Research On The Deviation Compensation Least-square Parameters Estimation Method

Industrial application background of subject was complex mechanical system dynamics modeling and design. Funding by the National Science Fundamental Project which named"Direct identification of non-linear continuous-time dynamic parametric models via wavelet modulating functions"(Project No: 50875028), Develop deviation compensation least-square parameters estimation method for the research content. The thesis has chiefly carried on research of the following several aspects:(1) Introduces the fundamental principle of modulation function method and four modulation function nature and features. And the identified parametric model of continuous time system based on modulating function has been established.(2) Introduces a kind of improved deviation compensation least-square algorithm. The identification principle and recursive process of this method were studied in discrete system. Use this method in the example of model parameter identification of discrete systems disturbed by color noise that proved the validity of the proposed algorithm.(3) Hermite modulation function method equivalent model was established for continuous system. Identify linear continuous model with deviation compensation principle in Identification of discrete model, Take linear continuous system under color noise as example ,which proof that deviation compensation least square method based on modulation function have good effect on noise suppression in identifying model parameters of linear continuous system.(4) The method proposed in this thesis are used in 2/2 and 1/2 order dynamical parameter of closed loop HAGC system to do the parameter identification of the model. Compared with measured data, verify the practicability of the method of this paper.Research shows that: Deviation compensation principle both for discrete model identification and continuous model identification, all have better anti-interference ability, and has extensive applicability.