{"title":"Machine Learning-Based Modeling and Predictive Control of Combustion\u0000 Phasing and Load in a Dual-Fuel Low-Temperature Combustion\u0000 Engine","authors":"Mohit Punasiya, A. Sarangi","doi":"10.4271/03-17-04-0030","DOIUrl":"https://doi.org/10.4271/03-17-04-0030","url":null,"abstract":"Reactivity-controlled compression ignition (RCCI) engine is an innovative\u0000 dual-fuel strategy, which uses two fuels with different reactivity and physical\u0000 properties to achieve low-temperature combustion, resulting in reduced emissions\u0000 of oxides of nitrogen (NOx), particulate matter, and improved fuel\u0000 efficiency at part-load engine operating conditions compared to conventional\u0000 diesel engines. However, RCCI operation at high loads poses challenges due to\u0000 the premixed nature of RCCI combustion. Furthermore, precise controls of\u0000 indicated mean effective pressure (IMEP) and CA50 combustion phasing (crank\u0000 angle corresponding to 50% of cumulative heat release) are crucial for\u0000 drivability, fuel conversion efficiency, and combustion stability of an RCCI\u0000 engine. Real-time manipulation of fuel injection timing and premix ratio (PR)\u0000 can maintain optimal combustion conditions to track the desired load and\u0000 combustion phasing while keeping maximum pressure rise rate (MPRR) within\u0000 acceptable limits.\u0000\u0000 \u0000In this study, a model-based controller was developed to track CA50 and IMEP\u0000 accurately while limiting MPRR below a specified threshold in an RCCI engine.\u0000 The research workflow involved development of an imitative dynamic RCCI engine\u0000 model using a data-driven approach, which provided reliable measured state\u0000 feedback during closed-loop simulations. The model exhibited high prediction\u0000 accuracy, with an R2 score exceeding 0.91 for all\u0000 the features of interest. A linear parameter-varying state space (LPV-SS) model\u0000 based on least squares support vector machines (LS-SVM) was developed and\u0000 integrated into the model predictive controller (MPC). The controller parameters\u0000 were optimized using genetic algorithm and closed-loop simulations were\u0000 performed to assess the MPC’s performance. The results demonstrated the\u0000 controller’s effectiveness in tracking CA50 and IMEP, with mean average errors\u0000 (MAE) of 0.89 crank angle degree (CAD) and 46 kPa and Mean absolute percentage\u0000 error (MAPE) of 9.7% and 7.1%, respectively, while effectively limiting MPRR\u0000 below of 10 bar/CAD. This comprehensive evaluation showcased the efficacy of the\u0000 model-based control approach in tracking CA50 and IMEP while constraining MPRR\u0000 in the dual-fuel engine.","PeriodicalId":47948,"journal":{"name":"SAE International Journal of Engines","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139526196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}