Abnormality determination device

An abnormality determination device mounted on a vehicle provided with a transmission configured to transmit power by rotation of a shaft, the abnormality determination device comprising:

• a memory configured to store mapping data that is data that defines mapping learned by machine learning, the mapping being a mapping of which an input variable is a variable indicating time-series data of a rotation speed of the shaft and of which an output variable is a plurality of state variables indicating a state of the shaft,
• the state of the shaft including a first state where the shaft is in a normal state, a second state where the shaft is in a divergent vibration state, a third state where the shaft is in a torsional vibration state, and a fourth state is in a bending vibration state, and
• the state variables including four state variables indicating that the state of the shaft is the first state, the second state, the third state, and the fourth state, respectively;
• a display configured to notify a state of the transmission;
• and
• a processor that is connected to and controls an internal combustion engine, a motor generator and an inverter of the vehicle, wherein
• the processor is configured to execute
• an acquisition process of acquiring the variable indicating the time-series data and using the variable as a value of the input variable,
• an input process of inputting the input variable to the mapping,
• a determination process of determining whether an abnormality has occurred in the transmission by determining the state of the shaft based on a maximum value of the four state variables output from the mapping,
• a storage process of storing the results of the determining the state of the shaft in the memory, and
• a notification process of operating the display based on the state of the shaft stored in the memory.