Method and device for determining hidden variables of Gaussian mixture model, equipment and medium

Abstract

The embodiment of the invention provides a hidden variable determination method and device of a Gaussian mixture model, equipment and a medium. The method comprises the following steps: S1, setting an initial value of a hidden variable based on a training sample set of a Gaussian mixture model; s2, determining a sample weight matrix corresponding to the training sample set according to the number of samples corresponding to each working condition in the training sample set; s3, determining a responsivity matrix corresponding to the iteration process according to the sample weight matrix and the current value of the hidden variable; s4, updating the current value of the hidden variable according to the responsivity matrix corresponding to the iteration process; s5, judging whether the current value of the updated hidden variable meets a convergence condition or not; if yes, ending the method, and taking the updated current value of the hidden variable as the optimal solution; otherwise, returning to the step S3 to execute the next iteration process. According to the invention, the problem of sample imbalance in an industrial scene can be better solved.

Description

technical field [0001] The invention relates to the technical field of industrial forecasting, in particular to a method, device, equipment and medium for determining latent variables of a Gaussian mixture model. Background technique [0002] In the application practice of industrial predictive maintenance, when the Gaussian mixture model is used to fit the normal operation mode in the actual production scenario, the number of samples is unbalanced in most cases. This is because there are often different working conditions in production, and the running time of each working condition is not completely equivalent, so some working conditions will generate more samples, while some working conditions will generate fewer samples, so caused the problem of sample imbalance. For the case with a small number of samples, the early warning effect of the model will be biased. [0003] At present, the main solution to the above problem is the synthetic minority class oversampling techn...

AI Technical Summary

Problems solved by technology

[0004] The invention provides a hidden variable determination method of a Gaussian mixture model, a device, a device, and a medium, which can better solve the problem of unbalanced samples in industrial scenarios

Images