In the United States, all residential scale unitary heat pumps must be rated according to AHRI (Air-Conditioning, Heating, & Refrigeration Institute) standards to provide rated full-load heating and cooling capacities at a total of three outdoor temperatures, a Heating Seasonal Performance Factor (HSPF), a Seasonal Energy Efficiency Ratio (SEER) for cooling, and a full-load cooling Energy Efficiency Ratio (EER) at a single outdoor temperature. This is the only performance data that is independently verified. However, building performance simulation models require heat pump capacity and efficiency performance characteristics at any combination of compressor speed, outdoor air conditions, and indoor air conditions. This paper describes a unified modeling approach for single-speed, two-speed, and variable-speed heat pump technologies that are rated under the AHRI standard. A major focus of the work is the data analysis of the voluntary performance data provided by manufacturers to the Northeast Energy Efficiency Partnerships (NEEP) Cold Climate Air Source Heat Pump List for over 130,000 variable capacity heat pump products. This analysis informs assumptions for the modeled performance of variable-speed units when just the verified AHRI ratings are available. For single- and two-speed heat pump systems, a congruent approach was developed based on previous analysis of a sample of manufacturer-provided expanded performance data.

This paper presents a hybrid simulation platform for training and evaluating reinforcement learning (RL) controllers for the energy optimization of commercial buildings. The platform integrates dynamic models of HVAC systems and building envelopes, providing a realistic and flexible simulation environment for RL algorithm development. The platform's modular design allows for easy customization of HVAC system and building configurations, reducing the effort required to create simulation models. The simulator is sufficiently accurate and relatively simple, enabling efficient training and evaluation of RL agents with minimal computational resources. A case study demonstrates the effectiveness of the platform in training an RL controller to optimize energy consumption while maintaining indoor thermal comfort. The developed platform has the potential to accelerate the adoption of RL algorithms for building energy optimization, enabling significant energy savings and improved indoor environmental quality. The platform's features, including template-based modeling and flexible simulation capabilities, make it an attractive solution for researchers and practitioners seeking to develop and deploy RL controllers for building energy systems.

Accurate building load forecasting is critical for improving energy efficiency in smart construction systems and data-driven facility management. While deep learning models, such as long short-term memory (LSTM), have shown promising performance in capturing complex temporal patterns, their effectiveness relies heavily on the quality and relevance of the input features. In this paper, we propose a Principal Variate Selection (PVS) forecasting pipeline that integrates Bayesian Optimization to identify the optimal feature subset for multivariate LSTM models. Although multivariate models generally outperform univariate ones, incorporating all available features can lead to exponential increases in computational costs, thereby degrading model performance. Our approach ensures superior predictive accuracy compared to univariate baselines while reducing the dimension of inputs, thereby improving both accuracy and training efficiency. The pipeline supports deployment in scenarios where data acquisition is costly or constrained, such as early-stage smart building projects or resource-limited environments. This work presents a generalizable method for enhancing prediction quality while optimizing building load data collection, thereby providing actionable insights for achieving more precise and reliable load forecasts that address specific energy management needs.

Small and medium-sized commercial buildings (SMCBs) represent 94% of U.S. commercial buildings but encounter substantial obstacles in adopting Building Energy Management (BEM) systems. Current approaches exhibit fundamental limitations: vendor-specific API platforms restrict interoperability through proprietary ecosystems; commercial automation software demands extensive technical expertise and licensing costs; open-source IoT solutions lack native support for building automation protocols and semantic models. This paper introduces a configuration-driven web interface framework addressing the gap between smart device advancements and accessible BEM software infrastructure for SMCBs. The framework leverages VOLTTRON middleware integrated with an automated converter that processes unified YAML configurations into heterogeneous system files, reducing required configuration artifacts from six separate files to a single unified specification. The system architecture enables vendor-agnostic operation through BACnet and Modbus protocols while supporting semantic building model integration via automated Brick Schema parsing. Configuration-driven interfaces automatically adapt to diverse HVAC types without custom development. Simulation-based validation using BOPTEST demonstrates automatic interface generation between fan coil and hydronic systems, with the automated converter successfully generating all platform-specific outputs from the single YAML input. The result demonstrates the framework's capability to streamline BEM system deployment through reduced configuration complexity. This work bridges simulation capabilities with operational deployment, demonstrating how virtual testbeds validate generalizable software frameworks for real-world building automation.