Abstract—Despite the significant effort spent the efficient and reliable delivery of energy to individual households remains an unsolved challenge. One of the key roadblocks is the high complexity of the smart grid system. In this paper we propose a new architecture for the management of the energy flow between smart grid and households. The proposed design strives to dynamically balance and optimize the amount of energy between the grid and the various smart homes. The preliminary experimental results show that the proposed design can capture the major characteristics of a smart home system, and when combined with higher level optimization tools has the potential to provide significant energy saving. Index Terms—Energy management, smart grid systems, renewable energy resources, intelligent connector. I. INTRODUCTION For the past fifty years, the electricity grid has been the key source of energy for households around world. However, ever since, the infrastructure for the energy generation and delivery has seen little improvement [1]. Looking forward, we face several major roadblocks before we can connect renewable resources into the current electricity grid. First, the current grid system is dominated by ageing transmission lines, transformers, and traditional power plant stations that makes the grid very unreliable. Second, it follows an outdated one-way grid-to-buildings energy distribution paradigm. Third, we receive very little information from the grid itself. The power plants collect information through sensors, however each house-hold and building have little information on their daily usage until monthly bills are received. The current grid system suffers from a poor information acquisition scheme. Recently, with the development of smart meters, appliances, and renewable energy sources, the requirements on grids have changed significantly. For example, the new renewable energy requires a two-way structure that allows energy to flow in and out of the grid continuously and with changing amplitude. Most appliances are only equipped with local sensors that cannot send their data to a central location for analysis (i.e. a central control board, a computer, or a community level server). A new real time monitoring, sensing, analyzing, and control framework Manuscript received April 15, 2013; revised June 28, 2013. This work was in part supported by the AzRise program at the Arizona Solar Institute. D. Nguyen, H. Zhou, and J. M. Roveda are with the University of Arizona at Tucson, Tucson, AZ 85721 USA (e-mail: dcnguyen@ email.arizona.edu, hezhou@email.arizona.edu, wml@ece.arizona.edu). H. Chang is with Maxim Integrated, Chandler, AZ 85225 USA (e-mail: gudxors@email.arizona.edu). C. Talarico is with the Department of Electrical and Computer Engineering at Gonzaga University, Spokane, WA 99202 USA (e-mail: talarico@gonzaga.edu). A. Annamalai is with Intel, Hillsboro, OR 97124 USA (e-mail: anitaannmalai@gmail.com). needs to be put in place for a household to be automatic and smart. Last, the current electricity grid only regulates the supply side (grid side). Once equipped with renewable energy resources, the grid is no longer the center of the energy system. The energy system becomes a vast network of distributed resources. Under this situation, a highly efficient grid requires the demand side and supply side being balanced at all times. Fig. 1. General architecture for the new smart grid design This paper discusses a new real time monitoring, sensing, analyzing, and control framework that facilitates data flow and energy flow management and integrates both power electronics hardware and software for smart grid and smart buildings. We intend to build a two-way structure that allows energy to flow in and out of the grid continuously and with changing amplitude. The proposed new smart system design strives to balance the demand side and supply side at all times. Instead of regulating the supply side (grid side) as in the traditional system, we need a two-side regulation scheme: we should be able to reconfigure household automatically, as well as regulating the grid. Fig. 1 demonstrates the general flow of the proposed project. The Intelligent Connector (I-Connector) controls the hardware system in the household. It receives optimization results from the simulation based Optimization with Hierarchical Options (SoHo). The Interface Model (I-model) [2], [3] prunes data from sensors and smart meters at the households to provide a fast optimization framework. The contributions of this paper include: 1) A new smart connector (I-Connector) that allows multiple direction energy flow and two-way regulations. In addition to the exiting peaking power tracking scheme, the proposed smart connector allows reconfigurable connection and stable output considering photovoltaic (PV) panel output uncertainty. 2) A new simulation based Optimization with Hierarchical Options (SoHo) scheme to schedule multiple buildings Intelligent Connector (I-Connector) Design and Demand Response in Smart Grid Dung Nguyen, Hyungtaek Chang, Claudio Talarico, Anita Annamalai, He Zhou, and Janet M. Roveda Journal of Clean Energy Technologies, Vol. 2, No. 1, January 2014 95 DOI: 10.7763/JOCET.2014.V2.99