Open Access
Subscription Access
Applying Capacitance/Inductance Measurements for Characterizing Oil Debris and pH
Lubricating oil is important in internal-combustion engines. The present study investigates about constructing capacitor for measurement of pH and debris in oil. In this paper, Adaptive Neuro Fuzzy Inference System (ANFIS) and Artificial Neural Networks (ANN) for predicting oil debris and pH are used. For measurement of pH and oil debris, two capacitor were constructed using two aluminum and two copper plate electrodes have a distance of 20 mm, with each other respectively. The best ANN model for predicting pH and oil debris was found as two hidden layers network with 1-5-1-1 structure. RMSE-test, RMSE-train and time for the best ANN model were reported as 0.17, 0.21 and 853 ms, and 0.18, 0.18, 853 ms respectively. The corresponding RMSE-test, RMSE-train and time values for the best ANFIS topology for predicting pH and oil debries were 0.16, 0.68, 17 and 6.7, 22.4, 23 respectively. It is concluded that ANN with lower RMSE-test was better than ANFIS for predicting.
Keywords
Aluminum Plate Electrodes, Artificial Neural Networks, Adaptive Neuro Fuzzy Inference System, Copper Plate Electrodes, Lubricating.
User
Information
Abstract Views: 139
PDF Views: 0