Predictive Failure Analysis of Spindle Motor & Cutting Oil Condition Monitoring of Grinding Machine using Artificial Intelligence Models

With the advent of Industry 4.0, manufacturing industries are competing to adopt intelligent machining systems to deliver high performance in minimum possible time. Machine downtimes results in hefty production loss(s). To avoid them, we must identify their causes (spindle failure in our case) and predict their occurrences. We used vibration analysis methods to analyze wear-&-tear of spindle during running condition. Our objective is to design a low-cost smart maintenance model to predict the failure before its occurrence using Support Vector Machine model. Additionally, condition-monitoring of cutting fluid by analyzing variation in temperature of rotating spindle, pH due to Sulphur concentration has been included. Learning Vector Quantization algorithm is used for data analysis of sensors monitoring the health of cutting fluids and filter oil. Time Series Analysis has been performed on pH data to predict the pH of cutting fluid for prevention of cutting fluid deterioration. The model is tested for a good range of working conditions and results are found more promising than existing system.