Cascade steepest descent learning algorithm for multilayer feedforward neural network

In the article, a new and efficient multilayer neural networks learning algorithm is presented. The key concept of this new algorithm is the two-stage implementation of the steepest descent method. At the first stage, it is used to search the optimal learning constant η and momentum term α for each weights updating process. At the second stage, the Delta learning rule is then employed to modify the connecting weights in terms of the optimal η and α. Computer simulations show that the proposed new algorithm outmatches other learning algorithms both in convergence speed and success rate. On real industrial application, a self-tuning neural-network based PID controller for precise temperature control of an injection mode barrel system by using the developed algorithm is developed. Experiments show that the proposed self-tuning PID controller can precisely control the barrel temperature within ±0.5°C