Local Linear Models adaptation for a 4 Inj — 2PP common-rail pressure system

The implementation of a Neuro-Fuzzy nonlinear adaptive structure, with Local Linear Models (LLM), designed for fuel pressure estimation in diesel common-rail (CR) hydraulic system, represents the main topic. Hydraulic systems, in general, are nonlinear and engineers have often struggled to find the best solution to approximate the input-output dependencies. Powerful tools are necessary for splitting the input space in smaller pieces where linear approximations can be considered satisfactory. Neuro-Fuzzy networks, combined with LLM represent the best solution in this case. Using appropriate numerical models, these architectures can be implemented in a real-time environment designed for on-line adaptation of the linear models parameters. The paper demonstrates that the LLMs, and hence, the whole dynamic models parameters of the CR´s NeuroFuzzy developed architecture, can be adapted in an on-line environment. The practical results are favorable.