Optimization of solenoid driver and controller for gaseous fuel high-pressure direct injector using model-based approach

• Main Author: Mohamad Hafidzul Rahman, Alias
• Format: Thesis
• Language: English
• Published: 2022
• Subjects: TA Engineering (General). Civil engineering (General); TJ Mechanical engineering and machinery; TL Motor vehicles. Aeronautics. Astronautics
• Online Access: http://umpir.ump.edu.my/id/eprint/35940/1/Optimization%20of%20solenoid%20driver%20and%20controller%20for%20gaseous%20fuel%20high-pressure%20direct%20injector.ir.pdf; http://umpir.ump.edu.my/id/eprint/35940/

This study focuses on the Direct Injection (DI) system utilizing Compressed Natural Gas (CNG) as the fuel. A conventional Gasoline Direct Injector (GDI) was converted into a gaseous fuel application. One of the issues that arises is the fluctuation in injector mass flow rate. The crucial factor leading to the occurring problem is a non-optimal injector driver and controller. Thus, the purpose of this study is to identify the most influential parameters of the injector, construct an analytical and data-driven model of the injector, conduct the model-based optimization of the injector and verify the optimal injector setup via simulation and experiment. A standalone injector test rig was used as the experimental setup. A parametric study was conducted using a one-dimensional (1D), first principle injector model builds in MATLAB Simulink. Data-driven modelling using a one-stage plan and an Interpolating Radial Basis Function (RBF) model was generated based on data collected from the injector simulation. An optimization study was conducted using Normal Boundary Intersection (NBI) algorithm in MATLAB Model-Based Calibration (MBC) Toolbox to produce an optimal injector setup. Finally, a verification study was performed using the attained optimal injector setup in both experiment and simulation of the injector. Based on the results, the experimental result shows a similar injector mass flow rate trend compared to the theoretical calculation except for the mass flow rate fluctuation point. The most influential injector parameter is the nozzle diameter with a sensitivity value of 1489.71 g/s/m, while the least significant injector parameter is the spring constant with a sensitivity value of 0.000083 g/s/N/m. Data-driven modelling produced an RMSE of 0 and a validation RMSE of 0.0249. The simulation result of the mass flow rate for baseline versus optimization shows an increment of 15.64% compared to the experimental result for baseline versus optimization, which shows an increase of 35.79%. The results obtained from the study are important to increase the effectiveness of control strategies embedded in the development of a dedicated driver and controller for the gaseous fuel direct injector.