Research On Deepwater Drilling Rate Optimization For Intelligent Application

In the process of deepwater drilling,drilling rate is an essential parameter that influences the drilling efficiency and determines drilling cost.Therefore,precise prediction and effective optimization of drilling rate are of great significance to improve drilling efficiency and reduce drilling cost.Until now,the existing drilling rate prediction models and optimization methods are difficult to be applied in the actual scene of deepwater drilling due to the poor calculation accuracy and limited application scenarios.Combined with the characteristics of actual deepwater drilling,this paper realizes the optimization of deepwater drilling rate for intelligent application.The main work of this paper is as follows.First,for the accuracy and real-time of deepwater drilling rate prediction and optimization,this paper constructs a real-time drilling rate prediction and optimization system.It integrates intelligent control,data management,data processing,intelligent application,and other functions.It can complete the accuracy analysis of different kinds of data and summarize the main process of drilling rate prediction and optimization,which provides a reference for the intelligent application of deepwater drilling rate optimization.Secondly,in order to remove the abnormal data and noise in the deepwater drilling data,a time-series data processing method based on Savitzky-Golay filtering is proposed.It integrates 3σ,Kmeans,LOF,and Savitzky-Golay filtering methods.This method can effectively remove the noise and outlier data in actual deepwater drilling data.By comparing the drilling rate prediction results of different methods,the effectiveness of the time-series data processing method is verified.Finally,in order to solve the problem of drilling rate prediction and optimization in the absence of geological information,this paper proposes an LSTM fusion model,which is called ELF model.The ELF model can realize the accurate prediction of drilling rate.And the simulated annealing algorithm is applied to optimize the combination of drilling parameters.Finally,the drilling rate prediction and optimization model without geological information is applied in the real-time drilling platform,which can increase the average drilling rate by 28.31%.The effectiveness of drilling rate optimization is verified.