A photovoltaic power generation effective data extraction system

Abstract

The present invention provides a photovoltaic power generation effective data extraction system, belonging to the field of data extraction technology; the system comprises: a monitoring module: obtaining photovoltaic power generation monitoring data through a sensor; a data processing module: obtaining standard monitoring data after preprocessing monitoring data of any data type, and obtaining trend data of the corresponding time node of the standard monitoring data; a prediction model module: obtaining historical power generation data of any data type according to the history library of the photovoltaic system, substituting the trend data into the prediction model to obtain prediction data, and obtaining comparison deviation; an error analysis module: analyzing the discrete situation of the trend data by using the error deviation; a storage module: filtering out data exceeding a preset discrete rate in the trend data, extracting effective data after effective processing, and storing it in an effective database. The data of the photovoltaic power generation system is effectively extracted and analyzed, and the performance of the system is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of data extraction, and in particular to a photovoltaic power generation effective data extraction system. Background Art

[0002] At present, photovoltaic power stations usually use photovoltaic modules and inverters as rated capacity for data storage and operation. However, with the continuous operation of photovoltaic power stations every year, the power generation efficiency of photovoltaic modules and inverters continues to decline. The actual power generation capacity of photovoltaic power stations will show a step-by-step decline and dynamically changing uncertain data, resulting in the system being unable to accurately calculate the actual power generation capacity of photovoltaic power stations.

[0003] Therefore, the present invention proposes a photovoltaic power generation effective data extraction system. Summary of the invention

[0004] The present invention provides a photovoltaic power generation effective data extraction system, which is used to obtain sensor data of a photovoltaic power station, obtain trend data after preprocessing, build a prediction model based on historical data and obtain prediction data, perform deviation analysis on the trend data based on the prediction data, and filter and store effective data of the trend data according to a preset discreteness.

[0005] In one aspect, the present invention provides a photovoltaic power generation effective data extraction system, comprising:

[0006] Monitoring module: deploy sensors in the photovoltaic system and obtain monitoring data of photovoltaic power generation through sensors;

[0007] Data processing module: After preprocessing the monitoring data of any data type, standard monitoring data is obtained, and the standard time series of the standard monitoring data and the trend data of the corresponding time nodes are obtained by using the time series analysis method;

[0008] Prediction model module: obtain historical power generation data of any data type according to the history library of the photovoltaic system, substitute the trend data of the same data type as the acquired historical power generation data into the prediction model to obtain prediction data, and compare the prediction data with the acquired historical power generation data to obtain the comparison deviation;

[0009] Error analysis module: using contrast deviation to analyze the error deviation of trend data of any data type, and analyzing the discrete situation of the trend data;

[0010] Storage module: performs discrete determination on the discrete situation of the trend data, filters out data exceeding a preset discrete rate in the trend data, extracts valid data after effectively processing the remaining data in the trend data according to the data type, and stores it in a valid database.

[0011] On the other hand, the monitoring module comprises:

[0012] Demand determination unit: determines the name of the target data type to be monitored in the photovoltaic system according to the preset valid data type;

[0013] Preset unit: determine the preset data collection frequency of any data type name according to the data type name-data monitoring frequency mapping table;

[0014] Sensor unit: Determine the deployment locations of sensors corresponding to different data type names according to the layout of the photovoltaic system, and configure a unique first number for any deployment location;

[0015] Configuration unit: configures a unique second number for the sensor, matches the first number with the second number according to a one-to-one correspondence between a sensor of any data type name and a corresponding deployment location, and installs the corresponding sensor at the deployment location;

[0016] Monitoring unit: After all sensors are configured, they are started to obtain the original monitoring data of the corresponding data type in real time.

[0017] On the other hand, the data processing module includes:

[0018] Data preprocessing unit: obtains raw monitoring data of a sensor of any data type, performs unit unification processing on the raw monitoring data according to the data type and a preset standard unit to obtain first data; identifies and removes abnormal values ​​of the first data according to a statistical method to obtain standard detection data;

[0019] Time standardization unit: obtaining a time set consisting of each time node of the standard detection data, and performing proportional standardization on the time nodes according to a unit distance of a standard time coordinate system, wherein the abscissa of the standard time coordinate system represents the number of the time node, and the ordinate represents the time node;

[0020] ;in, represents the standard time node of the i-th time node in the time set, Indicates that the time set has a total of n time nodes, represents the i-th time node in the time set, Represents the ratio of the unit distance of the standard time coordinate system to the time difference between the i-th time node and the i-1-th time node. Indicates based on expectations;

[0021] Finally, the standard time nodes of the standard detection data and the standard time sets they constitute are obtained.

[0022] On the other hand, the data processing module includes:

[0023] Time trend unit: maps each standard time node of the standard time set of any standard detection data to the standard time coordinate system to obtain a first discrete set consisting of the standard time nodes;

[0024] A first discrete set is subjected to curve fitting processing to obtain a first curve, the first curve is analyzed based on a time series analysis method to obtain a trend indicator of the first curve, and parameter estimation is performed on the trend indicator and a corresponding curve derivative of any standard time node to obtain trend data of the standard time node.

[0025] On the other hand, the prediction model module comprises:

[0026] Historical data unit: Use the corresponding code of any data type as the screening condition, combine SQL statements to screen the historical database to obtain the historical power generation data of the data type;

[0027] Feature analysis unit: performs covariance analysis on the historical power generation data of any data type based on the feature matrix to obtain feature indicators and feature values ​​of the historical power generation data;

[0028] Model building unit: taking all data types and their characteristic indicators and characteristic values ​​as the basic model framework and building the first model, taking the historical power generation data of all data types as the training set, iteratively fitting the training set based on the first model to obtain the first model indicator parameters; taking the standard detection data corresponding to all data types as the test set, verifying the first model indicator parameters, and optimizing and modifying the first indicator model indicator parameters to obtain the second model indicator parameters, and building the prediction model based on the second model indicator parameters;

[0029] Deviation unit: predicts the trend data of any data type based on the prediction model to obtain the prediction data, compares the prediction data with the historical power generation data of the corresponding prediction data time period, and calculates the relative absolute value of the prediction data and the historical power generation data to obtain the comparison deviation of the data type.

[0030] On the other hand, the error analysis module comprises:

[0031] Error deviation unit: Based on the comparison deviation of any data type, the prediction error of the corresponding data type is obtained as:

[0032] ; Among them, r represents the prediction error of the current data type, m represents a total of m trend data, and d represents the comparison deviation of the current data type. Indicates the mean of the predicted data of the current data type, Indicates the parameter value of the jth trend data of the current data type;

[0033] Discrete unit: Utilize the trend data variance of any data type and the prediction error of the data of the corresponding data type to comprehensively analyze the degree of deviation of each trend data of the corresponding data type and obtain the discrete situation of the trend data of the corresponding data type.

[0034] On the other hand, the storage module comprises:

[0035] Screening unit: according to the preset discrete rate, the trend data with a discrete rate greater than the preset discrete rate in all data types are marked as invalid data, and the remaining trend data are regarded as initial data;

[0036] Effective processing unit: The initial data is subjected to outlier elimination according to the standard data range of the corresponding data type, and the obtained outliers are filled with corresponding characteristic values ​​to obtain valid data of the corresponding data type.

[0037] On the other hand, the storage module further includes:

[0038] Storage unit: Build a valid database according to the preset table structure, and store valid data of all data types in the valid database according to the composite primary key of data type-data time node.

[0039] On the other hand, the storage module further includes:

[0040] Update unit: If there is a new update requirement, update and maintain the data of the storage module and save the log of each operation.

[0041] On the other hand, the updating unit comprises:

[0042] The operation log content includes: update time, update content, updater ID, and version number.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention provides a photovoltaic power generation effective data extraction system, which is used to obtain sensor data of a photovoltaic power station, obtain trend data after preprocessing, build a prediction model based on historical data and obtain prediction data, perform deviation analysis on the trend data based on the prediction data, and filter and store effective data of the trend data according to a preset discreteness. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 It is a structural schematic diagram of a photovoltaic power generation effective data extraction system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0048] Embodiment 1:

[0049] like Figure 1 As shown, a photovoltaic power generation effective data extraction system provided by an embodiment of the present invention includes:

[0050] Monitoring module: deploy sensors in the photovoltaic system and obtain monitoring data of photovoltaic power generation through sensors;

[0051] Data processing module: After preprocessing the monitoring data of any data type, standard monitoring data is obtained, and the standard time series of the standard monitoring data and the trend data of the corresponding time nodes are obtained by using the time series analysis method;

[0052] Prediction model module: obtain historical power generation data of any data type according to the history library of the photovoltaic system, substitute the trend data of the same data type as the acquired historical power generation data into the prediction model to obtain prediction data, and compare the prediction data with the acquired historical power generation data to obtain the comparison deviation;

[0053] Error analysis module: using contrast deviation to analyze the error deviation of trend data of any data type, and analyzing the discrete situation of the trend data;

[0054] Storage module: performs discrete determination on the discrete situation of the trend data, filters out data exceeding a preset discrete rate in the trend data, extracts valid data after effectively processing the remaining data in the trend data according to the data type, and stores it in a valid database.

[0055] In this embodiment, the photovoltaic system is a technical system that uses solar energy to generate electricity.

[0056] In this embodiment, the sensor is a device that can detect physical quantities or chemical quantities and convert them into measurable signals, including: current sensors, voltage sensors, temperature sensors, etc.

[0057] In this embodiment, monitoring data refers to information related to the operating status of the photovoltaic system collected by sensors or other devices, such as power generation, current, voltage, environmental parameters, etc.

[0058] In this embodiment, preprocessing refers to the steps of cleaning, transforming and formatting raw data during data analysis and modeling to improve data quality and usability.

[0059] In this embodiment, the standard monitoring data refers to photovoltaic power generation monitoring data that has been preprocessed and formatted.

[0060] In this embodiment, the time series analysis method is a statistical analysis method used to analyze data that changes over time.

[0061] In this embodiment, the standard time series refers to a data set obtained by standardizing time series data.

[0062] In this embodiment, a time node refers to a specific time mark associated with a specific data point or event in time series data.

[0063] In this embodiment, trend data refers to the results obtained by analyzing and processing monitoring data and reflecting the direction and pattern of data changes.

[0064] In this embodiment, the history library refers to a database storing monitoring data and related information in the past period of time in the photovoltaic system.

[0065] In this embodiment, the historical power generation data refers to the data recorded by the photovoltaic system in the past period of time, including: power generation, power generation time, weather conditions, photovoltaic system status and other data.

[0066] In this embodiment, the data types include: power generation data, environmental data, status data, time data, etc.

[0067] In this embodiment, the prediction model is an algorithm model used to predict future power generation and performance.

[0068] In this embodiment, the predicted data refers to an estimated value of future performance generated by a prediction model, for example, the predicted power generation data at 00:00 in the future is: 550 kWh.

[0069] In this embodiment, the comparison deviation refers to the difference generated when the prediction data generated by the prediction model is compared with the historical power generation data actually obtained in the photovoltaic system.

[0070] In this embodiment, the discrete situation refers to the distribution characteristics between data points in the data set.

[0071] In this embodiment, the discrete determination is a process of analyzing the data points in the data set to determine the dispersion degree thereof.

[0072] In this embodiment, the preset discrete rate refers to a threshold value set in advance during data processing and analysis, which is used to determine the discrete degree of data.

[0073] In this embodiment, valid data refers to data that has been screened and processed, meets specific standards, and can be used for subsequent analysis, modeling, and decision-making.

[0074] In this embodiment, the effective database refers to a data set that, after screening and processing, contains high-quality, reliable data that meets the analysis requirements.

[0075] The working principle and beneficial effects of the above technical solution are: through sensor monitoring, data processing and analysis, prediction model and error analysis, the accuracy and prediction ability of photovoltaic power generation data are optimized, the system efficiency is improved, the maintenance cost is reduced, and intelligent management is realized.

[0076] Embodiment 2:

[0077] Based on the above embodiment 1, the monitoring module includes:

[0078] Demand determination unit: determines the name of the target data type to be monitored in the photovoltaic system according to the preset valid data type;

[0079] Preset unit: determine the preset data collection frequency of any data type name according to the data type name-data monitoring frequency mapping table;

[0080] Sensor unit: Determine the deployment locations of sensors corresponding to different data type names according to the layout of the photovoltaic system, and configure a unique first number for any deployment location;

[0081] Configuration unit: configures a unique second number for the sensor, matches the first number with the second number according to a one-to-one correspondence between a sensor of any data type name and a corresponding deployment location, and installs the corresponding sensor at the deployment location;

[0082] Monitoring unit: After all sensors are configured, they are started to obtain the original monitoring data of the corresponding data type in real time.

[0083] In this embodiment, the preset valid data types refer to key data indicators that need to be collected and analyzed in photovoltaic system monitoring, including: temperature, voltage, current, power and other types.

[0084] In this embodiment, the data type name-data monitoring frequency mapping table refers to a table representing the mapping relationship between the data type name and the data monitoring frequency.

[0085] In this embodiment, the photovoltaic system layout refers to the specific location arrangement of photovoltaic modules, sensors, inverters and other equipment at the installation site.

[0086] In this embodiment, the deployment location refers to the specific location where the sensor is installed.

[0087] In this embodiment, the first number refers to a unique identifier of each sensor deployment location.

[0088] In this embodiment, the second number refers to a unique identifier of each sensor.

[0089] The working principle and beneficial effects of the above technical solution are: through clear monitoring targets, reasonable sensor layout and real-time data collection, accurate monitoring and efficient management of photovoltaic systems can be achieved, and operational efficiency and decision-making support capabilities can be improved.

[0090] Embodiment 3:

[0091] Based on the above embodiment 1, the data processing module includes:

[0092] Data preprocessing unit: obtains raw monitoring data of a sensor of any data type, performs unit unification processing on the raw monitoring data according to the data type and a preset standard unit to obtain first data; identifies and removes abnormal values ​​of the first data according to a statistical method to obtain standard detection data;

[0093] Time standardization unit: obtaining a time set consisting of each time node of the standard detection data, and performing proportional standardization on the time nodes according to a unit distance of a standard time coordinate system, wherein the abscissa of the standard time coordinate system represents the number of the time node, and the ordinate represents the time node;

[0094] ;in, represents the standard time node of the i-th time node in the time set, Indicates that the time set has a total of n time nodes, represents the i-th time node in the time set, Represents the ratio of the unit distance of the standard time coordinate system to the time difference between the i-th time node and the i-1-th time node. Indicates based on expectations;

[0095] Finally, the standard time nodes of the standard detection data and the standard time sets they constitute are obtained.

[0096] In this embodiment, the preset standard unit refers to a standardized unit used to unify the original monitoring data of different sensors, such as volts, watts, kilowatt-hours, etc.

[0097] In this embodiment, unit unification processing refers to converting the original monitoring data collected by different sources or different types of sensors into a unified standard unit.

[0098] In this embodiment, the first data refers to data obtained after unit unification processing.

[0099] In this embodiment, statistical methods are used to identify and remove outliers, such as a -score method, an interquartile range method, a moving average method, and the like.

[0100] In this embodiment, an outlier refers to a value that is significantly different from other data points in a data set.

[0101] In this embodiment, the standard detection data is a data set that meets specific standards and requirements after data preprocessing and outlier processing.

[0102] In this embodiment, the standard time coordinate system is a coordinate system used to normalize and represent time nodes, the horizontal axis represents the number of the time node, and the vertical axis represents the time node.

[0103] In this embodiment, the unit distance refers to the standardized interval between time nodes in the standard time coordinate system.

[0104] The working principle and beneficial effects of the above technical solution are: through data preprocessing and time standardization, the monitoring data units are unified, outliers are removed, and time nodes are standardized to improve data quality and consistency, providing a reliable basis for subsequent analysis and decision-making.

[0105] Embodiment 4:

[0106] Based on the above embodiment 3, the data processing module includes:

[0107] Time trend unit: maps each standard time node of the standard time set of any standard detection data to the standard time coordinate system to obtain a first discrete set consisting of the standard time nodes;

[0108] A first discrete set is subjected to curve fitting processing to obtain a first curve, the first curve is analyzed based on a time series analysis method to obtain a trend indicator of the first curve, and parameter estimation is performed on the trend indicator and a corresponding curve derivative of any standard time node to obtain trend data of the standard time node.

[0109] In this embodiment, the first discrete set refers to a set formed after each standard time node of the standard detection data is mapped to the standard time coordinate system.

[0110] In this embodiment, curve fitting is a process of approximating a set of data points to a curve by mathematical methods.

[0111] In this embodiment, the first curve refers to a mathematical function obtained by performing curve fitting on the first discrete set.

[0112] In this embodiment, the curve derivative refers to the instantaneous rate of change of the first curve at a specific standard time node.

[0113] In this embodiment, parameter estimation refers to the process of estimating parameters in a model by using statistical methods, including least squares method, maximum likelihood estimation, etc.

[0114] The working principle and beneficial effects of the above technical solution are: by mapping standard time nodes to a coordinate system, performing curve fitting and time series analysis, extracting trend indicators, and ultimately generating trend data, intuitive and accurate support is provided for data change assessment and decision-making.

[0115] Embodiment 5:

[0116] Based on the above embodiment 3, the prediction model module includes:

[0117] Historical data unit: Use the corresponding code of any data type as the screening condition, combine SQL statements to screen the historical database to obtain the historical power generation data of the data type;

[0118] Feature analysis unit: performs covariance analysis on the historical power generation data of any data type based on the feature matrix to obtain feature indicators and feature values ​​of the historical power generation data;

[0119] Model building unit: taking all data types and their characteristic indicators and characteristic values ​​as the basic model framework and building the first model, taking the historical power generation data of all data types as the training set, iteratively fitting the training set based on the first model to obtain the first model indicator parameters; taking the standard detection data corresponding to all data types as the test set, verifying the first model indicator parameters, and optimizing and modifying the first indicator model indicator parameters to obtain the second model indicator parameters, and building the prediction model based on the second model indicator parameters;

[0120] Deviation unit: predicts the trend data of any data type based on the prediction model to obtain the prediction data, compares the prediction data with the historical power generation data of the corresponding prediction data time period, and calculates the relative absolute value of the prediction data and the historical power generation data to obtain the comparison deviation of the data type.

[0121] In this embodiment, the corresponding code refers to a unique identifier used to identify a specific data type.

[0122] In this embodiment, the SQL statement refers to a statement for operating a database, and the operations include adding, deleting, modifying and checking.

[0123] In this embodiment, the feature matrix is ​​used to calculate each sample in the data set and its related features.

[0124] In this embodiment, covariance analysis is an analysis method used to measure the relationship and degree of variation between two or more variables.

[0125] In this embodiment, the feature index is a metric extracted from historical data during the data analysis and modeling process, and is used to describe key attributes of the data.

[0126] In this embodiment, the characteristic value in data analysis and modeling generally refers to a numerical value extracted through characteristic analysis or covariance analysis, which is used to describe the main characteristics of the data.

[0127] In this embodiment, the basic model framework refers to the preliminary model structure constructed during the data analysis and modeling process.

[0128] In this embodiment, the first model refers to a preliminary prediction model established in the model building unit based on all data types and their characteristic indicators and characteristic values.

[0129] In this embodiment, the training set refers to a data set used to train the first model in the model building unit.

[0130] In this embodiment, the first model indicator parameter number refers to the parameters obtained by fitting the data of the training set in the model construction unit, such as the learning rate, the regularization coefficient, etc.

[0131] In this embodiment, the test set is used to verify the performance of the model.

[0132] In this embodiment, the second model indicator parameter is a parameter obtained by optimizing and adjusting the model after the first model training and verification.

[0133] The working principle and beneficial effects of the above technical solution are: by screening historical power generation data, feature analysis and model construction, trend forecasting and historical data comparison can be achieved, forecasting accuracy and decision support capabilities can be improved, and continuous improvement and optimization can be promoted.

[0134] Embodiment 6:

[0135] Based on the above embodiment 1, the error analysis module includes:

[0136] Error deviation unit: Based on the comparison deviation of any data type, the prediction error of the corresponding data type is obtained as:

[0137] ; Among them, r represents the prediction error of the current data type, m represents a total of m trend data, and d represents the comparison deviation of the current data type. Indicates the mean of the predicted data of the current data type, Indicates the parameter value of the jth trend data of the current data type;

[0138] Discrete unit: Utilize the trend data variance of any data type and the prediction error of the data of the corresponding data type to comprehensively analyze the degree of deviation of each trend data of the corresponding data type and obtain the discrete situation of the trend data of the corresponding data type.

[0139] In this embodiment, the prediction error refers to the difference between the model prediction value and the actual observation value.

[0140] In this embodiment, the degree of deviation refers to the degree of difference between a data point and the overall mean or predicted value.

[0141] The working principle and beneficial effects of the above technical solution are: by calculating the prediction error and analyzing the discreteness of trend data, the accuracy and reliability of the model output are improved, anomalies are identified, data processing is optimized, and continuous improvement and adaptability of the model are promoted.

[0142] Embodiment 7:

[0143] Based on the above embodiment 6, the storage module includes:

[0144] Screening unit: according to the preset discrete rate, the trend data with a discrete rate greater than the preset discrete rate in all data types are marked as invalid data, and the remaining trend data are regarded as initial data;

[0145] Effective processing unit: The initial data is subjected to outlier elimination according to the standard data range of the corresponding data type, and the obtained outliers are filled with corresponding characteristic values ​​to obtain valid data of the corresponding data type.

[0146] In this embodiment, invalid data refers to data that is identified as not meeting a specific standard or preset condition during data processing.

[0147] In this embodiment, the initial data refers to a data set that meets a preset discrete rate standard after being processed by the screening unit.

[0148] In this embodiment, the standard data range refers to a set of preset limits used to determine whether the data is normal or valid during data processing and analysis.

[0149] The working principle and beneficial effects of the above technical solution are: by screening the discrete rate and eliminating outliers, ensuring data quality, optimizing data processing, and using eigenvalues ​​to fill in missing data, the accuracy and reliability of the data are improved, providing a solid foundation for subsequent analysis.

[0150] Embodiment 8:

[0151] Based on the above embodiment 7, the storage module further includes:

[0152] Storage unit: Build a valid database according to the preset table structure, and store valid data of all data types in the valid database according to the composite primary key of data type-data time node.

[0153] In this embodiment, a composite primary key in the combined primary key database is composed of two or more columns (fields) and can uniquely identify each row of records in the database table.

[0154] The working principle and beneficial effects of the above technical solution are: storing valid data by combining primary keys, improving management efficiency and retrieval speed, and ensuring data integrity.

[0155] Embodiment 9:

[0156] Based on the above embodiment 1, the storage module further includes:

[0157] Update unit: If there is a new update requirement, update and maintain the data of the storage module and save the log of each operation.

[0158] In this embodiment, the update requirement refers to the necessity of modifying, adding or deleting existing data.

[0159] The working principle and beneficial effects of the above technical solution are: ensuring data timeliness and traceability through updates and log records, and improving system stability.

[0160] Embodiment 10:

[0161] Based on the above embodiment 9, the updating unit includes:

[0162] The operation log content includes: update time, update content, updater ID, and version number.

[0163] The working principle and beneficial effects of the above technical solution are: improving transparency, responsibility tracking and historical backtracking through detailed operation logs to ensure the effectiveness of data management.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic power generation effective data extraction system, characterized in that: include: Monitoring module: deploy sensors in the photovoltaic system and obtain monitoring data of photovoltaic power generation through sensors; Data processing module: After preprocessing the monitoring data of any data type, standard monitoring data is obtained, and the standard time series of the standard monitoring data and the trend data of the corresponding time nodes are obtained by using the time series analysis method; including: Data preprocessing unit: obtains raw monitoring data of a sensor of any data type, performs unit unification processing on the raw monitoring data according to the data type and a preset standard unit to obtain first data; identifies and removes abnormal values ​​of the first data according to a statistical method to obtain standard monitoring data; Time standardization unit: obtaining a time set consisting of each time node of the standard monitoring data, and performing proportional standardization on the time nodes according to a unit distance of a standard time coordinate system, wherein the abscissa of the standard time coordinate system represents the number of the time node, and the ordinate represents the time node; ;in, represents the standard time node of the i-th time node in the time set, Indicates that the time set has a total of n time nodes, represents the i-th time node in the time set, Represents the ratio of the unit distance of the standard time coordinate system to the time difference between the i-th time node and the i-1-th time node. Indicates based on expectations; Finally, the standard time nodes of the standard monitoring data and the standard time sets they constitute are obtained; Time trend unit: maps each standard time node of the standard time set of any standard monitoring data to the standard time coordinate system to obtain a first discrete set consisting of the standard time nodes; Performing curve fitting processing on the first discrete set to obtain a first curve, analyzing the first curve based on a time series analysis method to obtain a trend index of the first curve, performing parameter estimation on the trend index and a corresponding curve derivative at any standard time node to obtain trend data of the standard time node; Prediction model module: obtain historical power generation data of any data type according to the history library of the photovoltaic system, substitute the trend data of the same data type as the acquired historical power generation data into the prediction model to obtain prediction data, and compare the prediction data with the acquired historical power generation data to obtain the comparison deviation; Error analysis module: uses comparative deviation to analyze the error deviation of trend data of any data type and analyzes the discrete situation of the trend data; including: Error deviation unit: Based on the comparison deviation of any data type, the prediction error of the corresponding data type is obtained as: ; Among them, r represents the prediction error of the current data type, m represents a total of m trend data, and d represents the comparison deviation of the current data type. Indicates the mean of the predicted data of the current data type, Indicates the parameter value of the jth trend data of the current data type; Discrete unit: using the trend data variance of any data type and the prediction error of the data of the corresponding data type, comprehensively analyze the deviation degree of each trend data of the corresponding data type, and obtain the discrete situation of the trend data of the corresponding data type; Storage module: performs discrete determination on the discrete situation of the trend data, filters out data exceeding a preset discrete rate in the trend data, extracts valid data after effectively processing the remaining data in the trend data according to the data type, and stores it in a valid database.

2. A photovoltaic power generation effective data extraction system according to claim 1, characterized in that: The monitoring module comprises: Demand determination unit: determines the name of the target data type to be monitored in the photovoltaic system according to the preset valid data type; Preset unit: determine the preset data collection frequency of any data type name according to the data type name-data monitoring frequency mapping table; Sensor unit: Determine the deployment locations of sensors corresponding to different data type names according to the layout of the photovoltaic system, and configure a unique first number for any deployment location; Configuration unit: configures a unique second number for the sensor, matches the first number with the second number according to a one-to-one correspondence between a sensor of any data type name and a corresponding deployment location, and installs the corresponding sensor at the deployment location; Monitoring unit: After all sensors are configured, they are started to obtain the original monitoring data of the corresponding data type in real time.

3. A photovoltaic power generation effective data extraction system according to claim 1, characterized in that: The prediction model module comprises: Historical data unit: Use the corresponding code of any data type as the screening condition, combine SQL statements to screen the historical database to obtain the historical power generation data of the data type; Feature analysis unit: performs covariance analysis on the historical power generation data of any data type based on the feature matrix to obtain feature indicators and feature values ​​of the historical power generation data; Model building unit: taking all data types and their characteristic indicators and characteristic values ​​as the basic model framework and building the first model, taking the historical power generation data of all data types as the training set, iteratively fitting the training set based on the first model to obtain the first model indicator parameters; taking the standard monitoring data corresponding to all data types as the test set, verifying the first model indicator parameters, and optimizing and modifying the first indicator model indicator parameters to obtain the second model indicator parameters, and building the prediction model based on the second model indicator parameters; Deviation unit: predicts the trend data of any data type based on the prediction model to obtain the prediction data, compares the prediction data with the historical power generation data of the corresponding prediction data time period, and calculates the relative absolute value of the prediction data and the historical power generation data to obtain the comparison deviation of the data type.

4. A photovoltaic power generation effective data extraction system according to claim 1, characterized in that: The storage module comprises: Screening unit: according to the preset discrete rate, the trend data with a discrete rate greater than the preset discrete rate in all data types are marked as invalid data, and the remaining trend data are regarded as initial data; Effective processing unit: The initial data is subjected to outlier elimination according to the standard data range of the corresponding data type, and the obtained outliers are filled with corresponding characteristic values ​​to obtain valid data of the corresponding data type.

5. A photovoltaic power generation effective data extraction system according to claim 4, characterized in that: The storage module further includes: Storage unit: Build a valid database according to the preset table structure, and store valid data of all data types in the valid database according to the composite primary key of data type-data time node.

6. A photovoltaic power generation effective data extraction system according to claim 1, characterized in that: The storage module further includes: Update unit: If there is a new update requirement, update and maintain the data of the storage module and save the log of each operation.

7. A photovoltaic power generation effective data extraction system according to claim 6, characterized in that: The operation log content includes: update time, update content, updater ID, and version number.

Images