Regulating air-fuel balance in combustion engines using adaptive learning in neural network

The paper proposes a monitoring system, which helps to regulate the proportion of air and fuel in vehicles to increase combustion engine efficiency and to reduce the amount of harmful gases thus emitted. A human operator generally performs the task of opening or closing the valve of fuel injection pump of locomotive engines to control amount of air going inside the combustion chamber. An artificial neural network has been trained to recognize the standard mix of fuel and air components in combustion process. The CMAC network has been trained using supervised learning for around 40 fuels and about 12 were used for testing the network performance. Mean square error calculation ensures the effectiveness of the network´s performance. Hence the network simulates human behavior, to achieve perfection and skill through AI. The paper presents a new approach for regulating air-fuel balance using neural network that is capable of autonomously learning the amount of air required to burn completely a given amount of fuel through the use of adaptive learning method. The approach is found to be extremely effective in identifying imbalance in combustion, detecting previously learnt air-fuel combination, and autonomously improving its performance over time by self-learning.