oa A New Control Strategy for Air/Fuel Regulation of Spark Ignition Engines to Maximize Energy Economy and Minimize Environment Pollution

The gasoline for cars and light trucks accounts for more than 40% of the total oil consumption worldwide. More important is the environment subjected to the pollution emitted by spark ignition engine's tailpipe. Spark ignition engines are commonly equipped with electronic control systems whose task is to provide desired air-fuel ratio (AFR) signal tracking to improve fuel energy economy and reduce exhaust emissions. The engine control system maintains the AFR to be close to the stoichiometric value as an index of maximum catalytic convertor efficiency. Unfortunately, performance of the catalytic convertor significantly depends on the precise value of the AFR. For instance, exceeding the stoichiometric value by 1% results in about 50% higher NOx emissions while receding the stoichiometric value by 1% drastically increases CO and HC pollutants. In addition to the emission concerns, regulated AFR according to the stoichiometric value can improve the fuel economy and provide efficient torque demands. However, the wide engine operating range, the inherent nonlinearities of the combustion process, the large modelling uncertainties and parameter variations along with the time-varying delay in the spark ignition engines make the design of the control system a challenging task.

In this paper, we present a new synthesis method to control air-fuel ratio (AFR) in spark ignition engines to maximize the fuel energy economy while minimizing environment pollutants (exhaust emissions). In this paper the time-varying delay is rendered into non-minimum phase characteristics with time-varying parameters. Application of parameter-varying dynamic compensators is invoked to retrieve unstable internal dynamics. Associated error dynamics is then utilized to construct a parameter-varying proportional-integral-derivative (PID) controller combined with a parameter-varying dynamic compensator to track the desired AFR command using the feedback from the universal exhaust gas oxygen sensor. The proposed method achieves desired dynamic properties independent of the matched and unmatched disturbances due to the dynamic compensator features. Results of applying the proposed method to experimental data on a Ford truck F-150 with a V8 4.6L engine demonstrate the closed-loop system excellent stability and performance against time-varying delay, canister purge disturbances and measurement noise for both port fuel injection engines and lean-burn engines.