---

A tri-generation cycle consisting of a homogeneous charge compression ignition (HCCI) engine and an ammonia-water absorption cogeneration cycle are proposed and analyzed. The energy of engine exhaust gases are utilized to run absorption cogeneration cycle. Also the energy of cooling water can be used in residential applications. A single zone model with capability to consider chemical kinetic talculations is developed for the HCCI engine. The results show that increasing the pump pressure ratio of the cogeneration cycle causes a decrease in the refrigeration output and an increase in first law efficiency. At a particular value of this pressure ratio the second law efficiency is maximized. It is shown that the contribution of engine in the total exergy destruction in the tri-generation system is much higher than those of the other components. With an ammonia concentration of 0.4 in the solution leaving the absorber and with an ambient temperature of 25oC, the maximum exergy efficiency occurs when the pump pressure ratio is 9.486. At this condition, the fuel energy saving ratio and CO2 emission reduction are 27.97% and 4.8%, respectively. It is also shown that the second law efficiency of the tri-generation system is 5.4% higher than the second law efficiency of the HCCI engine.

---

In this paper, an optimization-based nonlinear control strategy is applied to air path control of a turbocharged diesel engine. For this aim, the air-fuel ratio (AFR) and the pressure of exhaust manifold are controlled by calculating the air mass flow rates of turbocharger and exhaust gas recirculation. Controlling AFR which affects engine power, fuel consumption and exhaust emissions, is carried out by calculating the air mass flow rate with the assumption of known fuel path. For air path modelling, the mean value model which is a suitable method with low computational time is used to achieve the air path equations. Air mass flow is calculated by the developed control laws and applied by the turbocharger and exhaust gas recirculation. In the proposed control method, the nonlinear system response is firstly predicted by Taylor series expansion and then the optimal control law is developed by minimizing the difference between the desired response and the actual response. To compare the performance of the proposed optimal controller, a sliding mode controller has been also designed. The simulation results show that the rate of air mass and the pressure of exhaust manifold are close to their desired values and consequently the AFR is well controlled. Therefore, the designed controller with optimal inputs can successfully cope with the nonlinearities existing in engine dynamics model.