Font Size: a A A

Study On The Model Parameters Identification Methods Of The Dynamically Tuned Gyroscope

Posted on:2016-12-09Degree:DoctorType:Dissertation
Country:ChinaCandidate:J Y ZhaoFull Text:PDF
GTID:1108330485451984Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
Dynamically tuned gyroscope(DTG) is one of the most widely used high-precision gyroscope, and the system model analysis of DTG is the prerequisite for its further research and application. In this paper, according to the general system identification framework, based on DTG open-loop or closed-loop model sets and inherent colored noise analysis, the frequency response identification method, least square identification method and subspace identification method of the dynamically tuned gyroscope is deeply researched, which is corresponded with the transfer function model, discrete time model and state space model. The main content of this article is as follows:1. Firstly, the open-closed loop model and the inherent colored noise of DTG are studied. According to the Euler equation of DTG, the open-loop model is constructed at first. And for the explicated closed loop system, each of the circuit unit is simplified according to its frequency response property and the high order complex system is finally transmitted to a low order system. According to characteristics of mechanical structure in the gyroscope, the internal strong colored noises of DTG are analyzed.2. In the second part, this paper focuses on the study of DTG frequency response identification method. Using correlation analysis method and Levy identification method to identify the open-loop model, then one noise restraining method is combined with the Levy method, which could restrain the narrowband or single frequency noise based on the confidence level in frequency domain, enhancing the noise suppression capability to Levy method. According to the two-stage closed-loop identification method, the paper puts forward one two-stage Levy closed-loop identification method, which is applied in the DTG closed-loop identification.3. In the following section, DTG least-square identification method is fully studied. Colored noise is analyzed under the least square method of prejudice problem. On the premise of data preprocessing, this paper puts forward the DTG open-loop identification method with assistant variables. And based on the framework of optimal assistant variable method, this paper designs one hierarchical nearly optimal closed-loop assistant variable method, which could be applied in the BJ models.4. Subspace identification method for DTG is discussed in this part. Based on subspace identification framework and data Hankel matrices, orthogonal subspace identification method is built, which could eliminate the influence of DTG inherent colored noise in its open loop subspace identification process. Then, the recursive orthogonal decomposition closed-loop subspace identification strategy for DTG is analyzed. Based on the principle of matrix sliding window and elements corresponding law, the vector space projection could be recursively conducted with constant matrix dimension. With the propagator and matrix similarity transformation principle, the SVD decomposition could be avoided.5. Finally, the results of DTG identification are applied to fault diagnosis about disorder of gyro motor speed, digital circuit design and other practical cases.
Keywords/Search Tags:DTG, system identification, frequency response, least square method, subspace identification
PDF Full Text Request
Related items