Consensus-Based Coordinated Frequency Regulation for VSC-HVDC Interconnected Power Systems with Heterogeneous Resources

The electric grid is a vast, delicate machine that must stay in perfect balance. Every second, the amount of electricity being generated must match the amount being used. If this balance tips even slightly, the speed at which the system spins—the frequency—begins to wobble. In the past, this system relied on massive, spinning turbines in power plants. These heavy machines acted like giant flywheels, storing energy in their motion and naturally resisting sudden changes in speed. They provided a built-in cushion that kept the lights on steady even when demand spiked or a generator failed.

Today, the landscape is changing. As we shift toward cleaner energy, we are replacing those heavy turbines with solar panels and wind farms. These sources are connected to the grid through electronic converters rather than spinning metal. While they are excellent at producing power, they do not have that same heavy, physical inertia. They react differently to disturbances, and when many of them are connected together, the entire system becomes more fragile. Furthermore, modern grids often link different regions together using high-voltage direct current lines, which act as bridges between areas that do not share the same spinning rhythm. This creates a complex puzzle: how do you keep a system stable when the heavy flywheels are gone, the resources are different from one another, and the regions are connected by fast electronic bridges?

Researchers at Shanghai University of Electric Power and the State Grid Shanghai Municipal Electric Power Company have tackled this challenge by developing a new way to coordinate these diverse resources. They focused on a grid made up of three distinct areas, connected by high-voltage direct current links, and filled with a mix of traditional generators, modern wind and solar converters, and energy storage systems. The team realized that treating all these different resources the same way was the problem. A traditional generator, a wind farm, and a battery storage unit all react to frequency changes at different speeds and with different capabilities. To solve this, they created a two-layer control system that allows these heterogeneous resources to talk to each other and work together without a central commander telling them exactly what to do.

The first layer of this system operates like a rapid response team within each local area. When a disturbance occurs, such as a sudden drop in power, the resources in that area must react instantly to prevent the frequency from crashing. The researchers designed an algorithm that lets these different resources share information locally. Instead of waiting for a central office to calculate the solution, the resources negotiate with their neighbors. They agree on how much each one should contribute based on their specific capabilities and costs. This happens in milliseconds. The system ensures that the fastest resources, like certain types of converters, jump in immediately to arrest the fall, while slower resources follow up to support the recovery. This local coordination happens so quickly that it mimics the natural resistance of the old heavy flywheels, even though the physical weight is gone.

Once the immediate crisis is managed, the second layer of the system takes over to restore the grid to its perfect state. Even after the initial reaction, the frequency might not return to its exact target, and the power flowing between the different connected areas might be unbalanced. This is where the second layer comes in. It treats each entire region as a single agent that communicates with the other regions. Using the high-voltage direct current links, the system adjusts how much power is sent from one area to another. It also coordinates the secondary adjustments needed to bring the frequency back to normal and ensure that no single region is doing all the work. This layer ensures that the long-term recovery is not only stable but also economical, distributing the effort in a way that minimizes the overall cost of keeping the grid running.

To test their idea, the researchers built a detailed simulation of a three-area power system based on a standard model used by engineers worldwide. They introduced a sudden surge in demand in one area, equivalent to adding a significant amount of new load to the grid. They compared their new two-layer method against traditional control strategies. The results showed that their approach was significantly more effective. The frequency dropped less sharply, recovered faster, and settled back to the correct value with less oscillation. Perhaps most importantly, the total cost of regulating the frequency was lower. The system managed to keep all resources, from the traditional generators to the electronic converters, within their safe operating limits, preventing any single unit from being overworked or damaged.

The study also examined what would happen if the communication between these resources failed, a realistic concern for any distributed system. They simulated a scenario where a connection between two resources in one area was cut off. Even with this broken link, the system remained stable. The remaining resources adjusted their behavior to compensate for the missing information, and the frequency still recovered smoothly. This demonstrated that the method is robust; it does not rely on a perfect, unbroken network to function. It can handle the messy reality of a modern grid where connections might fail or where resources behave unpredictably.

The findings suggest that as power grids continue to evolve and rely more on renewable energy and electronic connections, the old methods of central control are no longer sufficient. The key to stability lies in allowing diverse resources to coordinate locally and quickly, while still maintaining a broader view of the entire interconnected system. By treating the grid as a collection of agents that can negotiate and adapt, rather than a machine that must be commanded from the top, engineers can maintain security and efficiency. This research provides a blueprint for how future grids can integrate high levels of renewable energy without sacrificing the reliability that keeps the lights on. The work confirms that with the right coordination, a grid made of many different parts can function as a single, resilient whole.