China has put another marker down in grid-scale storage. A 1 GW / 4 GWh lithium iron phosphate battery energy storage project in Inner Mongolia connected to the grid on August 3, according to Chinese industry and local government reports carried by pv magazine. The project, known as the DongSu Substation New Energy Storage Special Action project, is being described locally as the world's largest single-site grid-forming LFP battery plant. The milestone matters because it combines three threads that usually move separately: four-hour storage, LFP chemistry, and grid-forming operation. Batteries are already common as fast-response resources. A gigawatt-scale project built to support voltage, frequency, and renewable integration in a major resource region pushes the technology into a heavier grid role. AI-generated image DongSu puts four-hour LFP storage at gigawatt scale in a renewable-heavy region of northern China. 1 GW Project power rating 4 GWh Storage duration at scale CNY 3B Reported investment 1 TWh Expected annual discharge What Went Online The DongSu project is located near Mandulatu in Sonid Left Banner, Xilingol League, in the Inner Mongolia Autonomous Region. The site covers roughly 28 hectares and was backed by Wanbang Digital Energy with reported investment of about CNY 3 billion, or roughly $445 million. Local sources said the first energization and grid connection were completed successfully on the first attempt. The headline capacity is large by any regional standard. A 1 GW battery can inject or absorb power at the scale of a major power plant. Four hours of duration means the system can move enough energy to matter across daily renewable ramps, curtailment periods, and evening peaks. At full discharge, 4 GWh is enough stored electricity to change the shape of a regional supply curve for several hours. The project also uses LFP cells, the chemistry that has become the default for utility storage because of cost, cycle life, and thermal stability. LFP does not solve every storage problem, but it is the chemistry most capable of supporting massive deployment today. Pairing it with grid-forming controls is what makes DongSu more than another capacity addition. AI-generated image LFP chemistry remains the workhorse for large storage projects that prioritize cost, cycle life, and safety. Why Grid-Forming Matters Conventional grid-following inverters depend on an existing grid signal. They synchronize with the voltage and frequency already present on the system. Grid-forming inverters are meant to do more. They can help establish a stable voltage and frequency reference, which becomes more important as coal and gas turbines retire or run less often. That distinction is moving from engineering conference topic to procurement requirement. Power systems built around wind, solar, and batteries need fast resources that can do some of the stabilizing work once provided by spinning machines. Grid-forming storage can provide synthetic inertia, voltage support, fault ride-through, black-start support in some designs, and faster recovery after disturbances. Inner Mongolia is a logical place to test that at scale. The region has huge wind and solar resources, long transmission paths, and industrial loads that depend on reliable power. A large battery that can shape output and support grid strength is useful when renewable production rises faster than local demand or transmission capacity. The CurrentCells read DongSu is important because it treats a battery as grid equipment, not just energy inventory. The project tests whether LFP storage can provide both bulk shifting and stability services at power-plant scale. A China Storage Signal China is already the center of global battery manufacturing. DongSu shows how that manufacturing depth is feeding domestic grid deployment. Large LFP systems can move from factory lines into provincial grids quickly when cell supply, inverter supply, engineering capacity, land, and policy direction are aligned. The project also shows a different rhythm from many Western storage markets. In the United States and Europe, battery stories often revolve around tax credits, interconnection queues, capacity auctions, tolling agreements, and fire-code reviews. Those pieces matter in China too, but large provincial projects can move with more centralized coordination once local governments and grid companies make storage a priority. That does not mean every large Chinese storage project will perform equally well. The key test is not whether a battery can connect to the grid. It is whether the asset is dispatched often, earns or receives enough value to justify the capital, and actually reduces renewable curtailment or reliability risk. The reported expectation that DongSu will discharge around 1 TWh a year gives the market a useful benchmark to watch. AI-generated image Grid-forming storage depends on inverters, controls, and grid studies as much as battery containers. The Competitive Pressure For global storage developers, DongSu raises the scale bar. A single 4 GWh site is larger than many national storage programs and comparable to multi-project portfolios in emerging markets. It makes 100 MW and 200 MW projects look like routine infrastructure rather than frontier deployments. The scale also affects suppliers. Battery containers, power conversion systems, transformers, thermal management, fire protection, and software all have to work as an integrated plant. A weak component can create outage risk across a very large asset. As projects grow, warranties and operations contracts become as important as purchase price. Western markets are moving in the same direction, but often through scattered portfolios rather than single mega-sites. Scotland's Coalburn 1 recently reached commercial operations at 500 MW / 1 GWh, while the U.S. fleet is already measured in tens of gigawatts. DongSu adds a China-scale example focused on grid-forming capability and four-hour duration. What Comes Next The first thing to watch is utilization. A 4 GWh battery that cycles regularly can absorb renewable output that might otherwise be curtailed and return it during stronger demand hours. Low utilization would signal a planning or market-design problem, even if the engineering milestone remains real. The second marker is how grid operators evaluate performance. Grid-forming claims need data from faults, ramps, reserve events, voltage deviations, and restart procedures. A large battery can look impressive on a nameplate and still face limits if its control settings are conservative or if grid rules do not fully value stability services. The third marker is replication. If DongSu becomes a template for other renewable bases in northern and western China, grid-forming LFP storage could move rapidly from one flagship project to a standard design. That would pressure other markets to clarify technical requirements and revenue mechanisms for batteries that do more than arbitrage power prices. AI-generated image Operating data will decide whether mega-scale grid-forming batteries become a repeatable template. The Bottom Line DongSu is a storage capacity story, a chemistry story, and a grid-controls story at the same time. The project gives China a 1 GW / 4 GWh LFP system in a renewable-heavy region, with local sources framing it as the largest single-site grid-forming LFP battery plant now connected. For the battery industry, the lesson is plain. The next stage of grid storage will be judged less by announcements and more by the operating behavior of huge assets. Batteries need to shift energy, stabilize weak grids, help renewable regions export cleaner power, and stay available through hard operating conditions. DongSu puts all of those expectations into one very large field test. Sources pv magazine: World's largest LFP grid-forming goes online in China Energy-Storage.news: Scatec starts full operations at Obelisk solar-plus-storage project pv magazine: