Well control remote simulation system adaptive state feedback control method 

Ying Jia, Zhengsheng Chang, Jianjun Xu, Yanhui Li

Abstract: In order to reduce the oil and gas overflow in the process of drilling blowout accidents, effective protection of well control equipment and control right is the key to ensure the safety of well control.Need to drilling workers, drilling well control simulation training.To drilling well control of the remote machine simulation process, need to have the corresponding well control physical equipment and physical simulation drill platform operation, in order to create a real drilling environment for training personnel, and can be objective to display the level of training personnel operating the information and skills to achieve the purpose of simulation training.In order to achieve this purpose, the method of adaptive state feedback control is applied to the well control remote control simulation system.According to the system characteristics, design the adaptive steady-state controller, combined with the PID adaptive control, proposed an adaptive well control remote pressure oil preparation simulation of steady state feedback control algorithm.Method to verify the feasibility of taking the actual project as the background of Da qing oilfield drilling company simulation calculation and experiment results show that this method can make the operator and simulated under the environment of real close operation well control equipment, commonly used to implement inspection and operation of remote console and content of real operation training and other functions.

: Download (free)  CSE-PDF

Xingyuan Wang, Haijie Dong, Qingao Qin

Abstract: The new Hsp90 inhibitors, aminopyrazole-substituted resorcylate compounds, has extensive and effective Hsp90 inhibitory activity and was designed to develop new antibacterial drugs. Quantitative structure-activity relationship method was used to predict fungal selectivity of new Hsp90 inhibitors. The correlation coefficient R of the best linear model was 0.89. Two nonlinear models were established by gene expression programming and back propagation artificial neural network. The correlation coefficient and mean square error of the train set of gene expression programming model were 0.88 and 0.10, the test set were 0.89 and 0.11 respectively. In the back propagation artificial neural network model, the correlation coefficient of the train set was 0.95, the correlation coefficient of the test set was 0.95, the correlation coefficient of the verification set was 0.95, and the mean square error was 0.0069. The best satisfied model will provide reference for the design and development of new Hsp90 inhibitors.

: Download (free)  CSE-PDF