Bridge health monitoring and evaluation system

To understand the health status of the bridge, this paper explores how to use the bridge health monitoring and evaluation system for bridge health monitoring, management, and evaluation. It also expounds in detail the role of other subsystems. In this system, a sensor system can monitor the bridge in real time to obtain the basic data of the normal operation of a bridge. At the same time, the daily manual inspection work is carried out. We will receive data through the data collection, transmission, and processing subsystem for the next operation. Finally, the evaluation subsystem can use the data to carry out a series of calculations and analyses, and then the bridge health status score can be obtained.


Introduction
In the 21st century, with the continuous development of the economy, people's demand for bridge transportation conditions is also increasing. By the end of 2019, there were 878,300 highways and bridges nationwide, covering 60,634,600 meters, an increase of 26,800 and 4,948,600 meters from the previous year. Among them, there were 5,716 extra-large bridges with a total length of 10,332,300 meters, and 108,344 bridges with a total length of 29,237,500 meters [1]. People's attention to the problem of bridge safety and applicability is constantly increasing. Because the bridge has a long life cycle, large scale and the operation of the check is difficult, and it is easily affected by the weather and climate conditions, bridge's structure will inevitably produce a variety of injuries [2,3,4,5,6]. But the resources consumed by relying only on human resources to do a good job of bridge health testing are too large and restrictive.
With the continuous development of science and technology, the health monitoring technology of Bridges has also occupied an important position [7] [8]. This technology facilitates the relevant technical maintenance personnel to understand the health state of Bridges at all times and do a good job in the maintenance of Bridges. Based on the bridge health monitoring system, some daily inspection is carried out to judge the health state of the bridge, which provides a scientific basis for the maintenance and management.
Therefore, the emergence of bridge health monitoring and evaluation system is very important, which is of great significance in the application of the bridge. It can not only effectively feedback the use of the bridge in real time, but also send out early warning signals in time when abnormal conditions occur on the bridge and present specific information to the relevant technicians in order to respond in the shortest time. It provides decision-making plans to avoid a wide range of quality or safety issues and saves a lot of human and material resources.

Sensor System
Because of the complexity of the bridge and the factors that affect the health of the bridge structure are diversified, leading to different locations and types of bridge diseases. Therefore, there are more requirements for the selection of sensor types, and full consideration should be given to the various factors. The type of sensor is required to ensure the possibility to collect disease information. From the specific application of the sensor, there are some sensor types following for monitoring: (1) Humidity thermometer: used to monitor atmospheric temperature and relative humidity; temperature and humidity measurement values in various periods (hour, day, year); relationship between main beam deflection and temperature and humidity changes.
(2) Temperature sensor: the temperature time history of steel components, concrete, and air; temperature measurement and temperature difference in various periods (hour, day, year); the relationship between displacement value and temperature change.
(3) Displacement meter: the measured value of the displacement of the support and the main girder in each time period (hour, day, year); the lateral displacement and rotation angle change of the bridge; the relationship between the displacement of the expansion joint and the support and environmental factors.
(4) Inclinometer: the relative displacement between the measuring points in each time period (hour, day, year); the relationship between the displacement of the stiffening beam, the main tower and the environmental factors.
(5) Vibrating cable gauge: acceleration response time history of each measuring point; acceleration amplitude of each frequency; power spectrum; amplitude and phase difference of accelerometers of each measuring point; vertical speed, torsion speed, displacement and rotation angle Time history; peak displacement and rotation angle of characteristic points, mean and standard deviation of displacement and rotation angle in different time periods; structural dynamic characteristics, such as natural frequency, modal (and participation factor), damping ratio, modal energy conversion rate; specific Simple harmonic vibration and its corresponding natural frequency under external excitation conditions.
(6) Strain sensor: the amplitude and fluctuation of strain, stress and internal force; verification of the design value of strain, stress, force and bending moment at the local interface of the component; continuous time-history strain value of each measuring point; structural internal force, stress, strain changes with external temperature, traffic load, and other loads; predict extreme values based on statistical distribution; fatigue factor analysis based on rain flow counting.

The system of data acquisition and transmission
The system of data acquisition and transmission is mainly used for the information which is collected and transmitted by the sensor. The sensor collected all sorts of temperature, humidity, natural frequency, the structure of the displacement, stress, and strain information, and they are converted into a digital signal, through a serial port server, switch, and industrial control equipment, such as real-time data transmission, and uploaded to the monitoring platform of information management for the bridge health assessment. The design of the system focuses on the following aspects: (1) The data acquisition system can continue to work around the clock, and is not affected by severe weather.
(2) The transmission network structure must be safe and reliable, and its design must meet the corresponding technical standards.
(3) Electrical, installation specifications, machinery, and communication protocols should use relevant national specifications or compatible standards.
(4) The data acquisition and transmission system should have real-time self-diagnosis function, and be able to judge data abnormalities, sensor failures, functional abnormalities or subsystem failures.
(5) The remote control of data can be realized, and the sampling parameters can be changed (6) The system should have real-time data collection, storage and management functions.

The system of data processing
The data or discrete data collected through real-time monitoring of a single sensor can hardly reflect the true situation of the overall performance state of the bridge structure, so data processing is required to provide a basis for subsequent health assessment [18]. Data processing is divided into two parts, one is data preprocessing, and the other is data secondary processing. The former plays a role in real-time evaluation, processing the data collected by sensors through real-time monitoring, and uses mathematical statistics to calculate method to obtain the average value, highest value, lowest value, standard deviation and other data in the corresponding time. These data are used for the primary early warning comparison object and the data for reference in the final evaluation; the latter is the comparison between the monitoring data of each category further analysis [19].

Primary warning system for structural safety
The primary early warning system of the bridge structure is an indispensable subsystem in the system. The subsystem uses the model comparison method to determine whether to issue an alarm by comparing the real-time monitoring data of the sensor with the preset threshold. Relevant personnel are warned to check the bridge data to make a preliminary prediction of the health of the bridge [20]. Among them, the real-time monitoring data of the sensor used for comparison is the average value in each time period, and the predetermined threshold is the data of the unit component of the bridge within the acceptable range, and the data changes with the increase of the life of the bridge. The workflow of this subsystem is shown in the figure below.

. Evaluation method of bridge health state
For the assessment and maintenance of bridges, the safety assessment method of bridges is one of the important links [21]. The safety evaluation method is to evaluate the overall safety performance of the bridge by calculating the obtained data. It can not only determine whether the bridge is reliable, but also provide a scientific maintenance plan for bridge maintenance. The methods of safety assessment are also diverse, each one has its own advantages and disadvantages and scope of application [6]. The following are common evaluation methods.
(1) Routine comprehensive evaluation. This method uses weighted arithmetic average, weighted geometric average, and a hybrid evaluation method of comprehensive algorithm and geometric average, but it is not suitable for problems that are difficult to describe quantitatively.
(2) Fuzzy comprehensive evaluation method. This method is based on fuzzy mathematics and quantifies factors with unclear boundaries. This method introduces the concept of membership function and solves the contradiction between the ambiguity of things and the certainty of the algorithm. But the evaluators are subjective, and it is difficult to select fuzzy algorithm and determine the degree of membership.
(3) Appearance investigation and evaluation method. This is an assessment standard based on the " Highway Bridge and Culvert Maintenance Code", which requires comprehensive technical detection of the bridge, but its value depends largely on the detection experience of the evaluators themselves, and the evaluation value has great randomness.
(4) Finite element simulation of carrying capacity. It is to simplify several beams into a beam, to find out the bearing capacity of it, according to the transverse distribution coefficient of load to find out the bearing capacity of the whole bridge.
(5) Expert system evaluation method. This is the combination of expert knowledge and experience with the computer system for evaluation, which can use the knowledge and experience of experts to solve the uncertain factors in bridge design, construction and management, but at the same time, it is difficult to directly evaluate the complex bridge structure and numerous factors only by relying on the experience of experts.
(6) Fuzzy neural network method. It is a method that organically combines the mechanism of neural network with the reasoning mechanism of fuzzy logic, which can take the respective advantages of the two technologies, learn from each other and reduce the influence of human factors on the evaluation process [22].
(7) Grey relational degree evaluation method. This is based on the grey system theory, taking the time series as the object, and finally making the sorting method for each sequence.
(8) Analytic Hierarchy Process. By determining the relationship between the importance degree and the mutual influence of each bridge component, the multi-level evaluation and scoring can simplify a lot of complicated evaluation work, which is scientific, simple, and practical.

Assessment of bridge health status
The assessment of the bridge health status is based on a comprehensive calculation score based on its applicability, structural safety and other aspects, and finally the health status of the bridge and its score value are obtained. The scoring uses the analytic hierarchy process. The analytic hierarchy process is to hierarchize and adjust the various factors that affect the normal working state of the bridge, and summarize the factors that have similar effects on the state of the bridge units to form a layer, thereby establishing a multi-level comprehensive evaluation and evaluation system [18] [23].
The daily inspection data of the bridge and the structural health monitoring data were weighted and synthesized, and the overall health status assessment information of the bridge was obtained by integrating the indicators of each layer [24]. The specific calculation method is as follows: For the evaluation index of bridge health state, Hearn [25]  main components are based on the comparison between the current health status of the bridge structure and the completed structure to obtain the overall health status of the bridge, the scoring range is 0-9 points.
The overall health status score R of the bridge is obtained based on the following formula.
Among the number, i =1..., n is the element of the bridge that has an important influence on the overall health state of the bridge; i W is the weight of main components on the whole bridge; i K is the standard value of i W ; i R is a score of the worst condition on each bridge unit, which includes the analysis results of the bridge. The overall state of the bridge is better than the "initial state" Better than current norms No need to repair 8 Very good The overall state of the bridge is equivalent to the "initial state" Corresponding to current standards Without repair; List the items that need special attention during the next inspection cycle 7 Good, but it has a few small questions The overall state of the bridge is the same as the "initial state", and no defects are found That's better than the current minimum standards No immediate repair plan; Check for the possibility of further testing 6 Fine, there is a slight degradation of the structural unit The overall state of the bridge is the same as the "initial state". Although minor "damage" occurs in the secondary components, the overall performance is not affected Corresponding to the current minimum standards Repaired at the end of next quarter; Add it to the work program 5 Acceptable -All major components are in good condition, but there may be small section losses, cracks, or scour The overall state of the bridge is slightly lower than the "initial state", and no damage is found in the main components, while large damage appears in the secondary components, but it has little influence on the overall performance A little better than the acceptable minimum affluence Put in the current plan; Repair in the current quarter, and choose the first fix 4 Poor -section has a loss, degradation, peeling, or corrosion The overall state of the bridge is lower than the "initial state", and the main components are damaged, which has a great impact on the overall performance

The future research direction
At present, the bridge health monitoring and evaluation system still have many deficiencies, and some functions cannot meet the requirements of bridge construction. It needs to be further improved [26]; in order to improve the system, further research can be carried out in the following areas in the future.
(1) Optimize the layout of sensors. Because of the diversification of bridge disease information, the types and number of sensors required are also huge, but the number of sensors is also limited. How to reduce the number of sensors while achieving the purpose of complete information monitoring requires us to further think, namely How to optimize the placement of sensors, the placement of measuring points is the key.
(2) Collection of effective data. The data monitored through the sensors is undoubtedly a huge amount of various kinds. How to accurately identify the disease information we need from this information and use effective data to evaluate the integrity of the bridge? This is what we need to conduct further research.

Conclusion
The technology involved in the bridge health monitoring and evaluation system is complex and diverse. The basic data of the bridge can be obtained by real-time monitoring of the bridge through the sensor subsystem. Then, the system can realize the purpose of assessing the health status of the bridge after data transmission and processing. The development and progress of this system is of great significance to the operation and management of bridges, which means that the application of intelligent information technology has replaced traditional manual detection methods, saving manpower and material resources, and enabling resources to be reasonably configured and utilized. There are more possibilities for management, structural condition assessment and timely maintenance of bridges.
The future development and progress will depend on the joint efforts of us. We should improve the bridge health monitoring and evaluation system through continuous research.