Liandong Fu

The Fault Diagnosis for Electro-Hydraulic Servo Valve Based on the Improved Genetic Neural Network Algorithm

The paper analyzes the merits and drawbacks of the genetic algorithm and BP neural network, combines with the improved genetic algorithm and BP neural network to obtain a new algorithm. The new algorithm is used in the fault diagnosis of electro-hydraulic servo valve and justified its validity, accuracy and rapidity by experiment. The BP algorithm, the conventional GA-BP algorithm and the improved GA-BP algorithm are compared by the data of experiment. It is shown the superiority of the improved GA-BP algorithm in the fault diagnosis field

Application of immune neural network for electro-hydraulic servo valve fault diagnosis

Using the biological immunology principle unifies the neural network and the immunity algorithm to form the immunity neural network, which is applied to electro-hydraulic servo valve breakdown diagnosis. The result indicated that, the immunity neural network can identify many kinds of failures pattern recognition accurately by the smaller network scale, and has the high efficiency, good fault-tolerant performance and formidable auto-adapted ability.