High-speed railway infrastructure maintenance decision-making method, device, equipment and medium

Abstract

The application discloses a kind of to realize high-speed railway infrastructure maintenance decision-making method, device, equipment and medium, method includes obtaining line basic data, railway infrastructure disease prediction model is constructed based on the line basic data;Disease parameters are determined based on the railway infrastructure disease prediction model;The maintenance priority of each type of disease is calculated based on the disease parameters, and maintenance task sequencing is determined based on the maintenance priority;Based on maintenance task sequencing and pre-set parallel scheduling algorithm, maintenance tasks are arranged.The application can solve the technical problems of traditional periodic inspection and artificial experience judgment, lack of quantitative prediction of disease development trend.

Description

Technical Field

[0001] This invention relates to the field of railway maintenance and repair technology, and in particular to a method, device, equipment and medium for making maintenance decisions for high-speed railway infrastructure. Background Technology

[0002] Traditional railway maintenance relies mainly on regular inspections and manual experience, lacking quantitative prediction of disease development trends, leading to the following problems: (1) Delay in disease detection. Manual inspections have periodic blind spots, making it difficult to detect early symptoms of diseases in a timely manner and missing the best time for maintenance.

[0003] (2) Lack of trend prediction ability. It is impossible to quantitatively predict when the disease will reach the limit threshold, making it difficult to reasonably arrange preventive maintenance, resulting in frequent passive maintenance and emergency repairs.

[0004] (3) The influence of multiple factors is difficult to quantify. Track smoothness is affected by the cumulative total weight, bridge corrosion is affected by environmental factors, and roadbed settlement is affected by geological conditions. Traditional methods cannot comprehensively consider these factors.

[0005] (4) The data quality is inconsistent. Historical detection data has problems such as missing data, measurement errors, and inconsistent sampling frequencies, which require robust prediction methods.

[0006] (5) There are significant differences among different types of defects. Track smoothness exhibits cumulative deterioration, bridge corrosion exhibits exponential growth, roadbed settlement exhibits logarithmic decay, and tunnel water leakage exhibits sudden occurrence, requiring targeted prediction models. Summary of the Invention

[0007] This invention provides a method, apparatus, equipment, and medium for making maintenance decisions on high-speed railway infrastructure, which addresses the technical problem that traditional periodic inspections and manual experience-based judgments lack quantitative prediction of the development trend of defects.

[0008] According to one aspect of the present invention, a method for making maintenance decisions for high-speed railway infrastructure is provided, comprising: Acquire basic line data and construct a railway infrastructure defect prediction model based on the basic line data; Determine the disease parameters based on the railway infrastructure disease prediction model; The maintenance priority of various diseases is calculated based on the disease parameters, and the maintenance task order is determined based on the maintenance priority. Maintenance tasks are arranged based on maintenance task sorting and a pre-defined parallel scheduling algorithm.

[0009] Optionally, after obtaining the basic line data, the process further includes: The basic data of the line is preprocessed, and the preprocessing includes at least data cleaning, data interpolation and data normalization.

[0010] Optionally, the step of constructing a railway infrastructure defect prediction model based on the line basic data includes: Based on the basic data of the line, prediction models for various defects are constructed. The prediction models for various defects include at least a track smoothness deterioration prediction model, a track crack propagation prediction model, a bridge corrosion depth prediction model, a roadbed uneven settlement prediction model, and a tunnel water leakage prediction model.

[0011] Optionally, determining the defect parameters based on the railway infrastructure defect prediction model includes: Based on the railway infrastructure defect prediction model, the current defect index value, the time to reach the distance limit, and the defect development rate are determined.

[0012] Optionally, the step of calculating the maintenance priority for various diseases based on the disease parameters includes: The current disease index value, the time to reach the distance limit, and the disease development rate of each type of disease are input into a preset priority calculation formula to determine the maintenance priority of each type of disease.

[0013] Optionally, determining the order of maintenance tasks based on the maintenance priority includes: The maintenance priorities based on various diseases are input into a preset weight calculation formula to determine the ranking weight of each disease, and the maintenance tasks are ranked based on the ranking weight of each disease.

[0014] Optionally, the arrangement of maintenance tasks based on maintenance task sequencing and a preset parallel scheduling algorithm includes: Maintenance tasks are arranged in a cyclical manner, from the beginning to the end of the month; Obtain the resource pool for the day. The resource pool should include at least the personnel requirement table, the material requirement table, the track closure schedule, the entry and exit point table, the available personnel table, and the available material table. The tasks to be executed that day are traversed according to their priority based on the type of maintenance task. A parallel scheduling algorithm based on a greedy algorithm, along with preset constraints, yields the daily task plan.

[0015] According to another aspect of the present invention, a device for making maintenance decisions on high-speed railway infrastructure is provided, comprising: The model building unit is used to acquire basic line data and build a railway infrastructure defect prediction model based on the basic line data. A parameter prediction unit is used to determine disease parameters based on the railway infrastructure disease prediction model. The priority calculation unit is used to calculate the maintenance priority of various diseases based on the disease parameters, and to determine the order of maintenance tasks based on the maintenance priority. The task scheduling unit is used to arrange maintenance tasks based on the order of maintenance tasks and a preset parallel scheduling algorithm.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to execute the high-speed railway infrastructure maintenance decision-making method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the high-speed railway infrastructure maintenance decision-making method as described in any embodiment of the present invention.

[0018] The technical solution of this invention acquires basic line data, constructs a railway infrastructure defect prediction model based on the basic line data, thereby enabling early prediction of defect development trends; determines defect parameters based on the railway infrastructure defect prediction model, calculates maintenance priorities for various defects based on the defect parameters, and determines the order of maintenance tasks based on the maintenance priorities; and arranges maintenance tasks based on the maintenance task order and a pre-set parallel scheduling algorithm, using the scheduling algorithm and corresponding constraints, thereby improving maintenance efficiency and reducing resource waste.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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 effort.

[0021] Figure 1 This is a flowchart of a method for making maintenance decisions for high-speed railway infrastructure according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of a method for making maintenance decisions on high-speed railway infrastructure according to Embodiment 2 of the present invention; Figure 3 This is a structural diagram of a high-speed railway infrastructure maintenance decision-making device provided in Embodiment 2 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the high-speed railway infrastructure maintenance decision-making method according to an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0024] Example 1 Figure 1 This invention provides a flowchart for a method of making maintenance decisions for high-speed railway infrastructure, as described in Embodiment 1. Figure 1 As shown, the method includes: S101. Obtain basic line data and construct a railway infrastructure defect prediction model based on the basic line data.

[0025] The basic data for the railway line can include track smoothness data, track crack data, bridge corrosion data, subgrade settlement data, and tunnel water leakage data. Specifically, track smoothness data can include geometric parameters such as gauge deviation, horizontal deviation, vertical deviation, directional deviation, and triangular pits, with TQI (Total Quality Index) statistics calculated for 200m sections. Track crack data can include crack length, crack width, and crack depth, calculated per rail section. Bridge corrosion data can include steel reinforcement corrosion depth, corrosion area, and concrete carbonation depth, calculated per pier. Subgrade settlement data can include cumulative settlement, settlement rate, and differential settlement, calculated per track section. Tunnel water leakage data can include leakage area and leakage flow rate, calculated per tunnel segment.

[0026] Based on the basic data of the above-mentioned types of diseases, prediction models for each type of disease can be constructed to predict the development trend of each type of disease.

[0027] S102. Determine the disease parameters based on the railway infrastructure disease prediction model.

[0028] The parameters of various types of defects can be predicted based on the constructed railway infrastructure defect prediction model. These defect parameters characterize the current stage of the defect's development trend. For example, defect parameters may include defect index values, the time it takes for the defect to reach a preset threshold, and the defect development rate.

[0029] S103. Calculate the maintenance priority of various diseases based on the disease parameters, and determine the order of maintenance tasks based on the maintenance priority.

[0030] Disease parameters for various diseases can reflect the severity of each disease. Based on these parameters, the maintenance priority of each disease can be determined. The maintenance priority is then used to calculate the ranking weight of the maintenance tasks corresponding to each disease, thereby ranking the maintenance tasks for each disease. The higher the ranking, the more urgent the corresponding maintenance task.

[0031] S104. Based on the maintenance task sorting and the pre-set parallel scheduling algorithm, the maintenance tasks are arranged.

[0032] In this embodiment, personnel and materials can be scheduled according to a preset parallel scheduling algorithm, and daily maintenance tasks can be assigned to multiple personnel groups. These groups can execute maintenance tasks in parallel, thus determining the scheduling result of the maintenance tasks. Alternatively, a greedy algorithm can be used to schedule maintenance tasks in parallel based on the daily quantity of personnel and materials, corresponding constraints, and the order of maintenance tasks.

[0033] The technical solution of this invention acquires basic line data, constructs a railway infrastructure defect prediction model based on the basic line data, thereby enabling early prediction of defect development trends; determines defect parameters based on the railway infrastructure defect prediction model, calculates maintenance priorities for various defects based on the defect parameters, and determines the order of maintenance tasks based on the maintenance priorities; and arranges maintenance tasks based on the maintenance task order and a pre-set parallel scheduling algorithm, using the scheduling algorithm and corresponding constraints, thereby improving maintenance efficiency and reducing resource waste.

[0034] Example 2 Figure 2 This is a flowchart illustrating a method for making maintenance decisions on high-speed railway infrastructure, as provided in Embodiment 2 of the present invention. Figure 2 As shown, the method includes: S201. Obtain basic line data.

[0035] The basic data for the line includes track smoothness data, track crack data, bridge corrosion data, roadbed settlement data, and tunnel water leakage data.

[0036] Specifically, track smoothness data can include geometric parameters such as gauge deviation, horizontal deviation, vertical deviation, directional deviation, and triangular pits, as well as TQI indicators statistically analyzed in 200m sections.

[0037] Track crack data can include crack length, crack width, and crack depth, and is statistically analyzed by rail segment.

[0038] Bridge corrosion data can include the depth of steel bar corrosion, corrosion area, and concrete carbonation depth, and is statistically analyzed by pier.

[0039] Roadbed settlement data can include cumulative settlement, settlement rate, and differential settlement, and is statistically analyzed by track segment.

[0040] Tunnel water leakage data can include leakage area and leakage flow rate, and is statistically analyzed by tunnel segment.

[0041] S202. Preprocess the basic data of the line, the preprocessing including at least data cleaning, data interpolation and data normalization.

[0042] Data cleaning of the basic line data can be performed using the 3σ principle or box plot method to identify and remove outlier measurements. Data interpolation of the basic line data can include filling in missing data using linear interpolation, spline interpolation, or mean interpolation based on similar dates.

[0043] The formula for normalizing the basic data of the line is as follows: Data normalization formula: X'ij = (X ij - X i,min ) / (X i,max - X i,min ) Among them, X ij Let X be the original value of the i-th type of disease index measured in the j-th measurement. i,min and X i,max These are the historical minimum and maximum values ​​of the indicator, X'. ij This is the normalized value.

[0044] S203. Based on the basic data of the line, construct prediction models for various types of defects. The prediction models for various types of defects include at least a track smoothness deterioration prediction model, a track crack propagation prediction model, a bridge corrosion depth prediction model, a roadbed uneven settlement prediction model, and a tunnel water leakage prediction model.

[0045] In this embodiment, since track smoothness degradation data is typically influenced by multiple factors (such as cumulative total weight, time, environment, track structure, etc.), and the amount of long-term monitoring data may be limited, Gaussian process regression, as a non-parametric Bayesian method, does not require pre-defined specific function forms and can flexibly characterize complex nonlinear relationships through kernel functions, making it particularly suitable for modeling and prediction under small sample conditions. Furthermore, Gaussian process regression can capture the trend, periodicity (such as the impact of seasonal temperature changes on the track), and local fluctuation characteristics of TQI (Track Quality Index) as it increases with cumulative total weight by selecting appropriate kernel functions (such as radial basis functions, periodic kernels, etc.). In addition, this method easily integrates multi-source heterogeneous data (such as monitoring data, line design parameters, environmental data), improving the comprehensiveness of the prediction. Therefore, the Gaussian process regression prediction method is used to construct the smoothness degradation model.

[0046] Its Gaussian process regression model: f(x) ~ GP(μ(x), k(x,x')) Where μ(x) is the mean function and k(x,x') is the covariance function (kernel function). This embodiment employs a combined kernel function strategy to capture different patterns of TQI evolution. The radial basis function (RBF kernel) is used: k(x,x') = σ²f × exp(-||x-x'||² / (2l²)) + σ²n × δ(x,x') Where σ²f is the signal variance, l is the length scale, σ²n is the noise variance, and δ is the Kronecker function.

[0047] To achieve early warning capabilities 3-6 months in advance, this embodiment employs a recursive multi-step prediction strategy: Short-term forecast (within 1 month): Based on the latest testing data, predict the TQI value for the next 1-2 testing cycles for adjustment of recent maintenance plans; Medium-term forecast (1-6 months): Based on historical trends and seasonal patterns, predict the TQI evolution trajectory for the next 3-6 months for preventive maintenance planning; Long-term forecast (more than 6 months): Used for annual maintenance budget and resource planning, with greater uncertainty, requiring risk assessment in conjunction with confidence intervals.

[0048] Based on the predicted mean and confidence interval, a three-level early warning mechanism is established: ① Threshold of concern (TQI ≥ 4.0). Judgment logic: μ* ≥ 4.0 or P(μ* ≥ ​​4.0 | Data) > 50%; Response measures: Add to the list of key concerns, increase the frequency of detection (from once every 3 months to once a month), and analyze the causes of degradation.

[0049] ② Warning threshold (TQI ≥ 5.0). Judgment logic: μ* ≥ 5.0 or the probability of μ* ≥ 6.0 within 3 months is >30%; Response measures: Initiate the pre-design of preventive maintenance plan, check resource availability, and include the section in the monthly preventive maintenance candidate pool; Time window: It is recommended to complete preventive maintenance (grinding, tamping) within 3 months.

[0050] ③ Threshold limit (TQI ≥ 6.0). Judgment logic: μ* ≥ 6.0 or upper limit of confidence interval ≥ 6.0; Response measures: immediately generate a status repair task and initiate the emergency dispatch process; Time window: determined according to the development rate, generally requiring completion of the rectification within 1-2 months.

[0051] Based on the rate of disease development, further optimize the timing of maintenance: v<0.2 / month: Slow development, preventative maintenance can be carried out according to plan; 0.2 ≤ v < 0.5 / month: Medium-speed development, preventive maintenance needs to be arranged as soon as possible; v ≥ 0.5 / month: Rapid development, needs to be treated as an urgent task.

[0052] To ensure that the prediction model always reflects the latest orbital state evolution, this embodiment designs a model adaptive update mechanism: Incremental learning strategy: For each new batch of detection data, the model is retrained using a sliding time window method (e.g., retaining data from the most recent 24 months). For kernel function hyperparameters, the conjugate gradient method is used for rapid optimization to avoid training from scratch.

[0053] Concept drift detection: Monitor the prediction error sequence; if prediction deviations exceed 2σ consecutively, trigger model recalibration. If a major change in orbital structure is detected (such as after a major overhaul), clear historical data and remodel based on the new state.

[0054] Multi-model fusion: For the same segment, multiple candidate models with different kernel function combinations are maintained simultaneously. The Bayesian model averaging method is used to dynamically adjust the weights of each model based on recent prediction performance.

[0055] Since polynomial regression can approximate complex nonlinear functions (such as crack length curves after integration of the Paris formula) with arbitrary precision by increasing the order, it is particularly suitable for describing the monotonically increasing crack propagation trend with varying curvature. Therefore, polynomial regression can be used to construct a prediction model for track crack propagation.

[0056] Crack length prediction: L(t) = a0 + a1×t + a2×t² + a3×Q(t) Where L(t) is the crack length (mm) at time t, Q(t) is the cumulative total weight passing through, and a0, a1, a2, and a3 are regression coefficients.

[0057] Crack depth prediction:

[0058] Where d(t) is the crack depth (mm), d0 is the initial depth, N is the number of stress cycles, and k is the material constant (for rails, k≈0.03-0.05).

[0059] Crack disease assessment: The overall score is: S = w1×(L / Lmax) + w2×(d / dmax) + w3×(W / Wmax) Where S<0.3: minor crack, generating preventive repair task (welding repair); 0.3≤S<0.7: moderate crack, generating condition repair task (requiring replacement of rail section); S≥0.7: severe crack, generating emergency repair task (requiring immediate replacement).

[0060] The corrosion reaction begins after the concrete cover carbonizes or chloride ions penetrate to the surface of the reinforcing steel. Subsequently, the formation and accumulation of corrosion products often have an autocatalytic effect. For example, the volume expansion of rust (approximately 2-4 times the original iron volume) further leads to concrete cracking and spalling, thereby accelerating the intrusion of oxygen and moisture, forming a positive feedback loop that causes the corrosion rate to accelerate non-linearly over time. This process closely matches the characteristic described by the exponential growth model that "the growth rate is proportional to the current state quantity." Therefore, the exponential growth model can be used to construct a bridge corrosion depth prediction model.

[0061] Corrosion depth prediction: d(t) = d0 + α × (e (β×t) - 1) Where d(t) is the corrosion depth (mm) at time t, d0 is the initial corrosion depth, t is the service time (years), and α and β are environmental correlation coefficients. Environmental coefficient reference values: General atmospheric environment: α≈0.08, β≈0.15; Industrial atmospheric environment: α≈0.12, β≈0.20; Coastal environment: α≈0.15, β≈0.25; Rust damage assessment: Corrosion rate = d(t) / D_reinforcing bar × 100% Among them, corrosion rate <5%: light corrosion; 5% ≤ corrosion rate <15%: moderate corrosion; corrosion rate ≥ 15%: heavy corrosion.

[0062] Since secondary consolidation settlement is mainly caused by the viscous flow (creep) of the soil skeleton under continuous load, its settlement rate gradually decreases over time. This process conforms to a typical logarithmic relationship—the settlement is approximately proportional to the logarithm of time, reflecting the physical nature of the soil deformation rate gradually increasing as the internal structure adjusts and pore water pressure completely dissipates, followed by a gradual increase in deformation resistance. The settlement predicted by the logarithmic model increases with time, but the rate of increase gradually slows down, eventually approaching a stable value or increasing extremely slowly. This is consistent with the actual situation observed in engineering where subgrade settlement eventually tends to stabilize. Therefore, the logarithmic curve model can be used to construct the subgrade uneven settlement prediction model.

[0063] Settlement prediction: S(t) = S0 + Cα × ln(t / t0) Where S(t) is the cumulative settlement (mm) at time t, S0 is the initial settlement, Cα is the secondary consolidation coefficient, t0 is the reference time (usually 1 month), and t is the operating time (months).

[0064] Estimation of secondary consolidation coefficient: Cα = (S2 - S1) / ln(t2 / t1) Where S1 and S2 are the settlement amounts measured in two historical measurements, and t1 and t2 are the corresponding measurement times.

[0065] Differential settlement assessment: ΔS = |S(x1) - S(x2)| / L Where ΔS is the differential settlement rate, L is the distance between the two measuring points (m), and S(x1) and S(x2) are the settlement amounts of the two measuring points.

[0066] Subgrade disease assessment: ΔS < 2‰ and v < 0.5 mm / month: normal, regular monitoring required; 2‰ ≤ ΔS < 4‰ or 0.5 ≤ v < 1.0 mm / month: generate preventive repair task (requires grouting reinforcement); ΔS ≥ 4‰ or v ≥ 1.0 mm / month: generate condition repair task (requires subgrade treatment).

[0067] Tunnel leakage is influenced by multiple factors, including rainfall, groundwater level, and lining cracks, and is characterized by its suddenness. Tunnel leakage monitoring data is often collected at high frequencies (e.g., daily or weekly), but due to monitoring costs, the data length may be limited, and noise levels are high. This necessitates a prediction method that does not require complex parameters or large amounts of data and is robust to noise. The moving average method can calculate the average leakage volume over multiple consecutive time points (e.g., 7 days or 30 days), effectively filtering out monitoring errors, short-term minor disturbances, and other random noise, revealing masked short-term trends. Trend decomposition can model and predict trend and seasonal components separately (e.g., using linear or polynomial regression for the trend component and a periodic model for the seasonal component). By superimposing the prediction results of each component, the final comprehensive prediction value can be obtained. This method is more robust and accurate than directly modeling the original chaotic data. Therefore, the moving average method and trend decomposition method can be used to construct a tunnel leakage prediction model. Trend component extraction (moving average):

[0068] Where T(t) is the trend component, A(t) is the leakage area at time t, and n is the window length (taken as 6 months).

[0069] Prediction formula: A(t+Δt) = T(t) + k×Δt + Sm + ε Where k is the trend slope, Sm is the seasonal component for the corresponding month, and ε is the random disturbance term.

[0070] Leakage assessment: A < 0.5 m²: minor leakage, enhanced monitoring; 0.5 ≤ A < 2.0 m²: generate preventive repair task (requires crack grouting); A ≥ 2.0 m²: generate condition repair task (requires lining repair).

[0071] S204. Based on the railway infrastructure defect prediction model, determine the current defect index value, the time to reach the distance limit, and the defect development rate.

[0072] Among them, the current disease index value is the set threshold, such as the attention threshold, the warning threshold, the restriction threshold, etc.; the time to reach the distance limit is the time corresponding to the intersection of the time x-axis and the threshold y in the curve output by the prediction model; the development rate in the disease prediction model is the value of the increase of the index in a unit of time.

[0073] S205. Input the current disease index value, the time to reach the distance limit, and the disease development rate of each type of disease into a preset priority calculation formula to determine the maintenance priority of each type of disease.

[0074] Maintenance priority calculation: P = w1×(C / Cmax) + w2×[1 - (T / Tmax)] + w3×(v / vmax) Wherein, P is the maintenance priority (0-1, the larger the priority, the more urgent the priority), and the parameters are obtained according to the advice of relevant experts: C is the current disease index value, Cmax is the limit threshold, T is the time to reach the limit, Tmax is the maximum acceptable time window (6 months), v is the disease development rate, vmax is the maximum development rate, and w1, w2, and w3 are weighting coefficients, which can be set to w1=0.4, w2=0.4, and w3=0.2.

[0075] S206. The maintenance priority based on various diseases is input into a preset weight calculation formula to determine the ranking weight of various diseases, and the maintenance task ranking is determined based on the ranking weight of various diseases.

[0076] The preset weight calculation formula used for the task ranking weights of various diseases is as follows: W = λ1×P + λ2×(Tdeadline - Tcurrent) -1 +λ3×H +λ4×B Where W is the sorting weight (the larger the weight, the higher the priority), P is the maintenance priority, Tdeadline is the task deadline, Tcurrent is the current date, obtained from the relevant document knowledge base; H is the man-hours required for the task, B is the disease type weight; λ1, λ2, λ3, and λ4 are all weight coefficients.

[0077] S207. Based on the maintenance task sorting and the pre-set parallel scheduling algorithm, the maintenance tasks are arranged.

[0078] In one embodiment, step S207 specifically includes: S2071. Arrange maintenance tasks in a cyclical manner, from the beginning to the end of the month; S2072. Obtain the resource pool for the day. The resource pool shall include at least the personnel requirement table, the material requirement table, the track closure schedule, the entry and exit point table, the available personnel table, and the available material table. S2073. Traverse the tasks to be executed on the same day according to the priority of various maintenance tasks; S2074. A parallel scheduling algorithm based on a greedy algorithm, along with preset constraints, yields the daily task plan.

[0079] The high-speed railway maintenance task scheduling problem in this embodiment can be described as allocating a set of maintenance tasks to multiple parallel work teams under the constraints of limited track window time resources, human resources, material resources and space resources, determining the start time, end time, work team, and entry / exit points of each task, and satisfying all constraints, while optimizing one or more objective functions.

[0080] The objective function can be to maximize the number of tasks completed or to equalize resource utilization; the constraints can include time constraints, manpower constraints, human resource constraints, material resource constraints, and spatial resource constraints.

[0081] The specific steps of the scheduling algorithm may include: Maintenance tasks are scheduled in a cyclical manner from the beginning to the end of the month; the resource pool for the day is obtained, which includes at least a personnel requirement table, a material requirement table, a track maintenance window schedule, a track access point table, an available personnel table, and an available material table to determine the personnel and material resources for the day; the tasks to be executed for the day are traversed according to priority based on the priority of various maintenance tasks, such as traversing the planned maintenance / status maintenance / predicted maintenance task tables; based on the available personnel and material resources for the day, the optimal path for the personnel team to reach the task location is found to minimize the travel time of the personnel team, and the travel time and working time are calculated.

[0082] The methods for matching time windows with level crossings include: Single-point task walking time: Ttravel = 2 × |Rjunction - Rtask| / V Where Ttravel is the round-trip travel time (hours), Rjunction is the level crossing kilometer marker (kilometers), indicating the location of each level crossing on the track. Workers can only enter the track to work from the level crossing. Rtask is the task kilometer marker (kilometers), indicating the location of the task to be completed on the track. V is the travel speed (taken as 3 km / h).

[0083] Check whether the personnel working hours, material inventory, and maintenance window time meet the constraints. The maintenance window time constraint is as follows: Twork + Ttravel + Tsafety ≤ Twindow Where Twork is the maintenance work time, Ttravel is the walking time, Tsafety is the safety margin time (0.5 hours), and Twindow is the window time.

[0084] The employee working hours constraint is: Hused,i / Ntotal,i ≤ Hmonth Where Hused,i is the total working hours used for the i-th job, Ntotal,i is the total number of people in the job, and Hmonth is the maximum number of working hours per person per month (taken as 50 hours).

[0085] Personnel and material resource constraints can be determined based on the daily personnel and material demand lists; maintenance window time constraints can be determined based on the daily maintenance window schedule.

[0086] If the constraints are met, personnel and material resources are locked, and a greedy algorithm is used to assign maintenance tasks to each personnel group for parallel execution. Multiple personnel groups can work simultaneously within a time window. After all the time windows and personnel and material resources for the day have been allocated, the maintenance tasks for the next day are allocated. During the algorithm execution, multi-objective optimization (maximizing the number of tasks completed and balancing resource utilization to construct the objective function) can be used to obtain the final maintenance task plan.

[0087] In this embodiment of the invention, track smoothness deterioration prediction models, track crack propagation prediction models, bridge corrosion depth prediction models, roadbed uneven settlement prediction models, and tunnel water leakage prediction models are constructed using basic track data. These models predict the development trends of various defects and determine their urgency based on the defect parameters identified by each model. Subsequently, the current defect index values, distance limit arrival times, and defect development rates are input into a preset priority calculation formula to determine the maintenance priorities of each defect. This prioritizes maintenance tasks based on the urgency of each defect. Finally, based on the priority of maintenance tasks, a parallel scheduling algorithm, multiple constraints (such as track window time constraints, manpower time constraints, human resource constraints, material resource constraints, and spatial resource constraints), and a selected multi-objective optimization method are used to achieve intelligent scheduling of maintenance tasks, improving maintenance efficiency and reducing resource waste.

[0088] Example 3 Figure 3 This is a schematic diagram of a high-speed railway infrastructure maintenance decision-making device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: The model building unit 301 is used to acquire basic line data and build a railway infrastructure defect prediction model based on the basic line data. Parameter prediction unit 302 is used to determine disease parameters based on the railway infrastructure disease prediction model; Priority calculation unit 303 is used to calculate the maintenance priority of various diseases based on the disease parameters, and to determine the order of maintenance tasks based on the maintenance priority. The task scheduling unit 304 is used to arrange maintenance tasks based on the maintenance task sorting and a preset parallel scheduling algorithm.

[0089] The high-speed railway infrastructure maintenance decision-making device provided in the embodiments of the present invention can execute the high-speed railway infrastructure maintenance decision-making method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0090] Example 4 Figure 4A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0091] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0092] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0093] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a method for implementing maintenance decisions for high-speed railway infrastructure.

[0094] In some embodiments, a method for implementing high-speed railway infrastructure maintenance decisions can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for implementing high-speed railway infrastructure maintenance decisions described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform a method for implementing high-speed railway infrastructure maintenance decisions by any other suitable means (e.g., by means of firmware).

[0095] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0096] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0097] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0098] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0099] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0100] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0101] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0102] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for making maintenance decisions on high-speed railway infrastructure, characterized in that, include: Acquire basic line data and construct a railway infrastructure defect prediction model based on the basic line data; Determine the disease parameters based on the railway infrastructure disease prediction model; The maintenance priority of various diseases is calculated based on the disease parameters, and the maintenance task order is determined based on the maintenance priority. Maintenance tasks are arranged based on maintenance task sorting and a pre-defined parallel scheduling algorithm.

2. The method for making maintenance decisions for high-speed railway infrastructure according to claim 1, characterized in that, After obtaining the basic line data, the process also includes: The basic data of the line is preprocessed, and the preprocessing includes at least data cleaning, data interpolation and data normalization.

3. The method for making maintenance decisions for high-speed railway infrastructure according to claim 1, characterized in that, The construction of a railway infrastructure defect prediction model based on the aforementioned basic line data includes: Based on the basic data of the line, prediction models for various defects are constructed. The prediction models for various defects include at least a track smoothness deterioration prediction model, a track crack propagation prediction model, a bridge corrosion depth prediction model, a roadbed uneven settlement prediction model, and a tunnel water leakage prediction model.

4. The method for making maintenance decisions for high-speed railway infrastructure according to claim 3, characterized in that, The determination of defect parameters based on the railway infrastructure defect prediction model includes: Based on the railway infrastructure defect prediction model, the current defect index value, the time to reach the distance limit, and the defect development rate are determined.

5. The method for making maintenance decisions for high-speed railway infrastructure according to claim 4, characterized in that, The calculation of maintenance priorities for various diseases based on the disease parameters includes: The current disease index value, the time to reach the distance limit, and the disease development rate of each type of disease are input into a preset priority calculation formula to determine the maintenance priority of each type of disease.

6. The method for making maintenance decisions for high-speed railway infrastructure according to claim 5, characterized in that, The process of determining the order of maintenance tasks based on the maintenance priority includes: The maintenance priorities based on various diseases are input into a preset weight calculation formula to determine the ranking weight of each disease, and the maintenance tasks are ranked based on the ranking weight of each disease.

7. The method for making maintenance decisions for high-speed railway infrastructure according to claim 6, characterized in that, The maintenance tasks are arranged based on the maintenance task sorting and a pre-set parallel scheduling algorithm, including: Maintenance tasks are arranged in a cyclical manner, from the beginning to the end of the month; Obtain the resource pool for the day. The resource pool should include at least the personnel requirement table, the material requirement table, the track closure schedule, the entry and exit point table, the available personnel table, and the available material table. The tasks to be executed that day are traversed according to their priority based on the type of maintenance task. A parallel scheduling algorithm based on a greedy algorithm, along with preset constraints, yields the daily task plan.

8. A device for making decisions on the maintenance of high-speed railway infrastructure, characterized in that, include: The model building unit is used to acquire basic line data and build a railway infrastructure defect prediction model based on the basic line data. A parameter prediction unit is used to determine disease parameters based on the railway infrastructure disease prediction model. The priority calculation unit is used to calculate the maintenance priority of various diseases based on the disease parameters, and to determine the order of maintenance tasks based on the maintenance priority. The task scheduling unit is used to arrange maintenance tasks based on the order of maintenance tasks and a preset parallel scheduling algorithm.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for implementing high-speed railway infrastructure maintenance decision-making as claimed in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the method for making maintenance decisions on high-speed railway infrastructure as described in any one of claims 1-7.

Images