Closed-loop control
Perhaps the most important adjustment to the fuel injector pulse duration comes when the control is in the closed-loop mode. In the open-loop mode the accuracy of the fuel delivery is dependent on the accuracy of the measurements of the important variables. However, any physical system is susceptible to changes with either operating conditions (e.g., temperature) or with time (aging or wear of components).
In any closed-loop control system a measurement of the output variables is compared with the desired value for those variables. In the case of fuel control, the variables being regulated are exhaust gas concentrations of HC, CO, and NOx. Although direct measurement of these exhaust gases is not feasible in production automobiles, it is sufficient for fuel control purposes to measure the EGO concentration. These regulated gases can be optimally controlled with a stoichiometric mixture. The EGO sensor is, in essence, a switching sensor that changes output voltage abruptly as the input mixture crosses the stoichiometric mixture of 14.7.
The closed-loop mode can only be activated when the EGO (or HEGO) sensor is sufficiently warmed. That is, the output voltage of the sensor is high (approximately 1 volt) when the exhaust oxygen concentration is low (i.e., for a rich mixture relative to stoichiometry). The EGO sensor voltage is low (approximately 0.1 volt) whenever the exhaust oxygen concentration is high (i.e., for a mixture that is lean relative to stoichiometry).
The time-average EGO sensor output voltage provides the feedback signal for fuel control in the closed-loop mode. The instantaneous EGO sensor voltage fluctuates rapidly from high to low values, but the average value is a good indication of the mixture.
As explained earlier, fuel delivery is regulated by the engine control system by controlling the pulse duration (T ) for each fuel injector. The engine controller continuously adjusts the pulse duration for varying operating conditions and for operating parameters. A representative algorithm for fuel injector pulse duration for a given injector during the nth computation cycle, T(n), is given by:
where
Tb(n) is the base pulse width as determined from measurements of MAF rate and the desired air/fuel ratio
CL(n) is the closed-loop correction factor For open-loop operation, CL(n) equals 0; for closed-loop operation, CL is given by:
where I(n) is the integral part of the closed-loop correction P(n) is the proportional part of the closed-loop correction a and b are constants
These latter variables are determined from the output of the EGO sensor.
Whenever the EGO sensor indicates a rich mixture (i.e., EGO sensor voltage is high), then the integral term is reduced by the controller for the next cycle,
for a rich mixture.
Whenever the EGO sensor indicates a lean mixture (i.e., low output voltage), the controller increments I(n) for the next cycle,
for a lean mixture. The integral part of CL continues to increase or decrease in a limit-cycle operation.
The computation of the closed-loop correction factor continues at a rate determined within the controller. This rate is normally high enough to permit rapid adjustment of the fuel injector pulse width during rapid throttle changes at high engine speed. The period between successive computations is the computation cycle described above.
In addition to the integral component of the closedloop correction to pulse duration is the proportional term. This term, P(n), is proportional to the deviation of the average EGO sensor signal from its mid-range value (corresponding to stoichiometry). The combined terms change with computation cycle as depicted in fig below
In this figure the regions of lean and rich (relative to stoichiometry) are depicted. During relatively lean periods the closed-loop correction term increases for each computation cycle, whereas during relatively rich intervals this term decreases.
Once the computation of the closed-loop correction factor is completed, the value is stored in a specific memory location (RAM) in the controller. At the appropriate time for fuel injector activation (during the intake stroke), the instantaneous closed-loop correction factor is read from its location in RAM and an actual pulse of the corrected duration is generated by the engine control.