Application Research of Petroleum Basic Data Mining System Based on Intelligent Computing and Decision Tree Algorithm

Abstract

Recent improvements in data mining technologies, besides the IoT, enable the implementation of a strategy for boosting oil output from oil wells. As a regularly employed improved oil recovery technology, steam flood injection takes use of thermodynamic and gravitational capabilities to deploy and neutralize oil on-site to raise oil output. Instead of relying on conventional physics to model steam floods, this research proposes using a combination of a chimp optimization algorithm (ChOA) and a decision tree to better represent steam flood performance. We present a method for dealing with a particular type of petroleum time series data using ChOA in conjunction with decision trees and IoT. It is shown that the method is useful in predicting oil production in steam floods. Even more impressive is the 4.02 percent increase in oil output that may be achieved via the use of a new optimization system that offers the best possible steam allocation plan. Our objective has been to develop a cloud-based minimum viable product capable of data collection and storage and also training and deployment of a cloud ChOA model. Predictive maintenance, for example, might benefit from this workflow's ability to analyze time series data.