Zhezhao Zeng

Learn More
A novel algorithm for solving nonlinear equations is proposed. The computation is carried out by simple gradient descent rule with adaptive variable step-size. In order to make the algorithm be absolutely convergent, its convergence theorem was presented and proved. The convergence theorem indicates the theory criterion selecting the magnitude of the(More)
A neural-network algorithm for solving a set of nonlinear equations is proposed. The computation is carried out by simple gradient descent rule with variable step-size. In order to make the algorithm be absolutely convergent, its convergence theorem was presented and proved. The convergence theorem indicates the theory criterion selecting the magnitude of(More)
The problem of the offset thermal drift affects the measurement accuracy of pressure sensor in practical applications. To solve the problem, we proposed an approach of the non-linear compensation of pressure sensor using the neural network algorithm with Chebyshev basis functions. The convergence of the neural network algorithm is researched. To validate(More)
An algorithm for finding multiple roots of polynomials based on PID (Proportional-Integral-Derivative) neurons network is developed, which were not well solved by the other methods. The approach is especially suitable for finding the multiple roots of polynomials. Several examples are given to illustrate the efficiency of the new method and to give the(More)
A neural-network algorithm is proposed to find simultaneously roots of polynomials which were not well solved by the other methods. Its convergence was researched. The computation is carried out by simple steepest descent rule with adaptive variable learning rate. The specific examples showed that the proposed method can find the multiple roots of(More)
This paper introduces in detail the optimal design approach of high-order digital differentiator based on the algorithm of neural networks. The main idea is to minimize the sum of the square errors between the amplitude response of the ideal differentiator and that of the designed by training the weight vector of neural networks, then obtaining the impulse(More)
To improve the error compensation precision of sensors, a method of the error compensation of sensors based on the neural network algorithm is proposed. The convergence of the algorithm is researched. The theory gist to select learning rate is provided by the convergence theorem. To validate the validity of the algorithm, the simulation example of the error(More)
An optimal design approach of Hilbert convertor is researched in detail based on the neural-network algorithm. The main idea is to minimize the sum of the square errors between the amplitude-frequency response of the desired Hilbert convertor and that of the designed by training the weight vector of neural-network, then obtains the impulse response of(More)