Differential Pressure Prediction in Air Leak Detection Using RBF Neural Network

The objective of this paper is to increase the accuracy of the air leak detection through predicting the differential pressure using radial basis function(RBF)neural network. Air leak detection plays an important role in ensuring excellent performance of products. If the detection accuracy is low, unqualified products may be judged as qualified or otherwise. So it is important to increase the detection accuracy which is influenced by temperature, test pressure, chamber volume, balance time and so on. It is not easy to compensate the detection error directly, for the movement status of the gas in the chamber being complicated. This paper applies radial basis function(RBF)neural network to establish the function relationship between the leakage and influence factors and predict the differential pressure value which is used to determine whether the products leak or not. Experiment results using test data obtained on the differential-pressure-based air leak detection platform show that detection using RBF neural network has a higher accuracy than the detection not using it and RBF neural network for accuracy increase in air leak detection is feasibility.