How Singapore uses AI to cut electricity use

Chang Sau Sheong, CEO of energytech company SP Digital, discusses how they are cutting the nation’s power bills.

This week Elon Musk inserted a computer chip into a pig’s brain to give it augmented AI feedback.

That pig was more high-tech than many of the world’s power grids, which still supply electricity on demand without any augmentation. Singapore is leading the way, using AI to plug into its power grid and drive greater sustainability.

Chang Sau Sheong, CEO of energytech company SP Digital, tells GovInsider how the company is building AI to help the city become more energy efficient and use renewables.

Solar panels line roofs at the Sembcorp Marine shipyard in southwest Singapore. The panels are monitored by machine learning software that recommends the best time to use equipment to lower energy bills. “We do intelligent management of energy for the entire shipyard,” he adds.

During peak periods, it’s cheaper to use solar power stored in batteries than to draw power from the national grid.

SP Digital has created AI to manage the battery’s use and cut spikes in their energy bills. “This helps us to shave the peak off and flatten the peak to reduce energy consumption, and that reduces the cost for them. So we actually helped them to save quite a lot over the past couple of years.”

AI tracks how much energy is produced and used from the solar roofs. It also monitors the condition of the solar panels to reduce manual inspections. The tech is expected to cut 30 per cent of power drawn from the grid during peak production periods.

Commercial buildings and industrial complexes use 80 per cent of Singapore’s electricity, Chang says. Every little bit that they can cut has a huge impact.

His team is building AI to spot and flag leaks that a human might miss, like a gradual rise in electricity use over time. “We would be able to figure out that your water consumption has not increased, but your electricity consumption is increasing. That will flag it out to you and you can make changes.”

The models also learn to separate normal spikes in electricity use from the ones that are out of the ordinary. “Some spikes are actually normal usage. So you don’t want to be notified when actually this is a valid pattern.”

Nanyang Technological University uses this approach to manage energy use across its entire campus in Singapore. “We installed smart meters and we built a system behind it to collect all this data real time. We help them to keep track and also do things like anomaly detection.”

The real-time tracking from smart metres have replaced the previous monthly manual readings at the university. “We integrate smart meters for both electricity and water under a single system which allows the landlord to monitor on behalf of the tenants.”

SP Digital also built the tech used by Changi Airport’s Terminals 3 and 4 to monitor electricity use. “Using a combination of smart meters and advanced operational analytics, we helped Changi Airport to cut down their operational costs by more than 40 per cent,” he wrote in 2019.

Singapore is encouraging the use of electric vehicles which run on batteries and have a smaller carbon footprint. Only 0.14 per cent of cars in the city were fully electric in 2019. Increasing the number of charging points will be an important step in changing this, the Land Transport Authority has told GovInsider.

SP Digital’s parent company and the nation’s largest utility provider, Singapore Power, owns the largest EV charging network in the country with 250 charging locations in 2019. “The software that powers it is done by SP Digital,” Chang says.

The company’s utility payments app now includes a feature to locate charging points, and track and pay for charging. Chang, a former PayPal executive, had led a revamp of the app to introduce mobile payments for utility bills in 2017.

These projects are a start on Chang’s vision to build tech that powers a more sustainable Singapore. There will be more to come as the pressure to tackle climate change ratchets up.

Article was previously published on GovInsider by Medha Basu on 2 Sep 2020 How Singapore uses AI to cut electricity use