论文部分内容阅读
针对钢铁企业煤气系统中转炉煤气柜柜位难以建立模型进行预测的问题,结合概率神经网络、HP(Hodric-Prescott)滤波、Elman神经网络各自的性质建立了PNN-HP(2)-ENN模型,用于对转炉煤气柜柜位进行分类预测.将模型应用在企业实际数据中,实验结果表明,所建模型分类准确、耗时少、预测效果良好.与其它常用模型相比,此模型适合转炉煤气柜柜位的预测,能够为副产煤气的合理调度提供操作依据.
Aiming at the problem that it is difficult to establish a model for converter gas tank counter in steel enterprise gas system, PNN-HP (2) -ENN model is established based on the properties of probabilistic neural network, Hodric-Prescott filter and Elman neural network. The model is applied to the actual data of the enterprise.The experimental results show that the model is accurate and time-consuming classification, and the forecasting effect is good.Compared with other common models, this model is suitable for converter The forecast of gas cabinet counter can provide the operation basis for the reasonable dispatch of by-product gas.