Fuzzy logic estimator for dynamic weighing system

In the area of mass production, products are weighed using load cell based dynamic weighing systems. A load cell is an uncontrollable weighing device and the value of weight, for the passing product, is estimated by filtering the electrical signal from a load cell. Improvement in filtering increases the speed of weighing and enhances the measurement accuracy. In this paper a fuzzy logic estimator is proposed as weight filter for the dynamic weighing system. The structure of a fuzzy model is identified by the concept of fuzzy space clustering. The input-output space is clustered using a fuzzy C-means clustering scheme. Afterwards, the components of the centroid vectors of each cluster are moved onto the axes of their own coordinates and then the membership functions are formed on the basis that the triangle base is twice the standard deviation of the data set. In this way the fuzzy rules have been chosen in an optimal manner. The results achieved show that unconventional method of filtering fuzzy logic estimator may provide computationally effective alternative to the conventional method