Home Energy Management System for a domestic load center using Artificial Neural Networks towards Energy Integration
M. Pratapa Raju1, A. Jaya Laxmi2
1M. Pratapa Raju, Electrical Electronics Engineering, JNTUH college of Engineering, JNTUH, Hyderabad, India.
2Dr. A Jaya Laxmi, Electrical Electronics Engineering, JNTUH college of Engineering, JNTUH, Hyderabad, India.
Manuscript received on November 17., 2019. | Revised Manuscript received on November 24 2019. | Manuscript published on 30 November, 2019. | PP: 12548-12557 | Volume-8 Issue-4, November 2019. | Retrieval Number: D6761118419/2019©BEIESP | DOI: 10.35940/ijrte.D6761.118419
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Abstract: This paper titled ‘Home Energy Management System (HEMS) for a domestic load center using Artificial Neural Networks towards Energy Integration’ focuses on implementing intelligent integration of Distribution Generation Systems (DGS) for domestic load. The proposed energy integration by HEMS, is executed by implementing a Load Management Algorithm (LMA), which functions based on ANN forecast models. Demand Side Management (DSM) techniques are the back bone for LMA proposed in this paper. Historical temperature data and electrical load data of a domestic load center at Municipality of Birmingham, City of Alabama, USA, is considered to implement the proposed LMA. Two subroutine programs/algorithms are designed to implement the proposed LMA. First is of implementing Load Priority Techniques, to assign Load Priority for the loads for the hour based on the temperature forecast. Temperature forecast is done using Nonlinear Auto Regression with Exogenous Inputs (NARX) ANN time series model. And this is termed as Load Priority Assignment Algorithm (LPAA). Second is of predicting Threshold Power (PTh) for the hour as per load priority assigned by LPAA, this is termed as estimation of PTh. Big data driven Nonlinear Auto Regression with Exogenous Inputs (NARX) Neural Net based temperature forecasting model is implemented and used to assign hourly priority for all the appliances/loads. Big data driven Artificial Neural Network (ANN) based load power forecasting model is implemented and used to predict hourly PTh, LMA proposed for HEMS is implemented with step by step comparison of current load demand with PTh and transferring the loads between the energy sources, according to the load priority assignment from LPAA. Proposed LMA based HEMS contributes to enhance the penetration of non-conventional DGS into domestic load sector. Simulation of proposed HEMS is implemented in MATLAB-Simulink environment using MATLAB 2018b. Simulation model is embedded into target hardware: Arduino 2560 by enabling serial communication protocols between MATLAB-SIMULINK and ARDUINO. Hence the experimental setup of HEMS would be functional to prove simulation results and to transfer the loads in real time. Working Hardware model of the HEMS at domestic load center is fabricated using simple and cost effective components such as current sensor ACS712, Arduino Mega 2560, relays and relay driver circuit.
Keywords: Home Energy Management System (HEMS), Load Management Algorithm (LMA), Load Priority Assignment Algorithm (LPAA), Demand Side Management Techniques (DSM), Artificial Neural Networks.
Scope of the Article: Parallel and Distributed Algorithms.