基于预测模型的柴油机排气热管理标定方法研究  

Research on Calibration Method of Diesel Engine Exhaust Heat Management Based on Prediction Model

作　　者：王迎迎 裴玉姣 陈静 Wang Ying-ying;Pei Yu-jiao;Chen Jing(Weichai Power Company Ltd,Shandong Weifang 261061)

机构地区：[1]潍柴动力股份有限公司,山东潍坊261061

出　　处：《内燃机与配件》2024年第12期1-4,共4页Internal Combustion Engine & Parts

摘　　要：基于一款六缸重型柴油机,根据台架试验数据中排气温度确定热管理区域,采用空间填充法进行试验设计,并根据设计方法进行台架试验,搭建基于高斯过程回归的预测模型,以发动机NOx和Soot排放为约束条件,以排温为目标求解出燃油消耗率最低的控制参数,生成控制参数的MAP用于后期台架试验。通过台架试验结果在热管理区域显示,NOx和Soot排放满足要求,油耗升高率在合理范围内,在升温模式下稳态试验排气平均温度提升40℃左右,瞬态试验排气平均温度提升30℃左右,且WHTC瞬态循环时间缩短。Based on a six-cylinder heavy-duty diesel engine,the thermal management area is determined from the discharge temperature in the table frame test data,the experimental design is carried out using space filling method,and the bench test is carried out according to the design method to build a predictive model based on Gaussian process regression,The engine NOx and boot emissions are the binding conditions,and the control parameters with the lowest fuel consumption are solved with the goal of discharge temperature,and the map of the control parameters is generated for later stage tests.The results of the bench test show in the heat management area that NOx and Soot emissions meet the requirements,the increase in fuel consumption is within a reasonable range,the average temperature of the steady state test exhaust gas in the warming mode is raised by around 40℃,and the average temperature of the transient test exhaust gas is raised by about 30℃,and WHTC transient cycle times are reduced.

关 键 词：柴油机 神经网络 排气热管理 电控标定 

分 类 号：TK421.5[动力工程及工程热物理—动力机械及工程]