Liang Zhang

Interests

Research

Dr. Zhang's research is broadly focused in the area of computing and modeling for smart buildings and cities to promote building energy efficiency, sustainability, and resilience. His research has emphasized the following topics: 1) Generative AI in smart buildings, 2) machine learning control in buildings, 3) fault detection and diagnostics in buildings, 4) smart and connected communities, 5) grid-interactive efficient building, and 6) building energy modeling (building, community, and urban scale)

Teaching

Building Energy Modeling

Courses

Scholarly Contributions

Books

• Zhang, L. (2018). Data-driven whole building energy forecasting model for data predictive control. Drexel University.

Journals/Publications

• Ma, N., Labib, R., Amor, R., Chong, A., Fan, C., Forth, K., Fu, X., Fuchs, S., Hong, T., Klimenkova, N., & others, . (2026). Ten questions concerning large language models (LLMs) for building applications. Building and Environment, 114260.
• Haroon, S. M., Zhang, L., & Ryan, A. (2025). How Commute Time and EV Ownership Shape Residential Cooling Energy Load Profiles. Findings.
• Hong, T., & Zhang, L. (2025).

  AI for building energy modeling: A transformation

  . Building Simulation, 18(9).
• Jiang, G., Ma, Z., Zhang, L., & Chen, J. (2025). Prompt engineering to inform large language model in automated building energy modeling. Energy, 316, 134548.
• Kim, J., Frank, S., Buechler, R., Mishra, S., Petersen, A., Zhang, L., & Eslinger, H. (2025). Performance evaluation of automated data-driven feature extraction and selection methods for practical and scalable building energy consumption prediction models. Journal of Building Engineering, 103, 112045.
• Liu, M., Zhang, L., Chen, J., Chen, W., Yang, Z., Lo, L. J., Wen, J., & O’Neill, Z. (2025).

  Large language models for building energy applications: Opportunities and challenges

  . Building Simulation, 18(2).
• Shen, Y., Ryan, A., Pan, M., Fu, X., & Zhang, L. (2025). Behavioral and infrastructure influences on electric vehicle charging and grid impact. Transportation Research Part D: Transport and Environment, 149, 105011.
• Zhang, L., & Chen, Z. (2025). Opportunities of applying Large Language Models in building energy sector. Renewable and Sustainable Energy Reviews, 214, 115558.
• Zhang, L., Ford, V., Chen, Z., & Chen, J. (2025). Automatic building energy model development and debugging using large language models agentic workflow. Energy and Buildings, 327. doi:10.1016/j.enbuild.2024.115116
  More info

  Building energy modeling (BEM) is a complex process that demands significant time and expertise, limiting its broader application in building design and operations. While Large Language Models (LLMs) agentic workflow have facilitated complex engineering processes, their application in BEM has not been specifically explored. This paper investigates the feasibility of automating BEM using LLM agentic workflow. We developed a generic LLM-planning-based workflow that takes a building description as input and generates an error-free EnergyPlus building energy model. Our robust workflow includes four core agents: 1) Building Description Pre-Processing, 2) IDF Object Information Extraction, 3) Single IDF Object Generator Suite, and 4) IDF Debugging Agent. These agents divide the complex tasks into manageable sub-steps, enabling LLMs to generate accurate and reliable results at each stage. The case study demonstrates the successful translation of a building description into an error-free EnergyPlus model for the iUnit modular building at the National Renewable Energy Laboratory. The effectiveness of our workflow surpasses: 1) naive prompt engineering, 2) other LLM-based workflows, and 3) manual modeling, in terms of accuracy, reliability, and time efficiency. The paper concludes with a discussion on the interplay between foundational models and LLM agent planning design, advocating for the use of fine-tuned, specialized models to advance this field.
• Zhang, L., Fu, X., Li, Y., & Chen, J. (2025). Large language model-based agent Schema and library for automated building energy analysis and modeling. Automation in Construction, 176, 106244.
• Jiang, G., Ma, Z., Zhang, L., & Chen, J. (2024). EPlus-LLM: A large language model-based computing platform for automated building energy modeling. Applied Energy, 367, 123431.
• Khadka, S. (2024). Interpretable Machine Learning Control in Building Energy Systems. ASHRAE Transactions, 130, 544--551.
• Khadka, S., & Zhang, L. (2024). Scaling Data-Driven Building Energy Modelling using Large Language Models. arXiv preprint arXiv:2407.03469.
• Zhang, L., & Chen, Z. (2024). Large Language Model-Based Interpretable Machine Learning Control in Building Energy Systems. arXiv preprint arXiv:2402.09584.
• Zhang, L., & Chen, Z. (2024). Large language model-based interpretable machine learning control in building energy systems. Energy and Buildings, 313, 114278.
• Zhang, L., Chen, Z., & Ford, V. (2024). Advancing Building Energy Modeling with Large Language Models: Exploration and Case Studies. arXiv preprint arXiv:2402.09579.
• Zhang, L., Haroon, S. M., & Ryan, A. (2024). Py-Cosim: Python-Based Building Energy Co-Simulation Infrastructure. SSRN.
• Zhang, L., Kaufman, Z., & Leach, M. (2024). Physics-informed hybrid modeling methodology for building infiltration. Energy and Buildings, 320, 114580.
• Zhang, L., & Chen, Z. (2023). Opportunities and Challenges of Applying Large Language Models in Building Energy Efficiency and Decarbonization Studies: An Exploratory Overview. arXiv preprint arXiv:2312.11701.
• Zhang, L., & Wen, J. (2017). A Systematic Feature Selection Procedure for Data-driven Building Energy Forecasting Model Development. ASHRAE Annual Conference.
• Zhang, L., Chen, J., & Zou, J. (2023). Taxonomy, Semantic Data Schema, and Schema Alignment for Open Data in Urban Building Energy Modeling. arXiv preprint arXiv:2311.08535.
• Zhang, L., Leach, M., Chen, J., & Hu, Y. (2023). Sensor cost-effectiveness analysis for data-driven fault detection and diagnostics in commercial buildings. Energy, 263, 125577.
• Chen, J., Zhang, L., Li, Y., Shi, Y., Gao, X., & Hu, Y. (2022). A review of computing-based automated fault detection and diagnosis of heating, ventilation and air conditioning systems. Renewable and Sustainable Energy Reviews, 161, 112395.
• Zhang, L., & Leach, M. (2022). Evaluate the impact of sensor accuracy on model performance in data-driven building fault detection and diagnostics using Monte Carlo simulation. Building Simulation, 15(5).
• Zhang, L., Chen, Z., Zhang, X., Pertzborn, A., & Jin, X. (2023). Challenges and Opportunities of Machine Learning Control in Building Operations. Building SImulation, TBD.
• Bae, Y., Bhattacharya, S., Cui, B., Lee, S., Li, Y., Zhang, L., Im, P., Adetola, V., Vrabie, D., Leach, M., & others, . (2021). Sensor impacts on building and HVAC controls: A critical review for building energy performance. Advances in Applied Energy, 4, 100068.
• Li, Y., O'Neill, Z., Zhang, L., Chen, J., Im, P., & DeGraw, J. (2021). Grey-box modeling and application for building energy simulations-A critical review. Renewable and Sustainable Energy Reviews, 146, 111174.
• Zhang, L. (2021). Data-driven building energy modeling with feature selection and active learning for data predictive control. Energy and Buildings, 252, 111436.
• Zhang, L., & Wen, J. (2021). Active learning strategy for high fidelity short-term data-driven building energy forecasting. Energy and Buildings, 244, 111026.
• Zhang, L., Alahmad, M., & Wen, J. (2021). Comparison of time-frequency-analysis techniques applied in building energy data noise cancellation for building load forecasting: A real-building case study. Energy and Buildings, 231, 110592.
• Zhang, L., Leach, M., Bae, Y., Cui, B., Bhattacharya, S., Lee, S., Im, P., Adetola, V., Vrabie, D., & Kuruganti, T. (2021). Sensor impact evaluation and verification for fault detection and diagnostics in building energy systems: A review. Advances in Applied Energy, 3, 100055.
• Zhang, L., Plathottam, S., Reyna, J., Merket, N., Sayers, K., Yang, X., Reynolds, M., Parker, A., Wilson, E., Fontanini, A., & others, . (2021). High-resolution hourly surrogate modeling framework for physics-based large-scale building stock modeling. Sustainable Cities and Society, 75, 103292.
• Zhang, L., Wen, J., Li, Y., Chen, J., Ye, Y., Fu, Y., & Livingood, W. (2021). A review of machine learning in building load prediction. Applied Energy, 285, 116452.
• Bianchi, C., Zhang, L., Goldwasser, D., Parker, A., & Horsey, H. (2020). Modeling occupancy-driven building loads for large and diversified building stocks through the use of parametric schedules. Applied Energy, 276, 115470.
• Im, P., Bae, Y., Cui, B., Lee, S., Bhattacharya, S., Adetola, V., Vrabie, D., Zhang, L., & Leach, M. (2020). Literature Review for Sensor Impact Evaluation and Verification Use Cases-Building Controls and Fault Detection and Diagnosis (FDD).
• Zhang, L. (2020). A pattern-recognition-based ensemble data imputation framework for sensors from building energy systems. Sensors, 20(20), 5947.
• Zhang, L., Frank, S., Kim, J., Jin, X., & Leach, M. (2020). A systematic feature extraction and selection framework for data-driven whole-building automated fault detection and diagnostics in commercial buildings. Building and Environment, 186, 107338.
• Zhang, L., & Wen, J. (2019). A systematic feature selection procedure for short-term data-driven building energy forecasting model development. Energy and Buildings, 183, 428--442.
• Zhang, L., & Wen, J. (2018). Apply Active Learning in Short-term Data-driven Building Energy Modeling.
• Zhang, L., Wen, J., Cui, C., Li, X., & Wu, T. (2016). Experiment Design and Training Data Quality of Inverse Model for Short-term Building Energy Forecasting.
• Zhang, L., Xu, P., Mao, J., Tang, X. u., Li, Z., & Shi, J. (2015). A low cost seasonal solar soil heat storage system for greenhouse heating: Design and pilot study. Applied Energy, 156, 213--222.

Proceedings Publications

• Fu, X., Zhang, L., & Khadka, S. (2025). Fine-tuning Data Curation for LLM-Based Smart Search and Recommendation Systems for Building Decarbonization. In ASHRAE Annual Conference, 2025, 131.
  More info

  The building sector accounts for one-third of U.S. carbon emissions, making decarbonization a critical priority. Effective data management is essential for this effort, as vast amounts of building-related information—such as codes, standards, and operational data—must be analyzed to inform decision-making. Recent advancements in AI and large language models (LLMs) have the ability to process, interpret, and generate text, making them effective for accessing relevant insights. However, foundational LLMs (e.g., GPT, LLaMA, T5, BERT) rely on general datasets that lack the specialized knowledge required for building decarbonization. Fine-tuning these models with domain-specific data presents a promising solution, though challenges remain in addressing missing, duplicate, inconsistent, or unstructured information. This paper proposes a training data generation and validation method for fine-tuning the foundation model to power search and recommendation systems specifically for building decarbonization. Our method leverages various techniques and tools (e.g., LangChain language model framework, FAISS (Facebook AI Similarity Search) library for efficient document embedding search and vector storage, Unstructured open-source library and API for our document extraction, and Hugging Face Hub to access LLMs) to query relevant data for pre-processing, incorporate it into an LLM, enhance contextual understanding, and provide more comprehensive insights through Question-Answer (QA) pairs and optimized text for fine-tuning. From this processed content, we generate QA pairs and the optimized text to fine-tune the LLMs for building energy and decarbonization. The initial results show that LLaMA LLM versions outperform DeeSeek versions with similar parameters, and larger models generally perform better, with fine-tuning on QA pairs being more efficient than on optimized text, though in 33% of cases, QA pairs are more effective for fine-tuning. The fine-tuned LLM offers search and recommendation services guiding stakeholders in designing and operating (nearly) net-zero emission buildings.
• Sharma, A., Tandel, J., Li, X., Wang, L., Fariha, A., Zhang, L., Naqvi, S., Riaz, I. B., Cao, L., & Zou, J. (2025). DataMorpher: Automatic Data Transformation Using LLM-Based Zero-Shot Code Generation. In 2025 IEEE 41st International Conference on Data Engineering (ICDE).
• Jiang, G., Zhang, L., & Chen, J. (2024). Eplus-LLM: A Novel Automated Building Simulation Platform Using Natural Language. In 2024 ASHRAE Annual Conference, 130.
  More info

  Building Performance Simulation (BPS) through Building Energy Models (BEMs) serves as a critical tool for various applications aimed at enhancing building energy efficiency and sustainability. However, the establishment of BEMs often proves to be labor-intensive and time-consuming due to requisite expertise in building science, building equipment, as well as software usage. To address these challenges and facilitate user-friendly human-machine interaction for automated building simulation, we introduce Eplus-LLM (EnergyPlus-Large Language Model). This innovative approach is built upon a fine-tuned large language model (LLM) designed to assist modelers in building design and simulation tasks. Leveraging the attention mechanism within the LLM, Eplus-LLM enables modelers to engage in natural language interaction, allowing the model to comprehend the modeler’s demands and map human language into precise simulation models. By using scripts to call simulation software application programming interfaces (APIs), i.e., using Python to invoke the EnergyPlus engine without the need to utilize EnergyPlus for simulation, Eplus-LLM automates the creation of building models and generates simulation results efficiently. Validation results demonstrate that our proposed Eplus-LLM can generate BEMs encompassing different geometries and various internal load settings. The generated model structure and simulation results align seamlessly with experts’ modeling, affirming the effectiveness and robustness of our approach. In this study, we customize the LLM for the purpose of automated building modeling for the first time, directly constructing building models from natural language. This approach significantly improves the accessibility of building simulation, offering modelers a simple and efficient means to interact with the simulation software and obtain building models and simulation results. Moreover, our research has the potential to serve as a prototype for applications in other fields, with the prospect of being disseminated publicly or implemented in practical business settings.
• Sharma, A., Li, X., Guan, H., Sun, G., Zhang, L., Wang, L., Wu, K., Cao, L., Zhu, E., Sim, A., & others, . (2023). Automatic Data Transformation Using Large Language Model-An Experimental Study on Building Energy Data. In 2023 IEEE International Conference on Big Data (BigData).
• Zhang, L., & Leach, M. (2022). Sensor Cost-Effectiveness Analysis for Data-Driven Fault Detection and Diagnostics in Commercial Buildings. In 2022 ASHRAE Annual Conference, 128.
  More info

  Data-driven building fault detection and diagnostics (FDD) is heavily dependent on sensors. However, common sensors from building automation systems are designed to enable basic building control sequences and are not optimized to maximize accuracy in FDD. Installing additional sensors that provide more detailed building system information is key to maximizing the performance of FDD solutions. In this paper, we present a sensor cost analysis workflow to quantify the economic implications of installing new sensors for FDD using the concept of sensor threshold marginal cost (STMC) in the simulation (EnergyPlus) environment. STMC does not represent actual sensor cost. Rather, it represents a target cost based on the economic benefit that would be realized through improved FDD performance and one or more specified economic criteria. We calculate STMCs for multiple possible fault types and use fault prevalence information to aggregate STMCs into a single dollar value to determine the cost-effectiveness of a potential sensor investment. We conducted a case study using Oak Ridge National Laboratory's Flexible Research Platform (FRP) test facility as a reference. The case study demonstrates the feasibility of the analysis and highlights the key cost considerations in sensor selection for FDD. The concept of STMC is used to evaluate the cost effectiveness of single sensors and sensor groups. The results indicate that non-energy benefits can outweigh energy benefits, depending on how improved comfort and reduced maintenance are valued. Second, while performance improves as the candidate sensor set grows, diminishing returns are likely to make larger sensor sets less cost effective. To improve FDD performance cost effectively, selecting only the few most impactful sensor is critical, and our cost analysis workflow is designed to serve that function.
• Pradhan, O., Pertzborn, A., Zhang, L., & Wen, J. (2020). Development and Validation of a Simulation Testbed for the Intelligent Building Agents Laboratory (IBAL) Using TRNSYS (VC-20-C056). In 2020 ASHRAE Virtual Conference.
• Zhang, L., Wen, J., & Chen, Y. (2017). Systematic Feature Selection Process Applied in Short-Term Data-Driven Building Energy Forecasting Models: A Case Study of a Campus Building. In Dynamic Systems and Control Conference, 58295.
• Liu, C., Xu, P., Chen, W., Zhang, L., & Li, W. (2015). Study of Urban Patterns Optimization Employing CFD Method: A Case Study of Chenjiazhen Experimental Ecological Community, Chongming, Shanghai. In Building Resilient Cities in China: The Nexus between Planning and Science: Selected Papers from the 7th International Association for China Planning Conference, Shanghai, China, June 29--July 1, 2013.
• Zhang, L., Xu, P., & Li, Z. W. (2014). Relationship between Energy Consumption and Service Level: A Survey of Class A Office Buildings in Shanghai. In Advanced Materials Research, 953.
• Zhang, L., Xu, P., Mao, J. C., & Tang, X. u. (2014). Design and application of a seasonal solar soil heat storage system applied in greenhouse heating. In Applied Mechanics and Materials, 672.

Others

• Horsey, H., Parker, A., Farthing, A., Dahlhausen, M., Praprost, M., Bianchi, C., Robertson, J., Horowitz, S., Zhang, L., Ringold, E., & others, . (2024, 1).

  ComStock™ 2024 Release 1 [SWR-19-33 and SWR-20-32]

  .
• Bae, Y., Cui, B., Joe, J., Im, P., Adetola, V., Zhang, L., Leach, M., & Kuruganti, T. (2020, 1). Sensor Impact on Building Controls and Automatic Fault Detection and Diagnosis (AFDD).
• Frank, S. M., Lin, G., Jin, X., Singla, R., Farthing, A., Zhang, L., & Granderson, J. (2019, 1). Metrics and methods to assess building fault detection and diagnosis tools.
• Horsey, H., Parker, A., Farthing, A., Dahlhausen, M., Praprost, M., Bianchi, C., Robertson, J., Horowitz, S., & Zhang, L. (2020, 1). ComStock™.
• Im, P., Bae, Y., Cui, B., Lee, S., Bhattacharya, S., Adetola, V., Vrabie, D., Zhang, L., & Leach, M. (2020, 1). Sensor Impacts Evaluation and Verification: Expert Interview Responses.
• Wilson, E. J., Parker, A., Fontanini, A., Present, E., Reyna, J. L., Adhikari, R., Bianchi, C., CaraDonna, C., Dahlhausen, M., Kim, J., & others, . (2022, March). End-Use Load Profiles for the US Building Stock: Methodology and Results of Model Calibration, Validation, and Uncertainty Quantification.