Most breakdowns, approximately 89%, are unrelated to the lifespan of the equipment. For effective plant operation, early warning of issues not related to the age of the equipment is necessary. HanPHI enables a plant to perform predictive maintenance through the identification of breakdown symptoms in their early stage.
HanPHI models are built using robust mathematical algorithms and several recursive calculations are performed to verify the calculated results. The accuracy of the predicted result approaches 100% accuracy.
Yes. Every data point is scanned at buffered at very high speed. Each and every piece of data is timestamped with the quality. There is no loss of data even though several hundred thousand events occur at the same time.
HanPHI can read and analyze the historical normal conditions of all plant equipment to predict the expected normal condition of all plant equipment. HanPHI measures the difference between the real-time and expected normal operating condition to calculate the health index in terms of a percentage value. A higher health index represents a better health status of plant equipment.
It is ideal to construct the learning model by reflecting the normal plant operation for one year of data that takes seasonal change and equipment lifespan into account. However, HanPHI is able to generate a model adjusted to the purpose of a user by setting the learning period and range for as little as one day.