Automated Non-Intrusive Load Monitoring System Using Stacked Neural Networks and Numerical Integration

Abstract

Population growth and new consumer needs, among other factors, have lead to growing energy demand, without a concomitant increase in energy generation. This way, reduction and rationalization of energy consumption, especially by residential users, have become a global concern generating a need for developing techniques for efficient management and distribution of the available energy. Non-Intrusive Load Monitoring (NILM) techniques have provided valuable information about energy consumption for power generation companies as well as consumers. Such information is important for making decisions related to sustainable use of energy resources. This study proposes an automated system based on Artificial Neural Network for performing some of the NILM tasks. A stacked neural network was developed to extract features of power signals of appliances to identify those in operation during a given period. This information is then used to disaggregate individual appliance loads through the total aggregate signal, and consumption is calculated through numerical integration. The system was tested using real data from two databases about appliances with On/Off, multi-level, and variable consumption patterns collected in low frequency. The performance metrics, resulting from identification and disaggregation tasks, demonstrate the efficiency of the proposed system.

Real-Time Non-Intrusive Electrical Load Classification Over IoT Using Machine Learning

“…If a faster communication speed is required, the 4G/3G (such as SIM7600CE shield for Arduino, which supports 4G, 3G, and GPRS data transmission) module can be employed. [13]. They employed REDD and UK-DALE datasets and achieved F1-scores of 0.879 and 0.858, respectively, for the datasets.…”

“…However, the increasing computational complexity associated with the growing number of appliances makes the methods difficult to implement for real-time applications. Individual appliance identification using features of the low-frequency power-series signal was demonstrated by Corrêa et al [13] and Zhang et al [14]. Rafiq et al also identified a single appliance using low-frequency active power (P), apparent power (S), reactive power (Q), rms voltage (V), rms current (I), and power factor (PF) data [15].…”