基于免疫神经网络模型的油气浓度预测研究

Study on Trend Estimate of the Oil Gas Based on the Immune Algorithm Neural Network

将免疫算法与神经网络理论相结合,提出了免疫神经网络预报模型以预报油库油气浓度。该模型首先用历史数据对网络进行训练,然后利用训练好的模型进行油气浓度的趋势预测,最后结合某油气预报实例检验了免疫神经网络模型的可行性。结果表明,该智能预报模型能够较好地识别油气扩散的变化规律,预报精度明显高于神经网络模型。

An Immune Algorithm neural network model for predicting oil gas is presented by means of combining Immune Algorithm with neural network theory. First of all,the network is trained by history data,then the model is used to predict the general development trend of oil gas, finally the oil gas thickness is predicted. Results show that the Immune Algorithm neural network model can predict variety rule of oil gas better and has higher accuracy than that of neural network.