Interactive trouble condition sign mining for hydroelectric power plants

Kyushu Electric Power Co.,Inc. collects different sensor data and weather information (hereafter, operation data) to maintain the safety of hydroelectric power plants while the plants are running. It is very rare that trouble conditions occur in the power equipment. And in order to collect the trouble condition data, it is hard to construct an experimental power generation plant and make trouble conditions in this plant. The cost is too high to construct it. In this situation, we have to find trouble condition sign as a risk management. In this paper, we consider that the rise inclination of special unusual condition data gives trouble condition sign. And we propose an interactive trouble condition sign discovery method for hydroelectric power plants by using a one class support vector machine and a normal support vector machine. This paper also shows that the proposed method can find a trouble condition sign of bearing vibration from the real operation data.