Pressure control method and device innovative design for deep oil in-situ exploration and coring

Deep oil exploration coring technology cannot accurately maintain the in-situ pressure and temperature of samples, which leads to a distortion of deep oil and gas resource reserve evaluations based on conventional cores and cannot guide the development of deep oil and gas resources on Earth. The fundamental reason is the lack of temperature and pressure control in in-situ coring environments. In this paper, a pressure control method of a coring device is studied. The theory and method of deep intelligent temperature-pressure coupling control are innovatively proposed, and a multifield coupling dynamic sealing model is established. The optimal cardinality three term PID (Proportional-Integral-Differential) intelligent control algorithm of pressure system is developed. The temperature-pressure characteristic of the gas-liquid two-phase cavity is analyzed, and the pressure intelligent control is carried out based on three term PID control algorithms. An in-situ condition-preserved coring (ICP-Coring) device is developed, and an intelligent control system for the temperature and pressure of the coring device is designed and verified by experiments. The results show that the temperature-pressure coupling control system can effectively realize stable sealing under temperature-pressure fields of 140 MPa and 150 °C. The temperature-pressure coupling control method can accurately realize a constant pressure inside the coring device. The maximum working pressure is 140 MPa, and the effective pressure compensation range is 20 MPa. The numerical simulation experiment of pressure system control algorithm is carried out, and the optimal cardinality and three term coefficients are obtained. The pressure steady-state error is less than 0.01%. The method of temperature-pressure coupling control has guiding significance for coring device research, and is also the basis for temperature-pressure decoupling control in ICP-Coring.

Deep original information preservation by applying in-situ film formation technology during coring

Accurately obtaining the original information of an in-situ rock via coring is a significant guiding step for exploring and developing deep oil and gas resources.It is difficult for traditional coring technology and equipment to preserve the original information in deep rocks.This study develops a technology for insitu substance-preserved(ISP),moisture-preserved(IMP),and light-preserved(ILP)coring.This technology stores the original information in real time by forming a solid sealing film on the in-situ sample during coring.This study designed the ISP-IMP-ILP-Coring process and tool.In addition,an ISP-IMP-ILPCoring process simulation system was developed.The effects of temperature,pressure,and film thickness on the quality of the in-situ film were investigated by performing in-situ film-forming simulation experiments.A solid sealing film with a thickness of 2-3 mm can be formed;it completely covers the core sample and has uniform thickness.The film maintains good ISP-IMP-ILP properties and can protect the core sample in the in-situ environment steadily.This study verifies the feasibility of“film formation during coring”technology and provides strong support for the engineering application of ISP-IMP-ILPCoring technology.

Strip flatness prediction of cold rolling based on ensemble methods

Aiming at the problem of insufficient prediction accuracy of strip flatness at the outlet of cold tandem rolling,the prediction performance of strip flatness based on different ensemble methods was studied and a high-precision prediction ensemble model of strip flatness at the outlet was established.Firstly,based on linear regression(LR),K nearest neighbors(KNN),support vector regression,regression trees(RT),and backpropagation neural network(BPN),bagging,boosting,and stacking ensemble methods were used for ensemble experiments.Secondly,three existing ensemble models,i.e.,random forest,extreme random tree(ET)and extreme gradient boosting,were used to conduct experiments and compare the results.The research shows that bagging,boosting,and stacking three ensemble methods have the most significant improvement in the prediction accuracy of the regression trees model,which is increased by 5.28%,6.51%,and 5.32%,respectively.At the same time,the stacking ensemble method improves both the simple model and the complex model,and the improvement effect on the simple base model is the greatest,which is 4.69%higher than that of the base model KNN.Comparing all of the ensemble models,the stacking ensemble model of level-1(ET,AdaBoost-RT,LR,BPN)paired with level-2(LR)was discovered to be the best model(EALB-LR)and can be further studied for industrial applications.