Multi-data analysis equipment state monitoring and life prediction method and related device

Abstract

The invention provides a multi-data analysis equipment state monitoring and life prediction method and a related device, and the method comprises the steps: obtaining the partial discharge data, temperature data, water immersion data and environment temperature data of a cable joint of cable equipment in a window through employing a window sliding technology; carrying out preprocessing on the acquired data; performing abnormity monitoring on the partial discharge data and the temperature data in the window; if the short-term abnormal data appears, dynamically adjusting and updating the abnormal monitoring threshold value according to the proportion of the short-term abnormal data to the long-term data; quantifying the abnormal change trend of the data in the window, and carrying out early warning on the equipment state; and predicting through the prediction model to obtain the residual life of the cable equipment. According to the invention, through dynamic adjustment of the threshold value and trend analysis, accurate monitoring and fault early warning of the state of cable equipment are realized, the accuracy and efficiency of early warning are improved, and meanwhile, the residual life of the cable is predicted, so that the maintenance strategy is optimized, and the reliability and safety of a power system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of cable monitoring and fault warning in power systems, and particularly relates to a method for equipment status monitoring and life prediction based on multi-data analysis and related devices. Background Art

[0002] In a power system, as a key power transmission component, the accessories of a cable, especially the cable intermediate joint, become a high-fault area due to its complex structure and harsh installation environment. With the extension of the cable line, the number of intermediate joints increases, and the fault risk also rises accordingly. In the south of China, the cable intermediate joints also face problems such as water accumulation and poor heat dissipation, further increasing the possibility of faults.

[0003] In the power industry, partial discharge monitoring is an important means to evaluate the insulation status of a cable. However, there is currently no unified and scientific standard. Most monitoring systems rely on empirical alarm thresholds and are difficult to adapt to changes in the environment and operating conditions, resulting in insufficient accuracy of alarm signals. At the same time, the aging of cable insulation is a gradual process, and its trend analysis is crucial for predicting faults and planning maintenance. However, there is a lack of accurate trend analysis algorithms in the industry.

[0004] Existing technologies have various problems in cable monitoring and fault warning. First, when calculating the dynamic threshold, the algorithm is complex and the response time is long. As the amount of data increases, the calculation cost rises significantly. Second, the variability of the on-site environment leads to large errors in existing dynamic threshold judgment algorithms and a lack of effective correction and supplementation. In addition, there is a lack of a trend judgment algorithm for cable insulation defects, and it is impossible to accurately predict and evaluate the development of the insulation status. Finally, when predicting the life of a cable joint, the existing technologies have a complex structure, a large amount of data processing, insufficient real-time performance and accuracy, and it is difficult to meet the actual application requirements. Summary of the Invention

[0005] In view of this, the present invention provides a method for equipment status monitoring and life prediction based on multi-data analysis and related devices, aiming to achieve precise monitoring and fault warning of the insulation status of cable intermediate joints through dynamic threshold adjustment and trend analysis, improve the accuracy and efficiency of warning, predict the remaining life of the cable at the same time, and thus optimize the maintenance strategy and enhance the reliability and safety of the power system.

[0006] To achieve the above object, the technical solution provided by the present invention is as follows:

[0007] In the first aspect, the present invention provides a method for equipment status monitoring and life prediction based on multi-data analysis, including the following steps:

[0008] Using the window sliding technology, obtain the partial discharge data, temperature data, water immersion data, and ambient temperature data of the cable joint of the cable equipment within the window;

[0009] Preprocess the acquired data and fill in the missing values in the data;

[0010] Perform anomaly monitoring on the partial discharge data and temperature data within the window respectively according to the anomaly monitoring threshold. If the data within the set duration range is greater than the anomaly monitoring threshold, the data within the set duration range is determined as short-term anomaly data, and the remaining data within the window is long-term data;

[0011] If short-term anomaly data appears, dynamically adjust and update the size of the anomaly monitoring threshold according to the proportion of the short-term anomaly data to the long-term data;

[0012] Quantify the abnormal change trend of the data within the window, and use the quantified value of the abnormal change trend to give an early warning of the equipment status;

[0013] Based on the water immersion data within the window, as well as the temperature data of the cable joint and the ambient temperature data, perform prediction through a pre-constructed prediction model to obtain the remaining life of the cable equipment.

[0014] Furthermore, in dynamically adjusting and updating the size of the anomaly monitoring threshold, for different types of data, the size of the anomaly monitoring threshold is dynamically adjusted and updated according to the following steps:

[0015] For multiple groups of discontinuous short-term anomaly data that appear within the monitoring window, calculate the mean and standard deviation of different types of data in each group of short-term anomaly data respectively;

[0016] Calculate the first threshold of different types of data according to the mean and standard deviation of the short-term anomaly data, as follows:

[0017]

[0018] In the formula, is the first threshold; the subscript represents short-term, represents the data type, is the group number index of multiple groups of short-term anomaly data; and are the mean and standard deviation respectively; is the first coefficient;

[0019] Calculate the mean and standard deviation of each type of data in the long-term data;

[0020] Calculate the second threshold of different types of data according to the mean and standard deviation of the long-term data, as follows:

[0021]

[0022] In the formula, is the second threshold; the subscript Indicates long-term; is the second coefficient;

[0023] Calculate the proportion of each type of data in the short-term abnormal data to the total data respectively, and use the proportion as the weight of different types of data;

[0024] Calculate the dynamic threshold according to the following formula, and dynamically adjust and update the abnormal monitoring threshold according to the dynamic threshold:

[0025]

[0026] In the formula, is the dynamic threshold; is the corresponding weight; is the window data set; is the short-term abnormal data.

[0027] Furthermore, in calculating the dynamic threshold, it also includes correcting the calculated dynamic threshold, and the correction process includes the following steps:

[0028] Calculate the mean difference of the same type of data between the window data set and the long-term data and between the window data set and the short-term abnormal data respectively, and obtain the first type of absolute error and the second type of absolute error;

[0029] Calculate the skewness of different types of data in the short-term abnormal data;

[0030] Calculate the kurtosis of different types of data in the long-term data;

[0031] Calculate the first non-linear adjustment coefficient and the second non-linear adjustment coefficient according to the following formula, as follows:

[0032] ,

[0033] In the formula, and are the first non-linear adjustment coefficient and the second non-linear adjustment coefficient respectively; is an empirical value, i = 1, 2, 3, 4; is the skewness; is the kurtosis;

[0034] Calculate the correction coefficient of different types of data according to the following formula, as follows:

[0035]

[0036] In the formula, is the correction coefficient; is a preset coefficient; and are the first type of absolute error and the second type of absolute error respectively;

[0037] The dynamic threshold for different types of data is corrected according to the following formula, as follows:

[0038]

[0039] In the formula, is the corrected dynamic threshold.

[0040] Furthermore, for different types of data, the abnormal change trend of the data within the quantization window includes the following steps:

[0041] Calculate the absolute value of the change rate of the data at each adjacent moment;

[0042] Calculate the mean and standard deviation of the data, denoted as and respectively;

[0043] Calculate the change rate threshold of the data according to the following formula, as follows:

[0044]

[0045] In the formula, is the change rate threshold; is a preset coefficient;

[0046] Mark the moment when the absolute value of the change rate exceeds the change rate threshold as an abnormal moment;

[0047] Calculate the volatility of the data within one day near the abnormal moment and the volatility of the data within the window respectively;

[0048] Quantify the abnormal change trend of the data according to the following formula, as follows:

[0049]

[0050] In the formula, is the comprehensive quantization score value of the abnormal change trend; the subscript represents the data type; is the mean of the data within the window; is the volatility of the data within the window; is the mean of the data within one day near the th abnormal moment; is the mean of the data within one day near the th abnormal moment; is a determined value obtained from experiments, i = 1, 2, 3, 4.

[0051] Furthermore, use the quantization value of the abnormal change trend to give an early warning of the equipment status, including:

[0052] The comprehensive quantitative score value is normalized according to the following formula, as follows:

[0053] =

[0054] In the formula, is the comprehensive quantitative score value after normalization;

[0055] If the comprehensive quantitative score value after normalization is greater than the set threshold, it is determined that the device is abnormal.

[0056] Furthermore, when constructing the prediction model, the relationship between the water immersion data, temperature data, and ambient temperature data and the remaining life of the cable device is determined through the fitting process of the sine function. The construction process includes:

[0057] Considering the influence of temperature on the device life, the temperature influence coefficient is calculated based on the temperature data and ambient temperature data, as follows:

[0058]

[0059] In the formula, is the temperature influence coefficient; is the pre-exponential factor; is the activation energy; is the gas constant; is the temperature data; is the ambient temperature data;

[0060] Considering the dual influence of temperature on humidity and device life, the humidity influence coefficient is calculated based on the water immersion data, temperature data, and ambient temperature data, as follows:

[0061]

[0062] In the formula, is the humidity influence coefficient; is the basic humidity factor; is the coefficient of temperature difference; is the humidity level determined according to the water immersion data;

[0063] Considering the influence of the interaction between temperature and humidity on the device life, the comprehensive influence coefficient is calculated based on the temperature influence factor and humidity influence factor, as follows:

[0064]

[0065] In the formula, is the comprehensive influence coefficient;

[0066] Determine the parameters in the comprehensive influence coefficient through the fitting process of the sine function and ;

[0067] Construct and fit a prediction model based on each coefficient.

[0068] Furthermore, the mathematical expression of the prediction model is as follows:

[0069]

[0070] In the formula, is the remaining life of the cable equipment; is the life of the equipment without additional environmental pressure; is the temperature data and environmental temperature data; is the water immersion data; is the error term.

[0071] In a second aspect, the present invention provides a device status monitoring and life prediction device for multi-data analysis, including:

[0072] A data acquisition module, which is used to use the window sliding technology to acquire the partial discharge data, temperature data, water immersion data and environmental temperature data of the cable joints of the cable equipment within the window;

[0073] A preprocessing module, which is used to preprocess the acquired data and fill in the missing values in the data;

[0074] An anomaly monitoring module, which is used to respectively perform anomaly monitoring on the partial discharge data and temperature data within the window according to the anomaly monitoring threshold. If the data within the set duration range is greater than the anomaly monitoring threshold, the data within the set duration range is determined as short-term anomaly data, and the remaining data within the window is long-term data;

[0075] A dynamic threshold update module, which is used to dynamically adjust and update the size of the anomaly monitoring threshold according to the proportion of the short-term anomaly data in the long-term data when short-term anomaly data appears;

[0076] A device status warning module, which is used to quantify the abnormal change trend of the data within the window and use the quantified value of the abnormal change trend to warn the device status;

[0077] A device life prediction module, which is used to predict based on the water immersion data within the window, as well as the temperature data and environmental temperature data of the cable joints, through a pre-constructed prediction model to obtain the remaining life of the cable equipment.

[0078] In a third aspect, the present invention also provides a computer device, which includes a processor and a memory:

[0079] The memory is used to store a computer program and send the instructions of the computer program to the processor;

[0080] The processor executes a device status monitoring and life prediction method for multi-data analysis according to the instructions of a computer program as described in the first aspect.

[0081] In a fourth aspect, the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements a device status monitoring and life prediction method for multi-data analysis as described in the first aspect.

[0082] In summary, the present invention provides a device status monitoring and life prediction method for multi-data analysis. The method includes using a window sliding technique to obtain partial discharge data, temperature data, water immersion data, and ambient temperature data of a cable joint of a cable device within a window; preprocessing the obtained data to fill in missing values in the data; respectively performing anomaly monitoring on the partial discharge data and temperature data within the window according to an anomaly monitoring threshold. If data within a set duration range is greater than the anomaly monitoring threshold, the data within the set duration range is determined as short-term anomaly data, and the remaining data within the window is long-term data; if short-term anomaly data appears, the size of the anomaly monitoring threshold is dynamically adjusted and updated according to the proportion of the short-term anomaly data to the long-term data; quantifying the abnormal change trend of the data within the window, and using the quantified value of the abnormal change trend to give an early warning of the device status; based on the water immersion data within the window, as well as the temperature data of the cable joint and the ambient temperature data, predicting through a pre-constructed prediction model to obtain the remaining life of the cable device. The present invention realizes precise monitoring and fault early warning of the insulation state of the cable intermediate joint through dynamic threshold adjustment and trend analysis, improves the accuracy and efficiency of early warning, and at the same time predicts the remaining life of the cable, thereby optimizing the maintenance strategy and enhancing the reliability and safety of the power system.

[0083] The present invention also provides a device status monitoring and life prediction device, a computer device, and a computer-readable storage medium for multi-data analysis, which have similar effects when implemented and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0085] Figure 1 It is a flowchart of a device status monitoring and life prediction method for multi-data analysis provided by an embodiment of the present invention;

[0086] Figure 2Flow chart of dynamic threshold calculation and correction provided by the embodiments of the present invention;

[0087] Figure 3 Flow chart of abnormal change trend judgment provided by the embodiments of the present invention;

[0088] Figure 4 Block diagram of a device state monitoring and life prediction device for multi - data analysis provided by the embodiments of the present invention;

[0089] Figure 5 Block diagram of a computer device provided by the embodiments of the present invention. Detailed implementation manners

[0090] To make the objectives, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0091] Please refer to Figure 1 , this embodiment provides a method for device state monitoring and life prediction of multi - data analysis, including the following steps:

[0092] S1: Using the window sliding technology, obtain the partial discharge data, temperature data, water immersion data, and ambient temperature data of the cable joints of the cable equipment within the window.

[0093] S2: Pre - process the obtained data to fill the missing values in the data.

[0094] S3: Respectively perform abnormal monitoring on the partial discharge data and temperature data within the window according to the abnormal monitoring threshold. If the data within the set duration range is greater than the abnormal monitoring threshold, the data within the set duration range is determined as short - term abnormal data, and the remaining data within the window is long - term data.

[0095] S4: If short - term abnormal data appears, dynamically adjust and update the size of the abnormal monitoring threshold according to the proportion of the short - term abnormal data to the long - term data.

[0096] S5: Quantify the abnormal change trend of the data within the window, and use the quantified value of the abnormal change trend to give an early warning of the device state.

[0097] S6: Based on the water immersion data within the window, as well as the temperature data of the cable joints and the ambient temperature data, perform prediction through a pre - constructed prediction model to obtain the remaining life of the cable equipment.

[0098] The prediction method provided in this embodiment periodically collects multi-dimensional data during the operation of the device through the window sliding technology. After preprocessing the data, it monitors abnormal data based on the set threshold, then dynamically adjusts the threshold according to the situation of the abnormal data, quantifies the abnormal change trend and issues an early warning. Finally, it estimates the remaining life of the device through a prediction model in combination with specific data, realizing the comprehensive monitoring of the device status and life prediction. The prediction method of this embodiment comprehensively utilizes various types of operation data of the cable device, which can comprehensively reflect the device status. At the same time, it dynamically adjusts the abnormal monitoring threshold according to the proportion of short-term abnormal data in the long-term data, enhancing the monitoring adaptability. In addition, based on data such as water immersion, cable joint temperature, and ambient temperature, it uses a pre-constructed model to predict the remaining life of the cable, thereby optimizing the maintenance strategy, changing the blind or lagging maintenance method, and improving the reliability and safety of the power system.

[0099] In one embodiment, a data preprocessing method is provided for step S2, including the following steps:

[0100] S21: Obtain the historical data of the past 30 days within the window, and detect whether there are missing values in the historical data;

[0101] S22: For the detected missing values, calculate the mean values of the data at the moments before and after the missing values;

[0102] S23: Use the calculated mean values to fill the missing data;

[0103] S24: Perform time aggregation processing on the filled data, calculate the mean value of the data at hourly intervals, so as to generate a dataset for each hour.

[0104] By adopting the above technical solution, since the partial discharge and temperature data fluctuate and are missing due to external factors such as the environment, a comprehensive check is carried out on the data integrity to ensure that the data at each time point is available. Then, the missing data is filled with the mean values at the moments before and after to ensure the accuracy and continuity of the filled data. Finally, the filled high-frequency data is aggregated hourly to smooth the data fluctuations caused by external signals, further ensuring the data accuracy and saving the calculation time and overhead of subsequent algorithms.

[0105] The following takes the acquisition of partial discharge (amplitude and frequency) and temperature (ambient temperature and cable joint temperature) as an example to introduce some other embodiments of the present invention.

[0106] In one embodiment, an abnormal monitoring method is provided for step S3, including the following steps:

[0107] S31: Obtain the hourly dataset of the past 30 days within the window, where it includes partial discharge (amplitude and frequency) and temperature (ambient temperature and cable joint temperature), denoted as D;

[0108] S32: Initially set the anomaly detection threshold to the experimental threshold (a determined value obtained from experiments, with the experimental threshold for amplitude being 500 pC, the experimental threshold for frequency being 200, and the experimental threshold for temperature being 50 °C). The experimental threshold is defined as ( respectively being , representing amplitude, frequency, and the temperature of the cable joint);

[0109] S33: Independently partition and process the partial discharge and temperature data respectively. If the partial discharge or temperature data is greater than their respective experimental thresholds for 3 consecutive days, these data are marked as short - term anomaly data, denoted as , and the remaining data is marked as long - term data, denoted as . If no anomaly data appears, the data remains as long - term data, still denoted as ;

[0110] S34: After each algorithm iteration, update the anomaly monitoring threshold to a new dynamic threshold. After updating the dynamic threshold, continue to identify anomaly data according to the method of S33, thereby obtaining new short - term anomaly data and long - term data.

[0111] By adopting the above - mentioned technical solution, using the updated dynamic threshold, effectively monitor the partial discharge and temperature data. By identifying anomaly data, extract short - term anomaly data and long - term data, thereby realizing a comprehensive assessment of the equipment status.

[0112] In a further embodiment, for different types of data, the size of the anomaly monitoring threshold is dynamically adjusted and updated according to the following steps, including:

[0113] S41: Extract all the data in , denote the amplitude in as , the frequency as , and the temperature of the cable joint as ;

[0114] S42: During the anomaly detection process, if there is short - term anomaly data, extract all the data in the long - term data , denote the amplitude in as , the frequency as , and the temperature of the cable joint as , and independently perform the following calculations respectively;

[0115] S43: Calculate the mean value and the standard deviation of the amplitude in , , , is the number of short-term abnormal data in amplitude;

[0116] S44: Calculate the mean value of the frequency and the standard deviation in , , , is the number of short-term abnormal data in frequency;

[0117] S45: Calculate the mean value of the cable joint temperature and the standard deviation in , , , is the number of short-term abnormal data in cable joint temperature;

[0118] S46: Since short-term abnormal data may be discontinuous within a certain period, the short-term abnormal amplitude, frequency and cable joint temperature thresholds, ( are respectively , representing amplitude, frequency and cable joint temperature), are the determined values obtained from experiments;

[0119] S47: When there are short-term abnormal data, calculate the mean value of the amplitude and the standard deviation in the long-term data , , , is the number of long-term data in amplitude;

[0120] S48: Calculate the mean value of the frequency and the standard deviation in , , , is the number of long-term data in frequency;

[0121] S49: Calculate the mean value of the cable joint temperature and the standard deviation in , , , is the number of long-term data in frequency;

[0122] S410: Denote the long-term thresholds of amplitude, frequency and cable joint temperature as , , ( are respectively , representing amplitude, frequency, and cable joint temperature) is the determined value obtained from the experiment;

[0123] S411: Calculate the amplitude weight , , where n is the number of short-term abnormal amplitude data;

[0124] S412: Calculate the frequency weight , , where m is the number of short-term abnormal frequency data;

[0125] S413: Calculate the temperature weight , , where q is the number of short-term abnormal temperature data;

[0126] S414: Obtain the dynamic thresholds for amplitude, frequency, and cable joint temperature, and denote the dynamic thresholds as ( are respectively , representing amplitude, frequency, and cable joint temperature),

[0127] ;

[0128] By adopting the above technical solution, the dynamic thresholds for amplitude, frequency, and cable joint temperature are obtained respectively, so as to obtain the dynamic thresholds set in combination with the on-site environment of the equipment installation.

[0129] In a further embodiment, it further includes correcting the calculated dynamic thresholds, and the correction process includes the following steps:

[0130] S415: Extract the 30-day data set within the window All the data in it, denoted as The amplitude in it is , the frequency is , and the cable joint temperature is , and the following calculations are performed independently respectively;

[0131] S416: Since the equipment may be affected by external signals, which may cause data fluctuations. When there are short-term abnormal data, relying only on the distribution of short-term abnormalities and long-term data is not sufficient to comprehensively evaluate the overall characteristics. Therefore, calculate the absolute errors of the three characteristics of amplitude, frequency, and cable joint temperature to more comprehensively evaluate the data distribution. The following calculations are performed independently respectively, calculating the absolute errors of short-term abnormalities and long-term data of amplitude, frequency, and cable joint temperature, denoted as (respectively , representing amplitude, frequency, and cable joint temperature):

[0132] , ;

[0133] S417: Calculate the amplitude, frequency of short-term abnormal data, and the skewness of the cable joint temperature, denoted as , ( are respectively , representing the amplitude, frequency, and cable joint temperature)

[0134]

[0135] S418: Calculate the kurtosis of the amplitude, frequency of long-term data, and the cable joint temperature, denoted as ( are respectively , representing the amplitude, frequency, and cable joint temperature)

[0136] ;

[0137] S419: The fluctuations caused by external factors on the equipment may make the data distribution non-linear. Therefore, a non-linear adjustment coefficient is introduced. The following calculations are carried out independently. Calculate the non-linear adjustment coefficients of the amplitude, frequency, and cable joint temperature, denoted as , , ( are respectively , representing the amplitude, frequency, and cable joint temperature)

[0138] , , (i = 1, 2, 3, 4) are empirical values obtained from experiments, and their specific values are adjusted elastically;

[0139] S420: Calculate the correction coefficients of the amplitude, frequency, and cable joint temperature, denoted as , ( are respectively , representing the amplitude, frequency, and cable joint temperature)

[0140] , The experimental values are used to adjust the correction coefficients;

[0141] S421: Calculate the corrected thresholds of the amplitude, frequency, and cable joint temperature, denoted as , ( are respectively , representing the amplitude, frequency, and cable joint temperature)

[0142] ;

[0143] By adopting the above technical solution, dynamic thresholds are obtained for the amplitude, frequency, and cable joint temperature respectively, so as to obtain dynamic thresholds set in combination with the on-site environment of the equipment installation. To ensure the accuracy of the dynamic thresholds, a correction coefficient is calculated to correct the dynamic thresholds to ensure the accuracy of monitoring. The calculation and correction process of the dynamic thresholds is as Figure 2 shown.

[0144] In one embodiment, for different types of data, the abnormal change trend of the data within the quantization window includes the following steps:

[0145] S51: Calculate the absolute value of the change rate of each adjacent moment of the amplitude, frequency, and cable joint temperature, denoted as , ( are respectively , representing the amplitude, frequency, and cable joint temperature);

[0146] S52: Calculate the mean and standard deviation of the change rates of the amplitude, frequency, and temperature ( are respectively , representing the amplitude, frequency, and cable joint temperature), denoted as , ;

[0147] S53: Calculate the change rate thresholds of the amplitude, frequency, and cable joint temperature, ( are respectively , representing the amplitude, frequency, and cable joint temperature), denoted as , , p (1.5, 3), and the p range is the range determined by experiments ;

[0148] S54: It can be known from experiments that abnormal moments do not exist continuously within a period of time. Compare the change rates of the amplitude, frequency, and cable joint temperature with the change rate thresholds. If the change rate at a certain moment is greater than the change rate threshold ( are respectively , representing the amplitude, frequency, and cable joint temperature), it is considered that the equipment may be abnormal. Denote the change rate abnormal moment as , and calculate the mean and standard deviation of the amplitude, frequency, and cable joint temperature data in the recent day at the abnormal moment, denoted as , ( are respectively , representing the amplitude, frequency, and temperature);

[0149] S55: Calculate the kurtosis and skewness of the amplitude, frequency, and cable joint temperature at the abnormal t moment, denoted as 、 ( respectively );

[0150] S56: Calculate the volatility of the amplitude, frequency, and cable joint temperature in the past day at the abnormal moment, denoted as , , , . ; ; ;

[0151] S57: Calculate the kurtosis and skewness of the amplitude, frequency, and cable joint temperature in 30 days, denoted as , ( respectively );

[0152] S58: Calculate the volatility of the amplitude, frequency, and cable joint temperature in 30 days within the window, denoted as ( respectively , representing amplitude, frequency, and temperature). ; S59: Calculate the comprehensive score of the trend change degree of the amplitude, frequency, and cable joint temperature, denoted as ( respectively , representing amplitude, frequency, and temperature). , (i = 1, 2, 3, 4) are the determined values obtained from the experiment.

[0153] In a further embodiment, the device state is warned using the quantization value of the abnormal change trend, including:

[0154] S510: Normalize the comprehensive score, = ( respectively , representing amplitude, frequency, and cable joint temperature);

[0155] S511: Abnormal trend judgment, represents the comprehensive score of three characteristics of amplitude, frequency, and cable joint temperature ( respectively , representing amplitude, frequency, and cable joint temperature), , if the comprehensive score of any characteristic is greater than 0.5, it is determined that the device is abnormal.

[0156] By adopting the above technical solution, the abnormal performance of the identification device in the numerical dimension is identified in step four, and through further trend analysis, the device is secondarily detected to confirm whether there is an abnormal state. Through the secondary abnormal judgment, the accuracy and reliability of the monitoring results are ensured, thereby improving the safety of the device operation.

[0157] In one embodiment, when constructing the prediction model, the relationship between the water immersion data, temperature data, environmental temperature data and the remaining life of the cable device is determined through the fitting process of the sine function. The construction process includes:

[0158] S61: Extract all data from the dataset D of the past 30 days within the window. Denote the water immersion data as , the environmental temperature data as , and the cable joint temperature data as ;

[0159] S62: During the device monitoring process, when the temperature is too high, the damage rate of the device will be accelerated. To evaluate the impact of temperature on the device life, calculate the temperature impact coefficient , , is the pre-exponential factor with a value of , is the activation energy with a value of 80 , R is the gas constant with a value of 8.314 , is the cable joint temperature, is the environmental temperature;

[0160] S63: Considering the dual impact of temperature on humidity and device life, it is necessary to calculate the humidity impact factor to quantify the impact of humidity on device performance. Denote the humidity impact factor as . , : represents the basic humidity factor with a value of 0.75, is the coefficient of the temperature difference with a value of 2.5, is the cable joint temperature, environmental temperature, is the humidity level with a value of 0.8;

[0161] S64: Further considering the impact of humidity on device life, calculate the humidity impact coefficient ;

[0162] S65: Considering the combined impact of the interaction between temperature and humidity on device life, calculate its comprehensive impact coefficient , . For this purpose, introduce the sine function to consider the volatility impact of environmental factors, and determine and ;

[0163] S66: Construct and fit a prediction model, define the predicted life as A, , which is the life of the device without additional environmental stress, , and Figure 3 is the error term. The judgment process of the abnormal change trend is as

[0164] By adopting the above technical solutions, through the device life data output by the prediction model, risk assessment is carried out, effectively identifying and preventing potential failure risks. According to the risk assessment results, targeted preventive measures are formulated and implemented to improve the reliability and safety of the device.

[0165] Based on the above embodiments, compared with the prior art, the present invention has the following advantages:

[0166] (1) By identifying abnormal data, extracting short-term abnormal data and long-term data, retaining more effective information, comprehensively analyzing the data characteristics, realizing a comprehensive assessment of the device state, and obtaining a more accurate dynamic threshold.

[0167] (2) Applying a self-learning dynamic threshold to replace the empirical alarm threshold in the monitoring system to ensure the accuracy and efficiency of early warning.

[0168] (3) Setting a threshold correction algorithm to obtain a more accurate dynamic threshold.

[0169] (4) Considering the setting strategy of early warning from multiple dimensions to ensure the accuracy and efficient response of the alarm signal.

[0170] (5) Predicting the trend of the cable insulation aging operation state is crucial for analyzing potential faults and planning maintenance strategies.

[0171] Based on the same inventive concept, the embodiment of the present application also provides a device for monitoring the state and predicting the life of a device for multi-data analysis for implementing the method for monitoring the state and predicting the life of a device for multi-data analysis involved above. The implementation solution provided by this device to solve the problem is similar to the implementation solution described in the above method. Therefore, the specific limitations in the embodiment of the device for monitoring the state and predicting the life of a device for multi-data analysis provided below can refer to the limitations on the method for monitoring the state and predicting the life of a device for multi-data analysis in the above text, and will not be repeated here.

[0172] Please refer to Figure 4 , the embodiment of the present invention provides a device for monitoring the state and predicting the life of a device for multi-data analysis, including:

[0173] A data acquisition module, which is used to acquire the partial discharge data, temperature data, water immersion data, and ambient temperature data of the cable joints of cable equipment within the window by using the window sliding technology;

[0174] A preprocessing module, which is used to preprocess the acquired data and fill in the missing values in the data;

[0175] An anomaly monitoring module, which is used to separately monitor the partial discharge data and temperature data within the window according to the anomaly monitoring threshold. If the data within the set time range is greater than the anomaly monitoring threshold, the data within the set time range is determined as short-term anomaly data, and the remaining data within the window is long-term data;

[0176] A dynamic threshold update module, which is used to dynamically adjust and update the size of the anomaly monitoring threshold according to the proportion of short-term anomaly data in long-term data when short-term anomaly data appears;

[0177] An equipment status warning module, which is used to quantify the abnormal change trend of the data within the window and use the quantified value of the abnormal change trend to warn the equipment status;

[0178] An equipment life prediction module, which is used to predict the remaining life of the cable equipment based on the water immersion data within the window, as well as the temperature data of the cable joints and the ambient temperature data, through a pre-constructed prediction model.

[0179] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above-mentioned division of each functional unit and module is used as an example. In practical applications, the above-mentioned functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above-mentioned system can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0180] Refer to Figure 5 , an embodiment of the present invention also provides a computer device, including: a memory, a processor, and a computer program stored on the memory. When the computer program is executed on the processor, it implements a method for monitoring and predicting the equipment status of multi-data analysis as described in any one of the above methods.

[0181] The computer device may be a computing device such as a desktop computer, a notebook, a handheld computer, and a cloud server. The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that Figure 5 merely examples of computer devices, which do not constitute a limitation on computer devices, may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0182] The so-called processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), 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.

[0183] The memory may be an internal storage unit of the computer device in some embodiments, such as the hard disk or memory of the computer device. The memory may also be an external storage device of the computer device in other embodiments, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device. Further, the memory may also include both the internal storage unit and the external storage device of the computer device. The memory is used to store an operating system, application programs, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory may also be used to temporarily store data that has been output or will be output.

[0184] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it implements a device state monitoring and life prediction method for multi-data analysis as described in any one of the above methods.

[0185] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above embodiment methods of this application, a computer program can be used to instruct the relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0186] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0187] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0188] In the embodiments disclosed in this application, it should be understood that the disclosed device / terminal device and method can be implemented in other ways. For example, the device / terminal device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, 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 displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in an electrical, mechanical or other forms.

[0189] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for equipment condition monitoring and life prediction based on multi-data analysis, characterized in that: The steps include: Use window sliding technology to obtain partial discharge data and temperature data of cable joints of cable equipment in the window as well as water immersion data and ambient temperature data; Preprocess the acquired data and fill in missing values ​​in the data; According to the abnormal monitoring threshold, the partial discharge data and the temperature data in the window are respectively monitored for abnormalities. If the data in the set time range is greater than the abnormal monitoring threshold, the data in the set time range is determined as short-term abnormal data, and the rest of the data in the window is long-term data; If the short-term abnormal data appears, dynamically adjust and update the abnormal monitoring threshold according to the proportion of the short-term abnormal data to the long-term data; Quantify the abnormal change trend of the data in the window, and use the quantified value of the abnormal change trend to warn the device status; Based on the water immersion data in the window and the temperature data of the cable joint and the ambient temperature data, a prediction is performed using a pre-built prediction model to obtain the remaining life of the cable equipment.

2. The device condition monitoring and life prediction method based on multi-data analysis according to claim 1 is characterized in that: In dynamically adjusting and updating the size of the abnormality monitoring threshold, for different types of data, the size of the abnormality monitoring threshold is dynamically adjusted and updated according to the following steps: For multiple groups of discontinuous short-term abnormal data appearing in the monitoring window, respectively calculating the mean and standard deviation of different types of data in each group of short-term abnormal data; The first thresholds of different types of data are calculated according to the mean and standard deviation of the short-term abnormal data as follows: In the formula, is the first threshold; Indicates short term, Indicates the data type, is a group sequence number index of multiple groups of short-term abnormal data; and are the mean and standard deviation respectively; is the first coefficient; Calculate the mean and standard deviation of each type of data in the long-term data; The second thresholds of different types of data are calculated according to the mean and standard deviation of the long-term data as follows: In the formula, is the second threshold; Indicates long-term; is the second coefficient; Calculate the proportion of each type of data in the short-term abnormal data to the total data respectively, and use the proportion as the weight of different types of data; The dynamic threshold is calculated according to the following formula, and the abnormal monitoring threshold is dynamically adjusted and updated according to the dynamic threshold: In the formula, is the dynamic threshold; is the corresponding weight; is a window data set; is the short-term abnormal data.

3. The device condition monitoring and life prediction method based on multi-data analysis according to claim 2 is characterized in that: Calculating the dynamic threshold also includes correcting the calculated dynamic threshold, and the correction process includes the following steps: Calculate the data mean difference of the same type of data between the window data set and the long-term data and between the window data set and the short-term abnormal data respectively to obtain the first type absolute error and the second type absolute error; Calculating the skewness of different types of data in the short-term abnormal data; Calculating the kurtosis of different types of data in the long-term data; The first nonlinear adjustment coefficient and the second nonlinear adjustment coefficient are calculated according to the following formula: ， In the formula, and are the first nonlinear adjustment coefficient and the second nonlinear adjustment coefficient respectively; is the empirical value, i=1, 2, 3, 4; is the skewness; is the kurtosis; The correction coefficients for different types of data are calculated according to the following formula: In the formula, is the correction factor; is the preset coefficient; and are the first type absolute error and the second type absolute error respectively; The dynamic thresholds for different types of data are modified according to the following formula: In the formula, is the modified dynamic threshold.

4. The device condition monitoring and life prediction method based on multi-data analysis according to claim 1 is characterized in that: For different types of data, the abnormal change trend of the data in the quantification window includes the following steps: Calculate the absolute value of the rate of change of data at each adjacent moment; Calculate the mean and standard deviation of the data, denoted as and ; The data change rate threshold is calculated according to the following formula: In the formula, is the change rate threshold; is the preset coefficient; Recording the moment when the absolute value of the change rate exceeds the change rate threshold as an abnormal moment; Calculate the volatility of the data within one day and the volatility of the data within the window near the abnormal moment respectively; The abnormal change trend of the data is quantified according to the following formula: In the formula, is the comprehensive quantitative score of abnormal change trend; Indicates data type; is the mean of the data in the window; is the volatility of the data in the window; For the The mean of the data within one day around the abnormal moment; For the The volatility of the data within a day around an abnormal moment; are determined values ​​obtained from experiments, i=1, 2, 3, 4.

5. The device condition monitoring and life prediction method based on multi-data analysis according to claim 4 is characterized in that: Using the quantified value of the abnormal change trend to warn the device status includes: The comprehensive quantitative score value is normalized according to the following formula: = In the formula, is the comprehensive quantitative score value after normalization; If the normalized comprehensive quantitative score value is greater than a set threshold, the device is determined to be abnormal.

6. The device condition monitoring and life prediction method based on multi-data analysis according to claim 1 is characterized in that: When constructing the prediction model, the relationship between the water immersion data, the temperature data, the ambient temperature data and the remaining life of the cable equipment is determined through a sine function fitting process, and the construction process includes: Considering the impact of temperature on the life of the device, the temperature impact coefficient is calculated based on the temperature data and the ambient temperature data as follows: In the formula, is the temperature influence coefficient; is the pre-exponential factor; is the activation energy; is the gas constant; is the temperature data; is the ambient temperature data; Considering the dual effects of temperature on humidity and equipment life, the humidity influence coefficient is calculated based on the water immersion data, the temperature data, and the ambient temperature data as follows: In the formula, is the humidity influence coefficient; is the basic humidity factor; is the coefficient of temperature difference; is the humidity level determined based on the flooding data; Considering the impact of the interaction between temperature and humidity on the life of the equipment, the comprehensive impact coefficient is calculated based on the temperature impact factor and the humidity impact factor, as follows: In the formula, is the comprehensive influence coefficient; The parameters in the comprehensive influence coefficient are determined by the fitting process of the sine function. and ; Build and fit a predictive model based on the coefficients.

7. The device condition monitoring and life prediction method based on multi-data analysis according to claim 6 is characterized in that: The mathematical expression of the prediction model is as follows: In the formula, The remaining life of the cable equipment; The life of the equipment without additional environmental stress; The temperature data and the ambient temperature data; is the flooding data; is the error term.

8. A multi-data analysis device for equipment status monitoring and life prediction, characterized in that: include: A data acquisition module, used to acquire partial discharge data and temperature data of cable joints of cable equipment in the window, as well as water immersion data and ambient temperature data by using a window sliding technology; A preprocessing module is used to preprocess the acquired data and fill in missing values ​​in the data; An abnormality monitoring module, used for performing abnormality monitoring on the partial discharge data and the temperature data in the window according to the abnormality monitoring threshold value, and if the data in the set time range is greater than the abnormality monitoring threshold value, the data in the set time range is determined as short-term abnormal data, and the rest of the data in the window is long-term data; A dynamic threshold updating module, used for dynamically adjusting and updating the size of the abnormal monitoring threshold according to the proportion of the short-term abnormal data to the long-term data when the short-term abnormal data appears; The device status early warning module is used to quantify the abnormal change trend of the data in the window and use the quantified value of the abnormal change trend to warn the device status; The equipment life prediction module is used to predict the remaining life of the cable equipment through a pre-built prediction model based on the water immersion data in the window and the temperature data of the cable joint and the ambient temperature data, so as to obtain the remaining life of the cable equipment.

9. A computer device, characterized in that: The device comprises a processor and a memory: The memory is used to store a computer program and send instructions of the computer program to the processor; The processor executes a multi-data analysis equipment condition monitoring and life prediction method according to any one of claims 1 to 7 according to the instructions of the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for equipment status monitoring and life prediction using multi-data analysis according to any one of claims 1 to 7 is implemented.

Images