Gerwin Hoogsteen

Demand side management using profile steering

Many Demand Side Management (DSM) approaches use energy prices as steering signals. This paper shows that such steering signals may result in power quality problems and high losses. As an alternative, this paper proposes to use desired (e.g., flat) power profiles as steering signals and presents an efficient scheduling algorithm that can follow desired power profiles. This paper investigates the complexity of price and profile steering, and presents an algorithm for profile steering. The evaluation of this algorithm studies the results of a best possible uniform pricing and profile steering for a case of 121 houses, each with an electrical vehicle of which the power consumption can be controlled and shifted in time. In contrast to the other evaluated approaches, our profile steering algorithm results in a much flatter profile and keeps the voltage between 220V and 235V at each node. It reduces distribution losses by 57% compared to no control, and by 48% compared to uniform pricing.

Impact of peak electricity demand in distribution grids: A stress test

The number of (hybrid) electric vehicles is growing, leading to a higher demand for electricity in distribution grids. To investigate the effects of the expected peak demand on distribution grids, a stress test with 15 electric vehicles in a single street is conducted and described in this paper. The test is conducted in a neighbourhood where both transformers and households are equipped with measurement devices. A significant maximum power consumption increase (more than double) is observed at one transformer when both the electric vehicles and domestic loads stress the network. The observed voltage drop in the network is 17V. Analysis further shows that the hosting capacity is around 15%-20% for the investigated networks and that under voltage is unlikely to occur. The measurements are compared to a simulation and the results show that the simulations predict the actual measurement accurately.

Generation of flexible domestic load profiles to evaluate Demand Side Management approaches

Various Demand Side Management (DSM) approaches have been developed the last couple of years to avoid costly grid upgrades. However, evaluation of these DSM methodologies is usually restricted to a use-case specific example, making comparison between different DSM approaches hard. This paper presents a novel, open source, load profile generator to evaluate and compare DSM approaches. In addition to static load profiles for both active and reactive power, it also provides flexibility information for various classes of controllable domestic devices. Load profiles and flexibility information are generated using a simple behavioural simulation. The output data uses 1 minute intervals and incorporates device measurements. The generated profiles are in sound with the measurement data obtained in a field test on both the household level and aggregated neighbourhood level. The same dynamics, such as unbalanced loading and rapid power consumption fluctuations, are observed in the generated model.

Integrating LV network models and load-flow calculations into smart grid planning

Increasing energy prices and the greenhouse effect demand a more efficient supply of energy. More residents start to install their own energy generation sources such as photovoltaic cells. The introduction of distributed generation in the low-voltage network can have effects that were unexpected when the network was designed and could lead to a bad power quality. These developments ask for better insight in the effects of a planning for a fleet of households in a network. This paper presents the results of adding network models to planning strategies. Forward-backward load-flow calculations for a three phase low-voltage network are implemented to simulate the network. The results from load-flow calculations are used as feedback for demand side management. The results in this paper show that the implementation is both fast and accurate enough for integration purposes. Combining load-flow feedback and demand side management leads to improved worst-case voltage levels and cable usage whilst peakshaving optimization performance does not degrade significantly. These results indicate that load-flow calculations should be integrated with demand side management methodologies to evaluate whether networks support the effects of steering production and consumption. More sophisticated integration of network models are left for future work.

Non-intrusive appliance recognition

Energy conservation becomes more important nowadays. The use of smart meters and, in the near future, smart appliances, are the key to achieve reduction in energy consumption. This research proposes a non-intrusive appliance monitor and recognition system for implementation on an embedded system. The research focuses on computational efficiency of such a system as embedded systems usually have limited computation power. In this paper, research has been done on the effects on the accuracy of the sample frequency. The algorithm first detects an event in which an appliance is turned either on or off. Subsequently its profile is extracted. A hierarchical support vector machine (HSVM) is used to classify the appliance. The result is a complete algorithm that recognizes individual appliances within a household. Tests on this appliance recognizer show that the proposed algorithm can correctly detect appliances with reasonable accuracy.

Soft-islanding a group of houses through scheduling of CHP, PV and storage

In this paper we investigate the possibility of soft-islanding (near-autonomous operation) a group of houses from the electric power grid in the Netherlands. Energy balancing is possible through applying multi-mode smart grid scheduling for controllable energy generation, storage and consumption devices. The modeled neighborhood consists of modern, well-insulated terraced houses in a typical Dutch climate, each equipped with roof-mounted photovoltaic (PV) panels. The panels are sized to cover the daily electric demand during sunnier parts of the year where the heat demand is low. The system also includes a centrally placed combined heat and power (CHP) with hot water and electric storage, and controllable devices within the houses such as washing and dishwashing machines. The daily domestic hot water demand is supplied entirely by the central CHP. The investigation includes an estimation of system dimensions, e.g. PV, CHP and storage capacities based on daily supply and load profiles on top of the multi-level scheduling. Through simulations we demonstrate the technical feasibility for off-grid operation of this neighborhood.

Assessing the potential of residential HVAC systems for demand-side management

This paper investigates the potential of residential heating, ventilation and air conditioning systems to contribute to dynamic demand-side management. Thermal models for seven houses in Austin, Texas are developed with the goal of using them in a planning based demand-side management methodology. The thermal models form the base to determine the flexibility present in these houses with respect to cooling requirements. The linear models are shown to be reasonably accurate when used to predict indoor temperature changes. Furthermore, the resulting prediction errors can be largely attributed to human behavior. The considered thermal models are integrated in a planning-based demand-side management methodology while accounting for such prediction errors. The resulting methodology is capable of flattening the load profile of the considered houses considerably.

Considering grid limitations in profile steering

Demand side management is often envisioned as a promising technique to reduce peak loads, and match local generation and consumption in the future smart grid. To exploit flexibility offered by smart appliances, a methodology is required to steer the appliances towards better utilization of the grid. As conventional grid reinforcements are very costly, the main objective is often peak shaving to reduce stress on network assets. In this work, we consider a demand side management methodology based on profile steering. This approach generates schedules for controllable devices in a hierarchical way. We show that, when the available flexibility is unevenly distributed, profile steering leads to schedules that violate local grid constraints. We then introduce a modification that enables the methodology to consider local limits on the generated schedules. With this modification, we show that the generated schedules respect the local limits imposed without significantly altering the resulting total load profile.

A Cyber-Physical Systems Perspective on Decentralized Energy Management

Driven by the effects of climate change, our world is in a rapid transition towards a sustainable society powered by renewable energy, such as produced by solar panels and wind turbines. As a result, the supply of energy becomes less controllable, endangering the stability of the electricity system. On the other hand, the adoption of electric vehicles and heat pumps also provides an opportunity to avoid overloading as electricity can be produced and consumed locally. The scope of this thesis is to perform decentralized energy management, specifically within residential distribution grids where large scale adoption of distributed energy resources (DERs) is expected. A cyber-physical systems approach is taken to study the interaction between control systems, the operation of devices and the effect on the physical grid. The first contribution is a proactive control methodology called profile steering to decentralize the coordination between all DERs in a microgrid. The hierarchical structure of profile steering allows for a natural mapping between and radially operated low voltage grids. The model predictive nature of this approach resolves prediction errors in both the time and energy domain. A second contribution is a control methodology based on double-sided auctions, for real-time balancing in islanding situations. In such situations, DERs can assist conventional backup solutions in balancing a microgrid. Local information can be used to avoid overloading when communication networks fail. The third contribution is the developed simulation and demonstration framework to test these control methodologies in a cyber-physical systems context. A real-life stress-test resulted in a supply interruption due to grid overloading. The lack of controllability was a major cause for this, illustrating the importance of control in future distribution grids. The presented decentralized energy management approach is a valuable tool to unlock flexibility of DERs and provide means for a smooth energy transition.

Balancing islanded residential microgrids using demand side management

Now that the internet of things is emerging, control of domestic assets within smart microgrids is also gaining more interest. Various demand side management approaches are presented in literature to control these assets and steer the aggregated consumption profile towards a target profile. Furthermore, these microgrids may operate in islanded mode for short periods in which energy balance is important. This paper proposes a fast control method to control domestic assets in an islanded microgrid by means of 2-way communication and local measurement data. By using the potential of domestic assets, required capacity of backup storage and power ratings of backup generators can be decreased and/or the operation time of the islanded microgrid can be increased. Simulations of a use case show that the storage capacity can be significantly decreased (from 54.3 kWh to 29.6 kWh without loss of QoS) when households in a neighborhood are cooperating.