Research On ROP Increasing In Keshen Block Based On Big Data Analysis

The oil and gas resources in the Keshen block of the Tarim Oilfield are abundant,but due to the deep depth of the reservoir and the complex geological conditions,the ROP is greatly limited.At present,the factors that restrict the ROP are still not clear,and the traditional ROP increasing methods can no longer meet the production requirements.In view of the above problems,this paper uses the big data analysis method to study the ROP increasing technology of this block and obtains the following main results:(1)By using hierarchical clustering analysis and K-means method to perform cluster analysis on well logging data of each well in the Keshen block,the wells with similar geological characteristics are merged to make the ROP increasing method more targeted;(2)Correlativity coefficient method was used to analyze the correlations between the engineering parameters of each well in Keshen block,and the main control factors affecting the ROP were defined,which laid an important foundation for the establishment of the ROP prediction model.(3)By using the random forest regression method and the gradient boosing decision tree regression method,a prediction model of the ROP which conforms to the characteristics of various types of wells is established.Then optimize and verify the ROP prediction model,and achieved good application results;(4)The partial dependency analysis method is used to analyze the influence of each parameter in the prediction model of the ROP on the prediction results,and the analysis results can effectively guide the project production.