June 7, 2023
Moxa Teams Up with Artificial Intelligence Provider to Improve Solar Energy Plant Efficiency and Reduce Labor – Automation.com

Moxa Teams Up with Artificial Intelligence Provider to Improve Solar Energy Plant Efficiency and Reduce Labor – Automation.com

  • November 01, 2021
  • Moxa Technologies Inc
  • News

    Photon system increased solar power generated by 10%, while reducing cost of labor by 20-30%.

Moxa Teams Up with Artificial Intelligence Provider to Improve Solar Energy Plant Efficiency and Reduce Labor

Nov. 1, 2021 – Taiwan has set an ambitious goal for its energy policy that by 2025, 20% of power generated will be from renewable resources. To meet this objective, solar energy is targeted to supply 20GW of power annually which is fifteen times more than the 1.3 GW it currently produces.
Between 1.3GW and 20GW are many obstacles for Taiwan’s solar industry. Moxa and thingnario, an industrial IoT intelligent platform provider in Taiwan, are helping overcome these obstacles by combining their joint resources to develop the first intelligent solar energy monitoring system, dubbed “Photon”. This solution brings together industrial-grade network devices with artificial intelligence (AI) to improve solar power efficiency in challenging outdoor environments, and to reduce labor costs.
Photon addresses the reality that many solar plants lack effective operation and maintenance (O&M) programs, one of the most critical ways to ensure that the solar power system realizes the highest possible generation. An O&M program maintains the plant infrastructure and equipment with the goal of improving the equipment’s life by preventing excess depreciation and impairment. When power generation is interrupted, solar plants have to spend significant amounts of money to dispatch personnel to fix equipment.

Overcoming O&M bottlenecks

To develop the Photon solution, the owners of one of Taiwan’s largest solar plants shared with Moxa and thingnario the complex operations that take place when generating solar energy across the island’s 13,800 square miles for its 23 million citizens. They explained that each photovoltaic power station transmits between 20,000 and 50,000 pieces of field data per minute. This is far too much data for a traditional operating system to manage, creating a serious problem for O&M programs. In addition, a standard maintenance system is hardware-based and relies heavily on the capability of the inverter. Working under these circumstances, solar plants did not have a holistic view of operational status through a single platform.
Determining whether generated power was being lost is also a challenge for solar plants. Plus, when losses do occur, it is time-consuming to troubleshoot an entire system to pinpoint the root cause. Maintenance personnel had to be dispatched, a problem that was magnified when there were a large number of sites. As personnel resources are normally limited, it often results in delays in identifying and rectifying the problem. Even the smallest of things can cause power losses. For example, twenty percent of a site’s power can be lost due to bird droppings covering the panels. Maintenance personnel may have to travel to far off remote areas where solar panels are located to determine the problem, return to headquarters to pick up the necessary equipment and parts, and then return to make the repairs. All in all, this process from discovery to rectification can take weeks, or even months.

Development of Photon System

Engineers from thingnario worked with Moxa on how to develop a solution by utilizing Artificial Intelligence (AI). That AI solution, now known as the Photon System, incorporated smart monitoring functions and operational mechanisms to ensure that solar power equipment is monitored effectively. Photon’s AI engine analyzes large volumes of historical and real-time sensor data to recognize patterns so that it can predict how much power will be generated in the next 5 to 30 minutes. It integrates data from the inverter with weather patterns to determine how solar panels will be affected. When there is a large discrepancy between how much power was predicted to be generated and the actual amount generated, the system will send an alert to notify operators to identify issues and perform preventative maintenance. The real-time alerts have helped plant operators increase power generated by ten percent and reduce labor costs by almost thirty percent.

Ensuring durability, reliability and longevity

Before deploying Photon, thingnario had to ensure the industrial computers it installed met project requirements. After careful analysis, Moxa UC Series Arm-based industrial computers were deemed the most appropriate since they are capable of withstanding harsh environments, maintain operation in temperatures from -40° to 70°C, and have an proven reputation within the solar industry. The computers are equipped with multiple serial ports to simplify connections connections with inverters, pyrheliometers, and other meters. Low power consumption also helps to reduce maintenance complexity.

About Moxa

Moxa is a leader in edge connectivity, industrial computing, and network infrastructure solutions for enabling connectivity for the Industrial Internet of Things. With over 30 years of industry experience, Moxa has connected more than 71 million devices worldwide and has a distribution and service network to serve customers in more than 80 countries.

Did you enjoy this great article?

Check out our free e-newsletters to read more great articles..


Source: https://www.automation.com/en-us/articles/november-2021/moxa-artificial-intelligence-solar-energy-plant?listname=Automation%20&%20Control%20News%20&%20Articles

Leave a Reply

Your email address will not be published. Required fields are marked *