The double-curvature geometry of the system was optimized using a Genetic Algorithm, with experiments being performed in three stages: Stage 1 – Radiance only ; Stage 2 – Radiance + BSDF material ; Stage 3 – BSDF + Radiance 3-Phased Method. The second round of experiments focused on Stage 2 and Stage 3 only. Main conclusions drawn were: under direct beam radiation, the difference between the two methods is significant, with the 3-Phase Method performing up to 34% better in simulating the direct solar component. Without a strong direct solar component, the two methods have a closer performance, simulating diffuse light within a closer range, but Radiance+BSDF still has a significant error rate of up to 10%, thus proving unacceptable for this end. Rapid Prototyping was done initially done using 3D printing, with white plastic coated with 97% reflectance film, and its light transmission measured with the Integrating Sphere at LBNL. A final, industry-grade prototype was designed and budgeted, with an extra thin aluminum profile, 0.020 inches thick, built with 6063 Alloy, T6 temper, and coated with 3M D50A Specular Film Protected, with 98.5% visible light reflectance. A prototype sized for Flexlab, including casting dies, was budgeted at only $1200 by SAPA, but construction was not authorized. Final validation of simulation methods developed against physical measurements was thus not possible. Conclusions regarding product development: for optically complex systems, typical Rapid Prototyping methods do not display the necessary optical accuracy; however, measurements with the goniophotometer require uniform samples, and no larger than one square inch. Thus,hydroponic bucket for optically complex systems with non-uniform geometry, the only current option is building a full-scale prototype for physical testing.
If product development is to be made a priority, new alternatives should be developed to bridge between computer simulations and full scale, expensive and difficult to build prototypes. Whole Building Analysis with 3D Pareto Multi-criteria optimization: building variables under study were: floor stacking patterns, glazing and shading systems characteristics, and shading panes offset, for each orientation, and skylight dimensions. The multi-Objective Metrics and Goals applied were: illuminance levels , Cooling Energy Use Intensity and Heating Energy Use Intensity . Results suggested that Whole Building Multi-criteria Analysis is a fast and effective way of testing complex building alternatives at early design stages, providing useful, operative information for decision making in very short amounts of time, in comparison to standard parametric simulation. The large penetration of renewable energy sources requires flexible resources to manage the variability in generation and demand. Increased observability, improved modeling, and more detailed simulation algorithms are necessary to quantify the impact of intermittent generation to the power network, and – at the same time – efficiently plan the operation of the flexible resources. In order to study the interactions of such complex systems, co-simulation platforms are deemed necessary. The Virtual Grid Integration Laboratory is a modular co-simulation platform designed to study interactions between demand response strategies, building comfort, communication networks, and power system operation. It combines three different simulation tools and incorporates functions for the optimal management of both the grid and the flexible resources. First, research was conducted to determine the appropriate simulation tools for VirGIL and the interfaces to couple them. The Functional Mockup Interface has been selected for coupling the simulation tools. FMI provides a standardized interface, which allows for a very modular co-simulation architecture, where several different modules can be added, exchanged, and tested. FMI enables battery modeling, Electric Vehicle simulation, and advanced optimization functions to be coupled in the future.
DIgSILENT Power factory has been selected as the power system simulator. Using a widely used commercial power system simulation platform will help reduce the barriers to the industry for adopting such platforms, investigate and subsequently deploy demand response strategies in their daily operation. Modelica has been selected as the building modeling language. Modelica is an acausal modeling language, which allows for efficient simulation. Given the complexity of the building models and controls, simulation speed and accuracy are important factors. OMNET++, an established open-source network simulator, has been selected as the communication networks simulation tool. An FMI wrapper was developed for Power factory and OMNET++. VirGIL is the first tool worldwide that will connect Power factory, and OMNET in a co-simulation environment over FMI. Detailed Energy Plus models for existing buildings in LBNL campus were converted to RC models through the BRCM toolbox, and then converted to the Modelica language. Model reduction algorithms were used to reduce the model complexity but maintain the accuracy as much as possible. Studies to compare the accuracy have been carried out. All simulation tools are coupled through FMI to a master algorithm, which is implemented in Ptolemy II. To increase VirGIL’s simulation speed, novel simulation algorithms, notably the Quantized State Simulation algorithm, have been developed in Ptolemy. The first case study of VirGIL was on the LBNL distribution grid. Real data have been used for the power network and the modeling of Building 71. In the second year, we plan to simulate more complex systems, and extend VirGIL by adding advanced optimization algorithms for power system planning and building operation. Research shows that electricity sub-metering can lead to a 10%-30% reduction in electricity use in commercial buildings, and it is likely these savings are available in residential and industrial facilities too. The cost to install available electricity metering technology is very high resulting in virtually no market penetration and an inability to achieve the available savings. We aim to demonstrate the core technologies needed for electricity metering technology that has one-tenth the installed cost of today’s solutions. This new solution will provide sufficient accuracy and time resolution to enabling the retro- and continuous commissioning as well as distributed resource grid integration activities needed for a low carbon society. The core technologies behind power meter are voltage and current measurement, and we will demonstrate non-contact measurement of both quantities.
A suite of sensors will be installed on the surface of circuit breakers in electrical panels, and this installation can be done with minimal training and without an electrician. The sensors will measure the magnetic and electric fields passing through the face of the breaker thousands of times per second, and a set of inverse electromagnetic algorithms will estimate voltage and current in real-time. Each sensor unit will accurately estimate power under a variety of conditions and also compensate for external error sources automatically and without user driven calibration. Our most significant accomplishment is the successful demonstration of accurate voltage, current, and power measurement using a combination of commercially available sensors, custom electronics, and custom electromagnetic and statistical auto-calibration algorithms. We built a proof-of-concept sensor, installed it on various circuit breakers, and demonstrated measurement of voltage, current, and power under laboratory and real-world conditions. We also demonstrated the ability to reliably mitigate the impact of interfering fields from nearby currents. We have also designed an advanced sensor capable of measuring magnetic field in multiple locations and in multiple directions. This technology is critical for reducing the complexity of the installation, improving interference mitigation,stackable planters and enabling improved auto-calibration algorithms. The vector fields that are now measurable are useful for key aspects of advanced auto-calibration algorithms that require no reference information. We are currently applying these results to the design of new sensors and auto-calibration algorithms. The new sensors use a combination of commercially available components and CMOS multi-axis magnetic field sensors. The new algorithms utilize newly identified statistical properties of the primary and interfering signals as well as new inverse electromagnetic analysis techniques. The goal of the first year of this 2-year LDRD was to construct a mass spectrometer that will allow quantitative analysis of gas consumption and gas evolution from electrochemical or chemical systems. This spectrometer is envisioned to become a critical component of a comprehensive and systematic approach designed to understand fundamental reactions occurring in metal-air batteries, Li-ion batteries, CO2 reduction catalysis, oxygen reduction catalysis, and other systems where gas evolution occurs. The differential electrochemical mass spectrometer was constructed over the first 9 months of FY2014 and is now fully operational. The DEMS was custom-built to provide many unique features that allow in-situ, real-time quantitative analysis of gas consumption/evolution in electrochemical cells, which can then be related to Coulometry to more clearly understand electrochemical processes. This technique is extraordinarily powerful and sensitive and provides the capability to assign electrochemical currents to specific reactions. By combining this information with that provided by other spectroscopic and classic electrochemical characterization techniques, complex electrochemical processes can be clearly understood and more readily controlled. Construction and calibration of the DEMS was completed in June 2014, allowing it to have all of the useful features necessary to quantify gas consumption and evolution from electrochemical systems. We are currently exploring three avenues of research with this capability: new electrolyte systems for Li-air batteries , outgassing of Li-ion battery materials at high voltages, and magnesium corrosion in aqueous electrolyte systems to understand limitations of aqueous Mg-air batteries. Proposals have been submitted to the NSF and JCAP II in an attempt to secure funding for aqueous electrocatalysis research . Among the interesting scientific outcomes of our endeavors, we have identified that oxygen out gassing of high voltage non-stoichiometric, Li Li-ion battery cathode materials occurs above 4.6 V and continuously occurs if the electrode is held above that potential, implying that O2 evolution is a parasitic process that should be avoided. Furthermore, CO2 evolution at potentials much lower than 4.6V occurs in this system, implying that the electrode catalyzes an unwanted side reaction.
We have also quantified H2 evolution from Mg electrodes immersed in various aqueous electrolytes as a function of anodic and cathodic currents, providing useful insight into the currently poorly understood Mg-corrosion reaction. Vehicle-grid integration can simultaneously transform the electricity market and the automotive market. For the automotive market, VGI can: 1) allow vehicles to meet all corporate average fuel economy requirements and increasingly stringent emissions regulations, 2) move harmful vehicle emissions away from densely populated areas, and 3) provide revenue to offset the capital cost of vehicle electrification. For the electricity market, VGI can: 4) provide a distributed and growing source of grid energy storage, 5) provide better renewables integration, 6) provide a rapidly ramping resource for many electricity markets, and 7) encourage consumers to more closely scrutinize their home electricity bills just like with gasoline or diesel fuel prices. Despite these benefits, the widespread deployment of VGI faces many uncertainties and barriers within both the electricity market and the automotive market. This LDRD project has created the Vehicle-to-Grid Simulator to provide systematic quantitative methods to develop solutions to the electricity market and automotive market barriers to VGI. This LDRD is developing two versions of V2G-Sim, 1) V2G-Sim Analysis, and 2) V2G-Sim Operations. The V2G-Sim Analysis model couples sub-models for: a) driver behavior, b) automated generation of trip-specific drive cycles , c) vehicle power train models of energy usage during a trip, d) vehicle charging, and e) vehicle response to managed charging or V2G algorithms. V2G-Sim Analysis predicts the behavior for individual vehicles/drivers, and then aggregates individual vehicle profiles to generate grid impacts predictions for large numbers of plug-in vehicles . With these coupled sub-models, V2GSim Analysis provides a platform for scenario analysis of PEV deployment for transmission and distribution infrastructure planning, impacts analysis from various PEV managed charging or V2G algorithms, design of market and pricing structures, etc. V2G-Sim Operations builds upon V2G-Sim Analysis to provide temporally- and spatially-resolved forecasting of PEV charging demands and V2G opportunities allowing an aggregator or integrator to bid PEV services onto an electricity market and operate an electricity grid having many PEVs as a resource within the grid while ensuring each vehicle is sufficiently charged when it needs to be. An underappreciated aspect of climate change is how uncertainty about expected changes affects climate mitigation and adaptation measures. In the past, energy and water planners counted on a relatively stable assessment of climate, infrastructure and policy baselines. With climate change, planners face new uncertainties and forecasts of greater variability. Infrastructure plans must be revised which might include a new peripheral canal or a more decentralized electricity grid, mitigation measures might be considered, such as incorporating bio-fuels and improved batteries, and the valuation of water, as a product or as an energy commodity, must be reconsidered.