Flexible operation of electricity-intensive industrial processes

Abstract

The aim of this thesis is to study the flexibility of electricity-intensive industrial processes to adapt their electricity consumption to the available power.

Electricity-intensive processes may use hundreds of megawatts during normal operation. An industrial site typically generates electricity locally and distributes it to electrical loads comprising the processes and other equipment. However, if an electrical contingency occurs, the available power reduces and the power consumed by the electrical loads must also reduce. Currently, a process power management system handles an electrical contingency by disconnecting loads until the total consumption of the site is lower than the remaining available power. This operation is named load shedding. If the shed equipment was essential for the process, then the whole process shuts down.

The contribution of the thesis is to demonstrate that flexible process operation gives an alternative to load shedding. The concept is to reduce the power available to the process, so that the process can continue to operate but at a lower production rate. An electrical power outage occurs over a time scale of the order of 100ms. Therefore, in flexible operation, the power available to the process also reduces within 100ms. At the same time, the process control system must receive new set points so that the process moves to an operating point that is consistent with the available power.

The thesis studied flexible operation of liquefaction of natural gas and aluminium smelting to find the set points for the process controllers as a function of available electrical power. Results show that the reduction in production rate can be ameliorated by adjusting other secondary process variables. The results also show that a process with significant energy storage such as aluminium smelting can continue operating for short periods of power reduction even with a very significant loss of available power.

The final chapters of the thesis discuss how to generalize the study of flexible operation so that it can also be applied to other electricity-intensive processes.