Rail transit equipment fault prediction method and system

Abstract

The invention provides a rail transit equipment fault prediction method and system, and the method comprises the steps: obtaining the operation data and environment data of rail transit equipment in a current time period, and determining the time sequence correlation coefficient and statistical feature dispersion of the operation data, and an environment impact factor of the environment data affecting the fault of the rail transit equipment; determining a weight parameter of a time sequence prediction model according to the time sequence correlation coefficient, the statistical feature dispersion and the environmental impact factor, and determining a weight parameter of a classification model according to the weight parameter of the time sequence prediction model; inputting the operation data of the current time period into the time sequence prediction model, and obtaining the operation data of the rail transit equipment in the future time period according to the weight parameters of the time sequence prediction model; and inputting the operation data of the future time period into the classification model, obtaining the fault probability of the rail transit equipment according to the weight parameters of the classification model, and determining whether the rail transit equipment fails or not according to the fault probability. According to the invention, the accuracy and reliability of fault prediction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent operation and maintenance of rail transit, and particularly to a method and system for predicting faults of rail transit equipment. Background Art

[0002] At present, with the rapid development of rail transit, especially high-speed railways, the complexity of railway equipment and the difficulty of operation and maintenance are increasing continuously. The traditional railway operation and maintenance methods mainly rely on manual regular inspections, emergency handling after faults occur, and post-fault repairs. This method has many disadvantages:

[0003] 1. Low efficiency: The frequency of manual inspections is limited by human resources, and it is difficult to achieve real-time monitoring and dynamic perception of the status of rail transit equipment, which is extremely likely to lead to over-maintenance or under-maintenance.

[0004] 2. High cost: Frequent manual inspections and handling of sudden faults result in high operation and maintenance costs of rail transit.

[0005] 3. Safety hazards: Handling after a fault occurs is very likely to cause serious safety accidents, posing a huge threat to the safety of railway transportation.

[0006] In recent years, with the development and support of new-generation information technologies such as big data, Internet of Things (IoT), digital twin, artificial intelligence (AI), and railway-specific 5G communication protocol (5G-Railway, 5G-R), Predictive Maintenance (PM) has gradually become a research hotspot in the field of intelligent operation and maintenance of rail transit. By real-time monitoring the status of railway equipment and early screening, diagnosing, and predicting its potential fault hidden risks, the operation and maintenance efficiency of rail transit can be effectively improved, the cost can be reduced, and safety hazards can be reduced. However, existing models generally have poor generalization ability when facing the complex and changeable operation environment of railway equipment. For example, under the influence of factors such as different climate conditions, train operation density, and equipment aging degree, the fault prediction accuracy of the model will decrease significantly, making it difficult to meet the actual operation and maintenance requirements. Summary of the Invention

[0007] The present invention provides a method and system for predicting faults of rail transit equipment to solve the defect of low fault prediction accuracy of rail transit equipment in the prior art and achieve the improvement of the fault prediction accuracy of rail transit equipment.

[0008] The present invention provides a method for predicting faults of rail transit equipment, including:

[0009] Acquire operation data and environmental data of rail transit equipment in a current time period, determine a time series correlation coefficient and a statistical characteristic dispersion of the operation data, and an environmental impact factor of the environmental data on the rail transit equipment failure;

[0010] Determining weight parameters of a time series prediction model based on the time series correlation coefficient, the statistical feature dispersion, and the environmental influencing factor, and determining weight parameters of a classification model based on the weight parameters of the time series prediction model;

[0011] Inputting the operation data of the current time period into the time series prediction model, and obtaining the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model;

[0012] The operation data of the future time period is input into the classification model, the failure probability of the rail transit equipment is obtained according to the weight parameter of the classification model, and whether the rail transit equipment is faulty is determined according to the failure probability.

[0013] According to a rail transit equipment fault prediction method provided by the present invention, the weight parameters of the time series prediction model are determined according to the time series correlation coefficient, the statistical characteristic dispersion and the environmental influencing factor by the following formula, and the weight parameters of the classification model are determined according to the weight parameters of the time series prediction model:

[0014]

[0015] Among them, W LSTM and W RF are the current weight parameters of the time series prediction model and the classification model, α (t) and β (t) are the time series correlation coefficient and statistical characteristic dispersion corresponding to the current time period including the current time t, is the environmental impact factor.

[0016] According to a rail transit equipment fault prediction method provided by the present invention, the calculation formula of the time series correlation coefficient α is:

[0017]

[0018] Among them, x i is the operating data at the i-th moment in the current time period, is the mean value of the running data in the current time period, k is the lag order, N is the duration of the current time period, w i is the time attenuation weight at the i-th moment, and the calculation formula is:

[0019]

[0020] in, is the time attenuation coefficient, which is obtained by fitting historical data.

[0021] According to a rail transit equipment fault prediction method provided by the present invention, the calculation formula of the statistical characteristic dispersion β is:

[0022]

[0023] Among them, p i is the distribution probability density of the running data at the i-th moment in the current time period, Var(x) is the variance of the running data in the current time period, Var max is the historical maximum variance, δ is the volatility weight coefficient, and n is the length of the current time period.

[0024] According to a rail transit equipment fault prediction method provided by the present invention, the environmental impact factor The calculation formula is:

[0025]

[0026] in, is the aging coefficient of the rail transit equipment, a, b and c are fitting coefficients, H is the ambient humidity, H max is the preset maximum ambient humidity, G is the ambient wind speed, G max is the preset maximum ambient wind speed, T is the ambient temperature, T max The preset maximum ambient temperature.

[0027] According to a rail transit equipment fault prediction method provided by the present invention, after determining whether the rail transit equipment has failed according to the fault probability, the method further includes:

[0028] determining a current prediction error rate based on whether the rail transit equipment is faulty;

[0029] When the current prediction error rate is greater than a preset threshold, the allocation weight of the time series prediction model in the next time period is determined by the following formula:

[0030]

[0031] in, is the allocation weight of the time series prediction model in the next time period including time t+1, is the current weight parameter of the time series prediction model, E t is the current prediction error rate, is the learning rate;

[0032] Input the operation data of the rail transit equipment in the next time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model in the next time period;

[0033] Multiply the operation data in the future time period obtained according to the weight parameters of the time series prediction model in the next time period by the allocation weight of the time series prediction model in the next time period, and then input it into the classification model. Obtain the failure probability of the rail transit equipment according to the weight parameters of the classification model in the next time period, and determine whether the rail transit equipment fails at the next moment according to the failure probability.

[0034] According to a rail transit equipment failure prediction method provided by the present invention, determining whether the rail transit equipment fails according to the failure probability includes:

[0035] Determine the failure probability threshold value P according to the environmental data e , P e The calculation formula of is:

[0036]

[0037] Wherein, is the aging coefficient of the rail transit equipment, H is the environmental humidity, G is the environmental wind speed, T is the environmental temperature, is the equipment operation time factor, , , and are the influence coefficients of environmental humidity, environmental wind speed, environmental temperature and equipment operation time factor respectively, The calculation formula of is:

[0038]

[0039] Wherein, t is the already-operated time of the rail transit equipment, t max is the designed life of the rail transit equipment;

[0040] In the case that the failure probability of the rail transit equipment is greater than the failure probability threshold value, determine that the rail transit equipment fails and give an early warning to the rail transit equipment.

[0041] According to a rail transit equipment failure prediction method provided by the present invention, determine the early warning level of the rail transit equipment through the following formula:

[0042]

[0043] Wherein, P fis the failure probability of the rail transit equipment, L is the remaining life of the rail transit equipment, L max is the designed life of the rail transit equipment, H is the environmental humidity, H max is the preset maximum environmental humidity, is the weight coefficient, and round is the rounding function.

[0044] According to a rail transit equipment failure prediction method provided by the present invention, the remaining life R of the rail transit equipment is predicted through the Weibull distribution and the equipment degradation model, and the formula is as follows:

[0045]

[0046] Wherein, and are respectively the characteristic life and the shape parameter in the Weibull distribution parameters, and t is the current time.

[0047] The present invention also provides a rail transit equipment failure prediction system, including:

[0048] A calculation module, configured to obtain the operation data and environmental data of the rail transit equipment in the current time period, determine the time series correlation coefficient and the statistical feature dispersion degree of the operation data, and the environmental impact factor of the environmental data on the failure of the rail transit equipment;

[0049] A determination module, configured to determine the weight parameters of the time series prediction model according to the time series correlation coefficient, the statistical feature dispersion degree and the environmental impact factor, and determine the weight parameters of the classification model according to the weight parameters of the time series prediction model;

[0050] A first prediction module, configured to input the operation data in the current time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model;

[0051] A second prediction module, configured to input the operation data in the future time period into the classification model, obtain the failure probability of the rail transit equipment according to the weight parameters of the classification model, and determine whether the rail transit equipment fails according to the failure probability.

[0052] The rail transit equipment failure prediction method and system provided by the present invention predict the future trend of the equipment state by using the time series prediction model, judge whether the equipment is likely to fail by using the classification model, and improve the accuracy and reliability of the failure prediction through multi-model fusion; at the same time, the weight parameters of the time series prediction model and the classification model are dynamically determined according to the real-time operation data and environmental data of the rail transit equipment, which is applicable to the failure prediction of different types of data, and further improves the accuracy and universality of the failure prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0054] Figure 1 is a schematic flowchart of the method for predicting rail transit equipment failures provided by the present invention;

[0055] Figure 2 is a schematic diagram of the architecture of a digital railway predictive maintenance system based on multi-source heterogeneous data-driven in the method for predicting rail transit equipment failures provided by the present invention;

[0056] Figure 3 is a schematic flowchart of the data collection and preprocessing process in the method for predicting rail transit equipment failures provided by the present invention;

[0057] Figure 4 is a schematic flowchart of the fault prediction model training and inference process in the method for predicting rail transit equipment failures provided by the present invention;

[0058] Figure 5 is a schematic flowchart of the intelligent early warning and decision support process in the method for predicting rail transit equipment failures provided by the present invention;

[0059] Figure 6 is a schematic diagram of the structure of the rail transit equipment failure prediction system provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0061] The following will describe a method for predicting rail transit equipment failures of the present invention in conjunction with Figure 1 including:

[0062] Step 101: Obtain the operation data and environmental data of the rail transit equipment in the current time period, determine the time series correlation coefficient and statistical feature dispersion degree of the operation data, and the environmental impact factor of the environmental data on the failure of the rail transit equipment;

[0063] Step 102: Determine the weight parameters of the time series prediction model based on the time series correlation coefficient, statistical feature dispersion, and environmental impact factor, and determine the weight parameters of the classification model according to the weight parameters of the time series prediction model;

[0064] Step 103: Input the operation data of the current time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model;

[0065] Step 104: Input the operation data of the future time period into the classification model, obtain the failure probability of the rail transit equipment according to the weight parameters of the classification model, and determine whether the rail transit equipment fails according to the failure probability.

[0066] This embodiment is mainly applicable to the health management of key equipment such as railway tracks (covering components such as switch points and stock rails), catenaries (including parts such as insulators and messenger wires), rolling stocks (such as key parts like traction motors and axle box bearings), and signal systems (such as switch machines, signal lamps, and track circuits). Its core purpose is to achieve early screening of potential faults, prediction of remaining life, intelligent monitoring and early warning, and optimization of operation and maintenance decisions. By deeply integrating big data and artificial intelligence technologies, a closed-loop management mechanism of "data acquisition and monitoring - preprocessing and analysis - fault prediction - safety early warning - operation and maintenance decision-making" is constructed, so as to comprehensively improve the safe operation efficiency of rail transit and reduce potential safety hazards.

[0067] Install sensors (such as acceleration sensors, temperature sensors, current sensors, fiber optic sensors, etc.) at key parts of railway equipment (such as switch machines, catenary insulators, locomotive traction motors, and signal lamps) to collect equipment operation data in real time.

[0068] As Figure 2 shown, the digital railway predictive maintenance system architecture based on multi-source heterogeneous data-driven includes a data acquisition layer, an edge computing layer, a cloud analysis layer, and a decision output layer, showing that the data acquisition module transmits multi-source heterogeneous data back to the edge computing layer and the cloud analysis layer in real time through the railway 5G-R dedicated protocol, which is cleaned and fused by the data preprocessing module, the fault prediction module makes predictions based on the LSTM and RF models, the intelligent early warning module generates early warning information and pushes it to the operation and maintenance personnel through a visualization interface or a mobile terminal, and the decision support module provides operation and maintenance decision-making suggestions based on rules, clearly presenting the data interaction and functional collaboration relationships between modules.

[0069] As Figure 3As shown, the operating data (such as vibration, temperature, current, voltage, strain, displacement, etc.) and environmental data (such as temperature, humidity, wind speed, etc.) of rail transit equipment can be collected in real time through a photoelectro-mechanical-electrical composite sensor array (Photoelectro-Mechanical Sensor Array, PMSA) network, distributed optical fiber sensing technology (Distributed Optical Fiber Sensing, DOFS), and Internet of Things devices, ensuring that the collected data can comprehensively reflect the operating state of the equipment. Among them, the PMSA network includes MEMS acceleration sensors (accuracy requirement: ±0.1g), infrared temperature sensors (resolution requirement: 0.1°C), and Hall current sensors (accuracy requirement: ±0.5%), and its sampling frequency can be dynamically adjusted according to actual needs, with a range of 5Hz - 1kHz.

[0070] The collected data is transmitted via the Internet of Things gateway and synchronized to the cloud server and edge computing nodes to build a cloud-edge collaborative architecture (Cloud-Edge Collaborative Architecture, CECA). Real-time data transmission can be achieved through the dedicated railway 5G-R protocol, and the preprocessing delay of the edge node is ≤30ms (the required verification interval is 10 - 200ms). The cloud-edge collaborative architecture includes edge computing nodes and cloud servers.

[0071] During the data collection, transmission, and storage processes, the AES-256 encryption algorithm (compliant with FIPS 197 standard) is adopted, and digital signatures are performed through the national cryptographic SM2 algorithm to comprehensively ensure data security and privacy protection, effectively preventing the risks of data leakage and tampering.

[0072] As Figure 3 shown, for the collected raw data, it can be cleaned to remove outliers and noise caused by sensor failures or external interferences; multi-source data can be fused and normalized to generate a unified feature vector; wavelet packet decomposition (Wavelet Packet Decomposition, WPD) is used to process non-stationary signals, and key features that can accurately reflect the equipment state are extracted, and the feature extraction efficiency can be significantly improved, thus generating a high-quality data set, providing a reliable data basis for subsequent fault prediction. The preprocessed data is input into a dynamic weight multi-model prediction engine for fault prediction.

[0073] The time series prediction model in this embodiment can be a Long Short-Term Memory (LSTM) network, which is used to predict the future trend of the device state. The classification model can adopt machine learning models, such as deep learning, Random Forest (RF), Support Vector Machine, etc., to determine whether the device is likely to fail. A fault prediction model is constructed based on the time series prediction model and the classification model to achieve real-time prediction of device faults. By means of hybrid model integration, the prediction accuracy is effectively improved, and the generalization ability and adaptability of the model are enhanced.

[0074] Construct a dynamic weight allocation model for the time series prediction model and the classification model, and the weight parameters of the time series prediction model and the classification model are dynamically adjusted according to data characteristics.

[0075] Figure 4 Demonstrate the training process of the model, such as data input, model selection, parameter optimization, etc., and the inference process, such as real-time data input, fault probability output, etc. Use historical data to train the LSTM model, which can effectively capture the long-term dependence relationship of time series data such as the vibration, temperature, and current of the device, and accurately predict the change trend of the future state of the device. At the same time, combine the RF model to classify and predict the fault categories of the device, and early diagnose the device components with potential fault risks.

[0076] The data utilization rate of the existing technology is low, and the fusion and processing capabilities of multi-source heterogeneous data are insufficient, resulting in difficulty in effectively integrating the data collected by different device sensors, forming obvious data islands. For example, data from different sources such as track vibration data, catenary temperature data, and locomotive current data cannot be interconnected and analyzed collaboratively.

[0077] In this embodiment, the future trend of the device state is predicted by using the time series prediction model, and whether the device is likely to fail is judged by using the classification model. The accuracy and reliability of fault prediction are improved through multi-model fusion; at the same time, the weight parameters of the time series prediction model and the classification model are dynamically determined according to the real-time operation data and environmental data of the rail transit equipment, which is applicable to the fault prediction of different types of data, and further improves the accuracy and universality of fault prediction.

[0078] Based on the above embodiment, in this embodiment, the weight parameters of the time series prediction model are determined according to the time series correlation coefficient, statistical feature dispersion, and environmental impact factor through the following formula, and the weight parameters of the classification model are determined according to the weight parameters of the time series prediction model:

[0079]

[0080] Where, W LSTM and W RFare the current weight parameters of the time series prediction model and the classification model respectively, and α (t) and β (t) are the time series correlation coefficient and the statistical feature dispersion corresponding to the current time period including the current moment t respectively, is the environmental impact factor.

[0081] Dynamically allocate the weight parameters of LSTM and RF (random forest) based on α and β. represents the environmental impact factor, mainly quantifying the impact of environmental humidity H (unit: %RH), wind speed G (unit: m / s), and temperature T (unit: o °C) on equipment failure; when the environmental factor approaches 1, the proportion of LSTM weight parameters increases, which is suitable for scenarios with significant time series features (such as vibration signal analysis); when β is large (high data dispersion) and is small, the RF weight parameter increases, which is suitable for non-stationary signals. This dynamic weight parameter model is executed in real time by the FPGA module of the edge computing node through the cooperation of hardware such as "sensor array - edge computing - 5G-R transmission".

[0082] On the basis of the above embodiments, the calculation formula of the time series correlation coefficient α in this embodiment is:

[0083]

[0084] where x i is the operation data at the i-th moment in the current time period, is the mean value of the operation data in the current time period, k is the lag order (dynamically selected through the Akaike information criterion, and the value range is 1-10), N is the duration of the current time period, and w i is the time decay weight at the i-th moment, and the calculation formula is:

[0085]

[0086] where, is the time decay coefficient, which is obtained by fitting historical data.

[0087] α is the data time series correlation coefficient (the value range is 0~1, reflecting the periodicity or trend of the time series, and the closer the value is to 1, the stronger the time series correlation), and the correlation of time series data can be calculated through the autocorrelation function.

[0088] On the basis of the above embodiments, the calculation formula of the statistical feature dispersion β in this embodiment is:

[0089]

[0090] where p i is the distribution probability density of the operation data at the i-th moment in the current time period, Var(x) is the variance of the operation data in the current time period, Var max is the historical maximum variance, δ is the fluctuation weight coefficient determined through experimental data, and n is the duration of the current time period.

[0091] β is the statistical feature dispersion degree, and the dispersion degree of the data distribution can be calculated based on the Shannon entropy.

[0092] Based on the above embodiments, in this embodiment is the environmental impact factor (related to humidity H, wind speed G, and temperature T), and the calculation formula is:

[0093]

[0094] where is the aging coefficient of the rail transit equipment (fitted through historical operation and maintenance data, and the fitting error is required to be ≤2%), a, b, and c are fitting coefficients determined through historical data, H is the environmental humidity, H max is the preset maximum environmental humidity (representing the maximum value allowed by the design), G is the environmental wind speed, G max is the preset maximum environmental wind speed (representing the maximum value allowed by the design), T is the environmental temperature, T max is the preset maximum environmental temperature (representing the maximum value allowed by the design).

[0095] Based on the above embodiments, after determining whether the rail transit equipment is faulty according to the failure probability in this embodiment, it further includes:

[0096] Determine the current prediction error rate according to whether the rail transit equipment is faulty;

[0097] When the current prediction error rate is greater than the preset threshold, determine the allocation weight of the time series prediction model in the next time period through the following formula:

[0098]

[0099] where is the allocation weight of the time series prediction model in the next time period including the (t + 1)-th moment, is the current weight parameter of the time series prediction model, E t is the current prediction error rate, is the learning rate.

[0100] Establish a model update mechanism and the triggering conditions for model incremental learning: When the model prediction error rate threshold exceeds the preset tolerance (e.g., E > 5%), activate the incremental learning mechanism to ensure continuous optimization of the model. Introduce a dynamic feedback adaptive learning mechanism and establish real-time adjustment of the allocation weights based on the prediction error.

[0101] Based on the above embodiments, in this embodiment, determining whether the rail transit equipment is faulty according to the fault probability includes:

[0102] Determine the fault probability critical value P according to the environmental data e , P e The calculation formula of is:

[0103]

[0104] Wherein, is the aging coefficient of the rail transit equipment (fitted through historical operation and maintenance data, and the fitting error is required to be ≤ 2%), H is the environmental humidity, G is the environmental wind speed, T is the environmental temperature, is the equipment operation time factor, , , and are the influence coefficients of environmental humidity, environmental wind speed, environmental temperature and equipment operation time factor respectively, The calculation formula of is:

[0105]

[0106] Wherein, t is the operating time of the rail transit equipment, t max is the design life of the rail transit equipment; when the equipment operation time factor τ > 1, the equipment has exceeded its service life and needs to be focused on.

[0107] In the case where the fault probability of the rail transit equipment is greater than the fault probability critical value, determine that the rail transit equipment is faulty and give a warning to the rail transit equipment.

[0108] Based on the above embodiments, in this embodiment, the warning level of the rail transit equipment is determined by the following formula:

[0109]

[0110] Wherein, P f is the fault probability of the rail transit equipment, L is the remaining life of the rail transit equipment, L max is the design life of the rail transit equipment, H is the environmental humidity, H max is the preset maximum environmental humidity, is the weight coefficient, satisfying , round is the rounding function.

[0111] For example, the dynamic adjustment rule for the warning level is as follows:

[0112] Red warning: The failure probability P f > 80% and the remaining life L of the equipment < 30 days, or the failure probability P f > 90% and the equipment operation time factor τ > 90%. At this time, a red warning is triggered and a shutdown instruction is generated.

[0113] Orange warning: The failure probability P f > 60% and the environmental wind speed G > 15 m / s, or the failure probability P f > 70% and the environmental humidity H > 85%.

[0114] [[ID=2I]]Yellow warning: The failure probability P f > 40% and the equipment operation time exceeds 80% of the designed life, or the failure probability P f > 50% and Var(x) > 10.

[0115] The existing warning mechanism is not perfect, the warning system lacks intelligent warning strategies, and it is unable to dynamically adjust the warning threshold and level according to the equipment status. At the same time, the function of the decision support system is relatively simplistic, and it is difficult to provide comprehensive and accurate operation and maintenance suggestions and resource optimization plans.

[0116] As Figure 5 shown, according to the fault prediction results and the critical thresholds of the equipment (such as the vibration amplitude exceeding the set value, the temperature rising abnormally, or the current fluctuating too much), this embodiment generates warning information of different levels such as red and orange. And based on the failure probability and the critical threshold, the warning level is dynamically adjusted, and the warning information is pushed to the relevant operation and maintenance personnel in a timely manner through a visual interface (such as the dashboard of the track operation and maintenance monitoring center) and an intelligent mobile terminal (such as the mobile APP of the operation and maintenance personnel), so that they can quickly take corresponding measures to avoid the occurrence and expansion of faults and ensure the safe operation of the equipment.

[0117] In addition, a rule-based decision support system is also provided. According to the warning information and the historical operation and maintenance data of the turnout, it helps the operation and maintenance personnel formulate detailed track maintenance plans, such as arranging regular inspections of the turnout, replacing worn parts, optimizing the allocation of maintenance resources, etc., to improve the operation and maintenance efficiency and reduce costs. The system can reasonably arrange the working hours of maintenance personnel and the allocation of maintenance tools according to the fault prediction results and the remaining service life of the turnout, ensure the normal operation of the turnout, and at the same time minimize the maintenance cost and the impact on railway transportation.

[0118] Based on the above embodiments, the remaining life R of the rail transit equipment in this embodiment is predicted through the Weibull distribution and the equipment degradation model, and the formula is as follows:

[0119]

[0120] Wherein, and are respectively the characteristic life (63.2% of the equipment fails before this time) and the shape parameter (determining the failure mode: < 1 is early failure, = 1 is random failure, > 1 is wear-out failure) in the Weibull distribution parameters, obtained by fitting historical data through the Weibull distribution, and t is the current time.

[0121] Based on Monte Carlo simulation, combined with the Weibull Distribution and the equipment degradation model to predict the remaining life (error ≤ 5%).

[0122] Based on the maximum likelihood estimation method (MLE), the Weibull distribution parameters ( , ) are obtained by fitting historical data, and the iterative convergence condition is that the sum of squared residuals ≤ 0.01.

[0123] Derived from the fault prediction model, the steps of Monte Carlo simulation are as follows: generating random samples of equipment life based on the Weibull distribution (requiring the number of samples ≥ 1000); fitting the Weibull distribution parameters (shape parameter φ and characteristic life η) through historical data; calculating the confidence interval of the equipment remaining life (requiring a confidence level of 95%).

[0124] This embodiment predicts the remaining service life of the turnout based on Monte Carlo simulation, provides a scientific basis for operation and maintenance decisions, and reasonably arranges maintenance plans and resource allocation.

[0125] The advantages of the present invention are as follows:

[0126] 1. Real-time performance: Through the Internet of Things and edge computing technologies, real-time monitoring and prediction of equipment status are realized, potential fault risks can be discovered in a timely manner, and the efficiency of rail transit operation and maintenance can be improved.

[0127] 2. High precision: Based on big data and artificial intelligence technologies, through multi-model fusion (such as combining deep learning and traditional machine learning), the accuracy and reliability of fault prediction are improved, which can provide strong support for the intelligent operation and maintenance decision-making of digital rail transit.

[0128] 3. Intelligence: The intelligent early warning and decision support system can generate early warning levels and maintenance suggestions based on the results of fault prediction, combined with the historical data and maintenance experience of the equipment, and push them to the rail transit maintenance personnel through a visual interface or a mobile terminal, so that they can quickly take corresponding measures to avoid the further expansion of faults, optimize the maintenance process, and significantly reduce the maintenance cost.

[0129] 4. Scalability: The system supports multi-source data fusion and model update, is applicable to different types of railway equipment, and has good adaptability and scalability.

[0130] The rail transit equipment fault prediction system provided by the present invention will be described below. The rail transit equipment fault prediction system described below can be mutually referred to the rail transit equipment fault prediction method described above.

[0131] As Figure 6 shown, the system includes a calculation module 601, a determination module 602, a first prediction module 603, and a second prediction module 604, where:

[0132] The calculation module 601 is used to obtain the operation data and environmental data of the rail transit equipment in the current time period, determine the time series correlation coefficient and statistical feature dispersion of the operation data, and the environmental impact factor of the environmental data on the rail transit equipment fault;

[0133] The determination module 602 is used to determine the weight parameters of the time series prediction model according to the time series correlation coefficient, statistical feature dispersion and environmental impact factor, and determine the weight parameters of the classification model according to the weight parameters of the time series prediction model;

[0134] The first prediction module 603 is used to input the operation data of the current time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model;

[0135] The second prediction module 604 is used to input the operation data of the future time period into the classification model, obtain the fault probability of the rail transit equipment according to the weight parameters of the classification model, and determine whether the rail transit equipment is faulty according to the fault probability.

[0136] In this embodiment, the future trend of the equipment status is predicted by using a time series prediction model, and whether the equipment is likely to fail is judged by using a classification model. The accuracy and reliability of fault prediction are improved through multi-model fusion; at the same time, the weight parameters of the time series prediction model and the classification model are dynamically determined according to the real-time operation data and environmental data of the rail transit equipment, which is applicable to the fault prediction of different types of data, and further improves the accuracy and universality of fault prediction.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting faults in rail transit equipment, characterized in that, Including: Obtain the operation data and environmental data of the rail transit equipment in the current time period, determine the time series correlation coefficient and statistical feature dispersion degree of the operation data, and the environmental impact factor of the environmental data on the fault of the rail transit equipment; Determine the weight parameters of the time series prediction model according to the time series correlation coefficient, statistical feature dispersion degree and environmental impact factor, and determine the weight parameters of the classification model according to the weight parameters of the time series prediction model; Input the operation data of the current time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model; Input the operation data of the future time period into the classification model, obtain the fault probability of the rail transit equipment according to the weight parameters of the classification model, and determine whether the rail transit equipment is faulty according to the fault probability.

2. The method for predicting faults of rail transit equipment according to claim 1, characterized in that, Determine the weight parameters of the time series prediction model according to the time series correlation coefficient, statistical feature dispersion degree and environmental impact factor through the following formula, and determine the weight parameters of the classification model according to the weight parameters of the time series prediction model: ; Among them, W LSTM and W RF are the current weight parameters of the time series prediction model and the classification model respectively, α (t) and β (t) are the time series correlation coefficient and the statistical feature dispersion corresponding to the current time period including the current moment t respectively, is the environmental impact factor.

3. The method for predicting faults of rail transit equipment according to claim 1, wherein The calculation formula of the time series correlation coefficient α is: ; where x i is the operation data at the i-th moment in the current time period, is the mean value of the operation data in the current time period, k is the lag order, N is the duration of the current time period, w i is the time decay weight at the i-th moment, and the calculation formula is: ; Among them, is the time decay coefficient, which is obtained by fitting historical data.

4. The method for predicting faults of rail transit equipment according to claim 1, wherein, The calculation formula of the statistical feature dispersion degree β is: ; where p i is the distribution probability density of the operation data at the i-th moment in the current time period, Var(x) is the variance of the operation data in the current time period, and Var max is the historical maximum variance, δ is the fluctuation weight coefficient, and n is the duration of the current time period.

5. The method for predicting faults of rail transit equipment according to claim 1, wherein The environmental impact factor has the following calculation formula: ; Among them, is the aging coefficient of the rail transit equipment, a, b, and c are fitting coefficients, H is the environmental humidity, and H max is the preset maximum environmental humidity, G is the environmental wind speed, and G max is the preset maximum environmental wind speed, T is the environmental temperature, and T max is the preset maximum environmental temperature.

6. The method for predicting faults of rail transit equipment according to claim 1, wherein After determining whether the rail transit equipment is faulty according to the fault probability, it further includes: Determine the current prediction error rate according to whether the rail transit equipment is faulty; When the current prediction error rate is greater than the preset threshold, determine the allocation weight of the time series prediction model in the next time period through the following formula: ; Among them, is the allocation weight of the time series prediction model in the next time period including the (t + 1)th moment, is the current weight parameter of the time series prediction model, E t is the current prediction error rate, is the learning rate; Input the operation data of the rail transit equipment in the next time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model in the next time period; Multiply the operation data of the future time period obtained according to the weight parameters of the time series prediction model in the next time period by the allocation weight of the time series prediction model in the next time period and input it into the classification model, obtain the fault probability of the rail transit equipment according to the weight parameters of the classification model in the next time period, and determine whether the rail transit equipment is faulty at the next moment according to the fault probability.

7. The method for predicting faults of rail transit equipment according to any one of claims 1-6, characterized in that, Determining whether the rail transit equipment is faulty according to the fault probability includes: Determine the critical value of the failure probability P based on the environmental data e , P e The calculation formula of is as follows: ; Among them, is the aging coefficient of the rail transit equipment, H is the environmental humidity, G is the environmental wind speed, T is the environmental temperature, is the equipment operation time factor, , , and are the influence coefficients of environmental humidity, environmental wind speed, environmental temperature and equipment operation time factor respectively, The calculation formula of is: ; where t is the operating time of the rail transit equipment, and t max is the designed life of the rail transit equipment; In the case where the fault probability of the rail transit equipment is greater than the fault probability critical value, determine that the rail transit equipment is faulty and give an early warning to the rail transit equipment.

8. The method for predicting faults of rail transit equipment according to claim 7, wherein Determine the early warning level of the rail transit equipment through the following formula: ; Among them, P f is the failure probability of the rail transit equipment, L is the remaining life of the rail transit equipment, L max is the design life of the rail transit equipment, H is the environmental humidity, H max is the preset maximum environmental humidity, is the weight coefficient, and round is the rounding function.

9. The method for predicting faults of rail transit equipment according to claim 8, characterized in that, The remaining life R of the rail transit equipment is predicted through the Weibull distribution and the equipment degradation model, and the formula is as follows: ; wherein, and are the characteristic life and shape parameter in the Weibull distribution parameters respectively, and t is the current time.

10. A fault prediction system for rail transit equipment, characterized in that, Including: A calculation module, configured to obtain the operation data and environmental data of the rail transit equipment in the current time period, determine the time series correlation coefficient and statistical feature dispersion degree of the operation data, and the environmental impact factor of the environmental data on the fault of the rail transit equipment; A determination module, configured to determine the weight parameters of the time series prediction model according to the time series correlation coefficient, statistical feature dispersion degree and environmental impact factor, and determine the weight parameters of the classification model according to the weight parameters of the time series prediction model; The first prediction module is configured to input the operation data of the current time period into the time series prediction model, and obtain the operation data of the rail transit equipment in the future time period according to the weight parameters of the time series prediction model; The second prediction module is configured to input the operation data of the future time period into the classification model, obtain the failure probability of the rail transit equipment according to the weight parameters of the classification model, and determine whether the rail transit equipment fails according to the failure probability.

Images