Intelligent maintenance decision-making method and system based on bridge big data

Abstract

The invention provides an intelligent maintenance decision-making method and system based on bridge big data, and belongs to the technical field of bridge safety. The intelligent maintenance decision-making method comprises the steps of 1, collecting basic information, a detection report, a maintenance record and real-time state data of a bridge; 2, key features influencing the bridge performance are mined from the data collected in the step 1, the current state of the bridge is analyzed and evaluated, and the change trend of the bridge condition in a period of time in the future is predicted; 3, generating a plurality of possible maintenance strategy schemes based on the result of the step 2; step 4, carrying out cost benefit analysis on the maintenance strategy scheme generated in the step 3; 5, a multi-objective optimization algorithm is utilized, bridge performance improvement, maintenance cost, construction time and traffic influence factors are comprehensively considered, and an optimal scheme is selected from the maintenance strategy schemes evaluated in the step 4; and 6, formulating a detailed maintenance plan according to the optimal strategy and implementing the detailed maintenance plan.

Description

Technical Field

[0001] The present application relates to the field of bridge safety technology, and specifically to an intelligent maintenance decision-making method and system based on bridge big data. Background Art

[0002] In the field of bridge maintenance, traditional maintenance decision-making methods often rely on experience and manual judgment, and have problems such as low efficiency, high cost, and insufficient accuracy. With the development of big data and artificial intelligence technology, the field of bridge maintenance has also ushered in new opportunities and challenges. As an important part of infrastructure, the safety and stability of bridges are directly related to the safety and convenience of public transportation and people's lives. Therefore, how to use modern technical means to effectively monitor, evaluate and maintain bridges has become one of the current research hotspots.

[0003] With the continuous advancement of sensor technology, drone technology, machine learning algorithms and other technologies, the ability to obtain bridge-related data has been significantly improved, and a large amount of bridge structure data, inspection data and real-time status data has been accumulated and stored. However, how to extract useful information from these massive data, discover potential problems, and make scientific maintenance decisions is still a problem that needs to be solved in current research.

[0004] Therefore, it is necessary to propose an intelligent maintenance decision-making method and system based on bridge big data. By integrating big data technology, machine learning algorithms and multi-objective optimization technology, comprehensive monitoring and evaluation of bridge status can be achieved, providing scientific and accurate support for bridge maintenance decisions. Summary of the invention

[0005] In order to overcome a series of defects in the prior art, the purpose of this patent is to provide an intelligent maintenance decision-making method based on bridge big data for the above problems, including the following steps:

[0006] Step 1: Collect basic information, inspection reports, maintenance records, and real-time status data of the bridge;

[0007] Step 2: Extract key features that affect bridge performance from the data collected in step 1, analyze and evaluate the current state of the bridge, and predict the changing trend of the bridge condition in the future;

[0008] Step 3: Based on the results of step 2, generate multiple possible maintenance strategy solutions including maintenance time, scope, process and material usage;

[0009] Step 4, conducting a cost-benefit analysis on the maintenance strategy solution generated in step 3;

[0010] Step 5: Using a multi-objective optimization algorithm, comprehensively consider the bridge performance improvement, maintenance cost, construction time and traffic impact factors, and select the best maintenance strategy from the maintenance strategy solutions evaluated in step 4;

[0011] Step 6: Develop and implement a detailed maintenance plan based on the optimal strategy, and evaluate the implementation results to continuously optimize future maintenance decisions.

[0012] Furthermore, the basic information of the bridge includes design drawings, construction records, material information, and load parameters;

[0013] The bridge inspection report is the most recent inspection report, which is used to understand the structural health status, diseases, cracks and deformation data of the bridge;

[0014] Maintenance records include maintenance plans, materials used, and construction processes;

[0015] Real-time bridge status data includes:

[0016] Obtain high-resolution images of bridges through drone aerial photography for disease identification and deformation monitoring;

[0017] Bridge beam surface and beam bottom inspection data obtained by the cable-climbing robot;

[0018] Information about the underwater part of the bridge obtained by underwater robots;

[0019] Concrete quality, crack and cavity data obtained through sonar detection devices;

[0020] The bridge load size, vibration frequency and temperature change data are obtained through strain, vibration and temperature sensors installed at the main stress-bearing parts of the bridge.

[0021] Furthermore, the key features affecting bridge performance include the following steps:

[0022] Clean the data collected in step 1, including processing missing values, outliers, and noise data;

[0023] Standardize the cleaned data and convert data of different dimensions to the same scale;

[0024] Calculate the correlation coefficient between each characteristic and the bridge health status index;

[0025] According to the size of the correlation coefficient, the feature subset that is highly correlated with the bridge status is screened out;

[0026] The principal component analysis method is used to extract the principal components of the relevant features; the principal components that can explain most of the data variance are selected as the main factors affecting the bridge performance;

[0027] The contribution of each feature to the prediction results of the bridge performance evaluation model is judged, and the features with the greatest impact on bridge performance are selected as key features according to their importance.

[0028] Furthermore, the correlation coefficients between each feature and the bridge health status index are calculated using the Pearson correlation coefficient, and the specific formula is: Among them, x i and i are the i-th observation value of the characteristic variable and the bridge health status indicator, respectively. and are their respective average values;

[0029] The selection of the principal component is determined by the cumulative variance contribution rate, and the specific formula is:

[0030]

[0031] Among them, λ j is the eigenvalue corresponding to the jth principal component, p is the number of features in the feature subset, k is the number of principal components retained, and θ is the percentage of the total variance of the relevant feature subset that can be explained by the first k principal components selected.

[0032] Furthermore, the analysis and evaluation of the current state of the bridge and the prediction of the changing trend of the bridge condition in the future include the following steps:

[0033] Construct a three-dimensional fine finite element model to simulate the actual structure of the bridge, perform static and dynamic analysis on the three-dimensional fine finite element model, calculate stress, strain and displacement response, check the analysis results with the measured data, and update the model parameters;

[0034] Evaluate the current damage level and performance status of the bridge based on long-term structural health monitoring data;

[0035] Input the current state data of the bridge into the performance degradation model to predict the degradation trend in the future;

[0036] By integrating the results of finite element analysis, monitoring and evaluation, and degradation prediction, an uncertain knowledge reasoning decision-making model based on evidence theory is established to comprehensively evaluate the current status and future trends of the bridge.

[0037] Furthermore, the degradation trend in the future is predicted by the following formula:

[0038]

[0039] Among them, S(t) represents the structural performance state at time t; S0 is the initial state; r(S(τ), tT) is the performance degradation rate function that changes with time. The parameters of the degradation rate function are obtained by analyzing and estimating the historical performance state data, and then predicting the future performance state; t represents the current time, and τ represents the integral variable, which represents the time interval between a certain time point in the past and the current time t;

[0040] The uncertain knowledge reasoning decision model based on evidence theory is expressed by the following formula:

[0041]

[0042] Where A represents the possible state of the bridge; m(B) represents the degree of confidence that the true state of the bridge belongs to B for each possible state set B. The value of m(B) is obtained based on the results of finite element analysis, monitoring and evaluation, and degradation prediction; K represents the degree of conflict between different evidences. If all the evidences are consistent, then K = 0; if all the evidences are completely conflicting, then K = 1.

[0043] Further, step 4 includes the following steps:

[0044] Collect the direct and indirect costs of historical maintenance projects to build a cost estimation model, input maintenance engineering quantity, process and time parameters, and output the corresponding estimated cost;

[0045] Clarify the main objectives of bridge maintenance, determine the key indicators for benefit evaluation, and apply multi-criteria decision analysis to assign corresponding weights to each key indicator to reflect its relative importance in the overall decision;

[0046] Based on the three-dimensional fine finite element model and performance degradation model, the changes in bridge performance after the implementation of different maintenance strategies are simulated; the benefit indicators are quantitatively evaluated according to the simulation results, and a benefit evaluation model is constructed to output the benefit scores of each strategy under different indicators.

[0047] Further, step 5 includes the following steps:

[0048] Clearly define the key performance indicators that need to be optimized for bridge repairs, including structural safety improvement, repair costs, construction time and traffic impact;

[0049] Take maintenance time windows, construction process limitations and environmental protection requirements into consideration as practical constraints;

[0050] Construct a multi-objective optimization model to balance the key performance indicators and search for non-dominated solutions, that is, the Pareto optimal solution set that can achieve the acceptance level under all objectives;

[0051] A comprehensive evaluation is conducted on each solution in the Pareto solution set, and the most appropriate solution is selected as the optimal maintenance strategy based on the decision maker's preferences and needs.

[0052] An intelligent maintenance decision-making system based on bridge big data includes three modules: big data subsystem, bridge performance analysis subsystem and bridge maintenance decision optimization subsystem, among which:

[0053] The big data subsystem is responsible for collecting, transmitting, processing and storing various bridge-related data, providing data support for the entire system, including:

[0054] The data collection layer automatically obtains basic information, historical inspection reports and maintenance records of the bridge. At the same time, the status data of the bridge is monitored in real time through sensors installed on the bridge.

[0055] The data transmission layer is used to upload the collected data to the data processing layer;

[0056] The data processing layer cleans, converts and standardizes the collected data;

[0057] The data storage layer uses a distributed storage architecture to store processed data;

[0058] The bridge performance analysis subsystem is responsible for evaluating the current status of the bridge and predicting future trends, including:

[0059] Feature engineering module, used to extract key features affecting bridge performance, perform feature selection and construction;

[0060] A bridge performance assessment model that evaluates the current structural performance and safety status of a bridge based on a neural network;

[0061] Performance degradation prediction model, used to predict the remaining service life and the degradation trend of bridge performance in the future;

[0062] The bridge maintenance decision optimization subsystem is responsible for generating the optimal maintenance decision plan, including:

[0063] The maintenance strategy generation module generates a variety of possible maintenance strategy solutions based on the bridge performance analysis results and the maintenance solution library. The maintenance strategy includes maintenance time, maintenance scope, maintenance process and material usage.

[0064] Multi-objective optimization module, used to establish a multi-objective optimization model for maintenance decision-making;

[0065] The scheme evaluation module conducts a comprehensive evaluation of the optimized maintenance strategy scheme, including economic evaluation, feasibility evaluation and risk evaluation;

[0066] The maintenance plan feedback module records the implemented maintenance plans and maintenance results, and provides feedback data for future maintenance decisions.

[0067] Furthermore, the multi-objective optimization module uses genetic algorithms to solve multi-objective optimization problems and obtain the optimal maintenance strategy.

[0068] Compared with the prior art, the present application has at least the following technical effects or advantages.

[0069] This application proposes a maintenance intelligent decision-making method and system based on bridge big data. By collecting, analyzing and mining bridge-related data, combined with multi-criteria decision analysis and multi-objective optimization algorithms, it improves bridge maintenance efficiency, reduces maintenance costs, and improves bridge structure safety. It also establishes an intelligent maintenance decision-making system to provide decision makers with scientific and accurate maintenance decision-making plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a flow chart of an intelligent maintenance decision-making method based on bridge big data disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments.

[0072] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0073] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0074] Reference Figure 1 , an intelligent maintenance decision-making method based on bridge big data, comprising the following steps:

[0075] Step 1: Collect basic information, inspection reports, maintenance records, and real-time status data of the bridge;

[0076] Step 2: Extract key features that affect bridge performance from the data collected in step 1, analyze and evaluate the current state of the bridge, and predict the changing trend of the bridge condition in the future;

[0077] Step 3: Based on the results of step 2, generate multiple possible maintenance strategy solutions including maintenance time, scope, process and material usage;

[0078] Step 4, conducting a cost-benefit analysis on the maintenance strategy solution generated in step 3;

[0079] Step 5: Using a multi-objective optimization algorithm, comprehensively consider the bridge performance improvement, maintenance cost, construction time and traffic impact factors, and select the best maintenance strategy from the maintenance strategy solutions evaluated in step 4;

[0080] Step 6: Develop and implement a detailed maintenance plan based on the optimal strategy, and evaluate the implementation results to continuously optimize future maintenance decisions.

[0081] In step 1, by collecting bridge-related data, a comprehensive archive of the bridge can be established, including basic information, previous inspection reports, maintenance records, and real-time status data. This provides the necessary data support for subsequent analysis and decision-making, and helps to fully understand the condition and change trend of the bridge.

[0082] In step 2, by mining and analyzing bridge data, key features that affect bridge performance can be identified, the current status of the bridge can be evaluated in a timely manner, and future change trends can be predicted. This helps to identify potential problems and risks in a timely manner and provides a scientific basis for formulating effective maintenance strategies.

[0083] In step 3, based on the analysis results, a variety of possible maintenance strategy plans are formulated to help improve the flexibility and pertinence of maintenance decisions. In this way, the most appropriate maintenance plan can be selected under different circumstances to maximize the performance and service life of the bridge.

[0084] In step 4, cost-benefit analysis can be used to evaluate the economic benefits of each maintenance strategy and determine the most appropriate maintenance plan. This helps to achieve a balance between minimizing maintenance costs and maximizing bridge performance under limited resources.

[0085] In step 5, the multi-objective optimization algorithm is used to comprehensively consider the trade-offs of various factors and select the optimal maintenance plan. This ensures that while improving bridge performance, maintenance costs and impact on traffic are minimized, achieving overall optimization.

[0086] In step 6, by developing and implementing a detailed maintenance plan, the progress of maintenance work can be effectively managed and monitored. At the same time, evaluating the implementation effect will help to identify problems and improvement measures in a timely manner, continuously optimize future maintenance decisions, and ensure the safety and reliable operation of the bridge.

[0087] Furthermore, the basic information of the bridge includes design drawings, construction records, material information, and load parameters;

[0088] The bridge inspection report is the most recent inspection report, which is used to understand the structural health status, diseases, cracks and deformation data of the bridge;

[0089] Maintenance records include maintenance plans, materials used, and construction processes;

[0090] Real-time bridge status data includes:

[0091] Obtain high-resolution images of bridges through drone aerial photography for disease identification and deformation monitoring;

[0092] Bridge beam surface and beam bottom inspection data obtained by the cable-climbing robot;

[0093] Information about the underwater part of the bridge obtained by underwater robots;

[0094] Concrete quality, crack and cavity data obtained through sonar detection devices;

[0095] The bridge load size, vibration frequency and temperature change data are obtained through strain, vibration and temperature sensors installed at the main stress-bearing parts of the bridge.

[0096] As an important part of traffic arteries, monitoring and maintenance of bridge safety are crucial. In order to ensure the safe operation of bridges, it is necessary to obtain and master detailed information and real-time status data of bridges. This information can not only help understand the structural health of bridges, but also provide strong support for maintenance and reinforcement work.

[0097] First of all, the basic information of the bridge is the starting point for understanding the bridge. The design drawings record the structural design and construction requirements of the bridge in detail, providing the theoretical basis of the bridge. The construction records record the key nodes and quality control in the bridge construction process in detail, providing the historical background of the bridge construction. The material information includes the specifications and properties of the various materials used in the bridge, which is an important basis for evaluating the bearing capacity of the bridge. The load parameters reflect the bearing capacity of the bridge under different conditions and provide the safety standards for the use of the bridge.

[0098] In addition to basic information, bridge inspection reports are also an important way to understand the status of bridges. The latest inspection reports can provide the latest data on the health of bridge structures, including diseases, cracks and deformations. These data are an important basis for evaluating bridge safety and are also the basis for formulating maintenance and reinforcement plans.

[0099] Maintenance records are an important part of bridge maintenance. The maintenance plan records in detail the maintenance methods and steps for different diseases and damages, the material usage records the various materials and quantities used in the maintenance process, and the construction process describes in detail the implementation process of the maintenance work. These records provide comprehensive information on bridge maintenance work, which helps to summarize experience and lessons and improve the efficiency and quality of maintenance work.

[0100] Drone aerial photography technology can provide us with high-resolution images of bridges to help detect diseases and deformations. Climbing robots can perform inspections on the surface and bottom of bridge beams to obtain more detailed data. Underwater robots can provide us with information about the underwater part of the bridge, including the conditions of pile foundations, piers and other parts. Sonar detection devices can detect concrete quality, cracks and cavities, and provide more accurate information about the bridge structure. In addition, strain, vibration and temperature sensors installed at the main stress-bearing parts of the bridge can monitor the load size, vibration frequency and temperature change data of the bridge in real time, providing important information on the real-time status of the bridge.

[0101] In summary, by obtaining and mastering the basic information, inspection reports, maintenance records and real-time status data of bridges, we can fully understand the structural health status and safety performance of bridges. This information not only helps to timely discover and deal with bridge diseases and damages, but also provides strong support for maintenance and reinforcement work. In future work, we should continue to strengthen bridge information management and maintenance work to ensure the safe operation of bridges and provide more reliable and convenient transportation conditions for people's travel.

[0102] Furthermore, the key features affecting bridge performance include the following steps:

[0103] Clean the data collected in step 1, including processing missing values, outliers, and noise data;

[0104] Standardize the cleaned data and convert data of different dimensions to the same scale;

[0105] Calculate the correlation coefficient between each characteristic and the bridge health status index;

[0106] According to the size of the correlation coefficient, the feature subset that is highly correlated with the bridge status is screened out;

[0107] The principal component analysis method is used to extract the principal components of the relevant features; the principal components that can explain most of the data variance are selected as the main factors affecting the bridge performance;

[0108] The contribution of each feature to the prediction results of the bridge performance evaluation model is judged, and the features with the greatest impact on bridge performance are selected as key features according to their importance.

[0109] In order to explore the key characteristics of bridge performance in depth, a series of rigorous steps need to be followed. These steps not only involve data collection and processing, but also include data analysis, screening and evaluation to ensure that we can accurately identify the main factors affecting bridge performance.

[0110] First, the collected data is cleaned. This step is crucial because it is directly related to the accuracy and reliability of subsequent analysis. In the data cleaning process, missing values, outliers, and noisy data need to be processed to ensure the quality and consistency of the data. This can be achieved by filling missing values, removing outliers, or using smoothing techniques to remove noisy data.

[0111] Next, the cleaned data needs to be standardized. Since the dimensions and value ranges of different features may be different, direct analysis may lead to bias. Therefore, it is necessary to convert these data to the same scale for fair comparison and analysis. Standardization can be done through normalization, standardization or other appropriate methods.

[0112] After data cleaning and standardization, the correlation coefficients between each feature and the bridge health indicator can be calculated. The purpose of this step is to understand the degree of correlation between the features and bridge performance. By calculating the correlation coefficient, the impact of each feature on bridge performance can be quantified and provide a basis for subsequent feature screening.

[0113] According to the size of the correlation coefficient, feature subsets that are highly correlated with the bridge status can be screened out. These feature subsets represent the main factors affecting bridge performance and are crucial for subsequent analysis and modeling. The screening process can be carried out using methods such as threshold method and correlation sorting.

[0114] In order to further simplify the feature set and extract the main information, principal component analysis is used to extract the principal components of the relevant features. PCA is a commonly used dimensionality reduction technique that can convert multiple relevant features into a few unrelated principal components. By selecting principal components that can explain most of the data variance, a complex feature set can be simplified into a more representative feature subset.

[0115] Finally, it is necessary to judge the contribution of each feature to the prediction results of the bridge performance evaluation model. This can be achieved by calculating the importance score of the feature and analyzing the prediction performance of the model. According to the importance, the features that have the greatest impact on bridge performance can be selected as key features. These key features will provide strong support for our bridge performance evaluation and model building.

[0116] In summary, mining the key features that affect bridge performance is a complex and rigorous process. Through steps such as data cleaning, standardization, feature screening, principal component extraction, and feature evaluation, we can gradually gain a deeper understanding of the factors affecting bridge performance and provide a scientific basis for the safe operation and maintenance of bridges.

[0117] Furthermore, the correlation coefficients between each feature and the bridge health status index are calculated using the Pearson correlation coefficient, and the specific formula is: Among them, x i and i are the i-th observation value of the characteristic variable and the bridge health status indicator, respectively. and are their respective average values;

[0118] The selection of the principal component is determined by the cumulative variance contribution rate, and the specific formula is:

[0119]

[0120] Among them, λ j is the eigenvalue corresponding to the jth principal component, p is the number of features in the feature subset, k is the number of principal components retained, and θ is the percentage of the total variance of the relevant feature subset that can be explained by the first k principal components selected.

[0121] When discussing the assessment of bridge health, it is necessary not only to consider the overall condition of the bridge, but also to deeply analyze the correlation between each characteristic and the bridge health indicator. This correlation can be quantified by mathematical statistical methods, among which the most commonly used is the Pearson correlation coefficient. The Pearson correlation coefficient is an indicator that measures the strength and direction of the linear relationship between two variables, and its value range is between -1 and 1. When the correlation coefficient is 1, it means that the two variables are completely positively correlated; when the correlation coefficient is -1, it means that the two variables are completely negatively correlated; when the correlation coefficient is 0, it means that there is no linear relationship between the two variables.

[0122] In bridge health assessment, it is assumed that there is a set of characteristic variables, such as the material, age, and traffic volume of the bridge, as well as a bridge health indicator, such as the bridge's load-bearing capacity. In order to quantify the degree of correlation between these characteristics and the bridge health indicator, the Pearson correlation coefficient can be used for calculation. The calculated correlation coefficient value can be used to understand the strength and direction of the correlation between each characteristic and the bridge health indicator.

[0123] In addition, in order to better understand and analyze these characteristic variables, principal component analysis can also be used for dimensionality reduction. Principal component analysis is a commonly used statistical method that can transform a set of variables that may be correlated into a set of linearly uncorrelated variables, namely principal components, through orthogonal transformation. These principal components can retain most of the information of the original data while reducing the dimension of the variables and simplifying the data analysis process. The selection of principal components is usually determined by the cumulative variance contribution rate. By setting a suitable θ value, the principal component that best represents the original data can be selected, thereby achieving effective dimensionality reduction of the characteristic variables and accurate assessment of the health status of the bridge.

[0124] In summary, through the Pearson correlation coefficient and principal component analysis method, we can deeply understand the correlation between bridge characteristics and health status indicators, and achieve a comprehensive assessment of bridge health status. This not only helps to timely discover problems with bridges, but also provides a scientific basis for bridge maintenance and management, ensuring the safe operation of bridges.

[0125] Furthermore, the analysis and evaluation of the current state of the bridge and the prediction of the changing trend of the bridge condition in the future include the following steps:

[0126] Construct a three-dimensional fine finite element model to simulate the actual structure of the bridge, perform static and dynamic analysis on the three-dimensional fine finite element model, calculate stress, strain and displacement response, check the analysis results with the measured data, and update the model parameters;

[0127] Evaluate the current damage level and performance status of the bridge based on long-term structural health monitoring data;

[0128] Input the current state data of the bridge into the performance degradation model to predict the degradation trend in the future period;

[0129] By integrating the results of finite element analysis, monitoring and evaluation, and degradation prediction, an uncertain knowledge reasoning decision-making model based on evidence theory is established to comprehensively evaluate the current status and future trends of the bridge.

[0130] A series of scientific and systematic steps are required to analyze and evaluate the current state of the bridge. Among them, building a three-dimensional fine finite element model is a crucial first step. By using advanced modeling software and technology, the actual structure of the bridge can be accurately simulated, providing a solid foundation for subsequent analysis and evaluation. On this basis, static and dynamic analysis of the model can calculate the stress, strain and displacement response of the bridge under various working conditions, so as to fully understand the stress state and deformation of the bridge.

[0131] In order to verify the accuracy of the model, the analysis results need to be checked against the measured data. This process can be done by comparing the model calculation results with the measured data collected by the sensors installed on the bridge. If a large difference is found, the model parameters need to be adjusted to improve the accuracy of the model. In this way, a finite element model that accurately reflects the current state of the bridge can be obtained.

[0132] Next, the current damage level and performance status of the bridge can be evaluated based on long-term structural health monitoring data. Structural health monitoring is a method of obtaining information on the structural status of a bridge by real-time monitoring of the bridge. By analyzing the monitoring data, we can understand the stress, deformation and damage development of the bridge at different times and under different working conditions, so as to comprehensively evaluate the current status of the bridge.

[0133] After obtaining the current state data of the bridge, these data can be input into the performance degradation model to predict the degradation trend in the future. The performance degradation model is a model based on historical data and statistical methods, which can predict the performance degradation of the bridge in the future. By inputting the current state data of the bridge, the predicted values ​​of the stress, strain and displacement response of the bridge in the future can be obtained, so as to understand the future development trend of the bridge.

[0134] Finally, in order to comprehensively evaluate the current status and future trends of the bridge, it is necessary to integrate the results of finite element analysis, monitoring and evaluation, and degradation prediction. This can be achieved by establishing an uncertain knowledge reasoning decision model based on evidence theory. This model can comprehensively consider the uncertainty of various information sources, and comprehensively process the results of finite element analysis, monitoring and evaluation, and degradation prediction, so as to obtain a comprehensive and objective evaluation result.

[0135] In summary, a series of scientific and systematic steps are needed to analyze and evaluate the current status of bridges and predict future trends. By constructing a three-dimensional fine finite element model, conducting static and dynamic analysis, long-term monitoring data evaluation, performance degradation prediction, and establishing an uncertain knowledge reasoning decision model, we can fully understand the current status and future development trends of bridges and provide strong support for bridge maintenance and management. At the same time, the implementation of these steps also requires full use of modern scientific and technological means and methods to continuously improve the accuracy and efficiency of the evaluation.

[0136] Furthermore, the degradation trend in the future is predicted by the following formula:

[0137]

[0138] Among them, S(t) represents the structural performance state at time t; S0 is the initial state; r(S(τ), t-τ) is the performance degradation rate function that changes with time. The parameters of the degradation rate function are obtained by analyzing and estimating the historical performance state data, and then predicting the future performance state; t represents the current time, and τ represents the integral variable, which represents the time interval between a certain time point in the past and the current time t;

[0139] The uncertain knowledge reasoning decision model based on evidence theory is expressed by the following formula:

[0140]

[0141] Where A represents the possible state of the bridge; m(B) represents the degree of confidence that the true state of the bridge belongs to B for each possible state set B. The value of m(B) is obtained based on the results of finite element analysis, monitoring and evaluation, and degradation prediction; K represents the degree of conflict between different evidences. If all the evidences are consistent, then K = 0; if all the evidences are completely conflicting, then K = 1.

[0142] How to predict the degradation trend in the future by a given formula is a problem involving the change of structural performance status over time, which is especially important in the engineering field, such as the maintenance and management of bridges and buildings.

[0143] S(t) in the formula represents the structural performance state at time t. This is usually a quantitative indicator used to describe the integrity, safety or other related performance of the structure. S0 represents the initial state, that is, the performance state of the structure when it is first used. Over time, the structural performance state will change, and this change can be described by r(S(τ), t-τ).

[0144] r(S(τ), t-τ) is a time-varying performance degradation rate function. This function describes the rate of change of the structural performance state between a certain time point τ in the past and the current time t. The specific form of this function needs to be determined based on the actual situation, and its parameters usually need to be estimated by analyzing and processing a large amount of historical performance state data. In this way, this function can be used to predict future performance states.

[0145] In addition, in order to more accurately predict the future performance status, some other factors need to be considered, such as environmental factors, usage conditions, maintenance conditions, etc. These factors may have an impact on the performance status of the structure, so they need to be properly considered in the model.

[0146] Next, the uncertain knowledge reasoning decision model based on evidence theory is used to deal with uncertain knowledge, which is particularly suitable for situations involving multiple evidences and multiple possibilities. In the condition assessment of bridges, due to the existence of multiple uncertain factors, such as errors in monitoring data and imperfections in models, this model is very applicable.

[0147] In the model, A represents the possible states of the bridge. These states can be safe, dangerous, need maintenance, etc. m(B) represents the degree of confidence that the true state of the bridge belongs to B for each possible state set B. This confidence can be obtained through a variety of methods, such as finite element analysis, monitoring and evaluation, degradation prediction, etc.

[0148] K represents the degree of conflict between different pieces of evidence. If all the evidence is consistent, then K = 0; if all the evidence is completely conflicting, then K = 1. In practical applications, the value of K needs to be determined according to the specific situation to reflect the degree of conflict between different pieces of evidence.

[0149] In summary, the degradation trend in the future period can be predicted by the given formula, and the uncertainty knowledge can be processed by the uncertain knowledge reasoning decision model based on evidence theory, so as to more accurately evaluate the status of the bridge.

[0150] Further, step 4 includes the following steps:

[0151] Collect the direct and indirect costs of historical maintenance projects to build a cost estimation model, input maintenance engineering quantity, process and time parameters, and output the corresponding estimated cost;

[0152] Clarify the main objectives of bridge maintenance, determine the key indicators for benefit evaluation, and apply multi-criteria decision analysis to assign corresponding weights to each key indicator to reflect its relative importance in the overall decision;

[0153] Based on the three-dimensional fine finite element model and performance degradation model, the changes in bridge performance after the implementation of different maintenance strategies are simulated; the benefit indicators are quantitatively evaluated according to the simulation results, and a benefit evaluation model is constructed to output the benefit scores of each strategy under different indicators.

[0154] To further advance the cost-benefit analysis of bridge repair projects, the following key steps need to be implemented in Step 4:

[0155] First, it is necessary to collect detailed direct and indirect cost data of historical maintenance projects. This data will be used to build an accurate cost estimation model. By inputting maintenance engineering volume, process and time parameters, the model can output the corresponding estimated cost. This can not only more accurately predict future maintenance costs, but also provide a strong basis for formulating budgets and cost control strategies.

[0156] Secondly, the main goals of bridge maintenance need to be clarified. These goals include improving bridge safety, extending service life, improving traffic flow, etc. On this basis, key indicators for benefit evaluation need to be determined, such as cost-effectiveness ratio, return on investment, social benefits, etc. In order to more comprehensively reflect the relative importance of these indicators in the overall decision-making, multi-criteria decision analysis can be applied to assign corresponding weights to each key indicator.

[0157] Next, based on the three-dimensional fine finite element model and the performance degradation model, the performance changes of the bridge after the implementation of different maintenance strategies will be simulated. These simulations will take into account various factors, such as material properties, environmental factors, traffic flow, etc. By comparing the simulation results under different maintenance strategies, the benefit indicators can be quantitatively evaluated, thereby constructing a benefit evaluation model. This model will output the benefit scores of each strategy under different indicators to provide decision support for selecting the appropriate maintenance strategy.

[0158] In summary, by collecting historical cost data, clarifying maintenance goals, determining key benefit indicators, applying multi-criteria decision analysis and building a benefit evaluation model, we can more comprehensively analyze the cost-effectiveness of bridge maintenance projects. This will help to formulate more scientific and reasonable maintenance strategies and improve the overall benefits of maintenance projects.

[0159] Further, step 5 includes the following steps:

[0160] Clearly define the key performance indicators that need to be optimized for bridge repairs, including structural safety improvement, repair costs, construction time and traffic impact;

[0161] Take maintenance time windows, construction process limitations and environmental protection requirements into consideration as practical constraints;

[0162] Construct a multi-objective optimization model to balance the key performance indicators and search for non-dominated solutions, that is, the Pareto optimal solution set that can achieve the acceptance level under all objectives;

[0163] A comprehensive evaluation is conducted on each solution in the Pareto solution set, and the most appropriate solution is selected as the optimal maintenance strategy based on the decision maker's preferences and needs.

[0164] The safety, economy and efficiency of bridges are of vital importance to the entire society. In order to ensure the long-term stable operation of bridges, maintenance work is indispensable. During the maintenance process, how to formulate a scientific and reasonable maintenance strategy to ensure that the safety of bridges is improved, the maintenance cost is reduced, the construction time is shortened, and the impact on traffic is reduced is a key issue facing the current bridge maintenance field.

[0165] It is necessary to clearly define the key performance indicators that need to be optimized for bridge maintenance. These indicators include but are not limited to improving structural safety, reducing maintenance costs, shortening construction time, and minimizing traffic impact. Among them, improving structural safety is the core goal of maintenance work, ensuring that the bridge can withstand the invasion of various natural and human factors after maintenance and protect the safety of life and property. Reducing maintenance costs is related to economic benefits. How to reduce maintenance costs while ensuring quality is an issue that every decision maker needs to consider. Shortening construction time can reduce the impact on traffic and ensure smooth traffic.

[0166] When formulating maintenance strategies, it is also necessary to consider maintenance time windows, construction process limitations, and environmental protection requirements as practical constraints. The selection of maintenance time windows needs to take into account factors such as traffic flow and weather conditions to ensure that maintenance work can be carried out at the most appropriate time. Construction process limitations require us to consider existing construction technology and equipment when formulating maintenance plans to ensure the feasibility of the plan. At the same time, environmental protection is also an important part that cannot be ignored, and it is necessary to ensure that maintenance work will not have a negative impact on the environment.

[0167] On this basis, a multi-objective optimization model can be constructed to weigh various key performance indicators and search for non-dominated solutions, that is, the Pareto optimal solution set that can achieve an acceptable level under all objectives. This model can help us find the best balance between multiple objectives and ensure the comprehensive optimization of the maintenance strategy.

[0168] Finally, a comprehensive evaluation is conducted on each solution in the Pareto solution set, and the most appropriate solution is selected as the optimal maintenance strategy based on the decision maker's preferences and needs. This step requires decision makers to comprehensively consider various factors, including technical feasibility, economic benefits, environmental impact, etc., to ensure that the maintenance strategy finally selected is the most in line with the actual situation and needs.

[0169] In summary, the multi-objective optimization of bridge maintenance strategy is a complex and critical problem. By clearly defining key performance indicators, considering practical constraints, building a multi-objective optimization model, and conducting comprehensive evaluation, we can develop a scientific and reasonable maintenance strategy to ensure that the safety, economy, and efficiency of bridges are comprehensively improved.

[0170] An intelligent maintenance decision-making system based on bridge big data includes three modules: big data subsystem, bridge performance analysis subsystem and bridge maintenance decision optimization subsystem, among which:

[0171] The big data subsystem is responsible for collecting, transmitting, processing and storing various bridge-related data, providing data support for the entire system, including:

[0172] The data collection layer automatically obtains basic information, historical inspection reports and maintenance records of the bridge. At the same time, the status data of the bridge is monitored in real time through sensors installed on the bridge.

[0173] The data transmission layer is used to upload the collected data to the data processing layer;

[0174] The data processing layer cleans, converts and standardizes the collected data;

[0175] The data storage layer uses a distributed storage architecture to store processed data;

[0176] The bridge performance analysis subsystem is responsible for evaluating the current status of the bridge and predicting future trends, including:

[0177] Feature engineering module, used to extract key features affecting bridge performance, perform feature selection and construction;

[0178] A bridge performance assessment model that evaluates the current structural performance and safety status of a bridge based on a neural network;

[0179] Performance degradation prediction model, used to predict the remaining service life and the degradation trend of bridge performance in the future;

[0180] The bridge maintenance decision optimization subsystem is responsible for generating the optimal maintenance decision plan, including:

[0181] The maintenance strategy generation module generates a variety of possible maintenance strategy solutions based on the bridge performance analysis results and the maintenance solution library. The maintenance strategy includes maintenance time, maintenance scope, maintenance process and material usage.

[0182] Multi-objective optimization module, used to establish a multi-objective optimization model for maintenance decision-making;

[0183] The scheme evaluation module conducts a comprehensive evaluation of the optimized maintenance strategy scheme, including economic evaluation, feasibility evaluation and risk evaluation;

[0184] The maintenance plan feedback module records the implemented maintenance plans and maintenance results, and provides feedback data for future maintenance decisions.

[0185] In today's society, as the scale of bridge construction continues to expand and the operation time of bridges continues to extend, bridge maintenance has become a vital issue. Traditional bridge maintenance methods often rely on manual inspections and empirical judgments, which are not only inefficient, but also difficult to accurately evaluate bridge performance and predict future degradation trends. In order to solve this problem, an intelligent maintenance decision system based on bridge big data came into being.

[0186] This intelligent maintenance decision-making system consists of three core modules: big data subsystem, bridge performance analysis subsystem and bridge maintenance decision optimization subsystem. Each module has its own unique functions and roles, and together they constitute the complete framework of this intelligent maintenance decision-making system.

[0187] First of all, the big data subsystem is the cornerstone of the entire system. It is responsible for collecting, transmitting, processing and storing various types of data related to the bridge, providing data support for the entire system. The data collection layer achieves comprehensive coverage of bridge data by automatically acquiring basic information, historical inspection reports and maintenance records of the bridge, and monitoring the status data of the bridge in real time through sensors installed on the bridge. The data transmission layer is responsible for uploading the collected data to the data processing layer to ensure the timeliness and accuracy of the data. The data processing layer cleans, converts and standardizes the collected data to eliminate noise and outliers in the data and improve the quality and availability of the data. Finally, the data storage layer adopts a distributed storage architecture to store the processed data, ensuring the security and scalability of the data.

[0188] Secondly, the bridge performance analysis subsystem is responsible for evaluating the current state of the bridge and predicting future trends. The feature engineering module extracts key features that affect bridge performance, performs feature selection and construction, and provides effective data support for subsequent evaluation and prediction. The bridge performance evaluation model evaluates the current structural performance and safety status of the bridge based on a neural network, which can accurately reflect the actual condition of the bridge. The performance degradation prediction model is used to predict the remaining service life and the degradation trend of bridge performance in the future, providing a scientific basis for maintenance decisions.

[0189] Finally, the bridge maintenance decision optimization subsystem is responsible for generating the optimal maintenance decision plan. The maintenance strategy generation module generates a variety of possible maintenance strategy plans based on the bridge performance analysis results and the maintenance plan library. These plans include maintenance time, maintenance scope, maintenance process and material consumption, providing decision makers with a comprehensive selection space. The multi-objective optimization module is used to establish a multi-objective optimization model for maintenance decisions, comprehensively considering multiple factors such as economy, feasibility and risk, and finding the optimal maintenance strategy plan. The plan evaluation module conducts a comprehensive evaluation of the optimized maintenance strategy plan, including economic evaluation, feasibility evaluation and risk evaluation, to ensure the feasibility and effectiveness of the plan. The maintenance plan feedback module records the implemented maintenance plan and maintenance effect, provides feedback data for future maintenance decisions, and continuously improves and optimizes the maintenance decision plan.

[0190] In short, this intelligent maintenance decision system based on bridge big data realizes comprehensive monitoring and performance evaluation of bridge status through the application of big data technology, providing scientific basis and strong support for maintenance decisions. At the same time, the system also has the ability of self-learning and optimization, and can continuously adapt to changes and challenges in the operation of bridges, providing strong guarantees for the safe operation of bridges. In the future, with the continuous advancement of technology and the continuous expansion of its application scope, this intelligent maintenance decision system will play a more important role in the field of bridge maintenance.

[0191] Furthermore, the multi-objective optimization module uses genetic algorithms to solve multi-objective optimization problems and obtain the optimal maintenance strategy.

[0192] Genetic algorithm is an optimization algorithm that simulates the biological evolution process in nature. It searches for the optimal solution in the solution space by simulating the selection, crossover and mutation mechanisms in the biological evolution process. In multi-objective optimization problems, genetic algorithm can consider multiple objective functions at the same time, such as maintenance cost, equipment availability and production line stability, so as to find a set of solutions with the best comprehensive performance, namely Pareto optimal solution.

[0193] In the multi-objective optimization module, the specific implementation process of the genetic algorithm includes the following steps: first, define the appropriate objective function and constraints according to the characteristics of the problem; then, randomly generate a set of initial solutions as the initial population; next, evaluate the quality of each solution by calculating its fitness value; then, select excellent individuals according to the fitness value to perform crossover and mutation operations to generate a new population; finally, repeat the above process until the termination condition is met.

[0194] Through the application of genetic algorithms, the multi-objective optimization module can find a set of solutions with the best comprehensive performance in the multi-objective maintenance strategy problem. Therefore, the application of the multi-objective optimization module has important practical significance for improving maintenance efficiency, reducing operating costs and enhancing competitiveness.

Claims

1. A maintenance intelligent decision-making method based on bridge big data, characterized in that: The following steps are involved: Step 1: Collect basic information, inspection reports, maintenance records, and real-time status data of the bridge; Step 2: Extract key features that affect bridge performance from the data collected in step 1, analyze and evaluate the current state of the bridge, and predict the changing trend of the bridge condition in the future; Step 3: Based on the results of step 2, generate multiple possible maintenance strategy solutions including maintenance time, scope, process and material usage; Step 4, conducting a cost-benefit analysis on the maintenance strategy solution generated in step 3; Step 5: Using a multi-objective optimization algorithm, comprehensively consider the bridge performance improvement, maintenance cost, construction time and traffic impact factors, and select the best maintenance strategy from the maintenance strategy solutions evaluated in step 4; Step 6: Develop and implement a detailed maintenance plan based on the optimal strategy, and evaluate the implementation results to continuously optimize future maintenance decisions.

2. According to claim 1, a maintenance intelligent decision-making method based on bridge big data is characterized in that: The basic information of the bridge includes design drawings, construction records, material information and load parameters; The bridge inspection report is the most recent inspection report, which is used to understand the structural health status, diseases, cracks and deformation data of the bridge; Maintenance records include maintenance plans, materials used, and construction processes; Real-time bridge status data includes: Obtain high-resolution images of bridges through drone aerial photography for disease identification and deformation monitoring; Bridge beam surface and beam bottom inspection data obtained by the cable-climbing robot; Information about the underwater part of the bridge obtained by underwater robots; Concrete quality, crack and cavity data obtained through sonar detection devices; The bridge load size, vibration frequency and temperature change data are obtained through strain, vibration and temperature sensors installed at the main stress-bearing parts of the bridge.

3. The intelligent maintenance decision-making method based on bridge big data according to claim 1 is characterized in that: Uncovering the key features that affect bridge performance involves the following steps: Clean the data collected in step 1, including processing missing values, outliers, and noise data; Standardize the cleaned data and convert data of different dimensions to the same scale; Calculate the correlation coefficient between each characteristic and the bridge health status index; According to the size of the correlation coefficient, the feature subset that is highly correlated with the bridge status is screened out; The principal component analysis method is used to extract the principal components of the relevant features; the principal components that can explain most of the data variance are selected as the main factors affecting the bridge performance; The contribution of each feature to the prediction results of the bridge performance evaluation model is judged, and the features with the greatest impact on bridge performance are selected as key features according to their importance.

4. The intelligent maintenance decision-making method based on bridge big data according to claim 3 is characterized in that: The correlation coefficient between each characteristic and the bridge health status index is calculated by the Pearson correlation coefficient. The specific formula is: Among them, x i and i are the i-th observation value of the characteristic variable and the bridge health status indicator, respectively. and are their respective average values; The selection of the principal component is determined by the cumulative variance contribution rate, and the specific formula is: Among them, λ j is the eigenvalue corresponding to the jth principal component, p is the number of features in the feature subset, k is the number of principal components retained, and θ is the percentage of the total variance of the relevant feature subset that can be explained by the first k principal components selected.

5. The intelligent maintenance decision-making method based on bridge big data according to claim 1 is characterized in that: The analysis and evaluation of the current status of the bridge and the prediction of the changing trend of the bridge status in the future period include the following steps: Construct a three-dimensional fine finite element model to simulate the actual structure of the bridge, perform static and dynamic analysis on the three-dimensional fine finite element model, calculate stress, strain and displacement response, check the analysis results with the measured data, and update the model parameters; Evaluate the current damage level and performance status of the bridge based on long-term structural health monitoring data; Input the current state data of the bridge into the performance degradation model to predict the degradation trend in the future; By integrating the results of finite element analysis, monitoring and evaluation, and degradation prediction, an uncertain knowledge reasoning decision-making model based on evidence theory is established to comprehensively evaluate the current status and future trends of the bridge.

6. The intelligent maintenance decision-making method based on bridge big data according to claim 5 is characterized in that: The degradation trend in the future is predicted by the following formula: Among them, S(t) represents the structural performance state at time t; S0 is the initial state; r(S(τ), t-τ) is the performance degradation rate function that changes with time. The parameters of the degradation rate function are obtained by analyzing and estimating the historical performance state data, and then predicting the future performance state; t represents the current time, and τ represents the integral variable, which represents the time interval between a certain time point in the past and the current time t; The uncertain knowledge reasoning decision model based on evidence theory is expressed by the following formula: Where A represents the possible state of the bridge; m(B) represents the degree of confidence that the true state of the bridge belongs to B for each possible state set B. The value of m(B) is obtained based on the results of finite element analysis, monitoring and evaluation, and degradation prediction; K represents the degree of conflict between different evidences. If all the evidences are consistent, then K = 0; if all the evidences are completely conflicting, then K = 1.

7. The intelligent maintenance decision-making method based on bridge big data according to claim 1 is characterized in that: Step 4 includes the following steps: Collect the direct and indirect costs of historical maintenance projects to build a cost estimation model, input maintenance engineering quantity, process and time parameters, and output the corresponding estimated cost; Clarify the main objectives of bridge maintenance, determine the key indicators for benefit evaluation, and apply multi-criteria decision analysis to assign corresponding weights to each key indicator to reflect its relative importance in the overall decision; Based on the three-dimensional fine finite element model and performance degradation model, the changes in bridge performance after the implementation of different maintenance strategies are simulated; the benefit indicators are quantitatively evaluated according to the simulation results, and a benefit evaluation model is constructed to output the benefit scores of each strategy under different indicators.

8. The intelligent maintenance decision-making method based on bridge big data according to claim 1 is characterized in that: Step 5 includes the following steps: Clearly define the key performance indicators that need to be optimized for bridge repairs, including structural safety improvement, repair costs, construction time and traffic impact; Take maintenance time windows, construction process limitations and environmental protection requirements into consideration as practical constraints; Construct a multi-objective optimization model to balance the key performance indicators and search for non-dominated solutions, that is, the Pareto optimal solution set that can achieve the acceptance level under all objectives; A comprehensive evaluation is conducted on each solution in the Pareto solution set, and the most appropriate solution is selected as the optimal maintenance strategy based on the decision maker's preferences and needs.

9. An intelligent maintenance decision system based on bridge big data, used to implement the intelligent maintenance decision method according to claim 1, characterized in that: It includes three modules: big data subsystem, bridge performance analysis subsystem and bridge maintenance decision optimization subsystem, among which: The big data subsystem is responsible for collecting, transmitting, processing and storing various bridge-related data, providing data support for the entire system, including: The data collection layer automatically obtains basic information, historical inspection reports and maintenance records of the bridge. At the same time, the status data of the bridge is monitored in real time through sensors installed on the bridge. The data transmission layer is used to upload the collected data to the data processing layer; The data processing layer cleans, converts and standardizes the collected data; The data storage layer uses a distributed storage architecture to store processed data; The bridge performance analysis subsystem is responsible for evaluating the current status of the bridge and predicting future trends, including: Feature engineering module, used to extract key features affecting bridge performance, perform feature selection and construction; A bridge performance assessment model that evaluates the current structural performance and safety status of a bridge based on a neural network; Performance degradation prediction model, used to predict the remaining service life and the degradation trend of bridge performance in the future; The bridge maintenance decision optimization subsystem is responsible for generating the optimal maintenance decision plan, including: The maintenance strategy generation module generates a variety of possible maintenance strategy solutions based on the bridge performance analysis results and the maintenance solution library. The maintenance strategy includes maintenance time, maintenance scope, maintenance process and material usage. Multi-objective optimization module, used to establish a multi-objective optimization model for maintenance decision-making; The scheme evaluation module conducts a comprehensive evaluation of the optimized maintenance strategy scheme, including economic evaluation, feasibility evaluation and risk evaluation; The maintenance plan feedback module records the implemented maintenance plans and maintenance results, and provides feedback data for future maintenance decisions.

10. The intelligent maintenance decision-making system based on bridge big data according to claim 9 is characterized in that: The multi-objective optimization module uses genetic algorithms to solve multi-objective optimization problems and obtain the optimal maintenance strategy.

Images