Control-oriented modeling of hydrostatic power-split CVTs using Takagi-Sugeno fuzzy models

Power-split continuously variable transmissions (CVTs) represent a promising technology to improve the fuel economy of high-power o-road vehicles such as construction and agricultural machines. There are capable of keeping the combustion engine running within the optimal range in terms of fuel economy and performance. Power-split CVTs are a very specific type of CVT because this kind of transmissions are characterized by the combination of a traditional mechanical transmission and a continuously-variable transmission. A CVT powersplit design for high-power vehicles combines the advantages of pure hydrostatic drives at low speeds with the high eciency of power split drives at higher speeds. In this paper, a general control-oriented modeling approach based on Takagi-Sugeno (T-S) fuzzy models for the design of fuzzy observers for sensor-fault detection in power-split continuously CVTs is developed. It has been shown by simulation studies and experimental results that this approach can be used to reconstruct a selection of measurable values such as the hydrostatic pressure under varying load conditions and switched transmission modes.