Abstract: First, the significance of safety monitoring of modern long-span bridges is introduced. By comparing the advantages and disadvantages of several current displacement instruments, a new method for monitoring the displacement of bridges using GPS is proposed, and its principles and characteristics are described. The time and frequency domain methods of system identification are inadequate. The new method of parameter identification combined with time is discussed. Finally, the difficulty of the current damage detection method for long-span bridges is mentioned.
Keywords: Long-span bridge GPS system identification damage detection
1. The significance of bridge beam safety monitoring
With the advancement of science and technology and the needs of transportation, many long-span bridges have emerged at the historic moment. Especially, the suspension bridge is favored by people because of its large span, beautiful shape, and saving materials. It has become the first choice for long-span bridges. But with the increase of the span, from a few hundred meters to 3000m; the high-span ratio of the stiffening beam is getting smaller and smaller ((l / 40 ～ l / 300); the safety factor also drops, from the previous 4 ～ 5 It is 2 ～ 3. In addition, due to its high flexibility and low frequency, it is very sensitive to the effects of wind. Due to the lack of necessary monitoring and corresponding maintenance, a large number of bridge damage accidents have occurred around the world, causing huge losses to the national economy and life and property.
In October 1994, there was a 50m central break of the Shengshui Bridge across the Han River in Seoul, South Korea, of which 15m fell into the river, causing a major accident that killed 32 people and seriously injured 17 people. It is said that the cause of the sudden fracture of the bridge during the peak driving period is due to long-term overload operation and fatigue damage of steel beam bolts and rods.
The Tacoma Narrows bridge with a main span of 853m, completed in 1940, was only used for three months and caused a bridge collapse accident at a wind speed of 19m / s: In 1951, the Golden Gate Bridge with a main span of 1280m at a wind speed of 15 ~ Part of the bridge body is damaged due to vibration at 1520m / s, etc.
Of the approximately 500,000 existing highway bridges in the United States, more than 200,000 have different degrees of damage. In February 1967, the Silver Bridge across the Ohio River in the United States suddenly collapsed, killing 46 people.
The cable-stayed bridges built early in our country were severely corroded due to unreasonable cable protection. For example, the cable-stayed cables of the Yellow River Bridge in Jinan and the Haiyin Bridge in Guangzhou were forced to reach their design lifespan. All replacements have caused great economic losses and adverse social impacts.
In the past ten years, China has built a number of long-span bridges. In Shanghai alone, there are bridges with world-class standards such as Nanpu, Yangpu, and Xupu Bridges. In addition, the Tsing Ma Bridge in Hong Kong and the Humen Bridge in Humen are the first in China. The suspension bridges established in China have developed rapidly in recent years, especially in coastal areas, and there is an urgent need to build a large number of large-span bridges. In order to ensure the safety and durability of these huge bridges, which are closely related to national economy and people's livelihood, these bridges must be continuously monitored.
At present, the monitoring of bridges is getting more and more attention. Many researchers are devoting themselves to bridge monitoring research. Bridge safety monitoring is increasingly becoming a very active research direction in civil engineering disciplines [1,2,3].
2. The current status of bridge displacement monitoring instruments
Long-span bridges are greatly affected by wind loads, vehicle loads, temperature and earthquakes, but generally no earthquakes in coastal areas, mainly affected by typhoons, vehicle loads and temperatures. In order to ensure their safe operation under the above conditions, the bridges must be studied in the above The actual displacement curve under conditions, and the current research on wind is limited to theoretical and model experiments, the research on the real bridge under the action of wind is not enough, and the research on the vehicle is only carried out under a specific time and space. The main reason is that the test equipment is unreasonable, and the bridge cannot be continuously monitored in real time. At present, the instruments used for structural monitoring mainly include: theodolite, displacement sensor, acceleration sensor and laser test method.
The Yangpu Bridge in Shanghai uses the total station automatic scanning method, which continuously scans each measuring point for 7s a week. The disadvantage is that the measuring points are not synchronized and cannot be measured when there is large deformation.
Displacement sensor is a contact sensor, which must be in contact with the measuring point. Its disadvantage is that it is difficult to measure the difficult-to-access point and difficult to measure the lateral displacement.
The accelerometer has poor identification effect for low-frequency static displacement. It must be integrated twice in order to obtain displacement. The accuracy is not high and it cannot be real-time. The frequency of large suspension bridges is generally low.
The laser method has a high test accuracy, but it cannot be measured because the light spot cannot be captured when the bridge shakes greatly.
In addition to the above shortcomings, the torsion angle test of the bridge is also inadequate. In order to monitor the safety of the bridge, it is necessary to find a better test method. At present, there is a new method of using GPS for testing, field testing on the high-rise structure of the bridge [4 ~ 6], Guo Jingjun and the Imperial Building in Shenzhen in 1996, and the Qingma Bridge in Hong Kong in 1998, experimental research, especially It was a real bridge test at the Humen Bridge in Guangzhou in 1999, and it is now working normally. Overseas dodson, AH, 1997; brown, GJ, 1999 also used GPS to monitor the structure, and achieved success, but there is no precedent for using GPS to test bridges in China, and it is also limited to displacement monitoring abroad, using GPS for power It is not sufficient to analyze and study the mechanical behavior of bridges under the action of wind and vehicles. The following describes the principles and characteristics of using GPS monitoring.
GPS displacement monitoring principle: The bridge displacement monitoring system uses a satellite positioning system. It uses the real-time phase difference (RTK) (Real Time Kinematic) of the carrier phase of the receiving navigation satellite to measure the displacement of the bridge in real time. The principle is shown in Figure 1.
GPS RTK differential system is composed of GPS reference station, GPS monitoring station and communication system. The reference station transmits the received satellite differential information to the monitoring station in real time through the optical fiber. The monitoring station receives the satellite signal and the GPS reference station information, and the real-time difference can measure the three-dimensional space coordinates of the station in real time. This result will be sent to the GPS monitoring center. The monitoring center calculates the displacement and rotation angle of the bridge deck and pylon on the GPS differential signal of the receiver, and provides the bridge management department with a safety analysis.
GPS monitoring bridge displacement characteristics:
(L) Because GPS is the positioning of receiving satellites, as long as each point on the bridge can receive GPS differential signals from more than 6 GPS satellites and reference stations, GPS RTK differential positioning can be performed. There is no need to look through each monitoring station, they are independent observations.
(2) GPS positioning is less affected by the outside atmosphere and can be monitored during storms.
(3) High degree of automation in GPS measurement of displacement. From receiving signals, capturing satellites, to completing RTK differential displacement, the instrument can automatically complete. The measured 3D coordinates can be automatically stored in the monitoring center server for bridge safety analysis.
(4) GPS positioning speed is fast and high precision. GPS RTK can output positioning results at the fastest rate of 10 ～ 20Hi, the positioning accuracy plane is 10mm, and the elevation is 20mm.
Of course, the real-time monitoring of bridges by GPS also has shortcomings. Currently, only relatively large deformation displacements can be monitored. For small displacements, the positioning accuracy of GPS needs to be further improved, but the application prospect of GPS to other large structures is not excluded.
3. Theoretical research status of bridge overhead monitoring
Traditional inspection methods can monitor the appearance of the bridge and certain structural characteristics. The test results can also partially reflect the current state of the structure, but it is difficult to fully reflect the health status of the bridge, especially to make a systematic assessment of the bridge's safety reserves and degradation pathways. In addition, conventional detection techniques are difficult to detect the damage of hidden components. At present, one of the most promising methods that is generally accepted is the experimental modal analysis method that combines system identification, vibration theory, vibration testing technology, signal acquisition and analysis and other interdisciplinary technologies.
Two methods are commonly used in system parameter identification: frequency domain method and time domain method. The frequency domain method uses the applied excitation and the resulting response to obtain the frequency response function through FFT analysis, and then obtains the modal parameters using methods such as polynomial fitting. Since multiple averaging can be used to eliminate random errors on the frequency response function The accuracy of the frequency domain identification method is guaranteed, but the method has the following disadvantages: â‘ Based on the vibration mode is not coupled, therefore, only the real mode of the structure with classic damping can be identified. Structures like long-span suspension bridges have obvious non-classical damping properties. The application of frequency domain method is limited. â‘¡ FFT analysis is required, which brings the influence of skewness errors such as leakage on parameter identification. The recent environmental pulsation method can obtain the vibration mode parameters without knowing the excitation, and expands the application range of the method [7, 8]. The time-domain recognition method that appeared in the late 1970s made up for the shortcomings of the frequency-domain method. Random or free response data can be used to identify modal parameters. They do not need to perform FFT analysis, thus eliminating the error caused by FFT analysis. In particular, it can also obtain random decrement characteristics from unknown random excitation response signals, so this method becomes the only method that can identify the system based on online signals. But there are also some defects: because all the information of the measured signal is used in the parameter identification, instead of intercepting the effective frequency band, the number of modals contained in the signal is relatively large, but due to the experimental test link and other reasons, it makes Some of the modal information has not been collected enough, so that the incomplete information can only be regarded as noise, the current methods to eliminate noise mainly include extended recognition and least squares. At present, the on-line monitoring of bridges using the ITD method has achieved certain results [9,10] In summary, both the time-domain method and the frequency-domain method have their own shortcomings, and a comprehensive time-frequency method should be sought to improve the recognition accuracy. The emerging wavelet transform can synthesize time-frequency, and its application in bridge parameter identification can be discussed. In terms of structural damage detection and positioning, it can currently be divided into two types: model correction method and fingerprint analysis method.
1. Accurate finite element modeling is an important prerequisite for response prediction of large-scale bridges and earthquakes; it is also the basis for structural safety monitoring, damage detection, and optimal vibration control. However, despite the limited development that cannot be achieved, the finite element model of the actual complex structure still has errors. Finite element modeling provides a complete set of theoretical modal parameters for structural flight, but these parameters are often inconsistent with those obtained from structural modal experiments. Therefore, the structural theoretical model must be adjusted or revised so that the modified modal parameters are consistent with the experiment. This process is the finite element model revision.
The model correction method is mainly used in bridge monitoring to comprehensively compare the vibration response record of the experimental structure with the original model calculation results, using directly or indirectly measured modal parameters, acceleration time history records, frequency response functions, etc. Optimize the constraints and constantly modify the stiffness and mass information in the model to obtain information on structural changes and achieve structural damage identification and positioning. The main methods are:
(1) The matrix method is the earliest and most mature method to modify the entire matrix of the calculation model. It has the characteristics of high accuracy and easy execution. The main disadvantage is that the physical meaning of the modified model is ambiguous and lost. The banding characteristics of the original finite element model should be represented by Berman / Baruch's optimal method.
(2) Sub-matrix correction method, which defines the correction coefficient by the word matrix or unit matrix to be corrected, and corrects the structural stiffness by adjusting the correction coefficient of the U matrix Symmetry and sparsity.
(3) Sensitivity method to modify structural parameters The finite element model is modified by modifying the structural design parameter elastic modulus E and cross-sectional area A of the structure.
The first two methods described above correct the stiffness and mass matrix by solving a matrix equation or a constrained minimization problem, and assume that the changes in stiffness and mass are independent of each other. Therefore, this kind of method is not suitable for the modification of the finite element model related to the structural stiffness matrix and mass matrix changes. The change in mass of long-span bridges usually defies the change in structural rigidity, which is a typical nonlinear problem. Only the third method uses the sensitivity of observations to structural parameters to modify the structural parameters. The parameter correction based on sensitivity analysis can see the influence degree of each parameter on the structural vibration from the intermediate results of the sensitivity analysis; and, it can directly explain the modification of the physical quantity of the structure, without the need to reflect the modification by using the comparison of the general matrix However, when there are many parameters to be corrected, this method often results in parameter corrections that violate the physical meaning.
2. Fingerprint analysis method, looking for dynamic fingerprints related to the dynamic characteristics of the structure, and judging the true state of the structure through the changes of these fingerprints.
In online monitoring, frequency is the most easily obtained modal parameter, and the accuracy is very high, so it is easiest to identify whether structural damage occurs by monitoring the change in frequency. In addition, the vibration mode can also be used to find structural damage. Although the test accuracy of the vibration mode is lower than the frequency, the vibration mode contains more damage information. There are many ways to determine whether structural damage has occurred using the vibration shape; MAC, COMAC, CMS, DI, and compliance matrix methods.
However, a large number of models and actual structural experiments show that the natural frequency change caused by structural damage is very small, and the vibration mode changes obviously [11,12]. Generally, the damage makes the structural natural frequency change within 5% [11,12] However, Askegaard et al. Found after long-term observation of the bridge that even if there is no obvious change in the bridge during a year, its vibration frequency can change up to 10% [63], so it is generally believed that the natural frequency cannot be used directly As a fingerprint for bridge monitoring, although the vibration mode is sensitive to local stiffness, accurate measurement is difficult. MAC, COMAC, CMS and other dynamic fingerprints that depend on the vibration mode all encounter the same problem. There is no unified and effective index for the evaluation of bridge defect status. Some people have established a variety of bridge performance evaluation expert systems based on fuzzy theory, structural reliability theory, etc., but must first establish various specifications and expert databases.
4. Conclusion and Prospect
(L) Because long-span bridges are greatly affected by environmental factors and have low safety factors, they must be continuously monitored in real time.
(2) Due to the high positioning accuracy and speed of GPS, it has obvious advantages compared with other displacement monitoring instruments. It can be used for continuous real-time monitoring of long-span bridges, and its accuracy should be further improved to expand its application range. At present, GPS has been successfully installed on the Humen Bridge, realizing continuous real-time monitoring of the bridge.
(3) In terms of system identification, the advantages and disadvantages of the time domain and frequency domain methods are compared. In the future, a system identification study combining time and frequency should be conducted.
(4) In terms of model modification, a model modification method suitable for large-span bridges should be studied on the basis of sensitivity analysis.
(5) Due to the lack of unified and effective indicators for the assessment of the defect status of bridges, the fingerprint indicators suitable for long-span bridges should be studied in combination with experimental testing and finite element modeling.
Square Shape Biscui Containers
Biscuit Plastic Container,Airtight Biscuit Jar,Plastic Containers For Biscuits,Small Biscuit Container
Shantou Oudaya Packaging Co. Ltd. , https://www.imloudaya.com