提高电网的可靠性:用于电气故障分类的混合机器学习方法
Momotaz Begum1, Ariful Islam Shiplu1, Mehedi Hasan Shuvo1
1Department of Computer Science and Engineering, Dhaka University of Engineering & Technology (DUET), Gazipur, Bangladesh.
本研究介绍了一种混合机器学习模型,用于高效的电力输电线路故障分类. (随机森林+决策树+堆叠) 模型实现了93.64%的准确性,提高了电网可靠性和维护.
科学领域:
- 电气工程 电气工程
- 计算机科学 计算机科学
- 人工智能的人工智能
背景情况:
- 输电线路对于发电至关重要,但故障会导致停电和损坏.
- 准确和快速的故障分类对于智能电网和可靠的电力供应至关重要.
- 现有的方法可能缺乏实时电网监控的效率或解释性.
研究的目的:
- 开发和评估一个高效的机器学习模型来分类电力输电线路故障.
- 为了对错误分类性能进行经典和整体机器学习技术的基准测试.
- 提出一种实用,轻量化的混合模型,作为深度学习方法的替代方案.
主要方法:
- 调查的决策树 (DT),随机森林 (RF),天真湾 (NB),K-最近邻居 (KNN),支持矢量机 (SVM) 和AdaBoost.
- 组合技术包括:硬投票,软投票,堆叠和混合.
- 开发并测试了一种混合组合模型: (RF + DT + 堆叠).
主要成果:
- 混合 (RF + DT + 堆叠) 模型实现了高性能:93.64%的准确性,93.65%的精度,93.64%的回忆和93.64%的F1得分.
- 与其他评估模型相比,拟议的混合模型表现出卓越的性能,可解释性和计算效率.
- 该模型被证明是监控电网故障的实用和轻量级替代方案.
结论:
- 混合组合机器学习方法,特别是 (RF + DT + 堆叠),对于电气故障分类非常有效.
- 该模型增强了决策,优化了维护,并确保了输电网络的不间断能源供应.
- 该研究强调了定制机器学习解决方案的潜力,以提高电网弹性和运营效率.
更多相关视频
10:11Updated Technique for Reliable, Easy, and Tolerated Transcranial Electrical Stimulation Including Transcranial Direct Current Stimulation
Published on: January 3, 2020
12:18A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
相关概念视频
Finding Electric Potential From Electric Field
Determining Electric Field From Electric Potential
In general, regardless of whether the electric field is uniform, it points in the direction of decreasing potential because the force on a positive...
Electric Potential Energy in a Uniform Electric Field
Electrical Systems
To derive the transfer function, consider an RLC...
Electric Charges
The English physicist William Gilbert studied the phenomenon of static electricity in...
Electric Field
In the new picture, imagine that the first charge sets up an electric field independent of all other charges in the universe. When another charge comes in its vicinity, the second charge experiences an electric force depending on the electric field at that point. The source charge does not...
