Faculty of Chemical Engineering and Technology, University of Zagreb
Page 6 of 7
During the first part of 2019, we conducted a pilot project for predictive diagnostics of heat exchangers for INA/MOL Rijeka Refinery . On-line monitoring system was designed and established for the heat exchanger diagnostics at hydrocracking plant. Machine learning optimized models are developed for predicting degree of fouling. Monitoring system will be used for maintaining operation efficiency and predictive maintenance .
Organized by the Student Section of the Croatian Society of Chemical Engineering and Faculty of Chemical Engineering and Technology (LAM) the Summer School of Robotics was held from July 8 to 10. During the first day lectures on topic Industry 4.0 were given to introduce participants to the world of robotics, sensors, additive manufacturing and to its application in the industry. Second day, each of the participants had the opportunity to disassemble and assemble simple mBots and become familiar with the programming of robot functions. On the third day of school, the participants were divided into teams and participated in a mini-competition in which they applied the acquired knowledge. With 30 participants from various institutions, the interest of young professionals was considerable as well as an opportunity for networking and encouraging interdisciplinary work among students.
We spent a useful day at the Siemens Living Laboratory on April 16 in Vienna in company with our colleagues from Pliva/Teva TAPI and our kind hosts. The Living Lab is a unique pilot plant for the development and demonstration of automation solutions . The focus is on the process industry: in particular the pharmaceutical and chemical industries. In a combination of laboratory, production plant and control room, answers to current questions about Pharma 4.0 are given including extensive discussion on application in the pharmaceutical industry.
The purpose of current project in pharmaceutical API production is to optimize control performance during batch production. The main goal is to improve temperature control strategy in order to optimize overall batch process control. The results are reduced temperature oscillations and optimized heating/cooling sequences which bring improved and easier monitoring and control, a shorter batch run cycle and extends equipment’s service period.
On October 11 & 12 we visited Siemens sites in Karlsruhe and Frankfurt. It was an unique opportunity to see the state-of-the-art process control equipment production and met Siemens expert team to discuss current trends in the process industry and automation. Furthermore, we have visited the Höchst industrial park in Frankfurt with Siemens process and safety engineering laboratories and experienced R&D stuff. Our visit and contacts are step forward in the cooperation which has already begun in Zagreb by opening our process control laboratory and organizing education and trainings.
OSIsoft EMEA Users Conference was held in Barcelona in September with a series of customer presentations, thematic panels, demonstrations of software tools and interactive workshops. The emphasis was on advanced analytics and machine learning , enterprise infrastructure and business intelligence . Within the academic session we give a presentation titled Using the PI System in Chemical Engineering Education and Research. The emphasis was on predictive maintenance of the process equipment and our laboratory process control system. During the event we have established contacts with academic programme stuff and discussed with the PI experts on using of real plant data and solving industrial problems.
In June we were a part of the biggest world forum on chemical engineering and process industry – ACHEMA. During the event we, jointly with our USA partner PiControl Solutions , gave a presentation titled Multivariable Identification and PID/APC Optimization for Real Plant Application in the section: Plant control and optimization . The presentation covers closed-loop system identification, controller tuning and advanced process control using simple and affordable software package. The presentation was attended by more than 50 process control specialists, plant managers and other fair visitors.
In cooperation with our partners MKP Ltd. and industrial experts we are developing real-time monitoring and control laboratory crystallization system . The goal is implementation of PAT ( Process analytical technology ) techniques to employ real-time measurement of relevant process variables for the crystallization process control (e.g. supersaturation, number of crystal particles, crystal size distribution, polymorphic form). Real-time measurement will allow better understanding of the process and open up opportunity for advanced process control.
Due to reduced energy efficiency caused by fouled heat exchangers in industrial plants there is a need for detection of fouling formation. Models of artificial neural networks for detecting the fouling have been developed. The models are intended for fouling detection and heat transfer coefficient calculation for preventive and predictive maintenance of the industrial shell and tube heat exchangers. By monitoring the efficiency of the heat exchanger, it is possible to systematically plan the maintenance and determine the optimal cleaning time.
In a cooperation with our spin-off company Comprehensive Water Technology (CWT) we designed photobioreactor system for Ruđer Bošković Institute – Center for Marine Research in the town of Rovinj. Photobioreactors are installed for the purpose of algae production process development under the controlled condition (temperature, pH and light intensity). Produced algae are attended to be used as a raw material for biodiesel production. CWT designed control cabinet and the photobioreactor set-up, while LAM designed the control system by application of Siemens S7-1200 PLC and PC-station SCADA software. Developed system enables continuous monitoring, control and data acquisition for later analysis and research.
