Deflection Measurement Methods Based On The Velocities Of Pavement Deflections

Pavement Deflection is an important parameter to characterize the whole strength of subgrade and pavement on the road, which is to this day measured by FWD or Beckman Beam at home and abroad. There are obvious shortcomings of these methods.(1) Low efficiency. The measure speech just reaches to3-5km/h by interval of20m, which cannot meet the deflection measurement demand to network level.(2) Poor security. These devices, working on stop and go process model, are accident-prone while measuring on the highway especially.(3) Much influence on the traffic. it need traffic control when measuring.(4) Measuring results cannot meet the existing requirement of assessments. Static sampling mode cannot simulate pavement deflection caused by variable speed loads in normal traffic conditions, which results cannot reflect the actual dynamic driving characteristics; and meanwhile finite discrete data can not represent the whole road. Due to the disadvantages, the paper worked on high-speed deflection measurement system to measure deflection continuously at traffic speed of20-90km/h. The main contents are as follows:The paper firstly reviewed traditional measurement methods and technique way at home and abroad, and then summarized technique way of high-speed pavement deflection measurement and analyzed the advantages and disadvantages of calculation procedure using absolute displacement measurement and pavement deformation velocity, including measuring principle, processing theory and implementation, to determine technical method of pavement deformation velocity in the paper. Secondly, considering the actual condition, that is, foreign high-speed deflection measurements use inertial navigation systems (IMS) and servo mechanism to maintain measurement posture, while domestically the real condition cannot be satisfied. The paper proposed the deflection data processing model integrating dynamic angle and calibration factor, in which determine dynamic impact factor by the means of experiment, and worked out the realization method of the model.The data collection is the foundation. Without alternative and reference method and device for data collection, it is necessary to design a prototype system of deflection measurement to achieve experimental data and validation results. The paper planed to install sensor on4meter length rigid beams, in the places of100,300,750and3600millimeter away from the load wheel center. It is meet the measurement demand of small semi-rigid pavement deflection basin, and can also use to measure under the condition of bigger deflection basin, providing a guarantee the uniqueness of measurement posture. Pavement deformation speed is the smallest amount of which can be measured as we all known, then any error can lead to fails. By analyzing we learn that the angle between the fixed sensors is the biggest factor leads to errors. Research shows that show that even the error of0.005°will cause25%calculation error. The paper design a calibration method based on the angle between the fixed precision motion sensors. Calibration angle can satisfy accuracy requirement of0.0005°.The method to deal with abnormal data, which is treated as a kind of noise in the process of measurement and calculation, has a very important impact on its calculation results. Abnormal data included noise collected in the process of original measurement, such as environmental light disturb, pavement abnormalities, sensor noise, etc., and also includes noise made in the process of calculation and random noise caused by immeasurable amount, for example, noise was amplified too in the case of amplifying amplitude to meet the calculation demand, as well as in the case of subtle deformation of measuring beams. Using alternative method to handle measurable abnormal data, e.g. replaced with the average value or previous and next value, will not break the corresponding relationship between the measuring data. Using elimination strategies classifiably and step-by-step to handle immeasurable abnormal data, e.g. low-pass filtering the raw data and moving average filtering the data of moving speed, angle, sinking speed and deflection. Filtering and elimination strategies were carried out at every step, which is to provide base data for further manipulation on calculating pavement deformation speed and the deflection value.Extraction of road speed is the key to the research. After analyzing and validating of reflection of measuring system’s posture varying on measuring data, the paper proposed two methods to extract pavement deformation speed based on dynamic angle calculation and calibration equation calculation in condition of non-INS and non-servo mechanism. The method based of dynamic angle calculation can calculate measure posture of any data. To ensure accuracy, the relationship between beam’s rotation centres, IRI, corresponding wave. Considering influence of different regions, different subgrade, different pavement material and different measure environment, the paper proposed the calibration equation-based extraction method to precisely calibrate the horizontal velocity factor, the rotation speed factor and the compensation factor, avoiding the impact of measuring beams rotation center on the results. Integrating these two methods as a supplement to each other, could apply to measure deflection in different environment.Based on the proposed theories and methods, a large of measure data were used to completely verify accuracy of the theory, the stability and adaptability of the algorithm, versatility, data correlation, speed variation coefficient and suitable condition. The results showed that, measure methods and models can apply to different regions, different altitude and different pavement. Data repeatability is more than98%, while high correlation with Beckman Beam reached90%and influence on measure speed is small. Finally, the paper indicated the discussion and suggestion for the further research.