Simulation and big data challenges in tuning building energy models

Abstract

EnergyPlus is the flagship building energy simulation software used to model whole building energy consumption for residential and commercial establishments. A typical input to the program often has hundreds, sometimes thousands of parameters which are typically tweaked by a buildings expert to "get it right". This process can sometimes take months. "Autotune" is an ongoing research effort employing machine learning techniques to automate the tuning of the input parameters for an EnergyPlus input description of a building. Even with automation, the computational challenge faced to run the tuning simulation ensemble is daunting and requires the use of supercomputers to make it tractable in time. In this paper, we describe the scope of the problem, particularly the technical challenges faced and overcome, and the software infrastructure developed/in development when taking the EnergyPlus engine, which was primarily designed to run on desktops, and scaling it to run on shared memory supercomputers (Nautilus) and distributed memory supercomputers (Frost and Titan). The parametric simulations produce data in the order of tens to a couple of hundred terabytes. We describe the approaches employed to streamline and reduce bottlenecks in the workflow for this data, which is subsequently being made available for the tuning effort as well as made available publicly for open-science.

Development of a Building Information Modeling-Parametric Workflow Based Renovation Strategy for an Exemplary Apartment Building in Seoul, Korea

“…To streamline the BIM-parametric framework and include additional analysis levels for the development of renovation strategies of aged apartment buildings, different improvements are under development, particularly the automation of optimization cycles and direct data transfer between BIM, CAD software and parametric environmental analysis tools [94]. In terms of building information management, the automated transfer of construction data from the BIM model to the parametric software can be executed by employing the Revit parametric suite Dynamo [95].…”

Data modeling versus simulation modeling in the big data era: case study of a greenhouse control system

“…Researches about building energy modeling similarly apply big data to improve the simulation result. 18 However, it is important to manage and process the big data of buildings. 19 They focus on technical challenges to improve big data processing, but they do not approach the use of big data as a modeling means.…”

Scalable tuning of building models to hourly data

“…The Titan supercomputer at Oak Ridge National Laboratory (ORNL) allowed the use of 131,072 cores to calculate 524,288 simulations and write 44 TB of data to disk in 68 minutes [14]. Some of the latest advances in web-oriented database storage were used to allow queryable simulations generated from varying 156 inputs and reporting 96 outputs at 15-minute resolution (35 MB per simulation) for 8 million EnergyPlus simulations [15]. Measured data often are corrupt because of uncalibrated sensors or missing data, so statistical techniques have been refined for autonomous quality assurance and gap-filling [16].…”