A research paper published by Sichuan University and SPIC Green Energy Co., Ltd. in the journal "Chemical Progress" has disclosed the actual operational data of SPIC's Da'an 180,000-ton green ammonia project. The inherent volatility of large-scale wind and solar power generation, combined with the stringent requirements for continuous and stable operation in the ammonia synthesis process, creates a core technical challenge for flexible green ammonia production.

A multi-steady-state flexible production architecture for green ammonia driven by large-scale fluctuating green electricity has been proposed. Focusing on the multi-time-scale collaborative operation needs across the entire "source-electricity-hydrogen-ammonia" chain, the research systematically examines five core modules: production decision management, integrated electricity-hydrogen-ammonia dispatch, online real-time optimization, dynamic control of the hydrogen production cluster, and advanced control for ammonia synthesis. Technologies developed include multi-time-scale optimization scheduling, coordinated balancing of electricity-hydrogen-ammonia multi-energy flows, power allocation and start-stop management for the hydrogen production cluster, hybrid-driven online real-time optimization, and multi-variable nonlinear predictive control for ammonia synthesis, forming a comprehensive collaborative operation architecture.

This achievement was validated through a one-year engineering trial at a large-scale domestic project for producing green hydrogen and green ammonia from wind and solar power. The results demonstrate that through coordinated dispatch without electrochemical energy storage, the actual consumption rate of wind and solar power reached 96.8%, with grid power consumption accounting for less than 4.2%. The hydrogen production cluster achieved continuous power regulation from 8% to 110%, with PEM electrolyzers handling the vast majority of high-frequency fluctuation components. The electricity-to-hydrogen conversion efficiency improved by approximately 3.5% compared to operation without cluster control, while the average daily start-stop frequency of electrolyzers decreased by about 34%. Under scenarios of intense minute-level wind and solar power fluctuations, the PEM electrolyzer completed power reduction within 3 seconds, and the ALK electrolyzer followed at a rate of approximately 1.5% per second, with the cluster's overall response time being less than 10 seconds. The convergence time of the hybrid-driven real-time optimization model was reduced from 8 minutes to 10 seconds. The ammonia synthesis section achieved a wide load regulation range of 10% to 110%, with the root mean square error of the inlet temperature for the three-bed layers reduced from 3.2°C to 3.8°C under PID control to less than 1.2°C. The automatic APC (Advanced Process Control) operation rate exceeded 95%, and manual intervention frequency decreased by 95.6%. In typical cases, the median electricity consumption per ton of ammonia dropped by approximately 536.75 kWh, and the ammonia concentration at the synthesis reactor outlet increased from 9.52% to 16.18%. Under extreme conditions without hydrogen buffer storage, the measured ammonia synthesis load adjustment rate reached 4.38% per minute. These results confirm the engineering feasibility of the proposed technical system under large-scale fluctuating green electricity scenarios, providing a replicable technical model for the deep integration of renewable energy with continuous chemical processes and the large-scale development of the green ammonia industry.