Abnormal noise evaluation system and abnormal noise evaluation method
Abstract:
In an abnormal noise evaluation system, running noise of a vehicle is acquired, and the running noise is analyzed to generate analysis data. A rotational order component of the running noise of the vehicle is then extracted from the analysis data, and features for each rotational order of the running noise of the vehicle is generated based on the extracted rotational order component. Subsequently, a learning model is generated using as training data a combination of the features for each rotational order generated for a learning object that is of the same type as an object to be evaluated and an evaluation result given in advance to the learning object. Whether there is abnormal noise from the object to be evaluated is evaluated by applying the features for each rotational order generated for the object to be evaluated to the learning model, and an evaluation result is output.
Information query
Patent Agency Ranking
0/0