LoboVault Home

Stochastic Control for Smart Grid with Integrated Renewable Distributed Generators


Please use this identifier to cite or link to this item: http://hdl.handle.net/1928/16750

Stochastic Control for Smart Grid with Integrated Renewable Distributed Generators

Show full item record

Title: Stochastic Control for Smart Grid with Integrated Renewable Distributed Generators
Author: Jayaweera, Sudharman; Li, Ding
Subject: Smart grid control, load demand modeling, stochastic reference tracking, renewable distributed generation
Abstract: In this technical report we outline a control framework for a utility-maintained central power plant in a future smart-grid integrated with renewable distributed generators at customer premises having independent time-dependent household load demands. In traditional electric grid planning, the uncertainty that the utility has to deal with arises mainly due to the random consumption behavior of households. However, in future smart-grids there is to be a second source of uncertainty due to the inherent intermittent nature of integrated renewable distributed generations such as solar, wind and tidal resources at customer premises that would also be integrated to the electric grid. Different demand response and demand side management schemes have been proposed to affect customer load demands, but not much work has been focused on mitigating the uncertainty due to this ever-increasing penetration of renewable generation. Difficulty comes from the fact that renewable generation can be heavily affected by weather/climate conditions. One approach to address this is to develop sophisticated prediction models of the natural environment which, however, could proven to be be even harder than general weather forecast since renewable generation can be a function of many weather condition factors, not to mention other factors related to renewable generation facilities. Thus it is important for the utility to be able to adjust its power output in real time, taking in to account all the uncertainties mentioned above. Towards this end, we first propose a finite state non-stationary Markov chain model to represent household load demands. The proposed model is tested against real measured data to justify its validity. Next, we propose a quadratic stochastic reference tracking scheme for the utility generation control. Based on the state space representation of an assumed synchronous generator, we propose control rules corresponding to different assumptions on reference signals. These assumptions corresponds to different levels of knowledge available to the utility, which are known deterministic reference signal, stochastic reference signal with known moments and stochastic reference signal with known reference system dynamics. Simulation results are presented for tracking performance analysis and comparison.
Date: 2012-01-11
Series: EECE;12-0001
URI: http://hdl.handle.net/1928/16750

Files in this item

Files Size Format View Description
Tech_report_Jan2012.pdf 346.4Kb PDF View/Open Technical report

This item appears in the following Collection(s)

Show full item record

UNM Libraries

Search LoboVault


My Account