Wenming Cao

An adaptive controller for a class of nonlinear system using direction basis function

Adaptive controllers are investigated, using the direction basis function (DBF), for a class of nonlinear dynamical systems. Based on the criterion of Lyapunov stability, the DBF is designed which guarantees that the output of the controlled system asymptotically tracks the reference signals. Simulation shows the good tracking effectiveness of the adaptive controller.

The Research of Speaker-Independent Continuous Mandarin Digits Speech- Recognition Based on the Dynamic Search Method of High-Dimension Space Vertex Cover

In this paper, we present a novel algorithm of speaker-independent continuous Mandarin digits speech-recognition, which is based on the dynamic searching method of high-dimension space vertex cover. It doesn’t need endpoint detecting and segmenting. We construct a coverage area for every class of digits firstly, and then we put every numeric string into these coverage-areas, and the numeric string is recognized directly by the dynamic search method. Finally, there are 32 people in experiment, 16 female and 16 male, and 256 digits all together. All these digits are not learned. The correct recognition result is 218, and error recognition result is 26. Correct recognition rate is 85%

Hardware design of two weighted neural network and application for object recognition

In this paper, the design methodology of neural network hardware has been discussed, and two weighted neural network implemented by this method been applied for object recognition. It was pointed out that the main problem of the two weighted neural network hardware implementation lies in three aspects. At final, two weighted neural network implemented by this method is applied for object recognition, and the algorithm were presented. We did experiments on recognition of omnidirectionally oriented rigid objects on the same level, using the two weighted neural networks. Many animal and vehicle models (even with rather similar shapes) were recognized omnidirectionally thousands of times. For total 8800 tests, the correct recognition rate is 98.75%, the error rate and the rejection rate are 0.5 and 1.25% respectively.

Performance of the DBF neural controller for transient stability enhancement of the power system

To enhance the first swing stability a DBF neural network fastvalving controller is proposed. The solution approach is based on a recent fuzzy fastvalving control scheme. Disturbances in the PS are used for training the NN controller. The performance of the DBF neural controller is simulated in a single machine to an infinite bus power system.

The application of direction basis function networks for adaptive regulations of nonlinear continuous time systems

For a class of nonlinear continuous time systems xdot = f(x) + Bu + d, when the nonlinear function f(x) is bounded or satisfies linear growth condition with unknown growth coefficient, we first prove that x falls into a compact set, then two adaptive regulators are proposed based on the approximation capability of direct basis function networks . According to the Lyapunov stability theory, the control algorithms are proved to be globally stable, the state of closed-loop system is uniformly ultimately bounded and the control law is continuous.

Constructive transparent direction basis function network learning for non-linear control

A constructive direction basis function network (DBFN) learning method is applied. This approach uses the functional equivalence principles between DBFN and fuzzy systems in order to achieve a minimal structure network. Firstly, an initial network based on linguistic descriptions is constructed. Secondly, a constrained constructive adaptation law, based on a minimal resource allocating algorithm, is applied in order to adjust on-line the structure and parameters of the DBFN, keeping the transparency property and guaranteeing the linguistic interpretation. Thus, at any instant, knowledge from the network can be easily extracted, validating its structure. Experimental results in a benchmark process show the effectiveness of the presented approach.