In this work, the cycle-mean-value approach to marine engine process simulation is considered. At first, a basic principles model is employed where the engine crankshaft and turbocharger shaft speeds are obtained by integration of the angular momentum conservation differential equations. Any other engine variable can then be obtained by solving a set of nonlinearly-coupled, algebraic equations, corresponding to energy and mass conservation through the engine. Nonlinear data analysis is then performed on this process model. By approximating the torque maps, generated by the thermodynamic basic principles model, with neural nets, explicit functional relationships are obtained. Identification of the power-plant operating regimes through linearization and decomposition is finally performed. In effect, a supervisory power-plant controller structure, applicable to real-time control and diagnostics, is proposed, incorporating the nonlinear state-space description of the plant.

This content is only available via PDF.
You do not currently have access to this content.