Load identification method based on multivariate Gaussian discrimination mode

Abstract

The invention discloses a load identification method based on a multivariate Gaussian discrimination mode, and relates to the technical field of load identification. The method comprises: installing a load identification intelligent terminal at a user total incoming line, and collecting electrical data at the user total incoming line; performing event detection and event marking according to the electrical data to obtain a marked event; extracting electrical characteristics of the marked event, describing the electrical characteristics and transmitting the electrical characteristics to a mode identification module; identifying an equipment name by the mode identification module according to the electrical characteristics; and comparing the device name with a blacklist to obtain the dangerous load device. Electrical data are detected at the user main incoming line, the mode recognition module can accurately recognize the type of electrical equipment and recognize whether the electrical equipment belongs to dangerous loads or not in real time, once the dangerous loads are detected to be input, branch power supply can be rapidly cut off, electrical fire is controlled from the source, the safety of the electrical equipment is improved, and loss of personal safety and property safety is reduced.

Description

technical field [0001] The invention belongs to the technical field of load identification, in particular to a load identification method based on a multivariate Gaussian discriminant method. Background technique [0002] Electrical fires have always been the type of fire with the highest frequency of occurrence and the greatest loss in fire accidents. Most of the causes of electrical fires are caused by short-circuits caused by irregular power consumption behaviors and equipment failures. Therefore, ensuring the safety of power consumption has always been the power department, The focus of close attention by fire departments, enterprise / university management departments, and residential users. [0003] The development of load identification technology to the present stage mostly adopts two identification methods: intrusive and non-invasive. The intrusive identification method needs to install sensors at each load node. Although the data of each node can be obtained, the cos...

AI Technical Summary

Problems solved by technology

[0004] Most of the existing non-intrusive identification technologies are only aimed at the electrical type identification and electric energy measurement of each electrical equipment, collecting and analyzing the data of a certain day or month of the user, and cannot judge and identify electrical equipment in real time, and the danger of quick disconnection load, responding to user needs

Images