Photovoltaic power distribution operation early warning method, system and device and storage medium

Abstract

The invention discloses a photovoltaic power distribution operation early warning method, system and device and a storage medium, and relates to the technical field of photovoltaic power distribution, and the method comprises the steps: removing interference data in photovoltaic power station monitoring data; after interference data is removed, abnormal data in the monitoring data of the photovoltaic power station are determined through time sequence inspection and analysis; setting an early warning measurement index and an error weighting index of the abnormal data, and calculating fault early warning thresholds of the two indexes through variable transformation; and carrying out photovoltaic power station operation early warning according to the abnormal data and the fault early warning threshold. According to the method, the data quality is improved by removing the interference data, then the abnormal data is quickly and accurately identified by using the time sequence analysis technology, and the early warning index of the abnormal data is set, so that the early warning is more intelligent and flexible, adjustment can be performed according to different conditions, unnecessary false early warning and missing early warning are avoided, and the early warning efficiency is improved. And the early warning accuracy and reliability are improved.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic power distribution technology, and in particular to a photovoltaic power distribution operation early warning method, system, device and storage medium. Background Art

[0002] With the rapid development and large-scale application of photovoltaic power generation technology, the stable operation and efficient management of photovoltaic power stations have become particularly important. However, due to the complex environmental conditions during the operation of photovoltaic power stations, such as weather changes, equipment aging, and interference from data acquisition equipment, noise, anomalies or distortion often appear in the monitoring data, which makes it impossible to carry out accurate fault warnings, posing a hidden danger to the safety of power station operation. Summary of the invention

[0003] In order to achieve accurate fault warning of photovoltaic power stations and ensure the safe operation of photovoltaic power stations, the present application proposes a photovoltaic distribution operation warning method, system, equipment and storage medium. The present application improves data quality by removing interference data, and then uses time series analysis technology to quickly and accurately identify abnormal data and set warning indicators for abnormal data, making the warning more intelligent and flexible, and able to adjust according to different situations, avoiding unnecessary false warnings and missed warnings, and improving the accuracy and reliability of the warning.

[0004] This application is implemented through the following technical solutions: A photovoltaic power distribution operation early warning method, the early warning method comprising: Remove interference data from photovoltaic power station monitoring data; After removing the interference data, the abnormal data in the photovoltaic power station monitoring data is determined through time series test analysis; Set the warning metric and error weighting indicators for abnormal data, and calculate the fault warning thresholds of the two indicators through variable transformation; Based on the abnormal data and its fault warning threshold, a photovoltaic power station operation warning is performed.

[0005] In some implementations, the interference data removal process includes: Interference data is identified from the monitoring data of the photovoltaic power station, and the interference data identification process includes: firstly calculating the average value of the absolute value of the data of the monitoring data waveform segment; then identifying the interference data based on the set judgment criterion; wherein the set judgment is accurate as follows: if the absolute value of the data is greater than or equal to the product of the average value and the judgment threshold, the data is interference data, otherwise it is not interference data; The identified interference data is suppressed by using a median filtering method, specifically: the interference data is replaced by a median value of the data value in the filtering window.

[0006] In some embodiments, the time series test analysis process includes: When there are only two time series variables in the monitoring data, the EG two-step test method based on regression residuals is used to perform a stationarity test analysis on the time series variables, including: The unit root test method is used to test the stability of time series variables and determine the single integral order; Construct a stationary regression model to obtain a stationary residual sequence; A unit root test is performed on the stationary residual sequence. If the stationary hypothesis test does not hold, it indicates that the data is abnormal.

[0007] In some embodiments, the time series test analysis process includes: When the time series variable in the monitoring data is greater than 2, the Johansen test method is used to perform a stationary test analysis on the time series variable, including: Determine the number of mutually independent stationarity vectors; Constructing a vector error correction model for the considered time series variables; estimating parameters of the vector error correction model using a maximum likelihood method; The vector error correction model is subjected to a maximum eigenvalue test. If the stationarity hypothesis test fails, it indicates that the data is abnormal.

[0008] In some implementations, the fault warning threshold calculation process includes: The original index data are processed by variable transformation to transform them into data that conform to the standard normal distribution; the transformation parameters in the variable transformation are calculated by the maximum likelihood estimation method; The probability density function of the data that conforms to the normal distribution is obtained, and the threshold of the fault warning indicator is obtained by combining the confidence interval solution; The threshold value obtained by the solution is subjected to inverse transformation of the variables to obtain the fault warning threshold value corresponding to the original indicator data.

[0009] In some embodiments, the photovoltaic power station operation early warning process includes: The identified abnormal data is compared with its corresponding fault warning threshold. If the abnormal data is greater than the fault warning threshold, a warning is issued.

[0010] In a second aspect, the present application proposes a photovoltaic power distribution operation early warning system, the early warning system comprising: An interference removal module, which is used to remove interference data in photovoltaic power station monitoring data; An abnormality identification module, wherein the abnormality identification module determines abnormal data in the photovoltaic power station monitoring data after removing interference data through time series inspection analysis; An early warning indicator module, which is used to set early warning measurement indicators and error weighted indicators for abnormal data, and calculate fault early warning thresholds of the two indicators through variable transformation; And, a comparison module, wherein the comparison module performs photovoltaic power station operation warning according to the abnormal data and the fault warning threshold value.

[0011] In some embodiments, the early warning indicator module further includes: A variable transformation unit, wherein the variable transformation unit performs variable transformation on the original indicator data to convert it into data conforming to a standard normal distribution; the transformation parameters in the variable transformation are calculated using a maximum likelihood estimation method; A threshold value solving unit, wherein the threshold value solving unit obtains a probability density function of data that conforms to a normal distribution, and obtains a threshold value of a fault warning indicator by combining the confidence interval solution; And, a variable inverse transformation unit, which performs variable inverse transformation on the threshold value obtained by the solution to obtain a fault warning threshold value corresponding to the original indicator data.

[0012] In a third aspect, the present application proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned photovoltaic distribution operation early warning method when executing the computer program.

[0013] In a fourth aspect, the present application proposes a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the above-mentioned photovoltaic power distribution operation early warning method is implemented.

[0014] The present application proposes a photovoltaic distribution operation early warning method, system, device and storage medium, which first remove interference data in the monitoring data of the photovoltaic power station, can effectively filter out invalid or irrelevant data, reduce false alarms and errors, and ensure the quality of the monitoring data; then use time series analysis to quickly and accurately identify potential problems or abnormal conditions in the operation of the photovoltaic power station, so as to timely capture abnormal conditions that may affect the operation of the power station, and improve the sensitivity of fault detection; finally, it also sets early warning indicators for abnormal data, and triggers early warnings by calculating early warning thresholds, making the early warning more intelligent and flexible, and can be adjusted according to different situations, avoiding unnecessary false warnings and missed warnings, and further improving the reliability and accuracy of the early warning.

[0015] The present application proposes a photovoltaic power distribution operation early warning method, system, device and storage medium, which enable photovoltaic power stations to detect and handle potential problems in advance, thereby reducing downtime and the impact of failures, and improving the overall reliability and operational safety of the photovoltaic power station system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the embodiments of the present application, constitute a part of the present application, and do not constitute a limitation on the embodiments of the present application. In the drawings: Figure 1 A flowchart of the early warning method proposed in the embodiment of the present application; Figure 2 This is a functional block diagram of the early warning system proposed in the embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with examples and drawings. The illustrative implementation scheme of the present application and its description are only used to explain the present application and are not intended to limit the present application.

[0018] Embodiment: During the operation of a photovoltaic power station, the complex environmental conditions and the interference of data acquisition equipment make it impossible to provide accurate fault warnings, which brings potential safety hazards to the operation of the power station. In view of this, this embodiment proposes a photovoltaic power distribution operation warning method.

[0019] like Figure 1 As shown, the early warning method proposed in this embodiment includes the following steps: Step 100: removing interference data from photovoltaic power station monitoring data.

[0020] In the monitoring of photovoltaic power stations, invalid or irrelevant data caused by external environmental interference can easily cause false alarms of photovoltaic power stations. Therefore, it is necessary to effectively filter invalid or irrelevant data, reduce false alarms and errors, and ensure the quality of monitoring data, thereby providing a more accurate data basis for subsequent analysis. In this embodiment, the specific process of removing interference data is as follows: Step 101: Identify interference data from photovoltaic power station monitoring data.

[0021] First, calculate the average value of the absolute value of the data of the monitoring data waveform segment; the average value calculation formula is as follows:

[0022] in, Indicates the first The data value of the sampling point, is the number of sampling points, is the mean Then, based on the set judgment criteria, the interference data is identified; the judgment criteria are as follows:

[0023] in, is an outlier flag, if The value of It is interference data; otherwise, it is not interference data. is a given discrimination threshold.

[0024] Step 102: Suppress the identified interference data.

[0025] This embodiment may use but is not limited to a median filtering method to suppress interference data. The median filtering principle is:

[0026] in, The first The data value of the sampling point, is the median filter output result, is the median filter length (usually set to an odd number), Indicates the median value of the data within the window length.

[0027] In this way, the interference data monitored by the photovoltaic power station can be identified and suppressed, thereby avoiding false warnings caused by the interference data and providing a more accurate and reliable data basis for subsequent processes.

[0028] Step 200: After removing the interference data, abnormal data in the photovoltaic power station monitoring data is determined through time series test analysis.

[0029] This embodiment uses time series to determine whether there is abnormal data in the monitored data, and then issues an early warning based on the abnormal data.

[0030] The central idea of ​​time series is that even if two or more time series variables are non-stationary individually, their combination may produce a new stationary relationship. This stationary linear relationship is defined as stationarity, which means that multiple variables can be represented as a long-term equilibrium stable relationship. If there is a stationary linear combination between non-stationary series, then they can be said to be stationary.

[0031] In the definition of stationarity, it is assumed that there is a stationary relationship between two sequences, where the input time variable sequence is The corresponding response time variable sequence is , the regression model between them can be expressed as:

[0032] Where: is the stationary vector: is the regression residual, t Indicates the duration of the time series, a sequence with a stationary relationship and The regression residual series between It is a stationary sequence. As can be seen from the above formula, the ultimate goal of the stationary process is to identify the stationary vector to ensure the stationary residual sequence.

[0033] For a stationary time series variable , if the variable itself is stationary, it is called 0th order integral and is recorded as , Indicates 0th order integration. For non-stationary time series variables, if it passes through After the second difference operation, it becomes a stationary time series variable, which is called Order single, indicating time series variables Existence unit root, denoted by , express d The stationary analysis process is only valid for non-stationary variables with the same stationary order. Therefore, in order to obtain the stationary vector , we first need to determine the non-stationary order of the variables. Then, the stationary vector It can be calculated during the stationarity test analysis.

[0034] Stationarity tests can be divided into methods based on regression parameters and methods based on regression residuals. In addition, four key points should be noted in the process of stationarity analysis: First, the stationarity vector representing the relationship between non-stationary time series variables is not unique: Second, each time series variable should have the same non-stationarity, that is, the same order of stationarity to perform the stationarity analysis process; Third, for the number of variables Time series data of Linearly independent stationarity vectors: Fourth, each time series variable follows a similar sequence trend.

[0035] This embodiment uses the unit root test method to test the stability of the time series. The unit root test method is to prove that when a time series data has a unit root, the series is non-stationary. The stability test process is as follows: Starting from the form of the first-order autoregressive model, it can be deduced that the first-order autoregressive model can be expressed as:

[0036] Where: is the value at the current moment; is the value at the previous moment; The mean is 0 and the variance is Independent normally randomly distributed Gaussian white noise sequence; is a real number whose value determines the root of the characteristic equation of the above model. The root of the characteristic equation of any time series process will determine its stationarity. When the value of is greater than or equal to 1, the time series process is non-stationary. When the value of is less than 1, the time series process is stationary. When , there is a unit root in the process, and the above formula can be transformed into:

[0037] Through one difference operation, the non-stationary time series process is transformed into a stationary process, that is, , is a first-order single integration. When the time series process is fitted to (1) When the model (i.e., the first-order autoregressive model) is used, the stationarity information of the time series process can be obtained from the defined characteristic root parameters, which is usually obtained by testing When the time series data is fitted to a more complex time series model, the stationary test follows the above principle. The more complex time series model is expressed as:

[0038] in, is a difference operator, defined as ; is a coefficient of a real number: are the model coefficients: is the hysteresis number, which ensures becomes a white noise sequence. Model (i.e. p To transform the order autoregressive model into the above form, the following substitutions need to be made:

[0039] in, for The coefficients of the model. The characteristic equation of the model can be replaced by the above process to obtain the characteristic equation, namely:

[0040] in, is the root of the above formula. In the characteristic equation of the time series model, there is If at least one of the independent roots is a unit root, then we can get When there is only one unit root, consider the remaining characteristic equation Root, the above formula can be expressed as:

[0041] when When there is only one unit root, the values ​​of the other roots are all less than 1. At the same time, the following formula is the difference time series The characteristic equation of the model is:

[0042] Since all roots in the above equation are less than 1, the first-order difference is stable. is a first-order single integration, the condition in the time series model is necessary. If the form of the time series regression model includes a constant term or a trend term, additional assumptions and test statistics are required. The stationary test procedure depends on the estimation of the following model:

[0043] in, is a constant term, is the trend coefficient. In summary, the purpose of the stationary test process is to estimate the parameters in the characteristic equation of the time series model and test the null hypothesis 0. Statistic The inspection process is as follows:

[0044] in, for The least squares estimate of : for The result of hypothesis test is obtained by It is compared with the critical value in the stationary test statistics table. If the hypothesis is accepted, the time series If the hypothesis is rejected, continue Each order difference of is tested for stability until it is determined The single integer order of .

[0045] When conducting time series analysis, the stationarity test is whether there is a long-term equilibrium (or a stationary linear combination) between all elements of more than two variables. When there are only two time series variables, the EG two-step test based on regression residuals can be used; when the time series variables are greater than 2, the Johansen test can be used.

[0046] Among them, when the non-stationary order of the time series variable is determined, the variable is converted to the same order to obtain a stationary residual sequence. Therefore, the EG two-step test method based on regression residuals is adopted, and the test process is as follows: The first step is to estimate the stationary regression model and obtain a stationary residual sequence. The stationary regression model is expressed as:

[0047] In the formula: In the formula: is a time series variable; is the number of variables in the regression model; is the regression coefficient of the variable; is the stationary error. Assume that all time series variables are non-stationary and conform to , then a stationary residual sequence can be obtained:

[0048] in, is a stationary residual series.

[0049] When time series variables are stationary, they have a common trend of change and form a long-term stable relationship.

[0050] The second step is to perform a unit root test on the stationary residual series.

[0051] The stationarity test results of the residual sequence represent the stationarity test results of the time series variables, so:

[0052] Where: is the residual sequence at the current moment The first difference of is the residual sequence value of the previous moment; Represents the first-order difference of the residual sequence at the previous moment; is the model coefficient corresponding to the stationary equation; k represents a moment in time t; is the residual sequence of the stationary regression process. Once the residual sequence After the stationary test, it is confirmed to be a stationary series, and there is a stationary relationship between the time series variables. Under the assumption of non-stationarity, the estimated residual sequence is , and all parameters are 0. Therefore, it can be considered that the coefficient It shows whether the stationarity holds, and the test is transformed into:

[0053]

[0054] in: is the null hypothesis, indicating that the series is non-stationary; is the alternative hypothesis, indicating that the series is stationary. When it holds, there is no stationarity between the original time series variables: otherwise, the stationarity relationship holds.

[0055] The Johansen test is used to test whether the time series variable is stationary. If so, it further determines the number of independent stationary vectors and determines which stationary vector will produce the most stable linear combination of time series variables. The premise of the Johansen test is to use the maximum likelihood method to estimate the parameters of the vector error correction model VECM (Vector error-correction model) of the time series variable under consideration. VECM can be expressed as:

[0056] in, is a time series variable at time i The first difference of is a time series variable at time i A value of -1; is a time series variable at time i - j The first difference of For a variables to be analyzed dimensional vector, subscript It is related to the time series order; is the model order, i.e., the number of lag orders included in the model; is a normally distributed white noise sequence; is a deterministic trend item; are model parameters. The existence of an error correction model means that the included time series variables are stationary. If a true error correction model exists, The parameters in will describe the long-term balance between the variables, while The parameters in will account for the short-term adjustments needed to return to equilibrium after any data drift. Use maximum likelihood to estimate the parameter matrix In the basic principle of the Johansen test method, the problem is finally transformed into solving the generalized eigenvalue. The eigenvalue can be interpreted as the squared canonical correlation between the time series variables. The larger the eigenvalue, the more stable the stationary relationship. The optimal stationary vector corresponds to the maximum eigenvalue.

[0057] The above method can be used to determine whether there is abnormal data in the monitored data.

[0058] Step 300, setting the warning metric index and the error weighted index of the abnormal data, and calculating the fault warning thresholds of the two indicators through variable transformation.

[0059] When abnormal data is detected, this embodiment further sets the early warning metric index and error weighted index of the abnormal data, and calculates the fault early warning index of the two indicators. Among them, under normal conditions, the early warning metric index and the error weighted index are both positive and do not conform to a certain specific distribution, so their probability density functions are unknown, and the confidence interval cannot be used to determine the early warning threshold. To solve this problem, this embodiment uses variable transformation to process the indicator data in the normal state and convert it into data that conforms to the standard normal distribution.

[0060] Variable transformation transforms the dependent variable by parameters, so that there is a linear relationship between the transformed dependent variable and the independent variable, and the error components are independent and equal in variance. After variable transformation, the normality, additivity and homoscedasticity of the data can often be significantly improved. In addition, the calculation of variable transformation does not require prior information on data distribution and the calculation process is relatively simple.

[0061] Variable transformation is a family of transformations, including logarithmic transformation, square root transformation, inverse transformation and other common data transformation functions. The calculation formula of variable transformation is composed of the original data (Indicator data), transformation parameters And the transformed data It consists of three parts, and the specific calculation formula is as follows:

[0062] When changing parameters When , the variable is transformed into a logarithmic transformation; when When , the variable transformation is a square root transformation; when When , the variable transformation is the inverse transformation. When , the data needs to be translated, that is, , is the translation amount.

[0063] For the transformation parameters in variable transformation, the maximum likelihood estimation method is usually used for calculation. The specific derivation process is as follows: Assume that the distribution after transformation , then for the transformation parameters ,available and The likelihood function as follows:

[0064]

[0065] in: is the transformed random variable; is the regression coefficient vector; is the variance, The index at the current moment is The value at time; is the design matrix, which contains the information of independent variables; is the Jacobian determinant; is the value at the current moment; The index at the current moment is The value at time.

[0066] Let the likelihood function about and The derivative of is equal to zero, so we can get The maximum likelihood estimate of as follows:

[0067] get The maximum likelihood estimate of as follows:

[0068]

[0069] in, RSS represents the residual sum of squares.

[0070] In summary, we get the likelihood function The maximum values ​​are as follows:

[0071] Find the maximum value of the likelihood function The transformation parameters can be obtained The maximum value can be obtained by taking the logarithm of the above equation, as shown below:

[0072] Simplifying the above formula, we get:

[0073]

[0074] in, is the transformed data The arithmetic mean of RSS Represents the residual sum of squares. After confirmation, the above formula is a One-dimensional function, through the common optimization methods such as gradient descent method and quasi-Newton method, it is easy to calculate the maximum value of the likelihood function .

[0075] After the normal state warning metric is transformed, the data conforms to the normal distribution, making it easy to obtain the data probability density function , in determining the confidence interval Under the condition of Use the following formula to solve:

[0076] in, Fault warning indicator Less than threshold probability.

[0077] The threshold value obtained by the above process is the threshold of normally distributed data. In the online monitoring stage, the indicators of real-time data are not transformed, so it is necessary to After the inverse transformation of the variables, the corresponding distribution of the original indicator data is used as the early warning threshold in the online monitoring stage.

[0078] Step 400: Perform an operation warning of the photovoltaic power station based on the identified abnormal data and its fault warning threshold.

[0079] If the abnormal data is greater than the fault warning threshold (one or both are exceeded), an early warning is issued.

[0080] The early warning method proposed in this embodiment first removes the interference data in the monitoring data of the photovoltaic power station, which can effectively filter out invalid or irrelevant data, reduce false alarms and errors, and ensure the quality of the monitoring data, thereby providing a more accurate and reliable data basis for subsequent analysis. Then, by using time series analysis of abnormal data, potential problems or abnormal conditions in the operation of the photovoltaic power station can be identified more quickly and accurately, so as to timely capture abnormal conditions that may affect the operation of the power station and improve the sensitivity of fault detection. The early warning metric and error weighted index of the abnormal data are set, and the early warning is triggered by calculating the early warning threshold, so that the early warning is more intelligent and flexible, and can be adjusted according to different situations, avoiding unnecessary false warnings and missed warnings.

[0081] This embodiment proposes an early warning mechanism, so that potential problems in the photovoltaic power station can be discovered and handled in advance, thereby reducing downtime and failure impacts, and improving the overall reliability and operational safety of the photovoltaic power station system.

[0082] Based on the same technical concept as above, this embodiment also proposes a photovoltaic power distribution operation early warning system, such as Figure 2 As shown, the early warning system proposed in this embodiment includes: The interference removal module is used to remove interference data in the monitoring data of the photovoltaic power station.

[0083] The abnormality identification module determines the abnormal data in the photovoltaic power station monitoring data after the interference data is removed through time series inspection and analysis.

[0084] The early warning indicator module is used to set the early warning measurement indicators and error weighted indicators of abnormal data, and calculate the fault early warning thresholds of the two indicators through variable transformation.

[0085] And, a comparison module, which performs photovoltaic power station operation warning according to the abnormal data and its fault warning threshold. The details are as described in the above step 400, which will not be described in detail here.

[0086] Furthermore, the interference removal module also includes: The identification unit identifies interference data from the photovoltaic power station monitoring data. The specific process is as described in the above step 101, and will not be described in detail here.

[0087] and a filtering unit, the filtering unit being used to suppress the identified interference data. The specific process is as described in the above step 102, and will not be described in detail here.

[0088] Furthermore, the abnormality identification module also includes: The first time series test and analysis unit, when there are only two time series variables in the monitoring data, uses the EG two-step test method based on regression residuals to perform a stationarity test and analysis on the time series variables.

[0089] And, a second time series test and analysis unit, when the time series variable in the monitoring data is greater than 2, the second time series test and analysis unit uses the Johansen test method to perform a stationarity test analysis on the time series variable.

[0090] It should be noted that the specific implementation process of each functional unit of the abnormal identification module is as described in the above step 200, which will not be repeated here.

[0091] Furthermore, the early warning indicator module also includes: The variable transformation unit performs variable transformation on the original indicator data and converts it into data that conforms to the standard normal distribution; the transformation parameters in the variable transformation are calculated using the maximum likelihood estimation method.

[0092] The threshold value solving unit obtains the probability density function of the data that conforms to the normal distribution, and obtains the threshold value of the fault warning indicator by combining the confidence interval.

[0093] And, a variable inverse transformation unit, which performs variable inverse transformation on the threshold value obtained by the solution to obtain a fault warning threshold value corresponding to the original indicator data.

[0094] It should be noted that the specific implementation process of each functional unit of the early warning indicator module is as described in the above step 300, which will not be repeated here.

[0095] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0096] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0099] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A photovoltaic power distribution operation early warning method, characterized in that: The early warning method comprises: Remove interference data from photovoltaic power station monitoring data; After removing the interference data, the abnormal data in the photovoltaic power station monitoring data is determined through time series test analysis; Set the warning metric and error weighting indicators for abnormal data, and calculate the fault warning thresholds of the two indicators through variable transformation; Based on the abnormal data and its fault warning threshold, a photovoltaic power station operation warning is performed.

2. A photovoltaic power distribution operation early warning method according to claim 1, characterized in that: The interference data removal process includes: Interference data is identified from the monitoring data of the photovoltaic power station, and the interference data identification process includes: firstly calculating the average value of the absolute value of the data of the monitoring data waveform segment; then identifying the interference data based on the set judgment criterion; wherein the set judgment is accurate as follows: if the absolute value of the data is greater than or equal to the product of the average value and the judgment threshold, the data is interference data, otherwise it is not interference data; The identified interference data is suppressed by using a median filtering method, specifically: the interference data is replaced by a median value of the data value in the filtering window.

3. A photovoltaic power distribution operation early warning method according to claim 1, characterized in that: The time series test analysis process includes: When there are only two time series variables in the monitoring data, the EG two-step test method based on regression residuals is used to perform a stationarity test analysis on the time series variables, including: The unit root test method is used to test the stability of time series variables and determine the single integral order; Construct a stationary regression model to obtain a stationary residual sequence; A unit root test is performed on the stationary residual sequence. If the stationary hypothesis test does not hold, it indicates that the data is abnormal.

4. A photovoltaic power distribution operation early warning method according to claim 1, characterized in that: The time series test analysis process includes: When the time series variable in the monitoring data is greater than 2, the Johansen test method is used to perform a stationary test analysis on the time series variable, including: Determine the number of mutually independent stationarity vectors; Constructing a vector error correction model for the considered time series variables; estimating parameters of the vector error correction model using a maximum likelihood method; The vector error correction model is subjected to a maximum eigenvalue test. If the stationarity hypothesis test fails, it indicates that the data is abnormal.

5. A photovoltaic power distribution operation early warning method according to any one of claims 1 to 4, characterized in that: The fault warning threshold calculation process includes: The original index data are processed by variable transformation to transform them into data that conform to the standard normal distribution; the transformation parameters in the variable transformation are calculated by the maximum likelihood estimation method; The probability density function of the data that conforms to the normal distribution is obtained, and the threshold of the fault warning indicator is obtained by combining the confidence interval solution; The threshold value obtained by the solution is subjected to inverse transformation of the variables to obtain the fault warning threshold value corresponding to the original indicator data.

6. A photovoltaic power distribution operation early warning method according to claim 5, characterized in that: The photovoltaic power station operation early warning process includes: The identified abnormal data is compared with its corresponding fault warning threshold. If the abnormal data is greater than the fault warning threshold, a warning is issued.

7. A photovoltaic power distribution operation early warning system, characterized in that: The early warning system includes: An interference removal module, which is used to remove interference data in photovoltaic power station monitoring data; An abnormality identification module, wherein the abnormality identification module determines abnormal data in the photovoltaic power station monitoring data after removing interference data through time series inspection analysis; An early warning indicator module, which is used to set early warning measurement indicators and error weighted indicators for abnormal data, and calculate fault early warning thresholds of the two indicators through variable transformation; And, a comparison module, wherein the comparison module performs photovoltaic power station operation warning according to the abnormal data and the fault warning threshold value.

8. A photovoltaic power distribution operation warning system according to claim 7, characterized in that: The early warning indicator module also includes: A variable transformation unit, wherein the variable transformation unit performs variable transformation on the original indicator data to convert it into data conforming to a standard normal distribution; the transformation parameters in the variable transformation are calculated using a maximum likelihood estimation method; A threshold value solving unit, wherein the threshold value solving unit obtains a probability density function of data that conforms to a normal distribution, and obtains a threshold value of a fault warning indicator by combining the confidence interval solution; And, a variable inverse transformation unit, which performs variable inverse transformation on the threshold value obtained by the solution to obtain a fault warning threshold value corresponding to the original indicator data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the photovoltaic power distribution operation early warning method described in any one of claims 1-6 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the photovoltaic power distribution operation early warning method described in any one of claims 1 to 6 is implemented.

Images