Municipal road whole life cycle health management method

Abstract

The invention belongs to the technical field of road management, and provides a municipal road full-life-cycle health management method, which comprises the following steps of: acquiring structural strain, temperature, image and BIM data by using a distributed optical fiber, a temperature and humidity sensor and an unmanned aerial vehicle, fusing multi-source data according to time and space, and constructing a data set; recognizing the type and degree of a disease by using CNN, predicting the development of the disease by using LSTM, and evaluating the remaining bearing capacity and decommissioning state of a road in combination with full-life-cycle data; a fuzzy evaluation method is combined to obtain a road health state, and an optimal maintenance scheme is formulated according to a rule base and budget; when the monitoring index exceeds a safety threshold value, early warning is triggered and a management end is notified; by fusing multi-source data and combining an intelligent algorithm, structural potential safety hazards are recognized in advance, the disease development trend is predicted, passive maintenance is converted into active prevention, and full-life-cycle data closed-loop management is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of road management, and specifically relates to a full life cycle health management method for municipal roads. Background Art

[0002] Municipal roads are an important part of urban infrastructure, and their health status is directly related to urban traffic efficiency and public travel safety. With the acceleration of urbanization and the continuous increase in traffic load, problems such as aging road structures and frequent road diseases are becoming increasingly prominent, placing higher demands on the health management of roads throughout their life cycle.

[0003] However, traditional technologies rely on manual inspections and single-point sensors, resulting in low data collection efficiency, limited coverage, and the risk of missed detections. Manual inspections are limited by the inspector's experience and subjective judgment, making it difficult to accurately quantify the extent of damage. Single-point sensors can only obtain local data and cannot construct a comprehensive health profile of the road. Furthermore, independent data sources lack unified data standards and fusion mechanisms, forming data silos. This makes it difficult to effectively trace and comprehensively utilize data from the entire road lifecycle.

[0004] Furthermore, existing technologies are primarily focused on post-event repairs, lacking the ability to provide early warning and proactive prevention and control of road hazards. The inability to monitor road structural conditions in real time makes it difficult to detect structural safety hazards in advance, such as subgrade settlement and fatigue damage to the pavement structure. This leads to frequent accidents such as collapses and potholes. When accidents occur, emergency response mechanisms lack intelligence and coordination, resulting in delayed information transmission and low response efficiency, seriously impacting urban traffic order and the safety of public life and property.

[0005] In addition, existing road health management technologies are insufficient in predicting the development trends of diseases, cannot accurately grasp the evolution laws of diseases, and are difficult to formulate scientific and reasonable preventive maintenance strategies, making it difficult to effectively guarantee the safety and reliability of municipal roads.

[0006] To this end, technicians in this field have proposed a full life cycle health management method for municipal roads, aiming to achieve closed-loop data management throughout the entire life cycle. By integrating multi-source data and combining intelligent algorithms, structural safety hazards can be identified in advance, the development trend of diseases can be predicted, and passive maintenance can be transformed into active prevention. Summary of the Invention

[0007] In order to solve the above technical problems, the present invention provides a municipal road full life cycle health management method to solve the problems raised in the background technology.

[0008] A municipal road life cycle health management method includes the following steps:

[0009] S1. Real-time collection of structural strain data and temperature field data through distributed optical fiber and temperature and humidity sensors. Inspections are carried out by drones to obtain inspection image data. BIM data is obtained based on municipal road design parameters.

[0010] S2. Combining the structural strain data, temperature field data, inspection image data, and BIM data into multi-source data, and fusing the multi-source data according to time and space dimensions to form a multi-source data set;

[0011] S3. Using a CNN algorithm to identify the type and severity of the disease in the multi-source data set to obtain a disease identification result;

[0012] S4. Analyze the multi-source data set through the LSTM model to predict the disease development trend and obtain a disease prediction result;

[0013] S5. Integrate the full life cycle data to evaluate the remaining carrying capacity and functional suitability of the road, determine whether it has entered the retirement stage, and obtain the life cycle assessment results;

[0014] S6. Based on a fuzzy evaluation method, the disease identification result and the disease prediction result are combined with the life cycle assessment result to generate a road health assessment result;

[0015] S7. Based on the road health assessment results, match maintenance strategies from a rule base, combine maintenance budgets with traffic flow, generate an optimal maintenance plan, and implement maintenance measures;

[0016] S8. According to the road health assessment result, when a preset safety threshold is exceeded, an early warning signal is issued and sent to a terminal corresponding to the management personnel.

[0017] Preferably, the use of a CNN algorithm to identify the type and severity of diseases in the multi-source dataset to obtain a disease identification result includes:

[0018] Based on the inspection image data in the multi-source data set, a sliding window operation is performed on the inspection image data through a multi-layer convolution kernel to extract spatial features of different scales and obtain a feature map;

[0019] By reducing the dimension of the feature map through average pooling, the computational complexity is reduced and translation invariance is improved to obtain a deep feature map;

[0020] The deep feature map is flattened into a one-dimensional vector, mapped to the disease type classification space through a fully connected layer, and the probability of each disease category is output through the Softmax function:

[0021]

[0022] Among them, x is the input feature vector, z kis the k-th category score output by the fully connected layer, K is the total number of disease categories, p(c k |x) is the category c under the given input x k probability;

[0023] Based on the probability of each disease category, combined with target detection, the location and range of the disease are determined, the degree of the disease is quantified, and the disease identification result is obtained.

[0024] Preferably, the method of analyzing multi-source data sets through an LSTM model to predict disease development trends and obtain disease prediction results includes:

[0025] Aligning the multi-source data sets by timestamps to form a multidimensional time series;

[0026] Splitting the multidimensional time series into an input sequence and a target sequence through a sliding window;

[0027] Multi-layer LSTM is used to capture dependencies at different time scales. The input of the upper LSTM layer is the hidden state of the lower layer. The following formula is used to map the hidden state of the last LSTM layer to the prediction target through the fully connected layer:

[0028]

[0029] Among them, h T is the hidden state of the last layer of LSTM, W h is the weight matrix of the fully connected layer, used to map the hidden state to the prediction space, b h is the bias vector used to adjust the offset of the mapping, is the predicted value, i.e. the disease prediction result.

[0030] Preferably, the comprehensive life cycle data is used to evaluate the remaining carrying capacity and functional applicability of the road, determine whether it has entered the retirement stage, and obtain the life cycle assessment results, including:

[0031] The full life cycle data includes design life, actual service life and structural damage accumulation;

[0032] The structural damage accumulation D is a structural damage index calculated using the multi-source data set, and has a range of [0, 1], where 0 represents no damage and 1 represents complete failure;

[0033] The residual load capacity factor is expressed using the following formula: R load =1-α·D; where α is the sensitivity coefficient; when D=0, R load =1, indicating that the carrying capacity is not degraded; when D=1, R load =1-α;

[0034] The fuzzy comprehensive evaluation method is used to quantitatively evaluate the functional applicability, and the evaluation index set U={u1,u2,...,u n}; and determine the weight of each indicator W={ω1,ω2,...,ω n};

[0035] For each indicator u i , determine its membership degree μ based on the comprehensive life cycle data ij , forming a fuzzy membership matrix m is the number of levels;

[0036] The functional suitability index R is calculated using the following formula: func :

[0037] R func =W·R·C T

[0038] Where C is the rating vector;

[0039] A dual-threshold decision model is used to determine whether to enter the retirement stage, taking into account the remaining carrying capacity and functional applicability:

[0040]

[0041] Among them, β is used to avoid premature retirement due to accidental damage, T d is the design life, T use is the actual service life, T load is the carrying capacity threshold, T func Functional applicability threshold, A is the result of life cycle assessment.

[0042] Preferably, the fuzzy evaluation method is used to combine the disease identification result and the disease prediction result with the life cycle assessment result to generate a road health assessment result, including:

[0043] Based on the disease identification results and disease prediction results combined with the life cycle assessment results, they are converted into quantifiable evaluation indicators to obtain the disease status indicator U1 = {u 11 ,u 12 ,...,u 1m}, disease trend index U2={u 21 ,u 22 ,...,u 2n}, life cycle index U1={u 31 ,u 32 ,u 33 The life cycle indicators include residual carrying capacity, functional applicability and design life utilization rate;

[0044] The weights of the disease status indicator, disease trend indicator and life cycle indicator are determined by the hierarchical analysis method; the criterion layer weight is Q = [q1, q2, q3], where q1, q2, q3 correspond to the weights of disease status, disease trend and life cycle respectively; then the indicator layer weight Q1 = [q 11 ,q 12 ,...,q 1m ] is the weight of the disease status indicator, Q2=[q 21 ,q 22 ,...,q 2n ] is the disease trend index weight, Q3=[q 31 ,q 32 ,q 33 ] is the life cycle indicator weight;

[0045] Use the following formula to calculate the membership degree of each indicator and form a fuzzy relationship matrix:

[0046]

[0047] R2=[r 2ij ] n×k

[0048] R3=[r 3ij ] 3×k

[0049] Among them, R1 is the disease status fuzzy matrix, R2 is the disease trend fuzzy matrix, R3 is the life cycle fuzzy matrix, r ij Indicates the degree of membership of the i-th indicator to the j-th level;

[0050] Through multi-level fuzzy comprehensive evaluation, the first-level fuzzy evaluation is obtained by calculating the evaluation results of each criterion layer. The secondary fuzzy evaluation is achieved by integrating the results of each criterion layer:

[0051]

[0052] Finally, we get the fuzzy evaluation vector B=[b1,b2,...,b k ], indicating the degree to which the road health status belongs to each level; and using the maximum membership principle, select argmax j b j The corresponding level is the road health assessment result G.

[0053] Preferably, the GIS platform is based on the road health assessment result G, maintenance history data, and traffic flow information, and generates an optimal maintenance plan through a preset maintenance strategy rule base; the maintenance strategy rule base contains maintenance strategies corresponding to different disease types and degrees;

[0054] By using the preset safety threshold, based on the road health assessment results monitored in real time, when the safety threshold is reached or exceeded, an early warning signal is triggered.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] The present invention realizes the full-cycle traceability of roads "from birth to retirement" through a closed-loop data system, and integrates multi-source data to realize full-process automation, reduce manual intervention, and improve management efficiency. Combined with real-time monitoring and early warning mechanisms, it can identify structural safety hazards in advance, reduce the probability of sudden accidents such as collapse and potholes, predict the development of diseases in advance, transform passive maintenance into active prevention, and improve the safety and reliability of municipal roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flow chart of the municipal road full life cycle health management method of the present invention. DETAILED DESCRIPTION

[0058] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0059] As attached Figure 1 As shown:

[0060] Example 1: The present invention provides a method for managing the health of municipal roads throughout their life cycle, comprising the following steps:

[0061] S1. Real-time collection of structural strain data and temperature field data is achieved through distributed optical fibers and temperature and humidity sensors. Inspections are carried out through drones to obtain inspection image data. BIM data is obtained based on municipal road design parameters. By building a comprehensive perception system, distributed optical fibers monitor structural strain in real time, temperature and humidity sensors capture environmental impacts, drones obtain macroscopic disease images, and BIM data provides design benchmarks, achieving three-dimensional and automated data collection, and solving the problems of low efficiency and narrow coverage of traditional manual inspections.

[0062] S2. Structural strain data, temperature field data, inspection image data, and BIM data are combined into multi-source data, and the multi-source data are integrated according to time and space dimensions to form a multi-source data set;

[0063] S3. Use the CNN algorithm to identify the type and severity of diseases in the multi-source dataset and obtain disease recognition results. Based on the inspection image data in the multi-source dataset, a sliding window operation is performed on it through multi-layer convolution kernels to extract spatial features of different scales and obtain feature maps.

[0064] By reducing the dimension of the feature map through average pooling, the computational complexity is reduced and translation invariance is improved to obtain a deep feature map;

[0065] The deep feature map is flattened into a one-dimensional vector, mapped to the disease type classification space through a fully connected layer, and the probability of each disease category is output through the Softmax function:

[0066]

[0067] Among them, x is the input feature vector, z k is the k-th category score output by the fully connected layer, K is the total number of disease categories, p(c k |x) is the category c under the given input x k probability;

[0068] Based on the probability of each disease category and combined with target detection, the location and scope of the disease are determined, the degree of disease is quantified, and the disease identification result is obtained.

[0069] Convolutional neural networks are used to automatically identify defects such as cracks and potholes. Accurate diagnosis is achieved through feature extraction and classifier training. The type of defect is effectively identified and the degree of damage is quantified, thus replacing manual visual interpretation. The processing time for a single image is shortened from 30 minutes to 2 seconds, increasing efficiency by 900 times.

[0070] S4. Analyze multi-source data sets through the LSTM model to predict disease development trends and obtain disease prediction results; align multi-source data sets by timestamps to form a multi-dimensional time series;

[0071] Split the multidimensional time series into input sequence and target sequence through sliding window;

[0072] Multi-layer LSTM is used to capture dependencies at different time scales. The input of the upper LSTM layer is the hidden state of the lower layer. The following formula is used to map the hidden state of the last LSTM layer to the prediction target through the fully connected layer:

[0073]

[0074] Among them, h T is the hidden state of the last layer of LSTM, W h is the weight matrix of the fully connected layer, used to map the hidden state to the prediction space, b h is the bias vector used to adjust the offset of the mapping, The predicted value, i.e., the disease prediction result, is obtained by analyzing historical multi-source data through a long-short-term memory network to establish a disease evolution model and predict development trends over the next 3-12 months. This allows for early identification of accelerated disease development points, providing a scientific basis for allocating maintenance resources.

[0075] S5. Integrate full life cycle data to assess the remaining load-bearing capacity and functional suitability of the road, determine whether it has entered the retirement stage, and obtain the life cycle assessment results. Full life cycle data includes design life, actual service life, and accumulated structural damage.

[0076] Structural damage accumulation D is a structural damage index calculated from multi-source data sets, ranging from [0, 1], where 0 indicates no damage and 1 indicates complete failure;

[0077] The residual load capacity factor is expressed using the following formula: R load =1-α·D; where α is the sensitivity coefficient; when D=0, R load =1, indicating that the carrying capacity is not degraded; when D=1, R load =1-α;

[0078] The fuzzy comprehensive evaluation method is used to quantitatively evaluate the functional applicability, and the evaluation index set U={u1,u2,...,u n}; and determine the weight of each indicator W={ω1,ω2,...,ω n};

[0079] For each indicator u i , determine its membership degree μ based on comprehensive life cycle data ij , forming a fuzzy membership matrix m is the number of levels;

[0080] The functional suitability index R is calculated using the following formula: func :

[0081] R func =W·R·C T

[0082] Where C is the rating vector;

[0083] A dual-threshold decision model is used to determine whether to enter the retirement stage, taking into account the remaining carrying capacity and functional applicability:

[0084]

[0085] Among them, β is used to avoid premature retirement due to accidental damage, T d is the design life, T use is the actual service life, T load is the carrying capacity threshold, T func Functional applicability threshold, A is the result of life cycle assessment.

[0086] By evaluating the remaining carrying capacity and functional suitability of roads, determining whether they need to be retired and rebuilt, and scientifically defining the service life of roads, we can avoid waste of resources caused by excessive maintenance or premature retirement and effectively conduct life cycle assessments.

[0087] S6. Based on the fuzzy evaluation method, the disease identification results and disease prediction results are combined with the life cycle assessment results to generate the road health assessment results;

[0088] Based on the results of disease identification and disease prediction combined with the life cycle assessment results, they are converted into quantifiable evaluation indicators to obtain the disease status index U1={u 11 ,u 12 ,...,u 1m}, disease trend index U2={u 21 ,u 22 ,...,u 2n}, life cycle index U1={u 31 ,u 32 ,u 33 Life cycle indicators include residual load capacity, functional suitability and design life utilization;

[0089] The weights of disease status index, disease trend index and life cycle index are determined by analytic hierarchy process. The weight of criterion layer is Q = [q1, q2, q3], where q1, q2, q3 correspond to the weights of disease status, disease trend and life cycle respectively. Then the weight of indicator layer is Q1 = [q 11 ,q 12 ,...,q 1m ] is the weight of the disease status indicator, Q2=[q 21 ,q 22 ,...,q 2n ] is the disease trend index weight, Q3=[q 31 ,q 32 ,q 33 ] is the life cycle indicator weight;

[0090] Use the following formula to calculate the membership degree of each indicator and form a fuzzy relationship matrix:

[0091]

[0092] R2=[r 2ij ] n×k

[0093] R3=[r 3ij ] 3×k

[0094] Among them, R1 is the disease status fuzzy matrix, R2 is the disease trend fuzzy matrix, R3 is the life cycle fuzzy matrix, rij Indicates the degree of membership of the i-th indicator to the j-th level;

[0095] Through multi-level fuzzy comprehensive evaluation, the first-level fuzzy evaluation is obtained by calculating the evaluation results of each criterion layer. The secondary fuzzy evaluation is achieved by integrating the results of each criterion layer:

[0096]

[0097] Finally, we get the fuzzy evaluation vector B=[b1,b2,...,b k ], indicating the degree to which the road health status belongs to each level; and using the maximum membership principle, select argmax j b j The corresponding level is the road health assessment result G.

[0098] The fuzzy evaluation method is used to integrate the results of disease identification, prediction and life cycle assessment to quantify the road health status, solve the uncertainty problem of multi-index evaluation, and provide an intuitive health score, so that managers can quickly understand the overall condition of the road and give priority to high-risk sections.

[0099] S7. Based on the road health assessment results, maintenance strategies are matched from the rule base. The optimal maintenance plan is generated based on the maintenance budget and traffic flow, and maintenance measures are implemented. Based on the GIS platform, the optimal maintenance plan is generated using a pre-set maintenance strategy rule base, combining the road health assessment results (G), historical maintenance data, and traffic flow information. The maintenance strategy rule base contains maintenance strategies corresponding to different disease types and severity levels.

[0100] According to the health assessment results, maintenance strategies are matched from the rule base, and the maintenance sequence and process are optimized in combination with the maintenance budget and traffic flow, so as to realize intelligent matching and dynamic optimization of maintenance plans and reduce the impact of maintenance on traffic.

[0101] S8. Based on the road health assessment results, when the preset safety threshold is exceeded, an early warning signal is issued and sent to the corresponding terminal of the management personnel. When the health assessment results exceed the preset safety threshold, an early warning is automatically triggered and a message is sent to the management personnel terminal, shortening risk response time, improving emergency management capabilities, and ensuring the safety of road users.

[0102] As can be seen from the above, through multi-source data fusion, sensors + drones + BIM, integrated spatiotemporal monitoring is carried out to cover the entire structural layer; CNN-based automatic recognition supports simultaneous detection of multiple disease types, and LSTM time series analysis combines multiple factors such as temperature, humidity, and traffic load to reduce prediction errors; at the same time, fuzzy evaluation method is used to fuse multi-dimensional data to output a comprehensive health score to support full life cycle assessment; thus, a closed-loop system for municipal road health management is constructed, which significantly improves management efficiency, maintenance accuracy, and safety prevention and control capabilities.

[0103] Example 2: This example is basically the same as the previous example, except that: in step S7, the maintenance budget and traffic flow are combined to generate an optimal maintenance plan. The optimal maintenance plan includes phased implementation and nighttime construction to reduce traffic impact.

[0104] Maintenance equipment is managed through the Internet of Things, and equipment such as pavers and rollers are deployed to the work area. After maintenance is completed, the maintenance results are reviewed through sensors and inspections, maintenance history data is updated, and subsequent decision-making models are optimized.

[0105] When the monitoring data exceeds the safety threshold, an early warning is automatically triggered and the management department is notified via SMS or APP. According to the preset plan, traffic control, temporary repairs and other measures are initiated, and the distribution of emergency resources is visualized through the GIS platform to dynamically adjust the disposal plan.

[0106] It is important to note that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, it will be readily understood by those who consult this disclosure that many modifications are possible without departing substantially from the novel teachings and advantages of the subject matter described in this application. Other replacements, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to specific embodiments, but extends to a variety of modifications still falling within the scope of the appended claims.

[0107] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A municipal road full life cycle health management method, characterized in that: The following steps are involved: S1. Real-time collection of structural strain data and temperature field data through distributed optical fiber and temperature and humidity sensors. Inspections are carried out by drones to obtain inspection image data. BIM data is obtained based on municipal road design parameters. S2. Combining the structural strain data, temperature field data, inspection image data, and BIM data into multi-source data, and fusing the multi-source data according to time and space dimensions to form a multi-source data set; S3. Using a CNN algorithm to identify the type and severity of the disease in the multi-source data set to obtain a disease identification result; S4. Analyze the multi-source data set through the LSTM model to predict the disease development trend and obtain a disease prediction result; S5. Integrate the full life cycle data to evaluate the remaining carrying capacity and functional suitability of the road, determine whether it has entered the retirement stage, and obtain the life cycle assessment results; S6. Based on a fuzzy evaluation method, the disease identification result and the disease prediction result are combined with the life cycle assessment result to generate a road health assessment result; S7. Based on the road health assessment results, match maintenance strategies from a rule base, combine maintenance budgets with traffic flow, generate an optimal maintenance plan, and implement maintenance measures; S8. According to the road health assessment result, when a preset safety threshold is exceeded, an early warning signal is issued and sent to a terminal corresponding to the management personnel.

2. A municipal road life cycle health management method as claimed in claim 1, characterized in that: The method of using the CNN algorithm to identify the type and severity of diseases in the multi-source dataset to obtain disease identification results includes: Based on the inspection image data in the multi-source data set, a sliding window operation is performed on the inspection image data through a multi-layer convolution kernel to extract spatial features of different scales and obtain a feature map; By reducing the dimension of the feature map through average pooling, the computational complexity is reduced and translation invariance is improved to obtain a deep feature map; The deep feature map is flattened into a one-dimensional vector, mapped to the disease type classification space through a fully connected layer, and the probability of each disease category is output through the Softmax function: Among them, x is the input feature vector, z k is the k-th category score output by the fully connected layer, K is the total number of disease categories, p(c k |x) is the category c under the given input x k probability; Based on the probability of each disease category, combined with target detection, the location and range of the disease are determined, the degree of the disease is quantified, and the disease identification result is obtained.

3. A municipal road life cycle health management method as claimed in claim 1, characterized in that: The LSTM model is used to analyze multi-source data sets, predict disease development trends, and obtain disease prediction results, including: Aligning the multi-source data sets by timestamps to form a multidimensional time series; Splitting the multidimensional time series into an input sequence and a target sequence through a sliding window; Multi-layer LSTM is used to capture dependencies at different time scales. The input of the upper LSTM layer is the hidden state of the lower layer. The following formula is used to map the hidden state of the last LSTM layer to the prediction target through the fully connected layer: Among them, h T is the hidden state of the last layer of LSTM, W h is the weight matrix of the fully connected layer, used to map the hidden state to the prediction space, b h is the bias vector, used to adjust the offset of the mapping, is the predicted value, i.e. the disease prediction result.

4. A municipal road life cycle health management method as claimed in claim 1, characterized in that: The comprehensive life cycle data is used to evaluate the remaining carrying capacity and functional suitability of the road, determine whether it has entered the retirement stage, and obtain life cycle assessment results, including: The full life cycle data includes design life, actual service life and structural damage accumulation; The structural damage accumulation D is a structural damage index calculated using the multi-source data set, and has a range of [0, 1], where 0 represents no damage and 1 represents complete failure; The residual load capacity factor is expressed using the following formula: R load =1-α·D; where α is the sensitivity coefficient; when D=0, R load =1, indicating that the carrying capacity is not degraded; when D=1, R load =1-α; The fuzzy comprehensive evaluation method is used to quantitatively evaluate the functional applicability, and the evaluation index set U={u1,u2,...,u n }; and determine the weight of each indicator W={ω1,ω2,...,ω n }; For each indicator u i , determine its membership degree μ based on the comprehensive life cycle data ij , forming a fuzzy membership matrix m is the number of levels; The functional suitability index R is calculated using the following formula: func : R func =W·R·C T Where C is the rating vector; A dual-threshold decision model is used to determine whether to enter the retirement stage, taking into account the remaining carrying capacity and functional applicability: Among them, β is used to avoid premature retirement due to accidental damage, T d is the design life, T use is the actual service life, T load is the carrying capacity threshold, T func Functional applicability threshold, A is the result of life cycle assessment.

5. A municipal road life cycle health management method as claimed in claim 1, characterized in that: The fuzzy evaluation method is based on combining the disease identification results and disease prediction results with the life cycle assessment results to generate a road health assessment result, including: Based on the disease identification results and disease prediction results combined with the life cycle assessment results, they are converted into quantifiable evaluation indicators to obtain the disease status index U1 = {u 11 ,u 12 ,...,u 1m }, disease trend index U2={u 21 ,u 22 ,...,u 2n }, life cycle index U1={u 31 ,u 32 ,u 33 The life cycle indicators include residual carrying capacity, functional applicability and design life utilization rate; The weights of the disease status indicator, disease trend indicator and life cycle indicator are determined by the hierarchical analysis method; the criterion layer weight is Q = [q1, q2, q3], where q1, q2, q3 correspond to the weights of disease status, disease trend and life cycle respectively; then the indicator layer weight Q1 = [q 11 ,q 12 ,...,q 1m ] is the weight of the disease status indicator, Q2=[q 21 ,q 22 ,...,q 2n ] is the disease trend index weight, Q3=[q 31 ,q 32 ,q 33 ] is the life cycle indicator weight; Use the following formula to calculate the membership degree of each indicator and form a fuzzy relationship matrix: Among them, R1 is the disease status fuzzy matrix, R2 is the disease trend fuzzy matrix, R3 is the life cycle fuzzy matrix, r ij Indicates the degree of membership of the i-th indicator to the j-th level; Through multi-level fuzzy comprehensive evaluation, the first-level fuzzy evaluation is obtained by calculating the evaluation results of each criterion layer. The secondary fuzzy evaluation is achieved by integrating the results of each criterion layer: Finally, we get the fuzzy evaluation vector B=[b1,b2,...,b k ], indicating the degree to which the road health status belongs to each level; and using the maximum membership principle, select argmax j b j The corresponding level is the road health assessment result G.

6. A municipal road life cycle health management method as claimed in claim 1, characterized in that: Based on the GIS platform, the optimal maintenance plan is generated by combining the road health assessment result G, maintenance history data, and traffic flow information through a preset maintenance strategy rule base; the maintenance strategy rule base contains maintenance strategies corresponding to different disease types and degrees; By using the preset safety threshold, based on the road health assessment results monitored in real time, when the safety threshold is reached or exceeded, an early warning signal is triggered.

Images