CEM India - Abstract

Abstract Information :

The traditional and todays approach for emission monitoring is the implementation of Continuous Emission Monitoring System (CEMS). However, in the age of digitalization and an increasing amount of data and technical possibilities of data analysis, there are alternatives to this approach. Worldwide, there is increasing activity on the part of environmental control authorities to approve Predictive Emission Monitoring Systems (PEMS). Instead of measuring as in a CEMS system, a PEMS uses other process parameters to model and calculate the emission value. Although the use of PEMS is allowed, for example, by the Environmental Protection Agency in the USA (defined in Performance Specification 16 - Specifications and Test Procedures for Predictive Emission Monitoring Systems in Stationary Sources), its implementation has been slow. This is largely due to the time-consuming and complex creation of the models, which reduce the advantages of a PEMS. However, new technologies are creating new opportunities. The presentation will reduce a novel way to create PEMS systems, which allows to ensure a high accuracy of the model at a significantly reduced effort. Instead of the usual creation of a process model with the help of a process expert at high effort, a special AI algorithm is used here, which automates the modeling and is optimized at the end using process know-how. In the context of the presentation, the possibilities of PEMS systems as a supplement and alternative to CEMS will be explained and ways outlined as to how the path of local approval could be designed.