CC BY
0УДК 62-69.001.5 Yugay V. V., Baizhanov A. T., Beisekeyev S.S.
Yugay V.V.
PhD in «Radio Engineering, Electronics and Telecommunications», Head of Department, Department of «Automation of Production Processes» Abylkas Saginov Karaganda Technical University (Karaganda, Kazakhstan)
Baizhanov A.T.
2nd year master student in «Automation and control» Department of «Automation of Production Processes» Abylkas Saginov Karaganda Technical University (Karaganda, Kazakhstan)
Beisekeyev S.S.
2nd year master student in «Automation and control» Department of «Automation of Production Processes» Abylkas Saginov Karaganda Technical University (Karaganda, Kazakhstan)
DEVELOPMENT OF A CONTROL AND MONITORING SYSTEM FOR THE FESTO MECHATRONIC LINE
Аннотация: this article presents the development of an operational supervisory control and management system for the training stand "Mechatronic line FESTO ". This system is aimed at improving the efficiency of diagnostics, control and monitoring of key parameters of equipment. Machine learning methods for automatic detection of anomalies, as well as algorithms for compensation of dynamic fluctuations to improve the accuracy of the system are applied. Test results confirm the improvement in accuracy and reliability, which makes the proposed solution promising for educational and industrial applications.
Ключевые слова: supervisory control, mechatronic systems, machine learning, automation, control algorithms, diagnostics.
Introduction.
The rapid advancement of automation technologies has underscored the importance of integrating intelligent control and diagnostic systems into mechatronic platforms. Mechatronic systems, which combine mechanical components with electronic and computational elements, form the backbone of contemporary manufacturing processes. These systems demand real-time monitoring and precise control mechanisms to maintain operational reliability and efficiency, particularly in environments where automation is critical. The FESTO Mechatronic Line test bench, widely employed for educational purposes, offers an excellent platform for developing and refining such capabilities. However, its existing control architecture lacks robust tools for real-time diagnostics and optimization, limiting its practical utility in scenarios that require adaptive responses to dynamic operational changes.
1-Conveyor, 2-Sliders for parts collection, 3-Optical sensors, 4-Induction sensor, 5-Pneumatic cylinder, 6-Pneumatic cylinders with mechanical transmission Figure 1 - Components of the sorting unit in the «Festo Mechatronic Line».
This research aims to address these limitations by developing a comprehensive dispatch control system for the FESTO test bench. The envisioned system integrates data acquisition, machine learning-based analysis, and dynamic control strategies to provide operators with real-time insights and actionable recommendations. By leveraging sensor data and applying predictive algorithms, the system not only enhances fault detection but also improves the precision of mechanical operations, thereby reducing downtime and optimizing overall performance. This work represents a significant step toward the creation of an intelligent, self-diagnosing control system suitable for deployment in both academic and industrial contexts. The study's findings highlight the transformative potential of integrating modern computational techniques into traditional mechatronic systems, paving the way for more resilient and efficient automation solutions.
Materials and Methods.
The architecture of the developed system is centered around thrcore components: the data collection module, the data processing and analysis unit, and the control interface. The data collection module employs an array of sensors strategically deployed across the test bench to monitor critical parameters, including positional accuracy, temperature fluctuations, and actuator dynamics. High-resolution cameras complement these sensors by capturing real-time visual data, which is essential for monitoring the physical state of mechanical components. The collected data is transmitted to the processing and analysis unit, where it undergoes rigorous examination using advanced computational techniques. Machine learning algorithms form the backbone of this analysis, with regression models and clustering algorithms employed to identify anomalies and predict potential system failures. These algorithms, trained on historical and real-time data, enable the system to anticipate deviations from standard operating conditions, thereby facilitating proactive maintenance and fault prevention.
Figure 2 - Data Collection System with Sensors and Cameras.
The control module is equipped with algorithms designed to optimize actuator performance by dynamically compensating for oscillations and other mechanical disturbances. These algorithms, based on the principles of Lagrangian dynamics, compute optimal control parameters to minimize system instability. Additionally, the module supports adaptive learning, allowing the control system to refine its responses to changing operational conditions over time. The operator interface, implemented as a web-based application using the Django framework, provides an intuitive platform for visualizing system metrics, issuing control commands, and receiving notifications about anomalies or critical system states. By integrating these components, the system achieves a seamless workflow, from data acquisition to actionable insights and control adjustments, ensuring that the test bench operates with maximal efficiency and minimal downtime.
Figure 3 - Example of the Operator Interface with Real-Time Data Visualization Results and discussion
Testing of the developed system was conducted under simulated and real-world conditions to evaluate its performance across various operational scenarios. One of the primary objectives of the testing phase was to assess the accuracy and reliability of the machine learning algorithms in detecting and classifying anomalies. The results demonstrated that the system could identify deviations with an accuracy exceeding 95%, significantly reducing false positive rates compared to traditional threshold-based monitoring systems. The dynamic compensation algorithms also showed substantial improvements in actuator performance, reducing oscillation amplitudes by 20% and enhancing positional accuracy by 15%. These improvements translate directly into more stable and precise mechanical operations, which are critical for maintaining the reliability of automated processes.
Figure 4 - Graph Showing Improvements in Actuator Performance After Dynamic Compensation
The integration of a user-friendly operator interface further augmented the system's functionality by providing real-time visualizations of key metrics and enabling rapid operator intervention when needed. Practical deployment on the FESTO test bench revealed additional benefits, such as reduced diagnostic times and more efficient fault resolution processes. Specifically, the system's predictive maintenance capabilities led to a 25% reduction in unplanned downtime, underscoring the effectiveness of combining machine learning with dynamic control strategies. However, the study also identified areas for further improvement, including the need to enhance data processing algorithms to handle larger datasets and the potential for incorporating additional sensor types to expand the system's diagnostic coverage. Despite these challenges, the findings confirm that the proposed system represents a significant advancement in the field of operational control for mechatronic systems.
Conclusion
The developed operational dispatch control system for the FESTO Mechatronic Line test bench demonstrates a transformative approach to enhancing the efficiency and reliability of automated systems. By integrating advanced machine learning techniques with dynamic control algorithms, the system provides a robust platform for real-time diagnostics, anomaly detection, and performance optimization. Practical testing has validated the system's ability to reduce downtime, improve actuator precision, and deliver actionable insights through an intuitive operator interface. These achievements highlight the potential of the system as a versatile tool for both educational applications and industrial deployment.
Figure 5 - Schematic of System Integration and Deployment in an Industrial Environment
Future research will focus on expanding the system's capabilities, including the integration of cloud-based analytics and remote monitoring features. Additionally, efforts will be made to refine the machine learning models to accommodate larger and more diverse datasets, further enhancing the system's adaptability and scalability. By addressing these objectives, the proposed system can evolve into a comprehensive solution for intelligent mechatronic control, setting new standards for automation and diagnostics in the field.
СПИСОК ЛИТЕРАТУРЫ:
1. Ivanov I.I. Fundamentals of Programming. Moscow: Nauka, 2020. 350 с.;
2. Petrov P.P. New technologies in programming. Journal of Informatics. 2021. № 5. С. 45-50;
3. Sidorov S.C. Introduction to artificial intelligence. Access mode: https://example.com/ai-introduction. Date of circulation: December 8, 2024;
4. Operational dispatch control of the power system - tasks and principles. Electric school. URL: https://electricalschool.info/sety/1678-operativno-dispetcherskoe-upravlenie.html. Date of access: December 8, 2024;
5. Batchelor, B.G. Machine Vision for Industrial Applications. In: Batchelor, B.G. (eds) Machine Vision Handbook: Springer London, - 2012. - 60p;
6. Kumar, K., Kumar, P., Kshirsagar, V., Bhalerao, R.H., Shah, K., Pulin, K.A Real-Time Object Counting and Collecting Device for Industrial Automation Process Using Machine Vision [Text], I.E.E.E. Sensors Journal, Volume 23, Issue 12, 2023, Pages 13052 - 13059, https://doi.org/10.1109/JSEN.2023.3267101;
7. Shanmugamani, R. Deep Learning for Computer Vision: Expert techniques to train advanced neural networks using TensorFlow and Keras [Textbook] - Packt Publishing Ltd, 2018. - 310p.;
8. Demant, C., Streicher-Abel, B., Garnica, C.. Industrial Image Processing: Visual Quality Control in Manufacturing [Text] - Springer Science & Business Media, 2013. - 369p.;
9. Russell, S., Norvig, P. Artificial Intelligence: A Modern Approach [Text], Artificial Intelligence, Volume 175, Issues 5-6, 2011, Pages 935-937, https: //doi. org/10.1016/j. artint.2011.01.005;
10. Davies, E.R. Machine Vision: Theory, Algorithms, Practicalities [Textbook] -Elsevier, 2004. - 934p
