Non-Invasive Power Load Identification Method Based on Long Short-Term Memory Neural Network
The conventional non-intrusive power load identification method ignores the influence of transient features on load identification.Therefore,a non-intrusive power load identification method based on long short-term memory neural network was designed.Extract non-intrusive power load characteristics,collect current,voltage,and power signals,and obtain transient load characteristics and steady-state load characteristics.Based on the long short-term memory neural network to identify the power load category,the time characteristics are used to subdivide the load events,divide the interval of the load events,obtain the number of load types,and match the load events with the load sequence,so as to achieve the accurate identification of power loads.Comparative experiments are used to verify that the identification effect of the proposed method is better.
long short-time memory neural networknon-invasivepower loadidentification method
NETL
NSTL
万方数据
