论文部分内容阅读
回转干燥窑由于干燥速率难以在线测量,其干燥速率模型的建立一直是一大难题。在分析干燥速率建模的基础上,提出将最小二乘支持向量机运用到干燥速率建模,及其基于免疫-果蝇优化算法的最小二乘支持向量机回归参数优化方法(IAFOALSSVR)。首先利用预处理的干燥过程数据进行模型的训练,利用免疫-果蝇算法对模型参数进行寻优,然后获得最优参数并建立最优模型,通过使用该改进方法建立干燥速率模型与其他算法优化的模型进行对比,结果表明该优化方式在干燥速率建模精度上与其它智能算法相当,在计算效率上要优于其它算法。
Rotary kilns are difficult to measure on-line because of the drying rate. The establishment of a drying rate model has always been a major challenge. Based on the analysis of the drying rate modeling, a least-square support vector machine (LS-SVM) is proposed to model the drying rate and the optimization method based on immune-fruit fly optimization algorithm for regression parameters of least squares support vector machines (IAFOALSSVR) is proposed. Firstly, the model was trained by using the pretreatment drying process data, and the parameters of the model were optimized by immune-fruit fly algorithm. Then the optimal parameters were obtained and the optimal model was established. By using this improved method, the drying rate model and other algorithms were optimized The results show that this optimization method is comparable to other intelligent algorithms in modeling the drying rate and is superior to other algorithms in computational efficiency.