A Comparative Study to Detect Flowmeter Deviations Using One-Class Classifiers

The use of bicomponent materials has encouraged the proliferation of wind turbine blades to produce electric power. However, the high complexity of the process followed to obtain this kind of materials difficult the problem of detecting anomalous situations in the plant, due to sensors or actuators malfunctions. This work analyses the use of different one-class techniques to detect deviations in one flowmeter located in a bicomponent mixing machine installation. In this case, a comparative analysis is carried out by modifying the percentage deviation of the sensor measurements.