Today’s steam cracking furnace operator is under growing pressure to optimize plant profitability. This involves maximizing plant throughput on the one hand and, on the other, selecting the best feedstock and cracking severity. This is essential to minimize feed cost while augmenting the production of high-value products. Maintenance costs can be contained, for example, by extending coil lifetime without increasing the risk of coil ruptures. Eliminating operator errors is another key and constant challenge. Given the importance and frequency of field operator interventions with conventional furnaces, damages and repair trips can be a significant cost factor. 

Digitalization combined with artificial intelligence can deliver the insights needed to master many of these tasks while also supporting varying degrees of automation. Linde has developed an innovative suite of tools to support the optimization of steam cracking furnace operations. This suite builds on the experience we have gained managing over 1,000 process plants remotely and, more specifically, designing over 200 advanced process control solutions to support the growing digitalization trend. At the heart of this suite is the Linde Virtual Furnace.

The benefits of the LVF can be further extended by combining it with Linde’s thermal imaging module. This module continuously generates radiant coil temperature profiles by using Linde’s intelligent algorithms to process infrared images from cameras installed in the furnace wall. This avoids the risk of human error when operators take manual tube metal temperature measurements. In addition, more reliable measurements allow operators to extend the furnace runtime closer to the material limits. The automatic detection of hot spots together with temperature-controlled decoking avoid accidental overheating of the coils. This further increases their lifetime and saves maintenance costs. Linde’s real-time coil lifetime prediction algorithm predicts the risk of coil rupture based on the carburization levels. This information gives operators the insights they need to replace coils on demand, thus maximize replacement intervals while mitigating the risk of rupture.