Sandro Iacovella

Demand response of a heterogeneous cluster of electric water heaters using batch reinforcement learning

A demand response aggregator, that manages a large cluster of heterogeneous flexibility carriers, faces a complex optimal control problem. Moreover, in most applications of demand response an exact description of the system dynamics and constraints is unavailable, and information comes mostly from observations of system trajectories. This paper presents a model-free approach for controlling a cluster of domestic electric water heaters. The objective is to schedule the cluster at minimum electricity cost by using the thermal storage of the water tanks. The control scheme applies a model-free batch reinforcement learning (batch RL) algorithm in combination with a market-based heuristic. The considered batch RL technique is tested in a stochastic setting, without prior information or model of the system dynamics of the cluster. The simulation results show that the batch RL technique is able to reduce the daily electricity cost within a reasonable learning period of 40-45 days, compared to a hysteresis controller.

Cluster Control of Heterogeneous Thermostatically Controlled Loads Using Tracer Devices

Managing the aggregated demand of large heterogeneous clusters of thermostatically controlled loads (TCLs) is considered a sequential decision-making problem under uncertainty. Recent research indicates that using reduced-order models in combination with a broadcasted control signal offers a viable solution to the tradeoff between computational feasibility, and accurately describing the steady-state and transient cluster response. In this paper, we propose a novel control strategy based on tracer devices, which we define as a limited amount of virtual TCLs that represent the entire cluster of heterogeneous TCLs. These second-order model devices are identified in a nonintrusive manner, and capture both steady-state and transient population dynamics, as well as cluster heterogeneity. Additionally, the dispatch mechanism is included in the optimization, further improving the tracking performance. The parameterizable number of tracer devices enables a covering of the tradeoff domain. Both approaches have been evaluated in two scenarios. In the first small-scale scenario, improvements in price and power deviations are evaluated when using increasing numbers of tracer devices and integrating the dispatch dynamics. Results from the second large-scale scenario show that root mean square dispatch errors can be reduced by more than 10% when integrating the dispatch mechanism in the resulting high-fidelity model.

Distributed voltage control mechanism in low-voltage distribution grid field test

In this paper, we present a distributed voltage control mechanism that is being used in the large-scale field-test of the Linear project. The control system developed does not require a communication network between the different households. Only the locally measured household supply voltage is taken into account. The proposed control system is compatible with DSM infrastructure currently being developed, such as home gateways and smart meters. Moreover, the proposed control system can also be used as a fallback mechanism for other communication-based DSM control systems when communication fails or when the system has been compromised due to cyber security issues. Using Monte Carlo simulations on two accurately modeled field test grids and device models, the proposed approach and its various parameter set points are benchmarked against the optimal Dynamic Programming solution. Simulation results point out that on average the amount of over and under voltage occurrences can be lowered by more than 30 %.

Double-layered control methodology combining price objective and grid constraints

A major challenge consists of considering all stakeholders of the future Smart Grid, each with their specific and possibly opposing objectives. A distribution network operator aims at guaranteeing power quality criteria while consumers aspire lowering their power consumption bill. This fundamental issue currently delays the transition from small-scale research projects to a large-scale all-encompassing smart distribution grid. This paper describes a double-layered control methodology using the available flexibility of the majority of discrete smart appliances currently in use. The effect of striving for the objectives separately as well as in combination is examined. The results show that the targeted objective(s) strongly influence(s) the performance in terms of cost effectiveness as well as number of voltage issues.

Standalone LV distribution network voltage control mechanism

This paper describes a voltage stabilizing control mechanism using the available flexibility of smart devices within one household. The flexibility of all types of smart appliances is used, especially smart on/off devices. The main advantage of the developed control system is that it does not require a communication network between the different households, only locally available measurements, such as the household supply voltage, are taken into account. The control system will be rolled out in a real life pilot test. Simulation results point out that the amount of over and under voltage occurrences on average are lowered with 35%.

Development of a laboratory platform for distributed grid management applications

This paper describes practical experiences in the design and implementation of a laboratory platform for developing and evaluating distributed grid management applications. The applications considered here are focused primarily on the implementation of control mechanisms for demand response of a heterogeneous portfolio of loads and distributed energy resources connected at low voltage levels. Lessons learned from the development of interfaces to various devices for integration into complex multi-host distributed control system architectures are described. Additionally, the authors provide reflections on practical aspects of building such systems and propose a set of requirements for the development of larger pilot tests.

Development of an open-source smart energy house for K-12 education

Energy consumption in buildings represents about one-third of the world-wide energy consumption. Consumers often are not fully aware of energy-conserving measures they could take. Intelligent control of the heating and lighting systems in buildings is one way to increase energy-efficiency. Children and young adults influence domestic energy consumption, by using appliances such as TV and lighting. Often, they are not aware of the costs incurred. The goal of this research is to develop a educational platform for energy efficiency education aimed towards the full age range of K-12 education. A scaled model of a house is used, to explain the energy flows in the residential setting, well-known by the target audience. A model house is designed, with actual loads, using an Arduino Uno electronics platform as an interface to a PC. A reference program in the integrated development environment S4A allows visualizing the energy consumption in a simple manner. The children control a number of scaled household appliances interactively. A survey with the first 25 children (aged 10-12) suggests higher awareness of energy consumption.

Developing engineering-oriented educational workshops within a student branch

The development of two educational workshops, one on energy efficiency and one on human-machine interfaces, is detailed and discussed. Attraction to engineering is not created as much as lost at early ages through current education methods. Through positive, hands-on experiences with engineering in K-12 education, this trend can be turned. IEEE student branches have as part of their mission the education and creation of quality educational resources for the public. After searching in vain for suitable inexpensive material, the IEEE student branch Leuven decided to design and create two workshops on engineering topics for an audience of 10-12 year olds. Handling this as a repeatable project, the student branch found partners to create a low budget project for the attendees. Using the skillset of the specific partners and organizers optimally on the subtasks, a successful repeatable cooperation is realized. This paper discusses the environment in which the project is realized, the steps to it, and how it (indirectly) benefits the organizing student branch.

Multi-Agent platform for Grid and communication impact analysis of rapidly deployed demand response algorithms

This paper describes a rapid algorithm deployment platform for Smart Grid research. Accounting for the complex interplay of power system dynamics and communication delays in the network by means of rapid code deployment during algorithm design can improve the evaluation of Smart Grid control schemes and their impact on grid power quality. Our novel approach bridges the gap between the implementation of highly realistic multi-timeframe simulations, and expensive hardwired deployment. The architecture and behavior of the platform is presented for one specific Demand Response algorithm namely Dual Decomposition. Large numbers of distributed agents are efficiently managed by employing FIPA compliant Agent Management Specification (AMS) and Directory Facilitator (DF) functionalities, as well as an efficient SQL database monitoring and logging scheme. The architecture is deployed both on an actual laboratory setup and a virtual OPAL-RT environment. Simulations results show that system latency and computational load increase linearly for increasing numbers of distributed agents. This novel approach provides a realistic and pragmatic solution for evaluating distributed applications for grid management, market applications and advanced monitoring of power quality requirements.