METHOD AND SYSTEM FOR CALIBRATING MEASURED VALUES FOR AMBIENT AIR PARAMETERS USING TRAINED MODELS

• Main Authors: KANJ, Houssam, BAKR, Mohamed, ALI, Shirook
• Format: Patent
• Language: English
• Published: 13.06.2024
• Subjects: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

Embodiments described herein generally relate to a method and system for calibrating measured values for ambient air parameters using trained models. In at least one embodiment, the method includes calibrating measured values for an ambient parameter (AP) by generating training dataset points, each of the training dataset point comprising: (i) a first measured value of an AP, generated by a first sensor, at a time instance, and (ii) a second, time-paired measured value for the AP, generated by at least one reference sensor, at the time instance, wherein the first and second values are measured in a localized area surrounding the at least one reference sensor; training an AP-specific calibration model using the training dataset points to generate a trained AP-specific calibration model; applying the trained AP-specific calibration model to a new measured value of the AP, generated by a second sensor.