Multi regional power system transmission storage collaborative planning based on distributed optimization framework

The modern power grid is no longer a simple one-way street where electricity flows steadily from a few large power plants to homes and factories. It has become a complex, two-way network where the weather dictates the supply. Wind farms and solar arrays generate power only when the wind blows or the sun shines, often in remote regions far from the cities that need the energy most. This creates a fundamental mismatch: the energy is available in one place at one time, but the demand exists in another place at a different time. To keep the lights on, engineers must solve a dual puzzle. They need to build new power lines to move electricity across vast distances, and they need to install massive batteries to store that energy for later use. The challenge lies in figuring out how much of each to build and where to put them, without wasting money or leaving the grid vulnerable.

A team of researchers from the State Grid Shanxi Electric Power Company has tackled this puzzle by developing a new way to plan for the future of these interconnected power systems. Instead of treating the construction of new power lines and the installation of energy storage as separate decisions, they created a unified model that optimizes both simultaneously. Their approach recognizes that a new power line and a large battery can often do the same job: they both allow the system to handle the unpredictable swings of renewable energy. By using a sophisticated mathematical framework that allows different regions to solve their own planning problems while staying in sync with their neighbors, the researchers found a way to design a grid that is cheaper, more efficient, and better at absorbing clean energy than previous methods.

The researchers tested their method on a simulated four-region power system, a scenario designed to mimic the real-world complexity of a large national grid. In this system, some regions are rich in wind and solar power but have fewer people, while other regions have heavy industry and high electricity demand but less renewable generation. The team compared their new coordinated approach against three older strategies: one that only built new power lines, one that only installed batteries, and one where each region planned its own future without talking to the others. The results were clear. The coordinated plan, which decided on both transmission lines and storage batteries at the same time, produced the best outcome. It reduced the total annual cost of the system by tens of millions of dollars compared to the other strategies. Specifically, the new plan saved 36.5 million yuan per year compared to a plan that only built lines, 44.7 million yuan compared to a plan that only bought batteries, and a staggering 82.8 million yuan compared to the uncoordinated approach where regions acted alone.

Beyond just saving money, the coordinated plan significantly improved the system's ability to use renewable energy. In the simulations, the new method allowed the grid to accept and use 4.6 percentage points more wind and solar power than the "lines-only" strategy and 3.55 percentage points more than the "batteries-only" strategy. This is a crucial distinction because wasting renewable energy, known as curtailment, is a major economic and environmental loss. The researchers found that by placing batteries in the right locations and building just enough new power lines to connect them, the system could smooth out the daily fluctuations of the wind and sun. The batteries would charge when there was too much power and discharge when demand was high, while the power lines would move excess energy from a windy region to a sunny one, or from a region with low demand to one with high demand. This combination of moving energy across space and storing it across time created a much more stable and flexible grid.

The technical innovation behind this success was not just in the planning itself, but in how the researchers solved the massive mathematical problem required to find the best solution. Planning a grid for an entire country involves millions of variables: every possible location for a new line, every potential site for a battery, every hour of the day, and every possible weather scenario. Solving this all at once in a single computer program is incredibly difficult and slow, and it often requires every region to share its private data with a central authority, which is not always practical. The researchers used a technique called distributed optimization. Imagine a group of people trying to solve a giant jigsaw puzzle. Instead of one person trying to fit every piece at once, they split the puzzle into sections. Each person works on their own section, but they constantly check the edges where their section meets their neighbor's to make sure the pieces fit together.

In this study, each region of the power system solved its own planning problem independently, deciding locally where to build lines and batteries. However, they did not work in isolation. They exchanged only the information about the power lines connecting them to their neighbors, ensuring that the total amount of power flowing in and out matched up perfectly. This method allowed the team to run the calculations much faster. While a traditional, centralized computer model took over 82 seconds to find a solution, the new distributed method, running the regional calculations in parallel, finished in about 21.5 seconds. This speed-up is significant because it means grid planners can run many more simulations to test different scenarios, leading to more robust and reliable decisions. The method also respected the privacy of each region, as they did not need to reveal their internal costs or detailed network data to a central authority, only the necessary boundary information to keep the system balanced.

The study confirmed that relying on just one tool is insufficient for the future of energy. A plan that focuses solely on building more power lines tends to overbuild, spending too much on infrastructure that might not be fully utilized. Conversely, a plan that relies only on batteries can become too expensive because it tries to store every bit of excess energy locally, missing the opportunity to move that energy to where it is needed. The independent planning approach, where regions do not coordinate, leads to the worst results, with high costs and wasted energy because regions build redundant assets or fail to support each other when needed. The researchers' findings suggest that the most efficient path forward is a balanced mix. In their simulated four-region system, the optimal plan involved expanding five specific transmission corridors by amounts ranging from 130 to 260 megawatts, while simultaneously installing a total of 800 megawatts of power capacity and 3,200 megawatt-hours of energy storage across the four areas.

The specific configuration of these resources revealed a clear pattern. The regions with the highest renewable energy output, which often faced the risk of wasting that power, were the ones that benefited most from the addition of local storage. The batteries acted as a buffer, soaking up the excess energy during peak production times and releasing it during the evening peak when the sun had set. Meanwhile, the new transmission lines acted as a safety valve, allowing regions to export their surplus to neighbors who were experiencing a shortage. This coordination meant that the system did not need to build as many massive lines as the "lines-only" plan, nor did it need to install as many batteries as the "batteries-only" plan. The result was a system where the total cost was minimized, and the reliability was maximized.

The researchers also looked at how the system behaved on a typical day. They found that the power flows between regions were not static; they changed direction and magnitude throughout the day depending on the weather and demand. In the morning, a region might be importing power to meet the rising demand, while in the afternoon, as solar production peaked, that same region might switch to exporting power to its neighbors. The batteries played a critical role in this dance, charging when the grid was full of cheap, renewable energy and discharging when the grid needed power. The state of charge of these batteries, which represents how full they are, was carefully managed to stay within safe limits, ensuring they were ready for the next cycle. This dynamic interaction between storage and transmission lines allowed the grid to handle the variability of renewable energy with a level of smoothness that neither technology could achieve alone.

The study concludes that this coordinated approach is not just a theoretical exercise but a practical solution for the challenges facing modern power grids. By using a distributed optimization framework, grid planners can achieve results that are nearly as good as a perfect, centralized plan, but with the added benefits of speed, privacy, and scalability. The method allows different regions to maintain their autonomy while still working together to create a more efficient and resilient whole. As the world moves toward a future dominated by wind and solar power, the ability to plan for both the movement of energy across space and its storage across time will be essential. This research provides a clear roadmap for how to do that, showing that the key to a cheaper, cleaner, and more reliable grid lies in the intelligent coordination of its physical components. The findings suggest that the future of power planning is not about choosing between lines or batteries, but about understanding how they work together to solve the complex puzzle of renewable energy integration.