Prediction of Laboratory Equipment Demand Based on Grey Relation Analysis and Elman Neural Network

To deal with low accuracy problem of prediction of laboratory equipment demand, a new demand prediction model for laboratory equipment was established combining grey relation analysis and Elman (GRA-Elman) neural network.First, the correlation between various influence factors was calculated using grey relation analysis method. Then three factors with higher correlation were chosen as training data of Elman neural network. The original 6-12-1 Elman network model is reduced to a 3-6-1model. The algorithm was tested based on the actual demand of a certain equipment, and the prediction accuracies of GRA-Elman network model, single Elman network model and single BP network model were compared.The simulation results show that the training time of GRA-Elman network model is the shortest; the maximum relative error of its prediction is -0.67%, while those of single Elman network model and the single BP network model are -2.61% and -4.18% respectively. So the established GRA-Elman model has higher prediction precision and simpler structure.