Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Rule-based models”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

71 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Adaptation to climate change through strategic integration of long fallow into cropping system in a dryland Mediterranean-type environment

The crop-growing region of Western Australia characterized by a Mediterranean-type climate is projected to become warmer and drier. Appropriate selection of crop sequences will be of importance to cope with these climatic changes for this largest grain-producing region of Australia. Through linking a widely used crop model (APSIM), 26 General Circulation Models (GCMs) with one Shared Socioeconomic Pathway (SSP585) and economic analysis, we explored how the climate change would affect dryland wheat cropping and whether/how long fallow (the practice of leaving a field out of production for an entire growing season) could be integrated into wheat cropping system in Western Australia. The potential adaptation of long fallow into wheat system was assessed with four fixed rotations (fallow-wheat, fallow-wheat-wheat, fallow-wheat-wheat-wheat, and fallow-wheat-wheat-wheat-wheat) and four flexible sowing rule-based rotations (the land was fallowed if sowing rule was not met), compared with continuous wheat. The simulation results at four representing locations show that climate change would have negative impacts on both yield and economic return of continuous wheat cropping in Western Australia. Wheat after fallow out-yielded and out-profited wheat after wheat under future climate. But integrating fallow into wheat cropping systems with the above fixed rotations would lead to yield and economic loss. By contrast, cropping systems in which fallowing took place when sowing condition could not be met at a certain time would achieve comparable yield and economic return to continuous wheat, with wheat yield being only 5% less than continuous wheat and the gross margin being $12ha -1 more than continuous wheat averaged across locations. We highlight strategic integration of long fallow into cropping system in a dryland Mediterranean-type environment would have a great potential to cope with future climate change. These findings can be extended into other Mediterranean-type cropping regions in Australia and beyond.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Sex Differences in Odds of Brain Metastasis and Outcomes by Brain Metastasis Status after Advanced Melanoma Diagnosis

Sex differences in cancer are well-established. However, less is known about sex differences in diagnosis of brain metastasis and outcomes among patients with advanced melanoma. Using a United States nationwide electronic health record-derived de-identified database, we evaluated patients diagnosed with advanced melanoma from 1 January 2011–30 July 2022 who received an oncologist-defined rule-based first line of therapy (n = 7969, 33% female according to EHR, 35% w/documentation of brain metastases). The odds of documented brain metastasis diagnosis were calculated using multivariable logistic regression adjusted for age, practice type, diagnosis period (pre/post-2017), ECOG performance status, anatomic site of melanoma, group stage, documentation of non-brain metastases prior to first-line of treatment, and BRAF positive status. Real-world overall survival (rwOS) and progression-free survival (rwPFS) starting from first-line initiation were assessed by sex, accounting for brain metastasis diagnosis as a time-varying covariate using the Cox proportional hazards model, with the same adjustments as the logistic model, excluding group stage, while also adjusting for race, socioeconomic status, and insurance status. Adjusted analysis revealed males with advanced melanoma were 22% more likely to receive a brain metastasis diagnosis compared to females (adjusted odds ratio [aOR]: 1.22, 95% confidence interval [CI]: 1.09, 1.36). Males with brain metastases had worse rwOS (aHR: 1.15, 95% CI: 1.04, 1.28) but not worse rwPFS (adjusted hazard ratio [aHR]: 1.04, 95% CI: 0.95, 1.14) following first-line treatment initiation. Among patients with advanced melanoma who were not diagnosed with brain metastases, survival was not different by sex (rwOS aHR: 1.06 [95% CI: 0.97, 1.16], rwPFS aHR: 1.02 [95% CI: 0.94, 1.1]). This study showed that males had greater odds of brain metastasis and, among those with brain metastasis, poorer rwOS compared to females, while there were no sex differences in clinical outcomes for those with advanced melanoma without brain metastasis.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Analysis of precooling optimization for residential buildings

To reduce peak demand and/or energy cost for residential buildings, optimal precooling strategies are becoming important as an alternative to rule-based precooling strategies that are intuitive but may not be optimal. Since precooling optimization is heavily dependent on a variety of factors such as the home thermal properties, HVAC system, weather, thermal comfort criteria, and utility rate structure, the individual and collective impact of these factors on precooling performance needs to be analyzed. In addition, since the indoor air temperature is affected by heat transfer to and from the interior wall surface, performance analysis in view of the interior wall surface temperature is also essential. Therefore, in this paper, an optimal precooling strategy that accounts for the aforementioned factors and utilizes a second-order thermal network model, is proposed. With this strategy, the HVAC on/off control signal that minimizes 24-hour energy cost while maintaining thermal comfort, is determined. Through extensive simulations, it is found that the proposed optimal precooling strategy is able to adapt to changing conditions and that having a sufficiently low interior wall surface temperature during precooling is critical for avoiding expensive on-peak operation. Here, the reason for the latter is that such a temperature indicates that enough “cooling energy” has been stored. It is also found that weather has the most dominant impact on the precooling performance, followed by home thermal condition, with the rated cooling capacity and utility rate structure having the least impact.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Commercial building HVAC demand flexibility with model predictive control: Field demonstration and literature insights

Model Predictive Control (MPC) for building Heating Ventilation and Air Conditioning (HVAC) systems is beginning to gain traction in the market, with a few controls companies incorporating it into their product offerings. However, it remains difficult to assess whether the energy cost savings are enough to justify the cost of MPC implementation for a particular building, given the limited number of reported demonstrations. For small commercial and residential buildings with relatively uniform systems, standardized approaches can help lower implementation costs. In contrast, for large buildings or district systems, the potential magnitude of cost savings could justify more customized solutions. Estimating the cost-effectiveness of MPC becomes more challenging for medium and large commercial buildings, where a one-size-fits-all solution may not be suitable, and the potential energy cost savings may be insufficient to justify a customized solution. To make MPC technology more appealing, incorporating additional value streams beyond energy efficiency alone can significantly increase its attractiveness. One such revenue stream is demand flexibility, in response to dynamic electricity prices, where MPC can leverage the thermal mass of the building to shift the load and support the grid. Building on an extensive literature review of MPC field studies focused on cost savings and demand flexibility, this paper presents the results of implementing MPC control in a large office building HVAC system in Berkeley, CA. Four different dynamic electricity price profiles were integrated into the MPC objective function to shift building demand while maintaining comfort, and field testing was performed with each price profile across four seasons. The results show potential for 40–65 % demand decrease percentage and up to 61 % annual cost savings compared to the existing rule-based control strategy, under the tested dynamic price scenarios. This paper also presents a sensitivity analysis on the cost savings with respect to the price profile variability, discusses the implementation effort for the price-responsive MPC, and compares the cost savings found in this study to those found in literature on the basis of dynamic price variability, or so-called Electricity Price Relative Standard Deviation.

Zanetti, Ettore↗
DOE OSTI2022

Carbon emission responsive building control: A case study with an all-electric residential community in a cold climate

In the United States, buildings account for 35% of total energy-related carbon dioxide emissions, making them important contributors to decarbonization. Carbon intensities in the power grid are time-varying and can fluctuate significantly within hours, so shifting building loads in response to the carbon intensities can reduce a building’s operational carbon emissions. This paper presents a rule-based carbon responsive control framework that controls the setpoints of thermostatically controlled loads responding to the grid’s carbon emission signals in real time. Based on this framework, four controllers are proposed with different combinations of carbon accounting methods and control rules. To evaluate their performance, we performed simulation studies using models of a 27-home, all-electric, net zero energy residential community located in Basalt, Colorado, United States. The carbon intensity data of four future years from the Cambium data set are adopted to account for the evolving resource mix in the power grid. Various performance metrics, including energy consumption, carbon emission, energy cost, and thermal discomfort, were used to evaluate the performance of the controllers. Sensitivity analysis was also conducted to determine how the control thresholds and intervals affect the controllers’ performance. Simulation results indicate that the carbon responsive controllers can reduce the homes’ annual carbon emissions by 6.0% to 20.5%. However, the energy consumption increased by 0.9% to 6.7%, except in one scenario where it decreased by 2.2%. Compared to the baseline, the change in energy cost was between –2.9% and 3.4%, and thermal discomfort was also maintained within an acceptable range. The little impact on energy cost and thermal discomfort indicates there are no potential roadblocks for customer acceptance when rolling out the controllers in utility programs.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Online transfer learning strategy for enhancing the scalability and deployment of deep reinforcement learning control in smart buildings

In recent years, advanced control strategies based on Deep Reinforcement Learning (DRL) proved to be effective in optimizing the management of integrated energy systems in buildings, reducing energy costs and improving indoor comfort conditions when compared to traditional reactive controllers. However, the scalability and implementation of DRL controllers are still limited since they require a considerable amount of time before converging to a near-optimal solution. This issue is currently addressed in literature through the offline pre-training of the DRL agent. However this solution results in two main critical issues: (1) the need to develop a building surrogate model to perform the training task, and (2) the need to perform a fine-tuning process over several training episodes to obtain a near-optimal control policy. In this context, this paper introduces an Online Transfer Learning (OTL) strategy that exploits two knowledge-sharing techniques, weight-initialization and imitation learning, to transfer a DRL control policy from a source office building to various target buildings in a simulation environment coupling EnergyPlus and Python. A DRL controller based on discrete Soft Actor–Critic (SAC) is trained on the source building to manage the operation of a cooling system consisting of a chiller and a thermal storage. Several target buildings are defined to benchmark the performance of the OTL strategy with that of a Rule-Based Controller (RBC) and two DRL-based control strategies, deployed in offline and online fashion. The strategy adopted for OTL emulates the real world implementation with a simulation process by implementing the transferred DRL agent for a single episode in the target buildings. Target buildings have the same geometrical features and are served by the same energy system as the source building, but differ in terms of weather conditions, electricity price schedules, occupancy patterns, and building envelope efficiency levels. The results show that the OTL strategy can reduce the cumulated sum of temperature violations on average by 50% and 80% respectively when compared to RBC and online DRL while enhancing the energy system operation with electricity cost savings ranging between 20% and 40%. Furthermore, the OTL agent performs slightly worse than the offline DRL controller but it does not require any modeling effort and can be implemented directly on target buildings emulating a real-world implementation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Model-based predictive control of multi-stage air-source heat pumps integrated with phase change material-embedded ceilings

This paper presents a model-based predictive control strategy to optimize the operations of phase change material (PCM) ceiling panels coupled with a multi-stage air-source heat pump. A three-stage prototype heat pump unit has been built and tested in the laboratory, with the low and medium stages designed for space heating/cooling and the high compression stage dedicated to charging of the PCM energy storage. To facilitate optimal control of the integrated heat pump system, a mixed-integer linear programming formulation is derived through linearization of the heat pump model and a mixed-integer reformulation of the PCM dynamic governing equations. A predictive control strategy is synthesized based on the resultant control formulation and implemented in a receding horizon scheme that optimizes the PCM charging and the zone temperature schedules simultaneously to leverage both the passive (associated with building construction materials) and active (PCM) storage capacities of a building. The control strategy has been tested along with three benchmarking control scenarios using a co-simulation platform for a prototypical detached house in Atlanta, GA. In this study, test results showed that application of the proposed control strategy to the PCM-integrated heat pump could provide 27.1% electricity cost savings while a fine tuned rule-based control strategy could achieve cost savings of 20.4%, compared to a baseline case without PCM storage, under a time-of-use rate tariff.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Operator learning for energy-efficient building ventilation control with computational fluid dynamics simulation of a real-world classroom

Energy-efficient ventilation control plays an important role in reducing building energy consumption while ensuring occupant health and comfort. While Computational Fluid Dynamics (CFD) simulations provide detailed and physically accurate representations of indoor airflow, their high computational cost limits their use in real-time building control. In this work, we present a neural operator learning framework that combines the physical accuracy of CFD with the computational efficiency of machine learning to enable building ventilation control with the high-fidelity fluid dynamics models. Our method jointly optimizes the airflow supply rates and vent angles to reduce energy use and adhere to air quality constraints. We train an ensemble of neural operator transformer models to learn the mapping from building control actions to airflow fields using high-resolution CFD data. This learned neural operator is then embedded in an optimization-based control framework for building ventilation control. Experimental results show that our approach achieves significant energy savings compared to maximum airflow rate control, rule-based control, as well as data-driven control methods using spatially averaged CO 2 prediction and deep learning–based reduced-order models, while consistently maintaining safe indoor air quality. These results highlight the practicality and scalability of our method in maintaining energy efficiency and indoor air quality in real-world buildings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

Enhanced Oblique Decision Tree Enabled Policy Extraction for Deep Reinforcement Learning in Power System Emergency Control

Deep reinforcement learning (DRL) algorithms have successfully solved many challenging problems in various power system control scenarios. However, their decision-making process is usually regarded as black-boxes. Furthermore, how DRL models interact with human intelligence remains an open problem. Thus, this paper proposes a policy extraction framework to extract a complex DRL model into an explainable policy. This framework includes three parts: 1) DRL training and data generation. We train an agent for a specific control task and generate data, which contains the control policy of the agent. 2) Policy extraction. We propose an information gain rate based weighted oblique decision tree (IGR-WODT) for DRL policy extraction. 3) Policy evaluation. We define three metrics to evaluate the performance of the proposed approach. A case study for the under-voltage load shedding problem shows that the IGR-WODT presents a performance enhancement compared with DRL, weighted oblique decision tree, and univariate decision tree. The proposed policy extraction method could provide an intuitive explanation of the neural network decision-making process to the dispatchers when making final decisions on power grid operation. Also, the resulted rule-based controller could replace the deep neural network-based controller in many field edge devices with limited computing resources, providing comparable performance.

deep reinforcement learning↗
DOE OSTI2026

Emerging anomaly detection techniques for electronic health records: A survey

Background Anomaly detection in electronic health records (EHRs) is a cornerstone of biomedical informatics, with direct implications for patient safety, clinical decision-making, and the prevention of healthcare fraud. Once guided primarily by simple rule-based methods, the field has advanced rapidly, driven by increased computing power, richer and more detailed health data, and the rise of machine learning and deep learning techniques. The objective of this paper is to provide a comprehensive overview of modern approaches to detecting anomalies in EHRs, outlining their strengths, limitations, and relevance to key healthcare challenges. We review traditional statistical methods alongside newer ML- and DL-based strategies and hybrid models, with particular attention to how these techniques support transparency and build clinical trust. Methods This paper presents a thorough and critical survey through systematic review (PRISMA-based) of the latest anomaly detection strategies in time-sequence data domains within electronic health record systems. Results We explore a broad spectrum of methodologies, including statistical models, supervised and unsupervised learning approaches, hybrid frameworks, and state-of-the-art ML-based techniques that collectively advance the precision and scalability of detecting anomalies in complex clinical datasets. In addition to mapping current capabilities, we address the enduring challenges that hinder widespread implementation and provide a forward-looking perspective on the future of anomaly detection in the data-rich landscape of modern healthcare. Summary The advancement in AI-based approaches is reported along with the basic principles of the individual approaches and their applicability. The increased availability of high-quality data, advancements in DL approaches, and enhanced computation power are leading to more frequent adaptation of DL-based approaches. Emerging DL-based approaches that have been adapted in other domains or recently applied in the EHR domain are also discussed in detail. Although DL-based approaches can improve model predictions by incorporating comorbidities, their application is limited in low-frequency data domains (e.g., when the total available data remains in the single digits). Therefore, the user must carefully consider the application based on data availability.

Anomaly detection↗
DOE OSTI2026

Optimizing pressurized-water reactor equilibrium cycle using a novel loading pattern encoding and rule-based genetic crossover operators

This work presents an extended multi-batch approach applied in shuffling scheme optimization for equilibrium cycle for pressurized water reactors using Genetic Algorithms (GAs). A new ruled based GA crossover operator called Inherited Location and Batch (ILB) was introduced to enhance offsprings reproduction efficiency specialized for equilibrium cycle optimization problem. This approach was implemented within the Plant ReLoad Optimization (PRLO) framework and validated using a generic reactor model based on the AP1000 design, with core parameters calculated via the CASMO/SIMULATE software package. The ILB approach is then applied for both single and multi-objective problems in maximizing cycle length and core average exposure while minimizing the average enrichment of the 57 fresh fuel assemblies (FAs) per cycle. The optimal solutions are selected based on their dominance to the objectives from all feasible solutions. This research identified three optimal solutions satisfied safety constraints: The first solution minimizes feed enrichment costs with a cycle length of 338.8 days and core exposure of 25.39 MWd/MT; the second solution extends cycle length to 361.2 days, with the highest core exposure of 26.84 MWd/MT, using 3.75 wt% average fuel enrichment; the third solution balances both objectives with a cycle length of 349.6 days, core exposure of 25.82 MWd/MT with a slight enrichment increase compared to the first solution. Collectively, these findings underscore the efficiency and effectiveness of the proposed approach in achieving practical multi-objective optimal equilibrium cycle designs using GAs optimizer.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Enhanced Detection of Primary Biological Aerosol Particles Using Machine Learning and Single-Particle Measurement

Accurately identifying primary biological aerosol particles (PBAPs) using analytical techniques poses inherent challenges due to their resemblance to other atmospheric carbonaceous particles. Here, we present a study of an enhanced method for detecting PBAPs by combining single-particle measurement with advanced supervised machine learning (SML) techniques. We analyzed ambient particles from a variety of environments and lab-generated standards, focusing on chemical composition for traditional rule-based and clustering approaches and incorporating morphological features into the SML approaches, neural networks and XGBoost, for improved accuracy. This study demonstrates that SML methods outperform traditional methods in quantifying PBAPs, achieving significant improvements in precision, recall, F1-score, and accuracy, leading to an increased number of detected PBAPs by at least 19%. The adaptability of the proposed XGBoost-based SML model is showcased in comparison to traditional methods in categorizing PBAPs for blind data sets from different geographical locations. Two field case studies were investigated, over agricultural land and Amazonia rain forest, representing relatively low and high concentrations of PBAPs, respectively, where XGBoost consistently detected up to 3.5 times more PBAPs than traditional methods. Precise detection of PBAPs in the atmosphere could significantly improve the prediction of climatic impacts by them.

42 ENGINEERING↗
DOE OSTI2026

Automated Signal Timing Plan Reconstruction Using High-Resolution Event-Based Controller Data for Digital Twins

Transportation digital twins are essential tools for evaluating emerging technologies such as connected and automated vehicles, adaptive traffic signal control, and mobility optimization strategies. Realistic digital twins require accurate emulation of real-world signal controllers and detailed signal timing plans. However, signal timing plans are often unavailable or difficult to access, forcing researchers and modelers to rely on assumed fixed timings or halt their analysis. To overcome this challenge, we present a method that directly estimates signal timing plan parameters using high-resolution, event-based data from traffic signal controllers. The proposed method extracts key parameters, including cycle length, offset, phase sequence, coordinated phases, phase-specific minimum and maximum green durations, vehicle extensions, and splits under coordination. A rule-based deterministic signal timing reconstruction algorithm based on traffic signal operation rules, such as those outlined in the Signal Timing Manual, is developed and validated. We evaluate this method, which uses high-resolution controller event logs and verified signal timing plans, on 94 signalized intersections in Nashville, Tennessee, demonstrating their ability to generate accurate, simulation-ready signal timing plans for tools such as SUMO and Vissim.

Saroj, Abhilasha [ORNL] (ORCID:0000000191178063)↗
DOE OSTI2024

Digital Analytics, Causal Knowledge Acquisition and Reasoning for Technical Language Processing

Complex engineering systems such as nuclear power plants (NPPs) generate and collect large amounts of equipment reliability (ER) data elements that contain information on the status of components, assets, and systems. Some of this information is textual in form and can be found in documents such as incident reports (IRs) and work orders (WOs). Analyses of textual data in current NPPs-using natural language processing (NLP) methods-have been expanded over the last decade, and it is only recently that the true potential of such analyses has emerged. So far, applications of NLP methods have mostly been limited to classification and prediction, the goal being to identify the nature of the textual element (e.g., safety or non-safety related). Here, we target a more complex problem: automatically extracting knowledge from a textual element in order to assist system engineers in conducting system health assessments. Knowledge extraction is a very broad concept, and its definition may vary depending on the application context. Our methods are a blend of both rule-based and machine learning (ML) algorithms. For our purposes, knowledge extraction means identifying the systems or assets mentioned in a given textual element, as well as the type of event described (e.g., component failure or maintenance activity). In addition, we want to capture details such as measured quantities and the temporal/cause-effect relations between events. In this tool, we also demonstrate how textual data elements are preprocessed in order to handle typos, acronyms, and abbreviations. One main feature of these methods is that they are not based solely on data, but are in fact model-based. In other words, they also rely on MBSE models that are designed to capture-from a functional point of view-the architecture of the systems/assets under consideration. The main purpose of such models is to digitally emulate system engineers' knowledge of system and asset architecture and to identify dependencies among systems, assets, and components. Provided these models, analyses of textual and numeric ER data can be performed by first identifying the OPM model elements to which the ER data elements are referring. The relationships between ER data elements are then identified by checking for any temporal or logical dependencies.

Mandelli, Diego [Idaho National Laboratory (INL), ↗
DOE OSTI2022

CANShield: Signal-based Intrusion Detection for Controller Area Networks

Modern vehicles rely on complex cyber-physical systems made up of hundreds of electronic control units (ECUs) connected through controller area network (CAN) buses. However, the CAN bus attack surface is increasing due to advanced features in automobiles, making it prone to injection attacks. The ordinary injection attacks disrupt the typical timing properties of the CAN data stream, and the rule-based intrusion detection systems (IDS) can easily detect them. However, advanced attackers can inject false data to the signal level, maintaining the regular pattern/frequency of the CAN messages. Such attacks can bypass the rule-based IDS or any anomaly-based IDS built on binary payload data. To make the vehicles robust against such intelligent attacks, we propose CANShield, a signal-based intrusion detection framework for the CAN bus that consists of three modules. A data preprocessing module handles the high-dimensional CAN data stream at the signal level and make them suitable for any machine learning model. A data analyzer module consists of multiple deep autoencoder networks, each analyzing the time series data from a different perspective. Finally, an attack detection module uses an ensemble method to make the final decision. Evaluation results on a standard signal-based dataset show the effectiveness of the CANShield in detecting five advanced attacks.

Shahriar, Md Hasan↗
DOE OSTI2024

Energy Saving Estimation of ASHRAE Guideline 36 Supervisory Setpoint Reset Controls in a Commercial Large Office Building

Designing, commissioning, and retrofitting HVAC control systems for energy efficiency is crucial, but the use of ad-hoc control sequences by designers and contractors, based on scattered information, results in diverse and sub-optimal sequences. ASHRAE Guideline 36 (G36) addresses the challenge by providing standardized, rule-based HVAC control sequences that prioritize energy efficiency. However, there is limited evaluation of their energy performance at the building level, with only a few studies primarily focused on HVAC airside systems in small-to-medium-sized commercial buildings. In this study, the energy performance of ASHRAE Guideline 36 control sequences was assessed using a large office building emulator in Chicago. The emulator features a central plant system with multiple chillers and boilers as well as multiple variable air volume (VAV) systems with terminal reheat. To achieve a high-fidelity representation, we developed a Spawn-of-EnergyPlus-based model for the large office building, maintaining the DOE prototype large office building setup but substituting the HVAC system with its Modelica counterpart. This substitution ensures that the building thermal load, HVAC system's dynamics, and detailed control sequences are all accurately represented. The study involved evaluating and implementing control strategies outlined in ASHRAE Guideline 36-2021 to replace conventional controls. These strategies include the demand-based supply air temperature and duct static pressure setpoint reset and the request logic for demand-based reset of chilled/hot water supply temperature setpoints and pipe static pressure setpoints. Energy performance was evaluated under various load conditions, including cooling, heating, and transitional seasons, both for individual control strategies and in combination. The results indicate that the collective control strategies retrofit yield greater energy savings than the sum of individual strategies, highlighting the synergistic benefits of incorporating both airside and plant-side control retrofits. Additionally, energy savings of up to 41% in the heating season, 18% in the shoulder season, and 20 % in the cooling season were observed compared to baseline control while maintaining the thermal comfort level.

ASHRAE Guideline 36, Commercial buildings, Control↗
DOE OSTI2026

A Physics-Informed Reinforcement Learning Framework for Economic-Thermal Co-Optimization of Crypto Mining Data Centers: Preprint

The rapid expansion of cryptocurrency mining has created a new class of high-density data centers characterized by extreme thermal flux and high sensitivity to volatile economic markets. Traditional thermal management strategies, typically reliant on rule-based control, maintain static setpoints that fail to account for fluctuating electricity prices and cryptocurrency values - factors critical to mining profitability. To address this, we present a physics-informed reinforcement learning (PIRL) framework for economic-thermal co-optimization in crypto mining data centers. This framework consists of a proximal policy optimization (PPO) agent, a virtual testbed powered by high-fidelity physics-based models, and an interactive frontend dashboard. The PPO agent is trained using the virtual testbed and strict hardware safety limits. This physics-informed approach allows the agent to learn a stochastic policy that dynamically balances mining revenue against operational costs by co-optimizing HVAC cooling setpoints and IT computational hashrate. The simulation results demonstrate that the integrated framework achieved an 8.62% increase in net operational profit compared to traditional baseline strategies while strictly adhering to safety-critical temperature constraints (coolant supply temperature < 32 degrees C). This work provides a scalable template for the deployment of reinforcement learning in mission critical facilities where economic volatility and physical safety must be managed simultaneously.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗