Prediction maintenance method and system based on road infrastructure monitoring

Abstract

The invention discloses a prediction maintenance method and system based on road infrastructure monitoring, and relates to the related technical field of road operation and maintenance, and the method comprises the steps: collecting weekly traffic flow time sequence information and weekly people flow time sequence information; collecting annual temperature time sequence information, annual humidity time sequence information and annual precipitation time sequence information; when the first timing meets a first period, activating a first road infrastructure state prediction model to obtain a first road infrastructure state prediction result; when the second timing meets a second period, activating a second road infrastructure state prediction model to obtain a second road infrastructure state prediction result; obtaining a road infrastructure state fusion result through the road infrastructure state fusion model; and extracting an abnormal state type and abnormal triggering moment information. The technical problems that in the prior art, the detection period is long, and the maintenance response time is delayed are solved, and the technical effect of improving the road infrastructure state prediction accuracy and the maintenance timeliness is achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to road operation and maintenance, and specifically to a predictive maintenance method and system based on road infrastructure monitoring. Background Art

[0002] The maintenance and management of road infrastructure is the key to ensuring the efficient and safe operation of the transportation system. With the intensification of traffic flow and environmental changes, traditional road maintenance methods have been unable to meet the increasingly complex needs. At present, the operation and maintenance management of road infrastructure mainly relies on regular inspections and manual patrols. However, the coverage of manual patrols is limited, and it is impossible to monitor changes in road conditions in real time. Although regular inspections can detect some problems in time, due to the long detection cycle and delayed response time, the best maintenance time is often missed. It is also difficult to fully consider the environmental factors of road facilities, such as temperature, humidity, precipitation, etc., which all affect the service life and safety of road infrastructure, and thus affect the safety and reliability of roads.

[0003] Therefore, in the current related technologies, there are technical problems such as long detection cycle and delayed maintenance response time. Summary of the invention

[0004] This application solves the technical problems of long detection cycle and delayed maintenance response time existing in the prior art by providing a predictive maintenance method and system based on road infrastructure monitoring, thereby achieving the technical effect of improving the accuracy of road infrastructure status prediction and the timeliness of maintenance.

[0005] The present application provides a predictive maintenance method based on road infrastructure monitoring, the method comprising: collecting weekly traffic flow time series information and weekly passenger flow time series information; collecting annual temperature time series information, annual humidity time series information and annual precipitation time series information; whenever a first timing satisfies a first cycle, activating a first road infrastructure state prediction model, processing the weekly traffic flow time series information and the weekly passenger flow time series information, obtaining a first road infrastructure state prediction result, and re-counting the first timing from 0, wherein the first cycle is one week; whenever a second timing satisfies a second cycle, activating a second road infrastructure state prediction model, processing the annual temperature time series information, the annual humidity time series information and the annual precipitation time series information, obtaining a second road infrastructure state prediction result, and re-counting the second cycle from 0, wherein the second cycle is one year; obtaining a road infrastructure state fusion result for the first road infrastructure state prediction result and the second road infrastructure state prediction result through a road infrastructure state fusion model; extracting the abnormal state type and abnormal triggering time information of the road infrastructure state fusion result, and executing pre-maintenance configuration.

[0006] In a possible implementation, the first road infrastructure status prediction model is activated, the weekly vehicle flow time series information and the weekly pedestrian flow time series information are processed to obtain a first road infrastructure status prediction result, and the following processing is also performed: obtaining a road infrastructure initial state matrix; performing time domain aggregation on the weekly vehicle flow time series information and the weekly pedestrian flow time series information to obtain a first time domain binary flow, a second time domain binary flow, and up to the Qth time domain binary flow, wherein any one time domain has a vehicle flow identifier and a human flow identifier; the road infrastructure initial state matrix, the first time domain binary flow, the second time domain binary flow, and up to the Qth time domain binary flow are input into the first road infrastructure status prediction model, and the first road infrastructure status prediction result is output.

[0007] In a possible implementation, the weekly vehicle flow time series information and the weekly passenger flow time series information are aggregated in the time domain to obtain the first time domain binary flow, the second time domain binary flow, and up to the Qth time domain binary flow, and the following processing is also performed: the weekly vehicle flow time series information is aggregated in the time domain to obtain the first time domain vehicle flow, the second time domain vehicle flow, and up to the Nth time domain vehicle flow; the weekly passenger flow time series information is aggregated in the time domain to obtain the first time domain passenger flow, the second time domain passenger flow, and up to the Mth time domain passenger flow; the first time domain vehicle flow, the second time domain vehicle flow, and up to the Nth time domain vehicle flow are merged with the first time domain passenger flow, the second time domain passenger flow, and up to the Mth time domain passenger flow to obtain the first time domain binary flow, the second time domain binary flow, and up to the Qth time domain binary flow.

[0008] In a possible implementation, the road infrastructure initial state matrix, the first time-domain binary flow, the second time-domain binary flow, and up to the Qth time-domain binary flow are input into the first road infrastructure state prediction model, and the first road infrastructure state prediction result is output. The following processing is also performed: the road infrastructure initial state matrix, the first time-domain binary flow, and the first time-domain step length are input into the first road infrastructure state prediction model to obtain the first time-domain road infrastructure state prediction result; the first time-domain road infrastructure state prediction result, the second time-domain binary flow, and the second time-domain step length are input into the first road infrastructure state prediction model to obtain the second time-domain road infrastructure state prediction result; until the Q-1th time-domain road infrastructure state prediction result, the Qth time-domain binary flow, and the Qth time-domain step length are input into the first road infrastructure state prediction model to obtain the first road infrastructure state prediction result.

[0009] In a possible implementation, the first road infrastructure status prediction model is constructed, and the following processing is also performed: based on the road infrastructure type, weekly vehicle flow time series record information, weekly passenger flow time series record information and first road infrastructure status time series record information are collected; the weekly vehicle flow time series record information and the weekly passenger flow time series record information are time-domain aggregated to obtain first time-domain binary flow record information, second time-domain binary flow record information until the Yth time-domain binary flow record information; according to the first time domain, the second time domain until the Yth time domain, from the first road infrastructure status time series record information, the road infrastructure status at the start time of the first time domain, the road infrastructure status at the start time of the second time domain until the start time of the Yth time domain, and the road infrastructure status at the end time of the Yth time domain are extracted; the road infrastructure status at the start time of the first time domain is used as the basis for the prediction of the state of the first road infrastructure; The state of the road infrastructure at the starting moment of the second time domain and the second time domain binary flow record information are the first time domain input data of the long short-term memory neural network, the state of the road infrastructure at the starting moment of the second time domain and the second time domain binary flow record information are the second time domain input data of the long short-term memory neural network, and the state of the road infrastructure at the starting moment of the second time domain is the second time domain output supervision data of the long short-term memory neural network, until the road infrastructure state at the Yth time domain starting moment and the Yth time domain binary flow record information are the Yth time domain input data of the long short-term memory neural network, and the road infrastructure state at the Yth time domain end moment is the Yth time domain output supervision data of the long short-term memory neural network, to train the first road infrastructure state prediction model.

[0010] In a possible implementation, the second road infrastructure status prediction model is constructed and the following processing is also performed: based on the road infrastructure type, first sample data is collected, wherein the first sample data includes annual temperature time series record information, annual humidity time series record information, annual precipitation time series record information and second road infrastructure status time series record information; wherein the annual vehicle flow average of the first sample data is less than or equal to the vehicle flow threshold, and the annual pedestrian flow average of the first sample data is less than or equal to the pedestrian flow threshold; the second road infrastructure status prediction model is trained according to the annual temperature time series record information, the annual humidity time series record information, the annual precipitation time series record information and the second road infrastructure status time series record information.

[0011] In a possible implementation, the road infrastructure status fusion model is constructed, and the following processing is also performed: based on the road infrastructure type, second sample data is collected, wherein the second sample data includes second sample week traffic flow time series record information, second sample week passenger flow time series record information, second sample year temperature time series record information, second sample year humidity time series record information, second sample year precipitation time series record information and second sample road infrastructure status information; the second sample week traffic flow time series record information and the second sample week passenger flow time series record information are processed according to the first road infrastructure status prediction model to obtain first road infrastructure status identification data; the second sample week passenger flow time series record information, the second sample year temperature time series record information, the second sample year humidity time series record information and the second sample year precipitation time series record information are processed according to the second road infrastructure status prediction model to obtain second road infrastructure status identification data; the road infrastructure status fusion model is trained with the second sample road infrastructure status data as supervision and the first road infrastructure status identification data and the second road infrastructure status identification data as input.

[0012] The present application also provides a predictive maintenance system based on road infrastructure monitoring, including: a vehicle flow and passenger flow information collection module, used to collect weekly vehicle flow time series information and weekly passenger flow time series information; a meteorological data collection module, used to collect annual temperature time series information, annual humidity time series information and annual precipitation time series information; a first state prediction result acquisition module, used to activate the first road infrastructure state prediction model whenever the first timing meets the first cycle, process the weekly vehicle flow time series information and the weekly passenger flow time series information, obtain the first road infrastructure state prediction result, and re-count the first timing from 0, wherein the first cycle is one week; a second state prediction result acquisition module, used to activate the first road infrastructure state prediction model whenever the second timing meets the first cycle In the second period, the second road infrastructure status prediction model is activated, the annual temperature time series information, the annual humidity time series information and the annual precipitation time series information are processed to obtain a second road infrastructure status prediction result, and the second period is counted from 0, wherein the second period is one year; a status prediction result fusion module is used to fuse the first road infrastructure status prediction result and the second road infrastructure status prediction result through a road infrastructure status fusion model to obtain a road infrastructure status fusion result; a pre-maintenance configuration execution module is used to extract the abnormal status type and abnormal trigger time information of the road infrastructure status fusion result, and execute pre-maintenance configuration.

[0013] The predictive maintenance method and system based on road infrastructure monitoring proposed in this application is intended to collect weekly traffic flow time series information and weekly passenger flow time series information; collect annual temperature time series information, annual humidity time series information and annual precipitation time series information; whenever the first timing meets the first cycle, activate the first road infrastructure state prediction model to obtain the first road infrastructure state prediction result; whenever the second timing meets the second cycle, activate the second road infrastructure state prediction model to obtain the second road infrastructure state prediction result; obtain the road infrastructure state fusion result through the road infrastructure state fusion model; extract the abnormal state type and abnormal trigger time information, and perform pre-maintenance configuration. The technical problems of long detection cycle and delayed maintenance response time in the prior art are solved, and the technical effect of improving the accuracy of road infrastructure state prediction and the timeliness of maintenance is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic diagram of a predictive maintenance method based on road infrastructure monitoring provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a predictive maintenance system based on road infrastructure monitoring provided in an embodiment of the present application.

[0016] Explanation of the reference numerals: vehicle and passenger flow information collection module 10 , meteorological data collection module 20 , first state prediction result acquisition module 30 , second state prediction result acquisition module 40 , state prediction result fusion module 50 , pre-maintenance configuration execution module 60 . DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides a predictive maintenance method based on road infrastructure monitoring, such as Figure 1 As shown, the method includes: Step S100, collecting weekly vehicle flow time series information and weekly passenger flow time series information.

[0021] Preferably, the vehicle flow and pedestrian flow data within a week are periodically collected and recorded, organized and processed in a time series manner, and weekly vehicle flow time series information and weekly pedestrian flow time series information are obtained. Specifically, weekly vehicle flow time series information refers to recording the traffic flow on the road (usually in units of the number of vehicles passing through a specific road) by hour or finer time granularity within a week, such as the number of vehicles passing through a certain section of road per hour, vehicle type (such as cars, trucks, etc.), vehicle speed, etc., to reflect the traffic load of the road, traffic peak and trough periods, etc., which is helpful to predict the use of the road. Similarly, weekly pedestrian flow time series information refers to the recording of pedestrian flow on roads or in specific areas (usually in units of the number of pedestrians passing through a certain location) on an hourly or finer time scale within a week. For example, the number of pedestrians passing through a certain intersection or sidewalk per hour can also be further refined into pedestrian activity types (such as commuter peak, tourist flow, etc.), which helps to evaluate the load of road surfaces and infrastructure (such as sidewalks, traffic lights, etc.) and understand pedestrian activity patterns in different time periods, thereby assisting in predicting road usage.

[0022] Step S200, collecting annual temperature time series information, annual humidity time series information and annual precipitation time series information.

[0023] Preferably, meteorological data at different times of the year are collected and recorded, including temperature, humidity and precipitation, to form annual temperature time series information, annual humidity time series information and annual precipitation time series information. Specifically, annual temperature time series information refers to the temperature changes recorded at different times (such as daily, hourly, etc.) in a year. For example, data such as the highest temperature, lowest temperature, and average daily temperature of a certain area are collected. Annual temperature information can reflect seasonal climate changes and provide long-term data support for the impact of temperature on roads and infrastructure (such as pavements, bridges, tunnels, etc.). Temperature changes can affect the expansion and contraction of road materials and the freezing and thawing process, thereby affecting the structural stability and durability of roads. Annual humidity time series information refers to air humidity data recorded at different times (such as daily, hourly, etc.) in a year. For example, records Data such as hourly air humidity and daily average humidity, and annual humidity information are important for evaluating the impact of climate change, precipitation, and evaporation on road facilities in different seasons. A high humidity environment may cause road materials (such as asphalt and concrete) to become damp and corroded, affecting their structural stability and easily aggravating the aging of the road surface; annual precipitation time series information refers to precipitation data recorded in different time periods (such as daily, hourly, etc.) within a year. For example, daily precipitation or hourly precipitation data reflects the changes in precipitation in different seasons. Annual precipitation information is crucial for evaluating the long-term impact of precipitation on road infrastructure. Long-term precipitation (such as rain, snow, heavy rain, etc.) will affect the durability of roads, especially in rainy and snowy weather, where accumulated water may cause corrosion, subsidence, or structural damage to road facilities.

[0024] Step S300: Whenever the first timing satisfies the first cycle, the first road infrastructure status prediction model is activated, the weekly vehicle flow time series information and the weekly passenger flow time series information are processed to obtain the first road infrastructure status prediction result, and the first timing is recalculated from 0, wherein the first cycle is one week.

[0025] Preferably, the first time recording refers to the recording of time points according to the advancement of time during the operation of the system, which is usually a sign of a certain time trigger mechanism, such as updating the status at regular intervals. When the first time recording satisfies the first cycle, the first road infrastructure status prediction model is activated to process the weekly traffic flow time series information and the weekly passenger flow time series information, wherein the first cycle is one week, that is, whenever the time reaches a week, the first road infrastructure status prediction model is activated to process the weekly traffic flow time series information and the weekly passenger flow time series information. Specifically, the first road infrastructure status prediction model is based on the collected weekly traffic flow time series information and the weekly passenger flow time series information to predict the status of the road infrastructure, including the use status of the road, possible damage or the need for maintenance. For example, by analyzing the traffic and passenger flow data during peak traffic hours, it can be predicted which roads may be worn or damaged due to excessive use, and used as the first road infrastructure status prediction result. When a complete cycle (i.e., one week) ends, the first time recording is restarted from 0, that is, the statistical data of the previous cycle is cleared, and the new cycle data is recorded again to ensure that the data in each week can be accurately tracked and predicted.

[0026] Furthermore, step S300 also includes step S310, obtaining the initial state matrix of the road infrastructure; step S320, performing time domain aggregation on the weekly vehicle flow time series information and the weekly pedestrian flow time series information to obtain the first time domain binary flow, the second time domain binary flow, and up to the Qth time domain binary flow, wherein any time domain has a vehicle flow identifier and a human flow identifier; step S330, inputting the initial state matrix of the road infrastructure, the first time domain binary flow, the second time domain binary flow, and up to the Qth time domain binary flow into the first road infrastructure state prediction model, and outputting the first road infrastructure state prediction result.

[0027] Preferably, existing condition data of road infrastructure is obtained, including road surface damage, traffic flow, environmental impact, historical fault records, etc., and these data are represented as a numerical matrix of the current health status of road facilities, that is, an initial state matrix of road infrastructure is obtained. The initial state matrix of road infrastructure can contain different types of values, such as whether there are cracks, potholes, etc. in the road condition, the current road traffic flow, and information on the health status of road infrastructure such as road wear, damage to traffic facilities, etc.

[0028] Preferably, time-domain aggregation is performed on the weekly vehicle flow time series information and the weekly pedestrian flow time series information, which means aggregating and converting the weekly vehicle flow time series information and the weekly pedestrian flow time series information into binary flows at certain time intervals (such as hours, days, etc.), that is, aggregating and associating the vehicle flow and the pedestrian flow at the same time to obtain two identifiers, the vehicle flow identifier and the pedestrian flow identifier, wherein the vehicle flow identifier is the traffic flow in a certain time period, usually the number of vehicles or traffic density, and the pedestrian flow identifier is the pedestrian flow in the same time period, usually the number of pedestrians passing through a specific road section, thereby obtaining multiple time-domain binary flows, including a first time-domain binary flow, a second time-domain binary flow, and up to the Qth time-domain binary flow, wherein the first time-domain binary flow includes the vehicle flow and pedestrian flow data of a certain time period (such as the first hour), and the second time-domain binary flow includes the vehicle flow and pedestrian flow data of the second time period, covering Q time periods, indicating the change of traffic load in different time periods, and Q is a positive integer greater than 1.

[0029] Preferably, the initial state matrix of the road infrastructure and the binary flow data obtained through time domain aggregation (including binary flows in the first time domain, the second time domain, and up to the Qth time domain) are input into the first road infrastructure state prediction model. The model predicts the future health state of the road infrastructure (such as possible damage, failure, etc.) based on the input data (such as traffic flow, pedestrian flow, infrastructure initial state matrix, etc.), which may include predictions of the health state of the road, such as whether new damage is likely to occur and whether maintenance is required; risk assessment over a period of time in the future, predicting which road sections may have problems due to excessive traffic load, aging or environmental factors based on flow data and initial state; and then outputting the first road infrastructure state prediction result to help more accurately assess the health state of the road infrastructure and future maintenance needs.

[0030] Furthermore, step S320 also includes step S321, performing time-domain aggregation on the weekly vehicle flow time series information to obtain a first time-domain vehicle flow, a second time-domain vehicle flow, and up to the Nth time-domain vehicle flow; step S322, performing time-domain aggregation on the weekly pedestrian flow time series information to obtain a first time-domain pedestrian flow, a second time-domain pedestrian flow, and up to the Mth time-domain pedestrian flow; step S323, fusing the first time-domain vehicle flow, the second time-domain vehicle flow, and up to the Nth time-domain vehicle flow with the first time-domain pedestrian flow, the second time-domain pedestrian flow, and up to the Mth time-domain pedestrian flow to obtain the first time-domain binary flow, the second time-domain binary flow, and up to the Qth time-domain binary flow.

[0031] Preferably, time domain aggregation of weekly traffic flow time series information refers to aggregating these traffic flow data according to specific time periods, and the traffic flow in each time period (time domain) will be processed into an aggregate value, so as to obtain more simplified and operational traffic information. For example, according to the data of a week, it is divided into several time periods (such as every hour), and then the traffic flow in each time period is aggregated, and finally the first time domain traffic flow, the second time domain traffic flow, and the Nth time domain traffic flow are obtained. Each time domain represents the sum of the traffic flow in a period of time, wherein the first time domain traffic flow refers to the sum of the traffic flow in the first time period (for example: the first hour or the first day), the second time domain traffic flow refers to the traffic flow in the second time period, and so on. N is a positive integer greater than 1.

[0032] Preferably, similarly, time domain aggregation of weekly pedestrian flow time series information refers to summarizing or processing these pedestrian flow data according to a specific time period. For example, the pedestrian flow data is divided into multiple time periods by hours, days, etc., and the flow data in each time period is summarized to finally obtain the first time domain pedestrian flow, the second time domain pedestrian flow, and the Mth time domain pedestrian flow, where each time domain represents the total pedestrian flow in the time period, wherein the first time domain pedestrian flow refers to the pedestrian flow in the first time period (for example: the first hour or the first day), the second time domain pedestrian flow refers to the pedestrian flow in the second time period, and so on. M is a positive integer greater than 1.

[0033] Preferably, after completing the time domain aggregation of vehicle flow and pedestrian flow, the first time domain vehicle flow, the second time domain vehicle flow to the Nth time domain vehicle flow and the first time domain pedestrian flow, the second time domain pedestrian flow to the Mth time domain pedestrian flow are fused, that is, the values ​​of vehicle flow and pedestrian flow in each time period are combined to form a binary flow, including combining the first time domain vehicle flow and the first time domain pedestrian flow to obtain the first time domain binary flow representing the comprehensive traffic conditions in the time period, wherein the binary flow represents the comprehensive data of vehicle flow and pedestrian flow simultaneously in a time period, and through time domain fusion, the impact of vehicle flow and pedestrian flow on road infrastructure can be simultaneously evaluated, thereby obtaining a more comprehensive road status prediction and providing a more comprehensive road health assessment.

[0034] Furthermore, step S330 also includes step S331, inputting the road infrastructure initial state matrix, the first time domain binary flow and the first time domain step into the first road infrastructure state prediction model to obtain the first time domain road infrastructure state prediction result; step S332, inputting the first time domain road infrastructure state prediction result, the second time domain binary flow and the second time domain step into the first road infrastructure state prediction model to obtain the second time domain road infrastructure state prediction result; step S333, until the Q-1th time domain road infrastructure state prediction result, the Qth time domain binary flow and the Qth time domain step are input into the first road infrastructure state prediction model to obtain the first road infrastructure state prediction result.

[0035] Preferably, the first time domain step, the second time domain step to the Qth time domain step refer to the time span (i.e., step) of each time domain, which is used to control the prediction period of the model, and may be a specific time period (such as hours, days, etc.). The initial state matrix of the road infrastructure, the binary flow of the first time domain, and the first time domain step are used as input data to provide to the first road infrastructure state prediction model, and the road state of the first time period is evaluated, and the output result is the prediction result of the road infrastructure state in the first time domain, including wear, cracks, etc.; the prediction result of the road infrastructure state in the first time domain is then used as input, together with the binary flow of the second time domain and the second time domain step, to input into the first road infrastructure state prediction model, and the road state of the second time period is evaluated, and the output result is the prediction result of the road infrastructure state in the second time domain; according to the same logic, the prediction result of the previous time domain is used as one of the inputs of the current time domain, and then combined with the binary flow and step of the current time domain, the state of the current time domain is predicted, and finally processed to the Qth time domain, and the prediction result of the first road infrastructure state is output, so as to dynamically predict the health state of the road infrastructure.

[0036] Further, step S300 also includes step S340, based on the road infrastructure type, collecting weekly vehicle flow time series record information, weekly passenger flow time series record information and first road infrastructure status time series record information; step S350, performing time domain aggregation on the weekly vehicle flow time series record information and the weekly passenger flow time series record information to obtain first time domain binary flow record information, second time domain binary flow record information until the Yth time domain binary flow record information; step S360, according to the first time domain, the second time domain until the Yth time domain, from the first road infrastructure status time series record information, extract the road infrastructure status at the start time of the first time domain, the road infrastructure status at the start time of the second time domain until the start time of the Yth time domain, and the road infrastructure status at the end time of the Yth time domain; step S370, based on the road infrastructure at the start time of the first time domain, the road infrastructure status at the start time of the second time domain The facility status and the first time domain binary flow record information are used as the first time domain input data of the long short-term memory neural network, the road infrastructure status at the starting moment of the second time domain is used as the first time domain output supervision data of the long short-term memory neural network, the road infrastructure status at the starting moment of the second time domain and the second time domain binary flow record information are used as the second time domain input data of the long short-term memory neural network, and the road infrastructure status at the starting moment of the second time domain is used as the second time domain output supervision data of the long short-term memory neural network, until the road infrastructure status at the Yth time domain starting moment and the Yth time domain binary flow record information are used as the Yth time domain input data of the long short-term memory neural network, and the road infrastructure status at the Yth time domain end moment is used as the Yth time domain output supervision data of the long short-term memory neural network, to train the first road infrastructure status prediction model.

[0037] Preferably, according to the type of road infrastructure, weekly traffic flow time series record information (recording the changes in traffic volume over time within a week), weekly passenger flow time series record information (recording the changes in passenger flow over time within a week) and first road infrastructure status time series record information (recording the status of the road at different time points, including the degree of wear and tear, cracks, etc.) are collected, and then the weekly traffic flow time series record information and the weekly passenger flow time series record information are aggregated in the time domain, that is, the weekly traffic flow and weekly passenger flow time series record information are aggregated according to time periods into first time domain binary flow record information, second time domain binary flow record information, until the Yth time domain binary flow record information, each binary flow record information contains comprehensive data of vehicle flow and passenger flow in the corresponding time domain.

[0038] Preferably, according to the first time domain, the second time domain, and up to the Yth time domain, the road infrastructure status at the starting moment of each time domain is extracted from the time series record information of the first road infrastructure status, including the road infrastructure status at the starting moment of the first time domain, the road infrastructure status at the starting moment of the second time domain, and the road infrastructure status at the end moment of the Yth time domain, for supervision model training to guide the model to learn to predict changes in road status in each time period; then a prediction model is constructed based on a long short-term memory neural network (LSTM), with the road infrastructure status at the starting moment of the first time domain and the first time domain binary flow record information as the first time domain input data of the long short-term memory neural network, and the road infrastructure status at the starting moment of the second time domain as the output supervision data, to guide the long short-term memory neural network to learn how to predict the road status of the next time domain based on the initial road status and flow data.

[0039] Preferably, the road infrastructure state at the start time of the second time domain and the second time domain binary flow record information are used as the second time domain input data of the long short-term memory neural network, and the road infrastructure state at the start time of the second time domain is used as the second time domain output supervision data of the long short-term memory neural network, until the road infrastructure state at the start time of the Yth time domain and the Yth time domain binary flow record information are used as the Yth time domain input data of the long short-term memory neural network, and the road infrastructure state at the end time of the Yth time domain is used as the Yth time domain output supervision data of the long short-term memory neural network, which is used to capture the time dynamic characteristics of the road state, and by using the start and end states of each time domain as supervision data to guide training, it is ensured that the prediction results output by the model are consistent with the actual recorded state, and learn to predict the state of the next time period derived from the current state and flow conditions. Among them, the long short-term memory neural network is suitable for processing long time series data, can memorize remote historical information (such as initial state) and combine it with the current input for prediction, and is suitable for predicting road states with strong time dependence. The road infrastructure state prediction model can dynamically predict the future state of the road infrastructure, well combine short-term traffic load with long-term state changes, and provide strong data support and decision-making basis for subsequent prediction and maintenance.

[0040] Step S400, whenever the second timing satisfies the second period, activate the second road infrastructure status prediction model, process the annual temperature time series information, the annual humidity time series information and the annual precipitation time series information, obtain the second road infrastructure status prediction result, and count the second period from 0, wherein the second period is one year.

[0041] Preferably, the second timing refers to a time mark when a certain time point reaches a one-year cycle, that is, the time reaches a complete annual cycle, which is a periodically triggered moment. When the second timing satisfies the second cycle, the second road infrastructure status prediction model is activated to process the annual temperature time series information, the annual humidity time series information and the annual precipitation time series information. Among them, the second cycle refers to a one-year time period, that is, the annual meteorological data is counted and processed again every year to predict the road infrastructure status. Specifically, the second road infrastructure status prediction model is used to process annual meteorological data, and the long-term trend of the road infrastructure status is performed through the annual meteorological data (such as annual temperature, humidity, precipitation, etc.). Prediction: At the end of each annual cycle, the model is activated and the temperature, humidity and precipitation data of the past year are used as input to predict the long-term health of the road. For example, it is predicted that due to temperature fluctuations, humidity changes or precipitation, some roads may experience aging, crack expansion, structural damage and other problems, and then the second road infrastructure status prediction result is obtained. When the second cycle (i.e. one year) ends, the second cycle is counted from 0. That is, at the end of the one-year cycle, all statistical information related to the annual meteorological data is reset, and the data collection and analysis for the new year begins, so that the impact of annual meteorological changes on road infrastructure can be continuously and dynamically evaluated, and maintenance plans can be adjusted in a timely manner.

[0042] Preferably, the first sample data is collected according to the type of road infrastructure, including annual temperature time series record information, annual humidity time series record information, annual precipitation time series record information and second road infrastructure status time series record information. The annual temperature time series record information records the temperature fluctuation in each time period within a year (such as monthly and daily average temperature, etc.), the annual humidity time series record information records the air humidity conditions in different time periods within a year (such as daily average humidity, humidity fluctuation, etc.), and the annual precipitation time series record information records the time series data of precipitation in a year, such as monthly or daily precipitation, which helps to analyze the long-term impact of meteorological conditions on roads. The second road infrastructure status time series record information records the long-term status of road facilities, including information such as road wear, cracks, and settlement, which is used to analyze the health changes of road facilities under different meteorological conditions.

[0043] Preferably, the mean vehicle flow and the mean passenger flow in the first sample data should meet the set threshold conditions, that is, the annual average vehicle flow cannot exceed a certain specific value, which means that the area corresponding to the data set has not encountered excessive traffic load, and the annual average passenger flow must also be lower than a preset threshold, which limits the scope of the sample data, so that the samples for model training are concentrated on road infrastructure with lower traffic loads, and the vehicle flow threshold and the passenger flow threshold avoid the road state under high traffic load conditions, so as to analyze the reaction and aging mode of the road under relatively normal or low load conditions; then the time series recording information of annual temperature, humidity, precipitation and the second road infrastructure state is used to train the second road infrastructure state prediction model.

[0044] Preferably, a prediction model is constructed based on a long short-term memory neural network, using annual temperature time series record information, annual humidity time series record information, and annual precipitation time series record information as input data, and using the second road infrastructure status time series record information as output supervision data, the prediction model is trained to obtain a second road infrastructure status prediction model, which can predict the status of road infrastructure according to meteorological conditions (such as temperature, humidity, and precipitation), and help predict possible changes in road facilities in the future. For example, higher humidity or precipitation may cause accelerated corrosion or aging of road facilities, while high temperature may cause expansion and cracking of road surface materials, thereby formulating more efficient preventive maintenance plans.

[0045] Step S500: Obtain a road infrastructure state fusion result by using a road infrastructure state fusion model to perform fusion of the first road infrastructure state prediction result and the second road infrastructure state prediction result.

[0046] Preferably, a road infrastructure status fusion model is used to fuse a first road infrastructure status prediction result and a second road infrastructure status prediction result obtained based on different data sources (such as vehicle flow, passenger flow, and meteorological data such as temperature, humidity, and precipitation). Usually, some mathematical or statistical methods (such as weighted average, data fusion algorithm, machine learning, etc.) are used to combine the two prediction results (the first prediction result and the second prediction result) to obtain a road infrastructure status fusion result, wherein the road infrastructure status fusion model not only considers the impact of short-term traffic load, but also the role of long-term environmental factors, so as to more accurately and comprehensively reflect the true health status of road facilities.

[0047] Further, step S500 also includes step S510, collecting second sample data based on the road infrastructure type, wherein the second sample data includes second sample week traffic flow time series record information, second sample week passenger flow time series record information, second sample year temperature time series record information, second sample year humidity time series record information, second sample year precipitation time series record information and second sample road infrastructure status information; step S520, processing the second sample week traffic flow time series record information and the second sample week passenger flow time series record information according to the first road infrastructure status prediction model to obtain first road infrastructure status identification data; step S530, processing the second sample week passenger flow time series record information, the second sample year temperature time series record information, the second sample year humidity time series record information and the second sample year precipitation time series record information according to the second road infrastructure status prediction model to obtain second road infrastructure status identification data; step S540, using the second sample road infrastructure status data as supervision and the first road infrastructure status identification data and the second road infrastructure status identification data as input to train the road infrastructure status fusion model.

[0048] Preferably, according to the type of road infrastructure, second sample data is collected, including second sample weekly traffic time series record information (traffic flow data within a week in the second sample), second sample weekly passenger flow time series record information (passenger flow data within a week in the second sample), second sample annual temperature time series record information (recording temperature changes in the second sample for one year), second sample annual humidity time series record information (humidity changes in the second sample for one year), second sample annual precipitation time series record information (precipitation data in the second sample for one year) and second sample road infrastructure status information (road facility status data of the second sample at different time points, including road health status, degree of damage, wear, cracks and other information), the first road infrastructure status prediction model is used to process the weekly traffic flow time series record information and the weekly passenger flow time series record information in the second sample, and the first road infrastructure status identification data is output, that is, the prediction result of the traffic flow data based on the first model, which is used to indicate the health status of the road in that period of time.

[0049] Preferably, a second road infrastructure status prediction model is used to process the second sample week passenger flow time series record information, the second sample year temperature time series record information, the second sample year humidity time series record information and the second sample year precipitation time series record information, and output second road infrastructure status identification data, that is, based on the prediction results of the second model, to evaluate the health status of the road under these meteorological conditions; finally, the first road infrastructure status identification data and the second road infrastructure status identification data are used as input data, and the second sample road infrastructure status data is used as supervision data for fusion training to obtain a road infrastructure status fusion model, which is used to combine the prediction results from the two prediction models (based on traffic flow and based on meteorological conditions), and combine short-term and long-term impacts, so as to improve the prediction accuracy of the future status of road facilities, obtain a more comprehensive and accurate road status prediction, and then generate a more accurate road maintenance prediction.

[0050] Step S600: extract the abnormal state type and abnormal triggering time information of the road infrastructure state fusion result, and perform pre-maintenance configuration.

[0051] Preferably, by analyzing the fusion results of the road infrastructure status, abnormal conditions are identified, and appropriate preventive maintenance measures are taken based on the abnormal information. Specifically, through the fused road infrastructure status prediction results, it is found which specific road problems or fault types may occur. The abnormal conditions can be problems in the road structure, surface damage, excessive wear, cracks, settlement, corrosion, etc. For example, cracks in the road surface caused by excessive traffic flow on a certain section of road, or abnormal conditions such as water accumulation and bridge corrosion caused by excessive precipitation are detected; the abnormal trigger time information refers to the specific time point when the problem or fault is found in the road infrastructure status prediction. , determine when an anomaly occurs based on the data cycle (such as a week or a year), which usually corresponds to specific traffic loads or meteorological conditions. For example, if an abnormality is found on a certain section of road during a specific time period (such as peak hours or bad weather), it may be because of excessive traffic, excessive precipitation or low temperature at this time, causing problems on the road; Preventive maintenance configuration refers to taking maintenance measures in advance based on the prediction results after the abnormal state and its triggering time are discovered. Among them, predictive maintenance configuration is a prediction-based maintenance strategy that can avoid safety accidents caused by infrastructure damage by identifying potential problems in advance, thereby reducing maintenance costs and improving the service life and safety of roads.

[0052] In the above, refer to Figure 1 The predictive maintenance method based on road infrastructure monitoring according to an embodiment of the present invention is described in detail. Figure 2 A predictive maintenance system based on road infrastructure monitoring according to an embodiment of the present invention is described.

[0053] The predictive maintenance system based on road infrastructure monitoring according to the embodiment of the present invention is used to solve the technical problems of long detection cycle and delayed maintenance response time in the prior art, and achieves the technical effect of improving the accuracy of road infrastructure state prediction and the timeliness of maintenance. The predictive maintenance system based on road infrastructure monitoring includes: a vehicle flow and pedestrian flow information collection module 10, a meteorological data collection module 20, a first state prediction result acquisition module 30, a second state prediction result acquisition module 40, a state prediction result fusion module 50, and a pre-maintenance configuration execution module 60.

[0054] The vehicle flow and pedestrian flow information collection module 10 is used to collect weekly vehicle flow time series information and weekly pedestrian flow time series information; the meteorological data collection module 20 is used to collect annual temperature time series information, annual humidity time series information and annual precipitation time series information; the first state prediction result acquisition module 30 is used to activate the first road infrastructure state prediction model whenever the first timing meets the first cycle, process the weekly vehicle flow time series information and the weekly pedestrian flow time series information, obtain the first road infrastructure state prediction result, and re-count the first timing from 0, wherein the first cycle is one week; the second state prediction result acquisition module 40 is used to activate the second road infrastructure state prediction model whenever the second timing meets the second cycle The road infrastructure state prediction model processes the annual temperature time series information, the annual humidity time series information and the annual precipitation time series information to obtain a second road infrastructure state prediction result, and starts counting the second period from 0, wherein the second period is one year; a state prediction result fusion module 50 is used to fuse the first road infrastructure state prediction result and the second road infrastructure state prediction result through the road infrastructure state fusion model to obtain a road infrastructure state fusion result; a pre-maintenance configuration execution module 60 is used to extract the abnormal state type and abnormal triggering time information of the road infrastructure state fusion result, and execute pre-maintenance configuration.

[0055] The specific configuration of the first state prediction result obtaining module 30 will be described in detail below. The first state prediction result obtaining module 30 further includes: obtaining the initial state matrix of the road infrastructure; performing time domain aggregation on the weekly vehicle flow time series information and the weekly passenger flow time series information to obtain the first time domain binary flow, the second time domain binary flow until the Qth time domain binary flow, wherein any time domain has a vehicle flow identifier and a human flow identifier; inputting the initial state matrix of the road infrastructure, the first time domain binary flow, the second time domain binary flow until the Qth time domain binary flow into the first road infrastructure state prediction model, and outputting the first road infrastructure state prediction result.

[0056] The specific configuration of the first state prediction result obtaining module 30 will be described in detail below. The first state prediction result obtaining module 30 further includes: performing time-domain aggregation on the weekly vehicle flow time series information to obtain the first time-domain vehicle flow, the second time-domain vehicle flow, and the Nth time-domain vehicle flow; performing time-domain aggregation on the weekly pedestrian flow time series information to obtain the first time-domain pedestrian flow, the second time-domain pedestrian flow, and the Mth time-domain pedestrian flow; fusing the first time-domain vehicle flow, the second time-domain vehicle flow, and the Nth time-domain vehicle flow with the first time-domain pedestrian flow, the second time-domain pedestrian flow, and the Mth time-domain pedestrian flow to obtain the first time-domain binary flow, the second time-domain binary flow, and the Qth time-domain binary flow.

[0057] The specific configuration of the first state prediction result obtaining module 30 will be described in detail below. The first state prediction result obtaining module 30 further includes: inputting the road infrastructure initial state matrix, the first time domain binary flow and the first time domain step length into the first road infrastructure state prediction model to obtain the first time domain road infrastructure state prediction result; inputting the first time domain road infrastructure state prediction result, the second time domain binary flow and the second time domain step length into the first road infrastructure state prediction model to obtain the second time domain road infrastructure state prediction result; until the Q-1th time domain road infrastructure state prediction result, the Qth time domain binary flow and the Qth time domain step length are input into the first road infrastructure state prediction model to obtain the first road infrastructure state prediction result.

[0058] The specific configuration of the first state prediction result obtaining module 30 will be described in detail below. The first state prediction result obtaining module 30 further includes: based on the road infrastructure type, collecting weekly traffic flow time series record information, weekly passenger flow time series record information and first road infrastructure state time series record information; performing time domain aggregation on the weekly traffic flow time series record information and the weekly passenger flow time series record information to obtain first time domain binary flow record information, second time domain binary flow record information until the Yth time domain binary flow record information; according to the first time domain, the second time domain until the Yth time domain, from the first road infrastructure state time series record information, extract the road infrastructure state at the start time of the first time domain, the road infrastructure state at the start time of the second time domain until the start time of the Yth time domain, and the road infrastructure state at the end time of the Yth time domain; with the road infrastructure state at the start time of the first time domain, the road infrastructure state at the first time domain, the road infrastructure state at the start time of the second time domain, the road infrastructure state at the end time of the Yth time domain, and the road infrastructure state at the end time of the Yth time domain; with the road infrastructure state at the start time of the first time domain, the road infrastructure state at the first time domain, the road infrastructure state at the second time domain, the road infrastructure state at the end time of the second time domain, and the road infrastructure state at the end time of the first time domain. The first road infrastructure state prediction model is trained by using the road infrastructure state at the starting time of the Yth time domain and the Yth time domain binary flow record information as the first time domain input data of the long short-term memory neural network, the road infrastructure state at the starting time of the second time domain and the second time domain binary flow record information as the first time domain output supervision data of the long short-term memory neural network, the road infrastructure state at the starting time of the second time domain and the second time domain binary flow record information as the second time domain input data of the long short-term memory neural network, and the road infrastructure state at the starting time of the second time domain is the second time domain output supervision data of the long short-term memory neural network.

[0059] The specific configuration of the second state prediction result obtaining module 40 will be described in detail below. The second state prediction result obtaining module 40 further includes: based on the road infrastructure type, collecting first sample data, wherein the first sample data includes annual temperature time series record information, annual humidity time series record information, annual precipitation time series record information and second road infrastructure state time series record information; wherein the annual vehicle flow mean of the first sample data is less than or equal to the vehicle flow threshold, and the annual pedestrian flow mean of the first sample data is less than or equal to the pedestrian flow threshold; training the second road infrastructure state prediction model according to the annual temperature time series record information, the annual humidity time series record information, the annual precipitation time series record information and the second road infrastructure state time series record information.

[0060] The specific configuration of the state prediction result fusion module 50 will be described in detail below. The state prediction result fusion module 50 further includes: based on the road infrastructure type, collecting second sample data, wherein the second sample data includes the second sample week traffic flow time series record information, the second sample week passenger flow time series record information, the second sample year temperature time series record information, the second sample year humidity time series record information, the second sample year precipitation time series record information and the second sample road infrastructure state information; processing the second sample week traffic flow time series record information and the second sample week passenger flow time series record information according to the first road infrastructure state prediction model to obtain the first road infrastructure state identification data; processing the second sample week passenger flow time series record information, the second sample year temperature time series record information, the second sample year humidity time series record information and the second sample year precipitation time series record information according to the second road infrastructure state prediction model to obtain the second road infrastructure state identification data; using the second sample road infrastructure state data as supervision, and using the first road infrastructure state identification data and the second road infrastructure state identification data as input, to train the road infrastructure state fusion model.

[0061] The predictive maintenance system based on road infrastructure monitoring provided by the embodiment of the present invention can execute the predictive maintenance method based on road infrastructure monitoring provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0062] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0063] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, 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 application should be included in the protection scope of this application.

Claims

1. A predictive maintenance method based on road infrastructure monitoring, characterized in that: include: Collect weekly traffic flow time series information and weekly passenger flow time series information; Collect annual temperature time series information, annual humidity time series information and annual precipitation time series information; Whenever the first time count satisfies a first period, a first road infrastructure state prediction model is activated, the weekly vehicle flow time series information and the weekly passenger flow time series information are processed to obtain a first road infrastructure state prediction result, and the first time count is recalculated from 0, wherein the first period is one week; Whenever the second timing satisfies the second period, the second road infrastructure state prediction model is activated, the annual temperature time series information, the annual humidity time series information, and the annual precipitation time series information are processed to obtain a second road infrastructure state prediction result, and the second period is counted from 0, wherein the second period is one year; Obtaining a road infrastructure state fusion result by using a road infrastructure state fusion model for the first road infrastructure state prediction result and the second road infrastructure state prediction result; The abnormal state type and abnormal triggering time information of the road infrastructure state fusion result are extracted, and the pre-maintenance configuration is performed.

2. The predictive maintenance method based on road infrastructure monitoring according to claim 1, characterized in that: Whenever the first timing satisfies the first cycle, the first road infrastructure state prediction model is activated, the weekly vehicle flow time series information and the weekly passenger flow time series information are processed, and a first road infrastructure state prediction result is obtained, including: Obtaining the initial state matrix of road infrastructure; Performing time-domain aggregation on the weekly vehicle flow time-series information and the weekly pedestrian flow time-series information to obtain a first time-domain binary flow, a second time-domain binary flow, and up to a Qth time-domain binary flow, wherein any time domain has a vehicle flow identifier and a pedestrian flow identifier; The road infrastructure initial state matrix, the first time-domain binary flow, the second time-domain binary flow, and up to the Qth time-domain binary flow are input into the first road infrastructure state prediction model, and the first road infrastructure state prediction result is output.

3. The predictive maintenance method based on road infrastructure monitoring according to claim 2, characterized in that: Performing time-domain aggregation on the weekly vehicle flow time-series information and the weekly passenger flow time-series information to obtain first time-domain binary flow, second time-domain binary flow, and up to Qth time-domain binary flow, includes: Performing time-domain aggregation on the weekly traffic flow time series information to obtain a first time-domain traffic flow, a second time-domain traffic flow, and finally an Nth time-domain traffic flow; Performing time-domain aggregation on the weekly crowd flow time series information to obtain a first time-domain crowd flow, a second time-domain crowd flow, and finally an Mth time-domain crowd flow; The first time-domain vehicle flow, the second time-domain vehicle flow, up to the Nth time-domain vehicle flow, and the first time-domain pedestrian flow, the second time-domain pedestrian flow, up to the Mth time-domain pedestrian flow are merged to obtain the first time-domain binary flow, the second time-domain binary flow, up to the Qth time-domain binary flow.

4. The predictive maintenance method based on road infrastructure monitoring according to claim 2, characterized in that: Inputting the road infrastructure initial state matrix, the first time-domain binary flow, the second time-domain binary flow, and up to the Qth time-domain binary flow into the first road infrastructure state prediction model, and outputting the first road infrastructure state prediction result, including: Inputting the road infrastructure initial state matrix, the first time-domain binary flow, and the first time-domain step size into the first road infrastructure state prediction model to obtain a first time-domain road infrastructure state prediction result; Inputting the first time-domain road infrastructure state prediction result, the second time-domain binary flow, and the second time-domain step length into the first road infrastructure state prediction model to obtain a second time-domain road infrastructure state prediction result; Until the Q-1th time-domain road infrastructure state prediction result, the Qth time-domain binary flow and the Qth time-domain step length are input into the first road infrastructure state prediction model, the first road infrastructure state prediction result is obtained.

5. The predictive maintenance method based on road infrastructure monitoring according to claim 1, characterized in that: The first road infrastructure status prediction model building step comprises: Based on the road infrastructure type, weekly vehicle flow time series record information, weekly passenger flow time series record information and first road infrastructure status time series record information are collected; Performing time-domain aggregation on the weekly vehicle flow time-series record information and the weekly passenger flow time-series record information to obtain first time-domain binary flow record information, second time-domain binary flow record information, and up to Yth time-domain binary flow record information; According to the first time domain, the second time domain, and up to the Yth time domain, extracting the road infrastructure state at the start time of the first time domain, the road infrastructure state at the start time of the second time domain, and up to the start time of the Yth time domain, and the road infrastructure state at the end time of the Yth time domain from the first road infrastructure state time series record information; The road infrastructure state at the starting moment of the first time domain and the first time domain binary flow record information are used as the first time domain input data of the long short-term memory neural network, the road infrastructure state at the starting moment of the second time domain is used as the first time domain output supervision data of the long short-term memory neural network, the road infrastructure state at the starting moment of the second time domain and the second time domain binary flow record information are used as the second time domain input data of the long short-term memory neural network, and the road infrastructure state at the starting moment of the second time domain is used as the second time domain output supervision data of the long short-term memory neural network, until the road infrastructure state at the Yth time domain starting moment and the Yth time domain binary flow record information are used as the Yth time domain input data of the long short-term memory neural network, and the road infrastructure state at the Yth time domain end moment is used as the Yth time domain output supervision data of the long short-term memory neural network, to train the first road infrastructure state prediction model.

6. The predictive maintenance method based on road infrastructure monitoring according to claim 1, characterized in that: The second road infrastructure status prediction model building step comprises: Based on the road infrastructure type, first sample data is collected, wherein the first sample data includes annual temperature time series record information, annual humidity time series record information, annual precipitation time series record information and second road infrastructure status time series record information; The annual mean vehicle flow of the first sample data is less than or equal to the vehicle flow threshold, and the annual mean passenger flow of the first sample data is less than or equal to the passenger flow threshold; The second road infrastructure status prediction model is trained based on the annual temperature time series record information, the annual humidity time series record information, the annual precipitation time series record information and the second road infrastructure status time series record information.

7. The predictive maintenance method based on road infrastructure monitoring according to claim 1, characterized in that: The road infrastructure status fusion model construction step includes: Based on the road infrastructure type, second sample data is collected, wherein the second sample data includes time series record information of vehicle flow in the second sample week, time series record information of passenger flow in the second sample week, time series record information of temperature in the second sample year, time series record information of humidity in the second sample year, time series record information of precipitation in the second sample year, and second sample road infrastructure status information; Processing the second sample week vehicle flow time series record information and the second sample week passenger flow time series record information according to the first road infrastructure state prediction model to obtain first road infrastructure state identification data; Processing the second sample week passenger flow time series record information, the second sample year temperature time series record information, the second sample year humidity time series record information, and the second sample year precipitation time series record information according to the second road infrastructure state prediction model to obtain second road infrastructure state identification data; The road infrastructure state fusion model is trained using the second sample road infrastructure state data as supervision and the first road infrastructure state identification data and the second road infrastructure state identification data as input.

8. A predictive maintenance system based on road infrastructure monitoring, characterized in that: The system is used to implement the predictive maintenance method based on road infrastructure monitoring according to any one of claims 1 to 7, and the system comprises: The vehicle and pedestrian flow information collection module is used to collect weekly vehicle flow time series information and weekly pedestrian flow time series information; Meteorological data collection module, used to collect annual temperature time series information, annual humidity time series information and annual precipitation time series information; A first state prediction result obtaining module is used for activating a first road infrastructure state prediction model whenever a first time count satisfies a first period, processing the weekly vehicle flow time series information and the weekly passenger flow time series information, obtaining a first road infrastructure state prediction result, and recounting the first time count from 0, wherein the first period is one week; A second state prediction result obtaining module is used to activate the second road infrastructure state prediction model whenever the second timing satisfies the second period, process the annual temperature time series information, the annual humidity time series information and the annual precipitation time series information, obtain the second road infrastructure state prediction result, and count the second period from 0, wherein the second period is one year; A state prediction result fusion module, used to fuse the first road infrastructure state prediction result and the second road infrastructure state prediction result through a road infrastructure state fusion model to obtain a road infrastructure state fusion result; The pre-maintenance configuration execution module is used to extract the abnormal state type and abnormal triggering time information of the road infrastructure state fusion result and execute the pre-maintenance configuration.

Images