Integrated control of dynamic facades and distributed energy resources for energy cost minimization in commercial buildings
Introduction
Distributed energy resources (DER) promise economic, environmental and utility system benefits. Small-scale, distributed electricity generation acts as a hedge against electricity price fluctuation and addresses power quality and reliability concerns caused by failures in the utility network, large voltage drops due to switching operations, and voltage deviations as loads vary on the network (Pepermans et al., 2005, Dreisen and Katieraei, 2008). In cases where environmental regulations are mandating increased use of clean renewable energy sources such as solar- and wind-generated power, distributed generation delivers power more efficiently when co-located with the load it serves. When distributed generation is combined with storage, controllable and uncontrollable loads, and an active distribution network that enables bidirectional power exchanges with the utility grid, the result is a system architecture that enables very efficient delivery of energy, reliable power supply with differentiation in power service quality, and the ability to operate some or all building systems autonomously off the grid in the event of power outages.
Some electricity end uses in buildings require high power quality and reliability, so real time load management and demand response strategies can be used to mitigate power fluctuations due to intermittent power generation from distributed energy resources or variability in supply from the grid. Cost and environmental impacts are additional factors for consideration; distributed energy sources can be used during peak periods when electricity from the grid is expensive.
Controllable window, shading, and daylighting technologies provide an interesting opportunity to manage end use demands in ways that both complements the peak output profile of solar photovoltaic (PV) electricity generation and counteracts the peak demands on the utility grid produced by daytime commercial activities. In California, for example, air conditioning and lighting loads constitute over 60% of the peak summer commercial building end use load profile (Brown and Koomey, 2003), both of which are heavily influenced by the building facade. Proper design and control of the facade can reduce solar and thermal loads that contribute to peak cooling energy demand in perimeter zones, while daylighting can reduce lighting demand by admitting sun or skylight, enabling lights to be dimmed or shut off when the solar resource is at its peak output. With active façade controls, tradeoffs between solar control and daylight admittance can be optimized to minimize total loads. In the event of complete islanding due to power outages or other reasons, this load management capability can help make off-grid buildings potentially more livable and resilient – which is ultimately synergistic to ambitious zero net energy goals.
Integrating the demand side control with supply side control systems has been investigated in previous studies. Wu and Xia (2015) discuss an approach to achieve optimal switching of supply side resources for demand side management, but demand in this case is defined by time-of-use rate charges and optimal control refers to methods to switch control of a hybrid PV-diesel-battery system to minimize energy cost. Similarly, Hanna et al. (2014) investigated control of supply side resources to minimize time-of-use energy cost and peak demand charges, focusing on optimizing the battery storage dispatch strategy based on solar and load forecasts. Wang et al. (2011) proposed a multi-agent based control framework for integrated demand–supply side control, using simulations to demonstrate how genetic algorithms can be used to determine the optimum setpoints for building equipment (i.e., temperature, illuminance, and CO2 levels for heating/cooling, lighting, and ventilation components, respectively) and use of microgrid sources (PV and battery) to balance power consumption and occupant comfort. Here, active control of building loads in coordination with supply side resources fit the definition of demand–supply side control used in this study. Similarly, Stluka et al. (2011) evaluated the technical challenges of achieving optimal demand and supply side control. To implement control, Stluka formulated a solution where the load is disaggregated into a fixed base load and a non-negative controllable load, and the controllable load is assumed to be fragmentable over the time horizon and non-discretionary. A sequential quadratic programming (SQP) solver was then used to find the optimal demand–supply solution for the nonlinear optimization task. The optimization was implemented to control heating and cooling equipment and combined heat and power (CHP) units serving a 850-bed hospital in the Netherlands then operated subsequently for eight years.
This paper examines the synergies between controllable façade technologies and distributed energy resources, namely photovoltaic generation and electricity storage, to determine whether facade technologies can play an enabling role in supporting a desired level of service at either minimum operating cost or minimum carbon footprint. The challenges of coupled demand–supply side control optimizations are described and solved using a two-part approach that is similar conceptually to the method used by Stluka. Performance outcomes of such control are investigated using building energy simulations. A full-scale field test is used to demonstrate feasibility under real weather conditions. Outcomes are used to frame the challenges and benefits of integrating control of dynamic building envelope technologies with the new model of supplying electricity in a liberalized market.
Section snippets
Approach to demand–supply control optimizations
The approach used in this study is based on prior work in order to expedite exploration of the technical and market related issues associated with integrated demand–supply side controls. Therefore, the proof-of-concept solution described in this study was not intended to be widely replicable. On the demand side, prior work with model predictive controls for dynamic façade, lighting, and HVAC systems was leveraged (Coffey et al., 2013). On the supply side, the Distributed Energy Resources
Modeling assumptions
A typical private office was modeled to evaluate the energy cost savings produced by a dynamic versus static façade in combination with distributed energy resources in Berkeley, California. The 3.0 m by 4.57 m (13.9 m2) office was modeled with large-area, south-facing electrochromic windows (6.78 m2, window-to-wall-area ratio (WWR) of 0.59), an opaque exterior wall with an assembly U-value of 0.20 W/m2 K, adiabatic thermal coupling of the interior walls, and dimmable light-emitting diode (LED)
Field test set-up
A field test was conducted at the Advanced Windows Testbed, Berkeley, California to evaluate the feasibility of implementing integrated control of demand and supply side resources and to measure the performance of the system under real sun and sky conditions. The testbed consists of three identical, side-by-side private offices facing due south and is fully instrumented so that lighting and HVAC energy use, control operations, source energy, and comfort can be measured (Fig. 2). Each of the
Discussion
The simulations and field test demonstrate how dynamic façade technologies can play an enabling role in combination with distributed energy resources to achieve near zero net energy goals and a desired level of electrical service at minimum operating cost. The cost-benefit of doing so will be dependent on many factors, including the cost of the dynamic façade and DER technologies, climate, availability of solar resources at the site, utility grid time-of-use rate schedule, etc. These solutions
Conclusions
A proof-of-concept simulation and field study was conducted to investigate the upside potential of integrating dynamic façade technologies with distributed energy resources. Integrated control was achieved by coupling pre-existing modeling tools to generate an optimized schedule for controlling the dynamic façade, charging and discharging electric storage, and use or sale of electricity produced by a solar photovoltaic generation system. These tools were used to estimate the annual cost, carbon
Acknowledgments
We would like to acknowledge the contributions of our LBNL colleagues: Michael Wetter, Dennis DiBartolomeo, Darryl Dickerhoff, Wei Feng, and Anothai Thanachareonkit. We would also like to thank Sage Electrochromics and Saint-Gobain North America for their in-kind support for the field test.
Implementation and testing of the loads-to-DERCAM control module was conducted as part of Christoph Gehbauer’s master’s thesis under the supervision of Professor Hubert Fechner at the Renewable Urban Energy
References (21)
- et al.
Electricity use in California: past trends and present usage patterns
Energy Policy
(2003) - et al.
Energy dispatch schedule optimization for demand charge reduction using a photovoltaic-battery storage system with solar forecasting
Sol. Energy
(2014) - et al.
Distributed generation: definition, benefits and issues
Energy Policy
(2005) - et al.
A comparison of deterministic and probabilistic optimization algorithms for nonsmooth simulation-based optimization
Build. Environ.
(2004) - Coffey, B., McNeil, A., Nouidui, T., Lee, E.S., 2013. Automated production of optimization-based control logics for...
- et al.
Design for distributed energy resources
IEEE Power Energy Mag.
(2008) - EEX, 2014. European Emission Allowances Auction (EUA) | global environmental exchange, primary market auction....
- E3, 2010. Energy+Environmental Economics, GHG tool for buildings in California, December 2010, v.3....
- Gehbauer, C., 2014. A Linear Optimization Framework to Operate Shading Technologies (Thesis). Master of Science in...
- Griffith, B., Long, N., Torcellini, P., Judkoff, R., Crawley, D., Ryan, J., 2007. Assessment of the technical potential...
Cited by (24)
Analysis of the energy and visual performance of a building with photochromic windows for a location in southern Italy
2022, Building and EnvironmentCitation Excerpt :In addition to PCW, thermochromic and electrochromic windows also belong to the category of smart windows [13,14]. Many scientific papers aim to study the building response with these types of windows [15–18]. Typically, electrochromic glasses contain two conductive substrates separated by a solid, gel or liquid electrolyte.
Sustainability performance by ten representative intelligent Façade technologies: A systematic review
2022, Sustainable Energy Technologies and AssessmentsBiomimetic adaptive electrochromic windows for enhancing building energy efficiency
2021, Applied EnergyCitation Excerpt :The energy required to switch BAF from a clear state to a tinted state is 2.7 W/m2 while the energy required to maintain the window to a given tint level is 1.5 W/m2 [43]. The tinting of EC coating lags the command by around 10 minutes [43]. In EnergyPlus simulations, it is assumed that people work from 8:00 to 17:00 on working days.
Effect of electrochromic windows on energy consumption of high-rise office buildings in different climate regions of Iran
2021, Solar EnergyCitation Excerpt :Several previous studies concentrated on the effect of EC windows on heating, cooling and electricity demands of buildings separately (Chambers et al., 2019; Frattolillo et al., 2019; Nageib et al., 2020; Oh et al., 2018b; Tavares et al., 2016). On the other hand, there are several previous researches that evaluated the effect of EC windows on energy consumption and visual comfort simultaneously (Cannavale et al., 2018; Detsi et al., 2020; Dussault and Gosselin, 2017; Favoino et al., 2015; Gugliermetti and Bisegna, 2003; Jonsson and Roos, 2010; Lee et al., 2015; Nageib et al., 2020; Oh et al., 2019; 2018a; Papaefthimiou et al., 2006; Scorpio et al., 2020) or just focused on daylighting performance of EC windows (Aste et al., 2018; Karlsson and Roos, 2000; Page et al., 2007; Piccolo et al., 2009; Wu et al., 2019; Zinzi, 2006). It can be concluded that EC windows can reduce cooling demand and increase heating and lighting demands of a building.
Enhancing building energy efficiency by adaptive façade: A computational optimization approach
2020, Applied EnergyCitation Excerpt :The electrochromic window is subdivided into three zones, and SHGC and Tvis of each zone are controllable [33]. After one week, Lee et al. [33] found that the room with the electrochromic window reduced energy consumption by 50% compared to a benchmark room with a conventional low-emittance window. Ajaji and Andre investigated a control strategy for the electrochromic windows of a building in Brussels [34].