Application Research Of Predictive Control Algorithm Based On Improved Particle Swarm Algorithm

In industrial sites,thermal objects often have large delays,nonlinear,and large inertia characteristics.Traditional controllers are often difficult to control and control quality is not ideal.Therefore,the researchers proposed to solve this nonlinear and time-varying problem of thermal objects under variable load conditions with a constrained predictive control algorithm,while limiting the output amplitude of the controller to maintain Stability and safety of the controller output.Dynamic Matrix Control(DMC)is a new type of control algorithm proposed for similar controlled objects in the past 30 years.In the continuous innovation,the algorithm has achieved rich theoretical and applied results and formed a model.Features such as prediction and rolling optimization,and strong robustness;however,the algorithm has a large amount of computation in engineering applications and actual calculations.Based on the above reasons,this paper combines the improved particle swarm optimization algorithm with the stepped constrained predictive control algorithm in the design of the algorithm to form a new control algorithm,which simplifies the calculation of the algorithm and improves the computational efficiency of the algorithm.Firstly,by reading the references in recent years,this paper learns and grasps the basic principles and formula derivation of the dynamic matrix control algorithm.In order to simplify the calculation of the algorithm and improve the computational efficiency,combines the particle swarm optimization algorithm based on genetic operators with the stepwise constrained predictive control,and theoretically deduces it to obtain a new one.Control Strategy.Through simulation,the influence of each parameter on the control quality is obtained.Further,the feasibility and practicability of the new algorithm are verified by the physical experiment water tank platform,thus paving the way for industrial field application.Secondly,through field investigations,learn and master various control strategies for large delay objects,as well as the advantages and disadvantages of various control strategies.Establish mathematical models for the flue gas baffle and water spray temperature control,respectively,and aim at the difficult control Design a predictive controller with constraints.Finally,combined with Guodian's secondary reheat outlet steam temperature optimization project,the designed controller was applied to a two-stage reheat outlet steam temperature system.In order to further study the influence of the flue gas baffle on the outlet steam temperature,The collected model is established in the DCS system.Through reasonable experiments and observation of the output curve of the controlled object,it is concluded that the improved predictive control algorithm is superior to the traditional PID controller in terms of adjustment time and control quality.