| Residents’ electricity safety is one of the common core issues of smart grid and smart city research.Unsafe electricity behaviors may trigger electrical fires and pose a significant threat to life and properties.Monitoring of electrical utilization safety has become an important means to improve electrical utilization safety.Existing electrical utilization safety monitoring schemes have problems such as poor performance in targeting and prediction.It is difficult to find electrical safety issues in a timely and effective manner.Thanks to the maturity of the Internet of Things(IoT)technology,the acquisition,communication and collection of massive electricity consumption data have gradually become possible.The extraction of electricity behavior information and safety hazard clues from electricity consumption data can provide a new means for achieving comprehensive and detailed electricity safety monitoring.This thesis proposes a non-intrusive resident load safety monitoring method,which can realize the in-depth monitoring of resident loads and electrical appliance leakage current.The main research contents of this thesis are as follows:Research on non-intrusive resident load monitoring method.The event detection method is used to locate the appliances’ switching events.Then their power characteristics and harmonic characteristics are extracted.Based on their characteristics,a two-stage clustering method is proposed to maximize the use of feature information to complete the clustering analysis of appliances’ switching events.According to the clustering results,a load profiling method has been established to quantitatively describe the characteristics of electrical appliances.The results of case study show that the proposed method can analyze the behaviors of electrical appliances and know their composition in residential households only through the monitoring of aggregated electrical data.So the perception of residential loads can be realized.Research on safety monitoring of residents’ load based on leakage current.A leakage current monitoring system solution involving cooperation of cloud and detector has been designed,and an online identification method based on load profiles is also proposed to realize real-time identification of appliances’ switch events.Attaching the residual current to the non-intrusive load monitoring,the leakage current of the appliance is identified by linking the change of the residual current with the behavior of the appliance,which helps to monitor the safety status of the residential load.The results of case study show that the proposed method can accurately locate the leakage current appliances,and provides an effective and feasible technical solution for household electrical safety monitoring. |
