Research On Application Of Improved Gray Markov Model In Deformation Monitoring

With the accelerating progress of urbanization,the number of high-rise and super-high-rise buildings in cities has increased.The safety monitoring of buildings has also received increasing attention.During the construction and operation of large and high-rise buildings,in order to prevent the occurrence of unexpected conditions,it is necessary to monitor them and analyze their deformation trends in order to grasp the deformation laws of buildings and ensure the safety of buildings.In recent years,with the development of scientific theories and the needs of practical projects,many prediction models have emerged,such as gray models,Markov models,neural network models,time series analysis models,and so on.Although their research angles and depths are different,they have been widely used in the field of deformation data analysis and forecasting.From the perspective of applied research and practical work,a single model is no longer suitable for complex deformation analysis and forecasting.The organic combination of multiple theories and methods will be an effective way to solve the problem.The single gray model has low precision and poor prediction effect in building settlement prediction.Based on the author's learning and thinking,this paper finally proposes an improved gray-Markov combinatorial model based on wavelet denoising.The establishment of the model is:Firstly,use the wavelet transform theory to preprocess the data to obtain pure denoised data.Then use Matlab to build a metabolic gray model and predict the denoising data.Finally,the relevant predicted values obtained by the metabolic gray model are divided into different state intervals,and then use the Markov model to determine the final settlement prediction value.To verify the feasibility of this method,this paper establishes a gray-Markov combination model for prediction combining the measured data of a district in Guangzhou,then using the model to predict the monitoring data,the results show that the combined model is more accurate than a single gray model.In order to verify the reliability of the combined model,combined with the measured data of the H building in Qingdao,this paper establishes a combination model for prediction,then compare the predicted value of the combined model with the grey model and the grey model prediction of denoising.The results show that the final combined model accuracy and prediction results are better than the latter two models,and their stability and reliability are greatly improved.