Research On Model Predictive Control Of The Collecting System Of Linear Fresnel Photothermal Power Plant

It is urgent for human beings to research the usage of renewable energy and related technologies,with the depletion of conventional fossil energy sources and the escalation of environmental pollution problems.As the renewable energy source,solar energy resources have been highly thought by countries all over the world due to their high economic,universality and rich.The linear Fresnel solar thermal power generation system is a kind of medium-high temperature heat collecting system.Comparing with the trough solar thermal power generation system used for large-scale commercialization,it is inexpensive,is easy to manufacture and has low wind load on the mirror because of the use of plane mirrors in the heat collecting field.Therefore,it has great potential for solar high-temperature heat utilization.Nowadays,the research on linear Fresnel solar system has become one of the frontier topics in the field of thermal power generation.Based on the research status of solar thermal power generation at home and abroad,this paper firstly introduces four kinds of thermal power generation technologies(tower,trough,dish,linear Fresnel)used widely at present,and analyzes the characteristics of different power generation methods.In particular,the operation mode of the concentrating collector system,the heat storage system,the steam power generation system and the heat collecting system of the linear Fresnel solar thermal power generation concerned in this paper are explained.The heat collecting system is a complex nonlinear system including various control loops,such as circulating pump flow,steam separator water level,mirror field outlet steam temperature and pressure control loops.The control process is affected by complex meteorological conditions and it has the characteristics of large inertia,long lag time and time-varying parameters.The traditional PID control scheme can't meet the process requirements,so a better tracking property and more stable control algorithm is needed to complete the control of system-related parameters.As a kind of model predictive control method,the generalized predictive control algorithm can maintain the characteristics of the adaptive control algorithm,and overcome the effects of various uncertainties and complex changes.Even if the system model does not match,it still has a good control effect.The use of multi-step prediction,feedback correction and rolling optimization make the stability and robustness of the control system better,which is more suitable for industrial production control.In this study,the design of the model predictive controller is completed by the MATLAB simulation platform,and the performance of the control algorithm is more excellent comparing with the PID control.After completing the selection of the control algorithm,combining with the importance of the outlet steam parameters in the thermal power generation system,the control problem of the outlet steam parameters under the framework of the distributed control system is discussed and the design of the local controller is carried out.The control schemes in thethrough mode and the recirculation mode are studied separately,and the applicability and reliability of the algorithm in the adjustment process of the outlet steam temperature and pressure are verified by experiments.The drive system is the key component of the solar thermal power generation system.This paper introduces the drive control unit that constitutes the system,and explains its mode of action.Finally,considering the complexity of the actual working conditions,the safety performance of the collector tube is discussed,and the corresponding protection measures are briefly introduced.