基于深度学习的电力设备状态检测的研究

Research on State Detection of Power Equipment Based on Deep Learning

针对设备状态巡检业务应用对图像的智能化程度、可靠性、可用性等需求提出更高要求的情况,提出基于深度学习的电力设备状态检测方案。方案基于异构平台的多类电力设备的状态检测及效率提升技术,实现典型电力设备状态的高效识别,通过低功耗电力设备状态图像采集及分析装置,为应用提供多种形态检测方式,实现多类电力设备的状态检测。此方案改善了电力设备巡检模式,提升了设备状态管控力和运检决策水平,加快了管理决策速度,进一步提升了电力生产管理水平。

According to the requirement of intelligence, reliability degree, reliability and availability of image in the application of equipment state inspection service, a state detection scheme of power equipment based on deep learning is put forward. A multi-class object classification model based on deep learning multi-feature fusion is used to classify large-scale objects in the scheme. State detection and efficiency improvement technology of multi-type power equipment based on heterogeneous platform to realize efficient identification of typical power equipment state. The state image acquisition and analysis device of low power equipment is developed to provide a variety of morphological detection methods for applications and to realize the state detection of multi-class power equipment. This scheme improves the inspection mode of power equipment, improves the control power of equipment status and the level of operation inspection decision, speeds up the speed of management decision, and further improves the level of power production management.