An insulator flashover prediction method and device based on environmental factors

Abstract

The present invention discloses a method and device for predicting insulator flashover based on environmental factors, including obtaining historical environmental factor data affecting insulator flashover, preprocessing the historical environmental factor data to obtain preprocessed historical environmental factor data; constructing a risk probability change function for a single insulator according to the preprocessed historical environmental factor data; solving the risk probability change function for a single insulator to obtain an influence factor vector; constructing an insulator flashover prediction model based on environmental factors according to the influence factor vector and training the model; using the insulator flashover prediction model based on environmental factors to predict the flashover of the insulator to be predicted, obtaining the flashover risk degree of each insulator under the action of current environmental factors; and judging whether to carry out a preventive maintenance plan for the corresponding insulator according to the size of the flashover risk degree. The catenary insulator flashover prediction model of the present invention has a simple structure, easy-to-implement functions, and accurate and effective prediction.

Description

Technical Field

[0001] The present invention relates to the technical fields of catenary fault prevention and data mining, and particularly relates to a method and device for predicting insulator flashover based on environmental factors. Background Art

[0002] With the vigorous development of China's high-speed rail construction, the safety and reliability of the catenary system have attracted much attention. Insulators play the roles of insulation and mechanical support in the catenary system. Affected by the complex and changeable outdoor environmental factors, insulator flashover accidents occur continuously, which in turn trigger catenary system failures and cause consequences such as train delays.

[0003] To reduce catenary faults caused by insulator flashover, the human and material resources invested by each power supply section have increased year by year. For this fault, the existing solutions mainly involve replacing damaged insulators after analyzing the occurrence of catenary insulator flashover. However, there is a lack of accurate and effective insulator flashover prediction methods in the existing technology. Summary of the Invention

[0004] The technical problem to be solved by the present invention is the lack of accurate and effective insulator flashover prediction methods in the existing technology. The purpose is to provide a method and device for predicting insulator flashover based on environmental factors. The present invention starts from environmental factors for the first time, combines the historical fault records of the catenary, and realizes the prediction of the flashover probability of a single insulator. The catenary insulator flashover prediction model of the present invention has a simple structure and easy-to-implement functions, and the prediction is accurate and effective; moreover, the target object can be refined to a single insulator body, and the actual operability is strong; the present invention combines sampling methods, saves human and material resources compared with the existing maintenance methods, reduces the risk of catenary faults caused by insulator flashover with less resource consumption, and can significantly improve the maintenance efficiency of railway operation departments.

[0005] The present invention is realized through the following technical solutions:

[0006] In the first aspect, the present invention provides a method for predicting insulator flashover based on environmental factors, and the method includes:

[0007] Obtain historical environmental factor data affecting insulator flashover, and preprocess the historical environmental factor data to obtain preprocessed historical environmental factor data;

[0008] According to the preprocessed historical environmental factor data, construct a risk probability change function for a single insulator; solve the risk probability change function for a single insulator to obtain an influence factor vector; according to the influence factor vector, construct a prediction model for insulator flashover based on environmental factors and perform model training;

[0009] Use the insulator flashover prediction model based on environmental factors to predict the flashover of the insulator to be predicted, and obtain the flashover risk degree of each insulator under the action of current environmental factors; judge whether to carry out the preventive maintenance plan for the corresponding insulator according to the magnitude of the flashover risk degree.

[0010] The working principle is as follows: Due to the lack of accurate and effective insulator flashover prediction methods in the prior art, and the lack of research on the cause mechanism of insulator flashover in the catenary system at the present stage. The present invention starts from environmental factors for the first time and combines the historical fault records of the catenary to realize the prediction of the flashover probability of a single insulator. The present invention includes collecting and processing historical environmental factor data affecting insulator flashover, and using the proportional hazards regression COX model to construct a risk probability change function for a single insulator; solving the risk probability change function for the single insulator to obtain an influence factor vector, and using the influence factor vector as the environmental factor weight coefficient value of the prediction model; constructing an insulator flashover prediction model based on environmental factors according to the influence factor vector and training the model; finally, using the insulator flashover prediction model based on environmental factors to predict the flashover of the insulator to be predicted. Due to the complexity of railway operation conditions, the catenary working conditions in different regions have great differences. Collecting the relevant factors causing insulator flashover and analyzing their correlation with insulator flashover, and then giving early warnings with different treatments for insulators in different regions is an accurate and effective method.

[0011] The catenary insulator flashover prediction model of the present invention has a simple structure, easy-to-implement functions, accurate and effective prediction; and the target object can be refined to a single insulator body, with strong practical operability; the present invention combines sampling methods, saves manpower and material resources compared with the existing maintenance methods, reduces the risk of catenary faults caused by insulator flashover with less resource consumption, and can significantly improve the maintenance efficiency of railway operation departments.

[0012] Further, the environmental factors affecting insulator flashover include the first type of environmental factors and the second type of environmental factors;

[0013] The first type of environmental factors are meteorological pollution factors that cause dirt on the insulator surface;

[0014] The second type of environmental factors are factors that increase the flashover risk of dirt on the insulator.

[0015] Further, the first type of environmental factors include atmospheric salt density, atmospheric ash density, atmospheric temperature, and atmospheric relative humidity;

[0016] The second type of environmental factors include insulator wear degree and insulator service time.

[0017] Further, the preprocessing includes:

[0018] For the i-th type of environmental factor in the historical environmental factor data, zero-score standardization is used for preprocessing; the preprocessing formula is:

[0019]

[0020] where mean() represents calculating the mean value, and std() represents calculating the standard deviation; Q(t) is the t-th sample data, and T i (t) is the t-th training data.

[0021] Furthermore, the formula for the single insulator risk probability change function h(t, X) is:

[0022] h(t, X) = h0(t, X) exp(β1X1 + β2X2 +... + β n X n )

[0023] where X = (X1, X2,..., X n ) is the environmental factor vector of a single insulator; β i is the influence factor corresponding to the i-th environmental factor, reflecting the influence degree of this factor on insulator flashover; h0(t, X) is the function of the insulator failure risk changing with time t when all environmental factors are 0.

[0024] Furthermore, solving the single insulator risk probability change function to obtain the influence factor vector includes:

[0025] Performing likelihood and log-likelihood processing on the single insulator risk probability change function to obtain the log-likelihood function of the single insulator risk probability change function;

[0026] Using the Newton Raphson optimization algorithm or other mathematical optimization algorithms to perform maximum partial likelihood processing on the log-likelihood function of the single insulator risk probability change function to obtain the influence factor vector.

[0027] Furthermore, performing likelihood and log-likelihood processing on the single insulator risk probability change function to obtain the log-likelihood function of the single insulator risk probability change function includes:

[0028] Performing likelihood processing on the single insulator risk probability change function to obtain the likelihood function L;

[0029]

[0030] where N represents the total number of training sample insulators, X in represents the value of the n-th environmental factor of the i-th insulator, Xjn represents the value of the nth environmental factor of the jth insulator, where j represents the insulator sample whose environmental factor acquisition time is later than that of the ith insulator, and t i represents the acquisition time of the environmental factors of the ith insulator.

[0031] Perform logarithmic likelihood processing on the likelihood function L to obtain the log-likelihood function ln(L);

[0032]

[0033] where X in represents the value of the nth environmental factor of the ith insulator, and X jn represents the value of the nth environmental factor of the jth insulator, where j represents the insulator sample whose environmental factor acquisition time is later than that of the ith insulator, and t i represents the acquisition time of the environmental factors of the ith insulator; N is the total number of insulator samples.

[0034] Furthermore, the method further includes verifying the insulator flashover prediction model based on environmental factors through the Concordance Index (hereinafter referred to as C-index), and judging whether to reconstruct the insulator flashover prediction model according to the verification result; specifically:

[0035] When the concordance index C-index is less than or equal to the preset threshold, the insulator flashover prediction model based on environmental factors is completely invalid, discard the model and rebuild it;

[0036] When the concordance index C-index is greater than the preset threshold, the insulator flashover prediction model based on environmental factors is effective, and retain the model;

[0037] When the concordance index C-index is equal to 1, the insulator flashover prediction model based on environmental factors is completely correct;

[0038] The range of change of C-index is from 0 to 1.

[0039] Furthermore, judging whether to carry out the corresponding insulator preventive maintenance plan according to the magnitude of the flashover risk degree; including:

[0040] When the insulator with a flashover risk degree greater than 5 is recorded as a high-fault-risk insulator, arrange a preventive maintenance plan for the high-fault-risk insulator;

[0041] When the insulator with a flashover risk degree less than or equal to 5 is recorded as a low-fault-risk insulator, do not arrange a preventive maintenance plan for the low-fault-risk insulator.

[0042] In a second aspect, the present invention further provides an insulator flashover prediction device based on environmental factors, which supports the above-mentioned insulator flashover prediction method based on environmental factors; the device includes:

[0043] An acquisition unit, configured to acquire historical environmental factor data affecting insulator flashover;

[0044] A preprocessing unit, configured to preprocess the historical environmental factor data to obtain preprocessed historical environmental factor data;

[0045] A single insulator risk probability change function construction unit, configured to construct a single insulator risk probability change function according to the preprocessed historical environmental factor data;

[0046] An influence factor vector calculation unit, configured to solve the single insulator risk probability change function to obtain an influence factor vector;

[0047] An insulator flashover prediction model construction and training unit based on environmental factors, configured to construct an insulator flashover prediction model based on environmental factors according to the influence factor vector and perform model training;

[0048] A prediction unit, configured to use the insulator flashover prediction model based on environmental factors to perform flashover prediction on the insulator to be predicted, and obtain the flashover risk degree of each insulator under the action of current environmental factors; according to the magnitude of the flashover risk degree, determine whether to carry out a preventive maintenance plan for the corresponding insulator.

[0049] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0050] 1. For the first time, the present invention starts from environmental factors and combines historical fault records of the catenary to realize the prediction of the flashover probability of a single insulator, and the prediction is accurate and effective.

[0051] 2. The catenary insulator flashover prediction model of the present invention has a simple structure and easy-to-implement functions; moreover, the target object can be refined to a single insulator body, and the actual operability is strong; the present invention combines sampling methods, which saves manpower and material resources compared with the existing maintenance methods, reduces the risk of catenary faults caused by insulator flashovers with less resource consumption, and can significantly improve the maintenance efficiency of railway operation departments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of this application, and do not limit the embodiments of the present invention. In the drawings:

[0053] Figure 1 It is a flowchart of an insulator flashover prediction method based on environmental factors of the present invention.

[0054] Figure 2 This is a detailed flowchart of a method for predicting insulator flashover based on environmental factors according to the present invention.

[0055] Figure 3 This is a detailed diagram of the environmental factors of insulator flashover according to the present invention.

[0056] Figure 4 This is a schematic structural diagram of a device for predicting insulator flashover based on environmental factors according to the present invention. Detailed implementation manners

[0057] To make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with embodiments and drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0058] Embodiment 1

[0059] As Figures 1 to 3 shown, a method for predicting insulator flashover based on environmental factors according to the present invention, as Figure 1 and Figure 2 shown, the method includes:

[0060] Obtain historical environmental factor data affecting insulator flashover, and perform positive and negative sample data division and data preprocessing on the historical environmental factor data to obtain preprocessed historical environmental factor data;

[0061] According to the preprocessed historical environmental factor data, construct a risk probability change function for a single insulator; solve the risk probability change function for the single insulator to obtain an influence factor vector; according to the influence factor vector, construct a prediction model for insulator flashover based on environmental factors and perform model training;

[0062] Use the prediction model for insulator flashover based on environmental factors to predict the flashover of the insulator to be predicted, and obtain the flashover risk degree of each insulator under the action of the current environmental factors; according to the magnitude of the flashover risk degree, determine whether to carry out a preventive maintenance plan for the corresponding insulator.

[0063] In this embodiment, the environmental factors affecting insulator flashover include first-class environmental factors and second-class environmental factors;

[0064] The first-class environmental factors are meteorological pollution factors that cause dirt on the surface of the insulator;

[0065] The second-class environmental factors are factors that increase the flashover risk of the dirt on the insulator.

[0066] Specifically, the first type of environmental factors include atmospheric salt density, atmospheric ash density, atmospheric temperature, and atmospheric relative humidity;

[0067] The second type of environmental factors include insulator wear degree and insulator service time.

[0068] Among them, the detailed diagram of the insulator flashover environmental factors combining the above two types of factors is as Figure 3 shown.

[0069] In this embodiment, the division of the historical environmental factor data into positive and negative sample data includes:

[0070] Based on the historical environmental factor data, the posts prone to insulator flashover faults are used as positive samples and marked as 1; the posts without insulator flashover faults are used as negative samples and marked as 0 for comparison.

[0071] In this embodiment, the data preprocessing includes:

[0072] For the i-th type of environmental factors in the historical environmental factor data, zero-score standardization is used for preprocessing; the preprocessing formula is:

[0073]

[0074] where mean() represents calculating the mean value, std() represents calculating the standard deviation; Q(t) is the t-th sample data, and T i (t) is the t-th training data.

[0075] In this embodiment, the training and learning process of the influence factor vector (i.e., the environmental factor weight of the model) is as follows:

[0076] The degree of influence of each environmental factor on the insulator flashover in the catenary system is determined by the proportional hazards regression COX model. Assuming that the environmental factor "service time" in the previous steps is t, the risk probability change function h(t,X) of a single insulator is described by the following formula:

[0077] h(t,X) = h0(t,X)exp(β1X1 + β2X2 +... + β n X n )

[0078] where X = (X1, X2,..., X n ) is the environmental factor vector of a single insulator; β i is the influence factor corresponding to the i-th environmental factor, reflecting the degree of influence of this factor on the insulator flashover; h0(t,X) is the change function of the insulator failure risk with time t when all environmental factors are 0.

[0079] Let the total amount of sample data be N, and the time of occurrence of the fault of the i-th insulator be t i Then, the likelihood function L of the risk probability change function of the single insulator is as follows:

[0080]

[0081] Perform logarithmic likelihood processing on the likelihood function L to obtain the logarithmic likelihood function ln(L);

[0082]

[0083] where N represents the total number of insulator samples in the training sample, X in represents the value of the n-th environmental factor of the i-th insulator, X jn represents the value of the n-th environmental factor of the j-th insulator, j represents the insulator sample whose environmental factor collection time is later than that of the i-th insulator, and t i represents the environmental factor collection time of the i-th insulator.

[0084] Use the Newton Raphson optimization algorithm or other mathematical optimization algorithms to perform maximum partial likelihood processing on the logarithmic likelihood function of the risk probability change function of the single insulator, put the divided positive and negative sample data into the model for training, and obtain the influence factor vector The model completes training.

[0085] So far, after obtaining the influence factor vector (i.e., the environmental factor weight of the model), verify the insulator flashover prediction model based on environmental factors through the Concordance Index (hereinafter referred to as C-index), and judge whether to reconstruct the insulator flashover prediction model according to the verification result; specifically:

[0086] When the concordance index C-index is less than or equal to the preset threshold CI′, the insulator flashover prediction model based on environmental factors is completely invalid, discard the model and rebuild it;

[0087] When the concordance index C-index is greater than the preset threshold CI′, the insulator flashover prediction model based on environmental factors is valid, and the model is retained;

[0088] When the concordance index C-index is equal to 1, the insulator flashover prediction model based on environmental factors is completely correct;

[0089] where the C-index ranges from 0 to 1.

[0090] In this embodiment, the insulator flashover prediction model based on environmental factors is used to predict the flashover of the insulator to be predicted, and the flashover risk degree of each insulator under the action of current environmental factors is obtained, including:

[0091] (1) Preprocessing the relevant data of the insulator to be predicted;

[0092] According to the method of dividing the historical environmental factor data into positive and negative sample data and preprocessing the data, the corresponding temperature, relative humidity, salt density, ash density, wear degree, service time, etc. of the insulator to be predicted are obtained, and standardized processing is carried out according to the preprocessing formula.

[0093] (2) Calculating the flashover risk degree of each insulator under the action of current environmental factors;

[0094] Substitute the standardized weight of each insulator environmental factor (influence factor vector) into the following formula to obtain the relative risk degree of the insulator:

[0095] R = exp(β1X1 + β2X2 +... + β n X n )

[0096] Calculate the insulator flashover risk vector where R i represents the flashover risk degree of the i-th insulator.

[0097] Insulator flashover risk vector is the flashover risk degree vector of each insulator under the influence of various environmental factors.

[0098] In this embodiment, according to the magnitude of the flashover risk degree, it is judged whether to carry out the preventive maintenance plan for the corresponding insulator, including:

[0099] When the flashover risk degree of the insulator is greater than 5, it is recorded as a high-fault-risk insulator, and a preventive maintenance plan is arranged for the high-fault-risk insulator; the preventive maintenance plan is to clean the insulator or replace the insulator, etc.;

[0100] When the flashover risk degree of the insulator is less than or equal to 5, it is recorded as a low-fault-risk insulator, and no preventive maintenance plan is arranged for the low-fault-risk insulator.

[0101] The working principle is as follows: Due to the lack of accurate and effective insulator flashover prediction methods in the existing technology and the lack of research on the cause mechanism of insulator flashover in the catenary system at the present stage. The present invention starts from environmental factors for the first time, combines the historical fault records of the catenary, and realizes the prediction of the flashover probability of a single insulator. The present invention includes collecting and processing historical environmental factor data affecting insulator flashover, and constructing a risk probability change function of a single insulator by using the proportional hazards regression COX model; solving the risk probability change function of the single insulator to obtain an influence factor vector, and using the influence factor vector as the environmental factor weight coefficient value of the prediction model; constructing an insulator flashover prediction model based on environmental factors according to the influence factor vector and training the model; finally, using the insulator flashover prediction model based on environmental factors to predict the flashover of the insulator to be predicted. Due to the complexity of railway operation conditions, the catenary working conditions in different regions have great differences. Collecting the relevant factors causing insulator flashover and analyzing their correlation with insulator flashover, and then giving early warnings with differential treatments for insulators in different regions is an accurate and effective method. Specifically, differential treatment means that the environmental factor data of insulators installed in different anchor sections are significantly different, and the risk degrees evaluated by the prediction model of the present invention will show corresponding differences. Preventive maintenance is carried out on insulators with high risk degrees, and insulators with low risk degrees are not processed.

[0102] The catenary insulator flashover prediction model of the present invention has a simple structure, easy-to-implement functions, accurate and effective prediction; and the target object can be refined to a single insulator body, with strong practical operability; the present invention combines sampling methods, saves manpower and material resources compared with the existing maintenance methods, reduces the risk of catenary faults caused by insulator flashover with less resource consumption, and can significantly improve the maintenance efficiency of railway operation departments.

[0103] Embodiment 2

[0104] As Figure 4 shown, the difference between this embodiment and Embodiment 1 is that this embodiment provides an insulator flashover prediction device based on environmental factors, and this device supports the insulator flashover prediction method based on environmental factors described in Embodiment 1; this device includes:

[0105] An acquisition unit for acquiring historical environmental factor data affecting insulator flashover;

[0106] A preprocessing unit for dividing the historical environmental factor data into positive and negative sample data and preprocessing the data to obtain preprocessed historical environmental factor data;

[0107] A single insulator risk probability change function construction unit for constructing a single insulator risk probability change function according to the preprocessed historical environmental factor data;

[0108] An influence factor vector calculation unit, which is used to solve the risk probability change function of the single insulator to obtain an influence factor vector;

[0109] An insulator flashover prediction model construction and training unit based on environmental factors, which is used to construct an insulator flashover prediction model based on environmental factors and perform model training according to the influence factor vector;

[0110] A prediction unit, which is used to use the insulator flashover prediction model based on environmental factors to predict the flashover of the insulator to be predicted, and obtain the flashover risk degree of each insulator under the action of the current environmental factors; according to the magnitude of the flashover risk degree, it is judged whether to carry out the preventive maintenance plan for the corresponding insulator.

[0111] The execution process of each unit can be carried out according to the process steps of a method for predicting insulator flashover based on environmental factors described in Embodiment 1, and will not be elaborated one by one in this embodiment.

[0112] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0114] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device realizes the functions in Figure 1 one process or multiple processes and / or blocks Figure 1The functions specified in one or more boxes.

[0115] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in one Figure 1 one process or more processes and / or boxes Figure 1 or more boxes.

[0116] The specific embodiments described above further elaborate on the object, technical solution and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting insulator flashover based on environmental factors, characterized in that, The method includes: Obtaining historical environmental factor data affecting insulator flashover, and preprocessing the historical environmental factor data to obtain preprocessed historical environmental factor data; According to the preprocessed historical environmental factor data, constructing a risk probability change function for a single insulator; solving the risk probability change function for a single insulator to obtain an influence factor vector; according to the influence factor vector, constructing an insulator flashover prediction model based on environmental factors and performing model training; Using the insulator flashover prediction model based on environmental factors to predict the flashover of the insulator to be predicted, obtaining the flashover risk degree of each insulator under the action of current environmental factors; judging whether to carry out a preventive maintenance plan for the corresponding insulator according to the size of the flashover risk degree; The formula of the risk probability change function h(t,X) for a single insulator is: h(t,X) = h0(t,X)exp(β1X1 + β2X2 +... + β n X n ) Among them, X = (X1, X2,..., X n ) is the environmental factor vector of a single insulator; β i is the influence factor corresponding to the i-th environmental factor, reflecting the influence degree of this factor on the insulator flashover; h0(t, X) is the function of the insulator failure risk changing with time t when the environmental factors are all 0.

2. The insulator flashover prediction method based on environmental factors according to claim 1, wherein The environmental factors affecting insulator flashover include the first type of environmental factors and the second type of environmental factors; The first type of environmental factors are meteorological pollution factors that cause dirt on the surface of the insulator; The second type of environmental factors are factors that increase the flashover risk of the dirt on the insulator.

3. The insulator flashover prediction method based on environmental factors according to claim 2, characterized in that The first type of environmental factors include atmospheric salt density, atmospheric ash density, atmospheric temperature, and atmospheric relative humidity; The second type of environmental factors include insulator wear degree and insulator service time.

4. A method for predicting insulator flashover based on environmental factors according to claim 1, characterized in that The preprocessing includes: Preprocessing the i-th type of environmental factor data in the historical environmental factor data by zero-score standardization; the preprocessing formula is: Among them, mean() represents calculating the mean value, and std() represents calculating the standard deviation; Q(t) is the t-th sample data, and T i (t) is the t-th training data.

5. A method for predicting insulator flashover based on environmental factors according to claim 1, characterized in that The solving of the risk probability change function for a single insulator to obtain an influence factor vector includes: Performing likelihood and log-likelihood processing on the risk probability change function for a single insulator to obtain the log-likelihood function of the risk probability change function for a single insulator; Using the Newton Raphson optimization algorithm to perform maximum partial likelihood processing on the log-likelihood function of the risk probability change function for a single insulator to obtain the influence factor vector.

6. The insulator flashover prediction method based on environmental factors according to claim 5, characterized in that, The performing of likelihood and log-likelihood processing on the risk probability change function for a single insulator to obtain the log-likelihood function of the risk probability change function for a single insulator includes: Performing likelihood processing on the risk probability change function for a single insulator to obtain the likelihood function L; Performing log-likelihood processing on the likelihood function L to obtain the log-likelihood function ln(L); Among them, N represents the total number of training sample insulators, X in represents the value of the nth environmental factor of the ith insulator, X jn represents the value of the nth environmental factor of the jth insulator, where j represents an insulator sample whose environmental factor acquisition time is later than that of the ith insulator, and t i represents the environmental factor acquisition time of the ith insulator.

7. A method for predicting insulator flashover based on environmental factors according to claim 1, characterized in that, The method also includes verifying the insulator flashover prediction model based on environmental factors through a consistency index, and judging whether to reconstruct the insulator flashover prediction model according to the verification result; specifically: When the consistency index is less than or equal to a preset threshold, the insulator flashover prediction model based on environmental factors is invalid, discard the model and rebuild it; When the consistency index is greater than the preset threshold, the insulator flashover prediction model based on environmental factors is valid, and retain the model; When the consistency index is equal to 1, the insulator flashover prediction model based on environmental factors is completely correct.

8. A method for predicting insulator flashover based on environmental factors according to claim 1, characterized in that The judging whether to carry out a preventive maintenance plan for the corresponding insulator according to the size of the flashover risk degree includes: The insulator with a flashover risk degree greater than 5 is denoted as a high-fault-risk insulator, and a preventive maintenance plan is arranged for the high-fault-risk insulator; The insulator with a flashover risk degree less than or equal to 5 is denoted as a low-fault-risk insulator, and no preventive maintenance plan is arranged for the low-fault-risk insulator.

9. An insulator flashover prediction device based on environmental factors, characterized in that, The device supports a method for predicting insulator flashover based on environmental factors as described in any one of claims 1 to 8; the device includes: An acquisition unit for acquiring historical environmental factor data affecting insulator flashover; A preprocessing unit for preprocessing the historical environmental factor data to obtain preprocessed historical environmental factor data; A single-insulator risk probability change function construction unit for constructing a single-insulator risk probability change function according to the preprocessed historical environmental factor data; An influence factor vector calculation unit for solving the single-insulator risk probability change function to obtain an influence factor vector; An insulator flashover prediction model construction and training unit based on environmental factors for constructing an insulator flashover prediction model based on environmental factors and performing model training according to the influence factor vector; A prediction unit for using the insulator flashover prediction model based on environmental factors to perform flashover prediction on the insulator to be predicted, obtaining the flashover risk degree of each insulator under the action of current environmental factors; and judging whether to perform a preventive maintenance plan for the corresponding insulator according to the magnitude of the flashover risk degree; The formula for the single-insulator risk probability change function h(t,X) is: h(t,X) = h0(t,X)exp(β1X1 + β2X2 +... + β n X n ) where X = (X1, X2,..., X n ) is the environmental factor vector of a single insulator; β i is the influence factor corresponding to the i-th environmental factor, reflecting the influence degree of this factor on insulator flashover; h0(t, X) is the function of the insulator failure risk varying with time t when the environmental factors are all 0.

Images