Method and system for intelligently judging defects of power equipment based on data driving

Abstract

The invention discloses an intelligent electrical equipment defect discrimination method and system based on data driving, and relates to the technical field of electrical equipment defect discrimination, and the method comprises the steps: obtaining a historical defect record of electrical equipment from a power grid platform, analyzing state parameters in defect discrimination data, and carrying out the abnormal degree of defect discrimination of the electrical equipment; analyzing the parameter influence degree of operation parameters in the equipment operation record on abnormal state parameters in the parameter abnormal data; predicting the change trend of the state parameters of the power equipment in the current period, obtaining parameter prediction data, and carrying out equipment defect judgment on the power equipment in combination with defect judgment data; the method comprises the following steps: acquiring characteristic defect root cause data of defective power equipment, acquiring parameter prediction data, analyzing defect root causes of the defective power equipment, sending the defect root cause data of the defective power equipment to a worker through a power grid platform, and prompting the worker to maintain the defective power equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment defect discrimination, and specifically to an intelligent discrimination method and system for power equipment defects based on data driving. Background Art

[0002] In the power grid, power equipment ensures the safe and coordinated operation of the power generation, transformation, transmission, and distribution links of the power grid, thereby realizing the safe and stable power supply to users. Therefore, judging and identifying the defects of power equipment is an essential part of ensuring the normal and efficient operation of the power grid. The traditional method for discriminating the defects of power equipment mainly involves professional personnel judging the equipment status of power equipment through on-site inspection and data collected by sensors, and finding out the power equipment with defects. However, this method not only requires a large amount of manual analysis, but also may lead to errors in the discrimination of power equipment defects due to the lack of professional skills of professional personnel. Moreover, there is currently no good method for analyzing the root causes of power equipment defects, making it impossible to detect and repair the equipment defects of power equipment in a timely manner, which has a great negative impact on the normal use of the power grid. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent discrimination method and system for power equipment defects based on data driving to solve the problems raised in the prior art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: An intelligent discrimination method for power equipment defects based on data driving, the method comprising: Step S100: Obtain the historical defect records of power equipment from the power grid platform, obtain the defect discrimination data of power equipment, analyze the abnormal degree of the state parameters in the defect discrimination data for defect discrimination of power equipment, and obtain parameter abnormal data; Step S200: Obtain the historical equipment operation records of power equipment, analyze the operation parameters in the historical equipment operation records, and obtain parameter influence data on the influence degree of the abnormal state parameters in the parameter abnormal data; Step S300: Obtain the parameter influence data, monitor the power equipment in the current period, predict the change trend of the state parameters of the power equipment in the current period, obtain parameter prediction data, and combine the defect discrimination data to perform equipment defect discrimination on the power equipment to obtain defective power equipment; Step S400: Obtain the characteristic defect root cause data of the defective power equipment, obtain the parameter prediction data, analyze the defect root cause of the defective power equipment to obtain defect root cause data, and send the defect root cause data of the defective power equipment to the staff through the power grid platform and prompt the staff to repair the defective power equipment.

[0005] Further, step S100 includes: Step S101: Obtain each historical defect record of the power equipment, and obtain the data of each status parameter of the power equipment from the historical defect records, where equipment defects have occurred in all the power equipment in the historical defect records; Step S1O2: Obtain the defect discrimination data of the power equipment from the power grid platform, where the defect discrimination data is the range of each status parameter when performing defect discrimination on the power equipment; When the maximum value and the minimum value of each status parameter in a certain historical defect record are both within the range of the defect discrimination data, it is determined that the defect discrimination data is incorrect for the defect discrimination of a certain historical defect record, mark a certain historical defect record, and record it as an abnormal historical defect record; Obtain the total number C of several unmarked historical defect records in the power equipment sum , obtain several status parameters whose maximum value or minimum value in the unmarked historical defect records is outside the range of the defect discrimination data, and gather them to obtain the characteristic parameter set of the unmarked historical defect records; Step S103: Calculate the defect discrimination ratio of each status parameter, where the defect discrimination ratio L of the a-th status parameter in the power equipment a =C (a,sum) / C sum , C (a,sum) is the total number of unmarked historical defect records containing the a-th status parameter in the characteristic parameter set; Calculate the average value L´ and the standard deviation σ´ of the defect discrimination ratios of each status parameter, and obtain the lower limit ratio threshold B of each status parameter min =L´ - k×σ´, where k is a preset coefficient. When the defect discrimination ratio of a certain status parameter is less than the lower limit ratio threshold, mark a certain status parameter and record it as a suspected abnormal status parameter; Step S104: Obtain the median D of the suspected abnormal status parameter in the corresponding range in the defect discrimination data △ , and calculate the absolute values of the differences between the maximum value and the minimum value of the suspected abnormal status parameter in the abnormal historical defect records of the power equipment and the median D △ respectively, and take the maximum value of the absolute values, and divide the maximum value by the median D △ , to obtain the parameter abnormal value e of the suspected abnormal status parameter in the abnormal historical defect records; When the parameter outlier e is greater than the preset threshold e´, the abnormal historical defect record is recorded as the target abnormal historical defect record of the suspected abnormal state parameter, and the ratio of the total number of the target abnormal historical defect records of the suspected abnormal state parameter to the total number of each abnormal historical defect record in the power equipment is obtained to get the abnormal ratio of the suspected abnormal state parameter; When the abnormal ratio is greater than the preset threshold, it is determined that there is an abnormality when the suspected abnormal state parameter in the defect discrimination data identifies defects in the power equipment, and the suspected abnormal state parameter is recorded as the abnormal state parameter; Obtain several abnormal state parameters of the power equipment and gather them to obtain the parameter abnormal data of the power equipment.

[0006] Further, step S200 includes: Step S201: Obtain each historical equipment operation record of the power equipment, preprocess the data in each historical equipment operation record, set the unit time length, and obtain the average value of each operation parameter in the historical equipment operation record every unit time length to obtain the parameter set of each operation parameter; Step S202: Obtain the parameter abnormal data of the power equipment, and obtain each abnormal state parameter from the parameter abnormal data; Obtain the average value of each abnormal state parameter from the historical equipment operation record every unit time length to obtain the parameter set of each abnormal state parameter; Calculate the parameter influence degree between each operation parameter and each abnormal state parameter in the historical equipment operation record. Among them, the parameter influence degree S between the f-th operation parameter and the g-th abnormal state parameter in the historical equipment operation record (f,g) ; Step S203: Calculate the data influence value R of the f-th operation parameter on the g-th abnormal state parameter (f,g) : , where z is the total number of each historical equipment operation record of the power equipment; S z (f,g) is the parameter influence degree between the f-th operation parameter and the g-th abnormal state parameter in the z-th historical equipment operation record of the power equipment; Step S204: Set the influence threshold R. When |R (f,g) |≥R, it is determined that the data of the f-th operation parameter has an influence on the g-th abnormal state parameter, and the f-th operation parameter is recorded as the influencing operation parameter of the power equipment. Obtain each influencing operation parameter of the power equipment and gather them to obtain the parameter influence data.

[0007] Further, step S300 includes: Step S301: Obtain parameter impact data. Obtain various operating parameters affecting the power equipment, obtain each historical operation record of the power equipment, preprocess the data in each historical operation record, and sort each historical operation record in chronological order; Construct a data prediction model for the power equipment. Divide each historical operation record into a training set and a test set according to a preset ratio; Obtain the input data and output data of the model data set in the model. The input data is the various state parameters and various operating parameters affecting the operation in a certain historical operation record, and the output data is the various state parameters in the next historical operation record of a certain historical operation record; Step S302: Use the training set to train the data prediction model, and use the test set to calculate the model prediction accuracy of the data prediction model. When the model prediction accuracy is greater than the preset threshold, it is determined that the data prediction model is constructed; Step S303: Monitor the power equipment in the current cycle. Every unit time period, collect the values of the various state parameters and various operating parameters affecting the operation of the power equipment, and input them into the data prediction model to predict the various state parameters of the power equipment to obtain parameter prediction data. The parameter prediction data includes the predicted parameter set of the various state parameters; When the maximum or minimum value of the state parameter in the parameter prediction data is not within the range of the state parameter of the defect discrimination data, it is determined that there is a risk of equipment defect in the power equipment in the current cycle, and the state parameter and the power equipment are respectively recorded as the defect state parameter and the defective power equipment.

[0008] Furthermore, step S400 includes: Step S401: Obtain the defective power equipment in the power grid. Obtain the characteristic defect root cause data of the defective power equipment from the power grid platform, obtain the root cause discrimination data of the various defect root causes of the defective power equipment from the characteristic defect root cause data, and obtain the state parameters to be detected when discriminating a certain defect root cause from the characteristic defect root cause data, and record them as the target state parameters of a certain defect root cause. Among them, the root cause discrimination data of a certain defect root cause is the data range where the various target state parameters of a certain defect root cause are located when the defective power equipment has a certain defect root cause; Step S402: Obtain the predicted data set of the various defect state parameters of the defective power equipment. When a certain defect state parameter is the target state parameter of a certain defect root cause, and the value of a certain element in the predicted data set of a certain defect state parameter is within the data range of the root cause discrimination data, it is recorded as a primary parameter compliance of a certain defect root cause, and obtain the total number of times M' of the parameter compliance of a certain defect root cause sum , obtain the total number of items M of the various target state parameters of a certain defect root cause sum; Calculate the occurrence probability P of a certain defect root cause as P = M´ sum / M sum , when the occurrence probability P is greater than the preset probability threshold, it is determined that there is a certain defect root cause in the defective power equipment in the current cycle, and a certain defect root cause is recorded as the target defect root cause; Step S403: Obtain and collect each target defect root cause of the defective power equipment in the current cycle to obtain defect root cause data, send the defect root cause data to the grid staff through the grid platform, and prompt the staff to dispatch personnel to repair the defective power equipment; The purpose of obtaining the target defect root cause of the defective power equipment in the above steps is to better repair the power equipment. Because even if it is predicted that the power equipment has a defect, but the cause of the defect of the power equipment is not clear, that is, the above target defect root cause. Due to the wide distribution of power equipment in the grid, different defect root causes require different maintenance personnel, maintenance tools and maintenance equipment, all of which will make it difficult to repair the power equipment. Therefore, obtaining the target defect root cause of the power equipment can not only save manpower and material resources, but also greatly improve the maintenance efficiency and reduce the maintenance time, effectively ensuring the normal operation of the grid.

[0009] In order to better implement the above method, a data-driven intelligent discrimination system for power equipment defects is also proposed. The system includes an abnormal analysis module, a parameter influence analysis module, an equipment defect discrimination module and a defect root cause analysis module; The abnormal analysis module is used to obtain the historical defect records of power equipment, obtain the defect discrimination data of power equipment, analyze the abnormal degree of the state parameters in the defect discrimination data for defect discrimination of power equipment, and obtain parameter abnormal data; The parameter influence analysis module is used to obtain the historical equipment operation records of power equipment, analyze the operation parameters in the equipment operation records, and obtain parameter influence data on the influence degree of the abnormal state parameters in the parameter abnormal data; The equipment defect discrimination module is used to monitor the power equipment in the current cycle, predict the change trend of the state parameters of the power equipment in the current cycle, obtain parameter prediction data, and combine the defect discrimination data to conduct equipment defect discrimination on the power equipment to obtain defective power equipment; The defect root cause analysis module is used to analyze the defect root cause of the defective power equipment according to the characteristic defect root cause data, obtain the defect root cause data, send the defect root cause data of the defective power equipment to the staff through the grid platform, and prompt the staff to repair the defective power equipment.

[0010] Further, the anomaly analysis module includes a record anomaly analysis unit and an anomaly analysis unit; The record anomaly analysis unit is used to obtain each historical defect record of the power equipment, and combine the defect discrimination data of the power equipment to analyze the record anomaly of each historical defect record to obtain an abnormal historical defect record; The anomaly analysis unit is used to analyze the abnormal state of defect discrimination of the power equipment for each status parameter in the defect discrimination data according to the abnormal historical defect record to obtain parameter anomaly data.

[0011] Further, the parameter impact analysis module includes a parameter set acquisition unit and a parameter impact analysis unit; The parameter set acquisition unit is used to obtain the historical equipment operation records of the power equipment and obtain the parameter set of each operation parameter from the historical equipment operation records; The parameter impact analysis unit is used to obtain the parameter set of the abnormal state parameters in the historical equipment operation records, and analyze the parameter impact degree of the operation parameters on the abnormal state parameters according to the parameter set of each operation parameter in the historical equipment operation records of the power equipment to obtain parameter impact data.

[0012] Further, the equipment defect discrimination module includes a model construction unit and an equipment defect discrimination unit; The model construction unit is used to construct a data prediction model of the power equipment according to each historical operation record of the power equipment; The equipment defect discrimination unit is used to use the data prediction model to predict each status parameter of the power equipment in the current cycle, and combine the defect discrimination data of the power equipment to perform equipment defect discrimination on the power equipment to obtain defective power equipment.

[0013] Further, the defect root cause analysis module includes a defect root cause analysis unit; The defect root cause analysis unit is used to obtain the characteristic defect root cause data of the defective power equipment, and combine the parameter prediction data to analyze the defect root cause of the defective power equipment in the current cycle and send it to the staff of the power grid platform to prompt the staff to perform equipment maintenance on the defective power equipment.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention realizes the intelligent discrimination of power equipment defects. By obtaining historical defect records with inaccurate discrimination of power equipment defects and combining defect discrimination data for defect discrimination of power equipment, the state parameters with inaccurate discrimination of power equipment are found, that is, abnormal state parameters, and the operating parameters that cause the abnormal state parameters to inaccurately discriminate power equipment defects are analyzed, so as to optimize the construction of the prediction model of state parameters, improve the accuracy of model prediction and power equipment defect discrimination, and also analyze the root causes of power equipment defects, obtain the defect root causes, and send the defect root causes to the staff, greatly improving the efficiency of power equipment maintenance and enabling the power grid to operate safely and stably. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a method flow chart of a data-driven intelligent discrimination method for power equipment defects of the present invention; Figure 2 is a module schematic diagram of a data-driven intelligent discrimination system for power equipment defects of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

[0017] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a data-driven intelligent discrimination method for power equipment defects, and the method includes: Step S100: Obtain historical defect records of power equipment from the power grid platform, obtain defect discrimination data of the power equipment, and analyze the abnormal degree of state parameters in the defect discrimination data for defect discrimination of the power equipment to obtain parameter abnormal data; Among them, step S100 includes: Step S101: Obtain each historical defect record of the power equipment, and obtain the data of each state parameter of the power equipment from the historical defect record, where equipment defects have occurred in all power equipment in the historical defect record; For example, each state parameter includes voltage, current, harmonic content, etc.; Step S1O2: Obtain defect discrimination data of the power equipment from the power grid platform, and the defect discrimination data is the range of each state parameter when defect discrimination of the power equipment is performed; If the maximum and minimum values of all status parameters in a certain historical defect record are within the range of the defect discrimination data, it is determined that the defect discrimination data is incorrect for the defect discrimination of a certain historical defect record, and a certain historical defect record is marked and recorded as an abnormal historical defect record; Obtain the total number C of several unmarked historical defect records in the power equipment sum , obtain several status parameters whose maximum or minimum value in the unmarked historical defect records is outside the range of the defect discrimination data, and collect them to obtain the characteristic parameter set of the unmarked historical defect records; Step S103: Calculate the defect discrimination ratio of each status parameter. Among them, the defect discrimination ratio L of the a-th status parameter in the power equipment a =C (a,sum) / C sum , C (a,sum) is the total number of unmarked historical defect records containing the a-th status parameter in the characteristic parameter set; Calculate the average value L´ and standard deviation σ´ of the defect discrimination ratios of each status parameter, and obtain the lower limit ratio threshold B of each status parameter min =L´ - k×σ´, where k is a preset coefficient. When the defect discrimination ratio of a certain status parameter is less than the lower limit ratio threshold, mark a certain status parameter and record it as a suspected abnormal status parameter; Step S104: Obtain the median D of the suspected abnormal status parameters in the defect discrimination data within the corresponding range △ , and calculate the absolute values of the differences between the maximum and minimum values of the suspected abnormal status parameters in the abnormal historical defect records of the power equipment and the median D △ respectively, and take the maximum value of the absolute values, and divide the maximum value by the median D △ to obtain the parameter abnormal value e of the suspected abnormal status parameters in the abnormal historical defect records; When the parameter abnormal value e is greater than the preset threshold e´, record the abnormal historical defect record as the target abnormal historical defect record of the suspected abnormal status parameter, and obtain the ratio of the total number of the target abnormal historical defect records of the suspected abnormal status parameter to the total number of each abnormal historical defect record in the power equipment to obtain the abnormal ratio of the suspected abnormal status parameter; When the abnormal ratio is greater than the preset threshold, it is determined that there is an abnormality when the suspected abnormal status parameter in the defect discrimination data is used for defect identification of the power equipment, and the suspected abnormal status parameter is recorded as an abnormal status parameter; Obtain several abnormal status parameters of the power equipment and collect them to obtain the parameter abnormal data of the power equipment; Step S200: Obtain the historical device operation records of the power equipment, analyze the operation parameters in the historical device operation records, and obtain the parameter influence data for the parameter influence degree of the abnormal state parameters in the parameter abnormal data; Among them, step S200 includes: Step S201: Obtain each historical device operation record of the power equipment, preprocess the data in each historical device operation record, set the unit time period, and obtain the average value of each operation parameter in the historical device operation record every unit time period to obtain the parameter set of each operation parameter; For example, each operation parameter includes temperature, noise, humidity, etc.; For example, the preprocessing of each historical device operation record includes cleaning the data, using interpolation (linear / spline) to process the missing values, and finally performing normalization or standardization processing; Step S202: Obtain the parameter abnormal data of the power equipment, and obtain each abnormal state parameter from the parameter abnormal data; Obtain the average value of each abnormal state parameter from the historical device operation record every unit time period to obtain the parameter set of each abnormal state parameter; Calculate the parameter influence degree between each operation parameter and each abnormal state parameter in the historical device operation record. Among them, the parameter influence degree S between the f-th operation parameter and the g-th abnormal state parameter in the historical device operation record (f,g) ; For example, the parameter influence degree S (f,g) The specific calculation formula is: , Among them, x （f,i） is the value of the i-th element in the parameter set of the f-th operation parameter in the historical device operation record; y （g,i） is the value of the i-th element in the parameter set of the g-th abnormal state parameter in the historical device operation record; x´ f is the average value of each element in the parameter set of the f-th operation parameter in the historical device operation record; y´ g is the average value of each element in the parameter set of the g-th abnormal state parameter in the historical device operation record; Step S203: Calculate the data influence value R of the f-th operation parameter on the g-th abnormal state parameter (f,g) : , Among them, z is the total number of each historical device operation record of the power equipment; S z (f,g)is the parameter influence degree between the f-th operation parameter and the g-th abnormal state parameter in the z-th historical device operation record of the power equipment; Step S204: Set the influence threshold R. When |R (f,g) |≥R, it is determined that the f-th operation parameter has an influence on the data of the g-th abnormal state parameter. The f-th operation parameter is recorded as the influencing operation parameter of the power equipment, and each influencing operation parameter of the power equipment is collected to obtain parameter influence data; Step S300: Obtain parameter influence data, monitor the power equipment in the current period, predict the change trend of the state parameters of the power equipment in the current period to obtain parameter prediction data, and combine the defect discrimination data to perform equipment defect discrimination on the power equipment to obtain defective power equipment; Among them, Step S300 includes: Step S301: Obtain parameter influence data, obtain each influencing operation parameter of the power equipment, obtain each historical operation record of the power equipment, preprocess the data in each historical operation record, and sort each historical operation record in chronological order; Construct a data prediction model for the power equipment, and divide each historical operation record into a training set and a test set according to a preset ratio; For example, the data prediction model constructed for the power equipment can be a linear model: linear regression model, etc.; time series model: such as ARIMA, exponential smoothing model; machine learning model: such as decision tree, random forest, support vector machine model, etc.; deep learning model: LSTM model, etc.; Obtain the input data and output data of the model data set in the model. The input data is each state parameter and each influencing operation parameter in a certain historical operation record, and the output data is each state parameter in the next historical operation record of a certain historical operation record; For example, in the training set, the input data is the parameter set of each state parameter and each influencing operation parameter in a certain historical operation record, and the output data is the parameter set of each state parameter in the next historical operation record of a certain historical operation record; Step S302: Use the training set to train the data prediction model, and use the test set to calculate the model prediction accuracy of the data prediction model. When the model prediction accuracy is greater than the preset threshold, it is determined that the data prediction model is constructed; For example, when the data prediction model is a linear regression model, the model prediction accuracy can be the mean square error, root mean square error, determination coefficient R 2 ; Among them, when the model prediction accuracy is the mean square error MSE, the specific formula is: , where n is the total number of model datasets in the test set; y i is the true value of the average of the state parameters after normalization in the i-th model dataset in the test set; y' i is the predicted value of the average of the state parameters after normalization in the i-th model dataset in the test set; Step S303: Monitor the power equipment in the current cycle. Every unit time interval, collect the values of various state parameters and various operating parameter influencing factors of the power equipment, and input them into the data prediction model to predict the various state parameters of the power equipment, obtaining parameter prediction data. The parameter prediction data includes the predicted parameter set of various state parameters; When the maximum or minimum value of a state parameter in the parameter prediction data is not within the range of the state parameters of the defect discrimination data, it is determined that there is a risk of equipment defects in the power equipment in the current cycle, and the state parameter and the power equipment are respectively recorded as the defect state parameter and the defective power equipment; Step S400: Obtain the characteristic defect root cause data of the defective power equipment, obtain the parameter prediction data, analyze the defect root cause of the defective power equipment to obtain the defect root cause data, and send the defect root cause data of the defective power equipment to the staff through the power grid platform, and prompt the staff to repair the defective power equipment; Among them, step S400 includes: Step S401: Obtain the defective power equipment in the power grid, obtain the characteristic defect root cause data of the defective power equipment from the power grid platform, obtain the root cause discrimination data of each defect root cause of the defective power equipment from the characteristic defect root cause data, and obtain the state parameters to be detected when discriminating a certain defect root cause from the characteristic defect root cause data, and record them as the target state parameters of a certain defect root cause. Among them, the root cause discrimination data of a certain defect root cause is the data range where the target state parameters of a certain defect root cause are located when the defective power equipment has a certain defect root cause; Step S402: Obtain the predicted data set of the various defect state parameters of the defective power equipment. When a certain defect state parameter is the target state parameter of a certain defect root cause, and the value of an element in the predicted data set of a certain defect state parameter is within the data range of the root cause discrimination data, it is recorded as a primary parameter compliance of a certain defect root cause, and obtain the total number of times M' of the parameter compliance of a certain defect root cause sum , obtain the total number of items M of the target state parameters of a certain defect root cause sum ; Calculate the occurrence probability P of a certain defect root cause = M' sum / M sum, when the occurrence probability P is greater than a preset probability threshold, it is determined that there is a certain defect root cause of the defective power equipment in the current cycle, and a certain defect root cause is recorded as the target defect root cause; Step S403: Obtain and collect each target defect root cause of the defective power equipment in the current cycle to obtain defect root cause data, send the defect root cause data to the grid staff through the grid platform, and prompt the staff to dispatch personnel to repair the defective power equipment.

[0018] To better implement the above method, a data-driven intelligent discrimination system for power equipment defects is also proposed. The system includes an abnormal analysis module, a parameter impact analysis module, an equipment defect discrimination module, and a defect root cause analysis module; The abnormal analysis module is used to obtain the historical defect records of power equipment, obtain the defect discrimination data of power equipment, analyze the abnormal degree of the state parameters in the defect discrimination data for defect discrimination of power equipment, and obtain parameter abnormal data; The parameter impact analysis module is used to obtain the historical equipment operation records of power equipment, analyze the operation parameters in the equipment operation records, and obtain parameter impact data on the impact degree of the abnormal state parameters in the parameter abnormal data; The equipment defect discrimination module is used to monitor the power equipment in the current cycle, predict the change trend of the state parameters of the power equipment in the current cycle, obtain parameter prediction data, and combine the defect discrimination data to perform equipment defect discrimination on the power equipment to obtain defective power equipment; The defect root cause analysis module is used to analyze the defect root cause of the defective power equipment according to the characteristic defect root cause data, obtain the defect root cause data, send the defect root cause data of the defective power equipment to the staff through the grid platform, and prompt the staff to repair the defective power equipment.

[0019] Among them, the abnormal analysis module includes a record abnormal analysis unit and an abnormal analysis unit; The record abnormal analysis unit is used to obtain each historical defect record of the power equipment, and combine the defect discrimination data of the power equipment to analyze the record abnormality of each historical defect record to obtain abnormal historical defect records; The abnormal analysis unit is used to analyze the abnormal state of each state parameter in the defect discrimination data for defect discrimination of the power equipment according to the abnormal historical defect records, and obtain parameter abnormal data.

[0020] Among them, the parameter impact analysis module includes a parameter set acquisition unit and a parameter impact analysis unit; A parameter set acquisition unit for acquiring the historical device operation records of power equipment and obtaining the parameter sets of various operation parameters from the historical device operation records; A parameter influence analysis unit for obtaining the parameter sets of abnormal state parameters in the historical device operation records and analyzing the influence degree of operation parameters on abnormal state parameters based on the parameter sets of various operation parameters in the historical device operation records of the power equipment to obtain parameter influence data.

[0021] Among them, the device defect discrimination module includes a model construction unit and a device defect discrimination unit; The model construction unit is used to construct a data prediction model for the power equipment according to each historical operation record of the power equipment; The device defect discrimination unit is used to use the data prediction model to predict the state parameters of the power equipment in the current period, and combine the defect discrimination data of the power equipment to discriminate the device defects of the power equipment to obtain defective power equipment.

[0022] Among them, the defect root cause analysis module includes a defect root cause analysis unit; The defect root cause analysis unit is used to obtain the characteristic defect root cause data of the defective power equipment, and combine the parameter prediction data to analyze the defect root cause of the defective power equipment in the current period and send it to the staff of the power grid platform to prompt the staff to repair the defective power equipment.

[0023] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. A data-driven intelligent discrimination method for power equipment defects, characterized in that, The method includes: Step S100: Obtain the historical defect records of the power equipment from the power grid platform, obtain the defect discrimination data of the power equipment, analyze the abnormal degree of the state parameters in the defect discrimination data for defect discrimination of the power equipment, and obtain parameter abnormal data; Step S200: Obtain the historical equipment operation records of the power equipment, analyze the operation parameters in the historical equipment operation records, and obtain parameter influence data for the influence degree of the abnormal state parameters in the parameter abnormal data; Step S300: Obtain the parameter influence data, monitor the power equipment in the current cycle, predict the change trend of the state parameters of the power equipment in the current cycle, obtain parameter prediction data, and combine the defect discrimination data to perform equipment defect discrimination on the power equipment to obtain defective power equipment; Step S400: Obtain the characteristic defect root cause data of the defective power equipment, obtain the parameter prediction data, analyze the defect root cause of the defective power equipment to obtain defect root cause data, send the defect root cause data of the defective power equipment to the staff through the power grid platform, and prompt the staff to repair the defective power equipment.

2. The intelligent discrimination method for power equipment defects based on data driving according to claim 1, wherein The step S100 includes: Step S101: Obtain each historical defect record of the power equipment, and obtain the data of each state parameter of the power equipment from the historical defect record, where equipment defects have occurred in the power equipment in the historical defect record; Step S1O2: Obtain the defect discrimination data of the power equipment from the power grid platform, where the defect discrimination data is the range of each state parameter when performing defect discrimination on the power equipment; When the maximum value and the minimum value of each state parameter in a certain historical defect record are both within the range of the defect discrimination data, it is determined that the defect discrimination data is incorrect for defect discrimination of the certain historical defect record, mark the certain historical defect record, and record it as an abnormal historical defect record; Obtain the total number C of several historical defect records in the power equipment that have not been marked sum , obtain several status parameters whose maximum or minimum values in the unmarked historical defect records are outside the range of the defect discrimination data, and pool them to obtain the characteristic parameter set of the unmarked historical defect records; Step S103: Calculate the defect discrimination ratio of each of the state parameters, where the defect discrimination ratio L of the a-th state parameter in the power equipment a =C (a,sum) / C sum , C (a,sum) is the total number of unmarked historical defect records containing the a-th state parameter in the feature parameter set; Calculate the average value L´ and the standard deviation σ´ of the defect discrimination ratios of the various state parameters, and obtain the lower limit ratio threshold B of the various state parameters min = L´ - k × σ´, where k is a preset coefficient. When the defect discrimination ratio of a certain state parameter is less than the lower limit ratio threshold, mark the certain state parameter and record it as a suspected abnormal state parameter; Step S104: Obtain the median D of the suspected abnormal state parameters in the defect discrimination data within the corresponding range △ , and respectively calculate the absolute values of the differences between the maximum and minimum values of the suspected abnormal state parameters in the abnormal historical defect records of the power equipment and the median D △ . Then take the maximum value of the absolute values, and divide the maximum value by the median D △ to obtain the parameter abnormal value e of the suspected abnormal state parameters in the abnormal historical defect records; When the parameter abnormal value e is greater than the preset threshold e´, record the abnormal historical defect record as the target abnormal historical defect record of the suspected abnormal state parameter, and obtain the ratio of the total number of the target abnormal historical defect records of the suspected abnormal state parameter to the total number of each abnormal historical defect record in the power equipment, to obtain the abnormal ratio of the suspected abnormal state parameter; When the abnormal ratio is greater than the preset threshold, it is determined that there is an abnormality in the defect identification of the suspected abnormal state parameter in the defect discrimination data, and record the suspected abnormal state parameter as an abnormal state parameter; Obtain several abnormal state parameters of the power equipment and gather them to obtain the parameter abnormal data of the power equipment.

3. A data-driven intelligent discrimination method for power equipment defects according to claim 2, characterized in that The step S200 includes: Step S201: Obtain the respective historical device operation records of the power device, preprocess the data in the respective historical device operation records, set a unit time period, and obtain the average values of the respective operation parameters in the historical device operation records every unit time period to obtain a parameter set of the respective operation parameters; Step S202: Obtain the parameter abnormal data of the power device, and obtain the respective abnormal state parameters from the parameter abnormal data; Obtain the average values of the respective abnormal state parameters from the historical device operation records every unit time period to obtain a parameter set of the respective abnormal state parameters; Calculate the parameter influence degree between each operation parameter and each abnormal state parameter in the historical device operation record, where the parameter influence degree S between the f-th operation parameter and the g-th abnormal state parameter in the historical device operation record (f,g) ; Step S203: Calculate the data influence value R of the f-th operating parameter on the g-th abnormal state parameter (f,g) : ， where z is the total number of all historical device operation records of the power device; S z (f,g) is the parameter influence degree between the f-th operation parameter and the g-th abnormal state parameter in the z-th historical device operation record of the power device; Step S204: Set an influence threshold R. When |R (f,g) | ≥ R, it is determined that the f-th operating parameter has an impact on the data of the g-th abnormal state parameter, and the f-th operating parameter is recorded as the influencing operating parameter of the power device. Collect the influencing operating parameters of the power device to obtain parameter influence data.

4. The intelligent discrimination method for power equipment defects based on data driving according to claim 3, wherein, The said step S300 includes: Step S301: Obtain the parameter influence data, obtain the respective operation parameters affecting the power device, obtain the respective historical operation records of the power device, preprocess the data in the respective historical operation records, and sort the respective historical operation records in chronological order; Construct a data prediction model for the power device, and divide the respective historical operation records into a training set and a test set according to a preset ratio; Obtain the input data and output data of the model data set in the model. The input data is the respective state parameters and the respective operation parameters affecting the operation in a certain historical operation record, and the output data is the respective state parameters in the next historical operation record of the said certain historical operation record; Step S302: Use the training set to train the data prediction model, and use the test set to calculate the model prediction accuracy of the data prediction model. When the model prediction accuracy is greater than a preset threshold, determine that the data prediction model is constructed; Step S303: Monitor the power device in the current cycle. Every unit time period, collect the values of the respective state parameters and the respective operation parameters affecting the operation of the power device and input them into the data prediction model to predict the respective state parameters of the power device to obtain parameter prediction data. The parameter prediction data includes a predicted parameter set of the respective state parameters; When the maximum or minimum value of the state parameter in the parameter prediction data is not within the range of the state parameter in the defect discrimination data, determine that there is a risk of device defect in the power device in the current cycle, and record the state parameter and the power device as the defect state parameter and the defective power device respectively.

5. The intelligent discrimination method for power equipment defects based on data driving according to claim 4, characterized in that The said step S400 includes: Step S401: Obtain the defective power devices in the power grid, obtain the characteristic defect root cause data of the defective power devices from the power grid platform, obtain the root cause discrimination data of the respective defect root causes of the defective power devices from the characteristic defect root cause data, and obtain the state parameters to be detected when discriminating a certain defect root cause from the characteristic defect root cause data and record them as the target state parameters of the said certain defect root cause. Among them, the root cause discrimination data of the said certain defect root cause is the data range where the respective target state parameters of the said certain defect root cause are located when the defective power device has the said certain defect root cause; Step S402: Obtain a predicted data set of various defect status parameters of the defective power equipment. When a certain defect status parameter is the target status parameter of the certain defect root cause, and the value of an element in the predicted data set of the certain defect status parameter is within the data range of the root cause discrimination data, it is recorded as a first parameter compliance of the certain defect root cause, and obtain the total number of times M' of parameter compliance of the certain defect root cause sum , and obtain the total number of items M of the target status parameters of each item of the certain defect root cause sum ; Calculate the occurrence probability P of a certain defect root cause, where P = M' sum / M sum , when the occurrence probability P is greater than a preset probability threshold, it is determined that the defective power equipment has the certain defect root cause in the current cycle, and the certain defect root cause is recorded as the target defect root cause; Step S403: Obtain each target defect root cause of the defective power equipment in the current cycle and aggregate them to obtain the defect root cause data. Send the defect root cause data to the grid staff through the grid platform, and prompt the staff to dispatch personnel to repair the defective power equipment.

6. A data-driven intelligent discrimination system for power equipment defects, which is used to execute a data-driven intelligent discrimination method for power equipment defects described in any one of claims 1-5, characterized in that, The system includes an anomaly analysis module, a parameter impact analysis module, a device defect discrimination module, and a defect root cause analysis module; The anomaly analysis module is used to obtain the historical defect records of power equipment, obtain the defect discrimination data of the power equipment, and analyze the abnormal degree of the status parameters in the defect discrimination data for defect discrimination of the power equipment to obtain parameter anomaly data; The parameter impact analysis module is used to obtain the historical equipment operation records of the power equipment, analyze the operation parameters in the equipment operation records, and analyze the parameter impact degree of the abnormal status parameters in the parameter anomaly data to obtain parameter impact data; The device defect discrimination module is used to monitor the power equipment in the current cycle, predict the change trend of the status parameters of the power equipment in the current cycle to obtain parameter prediction data, and combine the defect discrimination data to perform device defect discrimination on the power equipment to obtain defective power equipment; The defect root cause analysis module is used to analyze the defect root cause of the defective power equipment according to the characteristic defect root cause data to obtain defect root cause data, send the defect root cause data of the defective power equipment to the staff through the grid platform, and prompt the staff to repair the defective power equipment.

7. The intelligent discrimination system for power equipment defects based on data driving according to claim 6, wherein The anomaly analysis module includes a record anomaly analysis unit and an anomaly analysis unit; The record anomaly analysis unit is used to obtain each historical defect record of the power equipment, and combine the defect discrimination data of the power equipment to analyze the record anomaly of each historical defect record to obtain an abnormal historical defect record; The anomaly analysis unit is used to analyze the abnormal status of each status parameter in the defect discrimination data for defect discrimination of the power equipment according to the abnormal historical defect record to obtain parameter anomaly data.

8. An intelligent discrimination system for power equipment defects based on data driving according to claim 6, characterized in that The parameter impact analysis module includes a parameter set acquisition unit and a parameter impact analysis unit; The parameter set acquisition unit is used to obtain the historical equipment operation records of the power equipment and obtain the parameter set of each operation parameter from the historical equipment operation records; The parameter impact analysis unit is used to obtain the parameter set of the abnormal status parameters in the historical equipment operation records, and analyze the parameter impact degree of the operation parameters on the abnormal status parameters according to the parameter set of each operation parameter in the historical equipment operation records of the power equipment to obtain parameter impact data.

9. The intelligent discrimination system for power equipment defects based on data driving according to claim 6, wherein The device defect discrimination module includes a model construction unit and a device defect discrimination unit; The model construction unit is used to construct a data prediction model of the power equipment according to each historical operation record of the power equipment; The device defect discrimination unit is used to predict various state parameters of the power device in the current cycle by using the data prediction model, and combine the defect discrimination data of the power device to discriminate the device defects of the power device, so as to obtain defective power devices.

10. An intelligent discrimination system for power equipment defects based on data driving according to claim 6, characterized in that The defect root cause analysis module includes a defect root cause analysis unit; The defect root cause analysis unit is used to obtain the characteristic defect root cause data of the defective power device, and combine the parameter prediction data to analyze the defect root cause of the defective power device in the current cycle, and send it to the staff of the power grid platform to prompt the staff to repair the defective power device.

Images