Jonathan Donadee

Stochastic Optimization of Grid to Vehicle Frequency Regulation Capacity Bids

This paper investigates the application of stochastic dynamic programming to the optimization of charging and frequency regulation capacity bids of an electric vehicle (EV) in a smart electric grid environment. We formulate a Markov decision problem to minimize an EVs expected cost over a fixed charging horizon. We account for both Markov random prices and a Markov random regulation signal. We also propose an enhancement to the classical discrete stochastic dynamic programming method. This enhancement allows optimization over a continuous space of decision variables via linear programming at each state. Simple stochastic process models are built from real data and used to simulate the implementation of the proposed method. The proposed method is shown to outperform deterministic model predictive control in terms of average EV charging cost.

Stochastic co-optimization of charging and frequency regulation by electric vehicles

Smart electrical grid infrastructure, such as advanced metering, will enable demand side participation in electrical energy and ancillary services markets. Deterministic optimization models have been proposed for minimizing the cost of charging electric vehicles (EVs) in liberalized market settings. These models include revenues that EVs could earn by providing ancillary services such as secondary frequency regulation. Optimization models currently in the literature do not account for the uncertainty in the costs and benefits of providing regulation. We propose a stochastic dynamic programming method to optimize EV charging and frequency regulation decisions under uncertainty. Expected future costs are included in decision problems as convex piecewise-linear approximations of non-convex value functions. Simulations demonstrate the benefit of charging an EV using our method over an expected value dynamic programming scheme. We also show that the proposed method gives high quality solutions.

AGC Signal Modeling for Energy Storage Operations

Energy storage resources (ESRs) are being used for secondary frequency regulation in the bulk electric power grid. In order to optimize the economic scheduling of an ESR using look-ahead model predictive control, predictive models of the automatic generation control (AGC) signal and its effect on an ESRs state of charge are needed. In this letter, we suggest a straightforward and effective procedure for forecasting the next state of charge for an ESR that provides regulation service in a liberalized market setting.

Optimal operation of energy storage for arbitrage and ancillary service capacity: The infinite horizon approach

This paper considers the co-optimization of the operations of a grid scale energy storage resource (ESR) for both energy price arbitrage and sales of secondary frequency regulation capacity. We investigate the application of the infinite horizon Markov decision problem (MDP) framework to this problem. We formulate the ESRs decision optimization problem as an infinite horizon, average reward MDP. This problem is a proof-of-principle which considers the automatic generation control signal as the only random parameter. Example MDPs are solved using the policy iteration algorithm. The optimal operating policies and gains are described. Results show that the value of an ESR can be increased substantially by using it for more than one purpose simultaneously.

Optimal Autonomous Charging of Electric Vehicles with Stochastic Driver Behavior

This paper proposes the application of the Markov decision problem (MDP) framework for optimizing the autonomous charging of individual plug-in electric vehicles (EVs). Two infinite horizon average cost MDP formulations are described, one for plug-in hybrid electric vehicles (PHEVs) and one for battery only electric vehicles (BEVs). In both formulations, we assume no direct input from the driver to the smart charger about the drivers travel schedule. Instead, we use stochastic models of plug-in and unplug behaviors as well as energy required for transportation to represent a drivers charging requirements. We also assume that electric energy prices follow a Markov random process. These stochastic models can be built from historical data on vehicle usage. The objective of the MDPs is to minimize the sum of electric energy charging costs, driving costs, and the cost of any driver inconvenience. We demonstrate the solution of the MDPs with assumed parameter values and analyze the results. This work presents a new approach to minimizing EV charging costs while reducing the need for trip planning by a driver.

Estimating the rate of battery degradation under a stationary Markov operating policy

Rechargeable Li-Ion battery energy storage is becoming a vital component of many power systems. The infinite horizon Markov decision problem (MDP) framework has been proposed for optimal scheduling of battery charging and discharging under uncertainty in many applications, such as hybrid electric vehicles and bulk electric power grids. In this paper we explain how to determine the expected rate of battery capacity degradation from the solution of an infinite horizon MDP and a degradation severity factor map. We apply the proposed methods to an example MDP from literature.

Distribution pricing and tariff structure: The ongoing US reforms

It is hard to report at present any streamlined efforts toward reforming distribution pricing and tariffs to better reflect the technological changes and their value. In particular, at present there is no direct relation between the demand-response commitments to participating in peak load shaving, direct load control programs, on one side, and the prices reflecting their value. As a rule, the signals reflecting the value of the technologies underlying these functionalities are not in place.

Generation and Demand Characteristics of the Islands of Flores and São Miguel

This chapter presents the data on generation and consumption of energy which is used throughout this monograph. Energy consumption is described by patterns in the system load, as well as consumption by consumer type. Energy supply is described by data on the composition of installed generation equipment and their estimated levelized costs of energy (LCOE). The availability of existing and potential renewable energy resources is also described. Finally, the electric tariff structure for consumers is discussed.

Assessing the Ability of Different Types of Loads to Participate in Adaptive Load Management

This chapter attempts to find the right strategy to utilize the loads on Flores and Sao Miguel most efficiently, in order to make the power systems of those islands sustainable and clean. In exploring the right technologies for the island, we recognize the need for different demand response program schemes for different loads. We differentiate the different types of loads according to their physical attributes and define the right decision tools and time frames for each type. The time frames of control range from the scheduling and shifting of seasonal or weekly loads to rescheduling some flexible loads days ahead of consumption, and adjusting the energy consumption of some loads in near real time. Specific examples of the loads on both Flores and Sao Miguel are also given, along with detailed descriptions of the numeric values of the loads and their ability to adjust to system conditions or price.

Look-Ahead Model-Predictive Generation and Demand Dispatch for Managing Uncertainties

In this chapter, we focus on demand response in the timescales of day-ahead and real-time operation. We show the formulation, numerical examples, and the results of economic dispatch considering an elastic load with its demand function and the generation resources. We use two distinct algorithms for this demand and generation dispatch: Day-Ahead Scheduling and Real-Time Adjustment. We compare and analyze the results of these two different methods and discuss their implications for short-term system dispatch and operation.