Karanjit Kalsi

Aggregated Modeling and Control of Air Conditioning Loads for Demand Response

Demand response is playing an increasingly important role in the efficient and reliable operation of the electric grid. Modeling the dynamic behavior of a large population of responsive loads is especially important to evaluate the effectiveness of various demand response strategies. In this paper, a highly accurate aggregated model is developed for a population of air conditioning loads. The model effectively includes statistical information of the load population, systematically deals with load heterogeneity, and accounts for second-order dynamics necessary to accurately capture the transient dynamics in the collective response. Based on the model, a novel aggregated control strategy is designed for the load population under realistic conditions. The proposed controller is fully responsive and achieves the control objective without sacrificing end-use performance. The proposed aggregated modeling and control strategy is validated through realistic simulations using GridLAB-D. Extensive simulation results indicate that the proposed approach can effectively manage a large number of air conditioning systems to provide various demand response services, such as frequency regulation and peak load reduction.

Aggregate model for heterogeneous thermostatically controlled loads with demand response

Due to the potentially large number of Distributed Energy Resources (DERs) - demand response, distributed generation, distributed storage - that are expected to be deployed, it is impractical to use detailed models of these resources when integrated with the transmission system. Being able to accurately estimate the transients caused by demand response is especially important to analyze the stability of the system under different demand response strategies, where dynamics on time scales of seconds to minutes are important. On the other hand, a less complex model is more amenable to study stability of a large power system, and to design feedback control strategies for the population of devices to provide ancillary services. The main contribution of this paper is to develop an aggregated model for a heterogeneous population of Thermostatic Controlled Loads (TCLs) to accurately capture their collective behavior under demand response. The aggregated model efficiently includes statistical information of the population, systematically deals with heterogeneity, and accounts for a second-order effect necessary to accurately capture the transient dynamics in the collective response. The developed aggregated model is validated against simulations of thousands of detailed building models using GridLAB-D (an open source distribution simulation software) under both steady state and severe dynamic conditions.

Market-Based Coordination of Thermostatically Controlled Loads—Part I: A Mechanism Design Formulation

This paper focuses on the coordination of a population of thermostatically controlled loads (TCLs) with unknown parameters to achieve group objectives. The problem involves designing the device bidding and market clearing strategies to motivate self-interested users to realize efficient energy allocation subject to a peak energy constraint. This coordination problem is formulated as a mechanism design problem, and we propose a mechanism to implement the social choice function in dominant strategy equilibrium. The proposed mechanism consists of a novel bidding and clearing strategy that incorporates the internal dynamics of TCLs in the market mechanism design, and we show it can realize the team optimal solution. This paper is divided into two parts. Part I presents a mathematical formulation of the problem and develops a coordination framework using the mechanism design approach. Part II presents a learning scheme to account for the unknown load model parameters, and evaluates the proposed framework through realistic simulations.

Development and Validation of Aggregated Models for Thermostatic Controlled Loads with Demand Response

One of the salient features of the smart grid is the wide spread use of distributed energy resources (DERs) like small wind turbines, photovoltaic (PV) panels, energy storage (batteries, flywheels, etc), Plug-in Hybrid Electric Vehicles (PHEVs) and controllable end-use loads. The affect of these distributed resources on the distribution feeder and on transmission system operations needs to be understood. Due to the potentially large number of DERs that are expected to be deployed, it is impractical to use detailed models of these resources when integrated with the transmission system. This paper focuses on developing aggregated models for a population of Thermostatic Controlled Loads (TCLs) which are a class of controllable end-use loads. The developed reduced-order models are validated against simulations of thousands of detailed building models using an open source distribution simulation software (Grid LAB-D) under both steady state and dynamic conditions (thermostat setback program as a simple form of demand response).

Aggregated modeling of thermostatic loads in demand response: A systems and control perspective

Demand response is playing an increasingly important role in smart grid research and technologies being examined in recently undertaken demonstration projects. The behavior of load as it is affected by various load control strategies is important to understanding the degree to which different classes of end-use load can contribute to demand response programs at various times. This paper focuses on developing aggregated control models for a homogeneous population of thermostatically controlled loads. The different types of loads considered in this paper include, but are not limited to, water heaters and HVAC units. The effects of demand response and user over-ride on the load population dynamics are investigated. The controllability of the developed lumped models is validated which forms the basis for designing different control strategies.

Optimal control of distributed energy resources using model predictive control

In an isolated power system (rural microgrid), distributed energy resources (DERs), such as renewable energy resources (wind, solar), energy storage and demand response, can be used to complement fossil fueled generators. The uncertainty and variability due to high penetration of wind makes reliable system operations and controls challenging. In this paper, an optimal control strategy is proposed to coordinate energy storage and diesel generators to maximize wind penetration while maintaining system economics and normal operation performance. The problem is formulated as a multi-objective optimization problem with the goals of minimizing fuel costs and changes in power output of diesel generators, minimizing costs associated with low battery life of energy storage, and maximizing the ability to maintain real-time power balance during operations. Two control modes are considered for controlling the energy storage to compensate either net load variability or wind variability. Model predictive control (MPC) is used to solve the aforementioned problem and the performance is compared to an open-loop look-ahead dispatch problem under high penetration of wind. Simulation studies using different prediction horizons further demonstrate the efficacy of the closed-loop MPC in compensating for uncertainties in the system caused by wind and demand.

Distributed Hierarchical Control Architecture for Transient Dynamics Improvement in Power Systems

In this paper, a novel distributed hierarchical control architecture is proposed for large-scale power systems. The newly proposed architecture facilitates faster and more accurate frequency restoration during primary frequency control, by providing decentralized robust control to several selected pilot generators in each area. At the local level, these decentralized robust controllers are designed to quickly damp oscillations and restore frequency after large faults and disturbances in the system. Incorporating this supplementary governor control helps the system reach the nominal frequency without necessarily requiring secondary frequency control. Thus, at the area level, automatic generation control (AGC) actions are alleviated in terms of conducting frequency restoration. Moreover, at the area level, AGC coordinates with the decentralized robust controllers to successfully perform tie-line power balancing, while efficiently damping low-frequency inter-area oscillations. The interaction of local and area controllers is validated through detailed simulations.

Transactive Control of Commercial Buildings for Demand Response

Transactive control is a type of distributed control strategy that uses market mechanisms to engage self-interested responsive loads to achieve power balance in the electrical power grid. In this paper, we propose a transactive control approach of commercial building heating, ventilation, and air-conditioning (HVAC) systems for demand response. We first describe the system models, and identify their model parameters using data collected from systems engineering building (SEB) located on our Pacific Northwest National Laboratory campus. We next present a transactive control market structure for commercial building HVAC systems, and describe its agent bidding and market clearing strategies. Several case studies are performed in a simulation environment using building controls virtual test bed (BCVTB) and calibrated SEB EnergyPlus model. We show that the proposed transactive control approach is very effective at peak shaving, load shifting, and strategic conservation for commercial building HVAC systems.

Transactive Control and Coordination of Distributed Assets for Ancillary Services

The need to diversify energy supplies, the need to mitigate energy-related environmental impact, and the entry of electric vehicles in large numbers present challenges and opportunities to power system professionals. Wind and solar power provide many benefits, and to reap the benefits the resulting increased variability—forecasted as well as unforecasted—should be addressed. Demand resources are receiving increasing attention as one means of providing the grid balancing services. Control and coordination of a large number (~millions) of distributed smart grid assets requires innovative approaches. One such is transactive control and coordination (TC2)—a distributed, agent-based incentive and control system. The TC2 paradigm is to create a market system with the following characteristics: • Participation should be entirely voluntary. • The participant decides at what price s/he is willing to participate. • The bids and responses are automated. Such an approach has been developed and demonstrated by Pacific Northwest National Laboratory for energy markets. It is the purpose of this project to develop a similar approach for ancillary services. In this report, the following ancillary services are considered: • spinning reserve • ramping • regulation. These services are to be provided by the following devices: • refrigerators • water heaters • clothes dryers • variable speed drives. The important results are summarized below: The regulation signal can be divided into an energy-neutral high frequency component and a low frequency component. The high frequency component is particularly well suited for demand resources. The low frequency component, which carries energy non-neutrality, can be handled by a combination of generators and demand resources. An explicit method for such a separation is obtained from an exponentially weighted moving average filter. Causal filters (i.e., filters that process only present and past values of a signal) introduce delays that can be issues in some signal processing applications that treat the high frequency part as a noise to be eliminated. For regulation, the high frequency component is an essential part of the signal. The delay in the low frequency component is not a problem. A stochastic self-dispatch algorithm determines the response of the devices to the regulation signal. • In an ensemble of devices under normal operation, some devices turn on and some turn off in any time interval. Demand response necessitates turning off devices that would normally be on, or turning on devices that would normally be off. Over time, some of these would have turned off on their own. A formalism to determine expectation values under a combination of natural and forced attrition has been developed. This formalism provides a mechanism for accomplishing a desired power profile within a bid period. In particular, a method to minimize regulation requirement can be developed. The formulation provides valuable insights into control. • Some ancillary services—ramping to absorb unforecasted increase in renewable generation, and regulation down—require the demand resources to increase their energy use. Some resources such as HVAC systems can do this readily, whereas some others require enabling technology. Even without such technology, it is possible to arrange refrigerators and water heaters to have an energy debt and be ready to increase their energy use. A transactive bid mechanism of revolving debt can be developed for this purpose. Dramatic changes in control systems, architecture and markets are expected in the electrical grid. The technical capabilities of a large number of devices interacting with the grid are changing. While it is too early to describe complete solutions, TC2 has attractive features suitable for adapting to the changes. The analyses in this report and the activities planned for FY 14 and beyond are designed to facilitate this transition.

Calibrating multi-machine power system parameters with the extended Kalman filter

Large-scale renewable resources and novel smart-grid technologies continue to increase the complexity of power systems. As power systems continue to become more complex, accurate modeling for planning and operation becomes a necessity. Inaccurate system models would result in an unreliable assessment of system security conditions and could cause large-scale blackouts. This motivates the need for model parameter calibration, since some or all of the model parameters could either be unknown or inaccurate. In this paper, the extended Kalman filter is used to calibrate the parameters of a multi-machine power system in the presence of faults. The calibration performance is tested under varying fault locations, parameter errors, and measurement noise giving an insight into how many generators and which generators could be difficult to calibrate.