System for efficient use of surplus solar power

“The deciding factor was its versatility: we can install our own custom applications flexibly.”

— Satoshi Iwasaki, Executive Officer, i GRID SOLUTIONS Inc. —

Industry

Power (renewable energy)

Products

Edge Gateway (Indoor) AG10

Application

Built into an energy management platform for solar power

Challenges

• Buying a router and a server separately raises the unit cost and lowers quality
• The devices are installed outdoors, and it was hard to find one that handles both low and high temperatures, from -20°C to 60°C

Results

• One device provides both the LTE router and server functions, for a low-cost, high-quality system
• Custom applications run on a highly versatile and expandable device
• A wide operating temperature range allows outdoor installation

Overview

i GRID SOLUTIONS Inc. is an energy service provider that offers integrated decarbonization solutions through its virtual power plant (VPP), energy management, and power supply businesses. Building on its experience with visualizing power usage and forecasting electricity consumption, the company added power generation forecasting to create a surplus power forecasting model. With it, the company released the new “R.E.A.L. New Energy Platform®.”

The “R.E.A.L. New Energy Platform®” is an energy management platform that forecasts surplus power with high accuracy using AI and controls DER (Distributed Energy Resources) automatically and in fine detail. In recent years, more businesses, such as supermarkets and company sales offices, have installed solar panels for their own use. However, solar power output is unstable, and these businesses often struggle with unexpected surplus power or power shortages. To solve this problem, i GRID SOLUTIONS provides the R.E.A.L. New Energy Platform®, in which AI forecasts the balance of power supply and demand and controls generation and discharge accordingly.

The R.E.A.L. New Energy Platform® manages the generation equipment and storage batteries at a site, grid power, and even the charging and discharging of the EVs the site uses (for a supermarket, its delivery vehicles), all in one place, to make the best use of power. The platform measures power generation with data from solar power controllers and on-site sensors such as pyranometers. It manages the power stored in batteries and EVs and continuously monitors the site’s power demand. This data is collected in a system built in the cloud, where AI makes decisions and controls the charging and discharging of power equipment and adjusts the power usage of site equipment.

The role of amnimo products

In the R.E.A.L. New Energy Platform®, an application running on Ubuntu on the amnimo Edge Gateway AG10 works with the cloud system. Together, they send data from solar power facilities at many sites to the cloud and deliver control signals from the cloud, for example to charge and discharge storage batteries. The diagram below shows the on-site IoT system, built around the Edge Gateway AG10.

Solar power controllers, battery controllers, and sensors such as pyranometers connect to the Edge Gateway AG10 through interface converters. An application running on the AG10 sends on-site measurement data to the cloud and passes control signals from the cloud to the on-site equipment.

The Edge Gateway AG10, which handles data exchange with the equipment at each site, is installed near the solar panels, EV charging spots, and similar equipment.

Development challenges

The hardest part of developing the system was that the equipment connected at each power plant varied. For example, the interfaces that equipment uses to output data differ from site to site. In this system, we therefore connected remote I/O units between the edge gateway and the sensors. The remote I/O receives and converts signals such as RS-485, analog (4–20 mA), and pulse, and outputs them to the Edge Gateway AG10 over Ethernet.

For developing the custom application on the Edge Gateway AG10, amnimo also gave us quick support whenever we got stuck, so the work went smoothly.

Results

1. One device provides both LTE router and server functions, for a low-cost, high-quality system

At first, we considered separate products for the router and the server. But buying them separately raises the unit cost and also reduces ruggedness. Then we learned that the amnimo “Edge Gateway AG10” provides both functions in one device. It is a router with LTE support, yet it has enough computing resources to run a server function that collects data from each sensor and sends it to the cloud. This let us simplify the equipment configuration and build a low-cost, high-quality system.

2. High versatility and expandability

To run our custom applications on an edge device, the “Edge Gateway AG10,” which lets us develop programs on Ubuntu, was very attractive. I think its greatest strength is its versatility and expandability, which let us respond flexibly to change, including in device management.

3. Wide operating temperature range

Where the devices are installed, a water-resistant case keeps rain off. But the location is outdoors, so the surrounding temperature changes widely, just like the outside air. The amnimo “Edge Gateway AG10” runs applications reliably over a wide operating temperature range, so we feel it is very well suited to outdoor use like this. In fact, we needed a device that could handle both low and high temperatures, from -20°C to 60°C, and there were almost no options other than the “Edge Gateway AG10.”

Future plans

The platform currently operates at 390 sites across Japan (as of July 2022). We aim to sign contracts for 250 MW and 1,500 sites by 2024. Compared with wind power and biomass power, which tend to need a large initial investment, solar power is a field where demand is growing. Distributed generation management systems like the R.E.A.L. New Energy Platform® are a new type of service that has not yet spread in the market. As a front-runner in the industry, we want to meet the needs of each business.

Currently, we use another company’s AI service for AI processing, but we are also considering AI processing at the edge. We have heard that amnimo is developing an “AI Edge Gateway” that can run edge AI processing, so in the future we would like to consider using the “AI Edge Gateway” as well.