A new energy station terminal data collection method

Abstract

The present invention relates to the field of data acquisition technology, and specifically to a station-side acquisition method for a new energy station, comprising the following steps: acquiring sensor data of all nodes of the new energy station, and determining similar nodes based on inherent attribute data and change attribute data of the sensor data; presetting classification conditions, classifying similar nodes based on the acquisition time and change attribute data of the sensor data, and determining an acquisition analysis set and an acquisition verification set; setting a data acquisition strategy for the new energy station based on the divided acquisition analysis set and acquisition verification set; and setting an acquisition inspection strategy based on the analysis results of the data acquisition strategy; the present invention can transmit a large amount of acquired data in batches, effectively improve the transmission efficiency after data acquisition, and can also compare the data acquired in batches with each other to generate a data acquisition strategy to improve acquisition efficiency, while also performing quality inspection on the data acquisition terminal to ensure the accuracy of acquisition.

Description

Technical Field

[0001] The present invention relates to the field of data acquisition technology, and in particular to a new energy station terminal data acquisition method. Background Art

[0002] A new energy station is a collection of all equipment at the grid connection point of a wind farm or solar power station that is centrally connected to the power system, including transformers, busbars, lines, converters, energy storage, wind turbines, photovoltaic power generation equipment, reactive power regulation equipment, and auxiliary equipment. Based on the layout, equipment type, and operating characteristics of the new energy station, key points and equipment that need to be monitored are determined. For each monitoring point, the appropriate sensor type is selected to complete data collection within the new energy station.

[0003] In the existing technology, when collecting data in new energy stations, all sensor data is obtained indiscriminately without classifying the collection points. Some low-value data may interfere with the subsequent analysis of data from the same type of equipment. Summary of the Invention

[0004] The purpose of the present invention is to provide a station-side data collection method for a new energy station. The technical problem solved by the present invention is that when collecting data at a new energy station, all sensor data are obtained indiscriminately without classifying the collection points. There may be some low-value data that interferes with the subsequent analysis of data from the same type of equipment.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A new energy station terminal collection method includes the following steps:

[0007] Obtain sensor data from all nodes in the new energy station and identify similar nodes based on the inherent attribute data and change attribute data of the sensor data;

[0008] Preset classification conditions, classify similar nodes according to the collection time and change attribute data of sensor data, and determine the collection analysis set and collection verification set;

[0009] According to the divided collection analysis set and collection verification set, the data collection strategy of the new energy station is set; then, based on the analysis results of the data collection strategy, the collection inspection strategy is set.

[0010] As a further solution of the present invention: the inherent attribute data is the sensor model, serial number or target device model; the variable attribute data is the value collected by the sensor in real time.

[0011] As a further solution of the present invention: the process of determining the data collected by the same type of nodes is as follows:

[0012] Compare the sensor data of all nodes in the order of whether the inherent attribute data is the same and whether the changed attribute data is the same, and determine the data collected by the same type of nodes;

[0013] Nodes with identical intrinsic attribute data from all sensor data are extracted and defined as nodes of the same type to be determined;

[0014] Obtain the changed attribute data of all nodes of the same type to be determined, and calculate the deviation between the changed attribute data of all nodes of the same type to be determined and the average value of the changed attribute data;

[0015] If the deviation value is within the preset deviation range, the node to be determined as the same type is defined as a node of the same type.

[0016] As a further solution of the present invention: the preset classification conditions include a preset classification period, a preset classification model and a preset classification ratio.

[0017] As a further solution of the present invention: the process of determining the analysis set and the validation set is:

[0018] A classification cycle is preset, and the collection time of sensor data collected by all nodes of the same type within the classification cycle is obtained and averaged to obtain the average collection time of nodes of the same type;

[0019] Sort the nodes of the same type in the order of their mean acquisition time to obtain the time series of the nodes of the same type;

[0020] Input the change attribute data of each subset in the time series of similar nodes into the preset classification model, and output the classification features of each subset;

[0021] According to the classification characteristics and preset classification ratio of each subset, similar nodes are divided into collection analysis set and collection verification set.

[0022] As a further solution of the present invention: the construction process of the preset classification model is:

[0023] Obtaining the change attribute data of each subset within a preset classification period, and calculating the ontological stability of the change attribute data of each subset;

[0024] Obtaining the change attribute data of each subset within a preset classification period, and calculating the standard stability of the change attribute data of each subset;

[0025] The ontological stability and standard stability of the changing attribute data of each subset are weighted to obtain the classification characteristics of each subset.

[0026] As a further solution of the present invention: the ontological stability of the change attribute data of each subset is obtained by performing variance calculation on all the change attribute data of the subset within a preset classification period.

[0027] As a further solution of the present invention: the calculation process of the standard stability of the change attribute data of each subset is:

[0028] Calculate the average value of the changed attribute data of all similar nodes;

[0029] Then calculate the difference between each change attribute data in the subset and the average value of the change attribute data of all similar nodes within the preset classification period, obtain the deviation proportion of all change attribute data of the subset within the preset classification period, and then calculate the mean to obtain the standard stability of the change attribute data of each subset.

[0030] As a further solution of the present invention, the process of dividing the same type of nodes into a collection analysis set and a collection verification set is as follows:

[0031] Assign classification features to each subset of the same type of node time series, and use 70% of the subsets in the same type of node time series as the collection and analysis set to be determined;

[0032] Nodes whose classification features are less than the classification feature threshold in the collection and analysis set to be determined are extracted and divided into the collection and analysis set; all subsets of the same type of node time series except the collection and analysis set are divided into the collection verification set.

[0033] As a further solution of the present invention: the process of setting the collection and inspection strategy based on the analysis results of the data collection strategy is as follows:

[0034] Calculate the first deviation between the mean of the change attribute data of the collection and analysis set and the mean of the change attribute data of the collection and verification set in the current collection period; calculate the second deviation between the mean of the change attribute data of the collection and analysis set and the mean of the change attribute data of the collection and verification set in the preset classification period;

[0035] If the first deviation is greater than the second deviation, obtain the change attribute data of all similar nodes in the current collection period, and calculate the classification features of the change attribute data of each subset;

[0036] For nodes with classification features that exceed the preset classification period, the corresponding inherent attribute data are checked.

[0037] Beneficial effects of the present invention:

[0038] The present invention classifies all nodes of the new energy station through inherent and changing attribute data, thereby realizing property classification of all collected data in the new energy station; and then realizes functional classification of collected data of the same nature in the new energy station through the sequence of collection time and the stability of changing attribute data. The result of this classification can realize batch transmission of a large number of collected data, which can effectively improve the transmission efficiency after data collection. It can also compare the data collected in batches with each other, generate data collection strategies to improve collection efficiency, and also perform quality inspection on the data collection end to ensure the accuracy of collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 This is a flow chart of a new energy station terminal data collection method provided in the first embodiment of the present invention;

[0041] Figure 2 This is a structural diagram of a new energy station terminal data collection system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described 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 making creative efforts should fall within the scope of protection of the present invention.

[0043] Example 1

[0044] like Figure 1 As shown, a new energy station terminal data collection method provided by the first embodiment of the present invention includes the following steps:

[0045] Step 1: Obtain sensor data from all nodes in the new energy station and identify nodes of the same type based on the inherent attribute data and change attribute data of the sensor data;

[0046] In step 1: First, specifically, the method for acquiring sensor data of all nodes in the new energy station is as follows:

[0047] Based on the layout, equipment type, and operating characteristics of the new energy station, determine the key points and equipment that need to be monitored, such as the speed, power, and temperature of the wind turbine, the voltage, current, and irradiance of the photovoltaic array, and the power, temperature, and charge and discharge status of the energy storage system;

[0048] For each monitoring point, select the appropriate sensor type, such as temperature sensor, humidity sensor, current sensor, voltage sensor, wind speed sensor, irradiance sensor, etc., to ensure that the required data can be accurately collected;

[0049] A data collector is installed near each monitoring point or device to receive data transmitted by the sensor, namely sensor data.

[0050] Secondly, the interpretation of intrinsic attribute data is as follows: the parameters determined during the design, manufacturing or calibration phase of the sensor, which are used to characterize its technical specifications and operating characteristics; these data usually do not change frequently (unless the sensor is damaged or reconfigured), but are crucial for data acquisition, processing and interpretation; and the target equipment monitored by the sensor, whose properties are also fixed;

[0051] Exemplarily, the inherent attribute data is a sensor model, a serial number, or a target device model;

[0052] The explanation of changing attribute data is: parameters dynamically collected by sensors during real-time monitoring and updated in real time as the measurement environment or object status changes;

[0053] Exemplarily, the changing attribute data is a value collected by a sensor in real time, such as a real-time temperature value;

[0054] Thirdly, specifically, the process of determining the data collected by the same type of nodes is as follows:

[0055] Compare the sensor data of all nodes in the order of whether the inherent attribute data is the same and whether the changed attribute data is the same, and determine the data collected by the same type of nodes;

[0056] Nodes with identical intrinsic attribute data from all sensor data are extracted and defined as nodes of the same type to be determined;

[0057] Obtain the changed attribute data of all nodes of the same type to be determined, and calculate the deviation between the changed attribute data of all nodes of the same type to be determined and the average value of the changed attribute data;

[0058] If the deviation value is within the preset deviation range, the nodes to be determined as the same type are defined as the same type nodes;

[0059] If the deviation value is not within the preset deviation range, the node to be determined as the same type is defined as a different type node;

[0060] For example, nodes with the same sensor model, serial number, and target device model in all temperature sensing data are extracted and defined as nodes of the same type to be determined;

[0061] Obtain the real-time temperature values of all nodes of the same type to be determined, and calculate the deviation between the real-time temperature values of all nodes of the same type to be determined and the average real-time temperature value;

[0062] If the deviation value is within the preset deviation range, the nodes to be determined as the same type are defined as the same type nodes;

[0063] If the deviation value is not within the preset deviation range, the node to be determined as the same type is defined as a different type node;

[0064] Step 2: Preset classification conditions, classify similar nodes according to the collection time and change attribute data of the sensor data, and determine the collection analysis set and collection verification set;

[0065] The preset classification conditions include a preset classification period, a preset classification model, and a preset classification ratio;

[0066] In step 2, specifically, the process of classifying the data groups collected by similar nodes and determining the analysis set and verification set is as follows:

[0067] A classification cycle is preset, and the collection time of sensor data collected by all nodes of the same type within the classification cycle is obtained and averaged to obtain the average collection time of nodes of the same type;

[0068] Sort the nodes of the same type in the order of their mean acquisition time to obtain the time series of the nodes of the same type;

[0069] Input the change attribute data of each subset in the time series of similar nodes into the preset classification model, and output the classification features of each subset;

[0070] According to the classification characteristics and preset classification ratio of each subset, the nodes of the same type are divided into the collection and analysis set and the collection and verification set;

[0071] It should be explained that the preset classification period and the preset classification ratio can be preset by those skilled in the art according to actual needs. Preferably, the preset classification period can be set to 0.5-1s, and the preset classification ratio can be set to 60-80%. For example, when the preset classification ratio is 70%, the nodes of the same type accounting for 70% are used as the collection and analysis set, and the remaining nodes of the same type accounting for 30% are used as the collection and verification set.

[0072] In more detail, the construction process of the preset classification model is:

[0073] Obtaining the change attribute data of each subset within a preset classification period, and calculating the ontological stability of the change attribute data of each subset;

[0074] and, obtaining the change attribute data of each subset within a preset classification period, and calculating the standard stability of the change attribute data of each subset;

[0075] The weighted influence coefficient of the ontological stability of the changing attribute data of each subset and the standard stability of the changing attribute data of each subset are calculated to obtain the classification characteristics of each subset. It should be explained that the weighted influence coefficient of the ontological stability of the changing attribute data of each subset is 0.689, and the weighted influence coefficient of the standard stability of the changing attribute data of each subset is 0.311.

[0076] For example, the calculation process of the ontology stability of the change attribute data of each subset is as follows:

[0077] Obtain the change attribute data of any subset within a preset classification period, perform variance calculation on all the change attribute data of the subset within the preset classification period, and obtain the ontological stability of the change attribute data of each subset;

[0078] The calculation process of the standard stability of the changing attribute data of each subset is:

[0079] Obtain the change attribute data of all subsets within the preset classification period and perform mean processing to calculate the average value of the change attribute data of all nodes of the same type;

[0080] Then calculate the difference between each change attribute data in the subset and the average value of the change attribute data of all similar nodes within the preset classification period to obtain the deviation ratio of the change attribute data of any subset within the preset classification period. Calculate the average of all the change attribute data deviation ratios of the subset within the preset classification period to obtain the standard stability of the change attribute data of each subset.

[0081] More specifically, the process of dividing similar nodes into the collection analysis set and the collection verification set according to the classification characteristics and preset classification ratio of each subset is as follows:

[0082] Assign classification features to each subset of the same type of node time series, and use 70% of the subsets in the same type of node time series as the collection and analysis set to be determined;

[0083] Extract the nodes whose classification features are less than the classification feature threshold in the collection and analysis set to be determined, and divide them into the collection and analysis set; divide all subsets of the same node time series except the collection and analysis set into the collection verification set;

[0084] The classification feature threshold is defined as the classification feature of the subset corresponding to the ranking of the preset classification ratio according to the classification features of each subset in the time series of the same type of nodes arranged from small to large.

[0085] Step 3: Set the data collection strategy for the new energy station based on the divided collection analysis set and collection verification set; then set the collection inspection strategy based on the analysis results of the data collection strategy;

[0086] In step 3, first, specifically, the process of setting the data collection strategy of the new energy station includes:

[0087] During the current collection period, the sensor data of all nodes in the collection and analysis set is obtained and sent to the terminal device first, and the sensor data in the collection and analysis set is intelligently analyzed. The intelligent analysis includes monitoring, fault analysis, and other analysis methods.

[0088] Obtain sensor data from all nodes in the collection and verification set, send it to the terminal device first, and perform verification analysis with the sensor data in the collection and analysis set;

[0089] The verification analysis process is as follows:

[0090] Obtain the change attribute data in the sensor data of all nodes in the collection and analysis set, and calculate the mean value of the change attribute data of the collection and analysis set;

[0091] Obtain the change attribute data in the sensor data of all nodes in the collection and verification set, and calculate the mean of the change attribute data of the collection and verification set;

[0092] Calculate the first deviation between the mean of the change attribute data of the collection and analysis set and the mean of the change attribute data of the collection and verification set in the current collection period; calculate the second deviation between the mean of the change attribute data of the collection and analysis set and the mean of the change attribute data of the collection and verification set in the preset classification period;

[0093] If the first deviation is less than or equal to the second deviation, execute the collection plan of dividing the same type of nodes into a collection analysis set and a collection verification set;

[0094] Secondly, specifically, based on the analysis results of the data collection strategy, the process of setting the collection and inspection strategy is as follows:

[0095] If the first deviation is greater than the second deviation, obtain the change attribute data of all similar nodes in the current collection period, and calculate the classification features of the change attribute data of each subset;

[0096] Nodes whose classification characteristics exceed the preset classification period are marked as abnormal nodes, and the inherent attribute data corresponding to the abnormal nodes are checked;

[0097] The technical solution of the embodiment of the present invention is as follows: the sensor data of all nodes of the new energy station are obtained, and the nodes of the same type are determined according to the inherent attribute data and the changing attribute data of the sensor data; the classification conditions are preset, and the nodes of the same type are classified according to the collection time and the changing attribute data of the sensor data to determine the collection analysis set and the collection verification set; the data collection strategy of the new energy station is set according to the divided collection analysis set and the collection verification set; and the collection inspection strategy is set according to the analysis result of the data collection strategy; the present invention classifies all nodes of the new energy station according to the inherent and changing attribute data to realize the property classification of all collected data in the new energy station; and then realizes the functional classification of the collected data of the same nature in the new energy station again according to the sequence of collection time and the stability of the changing attribute data. The result of this classification can realize the transmission of a large number of collected data in batches, which can effectively improve the transmission efficiency after data collection, and can also compare the data collected in batches with each other to generate a data collection strategy to improve the collection efficiency. At the same time, the data collection end is also quality inspected to ensure the accuracy of the collection.

[0098] Example 2

[0099] like Figure 2 As shown, a new energy station terminal data collection system provided by the first embodiment of the present invention includes the following modules:

[0100] Property classification module: obtains sensor data of all nodes in the new energy station and identifies nodes of the same type based on the inherent attribute data and change attribute data of the sensor data;

[0101] In a specific embodiment, intrinsic attribute data is interpreted as: parameters determined during the design, manufacturing, or calibration phase of a sensor, which are used to characterize its technical specifications and operating characteristics; these data generally do not change frequently (unless the sensor is damaged or reconfigured), but are crucial for data collection, processing, and interpretation; and the properties of the target device monitored by the sensor are also fixed;

[0102] The explanation of changing attribute data is: parameters dynamically collected by sensors during real-time monitoring and updated in real time as the measurement environment or object status changes;

[0103] The process of determining the data collected by the same type of nodes is as follows:

[0104] Compare the sensor data of all nodes in the order of whether the inherent attribute data is the same and whether the changed attribute data is the same, and determine the data collected by the same type of nodes;

[0105] Nodes with identical intrinsic attribute data from all sensor data are extracted and defined as nodes of the same type to be determined;

[0106] Obtain the changed attribute data of all nodes of the same type to be determined, and calculate the deviation between the changed attribute data of all nodes of the same type to be determined and the average value of the changed attribute data;

[0107] If the deviation value is within the preset deviation range, the nodes to be determined as the same type are defined as the same type nodes;

[0108] If the deviation value is not within the preset deviation range, the node to be determined as the same type is defined as a different type node;

[0109] Functional classification module: preset classification conditions, classify similar nodes according to the collection time and change attribute data of sensor data, and determine the collection analysis set and collection verification set;

[0110] The preset classification conditions include a preset classification period, a preset classification model, and a preset classification ratio;

[0111] In a specific embodiment, the process of classifying the data groups collected by the same type of nodes and determining the analysis set and the verification set is as follows:

[0112] A classification cycle is preset, and the collection time of sensor data collected by all nodes of the same type within the classification cycle is obtained and averaged to obtain the average collection time of nodes of the same type;

[0113] Sort the nodes of the same type in the order of their mean acquisition time to obtain the time series of the nodes of the same type;

[0114] Input the change attribute data of each subset in the time series of similar nodes into the preset classification model, and output the classification features of each subset;

[0115] According to the classification characteristics and preset classification ratio of each subset, the nodes of the same type are divided into the collection and analysis set and the collection and verification set;

[0116] In more detail, the construction process of the preset classification model is:

[0117] Obtaining the change attribute data of each subset within a preset classification period, and calculating the ontological stability of the change attribute data of each subset;

[0118] and, obtaining the change attribute data of each subset within a preset classification period, and calculating the standard stability of the change attribute data of each subset;

[0119] The ontological stability of the changing attribute data of each subset is weighted with the standard stability of the changing attribute data of each subset to obtain the classification characteristics of each subset;

[0120] Extract the nodes whose classification features are less than the classification feature threshold in the collection and analysis set to be determined, and divide them into the collection and analysis set; divide all subsets of the same node time series except the collection and analysis set into the collection verification set;

[0121] The classification feature threshold is defined as the classification feature of the subset corresponding to the ranking of the preset classification ratio according to the classification features of each subset in the time series of the same type of nodes arranged from small to large.

[0122] Strategy generation module: Set the data collection strategy for the new energy station based on the divided collection analysis set and collection verification set; then set the collection inspection strategy based on the analysis results of the data collection strategy;

[0123] In a specific embodiment: First, specifically, the process of setting the data collection strategy of the new energy station includes:

[0124] During the current collection period, the sensor data of all nodes in the collection and analysis set is obtained and sent to the terminal device first, and the sensor data in the collection and analysis set is intelligently analyzed. The intelligent analysis includes monitoring, fault analysis, and other analysis methods.

[0125] Obtain sensor data from all nodes in the collection and verification set, send it to the terminal device first, and perform verification analysis with the sensor data in the collection and analysis set;

[0126] The verification analysis process is as follows:

[0127] Obtain the change attribute data in the sensor data of all nodes in the collection and analysis set, and calculate the mean value of the change attribute data of the collection and analysis set;

[0128] Obtain the change attribute data in the sensor data of all nodes in the collection and verification set, and calculate the mean of the change attribute data of the collection and verification set;

[0129] Calculate the first deviation between the mean of the change attribute data of the collection and analysis set and the mean of the change attribute data of the collection and verification set in the current collection period; calculate the second deviation between the mean of the change attribute data of the collection and analysis set and the mean of the change attribute data of the collection and verification set in the preset classification period;

[0130] If the first deviation is less than or equal to the second deviation, execute the collection plan of dividing the same type of nodes into a collection analysis set and a collection verification set;

[0131] Secondly, specifically, based on the analysis results of the data collection strategy, the process of setting the collection and inspection strategy is as follows:

[0132] If the first deviation is greater than the second deviation, obtain the change attribute data of all similar nodes in the current collection period, and calculate the classification features of the change attribute data of each subset;

[0133] Nodes whose classification characteristics exceed the preset classification period are marked as abnormal nodes, and the inherent attribute data corresponding to the abnormal nodes are checked.

[0134] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope 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. A new energy station terminal collection method, characterized in that: The following steps are involved: Obtain sensor data from all nodes in the new energy station and identify similar nodes based on the inherent attribute data and change attribute data of the sensor data; Preset classification conditions, classify similar nodes according to the collection time and change attribute data of sensor data, and determine the collection analysis set and collection verification set; According to the divided collection and analysis set and collection and verification set, set the data collection strategy of the new energy station; Then, based on the analysis results of the data collection strategy, set the collection and inspection strategy; Obtaining the change attribute data in the sensor data of all nodes in the collection and analysis set or the collection and verification set, and calculating the mean of the change attribute data of the collection and analysis set or the collection and verification set; Calculate the first deviation or second deviation between the mean of the change attribute data of the collection analysis set and the mean of the change attribute data of the collection verification set in the current collection period or the preset classification period; if the first deviation is less than or equal to the second deviation, execute the collection plan of dividing the nodes of the same type into the collection analysis set and the collection verification set; If the first deviation is greater than the second deviation, obtain the change attribute data of all similar nodes in the current collection period, calculate the classification features of the change attribute data of each subset, and mark the nodes whose classification features exceed the preset classification period as abnormal nodes; The preset classification conditions include a preset classification period, a preset classification model and a preset classification ratio; The process of determining the analysis set and validation set is as follows: A classification cycle is preset, and the collection time of sensor data collected by all nodes of the same type within the classification cycle is obtained and averaged to obtain the average collection time of nodes of the same type; Sort the nodes of the same type in the order of their mean acquisition time to obtain the time series of the nodes of the same type; Input the change attribute data of each subset in the time series of similar nodes into the preset classification model, and output the classification features of each subset; According to the classification characteristics and preset classification ratio of each subset, similar nodes are divided into collection analysis set and collection verification set.

2. A new energy station terminal collection method according to claim 1, characterized in that: The inherent attribute data is the sensor model, serial number or target device model; the variable attribute data is the value collected by the sensor in real time.

3. A new energy station terminal collection method according to claim 1, characterized in that: The process of determining the data collected by the same type of nodes is as follows: Compare the sensor data of all nodes in the order of whether the inherent attribute data is the same and whether the changed attribute data is the same, and determine the data collected by the same type of nodes; Nodes with identical intrinsic attribute data from all sensor data are extracted and defined as nodes of the same type to be determined; Obtain the changed attribute data of all nodes of the same type to be determined, and calculate the deviation between the changed attribute data of all nodes of the same type to be determined and the average value of the changed attribute data; If the deviation value is within the preset deviation range, the node to be determined as the same type is defined as a node of the same type.

4. A new energy station terminal collection method according to claim 1, characterized in that: The construction process of the preset classification model is: Obtaining the change attribute data of each subset within a preset classification period, and calculating the ontological stability of the change attribute data of each subset; Obtaining the change attribute data of each subset within a preset classification period, and calculating the standard stability of the change attribute data of each subset; The ontological stability and standard stability of the changing attribute data of each subset are weighted to obtain the classification characteristics of each subset.

5. A new energy station terminal collection method according to claim 4, characterized in that: The ontological stability of the changing attribute data of each subset is obtained by calculating the variance of all the changing attribute data of the subset within a preset classification period.

6. A new energy station terminal collection method according to claim 5, characterized in that: The calculation process of the standard stability of the changing attribute data of each subset is: Calculate the average value of the changed attribute data of all similar nodes; Then calculate the difference between each change attribute data in the subset and the average value of the change attribute data of all similar nodes within the preset classification period, obtain the deviation proportion of all change attribute data of the subset within the preset classification period, and then calculate the mean to obtain the standard stability of the change attribute data of each subset.

7. A new energy station terminal data collection method according to claim 6, characterized in that: The process of dividing the same type of nodes into the collection analysis set and the collection verification set is as follows: Assign classification features to each subset of the same type of node time series, and use 70% of the subsets in the same type of node time series as the collection and analysis set to be determined; Extract nodes whose classification features are less than the classification feature threshold in the collection and analysis set to be determined, and divide them into the collection and analysis set; All subsets of the same type of node time series except the collection and analysis set are divided into the collection and verification set.

Images