Search NASA⌕ Search

Engineering topics

Huang, Yunzhi

Publications and source records attributed to Huang, Yunzhi.

Physics Informed Reinforcement Learning for Power Grid Control using Augmented Random Search

Wide adoption of deep reinforcement learning need to overcome several challenges in energy system domain, including scalability, learning from limited samples, and high-dimensional continuous state and action spaces. In this paper, we integrated physics-based information from the normal generator operation state formula in the reinforcement learning agent's neural network loss function, and applied an augmented random search agent to optimize the generator control under dynamic contingency. Simulation results demonstrated the reliability performance improvements in training speed, reward convergence, sampling efficiency, scalability, and transferability.

physics informed ML, Physics Informed Neural Netwo↗

Co-design Center for Exascale Machine Learning Technologies (ExaLearn)

We report rapid growth in data, computational methods, and computing power is driving a remarkable revolution in what variously is termed machine learning (ML), statistical learning, computational learning, and artificial intelligence. In addition to highly visible successes in machine-based natural language translation, playing the game Go, and self-driving cars, these new technologies also have profound implications for computational and experimental science and engineering, as well as for the exascale computing systems that the Department of Energy (DOE) is developing to support those disciplines. Not only do these learning technologies open up exciting opportunities for scientific discovery on exascale systems, they also appear poised to have important implications for the design and use of exascale computers themselves, including high-performance computing (HPC) for ML and ML for HPC. The overarching goal of the ExaLearn co-design project is to provide exascale ML software for use by Exascale Computing Project (ECP) applications, other ECP co-design centers, and DOE experimental facilities and leadership class computing facilities.

97 MATHEMATICS AND COMPUTING↗

ECP Report: Update on Proxy Applications and Vendor Interactions

The ExaLearn miniGAN team (Ellis and Rajamanickam) have released miniGAN, a generative adversarial network(GAN) proxy application, through the ECP proxy application suite. miniGAN is the first machine learning proxy application in the suite (note: the ECP CANDLE project did previously release some benchmarks) and models the performance for training generator and discriminator networks. The GAN's generator and discriminator generate plausible 2D/3D maps and identify fake maps, respectively. miniGAN aims to be a proxy application for related applications in cosmology (CosmoFlow, ExaGAN) and wind energy (ExaWind). miniGAN has been developed so that optimized mathematical kernels (e.g., kernels provided by Kokkos Kernels) can be plugged into to the proxy application to explore potential performance improvements. miniGAN has been released as open source software and is available through the ECP proxy application website (https://proxyapps.exascaleproject.ordecp-proxy-appssuite/) and on GitHub (https://github.com/SandiaMLMiniApps/miniGAN). As part of this release, a generator is provided to generate a data set (series of images) that are inputs to the proxy application.

97 MATHEMATICS AND COMPUTING↗

Coordination of Behind-the-Meter Energy Storage and Building Loads: Optimization with Deep Learning Models*

With the increasing penetration of renewable energy systems and energy storage systems in buildings, it is critical to optimize system operation. In this paper, we develop an optimization strategy to minimize the operation cost as well as maintain indoor thermal comfort, for a building integrated with battery and PV. A Recurrent Neural Network (RNN) is used to predict building thermal load and zone temperatures. A black-box optimization algorithm known as nonlinear optimization by mesh adaptive direct search (NOMADS) is employed in simulation to provide look-ahead optimal battery dispatch and zone temperature set-point schedules so that the operation cost is minimized. Field data collected from a medium sized office building at Pacific Northwest National Laboratory (PNNL) integrated with Photovoltaic (PV) system model and battery storage system model is used to demonstrate the proposed methodology. Compared with rule based methods, the optimization strategy obtained a lower cost of operation while satisfying comfort constraints.

Chen, Yimin↗

Building environment data collection systems

Building environment data collection systems and methods are described. According to one aspect, a system includes a plurality of sensor devices which store different types of environment data, a base system comprising a plurality of interface devices configured to receive the environment data from the sensor devices, storage circuitry configured to store the environment data, and communications circuitry configured to implement communications, and a user interface apparatus comprising communications circuitry configured to communicate with the communications circuitry of the base system, a user interface configured to receive user inputs during installation of the sensor devices, and processing circuitry configured to control the user interface to generate the displayed information and to process the user inputs received via the user interface, and wherein different ones of the sensor devices are configured to generate the different types of environment data as a result of the user inputs.

Hughes, Michael A.↗

Energy savings modelling of re-tuning energy conservation measures in large office buildings

Today, many large commercial buildings use sophisticated building automation systems (BASs) to manage a wide range of building equipment. While the capabilities of BASs have increased over time, many buildings still do not fully use the BAS’s capabilities and are not properly commissioned, operated or maintained, which leads to inefficient operation, increased energy use, and reduced lifetimes of the equipment. This paper investigates the energy savings potential of several common HVAC system re-tuning measures on a typical large office building, using the Department of Energy’s building energy modeling software, EnergyPlus. The baseline prototype model uses roughly as much energy as an average large office building in existing building stock, but does not utilize any re-tuning measures. Individual re-tuning measures simulated against this baseline include automatic schedule adjustments, damper minimum flow adjustments, thermostat adjustments, as well as dynamic resets (set points that change continuously with building and/or outdoor conditions) to static pressure, supply-air temperature, condenser water temperature, chilled and hot water temperature, and chilled and hot water differential pressure set points. Six combinations of these individual measures have been formulated – each designed to conform to limitations to implementation of certain individual measures that might exist in typical buildings. All the individual measures and combinations were simulated in 16 climate locations representative of specific U.S. climate zones. The modeling results suggest that the most effective energy savings measures are those that affect the demand-side of the building (air-systems and schedules). Many of the demand-side individual measures were capable of reducing annual total HVAC system energy consumption by over 20% in most cities that were modeled. Supply side measures affecting HVAC plant conditions were only modestly successful (less than 5% annual HVAC energy savings for most cities for all measures). Combining many of the re-tuning measures revealed deep savings potential. Some of the more aggressive combinations revealed 35-75% reductions in annual HVAC energy consumption, depending on climate and building vintage.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Energy savings and economics of advanced control strategies for packaged air conditioners with gas heat

This paper presents an evaluation of the potential energy savings from adding advanced control to existing packaged air conditioners. Advanced control options include air-side economizer, multi-speed fan control, demand control ventilation and staged cooling. The energy and cost savings from the different control strategies individually and in combination are estimated using the EnergyPlus detailed energy simulation program for four building types, namely, a small office building, a stand-alone retail building, a strip mall building and a supermarket building. For each of the four building types, the simulation was run for 16 locations covering all 15 climate zones in the U.S. The maximum installed cost of a replacement controller that provides acceptable payback periods to owners is estimated.

Wang, Weimin↗

Energy Savings and Economics of Advanced Control Strategies for Packaged Heat Pumps

Pacific Northwest National Laboratory (PNNL), with funding from the U.S. Department of Energy’s (DOE’s) Building Technologies Program (BTP), evaluated a number of control strategies for packaged cooling equipment that can be implemented in an advanced controller, which can be retrofit into existing packaged heat pump units to improve their operational efficiency. This report documents the results of that analysis.

Wang, Weimin↗

Energy Savings Modeling of Standard Commercial Building Re-tuning Measures: Large Office Buildings

Today, many large commercial buildings use sophisticated building automation systems (BASs) to manage a wide range of building equipment. While the capabilities of BASs have increased over time, many buildings still do not fully use the BAS's capabilities and are not properly commissioned, operated or maintained, which leads to inefficient operation, increased energy use, and reduced lifetimes of the equipment. This report investigates the energy savings potential of several common HVAC system retuning measures on a typical large office building prototype model, using the Department of Energy's building energy modeling software, EnergyPlus. The baseline prototype model uses roughly as much energy as an average large office building in existing building stock, but does not utilize any re-tuning measures. Individual re-tuning measures simulated against this baseline include automatic schedule adjustments, damper minimum flow adjustments, thermostat adjustments, as well as dynamic resets (set points that change continuously with building and/or outdoor conditions) to static pressure, supply air temperature, condenser water temperature, chilled and hot water temperature, and chilled and hot water differential pressure set points. Six combinations of these individual measures have been formulated - each designed to conform to limitations to implementation of certain individual measures that might exist in typical buildings. All of these measures and combinations were simulated in 16 cities representative of specific U.S. climate zones. The modeling results suggest that the most effective energy savings measures are those that affect the demand-side of the building (air-systems and schedules). Many of the demand-side individual measures were capable of reducing annual HVAC system energy consumption by over 20% in most cities that were modeled. Supply side measures affecting HVAC plant conditions were only modestly successful (less than 5% annual HVAC energy savings for most cities for all measures). Combining many of the retuning measures revealed deep savings potential. Some of the more aggressive combinations revealed 35-75% reductions in annual HVAC energy consumption, depending on climate and building vintage.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Energy Savings and Economics of Advanced Control Strategies for Packaged Air-Conditioning Units with Gas Heat

Pacific Northwest National Laboratory (PNNL) with funding from the U.S. Department of Energy's Building Technologies Program (BTP) evaluated a number of control strategies that can be implemented in a controller, to improve the operational efficiency of the packaged air conditioning units. The two primary objectives of this research project are: (1) determine the magnitude of energy savings achievable by retrofitting existing packaged air conditioning units with advanced control strategies not ordinarily used for packaged units and (2) estimating what the installed cost of a replacement control with the desired features should be in various regions of the U.S. This document reports results of the study.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Comparison of 2006 IECC and 2009 IECC Commercial Energy Code Requirements for Kansas City, MO

This report summarizes code requirements and energy savings of commercial buildings in climate zone 4 built to the 2009 IECC when compared to the 2006 IECC. In general, the 2009 IECC has higher insulation requirements for exterior walls, roof, and windows and have higher efficiency requirements for HVAC equipment (HVAC equipment efficiency requirements are governed by National Appliance Conversion Act of 1987 (NAECA), and are applicable irrespective of the IECC version adopted). The energy analysis results show that residential and nonresidential commercial buildings meeting the 2009 IECC requirements save between 6.1% and 9.0% site energy, and between 6.4% and 7.7% energy cost when compared to 2006 IECC. Analysis also shows that semiheated buildings have energy and cost savings of 3.9% and 5.6%.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Analysis of IECC (2003, 2006, 2009) and ASHRAE 90.1-2007 Commercial Energy Code Requirements for Mesa, AZ.

This report summarizes code requirements and energy savings of commercial buildings in Climate Zone 2B built to the 2009 IECC and ASHRAE Standard 90.1-2007 when compared to the 2003 IECC and the 2006 IECC. In general, the 2009 IECC and ASHRAE Standard 90.1-2007 have higher insulation requirements for exterior walls, roof, and windows and have higher efficiency requirements for HVAC equipment. HVAC equipment efficiency requirements are governed by National Appliance Conversion Act of 1987 (NAECA), and are applicable irrespective of the IECC version adopted. The energy analysis results show that commercial buildings meeting the 2009 IECC requirements save 4.4% to 9.5% site energy and 4.1% to 9.9% energy cost when compared to the 2006 IECC; and save 10.6% to 29.4% site energy and 10.3% to 29.3% energy cost when compared to the 2003 IECC. Similar analysis comparing ASHRAE Standard 90.1-2007 requirements to the 2006 IECC shows that the energy savings are in the 4.0% to 10.7% for multi-family and retail buildings, but less than 2% for office buildings. Further comparison of ASHRAE Standard 90.1-2007 requirements to the 2003 IECC show site energy savings in the range of 7.7% to 30.6% and energy cost savings range from 7.9% to 30.3%. Both the 2009 IECC and ASHRAE Standard 90.1-2007 have the potential to save energy by comparable levels for most building types.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Technical Support Document: 50% Energy Savings for Small Office Buildings

The Technical Support Document (TSD) for 50% energy savings in small office buildings documents the analysis and results for a recommended package of energy efficiency measures (EEMs) referred to as the advanced EEMs. These are changes to a building design that will reduce energy usage. The package of advanced EEMs achieves a minimum of 50% energy savings and a construction area weighted average energy savings of 56.6% over the ANSI/ASHRAE/IESNA Standard 90.1-2004 for 16 cities which represent the full range of climate zones in the United States. The 50% goal is for site energy usage reduction. The weighted average is based on data on the building area of construction in the various climate locations. Cost-effectiveness of the EEMs is determined showing an average simple payback of 6.7 years for all 16 climate locations. An alternative set of results is provided which includes a variable air volume HVAC system that achieves at least 50% energy savings in 7 of the 16 climate zones with a construction area weighted average savings of 48.5%. Other packages of EEMs may also achieve 50% energy savings; this report does not consider all alternatives but rather presents at least one way to reach the goal. Design teams using this TSD should follow an integrated design approach and utilize additional analysis to evaluate the specific conditions of a project.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗