Track circuit compensation capacitor centralized maintenance period determination method and device

Abstract

The invention discloses a track circuit compensation capacitor centralized maintenance period determination method and device. The method comprises the following steps: acquiring environment information, fault information and position information of each compensation capacitor; performing association processing on the environment information and the fault information to generate a feature vector of each compensation capacitor; dividing the compensation capacitor into a plurality of sub-domains based on the influence of classification variables in the feature vectors on the fault rate of the compensation capacitor; according to the historical fault data of each sub-domain, fitting to obtain a fault rate function of the compensation capacitor in each sub-domain; collecting maintenance cost data of the compensation capacitor in each sub-domain; constructing a target function taking the maintenance cost data, the fault quantity and the risk loss as optimization targets; and according to a minimization result of the target function, determining a target centralized maintenance period of each sub-domain. According to the invention, the accuracy and economical efficiency of track circuit compensation capacitor maintenance are improved, and the waste of maintenance resources and the fault risk are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of track circuit compensation capacitors, and in particular to a method and device for determining a centralized maintenance period for track circuit compensation capacitors. Background Art

[0002] This section is intended to provide a background or context for the presented embodiments of the invention. No admission is made that the description herein is prior art by virtue of its inclusion in this section.

[0003] In track circuit systems, compensation capacitors are core devices that ensure stable signal transmission. They compensate for the rail inductance parameters to ensure the effective transmission distance of the track circuit. According to current railway operation and maintenance standards, when the track section length exceeds 300 meters, compensation capacitors must be installed at fixed intervals (such as 60 / 80 meters). Due to the non-repairable nature of compensation capacitors, failures can only be repaired by replacement. Therefore, railway departments generally adopt a hybrid maintenance strategy of "fault replacement + regular centralized replacement", in which the centralized maintenance cycle is uniformly set at 5 years.

[0004] However, actual operation and maintenance data shows that the failure rate of compensation capacitors is significantly affected by environmental factors (such as tunnel moisture corrosion and extreme weather temperature fluctuations), line conditions (such as bridge vibration and curve stress), and equipment batch quality differences, resulting in significant differences in the actual life distribution of compensation capacitors in different regions. The unified centralized maintenance cycle of existing technologies has the following shortcomings:

[0005] 1. Waste of resources: In areas with low failure rates (such as dry, straight roads), equipment that has not reached its life cycle is replaced prematurely, resulting in wasteful equipment procurement and labor costs;

[0006] 2. Risks: In areas with high failure rates (such as high-vibration bridges and rainy tunnels), fixed cycles cannot respond to rising failure rate trends in a timely manner. Concentrated failures may trigger red light bands in track circuits, threatening driving safety.

[0007] 3. Extensive cost control: Existing strategies do not quantify the correlation between environmental factors and failure rates, nor do they incorporate multiple objectives such as maintenance costs and failure risk losses into the periodic decision-making model, resulting in insufficient O&M economic efficiency.

[0008] Therefore, there is an urgent need for a dynamic maintenance cycle decision-making method that can combine the actual service status of the equipment, environmental factors and multi-dimensional cost data to achieve precise and differentiated compensation capacitor operation and maintenance management. Summary of the Invention

[0009] An embodiment of the present invention provides a method for determining a centralized maintenance period for track circuit compensation capacitors, which is used to improve the accuracy and economy of track circuit compensation capacitor maintenance and reduce resource waste and failure risks caused by a unified maintenance period. The method includes:

[0010] obtain environment information, fault information and location information of each compensation capacitor; the environment information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition;

[0011] perform association processing on the environment information and the fault information to generate a feature vector of each compensation capacitor; the feature vector includes location coordinates, a road condition vector, a fault state mark, a fault time vector and a weather condition vector;

[0012] divide the compensation capacitor into a plurality of sub-domains based on the influence of classification variables in the feature vector on the failure rate of the compensation capacitor; the compensation capacitors in each sub-domain have the same failure rate distribution;

[0013] fit a failure rate function of the compensation capacitor in each sub-domain according to historical fault data of each sub-domain; the failure rate function is used to predict the failure rate of the compensation capacitor over time;

[0014] collect maintenance cost data of the compensation capacitor in each sub-domain; the maintenance cost data includes device purchase cost, transportation cost, personnel cost and fault risk loss cost; a target function is constructed with the maintenance cost data, the number of faults and the risk loss as optimization objectives; the target function calculates the comprehensive cost by counting the number of centralized maintenance, the number of fault replacement and the failure rate of the remaining time period in each sub-domain within a statistical time window;

[0015] determine a target centralized maintenance period of each sub-domain according to the minimization result of the target function.

[0016] The embodiment of the application also provides a track circuit compensation capacitor centralized maintenance period determination device to improve the accuracy and economy of track circuit compensation capacitor maintenance, and reduce resource waste and fault risk caused by unified maintenance period.

[0017] The device includes: a device information acquisition module, configured to obtain environment information, fault information and location information of each compensation capacitor; the environment information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition;

[0018] a feature vector generation module, configured to perform association processing on the environment information and the fault information to generate a feature vector of each compensation capacitor; the feature vector includes location coordinates, a road condition vector, a fault state mark, a fault time vector and a weather condition vector;

[0019] a sub-domain division module, configured to divide the compensation capacitors into a plurality of sub-domains based on the influence of the classification variables in the feature vectors on the failure rate of the compensation capacitors; the compensation capacitors in each sub-domain have the same failure rate distribution;

[0020] a failure rate function fitting module, configured to fit a failure rate function of the compensation capacitors in each sub-domain according to historical failure data of each sub-domain; the failure rate function is used to predict the failure rate of the compensation capacitors changing with time;

[0021] a target function determination module, configured to collect maintenance cost data of the compensation capacitors in each sub-domain; the maintenance cost data includes equipment purchase cost, transportation cost, personnel cost and failure risk loss cost; a target function is constructed with the maintenance cost data, the failure number and the risk loss as optimization targets; the target function is used to calculate the comprehensive cost by counting the centralized maintenance times, the failure replacement number and the failure rate of the remaining time period of each sub-domain in a statistical time window;

[0022] a target centralized maintenance cycle determination module, configured to determine the target centralized maintenance cycle of each sub-domain according to the minimization result of the target function.

[0023] The embodiment of the present application also provides a computer device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor; when the processor executes the computer program, the track circuit compensation capacitor centralized maintenance cycle determination method is realized.

[0024] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program; when the processor executes the computer program, the track circuit compensation capacitor centralized maintenance cycle determination method is realized.

[0025] The embodiment of the present application also provides a computer program product, which comprises a computer program; when the processor executes the computer program, the track circuit compensation capacitor centralized maintenance cycle determination method is realized.

[0026] In an embodiment of the present invention, environmental information, fault information and location information of each compensation capacitor are obtained; the environmental information includes road conditions and weather conditions, and the fault information includes equipment fault status, fault occurrence time and corresponding weather conditions; the environmental information and fault information are correlated to generate a feature vector of each compensation capacitor; the feature vector includes location coordinates, road condition vector, fault status mark, fault time vector and weather condition vector; based on the influence of the classification variables in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple subdomains; the compensation capacitors in each subdomain have the same failure rate distribution; according to each The historical failure data of the subdomains are fitted to obtain the failure rate function of the compensating capacitors in each subdomain; the failure rate function is used to predict the failure rate of the compensating capacitors over time; the maintenance cost data of the compensating capacitors in each subdomain are collected; the maintenance cost data include equipment purchase cost, transportation cost, personnel cost and failure risk loss cost; an objective function is constructed with maintenance cost data, number of failures and risk loss as optimization targets; the objective function calculates the comprehensive cost by counting the number of centralized maintenance times, the number of failure replacements and the failure rate of the remaining time period in each subdomain within the statistical time window; and the target centralized maintenance period for each subdomain is determined based on the minimization result of the objective function. The present invention collects data on road conditions, weather, location, and failure history of compensation capacitors to construct a multidimensional feature vector, dynamically correlating environmental variables with failure rates. This overcomes the limitation of traditional methods that ignore environmental differences. Variance analysis of categorical variables such as road conditions and weather conditions is used to divide subdomains, identify groups with significantly different failure rates, and lay the foundation for independent calculation of maintenance cycles for different subdomains. A comprehensive objective function is constructed that includes equipment procurement, transportation, labor, and risk losses to quantify the total cost, number of failures, and risk under different cycles. This balance between economy and safety is achieved by minimizing the objective function. Based on the subdomain failure rate function and cost model, an optimal centralized maintenance cycle is generated for each subdomain, avoiding premature replacement in low-failure-rate areas while providing timely intervention in high-failure-rate areas, fundamentally addressing the drawbacks of a "one-size-fits-all" strategy. This allows the site to appropriately adjust the centralized maintenance cycle of compensation capacitors based on the operating status of the compensation capacitor equipment, avoiding premature replacement of well-performing compensation capacitor equipment, reducing unnecessary equipment procurement and replacement costs, improving the accuracy and economy of track circuit compensation capacitor maintenance, and reducing the waste of resources and failure risks caused by a unified maintenance cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0028] Figure 1 A flowchart of a track circuit compensation capacitor centralized maintenance period determination method in an embodiment of the present application;

[0029] Figure 2 A compensation capacitor information collection and processing implementation flowchart in an embodiment of the present application;

[0030] Figure 3 A compensation capacitor sub-domain classification implementation flowchart in an embodiment of the present application;

[0031] Figure 4 A track circuit compensation capacitor centralized maintenance period decision flowchart in an embodiment of the present application;

[0032] Figure 5 A compensation capacitor centralized maintenance period decision method based on fault data in an embodiment of the present application;

[0033] Figure 6 A structure schematic diagram of a track circuit compensation capacitor centralized maintenance period determination device in an embodiment of the present application;

[0034] Figure 7 A computer equipment schematic diagram for track circuit compensation capacitor centralized maintenance period determination in an embodiment of the present application. DETAILED DESCRIPTION

[0035] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, further detailed description of the embodiments of the present application is given below in conjunction with the drawings. Herein, the illustrative embodiments of the present application and the description thereof are used to explain the present application, but not as a limitation of the present application.

[0036] The term "and / or" herein is merely used to describe an association relationship, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B and C can mean including any one or more elements selected from the set consisting of A, B and C.

[0037] In the description of the present specification, "include", "including", "have", "has", and the like are open terms, that is, mean to include but not limited to. The description of the terms "one embodiment", "one specific embodiment", "some embodiments", "for example", and the like means that the specific features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, or characteristics described can be combined in any one or more embodiments or examples in an appropriate manner. The order of the steps involved in each embodiment is used to illustrate the implementation of the present application, and the order of the steps is not limited. The order can be appropriately adjusted as needed.

[0038] The acquisition, storage, use, processing, etc. of data in the technical solutions of the present application comply with relevant regulations. The information collected in the present application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure, and application of relevant data comply with relevant standards, necessary security measures are taken, do not violate public order and good customs, and appropriate operation portals are provided for users to choose authorization or refusal.

[0039] It should be noted that in the embodiments of the present application, some existing industry solutions such as software, components, models, etc. may be mentioned, such as some existing software tools, components, algorithm models, or other widely known solutions in the technical field, which should be considered as exemplary, and the purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application. These references should be understood as typical examples, and the core purpose is to explain and verify the rationality and feasibility of the implementation of the technical solutions proposed in the present application. However, it does not mean that the applicant has or will necessarily use the solution. Such a reference does not imply that the applicant has actually adopted these existing solutions or will necessarily adopt these methods in the future implementation of the technology. In other words, these references only serve the purpose of illustration and help understand the relevance and transcendence of the innovative points of the present application over the prior art, and do not constitute an acknowledgment or dependence statement of a specific existing technical product.

[0040] The compensation capacitor is an important device to ensure the normal operation of the non-insulated track circuit, and is used to improve the transmission conditions of the track circuit signal on the steel rail line. When the length of the track circuit section is greater than 300m, compensation capacitors are required in principle according to the equal spacing principle (60 / 80m). Compensation capacitor devices are widely used in railways, and compensation capacitor operation and maintenance management is one of the important daily work of the relevant departments of the railway.

[0041] The compensation capacitor belongs to non-repairable equipment. When the compensation capacitor fails, due to its own structure and technical characteristics, it is not suitable to carry out traditional repair work, but to replace it directly. For a number of compensation capacitor devices used at the same time, as the use time increases, the number of compensation capacitor failures will increase year by year, so the operation and maintenance unit will replace all the compensation capacitor devices in a certain area according to a certain period, that is, the compensation capacitor centralized maintenance, and the maintenance period is generally 5 years. Therefore, the current domestic railway site generally adopts the combination of fault maintenance and centralized maintenance to manage the compensation capacitor equipment. However, in recent years, a large number of studies have pointed out that the service life of the compensation capacitor equipment is affected by a variety of factors, and the number of failures is significantly different under the influence of different factors such as line, weather conditions and road conditions. In the current operation and maintenance strategy, if all compensation capacitor devices are replaced with a unified centralized replacement cycle, for some devices in good operating condition and high quality, centralized replacement work is carried out when the failure rate is very low, which undoubtedly causes waste of resources and is contrary to the concept of cost reduction and efficiency improvement; For some areas with high failure rate of compensation capacitor, the frequency of centralized maintenance of compensation capacitor may need to be appropriately increased to avoid the impact of the normal transmission of track circuit caused by the centralized failure of compensation capacitor in extreme cases.

[0042] At the beginning of the line opening, the compensation capacitor equipment installed on the line is basically the same model produced by the same manufacturer. Based on the failure rate obtained from the failure information, the current service status of the batch of compensation capacitors can be effectively reflected. When considering maintenance work, all kinds of costs and workloads involved need to be weighed, and the actual service status of the compensation capacitor equipment is combined to deeply explore the cost investment under the flexible centralized maintenance cycle. Through accurate calculation, the optimal centralized maintenance cycle is determined to solve the operation and maintenance difficulties caused by uneven distribution of compensation capacitor life from the root.

[0043] Specifically, the unified centralized maintenance cycle of the prior art has the following defects:

[0044] 1. Resource waste: for areas with low failure rate (such as dry and straight lines), devices that have not reached the life cycle are replaced in advance, causing waste of device procurement and labor costs;

[0045] 2. Risk hidden danger: for areas with high failure rate (such as high-vibration bridges and rainy tunnels), the fixed cycle cannot respond to the rising trend of the failure rate, and the track circuit red light zone may be caused by centralized failure, which threatens the safety of train operation;

[0046] 3. Cost control is extensive: the existing strategy does not quantify the relevance of environmental factors to the failure rate, nor does it include multiple targets such as maintenance cost and failure risk loss in the cycle decision model, resulting in insufficient operation and maintenance economy.

[0047] Therefore, a dynamic maintenance cycle decision method capable of combining the actual service state of equipment, environmental factors and multi-dimensional cost data is urgently needed to realize precise and differentiated compensation capacitor operation and maintenance management.

[0048] To solve the above problems, the embodiment of the application provides a track circuit compensation capacitor centralized maintenance cycle determination method to improve the precision and economy of track circuit compensation capacitor maintenance, and reduce resource waste and fault risks caused by unified maintenance cycle. Figure 1 , Figure 1 The flowchart of the track circuit compensation capacitor centralized maintenance cycle determination method in the embodiment of the application can include the following steps:

[0049] Step 101: Obtain the environmental information, fault information and location information of each compensation capacitor; the environmental information includes road conditions and weather conditions, and the fault information includes equipment fault state, fault occurrence time and corresponding weather conditions;

[0050] Step 102: Perform association processing on the environmental information and fault information to generate a feature vector of each compensation capacitor; the feature vector includes location coordinates, road condition vectors, fault state markers, fault time vectors and weather condition vectors;

[0051] Step 103: Based on the influence of classification variables in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple sub-domains; the compensation capacitors in each sub-domain have the same failure rate distribution;

[0052] Step 104: According to the historical fault data of each sub-domain, a failure rate function of the compensation capacitor in each sub-domain is fitted; the failure rate function is used to predict the failure rate of the compensation capacitor over time;

[0053] Step 105: Collect the maintenance cost data of the compensation capacitor in each sub-domain; the maintenance cost data includes equipment purchase cost, transportation cost, personnel cost and fault risk loss cost; a target function is constructed with maintenance cost data, fault number and risk loss as optimization objectives; the target function calculates the comprehensive cost by counting the centralized maintenance times, fault replacement numbers and failure rates of the remaining time period of each sub-domain in the statistical time window;

[0054] Step 106: According to the minimization result of the target function, the target centralized maintenance cycle of each sub-domain is determined.

[0055] In the embodiment of the present application, the environmental information, fault information and position information of each compensation capacitor are acquired; the environmental information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition; the environmental information and fault information are associated to generate a feature vector of each compensation capacitor; the feature vector includes position coordinates, road condition vector, fault state mark, fault time vector and weather condition vector; based on the influence of classification variables in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple sub-domains; the compensation capacitors in each sub-domain have the same failure rate distribution; based on the historical fault data of each sub-domain, a failure rate function of the compensation capacitor in each sub-domain is fitted; the failure rate function is used to predict the failure rate of the compensation capacitor over time; the repair cost data of the compensation capacitor in each sub-domain is collected; the repair cost data includes device purchase cost, transportation cost, personnel cost and fault risk loss cost; a target function with repair cost data, fault number and risk loss as optimization objectives is constructed; the target function calculates the comprehensive cost by counting the number of centralized repairs, the number of fault replacements and the failure rate of the remaining time period of each sub-domain in a statistical time window; based on the minimization result of the target function, the target centralized repair period of each sub-domain is determined. The embodiment of the present application collects the road condition, weather, position and fault history data of the compensation capacitor, constructs a multi-dimensional feature vector, dynamically associates the environmental variables with the failure rate, and breaks through the limitations of traditional methods that ignore environmental differences; based on the variance analysis of classification variables such as road condition and weather condition, the sub-domains are divided, the groups with significant differences in failure rate are identified, and the foundation for independently calculating the repair period of different sub-domains is laid; by constructing a comprehensive target function including device purchase, transportation, labor and risk loss, the total cost, fault number and risk under different periods are quantified, and the balance between economy and safety is achieved by minimizing the target function; based on the failure rate function and cost model of the sub-domain, the optimal centralized repair period for each sub-domain is generated, avoiding premature replacement in low failure rate areas, and timely intervention in high failure rate areas, fundamentally solving the drawbacks of the "one-size-fits-all" strategy. Thus, the compensation capacitor centralized repair period is adjusted according to the state of the compensation capacitor device to avoid premature replacement of the compensation capacitor device in good operation, reduce unnecessary equipment procurement and replacement costs, improve the precision and economy of the track circuit compensation capacitor maintenance, and reduce resource waste and fault risk caused by uniform repair period.

[0056] The present application relates to the field of track circuit compensation capacitor, in particular to a method for determining the centralized repair period of track circuit compensation capacitor. Through the method, the centralized repair period of compensation capacitor in a specific area can be planned in combination with the fault information, environmental information and cost information of the compensation capacitor, providing a scientific basis for promoting condition-based maintenance.

[0057] In implementation, first, step 101 is performed: obtaining the environmental information, fault information and location information of each compensation capacitor; the environmental information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition.

[0058] In one embodiment, in the data collection stage, for the compensation capacitor device deployed in the track circuit system, the full life cycle data of each compensation capacitor is comprehensively collected through the sensor network deployed along the railway, the operation and maintenance record database and the meteorological monitoring device. Specifically, the environmental information is obtained through the line geographic information system and real-time monitoring device, including two types of parameters of road condition and weather condition at the location of the compensation capacitor:

[0059] The road condition includes the track structure characteristics of the section where the compensation capacitor is located, such as whether it is located inside the tunnel, bridge section, curve area or straight section, as well as the physical environmental indicators such as vibration intensity of the line and drainage performance of the track bed;

[0060] The weather condition includes the temperature, humidity, rainfall, historical records and real-time monitoring data of extreme weather events (such as heavy rain, snow and ice, high temperature) at the location of the compensation capacitor, which is stored in alignment with the device operation time axis in units of days.

[0061] The fault information is derived from the fault alarm record and on-site repair report of the railway electric system management system, specifically including:

[0062] Device fault state: records whether the compensation capacitor has occurred overcapacity, lead line fracture, shell damage and other fault types, and is stored in binary flag form;

[0063] Fault occurrence time: uses the device online operation timestamp as a reference to record the cumulative number of days of fault occurrence, which is used to associate the aging of the device with the action time of the external environment;

[0064] Corresponding weather condition: at the moment of fault occurrence, real-time weather data and previous continuous weather changes at the location are synchronously extracted, such as temperature and humidity fluctuations within 72 hours before the fault, rainfall duration, etc., which are used to analyze the correlation between fault inducement and environmental conditions.

[0065] The location information is obtained by the Beidou satellite navigation system positioning device, which records the latitude and longitude coordinates of each compensation capacitor, and maps it to the line mileage mark position combined with the line electronic map, which is used to associate the line slope and curve radius of the section where the compensation capacitor is located. The above data is preprocessed after collection, such as missing value filling and outlier correction, to ensure data integrity and consistency, providing reliable input for subsequent analysis.

[0066] In the implementation, in step 101, the environment information, fault information and position information of each compensation capacitor are acquired; the environment information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition; then in step 102, the environment information and fault information are associated to generate a feature vector of each compensation capacitor; the feature vector includes position coordinates, road condition vector, fault state mark, fault time vector and weather condition vector.

[0067] In the embodiment, the environment information and fault information are associated to generate a feature vector of each compensation capacitor, including:

[0068] The road condition of the compensation capacitor is converted into a binary coding vector, each type of road condition corresponds to an independent dimension, if the compensation capacitor belongs to the type of road condition, 1 is marked, otherwise 0 is marked;

[0069] The latitude and longitude position coordinates of the compensation capacitor are marked;

[0070] Whether the compensation capacitor has occurred fault is marked, if it has occurred fault, 1 is marked, otherwise 0 is marked;

[0071] The fault occurrence time is recorded, and 0 is marked when no fault occurs;

[0072] The weather condition when the fault occurs is converted into a binary coding vector, each type of weather corresponds to an independent dimension, if the fault occurs in the type of weather, 1 is marked, otherwise 0 is marked;

[0073] The marked vectors are spliced into a complete feature vector.

[0074] In an embodiment, in the data preprocessing stage, the collected environment information and fault information are structured and integrated to generate a feature vector that can comprehensively represent the state of the compensation capacitor device. The specific processing process is as follows:

[0075] Firstly, for the road condition where the compensation capacitor is located, a classification list covering all preset road condition types is established, such as tunnel, bridge, curve, straight section, high vibration area, water accumulation area, etc. For each compensation capacitor device, a binary coding vector is generated according to the actual road condition type it is located in, each dimension of the vector corresponds to a preset road condition category, if the device belongs to the category, the corresponding dimension is marked as 1, otherwise it is marked as 0, so as to realize the discrete mathematical expression of the road condition.

[0076] Secondly, the precise geographical position of the compensation capacitor is obtained by the positioning device, the longitude and latitude values are recorded, and the coordinates are mapped to the mileage marker position of the electronic map of the railway line, which is used to associate the line design parameters and the equipment operating environment. For the equipment failure state, a binary flag is used to represent it. If the equipment has failed during service, it is marked as 1, and if it has not failed, it is marked as 0. The failure time is recorded in terms of cumulative days from the date of the equipment going online, and the equipment that has not failed is set to 0, thereby reflecting the relationship between the service length and the failure time sequence of the equipment.

[0077] In weather condition processing, the types of weather that may affect equipment failure are defined in advance, such as heavy rain, high temperature, low temperature, ice and snow, and continuous overcast, etc. The weather state at the time of failure is determined, and a binary coding vector similar to the road condition is generated. Each dimension corresponds to a weather type. If the weather condition exists at the time of failure, it is marked as 1, otherwise it is marked as 0. For equipment that has not failed, the dimensions of the weather condition vector are all set to 0.

[0078] Finally, the above road condition coding vector, longitude and latitude coordinates, failure state flag, failure time and weather condition coding vector are spliced in a fixed order to form a multi-dimensional feature vector. The vector contains the spatial position of the equipment, the environmental exposure characteristics, the historical failure performance and the time sequence information, which provides a structured data basis for subsequent analysis of the quantitative correlation between different environmental factors and failure rate.

[0079] In the implementation, after the step 102 of associating the environmental information and the failure information to generate a feature vector for each compensation capacitor, the feature vector includes position coordinates, road condition vector, failure state flag, failure time vector and weather condition vector, the step 103 of dividing the compensation capacitor into multiple sub-domains based on the influence of the classification variable in the feature vector on the failure rate of the compensation capacitor is performed. The compensation capacitors in each sub-domain have the same failure rate distribution.

[0080] In one embodiment, dividing the compensation capacitor into multiple sub-domains based on the influence of the classification variable in the feature vector on the failure rate of the compensation capacitor includes:

[0081] Grouping different compensation capacitors according to the road condition vector, the failure state flag, the failure time vector and the weather condition vector in the feature vector;

[0082] Performing variance analysis on the failure rate of each group of compensation capacitors to calculate the ratio of inter-group variance to intra-group variance;

[0083] If the ratio exceeds a preset significance threshold, it is determined that the classification variable has an influence on the failure rate, and the variable is used as the basis for dividing the sub-domains;

[0084] Compensation capacitors with the same classification variable attribute are classified into the same sub-domain, and the failure rates of compensation capacitors in the same sub-domain obey the same distribution.

[0085] In an embodiment, in the sub-domain division process, first, all compensation capacitor devices in the region are multi-dimensionally grouped according to classification variables in the feature vector, such as road condition, failure state, failure time and weather condition. In specific implementation, the category marked as 1 in the road condition vector is taken as the grouping basis, for example, compensation capacitors located in tunnels, bridges and curves are classified into different groups respectively, and the weather condition vector at the time of failure is processed in the same way, for example, heavy rain weather group, high temperature weather group, etc. For the failure state mark and failure time data, hierarchical sampling is performed according to whether the device has failed and the period of failure (such as the initial, middle and late periods of service), to form a set of devices with similar failure characteristics.

[0086] After the preliminary grouping is completed, the failure rates of compensation capacitors in each group are statistically analyzed, and the variance analysis method is used to quantify the influence degree of the classification variable on the failure rate. The specific operation is as follows: the variance of the failure rates between groups is calculated to measure the dispersion degree of the failure rates of different groups; the variance of the failure rates within each group is calculated synchronously to reflect the fluctuation range of the failure rates of the devices within the group. The ratio of the inter-group variance to the intra-group variance is taken as a statistic, which is compared with a preset significance threshold. If the ratio exceeds the threshold, it indicates that the current classification variable has a significant influence on the failure rate, for example, the failure rate of compensation capacitors in the tunnel environment is significantly higher than that in other groups.

[0087] At this time, the classification variable is taken as the core basis for sub-domain division, and compensation capacitor devices with the same attribute value are classified into the same sub-domain, for example, all devices located in the tunnel form an independent sub-domain, and all devices that have failed in heavy rain form another sub-domain. The devices in the same sub-domain share the same classification variable attribute, and their failure rate change trend and distribution law are determined to be consistent. Subsequently, the same failure rate function can be used for modeling, thereby ensuring the applicability and calculation efficiency of the maintenance strategy within the sub-domain.

[0088] In specific implementation, after the compensation capacitors in each sub-domain have the same failure rate distribution, step 104 is performed: according to the historical failure data of each sub-domain, a failure rate function of the compensation capacitors in each sub-domain is fitted; the failure rate function is used to predict the failure rate of the compensation capacitors over time.

[0089] In an embodiment, according to the historical failure data of each sub-domain, a failure rate function of the compensation capacitors in each sub-domain is fitted, including:

[0090] The historical failure data of each sub-domain is fitted by a probability distribution model or a regression model;

[0091] A time-varying failure rate prediction function is generated for calculating the failure probability of the compensation capacitor in the sub-domain at any time point.

[0092] In an embodiment, in the failure rate modeling stage, for each historical failure data set of the sub-domain, a statistical analysis method is used to construct a mathematical model reflecting the change of the device failure rate over time. In specific implementation, first, the failure time series data of the compensation capacitor in the sub-domain is cleaned and arranged, and abnormal data points caused by recording errors or external interference are eliminated, and the complete time record of the device from being put into operation to failure is retained, and the current cumulative running time is taken as the censored data for the device that has not failed. According to the distribution characteristics of the sub-domain data, a suitable probability distribution model or a regression model is selected for parameter fitting: for the sub-domain showing obvious device aging law, a Weibull distribution model or a lognormal distribution model is preferentially used for fitting, and the shape parameter and the scale parameter are calculated by the maximum likelihood estimation method; for the sub-domain with linearly increasing failure rate over time, an exponential distribution model is used to describe its constant failure rate characteristics; for the sub-domain with non-linear fluctuation of failure rate, a polynomial regression model or a segmented regression model is selected to capture its change trend.

[0093] After the model fitting is completed, the goodness of fit of the model to the historical data is evaluated by statistical test methods, such as using residual sum of squares, Akaike information criterion, etc. for optimization, to ensure that the selected model can accurately represent the dynamic change law of the failure rate of the compensation capacitor in the sub-domain. Based on the finally determined model type and parameters, a failure rate prediction function is generated with the device running time as the independent variable, which can output the failure probability value of the compensation capacitor in the sub-domain at any running time point, providing a quantitative basis for subsequent maintenance cost calculation and cycle optimization.

[0094] In addition, for the sub-domain with multiple factor interactions, environmental factors (such as cumulative rainfall, temperature fluctuation amplitude) can be introduced as covariates based on the time variable to construct a proportional hazards model, further refining the accuracy of failure rate prediction.

[0095] In specific implementation, after step 104: fitting the failure rate function of the compensation capacitor in each sub-domain according to the historical failure data of each sub-domain; the failure rate function is used to predict the failure rate of the compensation capacitor over time, step 105 is performed: collecting the maintenance cost data of the compensation capacitor in each sub-domain; the maintenance cost data includes device purchase cost, transportation cost, personnel cost and failure risk loss cost; a target function is constructed with maintenance cost data, failure number and risk loss as optimization objectives; the target function calculates the comprehensive cost by counting the number of centralized maintenance, the number of failure replacement and the failure rate of the remaining time period in each sub-domain within the statistical time window.

[0096] In one embodiment, a target function is constructed to optimize maintenance cost data, the number of failures and risk loss, including:

[0097] A long-term statistical time window is set, and the number of centralized maintenance times and the length of the remaining time period of the sub-domain within the time window are calculated; the maintenance cost data is: the sum of the batch replacement cost corresponding to the number of centralized maintenance times, the single fault replacement cost in each centralized maintenance period and the single fault replacement cost of the remaining time period; the risk loss is: the sum of the economic loss of the track circuit red light band caused by failure within the centralized maintenance period and the remaining time period; the number of failures is: the sum of the expected number of failed devices corresponding to the number of centralized maintenance times and the expected number of failures in the remaining time period;

[0098] The maintenance cost data, risk loss and number of failures are combined according to the preset weight to form the target function.

[0099] In one embodiment, in the process of constructing the target function, first, a long enough statistical time window is set to simulate the maintenance cost and risk loss of the compensation capacitor in the long-term running scenario. For each sub-domain, according to the preset candidate centralized maintenance period, the number of centralized maintenance times that can be completely executed within the time window is calculated, and the length of the remaining uncovered time period after the last centralized maintenance is calculated.

[0100] The total maintenance cost is composed of three parts: one is the batch replacement cost corresponding to the number of centralized maintenance times, including the device purchase cost of all compensation capacitors in the sub-domain, centralized transportation and operation personnel cost; the second is the single replacement cost caused by device failure within each centralized maintenance period, which is calculated by the product of single fault maintenance cost and expected number of failures within the period; the third is the single replacement cost caused by device failure within the remaining time period, which is calculated according to the integral result of the product of the remaining time length and the failure rate function.

[0101] The risk loss is obtained by quantifying the economic impact of the failure of the compensation capacitor on the red light band of the track circuit, and the specific calculation method is: within each centralized maintenance period and the remaining time period, the failure rate function is integrated with the unit time loss cost function to obtain the total economic loss caused by failure in each time period. The total number of failures is the sum of the expected number of failed devices corresponding to all centralized maintenance periods within the statistical time window, plus the expected number of failures in the remaining time period, wherein the expected number of failures is calculated by integrating the failure rate function in the corresponding time interval.

[0102] Finally, the total maintenance cost, risk loss and total failure number are multiplied by the preset weight coefficient respectively, and then linearly superimposed to form the objective function of multi-objective optimization. The weight coefficient is dynamically adjusted according to the priority of the operation and maintenance management department in cost control, risk avoidance or failure number constraint, for example, increasing the cost weight when the equipment procurement budget is tight, and increasing the risk loss weight when the train safety requirement is strict, so as to realize the flexible adaptation of the maintenance cycle decision.

[0103] In a specific implementation, after step 105 of constructing the objective function with maintenance cost data, failure number and risk loss as optimization objectives, step 106 of determining the target set maintenance cycle of each sub-domain according to the minimization result of the objective function is performed.

[0104] In one embodiment, determining the target set maintenance cycle of each sub-domain according to the minimization result of the objective function comprises:

[0105] For each sub-domain, a plurality of pending set maintenance cycles are preset for the sub-domain;

[0106] The target function value corresponding to each pending set maintenance cycle is calculated, and the pending set maintenance cycle that minimizes the target function value is selected as the target set maintenance cycle of the sub-domain.

[0107] In one embodiment, in the maintenance cycle decision stage, a group of candidate set maintenance cycles are preset for each sub-domain, and the value range of the candidate cycle is determined based on historical operation and maintenance experience, equipment manufacturer recommended life and variation trend of the sub-domain failure rate function, for example, generating candidate values in the range of 1 to 10 years in units of years with a preset step length. For each candidate cycle, according to the length of the statistical time window, the complete set maintenance times and the remaining time period under the cycle are calculated, and then the target function model constructed is called, the failure rate function, maintenance cost parameters and candidate cycle value of the current sub-domain are input, and the corresponding target function calculation result is output, which represents the comprehensive operation and maintenance cost, risk loss and failure number weighted evaluation value under the candidate cycle.

[0108] By traversing all candidate cycles, a mapping relationship table of each cycle and the target function value is generated, from which the candidate cycle that minimizes the target function value is selected and determined as the optimal set maintenance cycle of the sub-domain. To verify the robustness of the decision result, the Monte Carlo simulation method can be further combined to perform sensitivity analysis within the fluctuation range of the failure rate parameter, ensuring that the selected cycle remains optimal under the influence of equipment life and environmental conditions uncertainty. The final output of the maintenance cycle set of each sub-domain provides the basis for differentiated and implementable set maintenance plans for the railway operation and maintenance department, realizing the upgrade of the maintenance strategy from experience-driven to data-driven.

[0109] A specific embodiment is given below to illustrate the specific application of the method of the present application. The purpose of this embodiment is to provide a track circuit compensation capacitor centralized maintenance cycle decision method to help the field flexibly formulate the compensation capacitor centralized maintenance cycle according to the needs. The maintenance strategy design aims to build a maintenance cycle model, use real line data as the driving, use the total cost, the total replacement number and the like as the objective function, explore the cost payment situation under different centralized maintenance cycles, select the appropriate maintenance cycle, and then provide a reference for the formulation of the real maintenance strategy.

[0110] The compensation capacitor centralized maintenance cycle decision method of the present application is based on the idea of multi-parameter fusion, and comprehensively considers various factors affecting the maintenance cycle of the compensation capacitor, including device failure rate, environmental factors (such as long tunnels, extreme weather and the like affecting the actual life of the compensation capacitor), cost factors (such as device cost, transportation cost, personnel cost, loss cost) and the like. The total process of this embodiment is shown in Figure 5 Figure 5 The total flowchart of a compensation capacitor centralized maintenance cycle decision method based on fault data in the embodiment of the present application is shown in

[0111] Step S1: For a total number of I compensation capacitor devices (which can be simply referred to as compensation capacitors) in a specific area, collect information including the road conditions d i , the device location p i , whether the device has ever failed h i , the device failure time t i , the fault weather condition q i and the like for each compensation capacitor device i (i≤I).

[0112] Step S2: Perform a pretreatment operation on the collected data, associate the compensation capacitor environmental information and fault information, and form a feature vector for each compensation capacitor device. The processing flow is shown in Figure 2 , and Figure 2 is a compensation capacitor information collection and processing implementation flowchart in the embodiment of the present application.

[0113] Step S3: Extract the features related to the compensation capacitor failure from the feature vector obtained in step S2, identify the influence of various environmental factor features on the compensation capacitor failure rate distribution, divide all the compensation capacitor devices into N sub-domains according to the N kinds of failure rate distributions met by the compensation capacitor devices in a specific area, and the processing flow is shown in Figure 3 , and Figure 3 is a compensation capacitor sub-domain classification implementation flowchart in the embodiment of the present application.

[0114] ​Step S4: fitting the compensation capacitor failure rate in the N sub-domains according to the collected failure data, and obtaining the failure rate of the compensation capacitor in each sub-domain n within a certain period.

[0115] Step S5: collecting the maintenance cost data of the I n compensation capacitor in the current sub-domain n, including the equipment purchase cost, the equipment transportation cost in the failure replacement operation, the personnel cost, and the loss cost involved in the track circuit red light band caused by the compensation capacitor failure.

[0116] Step S6: constructing a maintenance strategy model with the total cost, the total compensation capacitor failure number, and the risk loss as the objective function, and calculating the cost involved in the different centralized maintenance cycle plans for all sub-domains within a certain period.

[0117] Step S7: based on different compensation capacitor maintenance scenarios, comprehensively considering the compensation capacitor maintenance management status of each unit and the line conditions, calculating a reasonable maintenance cycle, and assisting in formulating a corresponding maintenance strategy, the processing flow of steps S5-S7 is shown in Figure 4 , Figure 4 which is a flow chart for the centralized maintenance cycle decision of the compensation capacitor of the track circuit in the embodiment of the application.

[0118] The specific processing process of step S2 is as follows:

[0119] Step S21: d i records the road condition of the compensation capacitor i. In order to obtain d i , the compensation capacitor road condition needs to be vectorized. Assuming that there are N d kinds of road conditions, the road condition d i can be vectorized as follows:

[0120]

[0121]

[0122] Step S22: p i records the longitude and latitude of the compensation capacitor i, and p i =[longitude latitude].

[0123] Step S23: h i records whether the compensation capacitor i device has failed. If the compensation capacitor i has failed, h i =1, otherwise 0.

[0124] Step S24: t i records the failure time of the compensation capacitor i, which is the number of days elapsed between the failure time and the online time of the compensation capacitor device. If the compensation capacitor i has not failed, t i= 0.

[0125] Step S25: q i Record the weather conditions when the compensation capacitor i fails, in order to obtain q i , the weather conditions need to be vectorized, assuming that there are N q weather conditions, the weather conditions can be vectorized as:

[0126]

[0127] Step S26: The compensation capacitor i feature vector can be represented as follows:

[0128]

[0129] The specific processing process of step S3 is:

[0130] Step S31: Group the compensation capacitor devices according to the feature vector, and determine whether there is a significant difference in the compensation capacitor failure rate between different groups through variance analysis. If the inter-group variance is significantly greater than the intra-group variance, it means that the classification variable (influencing factor) has a significant impact on the compensation capacitor failure rate, and the I compensation capacitors are divided into N sub-domains according to the classification result, and the number of compensation capacitor devices in each sub-domain is I n , and the failure rate of all compensation capacitors in each sub-domain is considered to conform to the same distribution, and the same centralized maintenance period is applicable.

[0131] Step S4 fits the compensation capacitor failure rate curve in each maintenance unit through the fault data set, which can include but is not limited to polynomial fitting, exponential distribution fitting, Weibull distribution fitting, etc. In any centralized maintenance period T n , for each sub-domain n, construct a function f with time as the independent variable, which can obtain the failure rate ξ under the time scale, which can be represented as follows:

[0132] ζ n (t) = f n (t), 0 < t ≤ T n

[0133] The specific processing process of step S5 is:

[0134] Step S51: Record the maintenance cost of replacing a part of the compensation capacitor that fails in the time period [0, t] in sub-domain n with a single replacement at any time scale t, which involves equipment purchase cost c v , transportation cost c t and personnel cost c s , which can be represented as follows:

[0135]

[0136] Step S52: Record the maintenance cost of the centralized replacement of the compensation capacitors in the whole sub-domain n, involving the equipment purchase cost c v , the centralized maintenance operation transportation cost c t2 , and the centralized maintenance operation personnel cost c s2 , which can be expressed as follows:

[0137]

[0138] Step S53: Record the risk loss cost involved in the red light band of the track circuit in the sub-domain n caused by the failure of the compensation capacitors in the time period [0, T], and construct a function F with time as the independent variable, which can obtain the risk loss cost at any time by combining the failure rate function f

[0139]

[0140] The specific processing process of step S6 is as follows:

[0141] Step S61: Determine a very long statistical time window H, then perform several centralized replacement operations in [0, H], and the remaining time should be less than one centralized replacement period, let the remaining time be R n , and its value should be:

[0142]

[0143] Similarly, the number of centralized maintenance operations Z n in [0, H] can be expressed as follows:

[0144]

[0145] Step S62: Q n Record the total number of compensation capacitor equipment replaced in [0, H] in the sub-domain n, including failure replacement and centralized replacement, and its value is:

[0146] Q n = I n · Z n · (1 + ζ n (T n )) + ζ n (R n )

[0147] Step S63: U is the objective function of the compensation capacitor maintenance period decision model, which is composed of the total compensation capacitor maintenance cost, risk loss cost and total compensation capacitor fault number in all sub-domains in the statistical time window H, and w1, w2 and w3 represent the weights of different categories respectively.

[0148]

[0149] Step S7: the maintenance period T is determined by the objective function U, and the weights w1, w2 and w3 can be set to meet the maintenance needs and concerns of different on-site units.

[0150] The embodiment can realize accurate cost control: the method provided by the application can assist on-site to appropriately adjust the compensation capacitor centralized maintenance period according to the compensation capacitor equipment operation state, avoid replacing the well-operating compensation capacitor equipment too early, reduce unnecessary equipment procurement and replacement cost, and reduce resource waste. For the area with high failure rate, the maintenance is reasonably arranged according to the accurate maintenance period decision, and excessive maintenance is avoided.

[0151] In addition, the weights of the objective function of the method provided by the application can be flexibly set, so that the maintenance period decision can be applied to on-site users who pay more attention to cost reduction, operation efficiency guarantee or workload reduction, so as to obtain the maintenance period decision most suitable for the user needs, and has strong universality and adaptability.

[0152] Of course, it can be understood that the above detailed process can also have other change examples, and the related change examples shall fall within the protection scope of the application.

[0153] The embodiment of the application also provides a track circuit compensation capacitor centralized maintenance period determination device, which is expressed in the following embodiment. Since the principle of solving the problem of the device is similar to that of the track circuit compensation capacitor centralized maintenance period determination method, the implementation of the device can be referred to the implementation of the track circuit compensation capacitor centralized maintenance period determination method, and the repeated parts will not be described herein.

[0154] The embodiment of the application also provides a track circuit compensation capacitor centralized maintenance period determination device, which is expressed in the following embodiment. Since the principle of solving the problem of the device is similar to that of the track circuit compensation capacitor centralized maintenance period determination method, the implementation of the device can be referred to the implementation of the track circuit compensation capacitor centralized maintenance period determination method, and the repeated parts will not be described herein. Figure 6 The structure diagram of the track circuit compensation capacitor centralized maintenance period determination device is shown in the following embodiment. Figure 6 As shown in the figure, the device comprises:

[0155] The device information acquisition module 601 is configured to acquire environmental information, fault information and position information of each compensation capacitor, wherein the environmental information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition.

[0156] The feature vector generation module 602 is configured to perform association processing on the environmental information and the fault information, and generate a feature vector of each compensation capacitor, wherein the feature vector includes position coordinates, a road condition vector, a fault state mark, a fault time vector and a weather condition vector.

[0157] The sub-domain division module 603 is configured to divide the compensation capacitor into a plurality of sub-domains based on the influence of classification variables in the feature vector on the fault rate of the compensation capacitor, and the compensation capacitors in each sub-domain have the same fault rate distribution.

[0158] The fault rate function fitting module 604 is configured to fit a fault rate function of the compensation capacitor in each sub-domain according to historical fault data of each sub-domain, and the fault rate function is used to predict the fault rate of the compensation capacitor changing with time.

[0159] The objective function determination module 605 is configured to collect maintenance cost data of the compensation capacitor in each sub-domain, wherein the maintenance cost data includes device purchase cost, transportation cost, personnel cost and fault risk loss cost, and an objective function is constructed with the maintenance cost data, fault number and risk loss as optimization objectives, and the objective function is used to calculate comprehensive cost by counting the number of centralized maintenance, the number of fault replacement and the fault rate of the remaining time period in each sub-domain within a statistical time window.

[0160] The target centralized maintenance cycle determination module 606 is configured to determine a target centralized maintenance cycle of each sub-domain according to the minimization result of the objective function.

[0161] In one embodiment, the association processing on the environmental information and the fault information to generate the feature vector of each compensation capacitor includes:

[0162] The road condition of the compensation capacitor is converted into a binary coding vector, each type of road condition corresponds to an independent dimension, and if the compensation capacitor belongs to the type of road condition, 1 is marked, otherwise 0 is marked.

[0163] The latitude and longitude position coordinates of the compensation capacitor are marked.

[0164] Whether the compensation capacitor has occurred a fault is marked, if the fault has occurred, 1 is marked, otherwise 0 is marked.

[0165] The fault occurrence time is recorded, and 0 is marked when no fault occurs.

[0166] The weather condition when the fault occurs is converted into a binary coded vector, each type of weather corresponds to an independent dimension, if it belongs to this type of weather when the fault occurs, it is marked as 1, otherwise 0;

[0167] The marked vector is spliced into a complete feature vector.

[0168] In one embodiment, based on the influence of the classification variable in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple sub-domains, including:

[0169] According to the road condition vector, fault state mark, fault time vector and weather condition vector in the feature vector, different compensation capacitors are grouped;

[0170] The failure rate of each group of compensation capacitors is subjected to variance analysis, and the ratio of inter-group variance to intra-group variance is calculated;

[0171] If the ratio exceeds a preset significance threshold, it is determined that the classification variable has an influence on the failure rate, and the variable is used as a sub-domain division basis;

[0172] Compensation capacitors with the same classification variable attribute are grouped into the same sub-domain, and the failure rates of compensation capacitors in the same sub-domain obey the same distribution.

[0173] In one embodiment, according to the historical failure data of each sub-domain, a failure rate function of the compensation capacitor in each sub-domain is fitted, including:

[0174] The historical failure data of each sub-domain is fitted by using a probability distribution model or a regression model;

[0175] A failure rate prediction function varying with time is generated, which is used to calculate the failure probability of the compensation capacitor in the sub-domain at any time point.

[0176] In one embodiment, a target function with maintenance cost data, failure number and risk loss as optimization objectives is constructed, including:

[0177] A long-term statistical time window is set, and the number of centralized maintenance and the length of the remaining time period of the sub-domain in the time window are calculated; the maintenance cost data is: the sum of the batch replacement cost corresponding to the number of centralized maintenance, the single failure replacement cost in each centralized maintenance period and the single failure replacement cost in the remaining time period; the risk loss is: the sum of the economic loss of the track circuit red light band caused by the fault in the centralized maintenance period and the remaining time period; the failure number is: the sum of the expected number of fault devices corresponding to the number of centralized maintenance and the expected number of faults in the remaining time period;

[0178] The maintenance cost data, risk loss and failure number are combined into the target function according to the preset weight.

[0179] In one embodiment, according to the minimization result of the target function, the target set maintenance period of each sub-domain is determined, including:

[0180] For each sub-domain, a plurality of pending set maintenance periods are preset for the sub-domain;

[0181] The target function value corresponding to each pending set maintenance period is calculated, and the pending set maintenance period that minimizes the target function value is selected as the target set maintenance period of the sub-domain.

[0182] Embodiments of the present application provide an embodiment of a computer device for implementing all or part of the above track circuit compensation capacitor set maintenance period determination method, and the computer device specifically includes the following content:

[0183] A processor, a memory, a communications interface, and a bus; wherein the processor, the memory, and the communications interface complete mutual communication through the bus; the communications interface is used for realizing information transmission between related devices; the computer device can be a desktop computer, a tablet computer, a mobile terminal, and the like, and the embodiments are not limited thereto. In the embodiments, the computer device can be implemented by referring to the embodiments of the track circuit compensation capacitor set maintenance period determination method and the embodiments of the track circuit compensation capacitor set maintenance period determination device, the contents of which are incorporated herein, and repeated descriptions are not repeated.

[0184] Figure 7 A schematic diagram of the computer device for track circuit compensation capacitor set maintenance period determination in the embodiments of the present application is shown in FIG. 1. As shown in FIG. 1, the computer device 1000 can include a central processor 1001 and a memory 1002; the memory 1002 is coupled to the central processor 1001. It is worth noting that the computer device 1000 is exemplary; other types of structures can also be used to supplement or replace the structure to realize telecommunication functions or other functions. Figure 7 Figure 7

[0185] In one embodiment, the track circuit compensation capacitor set maintenance period determination function can be integrated into the central processor 1001. The central processor 1001 can be configured to perform the following control:

[0186] Obtain the environment information, fault information, and location information of each compensation capacitor; the environment information includes road conditions and weather conditions, and the fault information includes device fault states, fault occurrence times, and corresponding weather conditions;

[0187] ​​The environment information and the fault information are associated to generate a feature vector of each compensation capacitor; the feature vector includes a position coordinate, a road condition vector, a fault state label, a fault time vector, and a weather condition vector;

[0188] Based on the influence of the classification variable in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into a plurality of sub-domains; the compensation capacitors in each sub-domain have the same failure rate distribution;

[0189] According to the historical fault data of each sub-domain, a failure rate function of the compensation capacitor in each sub-domain is fitted; the failure rate function is used to predict the failure rate of the compensation capacitor over time;

[0190] Maintenance cost data of the compensation capacitor in each sub-domain is collected; the maintenance cost data includes equipment purchase cost, transportation cost, personnel cost, and fault risk loss cost; a target function is constructed with the maintenance cost data, the number of faults, and the risk loss as the optimization target; the target function calculates the comprehensive cost by counting the number of centralized maintenance, the number of fault replacement, and the failure rate of the remaining time period in each sub-domain within a statistical time window;

[0191] According to the minimization result of the target function, a target centralized maintenance period of each sub-domain is determined.

[0192] In another embodiment, the track circuit compensation capacitor centralized maintenance period determination device can be configured separately from the central processor 1001, for example, the track circuit compensation capacitor centralized maintenance period determination device can be configured as a chip connected with the central processor 1001, and the track circuit compensation capacitor centralized maintenance period determination function is realized through the control of the central processor.

[0193] As shown in Figure 7 , the computer device 1000 can also include a communication module 1003, an input unit 1004, an audio processor 1005, a display 1006, and a power supply 1007. It is worth noting that the computer device 1000 does not necessarily include all the components shown in Figure 7 ; in addition, the computer device 1000 can also include components not shown in Figure 7 , which can refer to prior art.

[0194] As shown in Figure 7 , the central processor 1001, also known as a controller or an operation control, can include a microprocessor or other processor device and / or a logic device, which receives input and controls the operation of each component of the computer device 1000.

[0195] The memory 1002, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. Information related to the device described above can be stored, and in addition, programs for executing the related information can be stored. The central processing unit 1001 can execute the programs stored in the memory 1002 to achieve information storage or processing, etc.

[0196] The input unit 1004 provides input to the central processing unit 1001. The input unit 1004 is, for example, a key or a touch input device. The power supply 1007 is used to provide power to the computer device 1000. The display 1006 is used to display display objects such as images and text. The display can be, for example, an LCD display, but is not limited thereto.

[0197] The memory 1002 can be a solid state memory such as a read only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROM, etc. The memory 1002 can also be some other type of device. The memory 1002 includes a buffer memory 1021 (sometimes referred to as a buffer). The memory 1002 can include an application / function storage section 1022 for storing application programs and function programs or for storing a flow for executing the operation of the computer device 1000 by the central processing unit 1001.

[0198] The memory 1002 can also include a data storage section 1023 for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the computer device. A driver storage section 1024 of the memory 1002 can include various drivers of the computer device for communication functions and / or for executing other functions of the computer device such as a messaging application, an address book application, etc.

[0199] The communication module 1003 is a transmitter / receiver that transmits and receives signals via the antenna 1008. The communication module (transmitter / receiver) 1003 is coupled to the central processing unit 1001 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0200] Based on different communication technologies, multiple communication modules 1003, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc., can be provided in the same computer device. The communication module (transmitter / receiver) 1003 is also coupled to a speaker 1009 and a microphone 1010 via an audio processor 1005 to provide audio output via the speaker 1009 and to receive audio input from the microphone 1010 to enable typical telecommunication functions. The audio processor 1005 can include any suitable buffers, decoders, amplifiers, etc. In addition, the audio processor 1005 is coupled to the central processor 1001 to enable recording on-board via the microphone 1010 and to enable playing stored sounds on-board via the speaker 1009.

[0201] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the track circuit compensation capacitor centralized maintenance cycle determination method.

[0202] The embodiment of the present application further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the track circuit compensation capacitor centralized maintenance cycle determination method.

[0203] In the embodiment of the present application, the environmental information, fault information and position information of each compensation capacitor are acquired; the environmental information includes road condition and weather condition, and the fault information includes device fault state, fault occurrence time and corresponding weather condition; the environmental information and fault information are associated to generate a feature vector of each compensation capacitor; the feature vector includes position coordinates, road condition vector, fault state mark, fault time vector and weather condition vector; based on the influence of classification variables in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple sub-domains; the compensation capacitors in each sub-domain have the same failure rate distribution; according to the historical fault data of each sub-domain, a failure rate function of the compensation capacitor in each sub-domain is fitted; the failure rate function is used to predict the failure rate of the compensation capacitor over time; the repair cost data of the compensation capacitor in each sub-domain is collected; the repair cost data includes device purchase cost, transportation cost, personnel cost and fault risk loss cost; a target function with repair cost data, fault number and risk loss as optimization objectives is constructed; the target function calculates the comprehensive cost by counting the centralized repair times, fault replacement numbers and failure rates of the remaining time period of each sub-domain in a time window; according to the minimization result of the target function, the target centralized repair period of each sub-domain is determined. In the embodiment of the present application, the road condition, weather, position and fault history data of the compensation capacitor are collected, a multi-dimensional feature vector is constructed, the environmental variables are dynamically associated with the failure rate, and the limitations of traditional methods that ignore environmental differences are broken through; the sub-domains are divided based on variance analysis of classification variables such as road condition and weather condition, the groups with significant differences in failure rate are identified, and the foundation for independently calculating the repair period of different sub-domains is laid; by constructing a comprehensive target function including device purchase, transportation, labor and risk loss, the total cost, fault number and risk under different periods are quantified, and the balance between economy and safety is realized by minimizing the target function; based on the failure rate function and cost model of the sub-domain, the optimal centralized repair period of each sub-domain is generated, the replacement of the compensation capacitor in the low failure rate area is avoided, and the high failure rate area is intervened in time, thereby fundamentally solving the disadvantages of the "one-size-fits-all" strategy. Thus, the compensation capacitor centralized repair period is adjusted according to the state of the compensation capacitor device to avoid replacing the compensation capacitor device in good operation, reduce unnecessary equipment procurement and replacement costs, improve the precision and economy of the track circuit compensation capacitor repair, and reduce resource waste and fault risk caused by the unified repair period.

[0204] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0205] The present application is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0206] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0207] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0208] The specific embodiments described above are examples for purposes of explanation and illustration and are not intended to limit the scope of the application. The scope of the application is limited only by the claims that follow.

Claims

1. A method for determining the centralized maintenance period of track circuit compensation capacitors, characterized in that: include: Obtain environmental information, fault information and location information of each compensation capacitor; The environmental information includes road conditions and weather conditions, and the fault information includes equipment fault status, fault occurrence time and corresponding weather conditions; Correlating the environmental information with the fault information to generate a feature vector for each compensation capacitor; The feature vector includes position coordinates, road condition vector, fault status mark, fault time vector and weather condition vector; Based on the influence of the classification variables in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple sub-domains; the compensation capacitors in each sub-domain have the same failure rate distribution; According to the historical failure data of each sub-domain, a failure rate function of the compensation capacitor in each sub-domain is obtained by fitting; the failure rate function is used to predict the failure rate of the compensation capacitor over time; Collect maintenance cost data for compensation capacitors in each subdomain; the maintenance cost data includes equipment purchase cost, transportation cost, personnel cost, and failure risk loss cost; construct an objective function with maintenance cost data, number of failures, and risk loss as optimization targets; the objective function calculates the comprehensive cost by statistically analyzing the number of centralized maintenance times, the number of failure replacements, and the failure rate in the remaining time period for each subdomain within a time window; According to the minimization result of the objective function, the target centralized maintenance period of each subdomain is determined.

2. The method according to claim 1, wherein The environmental information and the fault information are correlated to generate a feature vector for each compensation capacitor, including: The road condition of the compensation capacitor is converted into a binary coded vector, where each road condition corresponds to an independent dimension. If the compensation capacitor belongs to this road condition, it is marked as 1, otherwise it is 0. Mark the latitude and longitude coordinates of the compensation capacitor; Mark whether the compensation capacitor has failed. If so, it is marked as 1, otherwise it is marked as 0. Record the time when the fault occurs, and mark it as 0 if no fault occurs; The weather conditions at the time of the fault are converted into binary coded vectors, where each type of weather corresponds to an independent dimension. If the fault occurs in that type of weather, it is marked as 1, otherwise it is 0. Concatenate the labeled vectors into a complete feature vector.

3. The method according to claim 1, wherein Based on the influence of the categorical variables in the feature vector on the failure rate of the compensation capacitor, the compensation capacitor is divided into multiple subdomains, including: Different compensation capacitors are grouped according to the road condition vector, fault state mark, fault time vector and weather condition vector in the feature vector; Perform variance analysis on the failure rate of each group of compensation capacitors and calculate the ratio of the between-group variance to the within-group variance; If the ratio exceeds a preset significance threshold, it is determined that the classification variable has an impact on the failure rate, and this variable is used as the basis for subdomain division; Compensation capacitors with the same classification variable attributes are classified into the same subdomain, and the failure rates of compensation capacitors in the same subdomain obey the same distribution.

4. The method according to claim 1, wherein Based on the historical failure data of each sub-domain, the failure rate function of the compensation capacitor in each sub-domain is fitted, including: For each subdomain’s historical fault data, a probability distribution model or regression model is used to fit the data. A time-varying failure rate prediction function is generated to calculate the failure probability of the compensation capacitor in the subdomain at any time point.

5. The method according to claim 1, wherein Construct an objective function with maintenance cost data, number of failures, and risk loss as optimization targets, including: A long-term statistical time window is set, and the number of centralized maintenance times for the subdomain within the time window and the length of the remaining time period are calculated. The maintenance cost data is: the sum of the batch replacement cost corresponding to the number of centralized maintenance times, the replacement cost for a single fault within each centralized maintenance cycle, and the single fault replacement cost for the remaining time period. The risk loss is: the sum of the economic losses caused by the red light band of the track circuit due to faults within the centralized maintenance cycle and the remaining time period. The number of faults is: the sum of the expected number of faulty devices corresponding to the number of centralized maintenance times and the expected number of faults in the remaining time period. The maintenance cost data, risk loss and number of failures are combined into the objective function according to preset weights.

6. The method according to claim 1, wherein According to the minimization result of the objective function, the target centralized maintenance period of each subdomain is determined, including: For each subdomain, multiple pending centralized maintenance periods are preset for the subdomain; Calculate the objective function value corresponding to each pending centralized maintenance cycle, and select the pending centralized maintenance cycle that minimizes the objective function value as the target centralized maintenance cycle of the subdomain.

7. A device for determining the centralized maintenance period of track circuit compensation capacitors, characterized in that: include: A device information acquisition module is used to obtain environmental information, fault information, and location information of each compensation capacitor, wherein the environmental information includes road conditions and weather conditions, and the fault information includes the device fault status, fault occurrence time, and corresponding weather conditions; A feature vector generating module, configured to associate the environmental information with the fault information to generate a feature vector for each compensation capacitor; The feature vector includes position coordinates, road condition vector, fault status mark, fault time vector and weather condition vector; a subdomain division module, configured to divide the compensation capacitor into a plurality of subdomains based on the influence of the classification variable in the feature vector on the failure rate of the compensation capacitor; the compensation capacitors in each subdomain have the same failure rate distribution; A failure rate function fitting module is used to fit the failure rate function of the compensation capacitor in each sub-domain based on the historical failure data of each sub-domain; the failure rate function is used to predict the failure rate of the compensation capacitor over time; An objective function determination module is configured to collect maintenance cost data for compensation capacitors within each subdomain; the maintenance cost data includes equipment acquisition costs, transportation costs, personnel costs, and failure risk loss costs; construct an objective function using the maintenance cost data, the number of failures, and the risk loss as optimization targets; and calculate the comprehensive cost by statistically analyzing the number of centralized maintenance visits, the number of replacements for failures, and the failure rate for the remaining time period for each subdomain within a time window. The target centralized maintenance period determination module is used to determine the target centralized maintenance period of each subdomain according to the minimization result of the objective function.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. 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 according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images