Font Size: a A A

A Study On Strategy Of Optimal Control In Procedure Of Chinese Traditional Medicine

Posted on:2007-12-26Degree:DoctorType:Dissertation
Country:ChinaCandidate:Z X HuangFull Text:PDF
GTID:1118360215499084Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
The advanced control of Chinese traditional medicine is important toimprove the technology level of Chinese traditional medicine industry andmaintain its unique characteristic and advantage. At present, process optimalcontrol becomes a bottleneck that limits the yield and quality of Chinesetraditional medicine, on account of behindhand control technique and lowlevel of automation in procedure of Chinese traditional medicine. Aiming atstrong non-linear, strong coupling, uncertainty, time-varying, large delay andmulti-restriction, using the traditional control theory or the single intelligentmethod is difficult to meet the requirements of multi-objective optimalcontrol. In this dissertation, optimal control strategy that combines steadyoptimization with control of dynamic parameters is proposed, and theproblems of state, yield and quality optimal control are solved efficiently. Themain content of study and innovation are ordered as follow:1. Based on qualitative analysis of production process of Chinesetraditional medicine, establishing the dynamic functional models ofcomplex extraction and concentration process of Chinese traditional medicine.Not only the basis of guidance for actual production operations is provided,but also the foundations of optimal design are established by dynamicsimulation of the main process variables.2. Aiming at too many controlling variables and strong coupling inextraction of Chinese Traditional Medicine, debasing the steady time of thetemperature in the extraction kettle, increasing the output of the volatile oil,decreasing the gross of energy consumption and enhancing the gross ofeffective liquid in extraction are taken as four optimal objectives, and amulti-objective optimal control scheme using the neural network is putforward, meanwhile genetic algorithm and variable structure algorithm areused to train the weight parameters and structure of neural network.3. A domain evolution model-based genetic algorithm is proposed, inorder to improve economic benefit in the recycling procedure of volatile oil.It defines three genetic operators that are migration, positive clustering andnegative clustering and imports fuzzy environment factor that are based onmulti-subgroup model. The simulation results validate that the search is bothrapid and efficient. It's an effective method for recycling volatile oil in procedure of Chinese traditional medicine.4. To the Double-effect and Energy-saving evaporator of concentrationprocedure of Chinese traditional medicine, high quality, high yield and lowconsumption are achieved by using SQP and systemic mathematic simulationto find the optimal operating condition. In SQP application, a methodcombining optimization with sequential modular approach is presented,establishing optimization model and procedure simulation model.5. It is found that the concentration control curve of Chinese traditionalmedicine is similar, adopting Iterative learning control which has robustnessproperty for tracking the curve to realize consistency control of Medicineliquid density. The simulation curves illustrate its tracking effect.6. The double-loop controller to evaporation speed is put forward basedon DMC predictive and PID serial structure. It not only has PID advantagesin convenience of adjusting control parameters, strong robustness andanti-jamming capability, but also has advantages of predictive control indealing with the controlled object with pure delay immediately, favorabletracking property and robustness to mismatched model.7. By reasonable system design, equipment selection, software designand optimization strategy, the optimal control equipment of Chinesetraditional medicine is developed with a view to the automation of wholeprocedure of Chinese traditional medicine.
Keywords/Search Tags:procedure of Chinese traditional medicine, multi-object optimization, genetic algorithm, sequential quadratic programming, iterative learning control, DMC predictive control
PDF Full Text Request
Related items