Power supply quality detection method based on artificial intelligence

Abstract

The invention provides a power supply quality detection method based on artificial intelligence, and relates to the technical field of power supply quality intelligent detection. The method comprises the following steps: acquiring historical quality detection data of same-type power supplies, and establishing a same-type power supply quality detection database; historical monitoring information of different detection parameters in the quality detection database of the same type of power supplies is extracted, feature information extraction aiming at the detection parameters is carried out, and detection feature comparison data is formed; and acquiring real-time detection data of the target power supply, and performing intelligent detection analysis in combination with the detection characteristic comparison data to form a real-time detection analysis result. According to the method, through automatic analysis and comparison of the detection parameters based on the power supply service life characteristics, the accuracy of power supply detection is ensured, and the efficiency of power supply detection is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent power quality detection, and in particular to a power quality detection method based on artificial intelligence. Background Art

[0002] As an energy supply unit, the power supply provides sufficient electric energy for power-consuming facilities and equipment to ensure the normal operation of the power-consuming facilities and equipment. In order to ensure that the power supply can continuously and stably supply electric energy, it is necessary to conduct real-time quality detection of the power supply. At present, the quality detection of the power supply is mainly a simple comparative analysis and judgment of the real-time detection parameters under the set threshold, and does not fully consider the change characteristics of the detection parameters affected by the service life of the power supply, resulting in low detection accuracy. On the other hand, most of the quality detection of the power supply still needs to be performed manually, and intelligent detection cannot be fully realized, which greatly reduces the efficiency of power quality detection.

[0003] Therefore, designing a power quality detection method based on artificial intelligence, which not only ensures the accuracy of power supply detection but also greatly improves the efficiency of power supply detection through automated analysis and comparison of detection parameters based on the power supply life characteristics, is a problem that needs to be solved urgently. Summary of the invention

[0004] The purpose of the present invention is to provide a power quality detection method based on artificial intelligence, by acquiring historical monitoring data of the same type of power supply over its service life cycle to establish a database for analyzing detection parameters based on service life characteristics, and by extracting and analyzing characteristic information of the detection parameters to establish basic comparison data of different detection parameters. At the same time, the real-time detection information of the target power supply is obtained, and the basic comparison data is automatically analyzed and compared with the basic comparison data in real time by extracting information based on service life characteristics to determine the power quality. On the one hand, the detection parameter data for service life characteristics is extracted based on historical monitoring data, which fully considers the changing characteristics of the detection parameters in the service life characteristics of the power supply, ensures the accuracy of the basic comparison data, and makes the data comparison more accurate and reasonable. On the other hand, the automatic analysis and comparison is combined with the historical detection big data, which greatly improves the efficiency of real-time analysis and comparison.

[0005] In a first aspect, the present invention provides a power quality detection method based on artificial intelligence, including obtaining historical quality detection data of power supplies of the same type and establishing a power quality detection database of the same type; extracting historical monitoring information of different detection parameters in the power quality detection database of the same type, performing feature information extraction on the detection parameters, and forming detection feature comparison data; obtaining real-time detection data of the target power supply, and combining the detection feature comparison data to perform intelligent detection analysis to form real-time detection analysis results.

[0006] In the present invention, the method establishes a database for analyzing detection parameters based on service life characteristics by acquiring historical monitoring data of the same type of power supply over its service life cycle, and establishes basic comparison data for different detection parameters by extracting and analyzing characteristic information of the detection parameters. At the same time, the real-time detection information of the target power supply is acquired, and the basic comparison data is automatically analyzed and compared with the basic comparison data in real time by extracting information based on service life characteristics to determine the quality of the power supply. On the one hand, the detection parameter data for service life characteristics is extracted based on historical monitoring data, which fully considers the changing characteristics of the detection parameters in the service life characteristics of the power supply, ensures the accuracy of the basic comparison data, and makes the data comparison more accurate and reasonable. On the other hand, the automatic analysis and comparison is combined with historical detection big data, which greatly improves the efficiency of real-time analysis and comparison.

[0007] As a possible implementation method, historical monitoring information of different detection parameters in the same type of power quality detection database is extracted, and feature information extraction for the detection parameters is performed to form detection feature comparison data, including: for the historical quality detection data of different power supplies of the same type, obtaining the object parameter life change curve formed by different detection parameters in the order of time dimension; for the same detection parameters, clustering the object parameter life change curves corresponding to different power supplies, and extracting feature information based on the order of time dimension to form parameter life feature information; and aggregating the parameter life feature information corresponding to different detection parameters to form detection feature comparison data.

[0008] In the present invention, feature information extraction based on detection parameters is performed on the established database. Considering that the feature data of the detection parameters need to be combined with the service life of the power supply, the feature information extraction, on the one hand, needs to form corresponding value change information on the time sequence of the power supply service life for different detection parameters, so as to provide a data basis for the subsequent feature information extraction based on the time dimension, so that the extracted data has the characteristics of the power supply service life. On the other hand, a specific extraction and analysis method of the feature information is performed. Considering that the extracted feature information has the characteristics of the service life, the extraction of the feature information also needs to be performed around the time parameter, that is, the extraction and analysis of the life feature information of different detection parameters is performed based on the time dimension sequence.

[0009] As a possible implementation method, for the same detection parameters, object parameter life change curves corresponding to different power sources are clustered, and feature information is extracted based on the time dimension sequence to form parameter life feature information, including: for the same detection parameters, all object parameter life change curves are clustered, and feature information is extracted based on the time dimension sequence to form parameter life feature information, including: for the same detection parameters, all object parameter life change curves are obtained to divide the service life stages based on parameter level changes to form parameter detection life stage information; according to the parameter detection life stage information, parameter detection feature information is extracted in combination with all object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information; parameter detection stage feature information corresponding to different detection parameters is aggregated to form parameter life feature information.

[0010] In the present invention, after obtaining the parameter value change information of the detection parameters during the service life of the power supply, considering that the parameter value change characteristics of the detection parameters are relatively large at different stages of the service life of the power supply, it is necessary to first divide the service life into different stages based on the data characteristics displayed by the detection parameters, and then extract the characteristic information of different detection parameters in stages to ensure that the extracted characteristic information of the detection parameters is more accurate and reasonable.

[0011] As a possible implementation method, for the same detection parameters, all object parameter life change curves are obtained to divide the service life stages based on parameter level changes to form parameter detection life stage information, including: determining the maximum object parameter value for different object parameter life change curves under the detection parameters; setting the parameter value adjustment direction, starting with the maximum object parameter value, gradually and continuously adjusting the parameter value, and judging the stage parameter value according to the following method: when the obtained parameter value corresponds to a time point greater than or equal to 2 and less than or equal to 4 on the object parameter life change curve, and the number of time points corresponding to the parameter values ​​continuously obtained in the parameter value adjustment direction on the object parameter life change curve increases in a jump, then the obtained parameter value is determined as the stage parameter value. numerical value; the earliest time point corresponding to the stage parameter value on the object parameter life change curve is determined as the early time point of object use; the latest time point corresponding to the stage parameter value on the object parameter life change curve is determined as the late time point of object use; the time period between the object start time point and the early time point of object use of the object parameter life change curve is determined as the early life stage of object use; the time period between the early time point and the late time point of object use of the object parameter life change curve is determined as the mid-life stage of object use; the time period between the late time point and the end time point of the object use of the object parameter life change curve is determined as the late life stage of object use.

[0012] In the present invention, the method of dividing the different life stages of the power supply life is determined by considering the number of values ​​of the detection parameters in the whole life cycle. It can be understood that in the initial stage and the final stage of the use of the power supply, the value change level of the detection parameter is relatively large, and the data change has a stable regularity in the middle stage, especially the detection parameters of the power supply, including but not limited to voltage, current, power, etc., have large fluctuations in the early stage of use, and the values ​​have the characteristics of continuous changes, stable small fluctuations in the middle stage, and continuous changes in values ​​similar to the initial stage in the late stage. Therefore, from the perspective of the entire service life cycle of the power supply, if the obtained detection parameter value has a jump change of more than 4 points on the curve, it can be considered that this value is the critical value for the life stage division. After all, in the early stage, due to the instability of the value of the detection parameter, there may be a large fluctuation jump on the curve, causing the jump value to return to the previous value. Therefore, a certain value in the early stage has a maximum of 2 intersections on the curve. The same is true for the later stage. There are a maximum of 2 intersections on the curve. Considering the entire service life cycle, there may be a maximum of 4 intersections. If the number of intersections suddenly jumps from 4 to more intersections, it can be considered that the intersection of the curve corresponding to this value may have a point on the fluctuation curve in the mid-term stable stage, which has exceeded the range of variation of the value taken during the initial and final periods. In this way, the time points of different life stages are determined, and different life stages are divided, completing the division of detection parameter data relative to different stages of the power supply life.

[0013] As a possible implementation method, according to the parameter detection life stage information, the parameter detection feature information is extracted in combination with all object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information, including: extracting the curve data of the life change curves of different object parameters under the detection parameters in the corresponding early life stage of the object to form the early life change curve of the object parameter; for different early life change curves of object parameters, determining the corresponding average change amount of the early object parameter ,in, , n represents the number of different detection parameters, i represents the number of different object parameter life change curves corresponding to the detection parameter numbered n, The maximum parameter detection value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The minimum parameter detection value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The time length value of the object parameter early life change curve obtained by the object parameter life change curve numbered i corresponding to the detection parameter numbered n; the average change of all early object parameters corresponding to the detection parameter , determine the average range of changes in the previous parameters corresponding to the detection parameters For different object parameter early life change curves, the corresponding early object time indicator function is determined according to the parameter detection value obtained at each time point and the ratio of the time span of the time point corresponding to each parameter detection value relative to the object starting time point to the time length value obtained by the object parameter early life change curve , r is the ratio of the previous detection time, which means the ratio of the time span of the time point corresponding to the parameter detection value relative to the object starting time point to the time length value obtained by the object parameter previous life change curve; according to different object time indicator functions , determine the range of parameter detection values ​​corresponding to the same detection time ratio, and form a mapping of the previous parameter value-time ratio range corresponding to the detection parameter ; Determine the maximum duration of the object's early life stage in the early life change curve of different object parameters, and mark it as the reference duration of the early parameters .

[0014] In the present invention, the extraction of the characteristic information of the detection parameters in the early stage of power use considers two aspects. One is that the value of the initial detection parameter changes greatly, and this change trend is within a certain range. If it exceeds this change range, it may be a situation where the detection parameter change is distorted due to power failure or other factors. On the other hand, the initial duration corresponding to different power supplies is different, and the change trend of the detection parameter is range-constrained. In order to make full use of big data for feature information extraction, it is necessary to determine the numerical value range of the detection parameter allowed at different time points under the standard reference time length. Thus, on the one hand, the average change of the detection parameter in the initial use is determined, and on the other hand, the detection parameter value is mapped by determining a reasonable same time ratio to perform a comprehensive application analysis of the initial parameter data between different power supplies. Of course, in order to ensure that the real-time data can be used to compare with the currently collected characteristic information data in the real-time detection process in the later stage, it is necessary to select a standard reference for the initial duration, which also provides an important data reference for the subsequent real-time comparison and analysis. It can be understood that the detection parameters start from values ​​that are significantly different from the stable values ​​in the mid-term in the early stage. The larger the deviation of the parameter values ​​at each time point is from the starting time point of the previous month, in order to obtain accurate comparison results in the subsequent early real-time comparative analysis, selecting the maximum value of the initial duration as the basic standard quantity of the time ratio is conducive to grasping the impact of the deviation of the initial detection parameters on the detection results.

[0015] As a possible implementation method, parameter detection feature information is extracted based on parameter detection life stage information and combined with all object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information, including: extracting curve data of different object parameter life change curves under the detection parameters in the corresponding object mid-term life stage to form object parameter mid-term life change curves; performing numerical range analysis of detection parameters on different object parameter mid-term life change curves to determine the parameter detection value range; performing detection value fluctuation feature analysis on different object parameter mid-term life change curves to form mid-term parameter fluctuation feature information.

[0016] In the present invention, for the extraction of detection parameter feature information in the mid-term stage of the power supply's service life, considering the mid-term detection parameter value change characteristics, it is necessary to extract two types of feature information, one is that the detection parameter has a certain volatility in the mid-term stage of the service life, and this volatility is reasonable and needs to be controlled, so the detection parameter needs to extract the corresponding value variation range in the mid-term stage, and the other is the fluctuation characteristics of the detection parameter in the mid-term stage. After all, the more the fluctuation amount and the higher the fluctuation frequency, the greater the noise of the detection parameter will be, which will have a greater impact on the stable operation of the power supply and the stable output of electric energy.

[0017] As a possible implementation method, the value range of the detection parameters is analyzed for different object parameter mid-term life change curves to determine the parameter detection range, including: extracting the minimum parameter detection value and the maximum parameter detection value of the detection parameters corresponding to the different object parameter mid-term life change curves to establish the object parameter detection value range; performing a AND operation on all object parameter detection value ranges to form the parameter detection value range corresponding to the detection parameter .

[0018] In the present invention, the range extraction of the detection parameters in the mid-term stage of power use mainly obtains the range values ​​of different power supplies in the mid-term, and then merges the ranges to obtain a wider range value to provide a basic reference for subsequent real-time analysis and comparison.

[0019] As a possible implementation method, the mid-term life change curves of different object parameters are analyzed based on the detection amount fluctuation characteristics to form mid-term parameter fluctuation characteristic information, including: setting a fluctuation amount judgment threshold, and determining the fluctuation amount of different object parameter mid-term life change curves in the following manner: if within the continuous time change range, the parameter detection values ​​of the detection parameters have equal intervals, and the difference in the parameter detection values ​​in the time period between the equal parameter detection values ​​is not less than the fluctuation amount judgment threshold, then the time period between the equal parameter detection values ​​is marked as a fluctuation period; extracting the maximum value of the fluctuation period in the mid-term life change curves of different object parameters. And the shortest value of the object's mid-life stage in the mid-life change curve of different object parameters , determine the parameter fluctuation rate corresponding to the detection parameter ,in, ; Determine the maximum duration of the object's mid-life phase in the mid-life change curve of different object parameters , and combined with the previous parameter reference time , calibrated as the mid-term parameter reference duration ,in, .

[0020] In the present invention, the fluctuation characteristic information of the detection parameters in the mid-term stage of power use mainly considers the frequency of the fluctuation. This frequency information needs to be obtained by extracting the corresponding number of fluctuations and the duration of the mid-term stage. The number of fluctuations is obtained through the characteristics of the fluctuations on the curve, that is, the fluctuations on the curve will have continuous value changes between the same values. The threshold for judging the amount of fluctuations can be set according to actual conditions, or it can be determined based on big data analysis. For the duration of the mid-term stage, considering that the fluctuation frequency is a limited boundary value, obtaining the shortest mid-term duration as a reference for extracting frequency characteristic information can maximize the extraction of this frequency limit. Of course, in real-time analysis, the real-time extracted data cannot determine the time position in the mid-term stage, so it is reasonable to provide a corresponding mid-term reference duration as a basis for judging the duration of the mid-term stage in subsequent real-time analysis.

[0021] As a possible implementation method, the mid-term life span change curves of different object parameters are analyzed based on the detection amount fluctuation characteristics to form mid-term parameter fluctuation characteristic information, including: extracting the curve data of the life span change curves of different object parameters under the detection parameters in the corresponding object use later life stage to form the object parameter later life span change curve; for different object parameter later life span change curves, determining the corresponding later object parameter average change amount ,in, , The maximum parameter detection value obtained by the object parameter later life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The minimum parameter detection value obtained by the object parameter later life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The time length value of the object parameter later life change curve obtained by the object parameter life change curve numbered i corresponding to the detection parameter numbered n; the average change of all later object parameters corresponding to the detection parameter , determine the average range of later parameters corresponding to the detection parameters For different object parameter late life change curves, the corresponding late object time indicator function is determined according to the parameter detection value obtained at each time point and the ratio of the time span of the time point corresponding to each parameter detection value relative to the object end time point to the time length value obtained by the object parameter late life change curve , k is the later detection time ratio, which means the ratio of the time span of the time point corresponding to the parameter detection value relative to the object starting time point to the time length value obtained by the object parameter life change curve; according to different later object time indicator functions , determine the range of parameter detection values ​​corresponding to the same detection time ratio, and form a later parameter value-time ratio range mapping corresponding to the detection parameter ; Determine the maximum value of the corresponding duration in the life span change curve of different object parameters, and mark it as the reference duration of the later parameters .

[0022] In the present invention, similarly, when there is a large change in the value of the detection parameter in the later stage of power use, the characteristic information of the detection parameter value on the average change amount and the time ratio can be used to analyze and judge. Here, for the mapping of the parameter value time ratio, considering that the accurate total power use time cannot be known in the case of real-time detection, the later parameter reference time is used as the time reference standard, and the time ratio is determined with the starting time point as a reference.

[0023] As a possible implementation method, the real-time detection data of the target power supply is obtained, and combined with the detection feature comparison data, intelligent detection analysis is performed to form real-time detection analysis results, including: obtaining the parameter target real-time detection change curve corresponding to different detection parameters of the target power supply; extracting the total real-time duration corresponding to the parameter target real-time detection change curve , and perform the following real-time detection analysis based on the parameter target real-time detection change curve and parameter detection stage characteristic information: ≤ , it is determined that the real-time monitoring corresponding to the detection parameters of the target power supply is in the early life stage of use, and the real-time average change of the early parameters is determined according to the parameter target real-time detection change curve And the corresponding real-time preliminary parameter value-time ratio mapping , where real-time pre-parameter value-time ratio mapping The time ratio is the time span of the time point corresponding to the real-time parameter detection value relative to the real-time start time point and the previous parameter reference time length The ratio of and judge: If ∈ , and real-time early parameter value-time ratio mapping The value at each time ratio belongs to the previous parameter value-time ratio range mapping If the value range of the corresponding time ratio is within the range, it is determined that the corresponding detection parameter is normal in real time, otherwise it is abnormal; < < , then it is determined that the real-time monitoring corresponding to the detection parameters of the target power supply is in the mid-life stage, and the extraction < Real-time parameter detection range of detection parameters within the time period , according to the fluctuation amount, the threshold is determined < The number of real-time fluctuation segments within the time period, and according to the number of real-time fluctuation segments and < The duration of the time period determines the real-time parameter volatility , and make a judgment: If ∈ ,and ≤ , then it is determined that the corresponding detection parameters are detected normally in real time, otherwise it is abnormal; when ≤ , it is determined that the real-time monitoring corresponding to the detection parameters of the target power supply is in the later life stage, and the real-time detection change curve of the parameter target is determined. ≤ The real-time average change of the later parameters corresponding to the time period And the corresponding real-time late parameter value-time ratio mapping , where real-time late parameter value-time ratio mapping The time ratio is the time span of the time point corresponding to the real-time parameter detection value relative to the real-time start time point and the reference time length of the later parameters The ratio of and judge: If ∈ , and real-time late parameter value-time ratio mapping The value at each time ratio belongs to the later parameter value-time ratio range mapping If the value range of the above corresponding time ratio is within the range, it is determined that the corresponding detection parameter real-time detection is normal, otherwise it is abnormal.

[0024] In the present invention, after extracting the characteristic information of the service life characteristics of the detection parameters, the target power supply can be subjected to real-time detection and analysis of the detection parameters to determine the quality of the power supply. Of course, the quality of the power supply is judged based on the normality of all detection parameters. Considering that different detection parameters may have different starting time points, it is necessary to combine the characteristic comparison data of different detection parameters for real-time analysis and judgment. First, the use stage of the target power supply is determined by the duration of the real-time detection, and then the characteristic information of the corresponding stage is extracted for analysis and judgment. Only when all the restrictions are met at the same time can the detection parameters be considered normal. Otherwise, the corresponding detection parameters are abnormal, and it is inferred that the quality of the power supply is abnormal.

[0025] The beneficial effects of the power quality detection method based on artificial intelligence provided by the present invention are as follows: This method obtains historical monitoring data of the same type of power supply over its service life cycle to establish a database for analyzing detection parameters based on service life characteristics, and establishes basic comparison data for different detection parameters by extracting and analyzing characteristic information of the detection parameters. At the same time, real-time detection information of the target power supply is obtained, and real-time automatic analysis and comparison with the basic comparison data is performed through information extraction based on service life characteristics to determine the quality of the power supply. On the one hand, based on historical monitoring data, detection parameter data for service life characteristics is extracted, which fully considers the changing characteristics of the detection parameters in the service life characteristics of the power supply, ensures the accuracy of the basic comparison data, and makes the data comparison more accurate and reasonable. On the other hand, combined with historical detection big data for automatic analysis and comparison, the efficiency of real-time analysis and comparison is greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments of the present invention are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 A step diagram of a power quality detection method based on artificial intelligence provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present invention will be described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0029] As an energy supply unit, the power supply provides sufficient electric energy for power-consuming facilities and equipment to ensure the normal operation of the power-consuming facilities and equipment. In order to ensure that the power supply can continuously and stably supply electric energy, it is necessary to conduct real-time quality detection of the power supply. At present, the quality detection of the power supply is mainly a simple comparative analysis and judgment of the real-time detection parameters under the set threshold, and does not fully consider the change characteristics of the detection parameters affected by the service life of the power supply, resulting in low detection accuracy. On the other hand, most of the quality detection of the power supply still needs to be performed manually, and intelligent detection cannot be fully realized, which greatly reduces the efficiency of power quality detection.

[0030] refer to Figure 1 , an embodiment of the present invention provides a power quality detection method based on artificial intelligence, which establishes a database for analyzing detection parameters based on service life characteristics by acquiring historical monitoring data of the same type of power supply over its service life cycle, and establishes basic comparison data for different detection parameters by extracting and analyzing characteristic information of the detection parameters. At the same time, real-time detection information of the target power supply is acquired, and real-time automatic analysis and comparison with the basic comparison data is performed through information extraction based on service life characteristics to determine the power quality. On the one hand, detection parameter data extraction based on service life characteristics is performed based on historical monitoring data, which fully considers the changing characteristics of detection parameters in the service life characteristics of the power supply, ensures the accuracy of basic comparison data, and makes data comparison more accurate and reasonable. On the other hand, automatic analysis and comparison is performed in combination with historical detection big data, which greatly improves the efficiency of real-time analysis and comparison.

[0031] The power quality detection method based on artificial intelligence specifically includes the following steps: S1: Obtain historical quality detection data of power supplies of the same type and establish a quality detection database for power supplies of the same type.

[0032] Establishing a database of similar power quality tests can provide an important data basis for subsequent comparative feature information extraction based on test parameters.

[0033] S2: extract historical monitoring information of different detection parameters from the same type of power quality detection database, extract feature information for the detection parameters, and form detection feature comparison data.

[0034] Extract historical monitoring information of different detection parameters in the same type of power quality detection database, extract feature information for the detection parameters, and form detection feature comparison data, including: for historical quality detection data of different power supplies of the same type, obtain object parameter life change curves formed by different detection parameters in time dimension sequence; for the same detection parameters, cluster the object parameter life change curves corresponding to different power supplies, and extract feature information based on time dimension sequence to form parameter life feature information; aggregate parameter life feature information corresponding to different detection parameters to form detection feature comparison data.

[0035] The established database is subjected to feature information extraction based on the detection parameters. Considering that the feature data of the detection parameters need to be combined with the service life of the power supply, the feature information extraction, on the one hand, needs to form corresponding value change information on the time sequence of the power supply service life for different detection parameters, so as to provide a data basis for the subsequent feature information extraction based on the time dimension, so that the extracted data has the characteristics of the power supply service life. On the other hand, it is a specific method for extracting and analyzing the feature information. Considering that the extracted feature information has the characteristics of the service life, the extraction of feature information also needs to be carried out around the time parameter, that is, the extraction and analysis of the life feature information of different detection parameters is carried out based on the time dimension sequence.

[0036] For the same detection parameters, the object parameter life change curves corresponding to different power supplies are clustered, and feature information is extracted based on the time dimension sequence to form parameter life feature information, including: for the same detection parameters, the object parameter life change curves corresponding to different power supplies are clustered, and feature information is extracted based on the time dimension sequence to form parameter life feature information, including: for the same detection parameters, all object parameter life change curves are obtained to divide the service life stages based on parameter level changes to form parameter detection life stage information; according to the parameter detection life stage information, parameter detection feature information is extracted in combination with all object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information; parameter detection stage feature information corresponding to different detection parameters is aggregated to form parameter life feature information.

[0037] After obtaining the parameter value change information of the detection parameters during the service life of the power supply, considering that the parameter value change characteristics of the detection parameters are relatively large at different stages of the power supply's service life, it is necessary to first divide the service life into different stages based on the data characteristics shown by the detection parameters, and then extract the characteristic information of different detection parameters in stages to ensure that the extracted characteristic information of the detection parameters is more accurate and reasonable.

[0038] For the same detection parameters, all object parameter life change curves are obtained to divide the service life stages based on parameter level changes to form parameter detection life stage information, including: for different object parameter life change curves under the detection parameters, the maximum object parameter value is determined; the parameter value adjustment direction is set, starting with the maximum object parameter value, and the parameter value is gradually and continuously adjusted, and the stage parameter value is judged according to the following method: when the obtained parameter value corresponds to a time point greater than or equal to 2 and less than or equal to 4 on the object parameter life change curve, and the number of time points corresponding to the parameter values ​​continuously obtained in the parameter value adjustment direction on the object parameter life change curve increases in a jump, the obtained parameter value is determined as the stage parameter value; the stage The earliest time point corresponding to the parameter value on the object parameter life change curve is determined as the early time point of object use; the latest time point corresponding to the stage parameter value on the object parameter life change curve is determined as the late time point of object use; the time period between the object start time point and the early time point of object use of the stage parameter value on the object parameter life change curve is determined as the early life stage of object use; the time period between the early time point of object use and the late time point of object use of the stage parameter value on the object parameter life change curve is determined as the mid-life stage of object use; the time period between the late time point of object use and the end time point of the stage parameter value on the object parameter life change curve is determined as the late life stage of object use.

[0039] The way of dividing the different life stages of the power supply life provided is to judge by considering the number of values ​​of the detection parameters in the whole life cycle. It can be understood that in the initial and final stages of the use of the power supply, the value change level of the detection parameters is relatively large, and the data changes have a stable regularity in the middle stage, especially the detection parameters of the power supply, including but not limited to voltage, current, power, etc., fluctuate greatly in the early stage of use, and the values ​​have the characteristics of continuous change, stable small fluctuations appear in the middle stage, and continuous changes appear in the later stage. Therefore, from the perspective of the entire service life cycle of the power supply, if the obtained detection parameter value has a jump change of more than 4 points on the curve, it can be considered that this value is the critical value for the life stage division. After all, in the early stage, due to the instability of the value of the detection parameter, there may be a large fluctuation jump on the curve, causing the jump value to return to the previous value. Therefore, a certain value in the early stage has a maximum of 2 intersections on the curve. The same is true for the later stage. There are a maximum of 2 intersections on the curve. Considering the entire service life cycle, there may be a maximum of 4 intersections. If the number of intersections suddenly jumps from 4 to more intersections, it can be considered that the intersection of the curve corresponding to this value may have a point on the fluctuation curve in the mid-term stable stage, which has exceeded the range of variation of the value taken during the initial and final periods. In this way, the time points of different life stages are determined, and different life stages are divided, completing the division of detection parameter data relative to different stages of the power supply life.

[0040] According to the parameter detection life stage information, the parameter detection feature information is extracted in combination with all object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information, including: extracting the curve data of the life change curves of different object parameters under the detection parameters in the corresponding early life stage of the object to form the early life change curve of the object parameter; for different early life change curves of object parameters, determining the corresponding average change amount of the early object parameter ,in, , n represents the number of different detection parameters, i represents the number of different object parameter life change curves corresponding to the detection parameter numbered n, The maximum parameter detection value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The minimum parameter detection value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The time length value of the object parameter early life change curve obtained by the object parameter life change curve numbered i corresponding to the detection parameter numbered n; the average change of all early object parameters corresponding to the detection parameter , determine the average range of changes in the previous parameters corresponding to the detection parameters For different object parameter early life change curves, the corresponding early object time indicator function is determined according to the parameter detection value obtained at each time point and the ratio of the time span of the time point corresponding to each parameter detection value relative to the object starting time point to the time length value obtained by the object parameter early life change curve , r is the ratio of the previous detection time, which means the ratio of the time span of the time point corresponding to the parameter detection value relative to the object starting time point to the time length value obtained by the object parameter previous life change curve; according to different object time indicator functions , determine the range of parameter detection values ​​corresponding to the same detection time ratio, and form a mapping of the previous parameter value-time ratio range corresponding to the detection parameter ; Determine the maximum duration of the object's early life stage in the early life change curve of different object parameters, and mark it as the reference duration of the early parameters .

[0041] The extraction of characteristic information of detection parameters in the early stage of power supply use considers two aspects. One is that the value of the initial detection parameter changes greatly, and this change trend is within a certain range. If it exceeds this change range, it may be a situation where the detection parameter change is distorted due to power supply operation failure or other factors. On the other hand, the initial duration corresponding to different power supplies is different, and the change trend of the detection parameter is range-constrained. In order to make full use of big data for feature information extraction, it is necessary to determine the numerical value range of the detection parameter corresponding to different time points under the standard reference time length. Therefore, on the one hand, the average change of the detection parameter in the initial use is determined, and on the other hand, the detection parameter value is mapped by determining a reasonable same time ratio to conduct a comprehensive application analysis of the initial parameter data between different power supplies. Of course, in order to ensure that the real-time data can be used to compare with the currently collected characteristic information data in the later real-time detection process, it is necessary to select a standard reference for the initial duration, which also provides an important data reference for the subsequent real-time comparison and analysis. It can be understood that the detection parameters start from values ​​that are significantly different from the stable values ​​in the mid-term in the early stage. The larger the deviation of the parameter values ​​at each time point is from the starting time point of the previous month, in order to obtain accurate comparison results in the subsequent early real-time comparative analysis, selecting the maximum value of the initial duration as the basic standard quantity of the time ratio is conducive to grasping the impact of the deviation of the initial detection parameters on the detection results.

[0042] According to the parameter detection life stage information, and in combination with all object parameter life change curves corresponding to the detection parameters, parameter detection feature information is extracted to form parameter detection stage feature information, including: extracting curve data of different object parameter life change curves under the detection parameters in the corresponding object mid-term life stage to form object parameter mid-term life change curves; performing numerical range analysis of detection parameters on different object parameter mid-term life change curves to determine the parameter detection value range; performing detection value fluctuation feature analysis on different object parameter mid-term life change curves to form mid-term parameter fluctuation feature information.

[0043] For the extraction of detection parameter feature information in the mid-term stage of the power supply's service life, considering the value change characteristics of the detection parameter in the mid-term, two types of feature information need to be extracted. One is that the detection parameter has a certain volatility in the mid-term stage of the service life, and this volatility is reasonable and needs to be controlled, so the detection parameter needs to extract the corresponding value change range in the mid-term stage. The other is the fluctuation characteristics of the detection parameter in the mid-term stage. After all, the more the fluctuation amount and the higher the fluctuation frequency, the greater the noise of the detection parameter will be, which will have a greater impact on the stable operation of the power supply and the stable output of electric energy.

[0044] Performing numerical range analysis of detection parameters on different object parameter mid-life change curves to determine parameter detection ranges, including: extracting minimum parameter detection values ​​and maximum parameter detection values ​​of detection parameters corresponding to different object parameter mid-life change curves to establish object parameter detection value ranges; performing AND operations on all object parameter detection value ranges to form parameter detection value ranges corresponding to detection parameters. .

[0045] The range extraction of detection parameters in the mid-term stage of power supply use is mainly to obtain the range values ​​of different power supplies in the mid-term, and then merge the ranges to obtain a wider range value to provide a basic reference for subsequent real-time analysis and comparison.

[0046] The mid-term life span change curves of different object parameters are analyzed based on the fluctuation characteristics of the detection amount to form the mid-term parameter fluctuation characteristic information, including: setting the fluctuation amount judgment threshold, and determining the fluctuation amount of the mid-term life span change curves of different object parameters in the following manner: if within the continuous time change range, the parameter detection values ​​of the detection parameters have equal intervals, and the difference of the parameter detection values ​​in the time period between the equal parameter detection values ​​is not less than the fluctuation amount judgment threshold, then the time period between the equal parameter detection values ​​is marked as a fluctuation period; extracting the maximum value of the fluctuation period in the mid-term life span change curves of different object parameters And the shortest value of the object's mid-life stage in the mid-life change curve of different object parameters , determine the parameter fluctuation rate corresponding to the detection parameter ,in, ; Determine the maximum duration of the object's mid-life phase in the mid-life change curve of different object parameters , and combined with the previous parameter reference time , calibrated as the mid-term parameter reference duration ,in, .

[0047] The fluctuation characteristic information of the detection parameters in the mid-term stage of power use mainly considers the frequency of fluctuations. This frequency information needs to be obtained by extracting the corresponding number of fluctuations and the duration of the mid-term stage. The number of fluctuations is obtained through the characteristics of the fluctuations on the curve, that is, the fluctuations on the curve will have continuous value changes between the same values. The threshold for judging the fluctuation amount can be set according to the actual situation or determined based on big data analysis. For the duration of the mid-term stage, considering the fluctuation frequency as a limited boundary value, obtaining the shortest mid-term duration as a reference for extracting frequency characteristic information can maximize the extraction of this frequency limit. Of course, in real-time analysis, the real-time extracted data cannot determine the time position in the mid-term stage, so it is reasonable to provide the corresponding mid-term reference duration as the basis for judging the duration of the mid-term stage in subsequent real-time analysis.

[0048] The mid-term life span change curves of different object parameters are analyzed based on the fluctuation characteristics of the detected amount to form the mid-term parameter fluctuation characteristic information, including: extracting the curve data of the life span change curves of different object parameters under the detected parameters in the corresponding late life stage of the object to form the late life span change curves of the object parameters; for different late life span change curves of the object parameters, determining the corresponding average change amount of the late object parameters ,in, , The maximum parameter detection value obtained by the object parameter later life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The minimum parameter detection value obtained by the object parameter later life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, The time length value of the object parameter later life change curve obtained by the object parameter life change curve numbered i corresponding to the detection parameter numbered n; the average change of all later object parameters corresponding to the detection parameter , determine the average range of later parameters corresponding to the detection parameters For different object parameter late life change curves, the corresponding late object time indicator function is determined according to the parameter detection value obtained at each time point and the ratio of the time span of the time point corresponding to each parameter detection value relative to the object end time point to the time length value obtained by the object parameter late life change curve , k is the later detection time ratio, which means the ratio of the time span of the time point corresponding to the parameter detection value relative to the object starting time point to the time length value obtained by the object parameter life change curve; according to different later object time indicator functions , determine the range of parameter detection values ​​corresponding to the same detection time ratio, and form a later parameter value-time ratio range mapping corresponding to the detection parameter ; Determine the maximum value of the corresponding duration in the life span change curve of different object parameters, and mark it as the reference duration of the later parameters .

[0049] Similarly, if the value of the detection parameter changes greatly after the power is used, the detection parameter value based on the average change and time ratio can be used to analyze and judge the characteristic information. Here, for the mapping of parameter value time ratio, considering that the exact total power usage time cannot be known in real-time detection, the later parameter reference time is used as the time reference standard, and the starting time point is used as a reference to determine the time ratio.

[0050] S3: Acquire real-time detection data of the target power supply, and combine the detection feature comparison data to perform intelligent detection analysis to form a real-time detection analysis result.

[0051] Acquire the real-time detection data of the target power supply, and combine it with the detection feature comparison data to perform intelligent detection analysis to form real-time detection analysis results, including: obtaining the parameter target real-time detection change curve corresponding to different detection parameters of the target power supply; extracting the total real-time duration corresponding to the parameter target real-time detection change curve , and perform the following real-time detection analysis based on the parameter target real-time detection change curve and parameter detection stage characteristic information: ≤ , it is determined that the real-time monitoring corresponding to the detection parameters of the target power supply is in the early life stage of use, and the real-time average change of the early parameters is determined according to the parameter target real-time detection change curve And the corresponding real-time preliminary parameter value-time ratio mapping , where real-time pre-parameter value-time ratio mapping The time ratio is the time span of the time point corresponding to the real-time parameter detection value relative to the real-time start time point and the previous parameter reference time length The ratio of and judge: If ∈ , and real-time early parameter value-time ratio mapping The value at each time ratio belongs to the previous parameter value-time ratio range mapping If the value range of the corresponding time ratio is within the range, it is determined that the corresponding detection parameter is normal in real time, otherwise it is abnormal; < < , then it is determined that the real-time monitoring corresponding to the detection parameters of the target power supply is in the mid-life stage, and the extraction < Real-time parameter detection range of detection parameters within the time period , according to the fluctuation amount, the threshold is determined < The number of real-time fluctuation segments within the time period, and according to the number of real-time fluctuation segments and < The duration of the time period determines the real-time parameter volatility , and make a judgment: If ∈ ,and ≤ , then it is determined that the corresponding detection parameters are detected normally in real time, otherwise it is abnormal; when ≤ , it is determined that the real-time monitoring corresponding to the detection parameters of the target power supply is in the later life stage, and the real-time detection change curve of the parameter target is determined. ≤ The real-time average change of the later parameters corresponding to the time period And the corresponding real-time late parameter value-time ratio mapping , where real-time late parameter value-time ratio mapping The time ratio is the time span of the time point corresponding to the real-time parameter detection value relative to the real-time start time point and the reference time length of the later parameters The ratio of and judge: If ∈ , and real-time late parameter value-time ratio mapping The value at each time ratio belongs to the later parameter value-time ratio range mapping If the value range of the above corresponding time ratio is within the range, it is determined that the corresponding detection parameter real-time detection is normal, otherwise it is abnormal.

[0052] After extracting the characteristic information of the service life characteristics of the detection parameters, the target power supply can be tested and analyzed in real time to determine the quality of the power supply. Of course, the quality of the power supply is judged based on the normality of all detection parameters. Considering that different detection parameters may have different starting time points, it is necessary to combine the characteristic comparison data of different detection parameters for real-time analysis and judgment. First, the use stage of the target power supply is determined by the duration of real-time detection, and then the characteristic information of the corresponding stage is extracted for analysis and judgment. Only when all the restrictions are met at the same time can the detection parameters be considered normal. Otherwise, the corresponding detection parameters are abnormal, and it is inferred that the quality of the power supply is abnormal.

[0053] In summary, the power quality detection method based on artificial intelligence provided by the embodiment of the present invention has the following beneficial effects: This method obtains historical monitoring data of the same type of power supply over its service life cycle to establish a database for analyzing detection parameters based on service life characteristics, and establishes basic comparison data for different detection parameters by extracting and analyzing characteristic information of the detection parameters. At the same time, real-time detection information of the target power supply is obtained, and real-time automatic analysis and comparison with the basic comparison data is performed through information extraction based on service life characteristics to determine the quality of the power supply. On the one hand, based on historical monitoring data, detection parameter data for service life characteristics is extracted, which fully considers the changing characteristics of the detection parameters in the service life characteristics of the power supply, ensures the accuracy of the basic comparison data, and makes the data comparison more accurate and reasonable. On the other hand, combined with historical detection big data for automatic analysis and comparison, the efficiency of real-time analysis and comparison is greatly improved.

[0054] In the embodiments of the present application, "indication" may include direct indication and indirect indication, and may also include explicit indication and implicit indication. The information indicated by a certain information is called information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated, such as but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated. The information to be indicated can also be indirectly indicated by indicating other information, wherein there is an association between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while the other parts of the information to be indicated are known or agreed in advance. For example, the indication of specific information can also be achieved by means of the arrangement order of each piece of information agreed in advance (for example, specified by the protocol), thereby reducing the indication overhead to a certain extent. At the same time, the common parts of each piece of information can also be identified and indicated uniformly to reduce the indication overhead caused by indicating the same information separately.

[0055] In addition, the specific indication method may also be various existing indication methods, such as but not limited to the above-mentioned indication methods and various combinations thereof. The specific details of the various indication methods can refer to the prior art and will not be repeated herein. As can be seen from the above, for example, when it is necessary to indicate multiple information of the same type, different indication methods may be used for different information. In the specific implementation process, the desired indication method can be selected according to specific needs. The embodiment of the present application does not limit the selected indication method. In this way, the indication method involved in the embodiment of the present application should be understood to cover various methods that can enable the party to be indicated to obtain the information to be indicated.

[0056] It should be understood that the information to be indicated can be sent as a whole, or divided into multiple sub-information and sent separately, and the sending period and / or sending time of these sub-information can be the same or different. The specific sending method is not limited in the embodiment of the present application. Among them, the sending period and / or sending time of these sub-information can be pre-defined, for example, pre-defined according to a protocol, or can be configured by the sending end device by sending configuration information to the receiving end device.

[0057] "Pre-definition" or "pre-configuration" can be implemented by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in the device, and the embodiments of the present application do not limit the specific implementation method. Among them, "saving" can mean saving in one or more memories. The one or more memories can be set separately or integrated in an encoder or decoder, a processor, or a communication device. The one or more memories can also be partially set separately and partially integrated in a decoder, a processor, or a communication device. The type of memory can be any form of storage medium, which is not limited by the embodiments of the present application.

[0058] The "protocol" involved in the embodiments of the present application may refer to a protocol family in the communication field, a standard protocol with a similar protocol family frame structure, or a related protocol used in future communication systems, and the embodiments of the present application do not make specific limitations on this.

[0059] In the embodiments of the present application, descriptions such as "when...", "in the case of...", "if" and "if" all mean that the device will make corresponding processing under certain objective circumstances. It does not limit the time, nor does it require the device to have a judgment action when implementing it, nor does it mean that there are other limitations.

[0060] In the description of the embodiments of the present application, unless otherwise specified, " / " indicates that the objects associated before and after are in an "or" relationship. For example, A / B can represent A or B; "and / or" in the embodiments of the present application is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. These three situations, where A and B can be singular or plural. In addition, in the description of the embodiments of the present application, unless otherwise specified, "multiple" refers to two or more than two. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple. In addition, in order to facilitate the clear description of the technical solution of the embodiments of the present application, in the embodiments of the present application, the words "first" and "second" are used to distinguish the same or similar items with basically the same functions and effects. Those skilled in the art will appreciate that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit the difference. At the same time, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0061] It should be understood that the processor in the embodiments of the present application may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0062] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0063] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0064] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0065] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0066] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0068] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0069] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0070] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0071] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0072] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0073] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A power quality detection method based on artificial intelligence, characterized in that: include: Obtain historical quality test data of the same type of power supply and establish a quality test database for the same type of power supply; Extract historical monitoring information of different detection parameters in the same type of power quality detection database, extract feature information of the detection parameters, and form detection feature comparison data; The real-time detection data of the target power source is obtained, and combined with the detection feature comparison data, intelligent detection analysis is performed to form a real-time detection analysis result.

2. The power quality detection method based on artificial intelligence according to claim 1 is characterized in that: The extracting of historical monitoring information of different detection parameters in the same type of power quality detection database, extracting feature information of the detection parameters, and forming detection feature comparison data includes: For the historical quality detection data of different power supplies of the same type, obtaining an object parameter life change curve formed by different detection parameters in sequence of the time dimension; For the same detection parameters, cluster the object parameter life change curves corresponding to different power supplies, and extract feature information based on the time dimension sequence to form parameter life feature information; The parameter life characteristic information corresponding to different detection parameters is collected to form the detection characteristic comparison data.

3. The power quality detection method based on artificial intelligence according to claim 2 is characterized in that: The method of clustering the object parameter life change curves corresponding to different power supplies for the same detection parameters and extracting feature information based on the time dimension sequence to form parameter life feature information includes: The method of clustering the object parameter life change curves corresponding to different power supplies for the same detection parameters and extracting feature information based on the time dimension sequence to form parameter life feature information includes: For the same detection parameters, obtain the life change curves of all the object parameters, divide the service life stages based on the parameter level changes, and form parameter detection life stage information; Detecting life stage information according to the parameters, and extracting parameter detection feature information in combination with all the object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information; The parameter detection phase characteristic information corresponding to different detection parameters is collected to form the parameter life characteristic information.

4. The power quality detection method based on artificial intelligence according to claim 3 is characterized in that: For the same detection parameter, obtaining the life change curves of all the object parameters and dividing the service life stages based on the parameter level changes to form parameter detection life stage information includes: For different object parameter life change curves under the detection parameters, determining the maximum object parameter value; Set the parameter value adjustment direction, starting with the maximum parameter value of the object, gradually and continuously adjust the parameter value, and determine the stage parameter value according to the following method: When the obtained parameter value corresponds to a time point greater than or equal to 2 and less than or equal to 4 on the object parameter life change curve, and the number of time points corresponding to the parameter values ​​continuously obtained in the parameter value adjustment direction on the object parameter life change curve increases in a jump, the obtained parameter value is determined as the stage parameter value; Determine the earliest time point corresponding to the stage parameter value on the object parameter life change curve as the object use early time point; Determine the latest time point corresponding to the stage parameter value on the object parameter life change curve as the object use later time point; Determine the period of time between the object start time point of the object parameter life change curve and the object early use time point as the object early use life stage; Determine the time period between the object use early time point and the object use late time point of the object parameter life change curve as the object use mid-life stage; The time period of the stage parameter value between the object late use time point and the object end time point of the object parameter life change curve is determined as the object late use life stage.

5. The power quality detection method based on artificial intelligence according to claim 4 is characterized in that: The detecting life stage information according to the parameters and extracting parameter detection feature information in combination with all the object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information includes: Extracting curve data of the life span change curves of the object parameters under the detection parameters at the corresponding early life span stage of the object to form an early life span change curve of the object parameters; For different early life change curves of the object parameters, determine the corresponding early average change of the object parameters ,in, , n represents the number of different detection parameters, i represents the number of different object parameter life change curves corresponding to the detection parameter numbered n, represents the maximum parameter detection value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, represents the minimum parameter detection value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, Indicates the time length value obtained by the object parameter early life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n; According to the average change of all the previous object parameters corresponding to the detection parameters , determine the average variation range of the previous parameters corresponding to the detection parameters ; For different object parameter early life change curves, the corresponding early object time indication function is determined according to the parameter detection value obtained at each time point and the ratio of the time span of the time point corresponding to each parameter detection value relative to the object starting time point to the time length value obtained by the object parameter early life change curve. , r is the early detection time ratio, which means the ratio of the time span of the time point corresponding to the parameter detection value relative to the starting time point of the object to the time length value obtained by the early life span change curve of the object parameter; According to different object time indication functions , determine the range of the parameter detection value corresponding to the same detection time ratio, and form a mapping of the previous parameter value-time ratio range corresponding to the detection parameter ; Determine the maximum duration of the object's early life stage in different early life change curves of the object parameters, and mark it as the early parameter reference duration .

6. The power quality detection method based on artificial intelligence according to claim 5 is characterized in that: The detecting life stage information according to the parameters and extracting parameter detection feature information in combination with all the object parameter life change curves corresponding to the detection parameters to form parameter detection stage feature information includes: Extracting curve data of the object parameter life change curves under the detection parameters at the corresponding mid-life stage of the object to form an object parameter mid-life change curve; Performing numerical range analysis of the detection parameters on different mid-life change curves of the object parameters to determine the parameter detection range; The mid-term life span change curves of different object parameters are analyzed based on the detected quantity fluctuation characteristics to form mid-term parameter fluctuation characteristic information.

7. The power quality detection method based on artificial intelligence according to claim 6 is characterized in that: The step of performing a numerical range analysis of the detection parameters on different mid-life change curves of the object parameters to determine the parameter detection range includes: Extracting the minimum parameter detection value and the maximum parameter detection value of the detection parameters corresponding to the mid-life change curves of the different object parameters, and establishing the object parameter detection value range; Perform a AND operation on all the object parameter detection value ranges to form the parameter detection value range corresponding to the detection parameter .

8. The power quality detection method based on artificial intelligence according to claim 7 is characterized in that: The mid-term life span change curves of different object parameters are analyzed based on the detected amount fluctuation characteristics to form mid-term parameter fluctuation characteristic information, including: The fluctuation amount judgment threshold is set, and the fluctuation amount of different mid-life change curves of the object parameters is determined in the following manner: If within the continuous time variation range, the parameter detection values ​​of the detection parameter are at equal intervals, and the difference of the parameter detection values ​​in the time period between the equal parameter detection values ​​is not less than the fluctuation amount judgment threshold, then the time period between the equal parameter detection values ​​is marked as a fluctuation period; Extract the maximum value of the fluctuation period in the mid-term life change curve of different object parameters and the shortest value of the duration of the mid-life stage of the object in the mid-life change curve of different object parameters , determine the parameter fluctuation rate corresponding to the detection parameter ,in, ; Determine the maximum duration of the mid-life stage of the object in different mid-life change curves of the object parameters , and combined with the reference duration of the previous parameters , calibrated as the mid-term parameter reference duration ,in, .

9. The power quality detection method based on artificial intelligence according to claim 8 is characterized in that: The mid-term life span change curves of different object parameters are analyzed based on the detected amount fluctuation characteristics to form mid-term parameter fluctuation characteristic information, including: Extracting curve data of the life span variation curves of the object parameters under the detection parameters at the corresponding later life span stage of the object to form a later life span variation curve of the object parameters; For different object parameter later life change curves, determine the corresponding later average change of object parameters ,in, , represents the maximum parameter detection value obtained by the object parameter later life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, represents the minimum parameter detection value obtained by the object parameter later life change curve of the object parameter life change curve numbered i corresponding to the detection parameter numbered n, Indicates the time length value obtained by the object parameter later life span change curve of the object parameter life span change curve numbered i corresponding to the detection parameter numbered n; According to the average change of all the later object parameters corresponding to the detection parameters , determine the average variation range of the later parameters corresponding to the detection parameters ; For different object parameter late life change curves, the corresponding late object time indication function is determined according to the parameter detection value obtained at each time point and the ratio of the time span of the time point corresponding to each parameter detection value relative to the object end time point to the time length value obtained by the object parameter late life change curve. , k is the late detection time ratio, which means the ratio of the time span of the time point corresponding to the parameter detection value relative to the starting time point of the object to the time length value obtained by the parameter life change curve of the object; According to different later object time indication functions , determine the range of the parameter detection value corresponding to the same detection time ratio, and form a later parameter value-time ratio range mapping corresponding to the detection parameter ; Determine the maximum value of the duration corresponding to the life span change curve of different object parameters, and mark it as the reference duration of the later parameters .

10. The power quality detection method based on artificial intelligence according to claim 9 is characterized in that: The real-time detection data of the target power source is obtained, and combined with the detection feature comparison data, intelligent detection analysis is performed to form a real-time detection analysis result, including: Obtaining a parameter target real-time detection change curve corresponding to different detection parameters of the target power source; Extract the total real-time duration corresponding to the parameter target real-time detection change curve , and perform the following real-time detection analysis based on the parameter target real-time detection change curve and the parameter detection stage characteristic information: when ≤ , it is determined that the real-time monitoring corresponding to the detection parameter of the target power source is in the early life stage of use, and the real-time average change of the early parameter is determined according to the real-time detection change curve of the parameter target And the corresponding real-time preliminary parameter value-time ratio mapping , wherein the real-time previous parameter value-time ratio mapping The time ratio is the time span of the time point corresponding to the real-time parameter detection value relative to the real-time starting time point and the reference time length of the previous parameter The ratio of and judge: like ∈ , and the real-time previous parameter value-time ratio mapping The value at each time ratio belongs to the previous parameter value-time ratio range mapping If the value range of the above corresponding time ratio is within the range, it is determined that the corresponding detection parameter real-time detection is normal, otherwise it is abnormal; when < < , it is determined that the real-time monitoring corresponding to the detection parameter of the target power source is in the mid-life stage, and the < Real-time parameter detection range of detection parameters within the time period , according to the fluctuation amount judgment threshold, determine < The number of real-time fluctuation segments within the time period, and according to the number of real-time fluctuation segments and < The duration of the time period determines the real-time parameter volatility , and make a judgment: like ∈ ,and ≤ , then it is determined that the corresponding detection parameter is normal in real time, otherwise it is abnormal; when ≤ , it is determined that the real-time monitoring corresponding to the detection parameter of the target power source is in the late life stage of use, and the real-time detection change curve of the parameter target is determined ≤ The real-time average change of the later parameters corresponding to the time period And the corresponding real-time late parameter value-time ratio mapping , wherein the real-time late parameter value-time ratio mapping The time ratio is the time span of the time point corresponding to the real-time parameter detection value relative to the real-time starting time point and the reference time length of the later parameter The ratio of and judge: like ∈ , and the real-time late parameter value-time ratio mapping The value at each time ratio belongs to the latter parameter value-time ratio range mapping If the value range of the above corresponding time ratio is within the range, it is determined that the corresponding detection parameter real-time detection is normal, otherwise it is abnormal.

Images