Research And Application Of Compressed Air Prediction Model Based On Neural Network

With the continuous development of industrial and agricultural society,the career of information technology has continued to grow up, and the pattern that human utilize energy begin to convert to multi-energy structure. Compressed air energy as an important driving force is an important part of the industrial production system operation scheduling, the forecast of the compressed air utilization is great significance to improve the enterprise economic benefit and social benefit.In this paper, the neural network focuses on the short-time forecast of the corporation daily compressed air. First, we need to deal with the high-dimensional data samples; Then, use BP neural network to predict the compressed air utilization, utilize the compressed air data that required in24hours as the original sequence, then train and predict these data to achieve a better result; Considering the defect of original BP neural network and improving prediction accuracy and efficiency of network training, we use the wavelet analysis and the genetic algorithm to optimize the BP neural network, and establish WNN predictive model and GA-WNN predictive model. And we analysis the various factors that may affect the compressed air, such as the test data, the number of input nodes and hidden nodes. The results showed that, Combining the genetic algorithm and the wavelet neural network as the predictive model is feasible and effective to predict the compressed air utilization, this model can be used in the industrial production and real-time scheduling.