Electrician equipment performance prediction method based on pre-training model

Abstract

The invention provides an electrical equipment performance prediction method based on a pre-training model, and the method comprises the following steps: S1, collecting a historical equipment data set and a target equipment data set, and building a test set; s2, selecting a deep learning network model, and training the selected deep learning network model by using the historical equipment data set; s3, utilizing the collected data set of the target equipment to perform fine tuning on the pre-training model by adopting a model fine tuning method; s4, testing the prediction capability of the fine-tuned model by using a test set, repeating the steps S1-S4 when the prediction capability is lower than a set condition, and executing the step S5 when the prediction capability is higher than the set condition; and S5, performing performance prediction on target equipment by using the trained and fine-tuned model. According to the electrical equipment performance prediction method based on the pre-training model, a neural network does not need to be trained from the beginning for a new task, and an accurate performance prediction result can be obtained under the condition of insufficient target equipment data.

Description

technical field [0001] The invention belongs to the technical field of equipment performance prediction, and in particular relates to a method for predicting the performance of electrical equipment based on a pre-training model. Background technique [0002] With the rapid development of the economy, the level of industrialization in our country has been continuously improved, which puts forward higher requirements for the design and optimization of electrical equipment. At this stage, with the continuous development of deep learning, machine learning methods continue to emerge, researchers will BP (Back Propagation) neural network, decision tree (Decision Tree, DT) and support vector machine (Support Vector Machine, SVM) and other traditional The machine learning method is applied to equipment performance analysis and optimization, which greatly saves computing time and computing resources while ensuring accuracy, and has achieved good results. However, deep learning often...

AI Technical Summary

Problems solved by technology

[0006] In view of this, the present invention aims to propose a method for predicting the performance of electrical equipment based on a pre-trained model to solve the problem of poor prediction results due to insufficient training samples in traditional deep learning

Images