Key takeaways:
VPPs are emerging as coordinated networks of distributed energy resources to support grid demand and flexibility
Layered architectures integrate data, control systems, and edge devices for real-time energy management
AI and predictive algorithms are increasingly used to optimize dispatch, forecasting, and system performance
Virtual Power Plants (VPPs) are emerging as independent market entities, and China’s National Development and Reform Commission (NRDC) has issued guidelines to accelerate their development, with targets of over 20 GW by 2027 and 50 GW by 2030. Jinko ESS has recently unveiled an integrated VPP solution at the Energy Storage International Conference and Expo (ESIE) 2026 in Beijing.
A VPP is a virtual network of individual energy sources, such as solar panels, batteries, and other devices, that work together to function like a traditional power plant. The idea is to supply electricity when it is most needed, thereby easing the demand on the grid. Jinko ESS, an energy storage equipment provider, is now offering a comprehensive storage and aggregation service. The idea is to transform distributed energy resources from passive grid connection to active participation in collaborating with other distributed resources and dispatching energy when required, reducing dependency on the grid.
As part of this solution’s launch, Jinko’s ‘Virtual Battery’ platform uses an advanced 3-layer architecture named Cloud-Station-Edge, in which each layer has a different function. The cloud integrates data from IoT devices with a centralized data platform and also interfaces with power trading markets and dispatch centers. This enables data collaboration between different energy resources. The station layer uses an Energy Management System (EMS) tool to precisely control devices. Finally, the edge layer aggregates all distributed resources, including PV, storage systems, flexible loads, EVs, etc.
The platform’s algorithm, according to the company, is powered by 4 intelligent features that optimize operations, including dynamic dispatch, machine learning to predict the state of equipment, deep learning for energy generation and load forecasting, and predictive control models to optimize costs and enhance resilience to fluctuations.
Moving forward, Jinko aims to improve its offerings by shifting from a data-driven model to AI-native intelligent operations.