Load online monitoring can provide real-time power consumption information to the grid and users. It is an effective method to support energy management and load forecasting work. Traditional method within intrusive mode is difficult to promote a wide range of applications, so non-intrusive load monitoring method (NILM) has important significance. Load identification is very important to NILM. Considering the residential load typical characteristic analysis, a non-intrusive residential load identification algorithm based on genetic optimization is proposed. The algorithm based on load characteristic, including active power and current effective value, uses three different encoding methods to structure fitness function, and ultimately determines the specific type of load by genetic optimization, and then the algorithm is verified effective by the actual sampling load data. Experimental results show that the algorithm can achieve residential load identification, and improves the speed of constriction and accuracy of the identification.
