Neural and Learning Approaches for Automotive Engine Control and Residential Energy System Optimization
2012-12-09T00:00:00Z (GMT) by
Due to the tightened legislation on emissions, fuel economy, and diagnostic standards in spark ignition internal combustion engines from governments, the automotive industry is striving to minimize the emission and, at the same time, to achieve better fuel economy and vehicle driveability. Using the actual data from a test vehicle with a V8 engine, specific neural networks are trained offline to simulate engine torque dynamics and exhaust air-fuel ratio dynamics for the purpose of the identification of controllers. Validation results demonstrate that neural network models simulate the engine processes with a high degree of accuracy. The goals of the present learning control design for automotive engines include improved performance, reduced emissions and maintained optimum performance under various operating conditions. More specifically, The goal of engine torque control is to track the commanded torque. The objective of the air-fuel ratio control is to regulate the engine air-fuel ratio at specified setpoints. We applied three different learning control algorithms to solve these problems: adaptive critic learning technique, neural sliding mode control technique and biological nonlinear adaptive control technique. Simulation results for the three control techniques indicate excellent tracking control has been achieved with proper control actions. Intelligent energy management systems can help to minimize energy costs for the residential customers and reduce emissions by efficiently using renewable energy resources and distributed energy storage systems. We applied a self-learning scheme for the control and management of residential energy system. Simulation results confirmed that our proposed learning scheme can greatly benefit the residential customers with the minimum electricity charge. This technique has the potential to revolutionize the residential energy management as it reduces and shifts demand automatically, provides valuable insights to customers which ultimately save the environment by reducing the carbon footprint of power companies.