Using Hardware in the Loop to Create a Microgrid Testbed

This post — part of a new special report — highlights how microgrid developers are using testing tools like hardware in the loop (HIL) to create microgrid testbeds, as well as what these testbeds can be used for.

Download the full report.

HIL to create microgrid testbeds

The electric power grid is the world’s largest man-made system. It is also a very complex machine with thousands of components and subsystems, like generating plants, transformers, switches, and miles and miles of wires.

That “machine” is undergoing a historic transformation. With digitalization and decentralization, power flows are becoming bi-directional, and digital sensors and electronic controls are making the grid more agile, more flexible, and more plastic. Also, the control of the grid is expanding from transmission level network to distribution network and all the way down to the consumer. Turning the consumer into a prosumer.

Often, all of those trends come together in a microgrid, which may comprise a generation source, such as solar panels, an energy storage device, a back-up generation source, such as a combined heat and power plant, inverters for the solar panels and battery systems, load control devices for captive power sinks, a utility system interface, controllers for the individual components, and an overall microgrid control system. Combining all those components is complex on its own, but a microgrid’s complexity is augmented by the fact that it operates within and interacts with the surrounding grid. Fundamentally, all the control functions that existed on the transmission level are now being democratized and are moving down to the distribution level and microgrids.

Best practice dictates microgrid testing

Given their complexity, best practice dictates that microgrids be tested before being connected to the grid. Not doing so, would be like flying a new aircraft without training. That is why pilots use flight simulators — not an aircraft carrying passengers — for training and testing. A flight simulator can duplicate real world conditions and replicate situations that could be too dangerous for a real-life training session. Even more to the point, that is why aircraft components are tested in the design phase before being deployed; the cost of failure during design is much less than in post-production.

Similarly, it would not be prudent to test a microgrid by plugging it into the grid or to test a new microgrid component by plugging it into a real-world microgrid. That is where HIL technology comes in. HIL technology builds on the principles of MBE. HIL is a technique that is used to develop and test embedded control systems using real-time simulation.

Actually, HIL involves connecting any piece of hardware into a computer model so that the device goes through realistic operation—just as it would in the field. This could be a light bulb, generator, battery, etc. If the device being plugged in is a controller, you have a special category referred to as controller-Hardware in the Loop (C-HIL). That’s what Typhoon specializes in because the controller is hard to model realistically, yet it’s often less expensive and easier to modify than the rest of the hardware. That is a useful feature at any stage of the development process.

Microgrid Digital Twin hardware and software example. (Image: Typhoon HIL)

Prior to adopting HIL testing, the EPC Power Corp., a designer and manufacturer of power conversion systems, was doing all its testing in a laboratory. But as the company began working with larger systems, it began to run into limitations. Testing larger systems would have required scaling up power levels in the lab proportionally, which would have required more square footage and more time.

For EPC Power, HIL was particularly useful for developing battery storage inverter control software. After adopting Typhoon HIL, “we were able to move the product line forward more quickly by doing preliminary testing on the HIL, and doing performance verification on the HIL,” Ryan Smith, CTO of EPC Power, said. “We could do a suite of tests in an hour that would take a week in the lab.”

HIL was also valuable in software-hardware integration, Smith says. “I would normally have allocated six weeks for that task, and that might even be aggressive. We did it in one day for the 500-kW inverter.”

Model fidelity

An important consideration with respect to real-world simulation is the model fidelity. Fidelity in this context refers to how accurately the computer model describes the real system. A crucial element of that accuracy is whether or not system response times are as fast as the real system functions, which is essential if the goal is to interact with a real system. Naturally, higher fidelity translates to more computing, which makes real-time simulation that much harder. Some systems, such as Typhoon HIL’s microgrid testbed, are capable of operating at half a microsecond. So, while the generator in a microgrid might operate in a slow timeframe, the inverter for the storage device is capable of operating at sub-microsecond speeds.

In order for the model to capture all interactions and possible issues, the fidelity needs to be as high as possible. Unfortunately, there is a tendency in the industry to default to the lowest, not the highest fidelity. That can become a problem when trying to test the real-time performance of an entire system. It could distort the responses and capabilities of the system. And it could fall short of the capabilities of a more robust testbed system.

HIL technology allows a controller to be put through the paces of thousands and thousands of possible scenarios without incurring the cost and time associated with actual physical tests, and that also facilitates faster and more accurate design iteration. Some of the tests, especially faults and other destructive tests, are impossible or impractical to perform in real life. HIL technology can also be used to comply with standards set by the Institute of Electrical and Electronics Engineers (IEEE) for testing microgrid controllers, IEEE2030.8-2018, which is used to verify the functions of a microgrid controller common to the control of all microgrids regardless of configuration or jurisdiction.

Using HIL modeling software, a microgrid power stage schematic can be designed that can run in real-time on a collection of HIL simulators that interface with protection relays, microgrid controllers and controllers of solar inverters, battery inverters, diesel gensets and fuel cells. The use of HIL in conjunction with controllers (C-HIL) allows controller software to be tested and vetted before crucial deployment to the field.

HIL reduces work time

It is difficult to quantify the savings from using HIL because the technology of every component is different, Qiang Fu, regional technology manager at Eaton, says. But using HIL throughout the design and validation phases of product development could cut man hours by 20% to 30%. And if a vendor were able to use one testing tool from design to deployment, that would be a great value to the industry, Fu says.

Another benefit of HIL is that the levels of fidelity can be adjusted at different stages of development. “That saves money because you don’t have to reinvent the wheel,” Fu said.

Also, HIL can be adapted to individual vendor specifications. Without HIL, engineers basically guess at the specs. An HIL system with a drop-down menu of commonly used vendor specs that can be plugged into an HIL platform would represent a big cost savings, according to Fu.

In a power system setting, HIL microgrid testbeds can be used for:

Time domain short circuit analysis, failure mode analysis, coordination study analysis
DER interoperability and microgrid functional testing
Testing of a microgrid against short circuits, phase losses, component failures and over voltage, low and over voltage ride through
Conduct of a sensitivity analysis of the entire network in real-time
Testing of a controller’s hardware, firmware, software and communications under all operating conditions that include faults in both islanded and grid connected modes

Over the next few weeks, the Microgrid Knowledge series on new microgrid simulation and testing techniques will cover the following topics: