Development of Arduino Microcontroller Based Non-Intrusive Appliances Monitoring System Using Artificial Neural Network

Non-Intrusive Load Monitoring (NILM) is a process of identifying the connected loads in a premises from the measurements obtained at the service entry. That is, through NILM one can tell the operating conditions ( ON or OFF) of the house appliances from the aggregated measurements taken at the service drop. The method is more advanced than the traditional method which require measuring sensors for every load of interest. In an effort to explore the applicability of NILM in home appliances' recognition, this paper presents the development of a home NILM using arduino microcontroller. Aggregated real power (P), rms current (Irms) and power factor (pf) of the connected appliances are used as an offline data to train a feed forward ANN whose output is the pattern of the connected loads. Four different home appliances are experimented in generating the training data and the ANN model is implemented in the arduino program to identify the loads. Experimental analysis on the monitoring system shows that it can accurately recognize the load patterns when the supply voltage is within the range of 240 V and the pattern may not be recognized when the input voltage deviated from 240 V. The developed system can be applied into home appliances management and control for efficient energy utilization, the operation of which can be assessed without entering into the consumer privacy.