Digital twin based reference architecture for petrochemical monitoring and fault diagnosis

The whole process of the petrochemical industry involves flammable and explosive dangerous goods. The timely discovery of abnormalities or failures in the petrochemical process is crucial to ensure production safety. This paper sets up the approach to build the Digital Twin System (DTs) of a petrochemical process. Specifically, we decompose the petrochemical process into five levels one by one and build a digital twin plug-in for each component of the component layers, and then inversely decouple the process to assemble the DTs layer by layer. As a specific experimental example, the characteristic DTs is proposed to build modules of temperature field and pressure field and flow field, these DT modules are driven by practical industrial sampling data from cracking furnace, and three characteristic DTS modules stated above are integrated to form DTS. Based on the digital twin technology and DTs, we propose the logical structure of chemical process status monitoring and fault diagnosis in detail, which improves the safety and controllability of the petrochemical process.

Presenting Trends In Petrochemical Process Control Systems

Petrochemical operators view the distributed control system process through operating graphics that are displayed on four or more cathode ray tube displays that are monitored 24/7. The graphics vary in both design and content, however there has not been extensive research done on the ability of vector arrows to display this trend information. The purpose of this study was to understand the current trend representation methods and to develop efficient trend representation displays. Results indicate that providing vector arrows for trend representation can be effective in getting the attention of the user and can help in reducing change blindness. Results also indicate that between 10 and 20 degrees of change a participant notices change and that any rate of change increase beyond that does not improve their recognition, illustrating the importance of capturing the rate of change within this range in representing trend data. We were able to demonstrate that use of vector arrows for at-a-glance initial trend analysis was easy to use and effective in petrochemical industry systems.

A Sparse Autoencoder-Based Unsupervised Scheme for Pump Fault Detection and Isolation

Pumps are one of the most critical machines in the petrochemical process. Condition monitoring of such parts and detecting faults at an early stage are crucial for reducing downtime in the production line and improving plant safety, efficiency and reliability. This paper develops a fault detection and isolation scheme based on an unsupervised machine learning method, sparse autoencoder (SAE), and evaluates the model on industrial multivariate data. The Mahalanobis distance (MD) is employed to calculate the statistical difference of the residual outputs between monitoring and normal states and is used as a system-wide health indicator. Furthermore, fault isolation is achieved by a reconstruction-based two-dimensional contribution map, in which the variables with larger contributions are responsible for the detected fault. To demonstrate the effectiveness of the proposed scheme, two case studies are carried out based on a multivariate data set from a pump system in an oil and petrochemical factory. The classical principal component analysis (PCA) method is compared with the proposed method and results show that SAE performs better in terms of fault detection than PCA, and can effectively isolate the abnormal variables, which can hence help effectively trace the root cause of the detected fault.

Simulation and Analysis of a Petrochemical Process (Deethanizer Column- MLE field) using HYSYS Aspen Simulator

Natural gas industry has a great strategic and economic importance; it becomes one of the most attractive business opportunities in the petroleum and petrochemical field. In order to produce high quality gas, many companies all over the world including Algeria create many plants for natural gas processing to clean raw natural gas, by using separation unites and other services to get a product that respect the commercial specifications. Natural gas fractions are separated by distillation column in our case is called deethanizer tower. In this work a dynamic simulation study of deethanizer column is developed and implemented in HYSYS simulator. Simulation results prove that the HYSYS is a powerful tool to simulate industrial processes such that the simulated results are close to the real ones.