Fast neural networks for diesel engine control design

摘要: Abstract Advanced engine control systems require accurate dynamic models of the combustion process, which are substantially nonlinear. This contribution presents application fast neural net for design purposes. After briefly introducing a special local linear radial basis function network (LOLIMOT) process building adequate is discussed in detail. These neuro-models then integrated into an upper-level emission optimization tool calculates cost exhaust versus consumption/torque and determines optimal settings. A DSP-based computer system allows at test stand.