IIT Mandi researchers develop algorithms to predict the functioning of vehicular Internal Combustion Engines
Mandi: Indian Institute of Technology Mandi researchers, along with Robert Bosch Engineering and Business Solutions Private Limited, Bangalore, have developed algorithms to predict the functioning of vehicular Internal Combustion (IC) engines so that their operation can be optimized for maximum fuel efficiency and minimum emissions.
A research team lead by Dr. Tushar Jain, Assistant Professor, School of Computing and Electrical Engineering, IIT Mandi, has published this research in the International Journal of Systems Science, Taylor & Francis. The paper is co-authored by Dr. Jain and his research scholar, Ms. Vyoma Singh, along with Dr. Birupaksha Pal from Robert Bosch Engineering and Business Solutions Private Limited, Bangalore.
The IC engine that is fuelled by petrol and diesel powers about 99.8% of global transport and, in doing so, generates about 10% of the world’s greenhouse gas (GHG) emissions. While alternatives including battery electric vehicles (BEVs) and other fuels like biofuels and hydrogen are slowly gaining ground, as of now, they are often used in conjunction with conventional IC engines. It is therefore imperative that IC engines designs are optimized in order to ensure the best fuel economy and minimal emissions over the entire lifespan of the engine.
“At any point of time, the working condition of the engine and other devices/systems inside the vehicle should be precisely known, for which, we need the information on several important engine parameters,” said Dr, Jain. If the information of all the relevant parameters were known, then by continuous monitoring and computation of these parameters, the driver could use the usual driving manoeuvres such as changing the gear appropriately to improve the vehicle’s performance.
From the technical viewpoint, designing the optimum performing engine depends on the precise knowledge of the system states and the engine parameters; for example, in petrol engines, an air-fuel ratio (AFR) value of 14.67 translates to complete combustion of the fuel, and thus minimal emissions and maximum power. While new vehicles out of the assembly line meet many of the requirements, as they age, the operational parameters change, and the vehicle’s operation becomes less than optimal.
“Due to the high frequency moving parts and operating conditions of the engine, it is difficult to place or install the sensors that are available in the market to measure all the key parameters continuously. We have developed a new algorithm for their online estimation, which will be used to develop advanced, sophisticated controllers for better engine performance” explains the lead researcher.
The proposed algorithm is based on the unscented Kalman filter and recursive least-squares mathematical techniques to accurately estimate the engine dynamics and parameters. The researchers have benchmarked the performance of their methodology by comparing it with that of the state-of-the-art estimation methods. The numerical stability and robustness of their proposed methodology are analyzed through rigorous Monte Carlo simulations and found to be superior to other methods.
The researchers have estimated the spark-ignition engine dynamics, namely the intake manifold pressure, engine speed, and the airflow rate past the throttle, along with the estimation of the engine parameters that determine the said dynamics accurately. The developed algorithm can be programmed and be a part of the Electronic Control Unit (ECU) installed in the vehicles.
The algorithm developed by the IIT Mandi team will help in on-board monitoring and control for IC engines. The application of the developed algorithm can be extended to determine other variables such as the State-of-Charge (SoC) in battery-operated vehicles in real-time as well.