虚拟电厂下基于基因表达式编程的云边协同分布式异常数据检测算法

A Cloud-edge Collaboration Distributed Anomaly Data Detection Algorithm Based on Gene Expression Programming in Virtual Power Plant

数据可靠互联对虚拟电厂中各类能源生产、传输以及调度等环节的安全稳定运行至关重要。然而人为失误、采集设备故障、网络恶意攻击等因素导致虚拟电厂各环节业务系统中异常稀疏数据频繁产生。现有基于统计学或机器学习的集中式异常数据检测方法存在依赖数据分布和先验知识、计算复杂度较高、效率低下等缺陷。为了解决上述问题,文章提出一种虚拟电厂下基于基因表达式编程的云边协同分布式异常检测算法。首先,基于云边协同机制,构建虚拟电厂云边协同分布式异常检测体系架构;其次,从算法原理、基于最小二乘的全局异常检测模型生成等方面设计基于基因表达式编程的分布式异常检测算法。基于3个真实数据集和3个开源数据集的仿真实验结果表明,与现有模型相比,提出的算法在异常数据检测的准确率、漏检率、误检率、平均耗时以及加速比方面均具有明显的优势。

Reliable data interconnection is crucial for the safe and stable operation of various energy production,transmission and dispatching processes in virtual power plants.However,human errors,collection equipment failures,malicious network attacks,and other factors lead to the frequent generation of anomalous and sparse data in the business systems of virtual power plants.Existing centralized anomalous data detection methods based on statistics or machine learning have defects such as relying on data distribution and a priori knowledge,high computational complexity,and low efficiency.To address the above problems,this paper proposes a cloud-edge collaboration distributed anomaly data detection algorithm based on gene expression programming(CSDAD-GEP)in virtual power plant.First,based on the cloud-edge collaboration mechanism,the cloud-edge collaboration distributed anomaly detection architecture in virtual power plant is constructed.Second,the distributed anomaly detection algorithm based on gene expression programming is designed from the aspects of algorithm principle and global anomaly detection model generation based on least squares.Experimental results on three real datasets and three open-source datasets demonstrate that,compared with the existing algorithms,the algorithm proposed in this paper has obvious advantages in terms of the accuracy rate of anomaly data detection,the leakage detection rate,the misdetection rate,the average time-cost,and the speed-up ratio.