The pule placement design principle pruvides a good way of determining a closed loop system with desired properties. The method has been used in adaptive contexts for many years. A drawback of the method is, however, ...
详细信息
The pule placement design principle pruvides a good way of determining a closed loop system with desired properties. The method has been used in adaptive contexts for many years. A drawback of the method is, however, that the polynomial formulation mav lead to robustness problems because of the sensitivity with respect to the coefficients of the polynomials. Further, unless all process zeros are cancelled a set of linear equations has to be solved at each step in time. In this paper an attempt is made to avoid these two problems. The sensitivity of the polynomial formulation is avoided by parameterizing the model in the poles and zeros, which are estimated on-line. With the plant expressed in this form it is possible to parameterize the controller such that the solution of the Diophantine equation corresponds to the solution of a triangular system of equations. The controller parameters can thus be computed recursively without extensive computations
Most adaptive controllers do not assume any a priori knowledge about the process to be cont rolled. In most situations there is, however, some knowledge about the process. This knowledge should be used to improve the ...
详细信息
Most adaptive controllers do not assume any a priori knowledge about the process to be cont rolled. In most situations there is, however, some knowledge about the process. This knowledge should be used to improve the performance compared to the “black box” approach of earlier adaptive controllers The a priori knowledge is in this paper used in the estimation of the process and also in the design of the controller. The use in the design is new compared to earlier approaches. The a priori knowledge is here used to design a nominal controller. The unmodelled part of the process is estimated and an adaptive stability augmentation signal is added to the nominal controller. Different types of a priori knowledge and design methods are discussed and the performances are investigated through analysis
An easily implemented and computationally efficient procedure is presented for the generation of autocorrelated pseudo-random numbers with specific probability distributions. A plot illustrates the relationship among ...
详细信息
An easily implemented and computationally efficient procedure is presented for the generation of autocorrelated pseudo-random numbers with specific probability distributions. A plot illustrates the relationship among the autocorrelations of the uniform, Rayleigh, and exponential distributions corresponding to a given autocorrelation in the normal generating distribution.
暂无评论