Novel Optimal Load Control for Power System Frequency and Voltage Regulation

Yaxin Wang, Donglian Qi, Jianliang Zhang, Jingcheng MeiJournal of Modern Power Systems and Clean Energy 2023 · DOI

TL;DR

This paper introduces Novel Optimal Load Control (NOLC), a framework that enables "smart loads" to stabilize both power grid frequency and voltage simultaneously. By framing the physical dynamics of the power network as a distributed optimization algorithm, the authors prove that the grid can essentially "solve" its own imbalance problems while minimizing the inconvenience (disutility) to end-users.

What Problem Does This Paper Address?

As we transition to modern power grids, two major shifts are occurring: the rise of intermittent renewable energy (like wind and solar) and the proliferation of smart, controllable loads (like electric vehicles and smart appliances). These changes make the grid more volatile. When supply and demand don't match, the system's frequency and voltage deviate from their nominal values, potentially leading to equipment damage or blackouts.

Traditionally, we rely on heavy-duty generators to fix these imbalances. However, generators are slow to respond due to mechanical constraints and are often carbon-intensive. While researchers have looked into using the "demand side" (the loads) to help, most existing solutions focus on either frequency or voltage, but rarely both. Furthermore, they often ignore the "disutility"—the cost or discomfort to the consumer when their power usage is adjusted.

Methods

The authors propose a mathematical framework called NOLC that treats the power grid as a distributed computer.

  1. Optimization Formulation: They define a global objective to minimize the total "disutility" of all controllable loads while maintaining power balance.
  2. The "Physics as Algorithm" Approach: The core innovation lies in proving that the natural swing equations of the grid (the physics of how generators and loads interact) are mathematically equivalent to a gradient descent algorithm solving the dual of their optimization problem.
  3. Control Laws: Based on this, they derive local control laws. Instead of needing a central controller to tell every appliance what to do, each smart load simply monitors its local frequency and voltage. It then adjusts its active power consumption based on frequency and its reactive power based on voltage.
  4. Stability Proof: Using Lyapunov stability theory, the authors prove that the system will always converge to the optimal steady-state point regardless of the initial disturbance.

Datasets

The researchers evaluated their approach using two standard power system benchmarks in MATLAB/Simulink:

  • IEEE 33-bus system: A standard radial distribution network used to test the basic performance of the NOLC against other methods.
  • New England 39-bus system: A more complex, multi-machine power system used to verify that the NOLC works in large-scale, realistic environments with multiple generators.

Key Insights and Ideas

  • The Grid is the Solver: The most elegant "aha moment" is the realization that we don't need to build a massive communication network to solve grid optimization. The physical laws of the network already perform the necessary "computation" in real-time.
  • Synergy of Frequency and Voltage: The paper shows that frequency and voltage are coupled. Controlling only one can sometimes negatively impact the other. By tackling both, NOLC provides a much "smoother" recovery after a disturbance.
  • User Comfort Matters: By including a "disutility function," the algorithm ensures that the burden of stabilizing the grid is shared across many loads in a way that minimizes the overall impact on consumers.
  • Plug-and-Play Compatibility: NOLC isn't meant to replace existing stabilizers like Traditional Generators (TGs) or Power System Stabilizers (PSS); it's designed to coordinate with them, filling the gap where mechanical systems are too slow.

Key Results

The simulation results demonstrated significant improvements over traditional methods:

  • Superior Recovery: Compared to traditional Optimal Load Control (which only looks at frequency), NOLC suppressed the voltage nadir (the lowest point of a drop) by 21.4%.
  • Accuracy: It improved steady-state error—the difference between the final settled value and the target value—by approximately 24.2%.
  • Scalability: The researchers found a clear "strength in numbers" effect: as the number of participating controllable loads increased, both the transient (initial drop) and steady-state performance of the grid improved linearly.
  • Stability: Small-signal analysis confirmed that the system remains stable even when there are slight variations or errors in the control parameters ($\alpha$ and $\beta$).

Limitations and Future Work

While the theoretical results are promising, the authors acknowledge a few hurdles:

  • Hardware Dependence: Real-world implementation requires a high density of smart meters and fast-acting sensors at the load level, which are still being rolled out in many regions.
  • Model Simplification: The current model assumes "lossless" transmission lines (ignoring line resistance). While this is a common approximation for long-distance transmission, it may need refinement for local distribution grids where resistance is more significant.
  • Future Directions: The team plans to expand this "network-dynamics-as-optimization" concept to even more complex scenarios, potentially including the coordination of energy storage systems and more diverse types of renewable sources.

Summary updated

Abstract

The sudden generation-consumption imbalance is becoming more frequent in modern power systems, causing voltage and frequency stability issues. One potential solution is load participation in primary control. We formulate a novel optimal load control (NOLC) problem that aims to minimize the disutility of controllable loads in providing primary regulation. In this paper, we show that the network dynamics, coupled with well-defined load control (obtained via optimality condition), can be seen as an optimization algorithm to solve the dual problem of NOLC. Unlike most existing load control frameworks that only consider frequency response, our load-side primary control focuses on frequency, voltage, and aggregate cost. Simulation results imply that the NOLC approach can ensure better frequency and voltage regulations. Moreover, the coordination between NOLC and other devices enabled in the system, the NOLC performance against the total size of controllable loads, and the NOLC effectiveness in a multi-machine power system are also verified in MATLAB/Simulink.
BibTeX
@article{wang2023novel,
  author = {Yaxin Wang and Donglian Qi and Jianliang Zhang and Jingcheng Mei},
  title = {{Novel Optimal Load Control for Power System Frequency and Voltage Regulation}},
  journal = {Journal of Modern Power Systems and Clean Energy},
  year = {2023},
  doi = {10.35833/mpce.2022.000506},
}
Figure