ENGINERIFY was assigned a project from Vestolit for dynamic modeling, output data validation for process controllers, and prediction of chlorine production from Chor-alkali cells for their PVC plant in Germany. This involved several activities including Data collection from the site, Model development, ANN for prediction and implementation.
Scope
The main objective of this project was data validation and improved Process control for chlorine production. To perform this project, the following activities were conducted:
Engineering data collection that includes operation data, design reports, and previous/ current anode/cathode design.
Database, drawings (PFDs and P&IDs), etc.
After a gathering of documents, review of collected data like design & operating data, and process control reports, etc.,
Research and development for mass and energy balance, rate equations, load calculation, design equations and constants, etc.
Updating the Data Sheet.
Making Simulation on Simulink/Matlab
Performing model validation, Data validation, and evaluation of model performance.
Implemented Artificial Neural Network to predict production and efficiency of the process.
Development of an Implementation Plan for the model to optimize process controllers.
Preparation of final Comprehensive Report
Process
A Comprehensive approach is set forth to ensure all essential components integral to the project scope are included.
Outcome
Reduction in byproduct formation
The improved overall efficiency of the process
Data validation was done successfully
Prediction of chlorine production and process controller performance
Suggested better options for process control application and operation.
This modeling offered a clear picture of the process control and presented a better plan to improve process smoothness, control, energy optimization, and prediction based on the company’s needs
Deliverables:
Enginerify successfully submitted the following deliverables
MATLAB/Simulink simulation and code files
Datasheets
Controller validation results and assessment
Artificial neural network code
Implementation guide
Detailed final report