【摘要】 现代化设备的状态维护要求对设备的运行状态进行预测。随着运行设备的非线形特点越来越明显，传统的数学建模预报方法已不能满足设备的复杂化和现代化要求。针对现代状况，采用灰色理论和神经网络建模预报。对于灰色预报，将原始序列作累加生成处理后，采用ＧＭ（１，１）模型进行建模预测;对于神经网络预报，采用ＢＰ算法对时间序列进行预报。对从大庆油田采回的实际现场数据进行灰色理论和神经网络的联合预报。实践证明，提高了预报精度，取得了较好的预报效果。更多还原

【Abstract】 he condition prediction is needed for the maintenance of the equipment. The oldmethod of identification of machinery system is not practicable because the non-linear character is becoming more and more prominent. Gray System and Neural Network is adoptedhere to the method of prediction. The method is found to be in agreement with experimentalda ta.更多还原