论文部分内容阅读
水泥强度是衡量水泥质量的重要指标之一。水泥生产过程中为控制其质量,需要尽早地预测产品的抗压强度,若出现数据异常能及早反馈于生产,减少经济损失。强度预测是一个多变量、非线性、大时滞问题;影响水泥的理化性能检测值与28D强度值虽然确定,但至今没有具体的函数表达式将其描述。传统的线性回归分析预测法[1],把水泥强度与各检测值之间的高度非线性关系简化为线性函数关系,忽视了水泥强度复杂的非线性和模糊性,预测结果的准确性很不理想。如果考虑的因素过多,会使回归分析的计算量过大,适用性和精度也会随之下降,难以在水泥生产中推广应用。因此,有必要利用新的预测方法进行强度的预测。本文针对以上问题,提出了预测水泥强度的人工神经网络(ANN)模型,并与多元回归对比分析表明,ANN建模有较高预测精度,为水泥生产的质量管理提供一定的技术支持,并对降低销售成本,提高经济效益,保护环境和低碳生产的实现具有重大的现实意义。
Cement strength is one of the important indicators to measure the quality of cement. In order to control its quality during the cement production process, it is required to predict the compressive strength of the product as soon as possible. If there is any abnormal data, it can be fed back to the production as early as possible to reduce the economic loss. The prediction of strength is a multivariable, nonlinear and time-lag problem. Although the physical and chemical properties and the 28D intensity of the cement have been determined, no concrete functional expression has so far been described. The traditional linear regression analysis and prediction method [1] simplifies the highly nonlinear relationship between the cement strength and each detection value to a linear function, neglects the complex nonlinearity and fuzziness of the cement strength, and the accuracy of the prediction result is very poor ideal. If too many factors are considered, the calculation of regression analysis will be too large, the applicability and accuracy will also decline, and it is difficult to popularize and apply in cement production. Therefore, it is necessary to use the new prediction method to predict the intensity. In order to solve the above problems, an artificial neural network (ANN) model for predicting the strength of cement is proposed in this paper. Compared with multivariate regression analysis, it shows that ANN model has higher prediction accuracy and provides some technical support for quality control of cement production. Reduce sales costs, improve economic efficiency, protect the environment and the realization of low-carbon production of great practical significance.