Research On Identification In The Incomplete Information System

The control theory was not researched adequately at the beginning of its development. Because the information missing caused by certain reasons was ignored, the systems were regarded as complete information systems when system performance was researched. In the background of big data ages, the higher and higher precision requirement of information systems identification was demanded for people. So the identification of system parameters under ignoring missing information could not meet the social demands. The parameter identification algorithms of incomplete information systems were studied as the faults of the one for complete information systems were discovered, and kinds of interpolation algorithm were used to reduce the influence on the precision of system parameter identification by some researchers, so that the accurate identification of parameters was realized and the convergence of the system was insured. Therefore, the spline interpolation particle swarm algorithm was explored in this paper for the more accurate identification of the incomplete information systems, and the convergence of the algorithm was analyzed by building a random variable in the framework of random process. Then the probability and effectiveness of the algorithm was verified in the actual DC speed regulating systems. The specific steps were concluded as follows:Firstly, an integrated algorithm combined PSO (Particle Swarm Optimization, Particle Swarm Optimization) method with mean value interpolation method was researched for the parameter identification of incomplete information systems. The random missing data was interpolated under the mean value interpolation method, the parameters of identification model were optimized iterative by using the PSO algorithm. The fitness function under a reasonable error criterion was discovered, whose purpose was to iterate the parameters speeds and sizes of each particle, so that a real-time, accurate identification of parameters for incomplete information systems was achieved. Upon there before basis, the convergence of the algorithm was further analyzed. That the parameter identification accuracy for incomplete information systems using the proposed method could be significantly increased under certain conditions was verified by the simulation.Secondly, the spline interpolation method was introduced to the parameter identification for incomplete information systems in order to achieve accurate identification of system parameters. Timestamps signal encoding was used to count the quantity of random missing information, and the missing data was interpolated in real time using cubic spline interpolation method to achieve full identification data acquisition. On these basis, PSO was further used in system parameters identification in order to improve the accuracy of identification, and then its convergence was derived theoretically.Finally, the proposed method was applied in the parameter identification for speed control system of speed feedback control DC motor. And the input and output data of the system under different missing rates was test and captured. Then the parameters of this system were identified using the spline interpolation identification algorithm. It was verified that the proposed method can effectively identify the speed feedback control system of DC motor. Also was a necessary experimental foundation of the proposed method introduced for its further application and promotion.