Using Google Trends as a proxy for occupant behavior to predict building energy consumption

@article{Fu2022UsingGT,
  title={Using Google Trends as a proxy for occupant behavior to predict building energy consumption},
  author={Chunlei Fu and Clayton Miller},
  journal={ArXiv},
  year={2022},
  volume={abs/2111.00426}
}

Limitations of machine learning for building energy prediction: ASHRAE Great Energy Predictor III Kaggle competition error analysis

TLDR
This analysis reveals the limitations for machine learning using the standard model inputs of historical meter, weather, and basic building metadata and forms the foundation for suggestions to reduce machine learning errors by collecting and using additional training data from onsite and web-based sources to improve the capability, accuracy, scalability, and usability of machine learning.

Building Energy Prediction Models and Related Uncertainties: A Review

Building energy usage has been an important issue in recent decades, and energy prediction models are important tools for analysing this problem. This study provides a comprehensive review of

Gradient boosting machines and careful pre-processing work best: ASHRAE Great Energy Predictor III lessons learned

The ASHRAE Great Energy Predictor III (GEPIII) competition was held in late 2019 as one of the largest machine learning competitions ever held focused on building performance. It was hosted on the

Priority of Water Allocation during Drought Periods: The Case of Jaguaribe Metropolitan Inter-Basin Water Transfer in Semiarid Brazil

Inter-basin water transfers are the root of many conflicts, and water scarcity accentuates them. Those conflicts involve the priority of water use between regions. The Jaguaribe Metropolitan system,

References

SHOWING 1-10 OF 63 REFERENCES

Limitations of machine learning for building energy prediction: ASHRAE Great Energy Predictor III Kaggle competition error analysis

TLDR
This analysis reveals the limitations for machine learning using the standard model inputs of historical meter, weather, and basic building metadata and forms the foundation for suggestions to reduce machine learning errors by collecting and using additional training data from onsite and web-based sources to improve the capability, accuracy, scalability, and usability of machine learning.

More Buildings Make More Generalizable Models - Benchmarking Prediction Methods on Open Electrical Meter Data

TLDR
This paper outlines a library of open-source regression techniques from the Scikit-Learn Python library and describes the process of applying them to open hourly electrical meter data from 482 non-residential buildings from the Building Data Genome Project, showing that there is no one size-fits-all modeling solution and that various types of temporal behavior are difficult to capture using machine learning.

Islands of misfit buildings: Detecting uncharacteristic electricity use behavior using load shape clustering

TLDR
A modular methodology for identifying potentially mislabeled buildings using distance-based clustering analysis based on hourly electricity consumption data and seeks to segment buildings according to their daily behavior and predict which ones are misfitsaccording to their assigned PSU label is proposed.

Prediction Hourly Building Energy Use : The Great Energy Predictor Shootout - Overview and Discussion of Results

Analysis of measured data from buildings has become increasingly important during the past half-decade for reasons ranging from the needs of diagnostic expert systems to predicting the efficacy of

Development of an ANN-based building energy model for information-poor buildings using transfer learning

TLDR
A transfer learning-based ANN model for one-hour ahead building energy prediction to fill the research gap and demonstrates that transfer learning can effectively improve the accuracy of BPNN-based building energy models for information-poor buildings with very limited training data.
...