Blackout prevention out of the cloud
Simulations supported by artificial intelligence can protect millions of pole mounted transformers from sudden, total failure.
Pole mounted transformers are a common sight in many countries. They reduce the voltage to 230 V just upstream of the customer’s power point. In India, for example, well over ten million of these transformers are in use. The grid operators and electricity customers there, however, have a massive problem with them: Every year, 15 to 20 percent of the transformers fail, causing blackouts and leading to huge repair costs and lost production.
By Hubertus Breuer and Frank Krull
Oil loss leads to fire risk
The loss of transformer oil is one of the most common causes of the high failure rate. This oil is used to cool the coils inside the transformers and keep them insulated from each other. In poorly maintained devices, where the housings are badly rusted or cracked, the oil can leak out. If the oil falls below a critical level, there’s the risk of the transformer overheating or of voltage flashover between the coils. In these cases, it isn’t uncommon for the transformers to burst into flames.
The expert group for system modeling at Siemens Corporate Technology in Bengaluru in southern India has now developed a solution to measure the oil level in pole-mounted transformer from any manufacturer using the digital twins of the transformers. “The twins make it possible to achieve a realistic oil-level simulation in the cloud using fast computers, so now grid operators can be warned in advance if one of their transformers reaches a critical state,” says Surya Bhamidipati who heads the expert group.
Twins that sound the alarm
Bhamidipati’s group hasn’t however created an individual digital twin for each of the many different pole-mounted transformers in use in India: “That would be infeasible, either financially or in terms of the time it would take,” he says. “Our approach is based on a normalized digital twin.” Simit Pradhan, a modeling expert in Bhamidipati’s group, adds: “That’s a twin that represents the essential principles of a pole-mounted transformer, but is not customized down to the level of a specific transformer. It only becomes an individual twin when we feed in the measured temperature values from the actual transformer.”
To obtain these temperature values, the transformers simply need to be retrofitted with four sensors and a router that sends the measured values to the cloud. Three of the sensors measure the temperature of the housing and one measures the ambient air temperature. These values can be used first to determine the shape, size, and rated power of the retrofitted transformer compared with an intact transformer, and then create an individual twin for it.
Artificial intelligence as a computing partner
Once the individual twin is in place, the measured temperature values can be used to simulate the oil level and indirectly determine how full the unit is. That won’t work with a traditional simulation, though, because the algorithms in those cases are based solely on physical laws. “The time and effort needed for the calculations would be very large,” Pradhan observes. “That’s why we combined these algorithms with statistically tractable techniques and methods from the world of artificial intelligence.” This is the only way to bring the calculation time and effort for the simulation down to the point where the oil level in the transformer can be simulated as often and as regularly as necessary to ensure reliable monitoring.
The ease of retrofitting isn’t the only benefit offered by the oil-level monitoring method created by Bhamidipati’s group. Compared with other monitoring solutions using sound or weight sensors, for example, it’s also quite inexpensive. Where other solutions can cost almost half as much as a new transformer, the solution using four temperature sensors, a router, and cloud-based simulation is available for about one-tenth of the cost of a new unit.
Endurance tests in Rajasthan and Kashmir
Recently Bhamidipati’s group successfully tested their oil-level monitoring system in a field trial outside the lab in India’s federal state of Goa, and they’re now moving on to a larger pilot project with Indian grid operator BSES Rajdhani Power Limited. This project involves retrofitting 40 pole-mounted transformers in the federal states of Rajasthan and Kashmir with the monitoring system to test how the solution works in different climatic conditions.
July 05, 2019
Hubertus Breuer is a science and technology journalist and works as a freelancer in Munich.
Frank Krull is a physicist and journalist and works in the communications department of Siemens.
Subscribe to our Newsletter
Stay up to date at all times: everything you need to know about electrification, automation, and digitalization.