Research On Data Processing And Prediction Of Deformation Monitoring Based On Wavelet Denoising Optimization

With the rapid development and advancement of science and technology,deformation monitoring technology is also constantly enriched and upgraded.However,most of the current operating units are still using traditional deformation monitoring methods and deformation monitoring data processing techniques.Even if the method of deformation monitoring is improved,it is mainly reflected in the improvement of monitoring equipment,and there are still many shortcomings in the data processing technology of deformation monitoring.If the rapid development of deformation monitoring methods is not supported by strong and effective data processing techniques,the effects of deformation monitoring and prediction will have great limitations,and it is difficult to accurately reflect the actual changes of the deformed objects.Therefore,while enriching and improving the deformation monitoring method,it is necessary to conduct a more in-depth,comprehensive and effective research on the deformation monitoring data processing technology.At present,the main research direction of deformation monitoring data processing technology is to use advanced and complete mathematical theory and signal processing methods to deeply analyze the nonlinear and complex deformation signals of the deformed object,and extract the deformation trend,law and amplitude of the deformed object.Furthermore,the stability analysis of the deformation of the deformed object is carried out to realize the deformation prediction and effective prevention of the deformed object.For the analysis of deformation signals of deformation objects,domestic and foreign scholars have proposed different methods from both theoretical and practical aspects.Among them,the combination of specific monitoring data for deformation analysis and forecasting has been widely used in major projects in China.Each mathematical model for deformation prediction has its own characteristics and advantages,and it also has inevitable defects.The research shows that in the deformation analysis,the pre-processing method is adopted to denoise the original observation data,and then the denoised data is used to establish the corresponding combined prediction model.The prediction accuracy is often better than the single nonlinear prediction method.Based on this,the paper firstly optimizes the nonlinear wavelet transform threshold denoising algorithm,and then uses the nonlinear wavelet transform threshold denoising optimization algorithm to denoise the original data.Finally,the wavelet transform is used to denoise the deformed data pair.The deformation object is analyzed by deformation and the wavelet optimization-GM(1,1)combination model is used to predict the deformation trend of the deformation object.This paper verifies the theoretical validity and engineering practicability of nonlinear wavelet transform threshold denoising optimization algorithm in building deformation monitoring data processing and prediction through theoretical analysis,experimental design and engineering application.The specific research contents and main results include:(1)The wavelet analysis method is introduced systematically,and the nonlinear wavelet transform threshold denoising algorithm is emphasized and a lot of optimization research is carried out.The Sym7 wavelet basis function,the best improvement method of threshold function and threshold estimate,a new wavelet denoising quality composite evaluation index M which can be effectively applied to the deformation monitoring data denoising processing are obtained;(2)For the deformation analysis of buildings,the denoising process of the original data is performed by the nonlinear wavelet transform threshold denoising optimization algorithm,and the deformed deformation curve is smoother,which effectively reduces the distortion of the original data.The adverse effects reflect the deformation trend,regularity and amplitude of the deformed object better;(3)A new wavelet denoising quality composite evaluation index M is used to evaluate the denoising quality of the building deformation monitoring data after wavelet denoising optimization,and the optimal decomposition and reconstruction scale is obtained,and the original is retained to the utmost extent.Useful signals,eliminating most of the noise,which has certain guiding significance for building deformation monitoring data processing;(4)The GM(1,1)model and the wavelet optimization-GM(1,1)combination model are used to predict the deformation trend of buildings.The results show that the wavelet optimization-GM(1,1)combination model has higher prediction accuracy.The prediction results are more stable and reliable,and the theoretical validity and engineering practicability of the nonlinear wavelet transform threshold denoising optimization algorithm are further verified.