monitored the catalyst performance and behaviour using distributed control system (DCS) or process monitoring tools, such as PI. The process of compiling data alone is already time-consuming, even before data review, opti- misation, or troubleshooting can start. Sometimes, the catalyst suppliers are consulted, and the turnaround time further delays the process. To make matters worse, the problem might have been overlooked due to overwhelming tasks or human fatigue/error, leading to sub-optimal cata- lyst performance or safety issues. As an example, the hydrogen consumption of a hydro- cracking unit is unusually high and keeps increasing, which is an indication of uncontrolled chemical reactions or local- ised temperature runaway. While this might be detected by a DCS operator during a normal/smooth refinery operation, the unit operator could be busy handling the feed fluctua - tion from upstream units, such as the vacuum distillation column, to the point that this unusual catalyst behaviour is missed, potentially leading to a full-scale temperature run- away, unit shutdown, production loss, or fatality. In modern days, refinery operators and engineers have the privilege of accessing online advanced data monitoring tools to assist with data gathering, processing, and analy- sis. In addition, the plant data can be further augmented with catalyst performance or kinetic databases. These assistant tools greatly unload the burden of rou- tine tedious tasks, so the operators and engineers can spend more time on more valuable duties, such as catalyst performance optimisation, troubleshooting, and ensuring safe operation. Some operational reasoning tools can also detect anomalies and suggest preventive actions to opera- tors, while providing a strategy for optimising catalyst per- formance on a real-time basis. Going back to the same hydrocracking example, an AI-based operational reasoning tool can continuously gather and process the plant data, then automatically define the hydrogen consumption baseline based on the feed rate, feed quality, catalyst life, and hydrocracking con- version. When the hydrogen consumption starts to deviate from the baseline, it can automatically notify the operator to take preventive action, i.e., reducing conversion or feed rate, without requiring intensive human monitoring. While originally intended for continuous process monitor- ing, optimisation, and troubleshooting, the online catalyst monitoring tool greatly helps improve overall manpower utilisation and refinery operation. A Melissa Clough Mastry, Director, Global Technology and Technical Services, BASF, melissa.mastry@basf.com, and Grant Severyn, Senior Technical Service Engineer, BASF, grant.severyn@basf.com The most significant advancement in online catalyst moni - toring is the convergence of plant data, catalyst perfor- mance analytics, and predictive modelling into a single decision-support ecosystem. Historically, refiners relied on periodic laboratory analyses and engineer interpretation to understand catalyst performance. Today, advances in connectivity, automation, and AI are enabling continuous performance monitoring and faster optimisation decisions.
One example is the growing use of automated fluid cata - lytic cracking (FCC) performance prediction tools. Rather than waiting for monthly reconciliations or catalyst evalu- ations, refiners can combine unit operating data, catalyst inventory information, and process models to continuously estimate catalyst activity, selectivity, and expected yields. A refinery targeting maximum propylene production, for example, can quickly evaluate whether a decline in olefin yield is being driven by catalyst age, feed quality changes, metals accumulation, or operating severity. In some cases, advanced predictive tools can identify a developing metals-related selectivity loss weeks before it becomes obvious in routine performance reviews, allowing refiners to adjust catalyst strategy proactively rather than reactively. What once required days of engineering analysis can increasingly be completed in near real time. The rapid expansion of data connectivity is also deliver- ing substantial benefits. Modern platforms can integrate laboratory results, process historians, advanced process control (APC) systems, and refinery planning tools through application programming interfaces (APIs), secure data transfers, and industrial analytics platforms to create a single source of truth for catalyst performance. The availability of information has improved as well. Digital customer portals such as BASF’s proprietary Catalyst Connect provides refiners with centralised access to shipment information, certificates of analysis (COAs), E-cat analytical results, and historical catalyst data. Instead of searching through emails, spreadsheets, and archived reports, engineers can access critical catalyst information through a single interface, improving visibility and accelerating troubleshooting and decision-making. AI and advanced analytics are further accelerating this transformation. The industry is increasingly leveraging AI-assisted mod- elling tools to evaluate large and complex FCC data sets, identify emerging performance trends, and rapidly gener- ate visualisations that help engineers focus on the most impactful opportunities. Rather than spending valuable time collecting and organising data, engineers can spend more time evaluating options and implementing improvements. Perhaps the most important development is that catalyst monitoring is evolving from reporting what happened to helping refiners decide what to do next. Many of these capa - bilities required to support this approach are already being applied, including automated performance predictions, digital customer portals, integrated refinery data streams, advanced modelling tools, and expert-driven analytics. When these tools are combined with catalyst expertise and refinery operating knowledge, they enable faster iden - tification of opportunities, more proactive troubleshoot - ing, and more informed catalyst and operating decisions. They also accelerate catalyst reformulation and optimisa- tion by rapidly evaluating catalyst performance, predicting outcomes, and identifying the most effective formulation adjustments to changing refinery objectives and operating conditions. The ultimate value is not more data, but better decisions that improve yields, profitability, and operational reliability.
6
PTQ Q4 2026
www.digitalrefining.com
Powered by FlippingBook