Building Occupancy Normalization via Sensor Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing people counting systems in buildings often provide inaccurate or erroneous occupancy data due to counting errors in multiple sensors, leading to implausible values such as non-zero occupancy at the end of the day or negative occupancy at certain times, which can result from exclusive reliance on sensor data without proper calibration and validation.
Innovation Solution
A system and method for determining normalized occupancy in a building by using sensors to count entries and exits at access points, with a controller that polls and processes data from people counting sensors to calculate accurate occupancy, includes historical calibration data to validate and adjust sensor readings, and manages resources based on normalized occupancy values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple people counting sensors are deployed to monitor occupancy in building zones, then coverage and data collection capability are improved, but counting errors and data accuracy deteriorate due to sensor errors accumulating and creating implausible values
Solution Approach 1:
The system uses historical calibration data as feedback to continuously validate and adjust sensor readings. The controller compares current sensor data against historical patterns and corrects deviations, ensuring accuracy is maintained even as coverage expands through multiple sensors.
Solution Approach 2:
Historical calibration data serves as an intermediary between raw sensor readings and final occupancy values. This intermediary layer processes and validates sensor data, filtering out counting errors before they propagate into inaccurate occupancy measurements.
2Speed
If sensor data is used exclusively to determine occupancy, then real-time monitoring capability is improved, but data reliability deteriorates due to counting errors producing implausible values such as negative occupancy or non-zero occupancy at end of day
Solution Approach 1:
The system performs preliminary calibration during off-hours when occupancy is known to be zero or minimal. This preliminary action establishes baseline accuracy and corrects sensor drift before regular occupancy monitoring begins, preventing implausible values without delaying real-time monitoring.
Solution Approach 2:
The controller continuously polls sensors and uses historical calibration data as feedback to validate readings in real-time. When sensor data deviates from expected patterns based on historical calibration, the system automatically adjusts or flags the data, maintaining reliability while preserving real-time monitoring capability.
3Measurement precision
If historical calibration data is collected and processed to validate sensor readings, then occupancy measurement accuracy is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The system performs self-calibration using its own historical data. The controller automatically polls sensors, collects data, compares readings against historical calibration patterns, and adjusts measurements without external intervention. This self-service approach improves accuracy while avoiding the complexity of external calibration systems.
Solution Approach 2:
Instead of requiring complex physical calibration equipment or manual adjustment mechanisms, the system creates a digital copy of calibration data stored in memory. This virtual calibration reference is repeatedly used to validate sensor readings, providing high accuracy through simple data comparison rather than complex hardware adjustments.
4Measurement precision
If sensor polling and data processing is performed continuously to maintain accurate occupancy data, then data accuracy is improved, but energy consumption and computational load increase
Solution Approach 1:
The controller polls sensors at periodic intervals rather than continuously, using historical calibration data to validate each reading. This periodic polling maintains measurement accuracy by regularly checking occupancy levels while significantly reducing energy consumption compared to continuous monitoring.
Data Source
AI summary
A device for determining normalized occupancy of one or more spaces in a building is disclosed. The device includes a memory and a processor coupled to the memory. The processor is configured to: poll one or more people counting sensors associated with access points to a defined region of the building to obtain counts data from the one or more people counting sensors for a specified time period and historical calibration data for the one or more people counting sensors; and process the counts data and the historical calibration data to determine a normalized occupancy of the defined region during the specified time period.