Cheol-Soo Park

• Ph.D.
• Professor (Full) at Seoul National University

This paper presents a prediction approach for indoor particulate matter (PM2.5, PM10) of two school gyms using a lumped model and an artificial neural network model. The aforementioned two models were developed based on the measurement data including indoor/outdoor PM2.5 & PM10 sensors, on/off status of energy recovery ventilators, and CCTV images...

This study aims to present that the thermal properties of transparent building envelopes, e.g. U-factor and SHGC, need to be treated as dynamic variables. The aforementioned thermal properties have been treated in a deterministic fashion in many simulation studies. To quantify parameter uncertainty caused by U-factor and SHGC, the authors conducted...

Accurate prediction of indoor daylight illuminance is crucial for daylight-based lighting controls. However, determining the illuminance using physics-based simulation tools requires significant amounts of information, e.g. grid of sensors, sky model, 3D geometry of a target building and surroundings, etc. In this study, the authors suggest a dayli...

This paper presents a real-time implementation of model predictive control (MPC) for HVAC systems in an ice-cream factory building. The target building consists of two large open spaces served by two HVAC systems. We developed four artificial neural network (ANN) models that predict the thermal states of the supply air and indoor air of the two the...

Parameter estimation methods can be classified into (1) manual (trial-and-error), (2) numerical optimization (optimization, sampling), (3) Bayesian inference (Bayes filter, Bayesian calibration), and (4) machine learning (generative model). Bayesian calibration has been widely used because it can capture stochastic nature of uncertain parameters. H...

It has been widely acknowledged that rational decision making at architectural design stage is important for energy efficient building design. In other words, the relationship between building energy use and design variables must be carefully analyzed. For this purpose, the global sensitivity analysis (GSA) can be a useful tool because GSA quantifi...

Although it is widely acknowledged that reinforcement learning (RL) can be beneficial for building control, many RL-based control actions remain unexplainable in the daily practice of facility managers. This paper reports a rule reduction framework using explainable RL to enhance the practicality of the control strategy. First, deep Q-learning was...

It is important to control the heating system by following real-time demand, while considering the dynamic changes and non-uniform distributions of indoor environments. This paper presents a model predictive control (MPC) scheme for predicting indoor air temperatures at multiple points in a large factory building that consists of large irregular sp...

In recent decades, many studies have attempted to develop a reliable occupancy model using either rule-based, stochastic, data-driven, or agent-based approaches. These are based on the hypothesis that occupant presence can become predictable provided sufficient knowledge and data are provided. However, a different view propounds that occupant prese...

This paper presents an integrated lighting control that employs a daylit illuminance prediction model for a large, open-spaced factory building in which interactions between multiple luminaires and points of the workplane exist. The prediction model was developed with Radiance and consists of daylight and electric lighting prediction components. Bo...

This study introduces a hybrid model that combines physics and machine learning (ML) models to describe the behaviour of variable refrigerant flow (VRF) systems. The standalone ML model was developed with identical data and conditions for comparison between the hybrid and ML models. A Bayesian neural network (BNN) was used for both the models, and...

The likelihood confidence interval (likelihood CI), the parameter's estimation range, can be useful for analysing parameter identifiability. In this regard, not only the analysis accuracy but also minimizing simulation runs are critical. The authors introduce the parameter identifiability analysis based on the Fisher information matrix (FIM). The F...

In the design or retrofit stage of buildings, sensitivity analysis (SA) has been widely utilized as a decision-making tool. SA analyses the influence of input on output and is used to assess important factors of energy-efficient building design. However, so far SA can only respond to a single output, or to a single criterion problem. Most of the de...

This study proposes a transfer learning (TL)-based inverse modelling to identify unknown building properties. This study examines the transfer from virtual buildings to existing buildings, especially for identifying wall U-value, HVAC efficiency and lighting power density (LPD). For this purpose, synthetic data were generated from simulation result...

Recently, the issue of parameter identifiability has been highlighted in Bayesian inference of building energy models. Parameter identifiability addresses the correspondence between the observed data and the model parameters, and this analysis verifies whether the model parameters can be estimated using observed data. The authors propose a model se...

For efficient energy management of buildings, it is important to classify building energy pattern where dominant design variables could be identified. In this regard, the authors present a classification method of building energy pattern using RANSAC and K-shape algorithm. A reference office building was chosen and populated using the Sobol sequenc...

Sensitivity analysis is important in rational decision making because it can identify meaningful design variables for energy-efficient design of new buildings or energy retrofit of existing buildings. However, it is often overlooked that sensitivity analysis itself is also influenced by uncertain parameters such as occupant behavior, infiltration,...

Data-driven simulation models can be used for optimal control and decision making in building systems only if they are developed accurately and reliably. An unsolved issue pertains to the data collected from BEMS that generally contain errors owing to sensors, malfunctioning systems, and other unknown reasons. These anomalies generally occur in a m...

It is widely acknowledged that Bayesian inference is only beneficial when prior information is properly defined. However, there is no clear rule for prior selection, and it is apparently a matter of subjective selection by the domain expert(s). In other words, because the posterior inference results can vary depending on how the prior is set, a pro...

The building energy performance indicator, called Energy Performance Index (EPI), has been used for the past decades in South Korea. It has a list of design variables assigned with weighting factors (a, b). Unfortunately, the current EPI method is not performance-based but very close to a prescriptive rating. With this in mind, this study aims to p...

Existing studies have treated variable refrigerant flow (VRF) control as a local control problem where control variables are determined using only local state information. This study investigates an integrated VRF control in which the VRF control actions are determined based on not only local information but also the dynamics of the room it serves....

A deep Q-network (DQN) was applied for model-free optimal control balancing between different HVAC systems. The DQN was coupled to a reference office building: an EnergyPlus simulation model provided by the U.S. Department of Energy. The building was air-conditioned with four air-handling units (AHUs), two electric chillers, a cooling tower, and tw...

Parameter identifiability is the concept of whether uncertain parameters can be correctly estimated from the observed data. The main cause of parameter unidentifiability in Bayesian inference is known as ‘overparameterization’. In this study, the likelihood confidence interval (CI) and the likelihood confidence region (CR) were introduced for quant...

This study describes the development of the optimal control strategies of eight parallel heat pumps in an existing building. The building consists of seven floors above ground and two floors underground with a total floor area of 22,440 m². The chilled water generated by each of the eight parallel heat pumps runs through a common primary pipe to mu...

The purpose of the present study was to investigate the relevance of building thermal performance and characteristics to building energy consumption. This paper reports an energy analysis of 4625 office buildings in Seoul, South Korea, using data from the Korean national building energy database and architectural database. The following four resear...

It has been widely acknowledged that occupants are one of the major contributors to account for the performance gap between predicted vs. actual building energy consumption. So far, occupancy has been represented as an average value model (deterministic), Markov chain model, survival model, or data mining model. Recent studies by Ahn et al. and Ahn...

This paper compares two modeling approaches for optimal operation of a turbo chiller installed in an office building: (1) a machine learning model developed with Artificial Neural Network (ANN) and (2) a hybrid machine learning model developed with the ANN model and available physical knowledge of the chiller. Before developing the ANN model of the...

Uncertainty analysis in building energy assessment has become an active research field because a number of factors influencing energy use in buildings are inherently uncertain. This paper provides a systematic review on the latest research progress of uncertainty analysis in building energy assessment from four perspectives: uncertainty data source...

This paper addresses an application of the Kalman filter for estimating a time-varying process disturbance in a building space. The process disturbance means a synthetic composite of heat gains and losses caused by internal heat sources e.g., people, lights, equipment), and airflows. It is difficult to measure and quantify the internal heat sources...

An objective measure of building energy performance is crucial for performance assessment and rational decision making on energy retrofits and policies of existing buildings. One of the most popular measures of building energy performance benchmarking is Energy Use Intensity (EUI, kwh/m²). While EUI is simple to understand, it only represents the a...

Occupant behavior is regarded as one of the major factors contributing to the discrepancy between simulation prediction and real energy use. Over the past several decades, occupants have been represented as fixed profiles of occupant presence in building energy simulation tools. Recently, stochastic models have been introduced to account for dynami...

In this paper, an application of heuristic vs. meta-heuristic approaches to the design of an office building is presented. The building was first optimized by a heuristic approach based on the designers’ expertise, prior experiences and intuitions with the use of a whole building simulation tool, EnergyPlus. Then, a meta-heuristic approach was comp...

The building sector contributes to about 40% of total energy consumption in South Korea. In particular, existing buildings older than 15 years account for 75% of the energy consumption by the entire building sector in South Korea. When assessing energy performance of existing buildings by the use of dynamic simulation tools, there are a variety of...

Building simulation tools have been widely used for performance assessment. However, many studies [1] have reported that a performance gap exists between the reality and simulation output, mainly caused by unknown simulation inputs. Therefore, model calibration needs to be introduced. Calibration attempts can fail for the following reasons: coarse...

In this paper, the authors present the multi-criterion stochastic optimal selection of a double glazing system for a given office building. Four elements are required for a multi-criterion stochastic optimal design: an accurate and fast simulation model, an optimization solver, correct handling of the multi-criterion decision problem, and considera...

Recently, BEMS(Building Energy Management Systems) are widely adopted in large existing buildings and there is a growing interest in applying model-assisted optimal control based on the BEMS data. Unfortunately, current BEMS are used only for measurement, data collection and rule-based operation. It would be ideal if a building's data-driven energy...

The current Building Energy Performance Simulation (BEPS) tools are based on first principles. For the correct use of BEPS tools, simulationists should have an in-depth understanding of building physics, numerical methods, control logics of building systems, etc. However, it takes significant time and effort to develop a first principles-based simu...

Building simulation tools have been developed to represent physical phenomena, but discrepancies between actual and modeled phenomena still persist. Various sources of the gap have been identified such as physical uncertainties, limitations of tools, subjective assumptions, simplification of the model, etc. One source of these discrepancies, which...

For better energy management of existing buildings, an accurate and fast prediction model is required. For this purpose, this study reports the development of a GP (Gaussian Process) model for an AHU fan of the real high-rise office building. The GP Model is a statistical data driven model, and requires far less inputs and demands less computing ti...

With the increasing focus on low energy buildings and the need to develop sustainable built environments, Building Energy Performance Simulation (BEPS) tools have been widely used. However, many issues remain when applying BEPS tools to existing buildings. This paper presents the issues that need to be solved for the application of BEPS tools to an...

For reliable building performance assessment, uncertainty caused by future climate change and future energy retrofit should be taken into account. This paper presents a case study of risk-based decision making driven by future uncertainty sources (climate, energy retrofit) in an existing office building. For the study, the EnergyPlus model was chos...

Recently, BIM based energy simulation has been highlighted due to expectancy for efficient reuse and sharing of building information. The BIM based energy simulation is based on a belief that an automatic conversion of BIM to a BEM (Building Energy Model) can be seamless and such conversion would be much more efficient than a manual conversion. Thi...

Although it is widely acknowledged that occupants play a critical role in building energy consumption, the characteristics of occupants are not well-represented in building simulation. Many statistic and data-mining technologies have been applied to develop a reliable occupant model. In contrast, rather than attempting to develop the occupant model...

Weather data are indispensable for building energy simulation. Most weather stations measure only global solar radiation and thus, the global radiation is usually divided into direct and diffuse radiation based on solar models. The solar models being currently used are expressed in regression equations. Several studies have reported that the differ...

Objective energy performance assessment of buildings is crucial for building stakeholders' rational decision making. One of the most popular building energy performance measures is Energy Use Intensity (EUI, kwh/m2.yr). This has been widely used since it is straightforward, simple and easy to understand. However, it has a severe drawback that it on...

This study reports the development of a Gaussian Process (GP) model based on BEMS data. The GP model is a data driven model and requires a few dominant inputs. It provides a quick prediction with a far less computation than the whole building simulation tools (e.g. EnergyPlus). The GP model developed in this study is capable of predicting the behav...

This paper presents a simulation study to reduce heating and cooling energy demand of a school building in Seoul Metropolitan Area, Korea. The aim of this study was to estimate the impact of passive vs. active approaches on energy savings in buildings using EnergyPlus simulation. By controlling lighting, the energy saving of the original school bui...