当前位置: X-MOL 学术Ind. Eng. Chem. Res. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Fast Locally Weighted PLS Modeling for Large-Scale Industrial Processes
Industrial & Engineering Chemistry Research ( IF 3.8 ) Pub Date : 2020-11-11 , DOI: 10.1021/acs.iecr.0c03932
Xinmin Zhang 1 , Chihang Wei 1 , Zhihuan Song 1
Affiliation  

Locally weighted partial least-squares (LW-PLS) is an efficient just-in-time (JIT) modeling method, which can handle process collinearity, nonlinearity, and time-varying characteristics. However, it is not suitable for modeling large-scale industrial processes due to its huge computational cost. To solve this issue, the present work proposes a novel fast LW-PLS (FLW-PLS) method that can handle large-scale process data well. FLW-PLS is designed based on the exact Euclidean locality-sensitive hashing algorithm. Unlike standard LW-PLS which implements a brute-force linear search of similar samples over a reference data set, FLW-PLS searches for similar samples in sublinear time. Thus, significant computational savings can be obtained by FLW-PLS. The effectiveness of the proposed FLW-PLS method was validated through the case studies of predicting the silicon content and phosphorus content in the ironmaking process. The application results have shown that when faced with large-scale data, FLW-PLS can significantly reduce the prediction time without significantly reducing the prediction accuracy in comparison to the conventional LW-PLS method.

中文翻译:

大规模工业过程的快速局部加权PLS建模

局部加权偏最小二乘(LW-PLS)是一种有效的实时(JIT)建模方法,可以处理过程共线性,非线性和时变特性。然而,由于其巨大的计算成本,它不适用于大型工业过程的建模。为了解决这个问题,目前的工作提出了一种新颖的快速LW-PLS(FLW-PLS)方法,该方法可以很好地处理大规模过程数据。FLW-PLS是基于精确的欧氏局部敏感哈希算法设计的。与标准LW-PLS在参考数据集上实现相似样本的强力线性搜索不同,FLW-PLS在亚线性时间内搜索相似样本。因此,FLW-PLS可以节省大量计算资源。通过预测炼铁过程中硅含量和磷含量的案例研究,验证了所提出的FLW-PLS方法的有效性。应用结果表明,与传统的LW-PLS方法相比,当面对大规模数据时,FLW-PLS可以显着减少预测时间,而不会显着降低预测精度。
更新日期:2020-11-25
down
wechat
bug