Performance prediction device

Abstract

To provide a performance prediction device capable of sufficiently predicting performance of a prediction object comprising a member including a plurality of types of materials.SOLUTION: A performance prediction device predicts performance of a prediction object comprising a member including a plurality of types of materials. The performance prediction device includes a prediction section for predicting, by using a prediction model, performance of the prediction object by acquiring information which expresses performance from information having data of a synthetic XRD spectrum of the plurality of types of materials. The prediction model is a learning model which is constructed by performing machine learning with the information having the data of the synthetic XRD spectrum of the materials as explanatory variables and with the information expressing the performance as target variables.SELECTED DRAWING: Figure 1

Claims

[Claim 1] A performance prediction device that predicts the performance of an object to be predicted, comprising components containing multiple types of materials, The system includes a prediction unit that predicts the performance of the target object by using a prediction model to obtain information that represents the performance from information having data of the combined XRD spectra of the multiple types of materials, The composite XRD spectrum of the aforementioned multiple types of materials is obtained by multiplying the intensity at each diffraction angle of the XRD spectrum of each type of material by the composition ratio of that type of material, thereby calculating multiple intensity data points at each diffraction angle that reflect the composition ratio of that type of material, and then summing these multiple data points. The performance prediction device is characterized in that the prediction model is a learning model constructed by performing machine learning with information having data of the synthetic XRD spectra of the multiple types of materials as explanatory variables and information representing the performance as the objective variable.

Images