Full-life-cycle management method for electromechanical equipment on mountainous road

Abstract

The invention discloses a full-life-cycle management method for electromechanical equipment of a mountainous road, and relates to the technical field of intelligent traffic infrastructure management, and the method comprises the steps: inputting a standardized equipment operation data set into a Bayesian network evaluation architecture, analyzing the posterior probability of each node under the dynamic constraint of environmental parameters through a multi-level fault propagation path, and carrying out the analysis of the posterior probability of each node; generating an equipment health assessment report; and fusing the equipment control execution state record with historical electromechanical equipment operation data, constructing an equipment full-life-cycle degradation curve through space-time correlation analysis, performing trend feature extraction and signal separation processing, and generating an equipment full-life-cycle analysis report. According to the method, the pertinence and timeliness of the maintenance decision are improved, and the refined monitoring and prediction of the full-cycle operation performance of the equipment are realized, so that a reliable basis is provided for scientifically formulating a maintenance plan, reducing the full-life-cycle cost and prolonging the service life of the equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation infrastructure management, and particularly relates to a full life cycle management method for mountain road electromechanical equipment. BACKGROUND

[0002] In recent years, the tunnel electromechanical equipment management technology field has made significant progress with the development of intelligent transportation systems. The current mainstream tunnel electromechanical equipment management system generally adopts a distributed sensor network architecture, which realizes real-time collection and transmission of key operating parameters by deploying various types of environmental monitoring sensors and equipment state acquisition terminals. This kind of system generally adopts a three-layer architecture design, including a field perception layer, a data transmission layer and a central processing layer, which can complete the basic device state monitoring and abnormal alarm function. In terms of data processing, the existing technology mostly adopts simple judgment rules based on threshold or traditional statistical analysis model to preliminarily process and analyze the collected equipment operation data. Especially in the aspect of environmental parameter monitoring, through multi-source sensor data fusion technology, the system can reliably monitor key parameters such as temperature, humidity, smoke concentration and the like, providing a basic guarantee for tunnel safety operation.

[0003] The existing technology still has obvious limitations in the full life cycle management of electromechanical equipment. The existing technology does not solve the problem of quantifying the influence of dynamic changes of environmental parameters on the equipment failure propagation path, resulting in that the accuracy of health assessment is greatly affected by environmental interference. The environment of mountain tunnels is complex and changeable, and the parameters such as temperature, humidity and air pressure fluctuate significantly. However, the existing evaluation model fails to fully consider the correlation between these dynamic environmental factors and equipment failure modes, resulting in deviation between the health status evaluation result and the actual working condition. At the same time, in the aspect of maintenance resource optimization, the existing system adopts a static priority scheduling strategy, which cannot be dynamically adjusted according to the actual degradation state of the equipment, resulting in that it is difficult to achieve optimal allocation under the condition of limited maintenance resources, affecting the overall maintenance efficiency. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a full life cycle management method for mountain road electromechanical equipment to solve the problems of health assessment being greatly affected by environmental interference and maintenance resource allocation being not optimal in the prior art.

[0006] To solve the above technical problems, the present application provides the following technical solutions: The present application provides a full life cycle management method for mountain road electromechanical equipment, which comprises collecting electromechanical equipment operation data in a mountain road tunnel, and performing cleaning and standardization processing on the electromechanical equipment operation data to output a standardized equipment operation data set. The standardized equipment operation data set is input into the Bayesian network evaluation architecture, the posterior probability of each node under the dynamic constraint of environmental parameters is analyzed through a multi-level fault propagation path, and an equipment health evaluation report is generated. According to the equipment health evaluation report, the intelligent scheduling of maintenance tasks and resource optimization of the mechanical and electrical equipment maintenance task queue and the limited resource set are performed, and an executable scheduling scheme is generated. The executable scheduling scheme is used for cooperative control of different types of mechanical and electrical equipment in the mountainous highway tunnel, and the control execution state record is recorded. The equipment control execution state record and the historical mechanical and electrical equipment operation data are fused, the equipment full life cycle degradation curve is constructed through space-time correlation analysis, and the trend feature extraction and signal separation processing are performed, and the equipment full life cycle analysis report is generated.

[0007] As a preferred scheme of the full life cycle management method of the mountainous highway mechanical and electrical equipment, wherein: the mechanical and electrical equipment operation data includes operation state parameters, energy consumption data, environmental monitoring data and fault alarm information.

[0008] As a preferred scheme of the full life cycle management method of the mountainous highway mechanical and electrical equipment, wherein: the mechanical and electrical equipment operation data is cleaned and standardized, and the standardized equipment operation data set is output, and the specific steps are as follows, The time stamp continuity, field integrity and abnormal value detection are performed on the mechanical and electrical equipment operation data, and the missing and abnormal records are removed to obtain the cleaned operation data. The unified unit, format and coding are performed on the cleaned operation data, and the normalization processing is performed to obtain the standardized operation data. The standardized operation data is grouped and stored according to the equipment type and time sequence to form the standardized equipment operation data set.

[0009] As a preferred scheme of the full life cycle management method of the mountainous highway mechanical and electrical equipment, wherein: the standardized equipment operation data set is input into the Bayesian network evaluation architecture, the posterior probability of each node under the dynamic constraint of environmental parameters is analyzed through a multi-level fault propagation path, and an equipment health evaluation report is generated, and the specific steps are as follows, The standardized equipment operation data set is input into the Bayesian network evaluation architecture, and is mapped to the physical layer node, the functional layer node and the operation risk node, and the three-level node mapping result is obtained. According to the three-level node mapping result, the historical fault data and the space-time weight relationship between nodes are combined, the causal dependence of each node is quantified, and the dynamic conditional probability table is constructed. Based on the dynamic conditional probability table, a multi-level fault propagation path analysis is performed, and the posterior probability of each node is calculated through forward propagation and backward correction iteration; According to the posterior probability of each node, the fault risk of the physical layer node and the functional layer node is mapped to the equipment health level, the equipment health state index is formed, and the maintenance priority is generated by combining the operation risk node probability sorting, and the equipment health evaluation report is output.

[0010] As a preferred scheme of the full life cycle management method of the mountain road electromechanical equipment, wherein: the Bayesian network evaluation architecture is determined based on the electromechanical equipment failure mode and effect analysis, the complex causal relationship between nodes is quantified through the dynamic conditional probability table, and the parameters are optimized and constructed by fusing time and space constraints.

[0011] As a preferred scheme of the full life cycle management method of the mountain road electromechanical equipment, wherein: according to the equipment health evaluation report, the maintenance task queue and the limited resource set are intelligently scheduled and resource optimized for the electromechanical equipment maintenance task, and an executable scheduling scheme is generated, the specific steps are as follows, According to the maintenance priority in the equipment health evaluation report, the maintenance task queue of the electromechanical equipment to be maintained is generated according to the risk level and the emergency degree; The attributes of the limited resource set are modeled, and the matching degree of each resource and the maintenance task is calculated to form a resource availability matrix; the limited resource set is constructed by counting the number of available maintenance personnel, skill level and the type and number of available tools and equipment; The maintenance task queue and the resource availability matrix are input into the intelligent scheduling algorithm, and the scheduling optimization is performed according to the task priority, resource matching degree and task dependency relationship to generate a preliminary scheduling scheme; The preliminary scheduling scheme is subjected to conflict detection, and the task order and resource allocation are adjusted through iteration to optimize the output executable scheduling scheme.

[0012] As a preferred scheme of the full life cycle management method of the mountain road electromechanical equipment, wherein: the executable scheduling scheme includes a maintenance task instruction sequence and a dynamic resource allocation parameter set; wherein: the maintenance task instruction sequence includes an execution time window, a responsible person ID and a required spare part code; the dynamic resource allocation parameter set includes personnel path planning, spare part allocation route and equipment downtime window.

[0013] As a preferred scheme of the full life cycle management method of the mountain road electromechanical equipment, wherein: the executable scheduling scheme is used to cooperatively control different types of electromechanical equipment in the mountain road tunnel, and the device control execution state record is obtained by recording, the specific steps are as follows, The executable scheduling scheme is parsed according to the device type and the maintenance sequence, and operation instruction lists are generated according to the start time, end time and allocated resources of each task; The operation instruction lists are sent to the corresponding mechanical and electrical equipment control nodes, and are cooperatively controlled according to the task sequence and execution time, and the device control execution process is recorded in real time to form preliminary device execution state information; The preliminary device execution state information is integrated and time-sequentially sorted to generate device control execution state records.

[0014] As a preferred scheme of the full life cycle management method of the mountain road mechanical and electrical equipment, the device control execution state records and the historical mechanical and electrical equipment operation data are fused, a device full life cycle degradation curve is constructed through space-time correlation analysis, trend feature extraction and signal separation processing are performed, and a device full life cycle analysis report is generated. The device control execution state records and the historical mechanical and electrical equipment operation data are aligned according to the device ID, time stamp and space position, and a unified data view is constructed using a multi-dimensional index structure; On the basis of the unified data view, the state similarity of the mechanical and electrical equipment at adjacent time points and space positions is calculated, and a device state evolution correlation matrix is formed by weight integration; The state changes of adjacent time points and space positions in the device state evolution correlation matrix are weighted and integrated to construct a device full life cycle degradation curve; A first-order differential processing algorithm is used to process the device full life cycle degradation curve, extract the short-term trend change of the mechanical and electrical equipment, and separate the long-term degradation trend and periodic fluctuation signal through a wavelet transform decomposition algorithm; The mechanical and electrical equipment short-term trend change, long-term degradation trend and periodic fluctuation signal are integrated to generate a device full life cycle analysis report.

[0015] As a preferred scheme of the full life cycle management method of the mountain road mechanical and electrical equipment, the multi-dimensional index structure uses the device ID, time stamp and space position in the device control execution state records and the historical mechanical and electrical equipment operation data as index dimensions, maps each device control execution state record to the corresponding multi-dimensional coordinate position, and adopts a tree structure to construct.

[0016] The present application has the beneficial effects that: through the multi-level Bayesian network fault propagation analysis based on complex causal relationship, the precise quantification and risk priority ranking of the health status of the mountainous highway electromechanical equipment under the dynamic environment constraint are realized, and the pertinence and timeliness of the maintenance decision are improved; and through the fusion of the equipment control execution state and the historical operation data to construct the whole life cycle degradation curve, the short-term trend, long-term degradation and periodic fluctuation are separated, the fine monitoring and prediction of the whole cycle operation performance of the equipment are realized, and thus reliable basis is provided for scientifically formulating the maintenance plan, reducing the whole life cycle cost and prolonging the service life of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Fig. 1 The flowchart of the whole life cycle management method of the mountainous highway electromechanical equipment.

[0019] Fig. 2 The flowchart of the composition and standardized processing of the electromechanical equipment operation data.

[0020] Fig. 3 The workflow diagram of the Bayesian network evaluation architecture.

[0021] Fig. 4 The flowchart of the intelligent scheduling of the maintenance task and the resource optimization. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0023] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0024] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or selective embodiment that excludes other embodiments.

[0025] REFERENCE Figs. 1-4For an embodiment of the present application, the embodiment provides a full life cycle management method of mountain road electromechanical equipment, comprising the following steps: S1, collecting electromechanical equipment operation data in a mountain road tunnel, and performing cleaning and standardization processing on the electromechanical equipment operation data to output a standardized equipment operation data set.

[0026] It should be noted that the electromechanical equipment operation data includes operation state parameters, energy consumption data, environmental monitoring data, and fault alarm information; The operation state parameters include real-time working condition information such as the start-stop state, operation mode, speed, voltage, current, and temperature of the equipment; The energy consumption data includes instantaneous power, cumulative power consumption, power factor, and load rate of the equipment during operation, and other energy usage information; The environmental monitoring data includes environmental indicators such as temperature, humidity, wind speed, wind direction, noise, dust concentration, and harmful gas concentration in the tunnel; The fault alarm information includes fault type, alarm time, alarm code, alarm level, and related sensor trigger records of the equipment.

[0027] S1.1, detect the time stamp continuity, field integrity, and abnormal value of the electromechanical equipment operation data, and eliminate missing and abnormal records to obtain cleaned operation data.

[0028] It should be noted that the collected electromechanical equipment operation data is arranged in chronological order, and whether the time stamp of the electromechanical equipment operation data is continuous is determined by checking whether the time stamp difference of adjacent records meets the expected collection interval (such as 1 second / time). For electromechanical equipment operation data records with non-continuous time stamps detected, the record index position is recorded and deleted in subsequent steps to maintain consistency of the time axis of the electromechanical equipment operation data; Each field of the electromechanical equipment operation data is checked for integrity, and it is confirmed piece by piece whether the fields of operation state parameters, energy consumption data, environmental monitoring data, and fault alarm information are all present with valid values. If it is found that the field values are null, illegal, or exceed the allowed range, the found electromechanical equipment operation data record is marked as a missing record and is eliminated; The abnormal value detection is performed on the electromechanical equipment operation data, and by comparison with the statistical distribution of the historical electromechanical equipment operation data, the values obviously deviating from the normal range in the operation state parameters, energy consumption data, environmental monitoring data and fault alarm information are identified, for example, the temperature example value in the operation state parameter is far higher than the upper limit of the normal working temperature, the corresponding electromechanical equipment operation data record is removed, and the remaining electromechanical equipment operation data meets the continuity and consistency requirements in time stamp, field value and value distribution, so as to form the cleaned operation data; wherein the normal range is obtained by analyzing the distribution characteristics (such as mean ± 3σ or IQR method) of the historical electromechanical equipment operation data.

[0029] S1.2, the cleaned operation data is uniformly converted in unit, format and coding, and normalized to obtain standardized operation data.

[0030] It should be noted that the physical quantity units of the operation state parameters, energy consumption data, environmental monitoring data and fault alarm information in the cleaned operation data are uniformly converted, for example, the temperature example value is uniformly converted to Celsius, the electric energy consumption example value is uniformly converted to kilowatt-hour, and the wind speed example value is uniformly converted to meters per second, so that the same type of physical quantity has consistent units. The field structure of the cleaned operation data is uniformly converted to field order, field type and field length, so that the structure of all records remains consistent. The classification variables in the cleaned operation data are converted according to the unified coding rule, for example, the operation mode is uniformly converted to digital coding, and the fault type is uniformly converted to letter and number combined coding, so as to ensure the comparability of the classification variables in different records. The numerical value type field of the cleaned operation data after the unit, format and coding are uniformly converted is scaled according to the normalization method, for example, the minimum and maximum normalization method is used to map the numerical value to the example value between 0 and 1, so as to eliminate the difference between different dimensions, and finally the standardized operation data is obtained.

[0031] S1.3, the standardized operation data is grouped and stored according to the device type and time sequence to form a standardized device operation data set.

[0032] It should be noted that the standardized operation data is classified according to the device type, and all records belonging to the same device type are classified into the same group. In each device type group, the standardized operation data is sorted according to the time sequence, so that the records are arranged in chronological order; Each device type group after sorting is arranged in a structured manner, so that the standardized operation data maintains uniform field order, field type and field length when stored; The completed device types are grouped and stored as a directly callable record set, forming a standardized device operation data set.

[0033] S2, input the standardized device operation data set into the Bayesian network evaluation architecture, analyze the posterior probability of each node under the dynamic constraint of the environmental parameter through the multi-level fault propagation path, and generate a device health evaluation report.

[0034] It should be noted that the Bayesian network evaluation architecture is based on the mechanical and electrical equipment failure mode and effects analysis to determine a three-level node system, quantize the complex causal relationship between nodes through dynamic conditional probability table, and construct by parameter optimization under the fusion of time and space constraints; More specifically: the operating state parameters, energy consumption data, environmental monitoring data and fault alarm information are taken as Bayesian network nodes, and the meaning of each Bayesian network node is defined; secondly, through historical operating state parameters, energy consumption data, environmental monitoring data and fault alarm information, the constraint-based method (such as conditional independence test) or score-based method (such as BIC score) is used to verify the conditional independence between variables through statistical test (such as chi-square test, mutual information), and the direct dependency relationship is identified; search strategy (such as hill climbing method, K2 algorithm) is used to iteratively optimize in the candidate structure, ensure that the parent node is always in front of the child node in the topological sorting, avoid loop formation, and finally generate the directed acyclic graph structure of the Bayesian network node; based on the standardized device operation data set, for each node (such as temperature, vibration, etc. State parameters), according to the combination of its parent nodes (such as environmental temperature, load, etc.), the standardized data is grouped; then the frequency of different states of the child node in each group is counted, the conditional probability distribution is calculated through maximum likelihood estimation or Bayesian smoothing (such as Laplace correction), and the conditional probability distribution and the directed acyclic graph structure are combined to form a complete Bayesian network architecture; finally, the operating state parameters, energy consumption data, environmental monitoring data and fault alarm information are inferred and evaluated using the formed Bayesian network architecture, and the construction of the Bayesian network evaluation architecture is completed.

[0035] It should also be noted that the prior art generally inputs the standardized device operation data set into the constructed Bayesian network, calculates the posterior probability of each Bayesian network node through static conditional probability, analyzes the fault propagation path and obtains the device health evaluation result, but in the processing process, the inference is usually made under the condition of fixed environmental parameters, and the dynamic constraint influence of environmental parameters changing with time or state is not fully considered. The standardized device operation data set is input into the Bayesian network evaluation architecture, and the posterior probability of each Bayesian network node is updated in real time under different environmental conditions through the multi-level fault propagation path analysis, and the device health evaluation report reflecting the influence of the actual operating environment is generated. Our invention incorporates dynamic constraints of environmental parameters into the Bayesian network evaluation process for posterior probability calculation, whereas existing technologies typically use fixed environmental parameters for inference. Our invention can improve the adaptability and accuracy of equipment health assessment results to the real operating environment, enhance the precision of fault propagation path analysis, and improve the ability to predict potential risks under environmental changes.

[0036] S2.1 Input the standardized equipment operation dataset into the Bayesian network evaluation architecture according to equipment type and time series, and map it to physical layer nodes, functional layer nodes and operational risk nodes to obtain the three-level node mapping results.

[0037] It should be noted that the standardized equipment operation dataset is input into the Bayesian network evaluation architecture in the order of equipment type and time series. In the Bayesian network evaluation architecture, the standardized equipment operation dataset and fault alarm information are sequentially mapped to the physical layer node, functional layer node and operational risk node. At the physical layer nodes, observations directly corresponding to sensor and electrical components in the standardized equipment operation dataset are used as evidence for the physical layer nodes. The relationship between physical layer nodes and their parent nodes is determined based on the directed acyclic graph structure in the Bayesian network evaluation architecture. The observation likelihood is calculated for each physical layer node according to the conditional probability distribution, expressed as: ; In the formula, It is a physical layer node In time index The observed likelihood is used for forward propagation calculations; It is a physical layer node In time index The value represents the physical layer node at the time index. The state or observation; It is a physical layer node The set of parent nodes, that is, the nodes that directly affect the physical layer. The set of upper-level nodes; It is a physical layer node In the parent node set The joint conditional probability under the given conditions is provided by the dynamic conditional probability table; It is a time index. It is a physical layer node index; Based on the observation likelihood, the posterior probability distribution of physical layer nodes is updated by the Bayesian forward propagation algorithm to obtain the posterior probability and evidence matching relationship of physical layer nodes under the time index and device type index, and output the physical layer node mapping result. At the functional layer nodes, after obtaining the mapping results of the physical layer nodes, the posterior probabilities of the physical layer nodes are aligned with the standardized equipment operation dataset within the same equipment type and the same time series window. Based on the directed acyclic graph structure in the Bayesian network evaluation architecture, the set of parent nodes of the functional layer nodes is determined as the relevant physical layer nodes and necessary observations. Joint evidence fusion and forward update are performed on the functional layer nodes according to the conditional probability distribution, and backward correction is performed on the time points with abnormal records. The posterior distribution and evidence source labeling of the functional layer nodes under the time index and equipment type index are obtained, and the functional layer node mapping results are output. At the operational risk node, after obtaining the functional layer node mapping results, the posterior probability of the functional layer node is aligned with the standardized equipment operation dataset by equipment type and time series within the same window. Based on the directed acyclic graph structure in the Bayesian network evaluation architecture, the set of parent nodes of the operational risk node is determined as the relevant functional layer nodes and necessary observations. The posterior probability distribution of the operational risk node is derived through the conditional probability distribution, and time series smoothing is performed to form the operational risk level and key evidence path labeling bound to the equipment type index and time index. The operational risk node mapping results are output, and finally the three-level node mapping results are obtained. It should also be noted that the training process of the Bayesian network evaluation architecture is as follows: the conditional probability distribution in the Bayesian network evaluation architecture is estimated using historical time series records from the standardized equipment operation dataset. First, the directed acyclic graph structure is determined based on the prior structure. Then, the conditional probability distribution of each Bayesian network node is learned by using maximum likelihood estimation or Bayesian estimation with prior. Hold-out validation or cross-validation is used to test the fitting effect on the standardized equipment operation dataset. After convergence, the conditional probability distribution is solidified for processing physical layer nodes, functional layer nodes, and operational risk nodes.

[0038] S2.2 Based on the three-level node mapping results, combined with historical fault data and spatiotemporal weight relationships between nodes, quantify the causal dependencies of each node and construct a dynamic conditional probability table.

[0039] It should be noted that the mapping results of physical layer nodes, functional layer nodes, and operational risk nodes are aligned with the historical fault records of corresponding equipment types and time series in the standardized equipment operation dataset. The fault propagation path time series is extracted using the fault occurrence time, fault type, and fault severity information in the historical fault records. Based on conditional probability statistics and time series analysis, the dependencies between nodes are quantified to form causal dependency weights. Based on the causal dependency weights and fault propagation path timings between each pair of parent and child nodes, and combined with the relative distribution and weights of network nodes in spatial location, the causal dependency weights are transformed into a preliminary dynamic conditional probability table. The preliminary dynamic conditional probability table is normalized and smoothed over time to reflect the dynamic adjustment of inter-node dependencies as the environment and operating status change. This generates a dynamic conditional probability table that can be used for forward and backward inference of Bayesian networks, enabling the quantification of causal relationships between nodes and the dynamic construction of conditional probabilities.

[0040] S2.3. Based on the dynamic conditional probability table, perform multi-level fault propagation path analysis, and calculate the posterior probability of each node through forward propagation and backward correction iteration.

[0041] It should be noted that after the standardized equipment operation dataset is generated through three-level node mapping and dynamic conditional probability tables, physical layer nodes, functional layer nodes, and operational risk nodes are taken as input under time series and equipment type indices. Forward propagation is performed on each node according to the dynamic conditional probability table. During forward propagation, the observation likelihood of each physical layer node is obtained based on the posterior probability of the node's parent node and the conditional probability of the observation at the current time point. The observation likelihood is propagated upward along the directed acyclic graph structure to the functional layer nodes, and then to the operational risk nodes. The conditional probability contribution of each node is accumulated during the propagation process. After forward propagation is completed, backward correction is performed on functional layer nodes and physical layer nodes at the time points where abnormal records exist. The posterior probability of each node is readjusted using time series smoothing methods to ensure that the posterior probability of the nodes remains continuous and consistent in the time series. During the iterative calculation process, forward propagation and backward correction are performed alternately until the posterior probabilities of physical layer nodes, functional layer nodes, and operational risk nodes converge, forming a set of node posterior probabilities bound to the device type index and time index.

[0042] S2.4 Based on the posterior probability of each node, the failure risk of physical layer nodes and functional layer nodes is mapped to the equipment health level to form equipment health status indicators. Combined with the probability ranking of operational risk nodes, maintenance priorities are generated, and an equipment health assessment report is output.

[0043] It should be noted that after the standardized equipment operation dataset undergoes three-level node mapping, construction of dynamic conditional probability tables, and forward propagation and backward correction to calculate the posterior probabilities of nodes, the posterior probabilities of physical layer nodes and functional layer nodes are mapped to the fault type and severity information corresponding to each node. According to the preset equipment health level rules, the fault risk of each physical layer node and functional layer node is converted into an equipment health level, forming an equipment health status index. Based on this, the posterior probabilities of operational risk nodes are sorted according to equipment type and time series to generate operational risk ranking results, so as to determine the priority of maintenance tasks; The equipment health status indicators and operational risk ranking results are used to form a structured equipment health assessment report, which includes equipment health status indicators of physical layer nodes and functional layer nodes, as well as maintenance priority information for operational risk nodes. It should also be noted that the equipment health level rules are set by defining the mapping relationship between different fault risk ranges and corresponding health levels. For example, the posterior fault probability of physical layer nodes or functional layer nodes is divided into low, medium and high levels, and the equipment health level of each node is determined by combining fault type and severity information.

[0044] S3. Based on the equipment health assessment report, perform intelligent scheduling and resource optimization of electromechanical equipment maintenance tasks and limited resource sets to generate an executable scheduling scheme.

[0045] S3.1 Based on the maintenance priority in the equipment health assessment report, generate a maintenance task queue according to the risk level and urgency of the electromechanical equipment to be maintained.

[0046] It should be noted that, based on the maintenance priority in the equipment health assessment report, the maintenance tasks for each piece of electromechanical equipment are sorted according to the risk level and urgency of each piece of equipment. The sorted tasks are then arranged one by one according to time sequence and equipment type to form a maintenance task queue. The maintenance task queue records the equipment ID, risk level, urgency, estimated maintenance time window, and related reference information for each maintenance task.

[0047] S3.2 Build attribute models for the finite set of resources and calculate the matching degree between each resource and the maintenance task to form a resource availability matrix.

[0048] It should be noted that attribute modeling is performed on the finite resource set, and attributes such as the number of available maintenance personnel, skill level, and types and quantities of available tools and equipment for each resource in the finite resource set are recorded and organized; in each maintenance task, the matching calculation is performed based on the skill level of the personnel required for the task, the type of tools and equipment, and the maintenance workload with the attributes of each resource in the finite resource set, so as to quantify the matching degree between each resource and each maintenance task; In the process of matching maintenance tasks with limited resources, the set of attributes for the maintenance task, including skill level, tool and equipment type, and maintenance workload, is matched with the first item in the set of limited resources. The attributes of the resources are compared using a matching function. Calculation yields the first Item resource for the first Matching degree of maintenance tasks The expression for the workload matching degree function is: Before calculating the workload matching function, the workload required for the maintenance task and the workload available for the resources are normalized. The expanded form of the workload matching function expression is: ; In the formula, It is the skill matching weight coefficient; It is a maintenance task Required personnel skill level indicators, such as level values ​​from 1 to 5, dimensionless; It is a resource Available personnel skill level indicators, and Dimensionless; It is a personnel skill matching function that outputs a dimensionless value of 0 to 1, representing the degree of matching between the skills required for the task and the skills available in the resource. It is the tool matching weight coefficient; It is a maintenance task Required tool or equipment type code, dimensionless; It is a resource Available tool or equipment type codes, dimensionless; It is a tool and equipment matching function that outputs a dimensionless value of 0 to 1, representing the degree of matching between the tools required for the task and the resource tools. It is the workload matching weight coefficient; It is a maintenance task The required workload, such as working hours, is measured in units of time. It is a resource The amount of work that can be done, such as available working hours, measured in units of time; It is a workload matching function, outputting a dimensionless value between 0 and 1. It is to maintain the task index. It is a resource index; According to maintenance task index As a row, resource index As columns, all matching scores are arranged into a two-dimensional matrix, where each element of the matrix... Indicates maintenance task With resources The degree of matching is used to form a resource availability matrix; It should also be noted that the limited resource set is constructed by statistically analyzing the number of available maintenance personnel, their skill levels, and the types and quantities of available tools and equipment.

[0049] S3.3 Input the task queue and resource availability matrix into the intelligent scheduling algorithm, optimize the scheduling based on task priority, resource matching degree and task dependency relationship, and generate a preliminary scheduling scheme.

[0050] It should be noted that the maintenance task queue and resource availability matrix are used as inputs. The task priority information is read from each maintenance task, and the resource matching degree corresponding to each maintenance task is extracted. At the same time, combined with the dependency constraints between maintenance tasks, the feasible resource allocation and execution time order of each maintenance task are determined according to the preset scheduling strategy. During the scheduling process, the resource allocation scheme that satisfies both task priority and resource matching degree maximization and follows task dependency relationships is selected through iterative search and optimization methods, thereby forming the initial scheduling scheme of maintenance tasks. It should also be noted that the scheduling strategy is set by comprehensively balancing the priority of maintenance tasks, resource matching degree, and task dependencies. It forms feasible scheduling rules by prioritizing high-risk or high-urgency tasks, prioritizing the allocation of resources with high matching degree, and ensuring that dependent tasks come first.

[0051] S3.4 Perform conflict detection on the preliminary scheduling scheme, and optimize it by iteratively adjusting the task order and resource allocation to output an executable scheduling scheme.

[0052] It should be noted that in the initial scheduling scheme, conflict detection is performed on the time arrangement and resource allocation of all maintenance tasks to check whether resources are repeatedly allocated or task dependencies are not met within the same time period. For the detected conflicts, the execution order of maintenance tasks and the resources allocated to each task are adjusted iteratively according to the priority of maintenance tasks, resource matching degree and task dependencies, and the conflict situation is recalculated until all conflicts are eliminated. After multiple rounds of iterative optimization, the final execution order and resource allocation scheme of each maintenance task are formed, and an executable scheduling scheme is output. It should also be noted that the executable scheduling scheme includes a maintenance task instruction sequence and a dynamic resource allocation parameter set; wherein: the maintenance task instruction sequence includes the execution time window, the responsible personnel ID, and the required spare parts code; the dynamic resource allocation parameter set includes personnel path planning, spare parts allocation route, and equipment downtime time window.

[0053] S4. Utilize an executable scheduling scheme to coordinate the control of different types of electromechanical equipment in mountainous highway tunnels, record the data, and obtain equipment control execution status records.

[0054] S4.1 Analyze the executable scheduling schemes one by one according to the equipment type and maintenance sequence, and generate an operation instruction list based on the start time, end time and allocated resources of each task.

[0055] It should be noted that the executable scheduling schemes are classified according to equipment type, and maintenance tasks of the same equipment type are grouped together. Within each group, the maintenance tasks are parsed one by one according to their execution order in the scheduling scheme. Each parsing operation includes extracting the start time, end time, and assigned maintenance personnel and equipment information of the maintenance task. Based on the start time, end time, and allocated resources of each maintenance task obtained from the analysis, corresponding operable execution instructions are generated, and a complete list of operation instructions is formed according to the order of maintenance tasks. Each instruction clearly specifies the maintenance task, the corresponding equipment type, the task execution time period, and the allocated resources, ensuring that maintenance personnel can complete the maintenance tasks in sequence according to the operation instructions.

[0056] S4.2. Send the list of operation instructions to the corresponding electromechanical equipment control nodes, coordinate control according to the task sequence and execution time, and record the equipment control execution process in real time to form preliminary equipment execution status information.

[0057] It should be noted that the list of operation instructions is sequentially issued to the corresponding electromechanical equipment control nodes according to the maintenance task order. Each instruction includes the start time, end time, and information on the assigned maintenance personnel and tools / equipment. The electromechanical equipment control nodes coordinate operations according to the task order and execution time specified in the instructions, while simultaneously recording the equipment control execution status of each maintenance task in real time, including the actual start time, end time, and resource usage. By recording the execution process of all tasks, preliminary equipment execution status information is formed.

[0058] S4.3 Integrate and sort the preliminary equipment execution status information by time series to generate equipment control execution status records.

[0059] It should be noted that the preliminary equipment execution status information should be integrated according to equipment type and maintenance task sequence, and the actual start time, actual end time, personnel performing the maintenance, and usage of tools and equipment for each maintenance task should be categorized and organized. The integrated task execution records are sorted according to time sequence, and all equipment control execution processes are arranged in chronological order to form a continuous and complete record of equipment control execution status.

[0060] S5. Integrate the equipment control execution status records with historical electromechanical equipment operation data, construct the equipment's full life cycle degradation curve through spatiotemporal correlation analysis, and perform trend feature extraction and signal separation processing to generate an equipment full life cycle analysis report.

[0061] It should be noted that in the existing technology, equipment operation status analysis usually relies on historical electromechanical equipment operation data, and mainly evaluates equipment performance through statistical methods, empirical curves or simple trend fitting. Fang Faming integrates real-time equipment control execution status records with historical electromechanical equipment operation data, and constructs a full life cycle degradation curve of the equipment through multi-dimensional spatiotemporal correlation analysis. Based on the degradation curve, he combines trend feature extraction and signal separation processing methods to decompose and quantify different types of degradation signals, thereby obtaining more complete and dynamic equipment performance evolution characteristics and generating a full life cycle analysis report of the equipment. The core differences between our invention and existing technologies are: first, it integrates real-time equipment control execution status records with historical electromechanical equipment operation data, rather than relying solely on historical data; second, it employs multi-dimensional spatiotemporal correlation analysis and signal separation processing methods to simultaneously process multi-source information and multiple types of degradation signals, while existing technologies are mostly limited to single-source, single-variable analysis; and third, the generated degradation curves not only reflect changes in a single indicator but also provide dynamic evolution characteristics throughout the equipment's entire lifecycle, which can directly support subsequent maintenance prioritization and scheduling optimization. Our invention can accurately capture the performance degradation patterns of equipment throughout its entire lifecycle, quantify the dynamic characteristics of various degradation signals, and provide multi-source, multi-dimensional equipment status analysis results by combining spatiotemporal correlation information. Compared with existing technologies, it can significantly improve the accuracy and comprehensiveness of equipment health assessment, provide more reliable data support for maintenance decisions, scheduling optimization, and lifespan prediction, reduce operation and maintenance costs, and improve the safety and reliability of equipment operation.

[0062] S5.1 Align the equipment control execution status records with historical electromechanical equipment operation data according to equipment ID, timestamp, and spatial location, and construct a unified data view using a multidimensional index structure.

[0063] It should be noted that the equipment control execution status records are matched with historical electromechanical equipment operation data according to the equipment ID, the timestamps of each record are sorted, and the equipment is aligned in combination with the spatial distribution information of the equipment; based on the index structure, the equipment control execution status records and historical electromechanical equipment operation data are uniformly mapped to the same data view, realizing the synchronous arrangement and fast retrieval of data from different sources, thereby forming a unified data view that can be used for subsequent analysis; It should also be noted that the multidimensional index structure uses the device ID, timestamp, and spatial location in the device control execution status records and historical electromechanical equipment operation data as index dimensions, maps each device control execution status record to the corresponding multidimensional coordinate position, and establishes it using a tree structure.

[0064] S5.2. Based on the unified data view, the similarity of the states of computer equipment at adjacent time points and spatial locations is used to form a correlation matrix of equipment state evolution by integrating them according to weights.

[0065] It should be noted that, based on a unified data view, the state records of each electromechanical device at consecutive time points are taken sequentially. The similarity between states at adjacent time points is obtained through cosine similarity, and combined with the state similarity of the device at different spatial locations, the similarity in time and space dimensions is weighted and integrated to form the state evolution relationship of each device as its state changes with time and space. Arrange the state evolution relationships of all devices according to device index and time index to construct a two-dimensional matrix. Each element in the two-dimensional matrix represents the degree of state evolution relationship of the corresponding device at a specific time and with adjacent time and spatial location, thus forming a device state evolution relationship matrix.

[0066] S5.3. Weighted integration of state changes at adjacent time points and spatial locations in the equipment state evolution correlation matrix to construct the equipment's full life cycle degradation curve.

[0067] It should be noted that the state records of each electromechanical device at continuous time points and spatial locations are sequentially extracted from the equipment state evolution correlation matrix to form the state sequence of each electromechanical device; The state sequence of electromechanical equipment is divided into continuous node segments according to time intervals and spatial locations to ensure that each node segment can fully reflect the state evolution characteristics of the equipment within a local time and space range; By associating state nodes with similarity or correlation between them, state nodes at adjacent time points and spatial locations are sequentially linked to form a chain of device state evolution. Based on the contribution of the equipment state in the equipment state evolution chain to the overall degradation trend, weights are assigned to different time and space nodes; for example, high, medium, and low weights can be assigned according to the magnitude of state changes, the frequency of anomalies, or the importance of key performance indicators. The state nodes in the equipment state evolution chain are integrated and mapped according to their weight order, so that the state changes of each electromechanical device throughout the entire observation period can continuously reflect the trend of equipment performance degradation, anomaly accumulation and potential failure. The integrated state mapping results are arranged vertically to generate the state evolution trajectory of each electromechanical device on the time axis, and the full life cycle degradation curve of the device is displayed through visualization or graphical means.

[0068] S5.4. Using the first-order differential processing algorithm, signal processing is performed on the degradation curve of the entire equipment life cycle to extract the short-term trend changes of electromechanical equipment, and the long-term degradation trend and periodic fluctuation signals are separated by the wavelet transform decomposition algorithm.

[0069] It should be noted that, based on the degradation curve of the entire equipment life cycle, the degradation state of each piece of electromechanical equipment is observed sequentially at consecutive time points. The degradation state at each time point is compared with the degradation state at the previous time point. The state change value at each time point is generated by the first-order differential processing algorithm, forming the short-term trend change of electromechanical equipment that represents the rise or fall of equipment performance in a short period of time. This can clearly reflect the performance fluctuation and local degradation trend of each piece of electromechanical equipment in a short period of time. For the full life cycle degradation curve of each electromechanical device, wavelet transform decomposition is performed sequentially according to the time series. The original degradation curve is decomposed into multiple signal components of different scales. Each signal component corresponds to different frequency characteristics of equipment performance changes. By selecting low-frequency components to extract long-term degradation trends, while high-frequency components are used to reflect periodic fluctuations and short-term disturbances, the long-term degradation trend is clearly separated, providing continuous and identifiable long-term trend information for equipment performance degradation analysis.

[0070] S5.5 Integrate short-term trend changes, long-term degradation trends and periodic fluctuation signals of electromechanical equipment to generate a full life cycle analysis report of the equipment.

[0071] It should be noted that the short-term trend changes, long-term degradation trends, and periodic fluctuation signals extracted from each piece of electromechanical equipment are integrated in chronological order. The changes in different signal components are comprehensively analyzed and summarized, and the short-term fluctuations, long-term degradation trends, and periodic patterns of equipment performance are clearly presented, forming a full life cycle analysis report for each piece of electromechanical equipment.

[0072] This embodiment also provides a computer device applicable to the full lifecycle management method for electromechanical equipment on mountain roads, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the full lifecycle management method for electromechanical equipment on mountain roads as proposed in the above embodiment.

[0073] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0074] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the full lifecycle management method for electromechanical equipment on mountain roads as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0075] In summary, this invention achieves precise quantification and risk prioritization of the health status of electromechanical equipment on mountain roads under dynamic environmental constraints through multi-level Bayesian network fault propagation analysis based on complex causal relationships, thereby improving the pertinence and timeliness of maintenance decisions. Furthermore, by integrating equipment control execution status and historical operating data to construct a full life cycle degradation curve, separating short-term trends, long-term degradation, and periodic fluctuations, it enables refined monitoring and prediction of equipment's full life cycle operating performance, thus providing a reliable basis for scientifically formulating maintenance plans, reducing total life cycle costs, and extending equipment service life.

[0076] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for full life-cycle management of electromechanical equipment for mountain roads, characterized in that: include, Collect operational data of electromechanical equipment in mountainous highway tunnels, clean and standardize the operational data, and output a standardized equipment operation dataset; The standardized equipment operation dataset is input into the Bayesian network evaluation architecture. The posterior probability of each node under the dynamic constraints of environmental parameters is analyzed through multi-level fault propagation paths to generate an equipment health assessment report. Based on the equipment health assessment report, intelligent scheduling and resource optimization of electromechanical equipment maintenance tasks are performed on the maintenance task queue and limited resource set to generate an executable scheduling scheme. By utilizing an executable scheduling scheme, different types of electromechanical equipment in mountainous highway tunnels are coordinated and controlled, and the control execution status records are obtained. By integrating equipment control execution status records with historical electromechanical equipment operation data, a full life cycle degradation curve of the equipment is constructed through spatiotemporal correlation analysis. Trend feature extraction and signal separation processing are then performed to generate a full life cycle analysis report of the equipment.

2. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 1, characterized in that: The electromechanical equipment operation data includes operating status parameters, energy consumption data, environmental monitoring data, and fault alarm information.

3. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 2, characterized in that: The process of cleaning and standardizing the electromechanical equipment operation data to output a standardized equipment operation dataset involves the following steps. The operation data of electromechanical equipment is checked for timestamp continuity, field integrity and outliers, and missing and abnormal records are removed to obtain cleaned operation data; The cleaned operational data is standardized by unifying the units, format, and encoding, and then normalizing it to obtain standardized operational data. Standardized operational data is grouped, organized, and stored according to equipment type and time series to form a standardized equipment operation dataset.

4. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 3, characterized in that: The process involves inputting a standardized equipment operation dataset into a Bayesian network evaluation architecture. Through multi-level fault propagation path analysis, the posterior probability of each node under dynamic constraints of environmental parameters is determined, generating an equipment health assessment report. The specific steps are as follows: The standardized equipment operation dataset is input into the Bayesian network evaluation architecture according to equipment type and time series, and mapped to physical layer nodes, functional layer nodes and operational risk nodes to obtain the three-level node mapping results; Based on the three-level node mapping results, combined with historical fault data and spatiotemporal weight relationships between nodes, the causal dependencies of each node are quantified and a dynamic conditional probability table is constructed. Based on the dynamic conditional probability table, a multi-level fault propagation path analysis is performed, and the posterior probability of each node is calculated iteratively through forward propagation and backward correction. Based on the posterior probability of each node, the failure risk of physical layer nodes and functional layer nodes is mapped to the equipment health level, forming an equipment health status index. Combined with the probability ranking of operational risk nodes, maintenance priorities are generated, and an equipment health assessment report is output.

5. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 4, characterized in that: The proposed Bayesian network evaluation architecture is based on the three-level node system determined by the failure mode and impact analysis of electromechanical equipment. It quantifies the complex causal relationships between nodes through dynamic conditional probability tables and integrates spatiotemporal constraints for parameter optimization.

6. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 4, characterized in that: Based on the equipment health assessment report, the process involves intelligent scheduling and resource optimization of electromechanical equipment maintenance tasks within the maintenance task queue and limited resource set, generating an executable scheduling scheme. The specific steps are as follows: Based on the maintenance priorities in the equipment health assessment report, the tasks of maintaining electromechanical equipment are divided into maintenance task queues according to risk level and urgency. A resource availability matrix is ​​formed by performing attribute modeling on a finite set of resources and calculating the matching degree between each resource and the maintenance task. The finite set of resources is constructed by statistically analyzing the number of available maintenance personnel, their skill levels, and the types and quantities of available tools and equipment. The task queue and resource availability matrix are input into the intelligent scheduling algorithm, which optimizes the scheduling based on task priority, resource matching degree and task dependency relationship, and generates a preliminary scheduling scheme. The initial scheduling scheme is subjected to conflict detection, and optimized by iteratively adjusting the task order and resource allocation, resulting in an executable scheduling scheme.

7. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 6, characterized in that: The executable scheduling scheme includes a maintenance task instruction sequence and a dynamic resource allocation parameter set; wherein: the maintenance task instruction sequence includes an execution time window, the responsible personnel ID, and the required spare parts code; the dynamic resource allocation parameter set includes personnel path planning, spare parts allocation routes, and equipment downtime time windows.

8. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 6, characterized in that: The method utilizes an executable scheduling scheme to coordinate and control different types of electromechanical equipment within mountainous highway tunnels, recording the results and obtaining equipment control execution status records. The specific steps are as follows: The executable scheduling schemes are parsed one by one according to equipment type and maintenance sequence, and an operation instruction list is generated based on the start time, end time and allocated resources of each task. The list of operation instructions is sent to the corresponding electromechanical equipment control nodes, and coordinated control is carried out according to the task sequence and execution time. The equipment control execution process is recorded in real time to form preliminary equipment execution status information. The preliminary equipment execution status information is integrated and sorted by time series to generate equipment control execution status records.

9. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 8, characterized in that: The process involves fusing equipment control execution status records with historical electromechanical equipment operation data, constructing a full lifecycle degradation curve for the equipment through spatiotemporal correlation analysis, extracting trend features, and performing signal separation processing to generate a full lifecycle analysis report. The specific steps are as follows: Align the equipment control execution status records with historical electromechanical equipment operation data according to equipment ID, timestamp, and spatial location, and construct a unified data view using a multidimensional index structure; Based on a unified data view, the similarity of the states of computer equipment at adjacent time points and spatial locations is used and integrated according to weights to form an equipment state evolution correlation matrix. By weighted integration of state changes at adjacent time points and spatial locations in the equipment state evolution correlation matrix, a full life cycle degradation curve of the equipment is constructed. Using a first-order differential processing algorithm, signal processing is performed on the degradation curve of the entire equipment life cycle to extract the short-term trend changes of electromechanical equipment, and the long-term degradation trend and periodic fluctuation signals are separated by a wavelet transform decomposition algorithm. By integrating short-term trend changes, long-term degradation trends, and cyclical fluctuation signals of electromechanical equipment, a full life cycle analysis report of the equipment is generated.

10. The method for full life-cycle management of electromechanical equipment for mountain roads as described in claim 9, characterized in that: The multidimensional index structure uses the device ID, timestamp, and spatial location in the device control execution status records and historical electromechanical equipment operation data as index dimensions, maps each device control execution status record to the corresponding multidimensional coordinate position, and establishes a tree structure.

Images