Research On Thermal Inertia Model Identification Based On Intelligent Optimization Algorithm

Model identification of thermal inertia is the foundation of researching large thermal control problems.Determination of the structure and parameters of the model is an important part of works on designing,commissioning and operating a control system.After the identification of the object model,the actual control system can be targeted to optimize the parameters,effectively improve the quality and efficiency of the control system and to ensure the safety of the production process.This paper takes model identification technology and PID parameter optimization techniques as the background to study typical thermal large inertia object based on actual operating data.Improved particle swarm algorithm was used to identify the model of thermal object,the main factors that affect the results of system identification and these problems need to pay more attention was analyzed.The optimization of PID controller parameters using the identification results was realized.PSO has the characteristics of high speed,flexibility and convenience.But it is easy to fall into local optimal solution and to be premature,its search precision is not high,therefore needs to research the improved method based on the particle swarm algorithm.First,chaotic search and simulated annealing algorithm based on particle swarm optimization algorithm were used to make the algorithm more diversity and jump out of local optimum.Secondly,combined with bacterial chemotaxis algorithm,repulsive operation is used to enhance the search ability of particles.The improved algorithm is validated by the representative function,and the results show that the algorithm convergence,stability,and global search capability have been significantly improved.Based on the cognition of the thermal model structure,the improved particle swarm optimization algorithm is applied to the model identification of the laboratory boiler temperature of the single input single output object and the Ultra Supercritical Boiler main steam temperature of multi input single output object.The identification reaults of improved particle swarm optimization algorithms were compared and analyzed.The results show that the bacterial chemotaxis particle swarm algorithm can reflect the dynamic characteristics of the object,increase the identification accuracy,shorten the identification time and improve the identification effect.