Built environment management systems and methods

By zoning buildings and using machine learning for continuous, micro-weather-based baselines, the system addresses the limitations of traditional methods, achieving accurate energy and carbon savings measurement and verification.

WO2026117467A1PCT designated stage Publication Date: 2026-06-04CARBONQUEST INC
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
CARBONQUEST INC
Filing Date
2025-11-21
Publication Date
2026-06-04

AI Technical Summary

Technical Problem

Traditional methods for measuring and verifying energy efficiency and carbon emissions in buildings are inadequate, as they rely on monthly utility bills and whole-building baselines, failing to account for nuanced changes and correlations rather than causations, which hinders financing and scalability of energy efficiency projects.

Method used

Implementing a system that breaks down buildings into zones, measures energy and carbon emissions continuously by hour, and uses machine learning to establish dynamic baselines based on micro-weather conditions, generating compact learned representations (CLRs) for time-series data to predict future consumption and detect anomalies.

Benefits of technology

This approach provides high-fidelity predictions and anomaly detection, enabling precise measurement of energy and carbon savings, facilitating financing and scaling of energy efficiency projects with a 20% margin of error.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2025056592_04062026_PF_FP_ABST
    Figure US2025056592_04062026_PF_FP_ABST
Patent Text Reader

Abstract

Computer-implemented methods for predicting baselines of time- series data are provided. The methods can include: using processing circuitry to generate a compact learned representation (CLR), calculating a similarity score; and using the processing circuitry to calculate a similarity score between a target CLR and a plurality of historical CLRs. Systems for dynamic baselining are also provided that can include: a data storage repository; a processing circuitry; and memory storing instructions that, when executed by the processor, cause the system to perform a computer-implemented method. Systems and methods for operating a building in the built environment are also provided, the systems and methods can include: a data storage repository; processing circuitry; and memory storing instructions that, when executed by the processor, cause the system to use the processing circuitry to generate a compact learned representation (CLR) for a time interval from a raw time-series data profile that includes External Environmental Data and Internal Environmental Data.
Need to check novelty before this filing date? Find Prior Art