Tokyo – February 13, 2018 – V-cube Robotics, Inc. (Headquarters: Shibuya, Tokyo; Representative Director and President: Taishin Demura; hereinafter “V-cube Robotics”), a provider of commercial drone solutions to enterprise and local governments, announced that it would begin offering SOLAR CHECK, a solar power plant inspection package service utilizing drones. Equipped with infrared thermography cameras, the drones autonomously fly around a facility, imaging the solar panels, which are then analyzed using AI, dramatically improving the efficiency of inspections and making it possible to discover discrepancies and report on the results of an inspection in a short amount of time.
About the service
The SOLAR CHECK drone’s flight path is set ahead of time to match the positions of the plant’s solar panels. It then autonomously flies along that path and images the panels from the air using its on-board thermography camera. The saved panel images are uploaded into the cloud, where they are automatically analyzed using deep learning to find discrepancies such as hotspots. The service also creates a report on the results of the inspection, showing the abnormal panels and where the discrepancy occurred.
Initial preparations include configuring the system as required for operation, test imaging and training, which V-cube Robotics perform after a customer applies for the service. From the second inspection on, the customer can perform the inspections themselves with an easy operation.
- Inspection time is dramatically reduced
Solar panel inspections that required each and every panel to be checked manually can now be done in a fraction of the time by imaging them with an autonomous drone. A manual inspection of a 2 MW solar power plant that used take 4 days can now be completed in 15 minutes with SOLAR CHECK.
- Accurate fault detection
SOLAR CHECK uses drones to take images matched to the angle of each panel, improving the accuracy of discrepancy detection, including hotspots, which are easily overlooked in conventional manual inspections. The service can be easily configured for steep hillsides and other areas that are difficult to inspect.
- Inspections can be made more often
Conventional solar panel inspections using drones required someone from outside of the organization to operate the drone, which restricted the frequency of inspections. SOLAR CHECK simplifies and automates drone operation, even making it possible for employees without special skills to perform inspections frequently. The service also allows companies to manage past inspection results, enabling them to see how their facilities are aging with time.
- Automatic analysis using AI, fast report creation
An issue with conventional drone inspections is that even though they reduced the imaging workload, matching images containing a discrepancy with the appropriate panel took an extraordinary amount of time. SOLAR CHECK uses deep learning to automatically analyze images for discrepancies and identify the affected panels, providing highly accurate reports in a short amount of time.*1
Image Recognition Using Deep Learning
SOLAR CHECK uses advanced image analysis technology that employs deep learning and other forms of AI offered by Datasection Inc. (hereinafter “Datasection”) to provide extremely accurate analyses.
One of Datasection’s strengths is its technology that processes large volumes of images instantly utilizing MLFlow, a platform it developed independently. The company applied this technology to the development of an algorithm tailored to detect discrepancies in solar panels specifically for the SOLAR CHECK service. It then developed this into a framework that improves detection accuracy using an AI algorithm that gets smarter the more it runs and that includes multiple deep networks in addition to existing image processing.
Example of How the Service is Used
Service Price Plan
- Individual estimates provided depending on the number of power plants and their size.
- Customers with compatible drones can choose to only use the SOLAR CHECK application
About Datasection, Inc.（http://www.datasection.co.jp）
Datasection carries out business marketing research, data provision and processing, image analysis, overseas social media, and BI & business data optimization based on the data and know-how it has accumulated over many years of analyzing social big data.
In recent years the company has also been devoting its energies to areas that make use of machine learning (primarily deep learning technologies), which includes AI image analysis, and is making steady progress based on the company’s strengths of image recognition and analysis.
About V-cube Robotics, Inc.（http://www.vc-robotics.com/）
Founded in October 2015, V-cube Robotics develops solutions combining video communications and drone computing technologies in the robotics field. The company is committed to creating environments that enable anyone to easily and safely control drones without professional knowledge or skills to perform real-time monitoring and communication from remote/multiple locations, allowing them to make required decisions instantly. V-cube Robotics aims to create a world where drones are active as an ordinary part of society by further expanding the possibilities of communications, and entrusting them with many day-to-day operations.
ENG: Please reference the following weblink.
V-Cube Robotics Begins Offering SOLAR CHECK, a Solar Power Plant Inspection Package Service