<?xml version="1.0" encoding="UTF-8"?><xml><records><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>17</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">David Blum</style></author><author><style face="normal" font="default" size="100%">K. Arendt</style></author><author><style face="normal" font="default" size="100%">Lisa Rivalin</style></author><author><style face="normal" font="default" size="100%">Mary Ann Piette</style></author><author><style face="normal" font="default" size="100%">Michael Wetter</style></author><author><style face="normal" font="default" size="100%">C.T. Veje</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">Practical factors of envelope model setup and their effects on the performance of model predictive control for building heating, ventilating, and air conditioning systems</style></title><secondary-title><style face="normal" font="default" size="100%">Applied Energy</style></secondary-title><short-title><style face="normal" font="default" size="100%">Applied Energy</style></short-title></titles><keywords><keyword><style  face="normal" font="default" size="100%">building simulation</style></keyword><keyword><style  face="normal" font="default" size="100%">hvac</style></keyword><keyword><style  face="normal" font="default" size="100%">Model predictive control</style></keyword><keyword><style  face="normal" font="default" size="100%">System identification</style></keyword></keywords><dates><year><style  face="normal" font="default" size="100%">2019</style></year><pub-dates><date><style  face="normal" font="default" size="100%">02/2019</style></date></pub-dates></dates><urls><web-urls><url><style face="normal" font="default" size="100%">https://linkinghub.elsevier.com/retrieve/pii/S0306261918318099https://api.elsevier.com/content/article/PII:S0306261918318099?httpAccept=text/xmlhttps://api.elsevier.com/content/article/PII:S0306261918318099?httpAccept=text/plain</style></url></web-urls></urls><volume><style face="normal" font="default" size="100%">236</style></volume><pages><style face="normal" font="default" size="100%">410 - 425</style></pages><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">&lt;p&gt;Model predictive control (MPC) for buildings is attracting significant attention in research and industry due to its potential to address a number of challenges facing the building industry, including energy cost reduction, grid integration, and occupant connectivity. However, the strategy has not yet been implemented at any scale, largely due to the significant effort required to configure and calibrate the model used in the MPC controller. While many studies have focused on methods to expedite model configuration and improve model accuracy, few have studied the impact a wide range of factors have on the accuracy of the resulting model. In addition, few have continued on to analyze these factors&#039; impact on MPC controller performance in terms of final operating costs. Therefore, this study first identifies the practical factors affecting model setup, specifically focusing on the thermal envelope. The seven that are identified are building design, model structure, model order, data set, data quality, identification algorithm and initial guesses, and software tool-chain. Then, through a large number of trials, it analyzes each factor&#039;s influence on model accuracy, focusing on grey-box models for a single zone building envelope. Finally, this study implements a subset of the models identified with these factor variations in heating, ventilating, and air conditioning MPC controllers, and tests them in simulation of a representative case that aims to optimally cool a single-zone building with time-varying electricity prices. It is found that a difference of up to 20% in cooling cost for the cases studied can occur between the best performing model and the worst performing model. The primary factors attributing to this were model structure and initial parameter guesses during parameter estimation of the model.&lt;/p&gt;</style></abstract></record><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>17</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Zhe Wang</style></author><author><style face="normal" font="default" size="100%">Tianzhen Hong</style></author><author><style face="normal" font="default" size="100%">Mary Ann Piette</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">Predicting plug loads with occupant count data through a deep learning approach</style></title><secondary-title><style face="normal" font="default" size="100%">Energy</style></secondary-title><short-title><style face="normal" font="default" size="100%">Energy</style></short-title></titles><keywords><keyword><style  face="normal" font="default" size="100%">deep learning</style></keyword><keyword><style  face="normal" font="default" size="100%">Long short term memory network</style></keyword><keyword><style  face="normal" font="default" size="100%">Occupant count</style></keyword><keyword><style  face="normal" font="default" size="100%">Plug loads</style></keyword><keyword><style  face="normal" font="default" size="100%">prediction</style></keyword><keyword><style  face="normal" font="default" size="100%">Predictive control</style></keyword></keywords><dates><year><style  face="normal" font="default" size="100%">2019</style></year><pub-dates><date><style  face="normal" font="default" size="100%">05/2019</style></date></pub-dates></dates><urls><web-urls><url><style face="normal" font="default" size="100%">https://linkinghub.elsevier.com/retrieve/pii/S0360544219310205</style></url></web-urls></urls><volume><style face="normal" font="default" size="100%">181</style></volume><pages><style face="normal" font="default" size="100%">29 - 42</style></pages><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">&lt;p&gt;Predictive control has gained increasing attention for its ability to reduce energy consumption and improve occupant comfort in buildings. The plug loads prediction is a key component for the predictive building controls, as plug loads is a major source of internal heat gains in buildings. This study proposed a novel method to apply the Long-Short-Term-Memory (LSTM) Network, a special form of Recurrent Neural Network, to predict plug loads. The occupant count and the time have been confirmed to drive the plug load profile and thus selected as the features for the plug load prediction. The LSTM network was trained and tested with ground truth occupant count data collected from a real office building in Berkeley, California. Results from the LSTM network markedly improve the prediction accuracy compared with traditional linear regression methods and the classical Artificial Neural Network. 95% of 1-h predictions from LSTM network are within ±1 kW of the actual plug loads, given the average plug loads during the office hour is 8.6 kW. The CV(RMSE) of the predicted plug load is 11% for the next hour, and 20% for the next 8 h. Lastly, we compared four prediction approaches with the office building we monitored: LSTM vs. ARIMA, with occupant counts vs. without occupant counts. It was found, the prediction error of the LSTM approach is around 4% less than the ARIMA approach. Using occupant counts as an exogenous input could further reduce the prediction error by 5%–6%. The findings of this paper could shed light on the plug load prediction for building control optimizations such as model-predictive control.&lt;/p&gt;</style></abstract></record><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>17</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Kaiyu Sun</style></author><author><style face="normal" font="default" size="100%">Tianzhen Hong</style></author><author><style face="normal" font="default" size="100%">Sarah C. Taylor-Lange</style></author><author><style face="normal" font="default" size="100%">Mary Ann Piette</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">A pattern-based automated approach to building energy model calibration</style></title></titles><dates><year><style  face="normal" font="default" size="100%">2015</style></year></dates><abstract><style face="normal" font="default" size="100%">&lt;p&gt;Building model calibration is critical in bringing simulated energy use closer to the actual consumption. This paper presents a novel, automated model calibration approach that uses logic linking parameter tuning with bias pattern recognition to overcome some of the disadvantages associated with traditional calibration processes. The pattern-based process contains four key steps: (1) running the original precalibrated energy model to obtain monthly simulated electricity and gas use; (2) establishing a pattern bias, either Universal or Seasonal Bias, by comparing load shape patterns of simulated and actual monthly energy use; (3) using programmed logic to select which parameter to tune first based on bias pattern, weather and input parameter interactions; and (4) automatically tuning the calibration parameters and checking the progress using pattern-fit criteria. The automated calibration algorithm was implemented in the Commercial Building Energy Saver, a web-based building energy retrofit analysis toolkit. The proof of success of the methodology was demonstrated using a case study of an office building located in San Francisco. The case study inputs included the monthly electricity bill, monthly gas bill, original building model and weather data with outputs resulting in a calibrated model that more closely matched that of the actual building energy use profile. The novelty of the developed calibration methodology lies in linking parameter tuning with the underlying logic associated with bias pattern identification. Although there are some limitations to this approach, the pattern-based automated calibration methodology can be universally adopted as an alternative to manual or hierarchical calibration approaches.&lt;/p&gt;</style></abstract><custom2><style face="normal" font="default" size="100%">LBNL-1004495</style></custom2></record><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>47</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Peng Xu</style></author><author><style face="normal" font="default" size="100%">Philip Haves</style></author><author><style face="normal" font="default" size="100%">Mary Ann Piette</style></author><author><style face="normal" font="default" size="100%">James E. Braun</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">Peak Demand Reduction from Pre-Cooling with Zone Temperature Reset in an Office Building</style></title><secondary-title><style face="normal" font="default" size="100%">2004 ACEEE Summer Study on Energy Efficiency in Buildings</style></secondary-title></titles><keywords><keyword><style  face="normal" font="default" size="100%">demand shifting (pre-cooling)</style></keyword></keywords><dates><year><style  face="normal" font="default" size="100%">2004</style></year><pub-dates><date><style  face="normal" font="default" size="100%">08/2004</style></date></pub-dates></dates><pub-location><style face="normal" font="default" size="100%">Pacific Grove, CA</style></pub-location><abstract><style face="normal" font="default" size="100%">&lt;p&gt;The objective of this study was to demonstrate the potential for reducing peak-period electrical demand in moderate-weight commercial buildings by modifying the control of the HVAC system. An 80,000 ft&lt;sup&gt;2&lt;/sup&gt; office building with a medium-weight building structure and high window-to-wall ratio was used for a case study in which zone temperature set-points were adjusted prior to and during occupancy. HVAC performance data and zone temperatures were recorded using the building control system. Additional operative temperature sensors for selected zones and power meters for the chillers and the AHU fans were installed for the study. An energy performance baseline was constructed from data collected during normal operation. Two strategies for demand shifting using the building thermal mass were then programmed in the control system and implemented progressively over a period of one month. It was found that a simple demand limiting strategy performed well in this building. This strategy involved maintaining zone temperatures at the lower end of the comfort region during the occupied period up until 2 pm. Starting at 2 pm, the zone temperatures were allowed to float to the high end of the comfort region. With this strategy, the chiller power was reduced by 80-100% (1 - 2.3 W/ft&lt;sup&gt;2&lt;/sup&gt;) during normal peak hours from 2 - 5 pm, without causing any thermal comfort complaints. The effects on the demand from 2 - 5 pm of the inclusion of pre-cooling prior to occupancy are unclear.&lt;/p&gt;</style></abstract><custom2><style face="normal" font="default" size="100%">LBNL-55800</style></custom2></record></records></xml>