受成品汽油调和配方质量“事前”精准预测评价需求驱动,针对罐式调和生产中存在的批次效应,提出一种多模型自适应集成建模的配方质量评价方法.首先考虑调和主料因产地属性差异引起的批次效应,以及多项式回归模糊C均值(PRFCM)聚类算法在多属性数据分析应用中的局限,通过仅关注配方中关键属性变量之间的非线性关系,并以马氏距离替代欧式距离,得到改进的IPRFCM算法.基于该算法,离线阶段对历史数据进行批次划分,并优选NGBoost算法建立各批次子模型;在线阶段计算当前配方的批次隶属度,并作为权重系数融合各子模型得到IPRFCM-NGBoost集成模型,实现对调和配方质量的动态自适应预测.经工业数据实验验证表明,对于调和配方质量预测中的多属性问题,IPRFCM的批次划分结果更清晰,融合系数更精准,且计算成本更低;相较传统机器学习或深度学习模型,对于有限的配方样本,IPRFCM-NGBoost具有更优的预测精度与泛化能力.
A multi-model adaptive integrated modeling method was proposed to evaluate the quality of finished gasoline blend in order to avoid the batch effect. Firstly, the batch effect caused by the difference of origin attributes of main materials and the limitations of PRFCM clustering algorithm in the application of multi-attribute data analysis are considered. By focusing only on the nonlinear relationship between the key attribute variables in the formulation and replacing the Euclidean distance with the Mahalanobis distance, an improved IPRFCM algorithm is obtained. Based on this algorithm, the historical data are divided into batches in the offline stage, and the NGBoost algorithm is selected to build batch sub-models. The batch membership degree of the current formula is calculated in the online stage, and the IPRFCM-NGBoost integrated model is obtained by fusing each submodel as the weight coefficient, which realizes the dynamic adaptive prediction of the quality of the blended formula. The experimental results of industrial data show that IPRFCM has clearer batch division results, a more accurate fusion coefficient, and lower calculation cost for the multi-attribute problem in the quality prediction of blended formula. Compared with traditional machine learning or deep learning models, IPRFCM-NGBoost has better prediction accuracy and generalization ability, especially when formulation sample sizes are limited.
[1] 何仁初,张朝晖,杨明磊,等.考虑碳排放因素的汽油调合在线优化[J].化工学报,2023,74(2):818-829.
[2] Stetco A,Zeng X J,Keane J.Fuzzy C-means++:fuzzy C-means with effective seeding initialization[J].Expert Systems with Applications,2015,42(21):7541-7548.
[3] Dagher I.Fuzzy clustering using multiple Gaussian kernels with optimized-parameters[J].Fuzzy Optimization and Decision Making,2018,17(2):159-176.
[4] Pratiwi N B I,Saputro D R S,Widyaningsih P.Fuzzy c-shells for clustering of data’s LQ45 in Indonesia based on earnings per share and price earning ratio[J].Journal of Physics:Conference Series,2021,1776(1):012054.
[5] Hathaway R J,Bezdek J C.Switching regression models and fuzzy clustering[J].IEEE Transactions on Fuzzy Systems,1993,1(3):195-204.
[6] Pang Y,Shi M,Zhang L,et al.PR-FCM:a polynomial regression-based fuzzy C-means algorithm for attribute-associated data[J].Information Sciences,2022,585:209-231.
[7] Liu Z,Zhang L,Elkamel A,et al.Multiobjective feature selection approach to quantitative structure property relationship models for predicting the octane number of compounds found in gasoline[J].Energy & Fuels,2017,31(6):5828-5839.
[8] Abdolkarimi V,Sari A,Shokri S.Robust prediction and optimization of gasoline quality using data-driven adaptive modeling for a light naphtha isomerization reactor[J].Fuel,2022,328:125304.
[9] 蒋伟,佟国香.基于改进PCA-RFR算法的汽油辛烷值损失预测模型的构建与分析[J].石油学报(石油加工),2022,38(1):220-226.
[10] 赵林,李希,谢永芳,等.基于自适应变量加权的汽油辛烷值预测方法[J].控制与决策,2022,37(10):2738-2744.
[11] Abdul Jameel A G,Van Oudenhoven V,Emwas A H,et al.Predicting octane number using nuclear magnetic resonance spectroscopy and artificial neural networks[J].Energy & Fuels,2018,32(5):6309-6329.
[12] 李炜,郑明杰,李亚洁,等.融合可解释机器学习的成品汽油调和配方质量预测评价与致因分析[J].石油学报 (石油加工),2024,40(1):126-136.
[13] 刘聪,谢莉,杨慧中.基于改进DPC的青霉素发酵过程多模型软测量建模[J].化工学报,2021,72(3):1606-1615.
[14] Zhu H,Li Y,Wang L,et al.A multi-model ensemble learning framework for imbalanced Android malware detection[J].Expert Systems with Applications,2023,234:120952.
[15] Zhao Z,Guo Y,Dong B,et al.A clustering-based multi-model ensemble projection of near-term precipitation changes over East China and its uncertainty[J].Environmental Research Letters,2023,18(9):094050.
[16] Fu N,Lai Z,Zhang Y,et al.An effective method based on multi-model fusion for research octane number prediction[J].New Journal of Chemistry,2021,45(21):9668-9676.
[17] Tian Z,Chen H.Multi-step short-term wind speed prediction based on integrated multi-model fusion[J].Applied Energy,2021,298:117248.
[18] Duan T,Avati A,Ding Y D,et al.NGBoost:natural gradient boosting for probabilistic prediction[J].International Conference on Machine Learning,2020:2690-2700.
[19] Chen S Z,Feng D C,Wang W J,et al.Probabilistic machine-learning methods for performance prediction of structure and infrastructures through natural gradient boosting[J].Journal of Structural Engineering,2022,148(8):04022096.
[20] 郑玉巧,郎启发,施成龙,等.基于PCA-FSEM方法的风力发电机可靠性研究[J].兰州理工大学学报,2024,50(1):35-40.
[21] 征原,谢云.基于划分的聚类个数与初始中心的确定方法[J].计算机技术与发展,2017,27(7):76-78.
