Energym: A Building Model Library for Controller Benchmarking

Abstract

We introduce the Python-based open-source library Energym, a building model library to test and benchmark building controllers. The incorporated building models are presented with a brief explanation of their function, location and technical equipment. Furthermore, the library structure is described, highlighting the necessary features to provide the benchmarking and control capabilities, i.e., standardized evaluation scenarios, key performance indicators (KPIs) and forecasts of uncertain variables. We go on to characterize the evaluation scenarios for each of the models and give formal definitions of the KPIs. We describe the calibration methodologies used for constructing the models and illustrate their usage through examples.

Beobench

“…We now discuss some recent frameworks, focusing on ones that target single-agent control, are open-source, under active development 2 , and accessible via a Python API. Frameworks that fit these criteria include BOPTEST 3 [2,3], Energym [16], Sinergym [8], and Reinforcement Learning Testbed for Power-Consumption Optimization [13]. For brevity, we will refer to the latter just as RL Testbed.…”

“…1. BOPTEST: We were able to count 3 distinct building models for BOPTEST: a. BESTEST Case 900 room model (used by testcases bestest_air, bestest_hydronic, bestest_hydronic_heat_pump) b. Multi-zone residential hydronic model (used by testcase multizone_residential_hydronic) c. Single-zone commercial building model (use by testcase singlezone_commercial_hydronic) Data for BOPTEST was taken from the BOPTEST repository 16 . There are 5 "testcases" based on 3 Buildings as listed above.…”

“…The focus on individual buildings could also lead to overfitting and weaken any subsequent performance claims. In the second phase of work, [2,3,8,13,16,20], testing frameworks were introduced, allowing different RL algorithms to be tested on a standard set of building environments. Given the complexity of developing high-fidelity building simulations, these frameworks each provide a limited set of simulated buildings.…”

“…2. Integrations of three popular building control frameworks into this package, namely BOPTEST [2,3], Energym [16], and Sinergym [8].…”

Benchmarking HVAC controller performance with a digital twin

“…Consequently, numerous evaluation metrics are described in the literature [2], which report a wide range of improvements on buildings' economic and environmental performances. As of 2022, lack of adequate benchmarking datasets and platforms continues to be a major challenge for evaluating the performance of BACS [6]. However, valuable contributions have greatly facilitated the reproducibility of experiments and evaluations.…”