Many products or intermediate products in industry have to be cleaned with industrial washers. For example in the metal processing industry, especially in the automotive industry, parts have to be cleaned from drilling chips and oil. For this a lot of water and chemical add ons are used. It is in the economical as well as the ecological interest of companies to use the washer liquid in an optimal way. On the one hand the washer liquid should not be changed too early, on the other hand high quality requirements in concerning the cleanness of the products have to be fulfilled. In the past, the change of the washer liquid was conducted to more or less subjective criteria. The aim of this work is the development of a sensor system, which is able to detect the optimal point in time to change the washer liquid. Depending on the application the particle pollution or oil pollution are of a higher interest. Therefore, it is necessary to measure these two kinds of pollutions independent from each other. This is realised with the help of an artificial neural network.