Intelligent control of the refining process for improved fiber quality

In the wet-processed fiberboard production, where virtually no binders are used, the refiner control plays a crucial role in ensuring the high quality of the fiberboard. In order to improve the refiner control, it is necessary to estimate quality parameters which are not directly available, and to determine the process set points leading to minimum production cost for the target quality. A Takagi-Sugeno-Fuzzy model for the refining process is generated, based on experts´ knowledge as well as gathered data. This nonlinear model is, on the one hand, implemented as an online soft sensor, and on the other hand it is used, together with a cost function, for process optimization. This enables the plant operator to continuously observe the fiberboard quality and to calculate new set points when the target quality changes, while keeping production cost at the minimum.