of data to calculate the expected failure rate, and the remaining life of the inverter is calculated accordingly to avoid the inverter from being detected and replaced at a too late stage, thereby reducing cost losses.

[0040] 1. An intelligent control method for a photovoltaic inverter, characterized in that it comprises the following steps:

[0041] S1: Obtain a calibration period T, and select a number of time points t from the calibration period T according to a preset time interval Δt 0 ,t 1 ,t 2 ,t 3 ,..., where t 0 The start time of the calibration cycle is used to obtain the load data W at the inverter. 0 ,W 1 ,W 2 , W 3 , ...;

[0042] S2: Calculate the average rated load R of the inverter pn , and its calculation formula is as follows:

[0043]

[0044]

[0045] Among them, W i represents the load data at the i-th time point, W i-1represents the load data at the i-1th time point, R j represents the average load between the i-th time point and the i-1-th time point, R pn represents the average calibrated load of the nth inverter in the calibration period T;

[0046] S3: Obtain the inverter fault count G within the calibration period sn , according to the fault number G of the nth inverter in the calibration period sn Calculate the failure rate G n =G sn / T, by averaging the load R pn and failure rate G n Set the Pearson correlation coefficient r;

[0047] S4: When r≠0, obtain the average calibration load R pn and failure rate G n Establish a linear regression model, calculate the regression coefficients a and b, and establish the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula of ;

[0048] S5: Through real-time load W n Substitute into the linear regression equation to calculate the expected inverter failure rate G yn , get the rated life T of the inverter 额 , expected failure rate G yn , failure rate G n and its calibration period T, calculate the expected remaining life M n , the calculation formula is as follows, where G yn ≤G n :

[0049]

[0050] Get the remaining lifespan M n And sort them from large to small, and generate a remaining life sorting table.

[0051] It is understandable that in the case of frequent failures of the inverter, its remaining life will be reduced accordingly, and this phenomenon directly leads to the actual failure rate being higher than the expected failure rate. In order to more accurately reflect the actual performance of the inverter, the present invention proposes a method for correction by calculating the deviation between the expected failure rate and the actual failure rate. Specifically, by making appropriate adjustments and corrections to the actual usage time, the remaining life data of the inverter can be obtained more accurately. This method not only significantly enhances the ability to evaluate the performance of the inverter, but also improves the effectiveness of maintenance and management, allowing operation and maintenance personnel to better formulate corresponding maintenance plans, thereby reducing the risk of failures and improving the operating efficiency of the equipment. Through this innovative evaluation mechanism, while ensuring the stable operation of the equipment, the maintenance cost can be effectively controlled, and ultimately the economic benefits can be maximized.

[0052] It can be known that the Pearson correlation coefficient r is calculated using the following formula:

[0053]

[0054] By averaging the load R pn and failure rate G n Set the value of Pearson correlation coefficient r. a and b are regression coefficients, and the calculation formula of a is The calculation formula of b is

[0055] It can be understood that the Pearson correlation coefficient r provides a quantitative indicator that can effectively measure the linear relationship between the average calibrated load and the failure rate, and help analyze the degree of correlation between the two. The value range of the correlation coefficient r is fixed between [-1,1]. When -1≤r<0, the two sets of data are negatively correlated, that is, when one variable increases, the other variable tends to decrease. When r=0, it means that the two sets of variables have no linear relationship. When 0<r≤1, the two sets of data are positively correlated, that is, when one variable increases, the other variable also tends to increase.

[0056] In a preferred embodiment of the present invention, in S1, all inverters are in the same semi-enclosed environment. In this semi-enclosed environment, the working environment of the inverters is kept in the same state through equipment such as thermostats, dehumidifiers, humidifiers, and ventilation systems.

[0057] It can be understood that the consistency of environmental conditions enables each group of inverters to operate in the same working environment, thereby ensuring the comparability and consistency of the performance of each group of inverters, which facilitates performance evaluation and optimization.

[0058] In a preferred embodiment of the present invention, in S1, the calibration period is the actual total operating time, excluding the time when the inverter is turned off, and each inverter is in a normal operating state during the calibration period.

[0059] It is understandable that more realistic data can be obtained by calibrating during the actual operating time, excluding the time when the inverter is not working. Such data is closer to the actual situation and helps to improve the accuracy of the calibration results.

[0060] In a preferred embodiment of the present invention, in S2, the average calibration load R pn , can also be calculated by the following method: obtain the load data change curve F(t) over time within the calibration period T, and the average calibration load R pn The calculation formula is as follows:

[0061]

[0062] Among them, t d The end time of the calibration cycle.

[0063] It is understood that in the implementation of the present invention, the length of the calibration cycle is proportional to the average calibration load R pn When the calibration period T is too short, the time points cannot be effectively divided according to the preset time interval Δt, which will result in the average calibration load R pn The calculation result is not accurate enough. In order to overcome this problem, the present invention provides another calculation method, which obtains load data within the calibration period T, and can divide the time points according to the preset time interval Δt when the calibration period is too short to obtain more accurate load data, thereby calculating a more reliable average calibration load R pn The average calibration load R is ensured under different calibration conditions in two ways. pn The calculation can reflect the real load state of the inverter in actual application.

[0064] In a preferred embodiment of the present invention, in S4, when r=0, the two sets of data are plotted into a scatter plot to obtain the position of each point (R pn ,G n ), filter out abnormal data by drawing a box plot, exclude the abnormal data, and repeat step S3.

[0065] It is understandable that scatter plots provide an intuitive display of data, making it easier to identify trends, patterns, and potential outliers. This visualization method makes data analysis more intuitive and easy to understand. By excluding abnormal data, the quality of the data set can be significantly improved, the impact of noise on the analysis results can be reduced, and the conclusions can be more reliable.

[0066] In a preferred embodiment of the present invention, in S4, the value range of the Pearson correlation coefficient r is [-1, 1], when r≠0, the average calibration load R is obtained. pn and failure rate G n , with an average rated load R pn is the independent variable, the failure rate G n Establish a linear regression model for the dependent variable and establish the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula, the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula is as follows:

[0067] G yn =aW n +b.

[0068] It is understandable that by establishing a linear regression model, real-time monitoring of the inverter load condition can be achieved, so as to timely warn of potential failure risks, thereby improving the reliability of the system. Furthermore, by calculating the deviation between the expected failure rate and the actual failure rate, the calibration cycle is corrected to obtain the remaining life of the inverter. This method effectively improves the maintenance and management efficiency of the inverter and ensures the stable operation of the system.

[0069] In a preferred embodiment of the present invention, in S5, the real-time load Wn is load data of the inverter obtained at the current time node.

[0070] It is understandable that real-time monitoring of load changes can be achieved. The benefit of this approach is that the impact of load fluctuations on the expected failure rate can be identified in a timely manner, thereby more accurately adjusting the relationship between the expected failure rate and the actual failure rate, and improving the reliability and safety of the system.

[0071] A photovoltaic inverter intelligent control system, characterized by comprising:

[0072] The data acquisition module acquires the calibration period T and selects several time points t from the calibration period T at a preset time interval Δt. 0 ,t 1 ,t 2 ,t 3 ,..., obtain the load data W at the inverter 0 ,W 1 ,W 2 , W 3 , ..., obtain the inverter fault times G within the calibration period sn , rated life T 额 , expected failure rate G yn, failure rate G n and its calibration period T, to obtain the remaining life M n And sort them from large to small, and generate a remaining life ranking table;

[0073] Data processing module: calculate the average rated load R of each inverter circuit pn , and its calculation formula is as follows:

[0074]

[0075] Calculate the failure rate G n , by averaging the load R pn and failure rate G n Set the Pearson correlation coefficient r. When r≠0, the average calibration load R pn and failure rate G n Establish a linear regression model, calculate the regression coefficients a and b, and obtain the linear regression equation formula for the expected failure rate. n Calculate the expected inverter failure rate G yn , calculate the expected remaining life M n , the calculation formula is as follows, where G yn ≤G n :

[0076]

[0077] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent control method for a photovoltaic inverter, characterized in that: The following steps are involved: S1: Obtain a calibration cycle T, and select a number of time points t0, t1, t2, t3, ... from the calibration cycle T according to a preset time interval Δt, where t0 is the start time of the calibration cycle, and obtain load data W0, W1, W2, W3, ... of the inverter; S2: Calculate the average rated load R of the inverter pn , and its calculation formula is as follows: Among them, W i represents the load data at the i-th time point, W i-1 represents the load data at the i-1th time point, R j represents the average load between the i-th time point and the i-1-th time point, R pn represents the average calibrated load of the nth inverter in the calibration period T; S3: Obtain the inverter fault count G within the calibration period sn , according to the fault number G of the nth inverter in the calibration period sn Calculate the failure rate G n =G sn / T, by averaging the load R pn and failure rate G n Set the Pearson correlation coefficient r; S4: When r≠0, obtain the average calibration load R pn and failure rate G n Establish a linear regression model, calculate the regression coefficients a and b, and establish the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula of ; S5: Through real-time load W n Substitute into the linear regression equation to calculate the expected inverter failure rate G yn , get the rated life of the inverter T 额 , expected failure rate G yn , failure rate G n and its calibration period T, calculate the expected remaining life M n , the calculation formula is as follows, where G yn ≤G n : Get the remaining lifespan M n And sort them from large to small, and generate a remaining life sorting table.

2. The intelligent control method of a photovoltaic inverter according to claim 1, characterized in that: In S1, the calibration period is the actual total operating time of the inverter.

3. The intelligent control method of a photovoltaic inverter according to claim 1, characterized in that: In S2, the average rated load R pn , can also be calculated by the following method: obtain the load data change curve F(t) over time within the calibration period T, and the average calibration load R pn The calculation formula is as follows: Among them, t d The end time of the calibration cycle.

4. The intelligent control method of a photovoltaic inverter according to claim 1, characterized in that: In S3, the Pearson correlation coefficient r is calculated using the following formula: Where h is the total number of inverters.

5. The intelligent control method of a photovoltaic inverter according to claim 6, characterized in that: In S4, when r=0, according to the average calibration load R pn and failure rate G n Draw a scatter plot and get the position of each point (R pn ,G n ), filter out abnormal data by drawing a box plot, exclude the abnormal data, and repeat step S3.

6. The intelligent control method of a photovoltaic inverter according to claim 1, characterized in that: In S4, the value range of the Pearson correlation coefficient r is [-1, 1]. When r≠0, the average calibration load R is obtained. pn and failure rate G n , with an average rated load R pn is the independent variable, the failure rate G n Establish a linear regression model for the dependent variable and establish the expected failure rate G yn The real-time load W of the nth inverter n The linear regression equation formula.

7. The intelligent control method of a photovoltaic inverter according to claim 6, characterized in that: In S5, the real-time load W n To obtain the load data of the inverter at the current time node.

8. An intelligent control system for a photovoltaic inverter, characterized in that: include: The data acquisition module obtains the calibration period T, and selects several time points t0, t1, t2, t3, ... from the calibration period T according to the preset time interval Δt, obtains the load data W0, W1, W2, W3, ... of the inverter, and obtains the number of faults G of the inverter within the calibration period sn , rated life T 额 , expected failure rate G yn , failure rate G n and its calibration period T, to obtain the remaining life M n And sort them from large to small, and generate a remaining life ranking table; Data processing module: calculate the average rated load R of each inverter circuit pn , and its calculation formula is as follows: Calculate the failure rate G n , by averaging the load R pn and failure rate G n Set the Pearson correlation coefficient r. When r≠0, the average calibration load R pn and failure rate G n Establish a linear regression model, calculate the regression coefficients a and b, and obtain the linear regression equation formula for the expected failure rate. n Calculate the expected inverter failure rate G yn , calculate the expected remaining life M n , the calculation formula is as follows, where G yn ≤G n :

Images