摘要
针对红外测温系统刚上电测温数据不稳定,测量值与真实值偏差较大,为了减少红外测温传感器受距离和环境温度影响,阐述距离和环境温度补偿,采用最小二乘法,建立温度-距离补偿模型,得出补偿后的温度值;采用遗传算法优化的BP神经网络算法,使测量误差显著降低。实验结果表明,经距离补偿后的平均相对误差为0.98%,经过环境温度补偿后的平均绝对误差为0.027℃,提高了红外测温系统的测量精度。
In order to reduce the influence of distance and ambient temperature on the infrared temperature sensor,the temperature measurement data of the infrared temperature measurement system is unstable when it is just powered on,and the deviation between the measured value and the real value is large.In this paper,the distance and ambient temperature compensation are studied,and the temperature-distance compensation model is established by using the least square method,and the temperature value after compensation is obtained.The BP neural network algorithm optimized by genetic algorithm can reduce the measurement error significantly.The experimental results show that the average relative error after distance compensation is 0.98%,and the average absolute error after ambient temperature compensation is 0.027℃,which improves the measurement accuracy of the infrared temperature measurement system..
作者
杨高祥
田军委
高青
王沁
YANG Gaoxiang;TIAN Junwei;GAO Qing;WANG Qin(School of Ordnance Science and Technology,Xi'an Technological University,Shaanxi 710021,China;School of Mechanical and Electrical Engineering,Xi'an Technological University,Shaanxi 710021,China;Shaanxi Electronic Information Group Co.Ltd.,Shaanxi 710026,China)
出处
《电子技术（上海）》
2021年第12期7-9,共3页
Electronic Technology
基金
未央科技局-产学研协同创新计划项目(202012)
榆林科技计划项目项目(2019-123)
关键词
红外测温
温度补偿
最小二乘法
神经网络
infrared temperature measurement
temperature compensation
least square method
neural network
