Smart Grid Resource Management

The traditional electrical utility industry is undergoing a major transformation that will enable demand management, reliability gains, resource efficiency, customer participation, and cleaner energy across the electrical grid. This transformation is being facilitated by the development of so-called Smart Grid (SG), which is integrated with advand metering, communication, information theory, control, and computer technology. Demand response management (DRM) is an important function in the energy management of SG as it can reduce peak power consumption. Recent interest in DRM has been directed towards real-time interaction with customers through smart meters.However, the successful implementation of DRM involes two major challenges. First, the communication network for DRM should be reliable in oder to maintain the high accuracy of the DRM conrol information and the electricity usage information. Second, the load scheduling mechanism of DRM should effectively cope with the new large-scale energy use scenario, such as the electric vehicle charging. Therefore, this thesis aims to improve the performance of DRM from the perspective of communication resourece management and load scheduling, respectively. Specifically, the major work and contribution relies on four aspects.First, the influence of communication reliability on DRM is analyzed. The load uncertainty brought by communication outage is first molded based on bounded uncertainty, Gaussian uncertainty and unknown uncertainty model, respectively. Then the effect of communication reliability on the DRM performance are derived. Both theoretical and simulation results shows that the DRM performance declines as communication outage probability increases, and it decreases fastest with the unknown distribution model. Second, since the effective use of DRM requires reliable communication nework with significant spectrum resources, a Cognitve Radio (CR) based joint spatial and temporal spectrum sharing is proposed to increase the spectrum utilization efficiency as well as the communication reliability. In particular, the spectrum sharing opportunities and communication outage probabilities for spatial and temporal spectrum sharing are determined. Then the joint spatial and temporal spectrum sharing technique is proposed to enhance the spectrum sharing opportunities in the space-time domain to increase the communication reliability for DRM. The performance improvement is confirmed via both theoretical analysis and simulation.Third, to satisfy different SG applications with different QoS requirements in terms of reliability and communication delay, a priority based Dynamic Spectrum Management (DSM) is proposed to allocate frequency resources to heterogeneous SG applications considering their priorities. DTV spectrum is dynamically used to create more spectrum utilization opportunities and limit the interference to the DTV system. Simulation results are presented which verify that the proposed approach can satisfy the QoS requirements of SG applications and provide a reliable network for critical real-time applications.Forth, the demand response management is implemented in the Vehicle-to-Grid (V2G) system to cope with new large-scale energy use scenario. First, a multi-aim EV charging/discharging scheme is studied in a Vehicle to Building (V2B) scenario.This charging/discharging scheme decreases the peak-to-average ratio (PAR) of the electricity usage in Smart Grid, as well as the cost of both EV users and the builing. A waterfilling based charging/discharing scheme is also proposed in the Vehicle to Vehicle (V2V) scenario to reduce the PAR and flatten the energy demand profile of EV users.Finally, conclusions and future work are summarized at the end of this thesis.