基于有监督深度学习的设备数字化模型的构建与实现

晋世仲

基于有监督深度学习的设备数字化模型的构建与实现

引用本文：晋世仲.基于有监督深度学习的设备数字化模型的构建与实现[J].计算技术与自动化,2026,(2):182-188

摘要点击次数: 44

全文下载次数: 32

作者	单位
晋世仲	（国家能源集团新能源技术研究院有限公司,北京 102209）

中文摘要:为提高设备数字化系统的安全性和可靠性，通过数据采集模块和数字化模型模块，采用改进型模糊控制算法，能自动调整传感器的运动参数，利用多种传感器实时采集设备数据。数字化模型模块采用改进型卷积神经网络算法和反向传播算法，大幅提升了模型的泛化能力和准确性。经过试验，该系统数据采集模块采集图片峰值信噪比为39.13，设备模型的颜色失真度评价为0.0382，设备模型的结构相似度评价为0.8962，满足设备数字化监测和管理需求。

中文关键词:有监督深度学习模糊控制算法传感器反向传播算法

Construction and Implementation of Device Digitization Model Based on Supervised Deep Learning

Abstract:In order to improve the security and reliability of the digital system for equipment, this article uses an improved fuzzy control algorithm through a data acquisition module and a digital model module, which can automatically adjust the motion parameters of sensors and collect equipment data in real-time using multiple sensors. The digital model module adopts improved convolutional neural network algorithm and back propagation algorithm, greatly improving the generalization ability and accuracy of the model. After testing, the peak signal-to-noise ratio of the collected images by the data acquisition module of the system is 39.13, the color distortion evaluation of the equipment model is 0.0382, and the structural similarity evaluation of the equipment model is 0.8962, which meets the requirements of digital monitoring and management of equipment.

keywords:supervised deep learning fuzzy control algorithm sensors back propagation algorithm

查看全文 查看/发表评论 下载pdf阅读器