Temperature estimation method based on neural network

Abstract

The invention discloses a temperature estimation method based on a neural network, and the method comprises the steps: 1), measuring the peripheral temperature data of a part, which is not convenient to measure, of a measured object in an actual working condition through a sensor under an experimental condition, taking the peripheral temperature data and the temperature data of the part of the measured object as sample data, and constructing a training set and a test set; 2) obtaining a large amount of temperature data at different time through the step 1), and performing preprocessing and normalization; 3) establishing a neural network model, and training the neural network model by using the data obtained in the experiment until the error of temperature estimation reaches a satisfactory error rate; and 4) in the actual working condition, utilizing the trained neural network model to accurately estimate the temperature of the part, which is inconvenient to measure, of the measured object in the actual working condition. Compared with the prior art, the method is simple to operate, good in practicability and high in accuracy.

Description

technical field [0001] The invention relates to the technical field of temperature measurement and data modeling, in particular to a neural network-based temperature estimation method. Background technique [0002] In the process of temperature measurement, if the temperature of the object to be measured can be directly measured, and temperature compensation and sensor calibration are added, good measurement temperature data can often be obtained. Moreover, there are a variety of temperature sensors on the market, which can adapt to different environments and have different measurement accuracy, and can be selected according to the production environment and use conditions. [0003] In many production plants and experiments, temperature measurement is very important. There are situations where the measured object needs to be accurately known but it is inconvenient to directly measure the temperature of a certain part in actual use. This can only be done by measuring the temp...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and propose a temperature estimation method based on neural network, by measuring the surrounding temperature of the measured object in the actual working condition, which is inconvenient to measure, and accurately estimate the measured temperature. Measure the temperature of the part of the object. This method does not directly measure the temperature of the part of the measured object that is inconvenient to measure in the actual working condition, but can be estimated by collecting the surrounding temperature of the measured object in the actual working condition that is inconvenient to measure. The more accurate temperature data solves the situation that it is not convenient to measure the temperature of the measured object under the production environment or use conditions, and it is easy to implement and has high accuracy.

Images