During chemical plant turnaround operations (plant startup and shutdown) and process upset management, off-spec product streams will be produced, which usually have to be sent for flaring. Flaring is a passive and end-of-pipe means to protect personnel and equipment safety in chemical process industry (CPI). However, flaring inevitably generates huge amount of CO, CO2, NOx, and some extremely hazardous volatile organic compounds (VOCs), which will cause severe air pollution problems and negative societal impacts. Meanwhile, it also results in tremendous raw material and energy loss that could generate much needed products from the industry. Thus, flare minimization has great significance on environmental, societal, and economic advancement toward the sustainability of chemical process industry.

Facing this challenge, one economic and effective way to minimize flaring during the plant turnaround operation is to use plant-wide dynamic simulation to critically check operating procedures before the real operation. The turnaround procedures are the steps taken by the plant operating personnel to operate a plant. Each of these steps involves the manipulation of the process plant through sophisticated control systems. The plant-wide dynamic simulation can identify benchmark allocations of flaring emissions during plant start-up, shutdown, and normal upset maintenance. It targets the emission source reduction during the operation; instead of expensive process modifications for industrial flare system.

The methodology starts with a development of steady-state simulation model for the target process area. The steady state simulation model is validated by plant data, which is from DCS historian. This validated steady state model is then transferred to a dynamic model by following the updated plant P&ID to set up the control strategy and input the control parameters and equipment dimensions. Again, the plant historian data are used to validate the plant-wide dynamic simulation model. After validation, the plant-wide dynamic model will be gradually turned to the ready to turnaround operation status. With the startup procedures, dynamic simulation will be conducted to identify possible unstable or risky operation conditions. When such unstable or risky conditions are identified, the plant-wide dynamic simulation model can be used to find new stable operation conditions. With the cooperation of the plant operation group, optimal plant turnaround operations can also be identified. The plant-wide simulation is based on first principles. Therefore, the developed methodology is applicable in general and not limited to one particular plant or process.

The dynamic simulation gives an insight into the process dynamic behavior that is not apparent through the use of steady-state simulation or industrial expertise alone. This information is crucial for plant turnaround operations in order to minimize the flaring. The efficacy of this methodology and its economic and environmental benefits for industrial sustainability are demonstrated by real field startup tests.