Study On Optimization And Control Methods For Membrane Separation Process

Membrane technology, a high and new technology, has received significant attraction in the circularly economic ages nowadays because of its multiple functions such as separation, concentration, purification and refining. Besides, membrane technology is an important means to help enterprises achieve green manufacture as it is highly efficient, energy-saving, environment-friendly and molecule-class filtration. In order to make full use of the advantages of membrane technology and provide the enterprises with the best economic benefits, based on the safe and steady operation of membrane device, optimum operation conditions of different processes are required to guarantee the membrane performances and attain the highest recovery and purity. Therefore, a new topic has been put forward in the field of control theory and control engineering. To realize the optimal membrane separation process, this paper develops investigations of optimization and control methods for the novel gas and liquid membrane separation processes and also presents the innovated methods. Main studies of this paper are as follows:Firstly, the two-stage membrane separation device arranged by TBM medium-low pressure modules is used to recovery the hydrogen from the refinery gases. As feed gas membrane in-let temperature is difficult to be steady, a fuzzy self-tuning PID control method is presented. The basic control loops consist of water cooler outlet temperature, cyclone separator level, feed gas pressure, permeate gas pressures and operating temperatures for two stages. The simulation program for the gas membrane separation process based on mathematical modeling is integrated in DCS as operation guidance, thus the operation of the device is optimized. Through successful application in Zhenhai Refinery, it is approved that the technical process is optimum, the control systems are safe, and the TBM membranes will be promoted in a new area.Secondly, a MIMO soft-sensor model for the gas membrane separation process has been established based on the radial basis function(RBF) neural networks. The quality target, i.e., permeate hydrogen concentration, and the economic target, i.e., hydrogen recovery rate, are estimated quantitatively by measuring the operating temperature, feed-side pressure, permeate-side pressure, residue-side pressure, feed-gas flux and feed-hydrogen concentration. It is concluded by simulation results that this novel approach with desired precision has abstained the complexity of the mechanisem model, and laid a foundation for advanced process control(APC) and optimal control of the gas membrane separation process.Thirdly, the principle of the gas membrane separation process and the operation characteristics of the separation device are studied deeply. Various control methods of feed or permeate gas flux in the gas membrane separation process are compared and analyzed. One control principle of the feed or permeate gas flux is applied: the feed gas flux or the permeate gas flux can be respectively auto-controlled through the control valve installed in the residue gas stream. One double-control method of the material balance is also used. Based on improved genetic algorithm(GA), the dynamic and optimal control for hydrogen recovery rate is realized. The methods are the same with all optimal controls for feed or permeate gas flux. The correctness and the effectiveness of the control philosophy are verified successfully in the process of extracting the high purity hydrogen from a continuous reformer in Luoyang Petrochemical and a hydrocracking in Jinling Petrochemical respectively.Fourthly, the concentration processes for liquid membrane separation are studied. The feed tank levels are difficult to be tracked efficiently in the random-rinsing and continuous-discharging nanofiltration process.The tracking control of feed tank levels is improved by the selective control and the transparent predictive functional control(PFC). It is proved by application in Gushen Group that the operation process is steady, the product quality is increased, the tracking control precision is ensured, the ideal recovery of oligosaccharides from streamed wastewater together with the optimal strategies can prompt membrane technology to be applied in environment protection and other separation processes.Fifthly, a concept of optimal membrane separation process is presented. It should meet the needs of separation performance and the best economic requirement to a certain industrial application. Based on the analysis of the membrane separation principle and the process characteristics, the total control idea and design principle are proposed to overcome the difficulties of the membrane separation process control, and a three-stage structure combined by basic control, advanced control and coordinated optimization is confirmed. One total-integrated optimal control system with advanced process control and real-time optimization is designed to guarantee the quality target of permeate gas hydrogen concentration and realize the maximum recovery rate for the hydrogen recovery process from refinery gas streams.