Development of a method for precise determination of catalyst activity and construction of Digital-Twin for prediction of deNOx catalyst deactivation

Abstract

The aim of the work was to develop a fast method of measuring the activity of the entire catalytic layer, a precise method of determining the lifetime (deactivation characteristics) of the catalyst and the construction of an SCR reactor computer tool (based on the so-called “Digital Twin” and machine learning) for determining the lifetime of deNOx catalysts depending on the history of their use and expected further operating conditions. In the first stages of work, a preliminary model of the SCR reactor was built, in which the scope and influence of the catalyst operating parameters on its deactivation was determined. For this purpose, the deactivation mechanisms described in the literature were used, as well as the results of experiments and laboratory tests. In addition, a sensitivity analysis was performed to define the most relevant factors influencing catalyst deactivation. Then, the methodology of measuring the activity of the catalytic layer in-situ was developed, the necessary measurement system was prepared, and comparative tests of catalyst samples were carried out on a laboratory stand and on a block boiler facility at the Siekierki CHP Plant. The SCR-DT software was developed, installed and launched on a server installed in the local network of PGNIG TERMIKA SA. The SCR-DT application accesses operational data on the operating conditions of catalytic layers from the JSMPT server in real time and can import data from xls files available on the local network (e.g. coal, ash and catalyst test data). Application of the SCR-DT software makes it possible to reduce the operating costs of catalysts through better maintenance management, extending the periods for their regeneration/replacement and reducing the amount of waste, thanks to the accurate prediction of their deactivation time and the development of recommendations for their optimal operation.