Supply chain
Thingnario uses AI to schedule PV module cleaning
Chloe Liao, Taipei; Adam Hwang, DIGITIMES

Thingnario has developed Photon, an AI-enabled PV monitoring solution that can determine the best time for cleaning modules of power stations and rooftop systems.

Operators of PV power stations and rooftop systems usually rely on labor-based inspection of PV modules and clean them twice a year, the startup firm said, adding such inspection takes much time and cannot be used to determine the best time for cleaning.

Photon works on a predicative model built using operating data collected from many existing PV power stations and rooftop PV systems, according to Thingnario. Photon monitors a target PV power station or rooftop PV system by collecting power generation and weather data to predict operating conditions via AI-based deep learning. If actual operations deviate from the prediction, Photon will tell the operator it is time to clean PV modules.

A 487KWp PV power station or rooftop PV system enabled by Photon can bring in electricity sales of over NT$1 million (US$32,400) half-yearly, Thingnario said. But manually-inspected systems of the same capacity can generate sales of only NT$880,000 half-yearly, Thingnario noted. In addition, Photon can recommend separate time to be taken in cleaning PV modules in different zones of a PV power-generating station to reach the most efficient cleaning on the whole, Thingnario indicated.

Photon's smart monitoring can detect abnormal operating conditions for timely repair or early maintenance, Thingnario said.

© 2018 DIGITIMES Inc. All rights reserved.
Please do not republish, publicly broadcast or publicly transmit content from this website without written permission from DIGITIMES Inc. Please contact us if you have any questions.