Development of a clustering-based morning start time estimation algorithm for space heating and cooling

The morning start time of heating and cooling equipment plays an important role in the energy and comfort performance of buildings. Existing algorithms to guide this decision require either many data types with a consistent labelling nomenclature or a detailed calibrated model. In this paper, a model-free clustering-based morning start time estimation algorithm is put forward. The algorithm inputs only four types of data: indoor and outdoor temperatures, and heating and cooling energy use, and does not require any information regarding the location of the temperature sensors. The algorithm consists of four steps. The first one employs clustering to form groups of zones with a similar temperature response. The second one searches for inflection points to identify cluster temperature change rates during morning start-up periods. The third one determines the start time based on previous morning start-up temperature change rates. The last one estimates the energy savings potential by using bivariate change point models. The algorithm was developed by using a dataset from a large office building. Through hierarchical clustering, the data from 142 temperature sensors were consolidated to only seven clusters. The median morning start-up temperature change rates in individual clusters were between 0.3°C/h and 0.8°C/h for heating, and between -0.5°C/h and -1.2°C/h for cooling. The savings potential by tuning daily start times based on this information was estimated as 3% and 7% for heating and cooling, respectively.

[1]  A. Athienitis,et al.  Model-Based Control of a Hydronic Radiant Slab for Peak Load Reduction , 2017, Building Simulation Conference Proceedings.

[2]  Ian Beausoleil-Morrison,et al.  Development and implementation of control-oriented models for terminal heating and cooling units , 2016 .

[3]  Yongjun Sun,et al.  Model-based optimal start control strategy for multi-chiller plants in commercial buildings , 2010 .

[4]  Amarjeet Singh,et al.  Poster Abstract: Analytics Driven Operational Efficiency in HVAC Systems , 2015, BuildSys@SenSys.

[5]  Martin Kozek,et al.  Ten questions concerning model predictive control for energy efficient buildings , 2016 .

[6]  V. R. Dehkordi,et al.  Near-optimal transition between temperature setpoints for peak load reduction in small buildings , 2015 .

[7]  A. B. Birtles,et al.  A new optimum start control algorithm , 1985 .

[8]  Elisa Guelpa,et al.  Peak-shaving in district heating systems through optimal management of the thermal request of buildings , 2017 .

[9]  Myoung-Souk Yeo,et al.  Application of artificial neural network to predict the optimal start time for heating system in building , 2003 .

[10]  John E. Seem,et al.  Model Selection for Predicting the Return Time from Night Setback , 2016 .

[11]  Rajesh Gupta,et al.  Brick : Metadata schema for portable smart building applications , 2018, Applied Energy.

[12]  Ardeshir Mahdavi,et al.  Assessment of thermal comfort under transitional conditions , 2014 .

[13]  Weiming Shen,et al.  Characterization of a building's operation using automation data: A review and case study , 2017 .

[14]  Selçuk Canbay.  Optimization of HVAC Control Strategies By Building Management Systems Case Study: Özdilek Shopping Center , 2003 .

[15]  Ji-Hoon Jang,et al.  Prediction of optimum heating timing based on artificial neural network by utilizing BEMS data , 2019, Journal of Building Engineering.

[16]  Xiufeng Pang,et al.  Monitoring-based HVAC commissioning of an existing office building for energy efficiency , 2013 .

[17]  Stanko Strmčnik,et al.  An algorithm for calculating the optimal reference temperature in buildings , 2009 .

[18]  Shengwei Wang,et al.  Optimal control strategy of central air-conditioning systems of buildings at morning start period for enhanced energy efficiency and peak demand limiting , 2018 .

[19]  William O'Brien,et al.  Building energy model reduction using model-cluster-reduce pipeline , 2018 .

[20]  Mohsen A. Jafari,et al.  Distributed air conditioning control in commercial buildings based on a physical-statistical approach , 2017 .

[21]  Lukas G. Swan,et al.  Methodology to determine the impact of simplified building models on model-predictive-control morning start optimization performance , 2018 .

[22]  Zheng O'Neill,et al.  Comparisons of inverse modeling approaches for predicting building energy performance , 2015 .

[23]  H. Burak Gunay,et al.  Data analytics to improve building performance: A critical review , 2019, Automation in Construction.

[24]  Vikas Chandan,et al.  A Data Driven Pre-cooling Framework for Energy Cost Optimization in Commercial Buildings , 2017, e-Energy.

[25]  Jin Woo Moon,et al.  Algorithm for optimal application of the setback moment in the heating season using an artificial neural network model , 2016 .