Separating the HVAC energy use from the total residential load can be used to improve energy usage monitoring and to enhance the house energy management systems (HEMS) for existing houses that do not have dedicated HVAC circuits. In this paper, a novel method is proposed to separate the HVAC dominant load component from the house load. The proposed method utilizes deep learning techniques and the physical relationship between HVAC energy use and weather. It employs novel long short-term memory (LSTM) encoder-decoder machine learning (ML) models, which are developed based on future weather data input in place of weather forecasts. In addition to being used in the proposed HVAC separation method, the LSTM models are employed also for accurate day-ahead HVAC and solar photovoltaic (PV) energy forecasts. To test and validate the proposed method, the SHINES dataset, a publicly available dataset spanning three years at 15-minute time resolution from a large-scale DOE experimental project, is used. Two computational case studies are constructed with a test HEMS to investigate the power regulating capability of smart home virtual operation as dispatchable loads or generators. Prediction results obtained with the proposed method show hourly and daily CV(RMSE) of 29.4% and 11.1%, respectively. These results are well within the bounds of error established by academia and the ASHRAE building model and calibration standards.

Building modeling, specifically heating, ventilation, and air conditioning (HVAC) load and equivalent energy storage calculations, represent a key focus for decarbonization of buildings and smart grid controls. In this paper, an ultra-fast one-minute resolution Hybrid Machine Learning Model (HMLM) is proposed as part of a novel contribution in the field of residential physics-based smart home surrogate modeling. Emulation of white box models, or digital twins, with editable parameters through machine learning (ML) meta-modeling serves as an alternative to wide-spread experimental big data collection. The HMLM employs combined k-means clustering with multiple linear regression (MLR) to emulate minutely HVAC power timestep-to-timestep with satisfactory nRMSE error of less than 10% across an entire year test set. An approach is also described to characterize HVAC systems as generalized storage (GES) devices to unify household appliance and virtual power plant (VPP) controls in accordance with industry Communication Technology Association (CTA) 2045 protocol and Energy Star metrics. Synthetic output data from experimentally calibrated EnergyPlus models for three existing smart homes managed by the Tennessee Valley Authority (TVA) is employed in residential case studies and a discussion provided for the large-scale application to hundreds of homes.

Decarbonization of existing electricity generation portfolios with large-scale renewable resources, such as wind and solar photo-voltaic (PV) facilities, is important for a transition to a sustainable energy future. This paper proposes an ultra-fast optimization method for economic dispatch of firm thermal generation using high granularity, one minute resolution load, wind, and solar PV data to more accurately capture the effects of variable renewable energy (VRE). Load-generation imbalance and operational cost are minimized in a multi-objective clustered economic dispatch problem with various generation portfolios, realistic generator flexibility, and increasing levels of VRE integration. The economic feasibility of thermal dispatch scenarios is evaluated through a proposed method of levelized cost of energy (LCOE) for clustered generation portfolios. Effective renewable economics is applied to assess resource adequacy, annual carbon emissions, renewable capacity factor, over generation, and cost to build between thermal dispatch scenarios with incremental increases in VRE penetration. Solar PV and wind generation temporally complement one another in the region studied, and the combination of the two is beneficial to renewable energy integration. Furthermore, replacing older coal units with cleaner and agile natural gas units increases renewable hosting capacity and provides further pathways to decarbonization. Minute-based chronological simulations enable the assessment of renewable effectiveness related to weather-related variability and of complementary technologies, including energy storage for which a sizing procedure is proposed. The generally applicable methods are regionally exemplified for Kentucky, USA, including eight scenarios with four major year-long simulated case studies and 176 subcases using high performance computing (HPC) systems.

As the smart grid involves more new technologies such as electric vehicles (EVs) and distributed energy resources (DERs), more attention is needed in research to general energy storage (GES) based energy management systems (EMS) that account for all possible load shifting and control strategies, specifically with major appliances that are projected to continue electrification such as the electric water heater (EWH). In this work, a methodology for a modified single-node model of a resistive EWH is proposed with improved internal tank temperature for user comfort modeling and capabilities for conservation voltage reduction (CVR) simulations as well as Energy Star and Consumer Technology Association communications protocol (CTA-2045) compliant controls, including energy storage calculations for “energy take”. Daily and weekly simulations are performed on a representative IEEE test feeder distribution system with experimental load and hot water draw (HWD) profiles to consider user comfort. Sequential controls are developed to reduce power spikes from controls and lead to peak shavings. It is found that EWHs are suitable for virtual power plant (VPP) operation with sustainable tank temperatures, i.e., average water temperature is maintained at set-point or above at the end of the control period while shifting up to 78% of EWH energy out of shed windows per day and 75% over a week, which amounts to up to 23% of the total load shifted on the example power system. While CVR simulations reduced the peak power of individual EWHs, the aggregation effect at the distribution level negates this reduction in power for the community. The EWH is shown as an energy constant load without consistent benefit from CVR across the example community with low energy reductions of less than 0.1% and, in some cases, increased daily energy by 0.18%.

Smart homes and virtual power plant (VPP) controls are growing fields of research with potential for improved electric power grid operation. A novel testbed for the co-simulation of electric power distribution systems and distributed energy resources (DERs) is employed to evaluate VPP scenarios and propose an optimization procedure. DERs of specific interest include behind-the-meter (BTM) solar photovoltaic (PV) systems as well as heating, ventilation, and air-conditioning (HVAC) systems. The simulation of HVAC systems is enabled by a machine learning procedure that produces ultra-fast models for electric power and indoor temperature of associated buildings that are up to 133 times faster than typical white-box implementations. Hundreds of these models, each with different properties, are randomly populated into a modified IEEE 123-bus test system to represent a typical U.S. community. Advanced VPP controls are developed based on the Consumer Technology Association (CTA) 2045 standard to leverage HVAC systems as generalized energy storage (GES) such that BTM solar PV is better utilized locally and occurrences of distribution system power peaks are reduced, while also maintaining occupant thermal comfort. An optimization is performed to determine the best control settings for targeted peak power and total daily energy increase minimization with example peak load reductions of 25+%.

The forecast and estimation of total electric power demand of a residential community, its baseload, and its heating ventilation and air-conditioning (HVAC) power component, which represents a very large portion of a community electricity usage, are important enablers for optimal energy controls and utility planning. This paper proposes a method that employs machine learning in a multi-step integrated approach. An LSTM model for total electric power at the main circuit feeder is trained using historic multi-year hourly data, outdoor temperature, and solar irradiance. New key temperature indicators, TmHAVC, corresponding to the standby zero-power operation for HVAC systems for summer cooling and winter heating are introduced using a V-shaped hourly total load curve. The trained LTSM model is additionally run with TmHVAC and zero irradiance inputs yielding an estimated baseload, which is representative of typical occupancy patterns. The HVAC power component is disaggregated as the difference between total and baseload power. Total power forecasts of an aggregated residential community as seen by major distribution lines are experimentally validated with a satisfactory MAPE error below 10% based on a 4-year dataset from a representative suburban community with more than 1800 homes in Kentucky, U.S. Discussions regarding the validity of the separation method based on combined considerations of fundamental physics, statistics, and human behavior are also included.