The Application Of Intelligent Optimization Algorithm In Process Control

Research on swarm intelligence optimization algorithm,such as the particle swarm algorithm,the genetic algorithm,the cuckoo search algorithm,etc.,has become a hot topic in the field of computer optimization.The cuckoo search algorithm with its simple structure and excellent global search ability is widely used in various fields.This paper studies the basic characteristics of the cuckoo search algorithm through the use of several standard test functions.At the same time,according to the specific industrial production problems,this paper propose the corresponding improvement of the cuckoo search algorithm,such as the CS-NLJ algorithm,the modified CS-NLJ algorithm,etc.Now,the improved cuckoo search algorithm has been applied in several fields of process control,including:(1)The studies of multi-input multi-output(MIMO)Hammerstein systems identification.This paper firstly proposes a novel CS-NLJ identification algorithm for the general type of MIMO Hammerstein systems in the presence of typical heavy-tailed noises.The proposed algorithm is based on the cuckoo search(CS)algorithm,and uses a new mutation rule based on the local nonlinear stochastic search NLJ algorithm.Further,by combining the global CS search and the local NLJ search,the global best individuals can be achieved.The simulation examples demonstrate that the proposed identification method can provide more accurate parameter estimates.(2)The identification of nonlinear Wiener model.This paper firstly applied the CS-NLJ algorithm to the identification of Wiener.It is necessary to point out that the structure of Wiener model using a unified model mentioned in the previous chapter.The dynamic linear part has been approximated by a unified two order dynamic model,the polynomial function has been used to fit the nonlinear part.Finally,two simulation examples are given to verify the effectiveness of the proposed algorithm.It should be noted that the two simulation models are based on two different types of heavy tailed noise.(3)The PID controller tuning method for time delay continuous systems with multi-objective and multi-constraint optimization.This paper firstly proposes a novel modified CS-based PID controller tuning method for the multi-objective and multi-constraint optimization problems.In the proposed swarm intelligent optimization algorithm,in order to improve the search performances,we apply the heavy-tailed-distribution(HTD)sequences into the basic cuckoo search(CS)algorithm.Thus,in terms of achieving global optimization,the modified CS algorithm can provides more accurate PID parameter estimates.The simulation experiments verify that the proposed algorithm is quite efficient.(4)The tuning of PID controller for unstable time delay process with multiple constraints.This paper firstly proposes a novel modified cuckoo search(CS)algorithm for the unstable time delay process with multiple constraints.The proposed algorithm is based on the cuckoo search(CS)algorithm,and uses a new random sequence based on the t-distribution.Thus,by applying the t-distribution sequence into the CS algorithm,the global search performances can be achieved.Simulation studies have been carried out on unstable first-order plus time delay(UFOPTD)process,unstable second-order plus time delay(USOPTD)process and unstable high-order plus time delay(UHOPTD)process to verify the effectiveness of the proposed algorithm.