Power grid asset health degree and operation cost correlation analysis method and system

Abstract

The invention provides a power grid asset health degree and operation cost correlation analysis method and system. The method comprises the steps of obtaining a power grid asset assessment sequence data set for analysis processing, generating a historical power grid asset portrait, performing feature parameter extraction and model training, obtaining an asset health degree assessment model, and obtaining a real-time asset health degree through model processing in combination with a real-time power grid asset portrait. Inputting a pre-trained health degree and potential operation cost mapping model for processing to obtain potential operation real-time prediction cost, and comparing the potential operation real-time prediction cost with preset potential operation permitted cost to obtain an operation decision strategy; according to the method, the correlation analysis of the health degree and the operation cost of the power grid assets is realized through the construction of the power grid asset portrait through multi-source data, the quantitative calculation of the health degree, the dual-model collaborative analysis, the full-life-cycle cost decomposition and prediction and the calculation of the asset investment operation return rate; therefore, intelligent association analysis from state monitoring to operation cost management is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid asset health management and cost control, in particular to a power grid asset health degree and operation cost correlation analysis method and system. BACKGROUND

[0002] The operation of multiple core assets in a power grid enterprise determines the operation and maintenance cost and operation efficiency of the enterprise. Traditional technology separates the management of asset health state and asset cost management, and the health state cannot be directly mapped and correlated to financial cost management, nor can it accurately predict the specific impact of the asset operation state on future costs. Moreover, cost management analysis is mostly based on maintenance fees or direct losses after the failure occurs, and then specific costs are analyzed, which cannot be predicted in advance before the failure occurs. At the same time, the management and decision-making of assets lack quantitative standards and targeted decision-making strategies. Therefore, there is an urgent need for an intelligent analysis method that can directly quantify the correlation between the health state of power grid assets and future potential operation costs.

[0003] In view of the above problems, an effective technical solution is currently needed. SUMMARY

[0004] The purpose of the present application is to provide a power grid asset health degree and operation cost correlation analysis method and system, which can construct a power grid asset portrait, health degree quantitative calculation, double model collaborative analysis, full life cycle cost decomposition and prediction, and asset investment and operation return on investment calculation through multi-source data, thereby realizing power grid asset health degree and operation cost correlation analysis, and realizing intelligent correlation analysis from state monitoring to operation cost management.

[0005] In a first aspect, the present application provides a power grid asset health degree and operation cost correlation analysis method, comprising the following steps: Obtain a power grid asset evaluation sequence data set in a historical preset time period, analyze and process the power grid asset evaluation sequence data set, and generate a historical power grid asset portrait; According to the feature parameter extraction of the historical power grid asset portrait, the asset health degree historical evaluation feature parameter is obtained, the asset health degree historical evaluation feature parameter is trained, and the asset health degree evaluation model is obtained; Obtain a real-time power grid asset portrait, extract feature parameters according to the real-time power grid asset portrait, obtain asset health degree real-time evaluation feature parameters, and input the asset health degree evaluation model for processing to obtain real-time asset health degree; Obtain historical asset health degree and corresponding historical operation evaluation cost in a preset time period, train the historical asset health degree and corresponding historical operation evaluation cost, and obtain a health degree and potential operation cost mapping model; The real-time asset health degree input is analyzed by the health degree and potential operation cost mapping model to obtain a potential operation real-time prediction cost, and compared with a preset potential operation permitted cost to obtain an operation decision strategy.

[0006] Optionally, in the power grid asset health degree and operation cost correlation analysis method described in the application, the historical preset time period power grid asset evaluation sequence data set is obtained, and the historical power grid asset portrait is generated by analyzing and processing the power grid asset evaluation sequence data set, including: The historical preset time period power grid asset evaluation sequence data set is obtained, including historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data; The historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data are data preprocessed, and the historical power grid asset portrait is generated in combination with a preset asset ID.

[0007] Optionally, in the power grid asset health degree and operation cost correlation analysis method described in the application, the historical power grid asset portrait is used for feature parameter extraction to obtain asset health degree historical evaluation feature parameters, and the asset health degree historical evaluation feature parameters are used for training to obtain an asset health degree evaluation model, including: The historical power grid asset portrait is used for feature parameter extraction to obtain asset health degree historical evaluation feature parameters, including running state feature parameters, running trend feature parameters, running environment influence feature parameters and running health feature parameters; The initial asset health degree evaluation model is trained according to the running state feature parameters, running trend feature parameters, running environment influence feature parameters, running health feature parameters and corresponding preset historical asset health degree to obtain an asset health degree evaluation model.

[0008] Optionally, in the power grid asset health degree and operation cost correlation analysis method described in the application, the real-time power grid asset portrait is obtained, the real-time power grid asset portrait is used for feature parameter extraction to obtain asset health degree real-time evaluation feature parameters, and the asset health degree real-time evaluation feature parameters are input into the asset health degree evaluation model for processing to obtain a real-time asset health degree, including: The real-time power grid asset portrait is obtained, the real-time power grid asset portrait is used for feature parameter extraction to obtain asset health degree real-time evaluation feature parameters, including real-time running state feature parameters, real-time running trend feature parameters, real-time running environment influence feature parameters and real-time running health feature parameters; Input the real-time running state feature parameters, real-time running trend feature parameters, real-time running environment influence feature parameters and real-time running health feature parameters into the asset health degree evaluation model for processing to obtain a real-time asset health degree.

[0009] Optionally, in the power grid asset health degree and operation cost correlation analysis method described in the present application, the historical asset health degrees and corresponding historical operation evaluation costs in a preset time period are obtained, and the health degree and potential operation cost mapping model is obtained by training according to the historical asset health degrees and corresponding historical operation evaluation costs, including: The historical asset health degrees and corresponding historical operation evaluation costs in a preset time period are obtained, wherein the historical operation evaluation costs include historical planned maintenance costs, historical failure repair costs and historical risk prediction costs; The historical planned maintenance costs, historical failure repair costs and historical risk prediction costs are processed by weighted summation to obtain historical potential operation evaluation costs; The initial health degree and potential operation cost mapping model is trained according to the historical asset health degrees and corresponding historical potential operation evaluation costs to obtain the health degree and potential operation cost mapping model.

[0010] Optionally, in the power grid asset health degree and operation cost correlation analysis method described in the present application, the real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation real-time prediction cost, and the potential operation real-time prediction cost is compared with a preset potential operation permitted cost to obtain an operation decision strategy, including: The real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation real-time prediction cost; The potential operation real-time prediction cost is compared with a preset potential operation permitted cost to obtain a real-time operation cost deviation rate; The real-time operation cost deviation rate is compared with a preset cost operation deviation rate threshold value to obtain a corresponding operation decision strategy; If the real-time operation cost deviation rate is less than or equal to the preset cost operation deviation rate threshold value, the operation decision strategy is to maintain the asset operation status; If the real-time operation cost deviation rate is greater than the preset cost operation deviation rate threshold value, the operation decision strategy is to implement asset maintenance.

[0011] Optionally, in the power grid asset health degree and operation cost correlation analysis method described in the present application, it further includes: Obtaining a maintenance scheme and maintenance investment data for implementing asset maintenance; According to the maintenance scheme, the predicted asset health degree is obtained by analyzing and processing a preset asset operation digital twin model; The predicted asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation correction predicted cost; According to the potential operation correction predicted cost and the potential operation real-time predicted cost in combination with the maintenance investment data, an asset investment operation return on investment is obtained; The predicted asset health degree is compared with the real-time asset health degree to obtain an asset health degree improvement rate; The asset health degree improvement rate and the asset investment operation return on investment are respectively compared with corresponding preset threshold values to obtain an asset health degree improvement evaluation state and an asset investment operation return on investment evaluation state; The asset health degree improvement evaluation state and the asset investment operation return on investment evaluation state are calculated; If it is a qualified state, asset maintenance is implemented according to the maintenance scheme; If it is an unqualified state, an early warning response is output In a second aspect, the present application provides an electric power grid asset health degree and operation cost correlation analysis system, which comprises a memory and a processor, the memory comprising a program of an electric power grid asset health degree and operation cost correlation analysis method, the program of the electric power grid asset health degree and operation cost correlation analysis method being executed by the processor to realize the following steps: A historical preset time period electric power grid asset evaluation sequence data set is obtained, and a historical electric power grid asset portrait is generated by analyzing and processing the electric power grid asset evaluation sequence data set; According to the historical electric power grid asset portrait, asset health degree historical evaluation feature parameters are obtained by extracting feature parameters, and an asset health degree evaluation model is obtained by training according to the asset health degree historical evaluation feature parameters; A real-time electric power grid asset portrait is obtained, asset health degree real-time evaluation feature parameters are obtained by extracting feature parameters according to the real-time electric power grid asset portrait, and the asset health degree real-time evaluation feature parameters are input into the asset health degree evaluation model for processing to obtain a real-time asset health degree; A historical asset health degree and a corresponding historical operation evaluation cost in a preset time period are obtained, and a health degree and potential operation cost mapping model is obtained by training according to the historical asset health degree and the corresponding historical operation evaluation cost; The real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation real-time predicted cost, and the potential operation real-time predicted cost is compared with a preset potential operation permitted cost to obtain an operation decision strategy.

[0012] Optionally, in the power grid asset health degree and operation cost correlation analysis system provided in the application, the historical power grid asset evaluation sequence data set of the preset time period is obtained, the historical power grid asset portrait is generated by analyzing and processing the power grid asset evaluation sequence data set, and the historical power grid asset portrait comprises: The historical power grid asset evaluation sequence data set of the preset time period comprises historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data. The historical monitoring record sequence data, the historical asset account record sequence data, the historical inspection record sequence data, the historical running environment record sequence data and the historical running condition record sequence data are preprocessed, and the historical power grid asset portrait is generated in combination with the preset asset ID.

[0013] Optionally, in the power grid asset health degree and operation cost correlation analysis system provided in the application, the historical power grid asset portrait is used for feature parameter extraction to obtain asset health degree historical evaluation feature parameters, and the asset health degree historical evaluation feature parameters are used for training to obtain an asset health degree evaluation model, and the historical power grid asset portrait comprises: The historical power grid asset portrait is used for feature parameter extraction to obtain asset health degree historical evaluation feature parameters, including running state feature parameters, running trend feature parameters, running environment influence feature parameters and running health feature parameters. The initial asset health degree evaluation model is trained according to the running state feature parameters, the running trend feature parameters, the running environment influence feature parameters, the running health feature parameters and the corresponding preset historical asset health degree to obtain the asset health degree evaluation model.

[0014] As can be seen from the above, the power grid asset health degree and operation cost correlation analysis method and system provided in the application realize the power grid asset health degree and operation cost correlation analysis through multi-source data construction of a power grid asset portrait, health degree quantitative calculation, double-model collaborative analysis, full-life cycle cost decomposition and prediction and asset investment operation return rate calculation, and further realize intelligent correlation analysis from state monitoring to operation cost management.

[0015] Other features and advantages of the application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the application. The objects and other advantages of the application will be realized and attained by means of the instrumentalities particularly pointed out in the written description and claims hereof. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly introduced as follows. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0017] Figure 1 The flowchart of the power grid asset health degree and operation cost correlation analysis method provided by the embodiments of the present application; Figure 2 The flowchart of the generation of historical power grid asset portrait of the power grid asset health degree and operation cost correlation analysis method provided by the embodiments of the present application; Figure 3 The flowchart of the obtaining of asset health degree evaluation model of the power grid asset health degree and operation cost correlation analysis method provided by the embodiments of the present application; Figure 4 The flowchart of the obtaining of health degree and potential operation cost mapping model of the power grid asset health degree and operation cost correlation analysis method provided by the embodiments of the present application; Figure 5 The high-level flowchart of the power grid asset health degree and operation cost correlation analysis method provided by the embodiments of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of the present application.

[0019] It should be noted that similar reference numerals and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0020] Please refer to Figure 1 , Figure 1is a flowchart of a power grid asset health degree and operation cost correlation analysis method in some embodiments of the present application. The power grid asset health degree and operation cost correlation analysis method is used in terminal equipment such as computers, mobile phones and the like. The power grid asset health degree and operation cost correlation analysis method comprises the following steps: S11, acquiring a power grid asset evaluation sequence data set in a historical preset time period, analyzing and processing the power grid asset evaluation sequence data set to generate a historical power grid asset portrait; S12, extracting feature parameters according to the historical power grid asset portrait to obtain asset health degree historical evaluation feature parameters, training according to the asset health degree historical evaluation feature parameters to obtain an asset health degree evaluation model; S13, acquiring a real-time power grid asset portrait, extracting feature parameters according to the real-time power grid asset portrait to obtain asset health degree real-time evaluation feature parameters, and inputting the asset health degree real-time evaluation feature parameters into the asset health degree evaluation model for processing to obtain a real-time asset health degree; S14, acquiring historical asset health degrees and corresponding historical operation evaluation costs in a preset time period, training according to the historical asset health degrees and the corresponding historical operation evaluation costs to obtain a health degree and potential operation cost mapping model; S15, inputting the real-time asset health degree into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation real-time prediction cost, and comparing the potential operation real-time prediction cost with a preset potential operation permitted cost to obtain an operation decision strategy.

[0021] It should be noted that, in order to accurately quantify the operation cost according to the health state of the power grid asset, first, a multi-source power grid asset evaluation sequence data set including physical operation state, environmental condition and operation condition is acquired to construct an accurate power grid asset portrait, then a feature parameter extraction is performed to pre-train an asset health degree evaluation model to accurately quantify the real-time asset health degree, at the same time, the operation cost in the whole life cycle is divided into planned maintenance cost, fault repair cost and risk prediction cost to pre-train a health degree and potential operation cost mapping model to realize dynamic correlation analysis of the asset health state and the cost, finally, the potential operation real-time prediction cost is compared with the preset potential operation permitted cost, and the corresponding operation decision strategy is generated according to the deviation, and the adaptability of the operation decision strategy is evaluated.

[0022] Please refer to Figure 2 , Figure 2 is a flowchart of a power grid asset health degree and operation cost correlation analysis method in some embodiments of the present application. According to the embodiment of the present application, the acquisition of the power grid asset evaluation sequence data set in the historical preset time period and the analysis and processing of the power grid asset evaluation sequence data set to generate a historical power grid asset portrait comprises: S21, acquire a power grid asset evaluation sequence data set in a historical preset time period, including historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data; S22, perform data preprocessing on the historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data, and generate a historical power grid asset portrait in combination with a preset asset ID.

[0023] It should be noted that, in order to comprehensively evaluate the operation of the power grid asset, firstly, a plurality of power grid asset evaluation data in a historical preset time period is collected to form a power grid asset evaluation sequence data set, wherein the historical monitoring record sequence data includes oil chromatogram monitoring data, partial discharge monitoring data, bushing dielectric loss monitoring data, temperature monitoring data, load current monitoring data and voltage waveform monitoring data, the historical asset account record sequence data includes equipment model feature data, installation date, manufacturer feature data, rated service life and asset technical transformation record data, the historical inspection record sequence data includes inspection log record data and preventive test record data, the historical running environment record sequence data includes running environment temperature average, running environment humidity average and asset pollution level, and the historical running condition record sequence data is running abnormal record data, such as short-circuit power impact record data. Then, a comprehensive and accurate historical power grid asset portrait is constructed based on the collected multi-source data of a plurality of time periods and the preset asset ID representing the asset identity.

[0024] Please refer to Figure 3 , Figure 3 is a flowchart of a method for obtaining an asset health degree evaluation model in some embodiments of the present application. According to the embodiment of the present application, the feature parameter extraction based on the historical power grid asset portrait obtains asset health degree historical evaluation feature parameters, and the asset health degree historical evaluation feature parameters are used for training to obtain an asset health degree evaluation model, which includes: S31, feature parameter extraction based on the historical power grid asset portrait to obtain asset health degree historical evaluation feature parameters, including running state feature parameters, running trend feature parameters, running environment influence feature parameters and running health feature parameters; S32, training of an initial asset health degree evaluation model according to the running state feature parameters, running trend feature parameters, running environment influence feature parameters, running health feature parameters and corresponding preset historical asset health degree to obtain an asset health degree evaluation model.

[0025] It should be noted that, in order to realize accurate quantification of asset health condition based on asset operation state, firstly, the characteristic parameters affecting asset health degree are extracted according to the constructed historical power grid asset portrait, wherein the operation state characteristic parameters are used to quantify the real-time physical state of the power grid asset, such as real-time content of characteristic gas, real-time gas production rate of characteristic gas, load overload multiple, load overload time length, the operation trend characteristic parameters are used to represent the change trend of the characteristic parameters in a preset time period, such as the amplitude change slope of the partial discharge signal and the signal duty cycle change trend data, the operation environment influence characteristic parameters are used to quantify the damage of external environment to the asset health degree, such as the pollution level, the lightning insulation damage risk value (obtained by mapping the number of lightning strikes), the metal component corrosion acceleration coefficient (obtained by weighting and correcting the preset initial corrosion coefficient according to the environmental humidity and temperature), the operation health characteristic parameters are used to represent the failure, maintenance and impact of the asset in the whole life cycle, such as the number of short-circuit impact values, and the specific characteristic parameters are preset and set by the person skilled in the art according to different assets, and can be dynamically adjusted, then, based on the extracted operation state characteristic parameters, operation trend characteristic parameters, operation environment influence characteristic parameters, operation health characteristic parameters and corresponding preset historical asset health degree, the initial asset health degree evaluation model is trained to obtain the asset health degree evaluation model.

[0026] According to the embodiment of the present application, the real-time power grid asset portrait is obtained, the characteristic parameters are extracted according to the real-time power grid asset portrait, the asset health degree real-time evaluation characteristic parameters are obtained, and the asset health degree evaluation model is input for processing to obtain the real-time asset health degree, comprising: The real-time power grid asset portrait is obtained, the characteristic parameters are extracted according to the real-time power grid asset portrait, the asset health degree real-time evaluation characteristic parameters are obtained, including real-time operation state characteristic parameters, real-time operation trend characteristic parameters, real-time operation environment influence characteristic parameters and real-time operation health characteristic parameters; The real-time operation state characteristic parameters, real-time operation trend characteristic parameters, real-time operation environment influence characteristic parameters and real-time operation health characteristic parameters are input into the asset health degree evaluation model for processing to obtain the real-time asset health degree.

[0027] It should be noted that, in order to evaluate the real-time health condition of the asset, the real-time power grid asset portrait is extracted for characteristic parameter extraction, and the pre-trained asset health degree evaluation model is analyzed and processed to obtain the real-time asset health degree.

[0028] Please refer to Figure 4 , Figure 4is a flowchart of an asset health degree and operation cost correlation analysis method in some embodiments of the present application. According to an embodiment of the present application, the historical asset health degree and the corresponding historical operation evaluation cost in a preset time period are obtained, the historical asset health degree and the corresponding historical operation evaluation cost are trained to obtain a health degree and potential operation cost mapping model, including: S41, obtaining historical asset health degree and corresponding historical operation evaluation cost in a preset time period, wherein the historical operation evaluation cost includes historical planned maintenance cost, historical failure repair cost and historical risk prediction cost; S42, the historical planned maintenance cost, the historical failure repair cost and the historical risk prediction cost are weighted and summed to obtain the historical potential operation evaluation cost; S43, training the initial health degree and potential operation cost mapping model according to the historical asset health degree and the corresponding historical potential operation evaluation cost to obtain the health degree and potential operation cost mapping model.

[0029] It should be noted that in order to realize the dynamic correlation between operation cost and asset health degree, the asset life cycle operation cost is divided into planned maintenance cost, failure repair cost and risk prediction cost in the embodiments of the present application. The planned maintenance cost refers to the cost generated by preventive maintenance to prevent asset failure. The failure repair cost refers to the direct and indirect costs generated by maintenance after failure. The risk prediction cost refers to the potential loss caused by the chain reaction of single asset failure. According to the preset asset ID, the asset type is determined, and then the corresponding weight value is determined. The historical potential operation evaluation cost is obtained by weighted sum processing. Then, based on the determined historical asset health degree and the corresponding historical potential operation evaluation cost, the initial health degree and potential operation cost mapping model is trained to obtain the health degree and potential operation cost mapping model.

[0030] According to an embodiment of the present application, the real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain potential operation real-time prediction cost, and compared with a preset potential operation permitted cost to obtain an operation decision strategy, including: The real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain potential operation real-time prediction cost; The potential operation real-time prediction cost is compared with a preset potential operation permitted cost to obtain a real-time operation cost deviation rate; The real-time operation cost deviation rate is compared with a preset cost operation deviation rate threshold to obtain a corresponding operation decision strategy; If the real-time operation cost deviation rate is less than or equal to the preset cost operation deviation rate threshold, the operation decision strategy is to maintain the asset operation status; If the real-time operation cost deviation rate is greater than the preset cost operation deviation rate threshold, the operation decision strategy is to implement asset maintenance.

[0031] It should be noted that the potential operation real-time predicted cost is obtained by mapping processing based on the determined real-time asset health degree through the pre-trained model, and is compared with the preset potential operation permitted cost to obtain the real-time operation cost deviation rate. The real-time operation cost deviation rate refers to the ratio of the potential operation real-time predicted cost minus the preset potential operation permitted cost to the preset potential operation permitted cost. If it is negative, that is, the potential operation real-time predicted cost is less than the preset potential operation permitted cost, the real-time operation cost deviation rate is recorded as 0. Then, through threshold comparison, it is determined that the operation strategy of the asset is to maintain the asset operation status or implement asset maintenance.

[0032] According to the embodiment of the present application, further comprising: Obtaining a maintenance scheme and maintenance investment data for implementing asset maintenance; Obtaining a maintenance scheme and maintenance investment data for implementing asset maintenance; Obtaining a maintenance scheme and maintenance investment data for implementing asset maintenance; According to the potential operation corrected predicted cost and the potential operation real-time predicted cost, the maintenance investment data is processed to obtain the asset investment operation return rate; Comparing the predicted asset health degree with the real-time asset health degree to obtain the asset health degree improvement rate; Comparing the predicted asset health degree with the real-time asset health degree to obtain the asset health degree improvement rate; Comparing the predicted asset health degree with the real-time asset health degree to obtain the asset health degree improvement rate; If it is a qualified state, asset maintenance is implemented according to the maintenance scheme; If it is an unqualified state, an early warning response is output.

[0033] It should be noted that after determining to implement maintenance, in order to determine whether the maintenance scheme achieves the expectation, first, based on the maintenance scheme, the preset asset operation digital twin model is simulated to determine the predicted asset health degree after implementation of the maintenance scheme, then the potential operation correction predicted cost is obtained by analyzing and processing the health degree and the potential operation cost mapping model, the asset investment operation return rate is obtained by processing the potential operation correction predicted cost and the potential operation real-time predicted cost in combination with the maintenance investment data, the asset investment operation return rate refers to the potential operation real-time predicted cost minus the potential operation correction predicted cost divided by the maintenance investment data, if it is negative, it is directly recorded as 0, at the same time, the asset health degree improvement rate is evaluated, that is, the predicted asset health degree minus the real-time asset health degree and the real-time asset health degree, if it is negative, it is directly recorded as 0, finally, by comparing the threshold, if it is greater than the corresponding threshold, the asset health degree improvement evaluation state and the asset investment operation return evaluation state are determined to be qualified state, otherwise, it is unqualified state, if the asset health degree improvement evaluation state and the asset investment operation return evaluation state are both qualified state, it is finally determined that the maintenance scheme is effective, and the asset maintenance is implemented, otherwise, any one is unqualified, then an early warning is output, whether to continue to implement the maintenance scheme is finally determined by the technical personnel.

[0034] Please refer to Figure 5 , Figure 5 is a high-level flowchart of an electric grid asset health degree and operation cost correlation analysis method in some embodiments of the present application.

[0035] It is worth mentioning that according to the embodiment of the present application, further comprising: According to the real-time asset health degree, a preset health degree and failure probability mapping table is queried to obtain the corresponding real-time failure probability; According to the preset asset ID, the failure impact loss load and the corresponding predicted unit time power failure loss and predicted power failure duration are obtained; The failure impact loss load, the predicted unit time power failure loss and the predicted power failure duration are integrated and accumulated to obtain an initial risk prediction cost; According to the real-time failure probability, the initial risk prediction cost is corrected to obtain a risk prediction cost.

[0036] It should be noted that the risk prediction cost has concealment and serious consequences, in order to accurately quantify, first, based on the determined real-time asset health degree, the preset health degree and fault probability mapping table is inquired to obtain the corresponding real-time fault probability, and the preset health degree and fault probability mapping table is analyzed and constructed by a person skilled in the art according to a large number of historical samples, and can be dynamically adjusted; then, according to the preset asset ID, the influence range after the asset failure is determined, such as A substation affecting a factory and a residential area, and then the fault influence loss load and the corresponding predicted unit time power failure loss and predicted power failure duration are determined, the fault influence loss load refers to the total load of the power affected by the asset, such as 30MW for a factory and 10MW for a residential area, the predicted unit time power failure loss refers to the predicted economic loss per KW·h, such as 100 yuan / KW·h for a factory and 10 yuan / KW·h for a residential area, and the predicted power failure duration is obtained according to the fault type combined with prior experience; the fault influence loss load, the predicted unit time power failure loss and the predicted power failure duration are multiplied and processed, and the data of a plurality of units in the influence range are summed and operated to determine the initial risk prediction cost; finally, the initial risk prediction cost is corrected according to the real-time fault probability to obtain the risk prediction cost, that is, the multiplication processing of the two.

[0037] The application further discloses a power grid asset health degree and operation cost correlation analysis system, comprising a memory and a processor, wherein the memory comprises a power grid asset health degree and operation cost correlation analysis method program, and the power grid asset health degree and operation cost correlation analysis method program is executed by the processor to realize the following steps: A power grid asset evaluation sequence data set of a historical preset time period is acquired, and the power grid asset evaluation sequence data set is analyzed and processed to generate a historical power grid asset portrait. Feature parameters are extracted according to the historical power grid asset portrait to obtain asset health degree historical evaluation feature parameters, and the asset health degree historical evaluation feature parameters are trained to obtain an asset health degree evaluation model. A real-time power grid asset portrait is acquired, feature parameters are extracted according to the real-time power grid asset portrait to obtain asset health degree real-time evaluation feature parameters, and the asset health degree real-time evaluation feature parameters are input into the asset health degree evaluation model for processing to obtain a real-time asset health degree. Historical asset health degrees and corresponding historical operation evaluation costs in a preset time period are acquired, the historical asset health degrees and the corresponding historical operation evaluation costs are trained to obtain a health degree and potential operation cost mapping model. The real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation real-time prediction cost, and the potential operation real-time prediction cost is compared with a preset potential operation permitted cost to obtain an operation decision strategy.

[0038] It should be noted that in order to realize accurate quantification of operation cost according to the health state of power grid assets, firstly, a multi-source power grid asset evaluation sequence data set including physical operation state, environmental condition and operation condition is acquired, and a precise power grid asset portrait is constructed, then, a pre-trained asset health degree evaluation model is extracted after feature parameter extraction, and the real-time asset health degree is accurately quantified, at the same time, the operation cost in the whole life cycle is divided into planned maintenance cost, fault repair cost and risk prediction cost, a health degree and potential operation cost mapping model is pre-trained, the dynamic correlation analysis of asset health state and cost is realized, finally, the potential operation real-time prediction cost is compared with the preset potential operation permitted cost, the corresponding operation decision strategy is generated according to the deviation, and the adaptability of the operation decision strategy is evaluated.

[0039] According to the embodiment of the application, the historical power grid asset evaluation sequence data set of the preset time period is acquired, and the historical power grid asset portrait is generated by analyzing and processing the power grid asset evaluation sequence data set, comprising: The historical power grid asset evaluation sequence data set of the preset time period includes historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data; The historical monitoring record sequence data, the historical asset account record sequence data, the historical inspection record sequence data, the historical running environment record sequence data and the historical running condition record sequence data are preprocessed, and the historical power grid asset portrait is generated in combination with the preset asset ID.

[0040] It should be noted that in order to comprehensively evaluate the operation of power grid assets, firstly, a plurality of power grid asset evaluation data in a historical preset time period is collected to form a power grid asset evaluation sequence data set, wherein the historical monitoring record sequence data includes oil chromatogram monitoring data, partial discharge monitoring data, sleeve dielectric loss monitoring data, temperature monitoring data, load current monitoring data and voltage waveform monitoring data, the historical asset account record sequence data includes equipment model feature data, installation date, manufacturer feature data, rated life and asset technical transformation record data, the historical inspection record sequence data includes inspection log record data and preventive test record data, the historical running environment record sequence data includes running environment temperature mean value, running environment humidity mean value and asset pollution grade, and the historical running condition record sequence data is running abnormal record data, such as short-circuit power impact record data, then, a comprehensive and accurate historical power grid asset portrait is constructed based on the collected multi-source data of multiple time periods and the preset asset ID representing asset identity.

[0041] According to the embodiment of the present application, the feature parameter extraction according to the historical power grid asset image obtains asset health degree historical evaluation feature parameters, and the asset health degree historical evaluation feature parameters are used for training to obtain an asset health degree evaluation model, which comprises: The feature parameter extraction according to the historical power grid asset image obtains asset health degree historical evaluation feature parameters, which comprise operation state feature parameters, operation trend feature parameters, operation environment influence feature parameters and operation health feature parameters. The initial asset health degree evaluation model is trained according to the operation state feature parameters, the operation trend feature parameters, the operation environment influence feature parameters, the operation health feature parameters and the corresponding preset historical asset health degree to obtain the asset health degree evaluation model.

[0042] It should be noted that, in order to realize accurate quantification of asset health based on asset operation state, firstly, the feature parameters influencing asset health degree are extracted according to the constructed historical power grid asset image, wherein the operation state feature parameters are used for quantifying real-time physical state of the power grid asset, such as real-time content of characteristic gas, real-time gas production rate of characteristic gas, load overload multiple and load overload time length, the operation trend feature parameters are used for representing variation trend of feature parameters in a preset time period, such as amplitude variation slope of partial discharge signal and signal duty cycle variation trend data, the operation environment influence feature parameters are used for quantifying damage of external environment to asset health degree, such as fouling grade, lightning insulation damage risk value (obtained by mapping lightning times) and metal component corrosion acceleration coefficient (obtained by weighting and correcting preset initial corrosion coefficient according to environmental humidity and temperature), and the operation health feature parameters are used for representing faults, maintenance and impact of asset in the whole life cycle, such as short-circuit impact times value, and specific feature parameters are preset and set by the person skilled in the art according to different assets and can be dynamically adjusted, then, the initial asset health degree evaluation model is trained based on the extracted operation state feature parameters, operation trend feature parameters, operation environment influence feature parameters, operation health feature parameters and corresponding preset historical asset health degree to obtain the asset health degree evaluation model.

[0043] According to the embodiment of the present application, the real-time power grid asset image is obtained, the feature parameter extraction is performed according to the real-time power grid asset image to obtain asset health degree real-time evaluation feature parameters, and the asset health degree real-time evaluation feature parameters are input into the asset health degree evaluation model for processing to obtain real-time asset health degree, which comprises: The real-time power grid asset image is obtained, the feature parameter extraction is performed according to the real-time power grid asset image to obtain asset health degree real-time evaluation feature parameters, which comprise real-time operation state feature parameters, real-time operation trend feature parameters, real-time operation environment influence feature parameters and real-time operation health feature parameters. The real-time asset health degree is obtained by inputting the real-time running state feature parameter, the real-time running trend feature parameter, the real-time running environment influence feature parameter and the real-time running health feature parameter into the asset health degree evaluation model.

[0044] It should be noted that, in order to evaluate the real-time health of the asset, the real-time power grid asset portrait is subjected to feature parameter extraction, and is analyzed and processed based on the pre-trained asset health degree evaluation model to obtain the real-time asset health degree.

[0045] According to the embodiment of the present application, the historical asset health degree and the corresponding historical operation evaluation cost in a preset time period are obtained, and the health degree and the potential operation cost mapping model is obtained by training according to the historical asset health degree and the corresponding historical operation evaluation cost, comprising: The historical asset health degree and the corresponding historical operation evaluation cost in a preset time period are obtained, wherein the historical operation evaluation cost comprises a historical planned maintenance cost, a historical failure repair cost and a historical risk prediction cost; The historical planned maintenance cost, the historical failure repair cost and the historical risk prediction cost are subjected to weighted summation processing to obtain a historical potential operation evaluation cost; The initial health degree and the potential operation cost mapping model is trained according to the historical asset health degree and the corresponding historical potential operation evaluation cost to obtain the health degree and the potential operation cost mapping model.

[0046] It should be noted that, in order to realize the dynamic correlation between the operation cost and the asset health degree, the operation cost of the asset in the whole life cycle is decomposed into the planned maintenance cost, the failure repair cost and the risk prediction cost in the embodiment of the present application, the planned maintenance cost refers to the cost generated by the preventive maintenance taken to prevent the asset from failure, the failure repair cost refers to the direct cost and the indirect cost generated by the repair after the failure, and the risk prediction cost refers to the potential loss caused by the chain reaction due to the failure of a single asset, the asset type is determined according to the preset asset ID, and then the corresponding weight value is determined, the weighted summation processing is performed to obtain the historical potential operation evaluation cost, and then the initial health degree and the potential operation cost mapping model is trained based on the determined historical asset health degree and the corresponding historical potential operation evaluation cost to obtain the health degree and the potential operation cost mapping model.

[0047] According to the embodiment of the present application, the real-time asset health degree is input into the health degree and the potential operation cost mapping model for analysis and processing to obtain the potential operation real-time prediction cost, and the operation decision strategy is obtained by comparing with the preset potential operation permitted cost, comprising: The real-time asset health degree is input into the health degree and the potential operation cost mapping model for analysis and processing to obtain the potential operation real-time prediction cost. obtain a real-time operation cost deviation rate by comparing the potential real-time operation predicted cost with a preset potential operation permitted cost; obtain a corresponding operation decision strategy by threshold comparison between the real-time operation cost deviation rate and a preset cost operation deviation rate threshold value; if the real-time operation cost deviation rate is less than or equal to the preset cost operation deviation rate threshold value, the operation decision strategy is to maintain the asset operation status; if the real-time operation cost deviation rate is greater than the preset cost operation deviation rate threshold value, the operation decision strategy is to implement asset maintenance.

[0048] It should be noted that the real-time asset health degree determined is mapped and processed by the pre-trained model to obtain the potential real-time operation predicted cost, which is compared with the preset potential operation permitted cost to obtain the real-time operation cost deviation rate. The real-time operation cost deviation rate refers to the ratio of the potential real-time operation predicted cost minus the preset potential operation permitted cost to the preset potential operation permitted cost. If it is negative, that is, the potential real-time operation predicted cost is less than the preset potential operation permitted cost, the real-time operation cost deviation rate is recorded as 0. Then, through threshold comparison, it is determined that the operation strategy of the asset is to maintain the asset operation status or implement asset maintenance.

[0049] According to the embodiment of the application, further comprising: obtain a maintenance scheme and maintenance investment data for implementing asset maintenance; obtain a predicted asset health degree by analyzing and processing the maintenance scheme through a preset asset operation digital twin model; obtain a potential operation corrected predicted cost by inputting the predicted asset health degree into the health degree and potential operation cost mapping model for analysis and processing; obtain an asset investment operation return rate by processing the potential operation corrected predicted cost and the potential real-time operation predicted cost in combination with the maintenance investment data; obtain an asset health degree improvement rate by comparing the predicted asset health degree with the real-time asset health degree; obtain an asset health degree improvement evaluation state and an asset investment operation return evaluation state by threshold comparison between the asset health degree improvement rate and the asset investment operation return rate and the corresponding preset threshold values, respectively; operate the asset health degree improvement evaluation state and the asset investment operation return evaluation state; if it is a qualified state, implement asset maintenance according to the maintenance scheme; if it is an unqualified state, output an early warning response.

[0050] It should be noted that after determining to implement the maintenance, in order to determine whether the maintenance scheme achieves the expectation, first, based on the maintenance scheme, the preset asset operation digital twin model is simulated to determine the predicted asset health degree after implementation of the maintenance scheme, then the health degree and the potential operation cost mapping model are analyzed and processed to obtain a potential operation correction predicted cost, the potential operation correction predicted cost and the potential operation real-time predicted cost are combined with the maintenance investment data for processing to obtain an asset investment operation return rate, the asset investment operation return rate is a potential operation real-time predicted cost minus the potential operation correction predicted cost divided by the maintenance investment data, if it is negative, it is directly recorded as 0, at the same time, the asset health degree improvement rate is evaluated, that is, the predicted asset health degree minus the real-time asset health degree and the real-time asset health degree, if it is negative, it is directly recorded as 0, finally, by comparing the threshold, if it is greater than the corresponding threshold, the asset health degree improvement evaluation state and the asset investment operation return evaluation state are determined to be a qualified state, otherwise, it is an unqualified state, if the asset health degree improvement evaluation state and the asset investment operation return evaluation state are both qualified states, it is finally determined that the maintenance scheme is effective, and the asset maintenance is implemented, otherwise, if any one is unqualified, an early warning is output, and whether to continue to implement the maintenance scheme is finally determined by a technician.

[0051] It is worth mentioning that, according to the embodiment of the application, further comprising: According to the real-time asset health degree, a preset health degree and failure probability mapping table is queried to obtain a corresponding real-time failure probability; According to the preset asset ID, a failure impact loss load and corresponding predicted unit time power failure loss and predicted power failure duration are obtained; The failure impact loss load, the predicted unit time power failure loss and the predicted power failure duration are integrated and accumulated to obtain an initial risk prediction cost; According to the real-time failure probability, the initial risk prediction cost is corrected to obtain a risk prediction cost.

[0052] It should be noted that the risk prediction cost has concealment and serious consequences, in order to accurately quantify, first, based on the determined real-time asset health degree, the preset health degree and fault probability mapping table is inquired to obtain the corresponding real-time fault probability, and the preset health degree and fault probability mapping table is analyzed and constructed by a person skilled in the art according to a large number of historical samples, and can be dynamically adjusted; then, according to the preset asset ID, the influence range of the asset after failure is determined, such as A substation affecting a factory and a residential area, and then the fault influence loss load and the corresponding predicted unit time power failure loss and predicted power failure duration are determined, the fault influence loss load refers to the total load of the power consumption units affected by the asset, such as 30MW for the factory and 10MW for the residential area, the predicted unit time power failure loss refers to the predicted economic loss per KW·h, such as 100 yuan / KW·h for the factory and 10 yuan / KW·h for the residential area, and the predicted power failure duration is obtained according to the fault type combined with prior experience; the fault influence loss load, the predicted unit time power failure loss and the predicted power failure duration are multiplied and processed, and the data of a plurality of units in the influence range are summed and operated to determine the initial risk prediction cost; finally, the initial risk prediction cost is corrected according to the real-time fault probability to obtain the risk prediction cost, that is, the multiplication processing.

[0053] The power grid asset health degree and operation cost correlation analysis method and system disclosed by the application realizes the power grid asset health degree and operation cost correlation analysis by constructing the power grid asset portrait through multi-source data, health degree quantitative calculation, double model collaborative analysis, full life cycle cost decomposition and prediction, and asset investment operation return rate calculation, thereby realizing intelligent correlation analysis from state monitoring to operation cost management.

[0054] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division mode, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed components can be through some interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0055] The units described above as separate components can or can not be physically separated, and the components shown as units can or can not be physical units; they can be located in one place or distributed on multiple network units; part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment scheme.

[0056] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be separately taken as one unit, or two or more units can be integrated in one unit; the integrated unit can be realized in the form of hardware or in the form of hardware plus software function unit.

[0057] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a readable storage medium, and the program, when executed, executes the steps including the above-mentioned method embodiments; and the foregoing storage medium includes a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disc or an optical disc, and various storage medium capable of storing program codes.

[0058] Alternatively, when the integrated unit of the present application is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product in essence or the part that contributes to the prior art, and the software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present application. The foregoing storage medium includes a mobile storage device, a ROM, a RAM, a magnetic disc or an optical disc, and various storage medium capable of storing program codes.

Claims

1. A method for grid asset health and operating cost correlation analysis, the method comprising: The method comprises the following steps: ​ obtaining a power grid asset evaluation sequence data set of a historical preset time period, analyzing and processing the power grid asset evaluation sequence data set to generate a historical power grid asset portrait; extracting feature parameters according to the historical power grid asset portrait to obtain asset health degree historical evaluation feature parameters, training according to the asset health degree historical evaluation feature parameters to obtain an asset health degree evaluation model; obtaining a real-time power grid asset portrait, extracting feature parameters according to the real-time power grid asset portrait to obtain asset health degree real-time evaluation feature parameters, and inputting the asset health degree real-time evaluation feature parameters into the asset health degree evaluation model for processing to obtain a real-time asset health degree; obtaining historical asset health degrees and corresponding historical operation evaluation costs in a preset time period, training according to the historical asset health degrees and the corresponding historical operation evaluation costs to obtain a health degree and potential operation cost mapping model; inputting the real-time asset health degree into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation real-time prediction cost, and comparing the potential operation real-time prediction cost with a preset potential operation permitted cost to obtain an operation decision strategy.

2. The power grid asset health and operating cost correlation analysis method of claim 1, wherein, The method comprises the following steps: obtaining a power grid asset evaluation sequence data set of a historical preset time period, analyzing and processing the power grid asset evaluation sequence data set to generate a historical power grid asset portrait, comprising: obtaining a power grid asset evaluation sequence data set of a historical preset time period, comprising historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data and historical running condition record sequence data; 3. The power grid asset health and operating cost correlation analysis method of claim 2, wherein, performing data preprocessing on the historical monitoring record sequence data, the historical asset account record sequence data, the historical inspection record sequence data, the historical running environment record sequence data and the historical running condition record sequence data, and generating a historical power grid asset portrait in combination with a preset asset ID. The method comprises the following steps: extracting feature parameters according to the historical power grid asset portrait to obtain asset health degree historical evaluation feature parameters, training according to the asset health degree historical evaluation feature parameters to obtain an asset health degree evaluation model, comprising:

4. The power grid asset health and operating cost correlation analysis method of claim 3, wherein, extracting feature parameters according to the historical power grid asset portrait to obtain asset health degree historical evaluation feature parameters, including running state feature parameters, running trend feature parameters, running environment influence feature parameters and running health feature parameters; training an initial asset health degree evaluation model according to the running state feature parameters, the running trend feature parameters, the running environment influence feature parameters, the running health feature parameters and the corresponding preset historical asset health degrees to obtain the asset health degree evaluation model. The method comprises the following steps: obtaining a real-time power grid asset portrait, extracting feature parameters according to the real-time power grid asset portrait to obtain asset health degree real-time evaluation feature parameters, and inputting the asset health degree real-time evaluation feature parameters into the asset health degree evaluation model for processing to obtain a real-time asset health degree, comprising: obtaining a real-time power grid asset portrait, extracting feature parameters according to the real-time power grid asset portrait to obtain asset health degree real-time evaluation feature parameters, including real-time running state feature parameters, real-time running trend feature parameters, real-time running environment influence feature parameters and real-time running health feature parameters; Input the real-time running state feature parameters, real-time running trend feature parameters, real-time running environment influence feature parameters and real-time running health feature parameters into the asset health degree evaluation model for processing to obtain real-time asset health degree.

5. The power grid asset health and operating cost correlation analysis method of claim 4, wherein, The historical asset health degree and the corresponding historical operation evaluation cost in the preset time period are obtained, and the historical asset health degree and the corresponding historical operation evaluation cost are used for training to obtain a health degree and potential operation cost mapping model, which includes: The historical asset health degree and the corresponding historical operation evaluation cost in the preset time period are obtained, wherein the historical operation evaluation cost includes historical planned maintenance cost, historical failure repair cost and historical risk prediction cost; The historical planned maintenance cost, historical failure repair cost and historical risk prediction cost are processed by weighted summation to obtain historical potential operation evaluation cost; The initial health degree and potential operation cost mapping model is trained according to the historical asset health degree and the corresponding historical potential operation evaluation cost to obtain a health degree and potential operation cost mapping model.

6. The power grid asset health and operating cost correlation analysis method of claim 5, wherein, The real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain potential operation real-time prediction cost, and the potential operation real-time prediction cost is compared with a preset potential operation permitted cost to obtain an operation decision strategy, which includes: The real-time asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain potential operation real-time prediction cost; The potential operation real-time prediction cost is compared with a preset potential operation permitted cost to obtain a real-time operation cost deviation rate; The real-time operation cost deviation rate is compared with a preset cost operation deviation rate threshold value to obtain a corresponding operation decision strategy; If the real-time operation cost deviation rate is less than or equal to the preset cost operation deviation rate threshold value, the operation decision strategy is to maintain the asset operation status; If the real-time operation cost deviation rate is greater than the preset cost operation deviation rate threshold value, the operation decision strategy is to implement asset maintenance.

7. The power grid asset health and operating cost correlation analysis method of claim 6, wherein, Further comprising: Obtaining a maintenance scheme and maintenance investment data for implementing asset maintenance; According to the maintenance scheme, the preset asset running digital twin model is analyzed and processed to obtain a predicted asset health degree; The predicted asset health degree is input into the health degree and potential operation cost mapping model for analysis and processing to obtain a potential operation corrected prediction cost; According to the potential operation corrected prediction cost and the potential operation real-time prediction cost in combination with the maintenance investment data, an asset investment operation return rate is obtained; The predicted asset health degree is compared with the real-time asset health degree to obtain an asset health degree improvement rate; The asset health degree improvement rate and the asset investment operation return rate are respectively compared with corresponding preset threshold values to obtain an asset health degree improvement evaluation state and an asset investment operation return evaluation state; The asset health degree improvement evaluation state and the asset investment operation return evaluation state are calculated; If it is a qualified state, asset maintenance is implemented according to the maintenance scheme; If it is an unqualified state, an early warning response is output.

8. A power grid asset health and operating cost correlation analysis system, characterized by, The method comprises a memory and a processor, the memory comprises a power grid asset health degree and operation cost correlation analysis method program, and the power grid asset health degree and operation cost correlation analysis method program is executed by the processor to realize the following steps: Obtain a power grid asset evaluation sequence data set of a historical preset time period, analyze and process the power grid asset evaluation sequence data set, and generate a historical power grid asset portrait; Extract feature parameters according to the historical power grid asset portrait, obtain asset health degree historical evaluation feature parameters, train the asset health degree historical evaluation feature parameters, and obtain an asset health degree evaluation model; Obtain a real-time power grid asset portrait, extract feature parameters according to the real-time power grid asset portrait, obtain asset health degree real-time evaluation feature parameters, input the asset health degree real-time evaluation feature parameters into the asset health degree evaluation model for processing, and obtain a real-time asset health degree; Obtain historical asset health degrees and corresponding historical operation evaluation costs in a preset time period, train the historical asset health degrees and the corresponding historical operation evaluation costs, and obtain a health degree and potential operation cost mapping model; Input the real-time asset health degree into the health degree and potential operation cost mapping model for analysis and processing, obtain a potential operation real-time prediction cost, compare the potential operation real-time prediction cost with a preset potential operation permitted cost, and obtain an operation decision strategy.

9. The power grid asset health and operating cost correlation analysis system of claim 8, wherein, The obtaining of the power grid asset evaluation sequence data set of the historical preset time period, the analysis and processing of the power grid asset evaluation sequence data set, and the generation of the historical power grid asset portrait comprise: The power grid asset evaluation sequence data set of the historical preset time period comprises historical monitoring record sequence data, historical asset account record sequence data, historical inspection record sequence data, historical running environment record sequence data, and historical running condition record sequence data; The historical monitoring record sequence data, the historical asset account record sequence data, the historical inspection record sequence data, the historical running environment record sequence data, and the historical running condition record sequence data are preprocessed, and a historical power grid asset portrait is generated in combination with a preset asset ID.

10. The power grid asset health and operating cost correlation analysis system of claim 9, wherein, The extraction of the feature parameters according to the historical power grid asset portrait, the obtaining of the asset health degree historical evaluation feature parameters, and the training of the asset health degree historical evaluation feature parameters to obtain the asset health degree evaluation model comprise: The extraction of the feature parameters according to the historical power grid asset portrait and the obtaining of the asset health degree historical evaluation feature parameters comprise running state feature parameters, running trend feature parameters, running environment influence feature parameters, and running health feature parameters; The initial asset health degree evaluation model is trained according to the running state feature parameters, the running trend feature parameters, the running environment influence feature parameters, the running health feature parameters, and the corresponding preset historical asset health degrees, and an asset health degree evaluation model is obtained.

Images