Information processing system to calculate damage in oil in an electric motor

An information processing system comprising:

• a vehicle including
• a motor,
• a vehicle control device,
• a first sensor configured to acquire a motor temperature of the motor, and
• a second sensor configured to acquire an oil temperature of oil that cools the motor;
• an information processing device including a processor;
• and
• an information processing terminal, wherein
• the vehicle control device is configured to collect, over a predetermined period, original data including the motor temperature, a first duration of the motor temperature, the oil temperature, and a second duration of the oil temperature, and transmit the original data to the information processing device, and
• the processor is configured to
• calculate, based on the original data, a plurality of features including (i) a product of the motor temperature and the first duration of the motor temperature and (ii) a product of the oil temperature and the second duration of the oil temperature,
• perform clustering, the clustering being machine learning that groups data of each interval obtained by dividing the original data into intervals of a certain period into a predetermined number of clusters,
• calculate, for each of the features, a relative frequency distribution of the original data,
• set a plurality of time windows for clipping data of a partial period of the original data such that a sum of periods of all the time windows is shorter than the predetermined period and such that a difference between a proportion of each of the clusters in extracted data and a proportion of each of the clusters in the original data is equal to or less than a cluster threshold,
• clip data from the original data according to the time windows,
• calculate, for each of the features, the relative frequency distribution of extracted data obtained by combining all the data clipped according to the time windows,
• calculate an error between the relative frequency distribution of the original data and the relative frequency distribution of the extracted data,
• change settings of the plurality of time windows and repeat following steps (i) to (iii) until the error is equal to or less than a first threshold:
• (i) the clipping, (ii) the calculation of the relative frequency distribution of the extracted data, and (iii) the calculation of the error,
• in a case where the error is equal to or less than the first threshold, calculate, using the extracted data, an index value indicating a degree of deterioration of the oil, the degree of deterioration representing a ratio of deterioration relative to an allowable limit of damage accumulated in the oil;
• determine whether the index value is equal to or greater than a second threshold, and
• in a case where the index value is equal to or greater than the second threshold, determine that occurrence of a failure of the motor is predicted, and transmit, to the information processing terminal, a message indicating that occurrence of the failure has been predicted.