Analysis Of Influencing Factors And Prediction Of Rock Entry Parameters By Rotary Drilling Rig

As the mainstream equipment of contemporary pile foundation engineering construction,rotary drilling rig has been widely used in many infrastructure constructions in China,but it is still a worldwide technical problem at this stage.At present,there are few studies on rock entry of rotary drilling rig.With the continuous development of big data analysis technology and artificial intelligence algorithm in recent years,it provides a new idea for rock breaking research of rotary drilling rig.In this thesis,the rock breaking process of rotary drilling rig is taken as the research object,and the influencing factors of rock breaking are qualitatively analyzed by relevant mechanics theory.Based on the analysis results of the influencing factors of rock breaking of rotary drilling rig,the rock breaking data is preprocessed and the correlation analysis is carried out,which provides a reference for the input and output selection of the rock entry parameter prediction model.Through the artificial intelligence algorithm,the drilling speed prediction model and the large cavity pressure prediction model are constructed to realize the accurate mapping of the multi-dimensional nonlinear rock entry mechanism and the accurate prediction of the rock entry parameters,which provides a certain reference for the future construction efficiency monitoring of the rotary drilling rig,the improvement of the pressurization method and the intelligent development.The main research contents of this thesis include :(1)The influencing factors of rock breaking of rotary drilling rig are analyzed.Based on the mechanics theory,a single pick machine-rock interaction model is established with a single pick as the object.The single pick machine-rock interaction model is combined with the output parameters of the host to study the influence of the host output parameters on the rock breaking ground when the rotary drilling rig breaks the rock,and the key influencing parameters are determined,which provides a basis for data processing and analysis of the rotary drilling rig.(2)Pretreatment and analysis of engineering data.Based on the qualitative analysis results of the influencing factors of drilling rock breaking,the key parameter extraction,drilling section division,abnormal data screening and elimination,drilling speed completion are carried out on the historical construction data,and the correlation between the rock entry parameters is analyzed to prepare for the construction of the prediction model.(3)The PSO-BP drilling speed prediction model and the deep LSTM large cavity pressure prediction model are established.Through the theoretical research of neural network algorithm,the initial weights and thresholds of error back propagation algorithm(BP)are optimized by using the global optimization ability of particle swarm optimization(PSO)to improve the learning ability and prediction accuracy of BP algorithm.The drilling speed prediction model based on PSO-BP algorithm is built to realize the nonlinear mapping between rock entry parameters and drilling speed and the prediction of drilling speed during rock breaking.At the same time,based on the long short-term memory network(LSTM),a deep LSTM large cavity pressure prediction model is constructed to improve the nonlinear mapping ability and prediction accuracy of the traditional LSTM,and realize the nonlinear mapping between the rock entry parameters and the large cavity pressure and the prediction of the large cavity pressure during rock breaking.Based on the construction data of Tangwanghe No.1 Bridge,the PSO-BP drilling speed prediction model and the deep LSTM large cavity pressure prediction model are verified and analyzed.