Study On Analysis Method Of Soft Soil Roadbed Settlement Data Based On Wavelet Technique

This dissertation applied system engineering method, started with consummating the supervision mechanism of soft-base monitoring system, carried through systemic study on scientific establishment of soft-base monitoring program. This dissertation set up a reasonable, high-efficient soft-base monitoring data supervision system to advance the reliability and time-effectiveness of scene monitoring data. In order to reduce the adverse impacts of inevitable monitoring system error and artificial factors on data analysis and to reflect the fact of roadbed settlement, on the basis of settlement monitoring sequence preprocessing, I adopted the better-developed wavelet technology in signal processing to pick up tendency information. I also introduced wavelet neural network nonlinear combination forecast method in soft base settlement forecast to furthest advance the veracity of soft-base settlement forecast. The main work of this dissertation is listed as follows:1. On the basis of analyzing the existent problems of China's soft base monitoring administration system, this dissertation brought forward the basic methods on establishing Internet-based high-grade highway soft-base monitoring information supervision system. I also studied the framework, design principles and key technology of the system. The developed network declaring system for freeway soft base monitoring was applied to One Ringroad speed main line of Foshan City in Guangdong Province, and gained good results.2. On the basis of error analysis for settlement monitoring of advanced highway soft-base, this dissertation summarized precision indicators, monitoring frequency, monitoring principles and quality guarantee measures that are fit for the settlement monitoring of advanced highway soft-base; analyzed the cause of abnormal data; pointed out that the MGM(1,n) model, a model that takes the impacts of relevant factors into consideration, should be applied into the interpolation of soft-base settlement data; discussed the revision method of the model in order to improve the precision of settlement data interpolation.3. This dissertation carried out comprehensive experiment research on wavelet denoising performance of soft-base settlement data. For assessing the wavelet denoising performance of different wavelet functions, threshold value selection methods and boundary exhibition patterns, this dissertation constructed the ideal signal for wavelet analysis based on the soft-base subside rule, and used known-noise to carry out experiment test.4. Aimed at the problems of high restriction and low forecast veracity in subside analysis of single model, I inducted combination forecast model in this dissertation. Applied the wavelet neural network combination forecast model to forecast soft-base settlement and compared the neural network combination forecast result with single models and BP neural network, in order to evaluate the wavelet neural network combination forecasting model's applicability in the field of soft-base settlement forecast.