Study On Identification Methods Based On Improved PSO Algorithm

In the design and implementation of advanced process control, a mathematical model of the controlled system is needed. In this paper, the modeling of industrial systems is discussed, especially for the problems encountered in modeling of nonlinear system, such as data preprocessing of identification, identification methods, etc. An improved particle swarm optimization (PSO) algorithm is proposed and employed to identify liner and nonlinear system.Additive noise and instantaneous disturbance may produce adverse influences to system identification. Excessive noise will seriously affect the identification results. Various methods of data preprocessing is introduced. When using particle swarm optimization, the algorithm may not converge due to the interference of noise. To solve this problem, author proposed an improved moving average filter algorithm. The new algorithm improved traditional moving average filter algorithm has large phase deviation which would lead an additional lag when identification. By dynamic phase correction, PSO method can identify the system model more effectively. The experimental results have shown that the new approach effectively enhance the accuracy of the identification result.In many studies, PSO has been successful in a variety of optimization problems. But the speed of convergence of standard PSO algorithm on high dimensional search space is unacceptable in practice. The concept of particle health was proposed, and gives an algorithm for particle health calculation in this paper. A new variation of PSO model (HPSO) proposed based on particle health can effectively reduce the probability of local optimum and enhance convergence speed especially for high dimensional search spaces. The proposed were tested by a variety of high-dimensional benchmark functions, and compared with Standard PSO algorithm and decreasing inertia weight variation (WPSO). We have found that the application of these modifications resulted in significant gain in speed and efficiency.Linear continuous system and nonlinear Wiener model were identified by HPSO algorithm, Input signal with strong noise were pre-processed. Simulation results show that the proposed algorithms can efficiently identify system parameters in strong noise and noiseless cases.