Optimization of compressor unit operation using artificial intelligence technologies
The work examined the possibility of using neural networks and machine learning algorithms to optimize fuel gas consumption by a compressor unit. The authors have studied the dependences of the proposed data from the automated process control system of the compressor station and their impact on fuel...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tomsk Polytechnic University
2023-12-01
|
| Series: | Известия Томского политехнического университета: Промышленная кибернетика |
| Subjects: | |
| Online Access: | https://indcyb.ru/journal/article/view/40/32 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The work examined the possibility of using neural networks and machine learning algorithms to optimize fuel gas consumption by a compressor unit. The authors have studied the dependences of the proposed data from the automated process control system of the compressor station and their impact on fuel gas consumption of the unit. The applied methods of correlation analysis and the study of the technological process of gas pumping made it possible to identify 18 features that have the greatest impact on the target parameter. These features were then used to build a simulation model – a neural network, which allows predicting fuel gas consumption based on input data. Based on the model, an optimizer was built using gradient descent, which, under established input data restrictions, finds the optimal operating parameters of the compressor unit in specified ranges that minimize fuel gas consumption. The developed algorithm allows reducing the gas consumption of a single unit by 2 %. |
|---|---|
| ISSN: | 2949-5407 |