林产化学与工业2026，Vol.46Issue(2)：21-30,10.DOI:10.20195/j.issn.0253-2417.2025073
基于NSGA-Ⅱ算法的煤粉-生物质掺混输送优化分析
Optimization Analysis of Coal Powder-Biomass Blending Transportation Based on NSGA-Ⅱ Algorithm
摘要
Abstract
Biomass-coupled pulverized coal as a fuel for power plants has broad application prospects,however,safety issues during pipeline transportation urgently need to be addressed.Taking the pipeline transportation of coal powder mixed with biomass powder as the research object,a numerical model was established by using the computational fluid dynamics(CFD)method,and the accuracy of the model was verified by comparing it with the experimental data in the literature.On this basis,four machine learning models,namely extreme gradient boost(XGBoost),random forest(RF),support vector regression(SVR),and multi-layer perceptron(MLP),were developed to fit and predict key parameters.Further combined with the NSGA-Ⅱ multi-objective optimization algorithm,the optimal operating conditions under multiple performance indicators were identified.The research results showed that the CFD numerical model had good accuracy,with an average relative error(MRE)of 13.2％and a root mean square error of pressure drop(RMSE)of 0.103 7.Among the four machine learning models,the XGBoost model performs the best,having the lowest RMSE on both the training set and the test set,and the coefficient of determination(R2)was higher than 0.98.The NSGA-Ⅱ multi-objective optimization algorithm had achieved the optimization of pipeline transportation conditions,and the optimal conditions obtained were as follows:the air velocity of pulverized coal was 18.61 m/s,the air velocity of biomass was 30 m/s,the inlet temperature of pulverized coal was 98.66 ℃,the inlet temperature of biomass was 45 ℃,and the proportion of biomass blending was 16％.Under these conditions,the average temperature rise of pulverized coal was-1.01 ℃,that of biomass was 32.24 ℃,the maximum temperature rise of pulverized coal was 4.31 ℃,the maximum temperature rise of biomass was 44.13 ℃,and the flow resistance was 10.57 kPa.关键词
机器学习/密相输送/XGBoost/NSGA-Ⅱ/CFDKey words
machine learning/dense-phase transport/XGBoost/NSGA-Ⅱ/CFD分类
化学化工引用本文复制引用
陈昕,罗博,程永新,汪晓,朱有健,韩勇,武潭..基于NSGA-Ⅱ算法的煤粉-生物质掺混输送优化分析[J].林产化学与工业,2026,46(2):21-30,10.基金项目
国家重点研发计划资助项目(2022YFB4202000) （2022YFB4202000）
