Cooling accounts for 12-38% of total energy consumption in schools in the US, depending on the region. In this study, stacking learning is utilized to predict chiller running capacity for four school buildings (regression) and to predict the chiller status for four another schools (classification) using a collection of interval chiller data and building demand. Singular and multiple measurement periods within one or more seasons are considered. A generalized methodology for modeling building energy systems is posited that informs selection of features, data balancing to attain the best model possible, ensemble-based stacked learning in order to prevent over-fitting, and final model development based upon the results from the stacked learning. The results show that ensemble-based stacked learning improves the model performance substantially; providing the most accurate results for both regression and classification. for both classification and regression. For, classification, the balanced accuracy is 99.79% while Kappa is 99.39%. For regression, the R-squared value, the mean absolute error (MAE) error, and the root mean squared error (RMSE) are 1.78 kW, 2.77 kW, and 0.983 respectively.

Traditional building energy audits are both expensive, in the range of USD $1.29/m 2-$5.37/m 2, and inconsistent in their prediction of potential energy savings. Automation to reduce costs of evaluating the energy effectiveness of buildings is strongly needed. A key element of such automation is a means to estimate the building envelope energy effectiveness. We present a method that addresses this need by using infrared thermography to characterize building wall envelope effectiveness. To date, thermal imaging approaches for estimating wall R-Values, based upon thermal-physical models of walls, require additional manual measurements and analysis which prohibit low-cost, large-scale implementation. To overcome this implementation challenge, a machine learning approach is used to predict wall R-Values for a set of residences with known thermal resistance by utilizing the measured wall imaging temperature, prior weather conditions, historical energy consumption data, and available building geometrical data. The developed model is shown to predict wall R-Values with a maximum test-set root mean squared error of 7% using as few as nine training houses. This result has significant implications for low-cost large-scale envelope energy effectiveness characterization.

Predicting cooling load is essential for many applications such as diagnosing the health of existing chillers, providing better control functionality, and minimizing peak loads. In this study, short-term chiller and total building demand are acquired for five different commercial buildings in the Midwest USA. Four different machine learning models are then used to predict the chiller demand using the total building demand, outdoor weather data, and day/time information. Two data collection scenarios are considered. The first relies upon use of multiple weeks of data collection that includes very warm periods and season transitional periods where the outdoor temperature ranged from very warm to cool conditions in order to envelope all cooling season weather conditions. The second scenario employs use of contiguous data for a several weeks during only the warmest period of the year. The results show that using two or more separate time periods to envelope most of the weather data yields a much more accurate model in comparison to use of data for only one time period. These research findings have importance to energy service companies which often do short term audits (measurements) in order to estimate potential savings from chiller system upgrades (controls or otherwise).

Energy savings based upon use of smart WiFi thermostats ranging from 10 to 15% have been documented, as new features such as geofencing have been added. Here, a new benefit of smart WiFi thermostats is identified and investigated; namely, as a tool to improve the estimation accuracy of residential energy consumption and, as a result, estimation of energy savings from energy system upgrades, when only monthly energy consumption is metered. This is made possible from the higher sampling frequency of smart WiFi thermostats. In this study, collected smart WiFi data are combined with outdoor temperature data and known residential geometrical and energy characteristics. Most importantly, unique power spectra are developed for over 100 individual residences from the measured thermostat indoor temperature in each and used as a predictor in the training of a singular machine learning models to predict consumption in any residence. The best model yielded a percentage mean absolute error (MAE) for monthly gas consumption ±8.6%. Applied to two residences to which attic insulation was added, the resolvable energy savings percentage is shown to be approximately 5% for any residence, representing an improvement in the ASHRAE recommended approach for estimating savings from whole-building energy consumption that is deemed incapable at best of resolving savings less than 10% of total consumption. The approach posited thus offers value to utility-wide energy savings measurement and verification.

Abstract: A conventional ground-coupled heat pump (GCHP) can be used to supplement heat

rejection or extraction, creating a hybrid system that is cost-eective for certainly unbalanced climes.

This research explores the possibility for a hybrid GCHP to use excess heat from a combined heat

power (CHP) unit of natural gas in a heating-dominated environment for smart cities. A design for

a multi-family residential building is considered, with a CHP sized to meet the average electrical

load of the building. The constant electric output of the CHP is used directly, stored for later use in a

battery, or sold back to the grid. Part of the thermal output provides the building with hot water,

and the rest is channeled into the GCHP borehole array to support the building’s large heating needs.

Consumption and weather data are used to predict hourly loads over a year for a specific multi-family

residence. Simulations of the energies exchanged between system components are performed, and a

cost model is minimized over CHP size, battery storage capacity, number of boreholes, and depth of

the borehole. Results indicate a greater cost advantage for the design in a severely heated (Canada)

climate than in a moderately imbalanced (Ohio) climate.

The multi-family residential building sector is the least energy efficient in the United States, thus allowing for ample opportunities for significant cost-effective energy and carbon savings. In the present study, we propose a district solar borehole thermal solar energy storage (BTES) system for both retrofit and new construction for a multi-family residence in the Midwestern United States, where the climate is moderately cold with very warm summers. Actual apartment interval power and water demand data was mined and used to estimate unit level hourly space and water heating demands, which was subsequently used to design a cost-optimal BTES system. Using a dynamic simulation model to predict the system performance over a 25-year period, a parametric study was conducted that varied the sizes of the BTES system and the solar collector array. A life-cycle cost analysis concluded that is it possible for an optimally-sized system to achieve an internal rate of return (IRR) of 11%, while reducing apartment-wide energy and carbon consumption by 46%. Both a stand-alone and solar-assisted ground-source heat pump system were designed and simulated for comparison to the BTES system, and found to be less economically favorable than the solar BTES system. Thus, the promise for district-scale adoption of BTES in multi-family residences is established, particularly for new buildings.

A novel geothermal heat pump (GHP) system with an integrated low- to moderate-temperature salt hydrate phase change material (PCM) storage tank for buildings in cold climates is proposed in this study. The purpose of the PCM storage tank is to dampen peak heating loads and to remove annual ground thermal load imbalances on the ground heat exchanger (GHX) to assist in achieving an optimally-sized GHX. As heat is extracted from the closed-loop system by heat pumps in heating mode, a significant portion of this heat is used to solidify a salt hydrate PCM. This heat of fusion is later released back into the heat transfer fluid, storing it in the PCM tank and GHX for later diurnal and seasonal use. To examine the merits of the proposed concept, electric utility meter data on 15-minute time intervals were mined from an actual apartment building and used to estimate space heating, cooling, and hot water heating loads. Those data were used in an hourly, dynamic 20-year life-cycle simulation model in TRNSYS to design an optimum combination of GHX and PCM storage, where each component was sized to balance the annual ground thermal loads. The system simulation results show significant potential for GHX size reduction with a PCM storage tank, but the system is quite sensitive to the PCM melt temperature due to significant hysteretic nature of the salt hydrate PCM heating and cooling curves. We also find that there is no unique optimum unless other factors are considered such as installation cost and physical constraints; many combinations of GHX size and PCM mass are capable of achieving the design goal with similar annual electric energy consumption. For the cases examined here, a PCM melt temperature of 27 °C yields the most favorable economic results, and a preliminary economic analysis suggests that with typical drilling cost and PCM tank cost values, the GHX size can be reduced by over 50 %.

Abstract

This article presents yet another methodology for the calculation of dimensionless thermal response factors for vertical borehole ground heat exchanger (GHX) arrays, which is a concept introduced by Eskilson (1987). The presented method is based on a well-known solution to an analogous problem in the field of well hydraulics. This solution method, known mathematically as an incomplete Bessel function, and known in the field of well hydraulics as the 'leaky aquifer function', describes the hydraulic head distribution in an aquifer with predominantly radial flow to a well combined with vertical 'leakage' from geologic layers above and below the pumped aquifer. The solution is adapted to model heat transfer from an array of arbitrarily-placed vertical boreholes of finite depth. With proper expression of parameters in the incomplete Bessel function, we show that g-functions of previous researchers can be approximated. The proposed method has been implemented into Matlab and Excel/VBA for g-function generation and monthly GHX simulation.