Light Emitting Diode Distance & Ranging Environmental Monitoring System (LEMS)
Poster-Academic
Poster Presentation
Prepared by
Contact Information: [email protected]; (403) 915-5869
ABSTRACT
The Light Emitting Diode Distance and Ranging Environmental Monitoring System (LEMS) is a robust, low power, snow and river level monitoring sensor with embedded machine learning, object recognition, and near-field object discernment. When installed for the purpose of in-situ snow pack monitoring, the device provides the snow depth at 16 points along a transect and is capable of maintaining millimeter accuracy during harsh alpine conditions where such near-field particulates induce significant error in ultrasonic distance and ranging devices. When integrated with additional MET sensors, the embedded machine learning algorithm estimates the state of the snow pack and computes the probability of the onset of melt. For river level monitoring, turbidity, level, and discharge are calculated. The object recognition algorithm not only discards false readings from debris, but can be set to warn the user if debris exceeds a predefined size. This is particularly useful in protecting infrastructure such as hydroelectric dams, weirs and water treatment plants. Used individually the LEMS is an invaluable tool for hydrologic monitoring, however when multiple units are connected as a network, the LEMS becomes a powerful early flood warning tool. By integrating the snow pack morphology with river level fluctuation, emergency response teams can better assess flood risks in near real-time, reducing the number of premature or unnecessary evacuations.
Poster-Academic
Poster Presentation
Prepared by
Contact Information: [email protected]; (403) 915-5869
ABSTRACT
The Light Emitting Diode Distance and Ranging Environmental Monitoring System (LEMS) is a robust, low power, snow and river level monitoring sensor with embedded machine learning, object recognition, and near-field object discernment. When installed for the purpose of in-situ snow pack monitoring, the device provides the snow depth at 16 points along a transect and is capable of maintaining millimeter accuracy during harsh alpine conditions where such near-field particulates induce significant error in ultrasonic distance and ranging devices. When integrated with additional MET sensors, the embedded machine learning algorithm estimates the state of the snow pack and computes the probability of the onset of melt. For river level monitoring, turbidity, level, and discharge are calculated. The object recognition algorithm not only discards false readings from debris, but can be set to warn the user if debris exceeds a predefined size. This is particularly useful in protecting infrastructure such as hydroelectric dams, weirs and water treatment plants. Used individually the LEMS is an invaluable tool for hydrologic monitoring, however when multiple units are connected as a network, the LEMS becomes a powerful early flood warning tool. By integrating the snow pack morphology with river level fluctuation, emergency response teams can better assess flood risks in near real-time, reducing the number of premature or unnecessary evacuations.
