Abstract
This paper describes recent work on the development of a wireless-based remote monitoring system for household energy consumption and generation in Madeira Island, Portugal. It contains three different main sections: (1) a monitoring system for consumed and produced energy of residencies equipped with photovoltaic (PV) systems, (2) developing a tool to predict the electricity production, (3) and proposing a solution to detect the PV system malfunctions. With the later tool, the user (owner) or the energy management system can monitor its own PV system and make an efficient schedule use of electricity at the consumption side. In addition, currently, the owners of PV systems are notified about a failure in the system only when they receive the bill, whereas using the proposed method conveniently would notify owners prior to bill issue. The artificial neural network was employed as a tool together with the hardware-based monitoring system which allows a daily analysis of the performance of the system. The comparison of the predicted value of the produced electricity with the actual production for each day shows the validity of the method.
Highlights
Over the last few years, the decrement in the manufacturing cost associated with photovoltaic (PV) systems has affected the electricity industry within various aspects
This paper describes recent work on the development of a wireless-based remote monitoring system for household energy consumption and generation in Madeira Island, Portugal
In a scenario in which a considerable portion of the energy of an electric grid is provided by the PV systems, the prediction of the changes on daily production might be necessary in order to schedule the spinning resources capacity [2]
Summary
Over the last few years, the decrement in the manufacturing cost associated with photovoltaic (PV) systems has affected the electricity industry within various aspects. In a scenario in which a considerable portion of the energy of an electric grid is provided by the PV systems, the prediction of the changes on daily production might be necessary in order to schedule the spinning resources capacity [2]. Neural Computing and Applications (2020) 32:15835–15844 analyzer for monitoring the PV systems, for instance, increasing the reliability of the electricity supply and improving the quality of scheduling the network operations, in isolated grids where reserves are limited. Thereafter, PA determines whether the system is functioning properly or not It receives two daily production inputs in kWh from the same installation: one of them is the daily production which comes from the SmartSolar-Box and the other comes from the ANN model. Since the predicted value contains an error (it might be estimated higher or less), a threshold on top of the estimated value is introduced
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
