How to control frequency fluctuations during PV plant deloading

Researchers in Egypt have developed new control strategies for managing frequency fluctuations during deloading of a photovoltaic plant. The proposed methodology combines a proportional–integral controller along with a rate of change of frequency controller.

A research group led by scientists from Egypt’s Cairo University has developed a novel control strategy for frequency fluctuations during the deloading of PV systems.

“This paper evaluates the performance of the conventional PV deloaded system controller and showcases its limitations,” the academic said. “An important contribution of this research is the proposal of a blended controller that enhances both frequency nadir and rate of change of frequency (RoCoF). Additionally, a modified controller is suggested to facilitate a safe transition back to normal deloading levels.”

The proposed strategy combines a proportional-integral (PI) controller along with a RoCof controller.

The group initially simulated a conventional deloading control as a reference. The simulation was conducted in the Simulink model, set to a 30% PV penetration rate and a load variation of 1% or 0.01 pu. The conventional control strategy, also known as PI droop, enables PV systems to reserve a portion of the available power for providing dynamic support in both short and long-term frequency scenarios.

“The conventional deloading control method exhibits certain limitations when it comes to providing support at different frequencies,” the tests have shown. “While there have been some enhancements noted in terms of frequency nadir, the level of change seen in the RoCof is quite restricted, resulting in the system not being able to revert to its usual operating state.”

For this reason, the group has proposed its blended control, which incorporates an RoCof controller that monitors how fast the frequency is changing, and when it recognizes a quick drop, it sends a signal to the PV system to increase its power output. By that, it ultimately slows down the frequency drop, while the PI controller adjusts PV power gradually and helps bring it back to the designated level.

In addition, the group has developed a modified version of its blended model by adding a recovery mechanism. Initially, the control action boosts the power output of the PV system, followed by a gradual decrease over the remaining duration of the event. The system was able to improve RoCof by 3.11%, compared to 0.2% for the non-modified blended method.

“The modified controller stands out for its ability to maintain the lowest frequency nadir and RoCof levels among all controllers considered, while still achieving an acceptable settling time,” the team concluded. “Additionally, this controller is capable of guiding the system back to its normal operating point, ensuring a more stable and reliable operation. The modified controller has relatively longer settling time, but it is reasonable with its features of returning the control system to the normal operating point.”

The method and the test results were presented in “Enhanced control strategies for managing frequency fluctuations during deloading of a photovoltaic system,” published in Ain Shams Engineering Journal.