Solar energy is a key renewable energy source; however, its intermittent nature and potential for use in distributed systems make power prediction an important aspect of grid integration. This research analyzed a variety of machine learning techniques to predict power output for horizontal solar panels using 14 months of data collected from 12 northern-hemisphere locations. We performed our data collection and analysis in the absence of irradiation data—an approach not commonly found in prior literature. Using latitude, month, hour, ambient temperature, pressure, humidity, wind speed, and cloud ceiling as independent variables, a distributed random forest regression algorithm modeled the combined dataset with an R² value of 0.94. As a comparative measure, other machine learning algorithms resulted in R² values of 0.50–0.94. Additionally, the data from each location was modeled separately with R² values ranging from 0.91 to 0.97, indicating a range of consistency across all sites. Using an input variable permutation approach with the random forest algorithm, we found that the three most important variables for power prediction were ambient temperature, humidity, and cloud ceiling. The analysis showed that machine learning potentially allowed for accurate power prediction while avoiding the challenges associated with modeled irradiation data.

The present research leverages prior works to automatically estimate wall and ceiling R-values using a combination of a smart WiFi thermostat, building geometry, and historical energy consumption data to improve the calculation of the mean radiant temperature (MRT), which is integral to the determination of thermal comfort in buildings. To assess the potential of this approach for realizing energy savings in any residence, machine learning predictive models of indoor temperature and humidity, based upon a nonlinear autoregressive exogenous model (NARX), were developed. The developed models were used to calculate the temperature and humidity set-points needed to achieve minimum thermal comfort at all times. The initial results showed cooling energy savings in excess of 83% and 95%, respectively, for high- and low-efficiency residences. The significance of this research is that thermal comfort control can be employed to realize significant heating, ventilation, and air conditioning (HVAC) savings using readily available data and systems.

A conventional ground-coupled heat pump (GCHP) can be used to supplement heat rejection or extraction, creating a hybrid system that is cost-effective 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.

Smart WiFi thermostats have moved well beyond the function they were originally designed for; namely, controlling heating and cooling comfort in buildings. They are now also learning from occupant behaviors and permit occupants to control their comfort remotely. This research seeks to go beyond this state of the art by utilizing smart WiFi thermostat data in residences to develop dynamic predictive models for room temperature and cooling/heating demand. These models can then be used to estimate the energy savings from new thermostat temperature schedules and estimate peak load reduction achievable from maintaining a residence in a minimum thermal comfort condition. Back Propagation Neural Network (BPNN), Long-Short Term Memory (LSTM), and Encoder-Decoder LSTM dynamic models are explored. Results demonstrate that LSTM outperforms BPNN and Encoder-Decoder LSTM approach, yielding and a MAE error of 0.5 °C, equal to the resolution error of the measured temperature. Additionally, the models developed are shown to be highly accurate in predicting savings from aggressive thermostat set point schedules, yielding deep reduction of up to 14.3% for heating and cooling, as well as significant energy reduction from curtailed thermal comfort in response to a high demand event.

Despite the climate commitments made by countries in the Paris Climate Agreement adopted in 2015 and reinforced during COP 21 and with notably less success during COP 22, world carbon emissions increased in both 2021 and 2022. It is increasingly unlikely that the world will achieve the targeted 50% carbon reduction by 2030, the reduction approximately needed for reducing global temperature rise since the beginning of the Industrial Revolution to less than 1.5 deg. C. At the same time, there remain nearly 2 billion people in the world who have no or highly unreliable access to power. In developed countries, access to both clean energy and energy efficiency investment in residences within low to moderate income communities has also lagged. This paper provides a review of the “Productive Use of Energy (PUE)”, which is a means to add value to solar energy mini- and micro-grids to ensure investment worthiness and add more value to the communities being served. In this context, it posits an opportunity to leverage Bitcoin mining as a common PUE strategy applicable to new solar installations. Several actual pilot cases are described to demonstrate this potential throughout the world and at multiple scales. These include: (i) existing micro-grids with significant stranded energy to generate income that could be used to reduce the cost per kWh for the community; (ii) new solar micro-grids optimized to meet community load and mining operations; (iii) dedicated solar-powered Bitcoin mining mini-grids developed solely to create a funding stream for self-investment by communities for their benefit; and (iv) a low-income residential solar-powered Bitcoin miner to reduce the energy cost burden for residents. Several of these scenarios show significant potential to aid investment worthiness.