In the quest for more efficient and cost-effective industrial automation, researchers have turned to a seemingly unlikely hero: the humble pneumatic actuator. These devices, known for their reliability and simplicity, are often paired with complex proportional valves to achieve precise positioning. However, this combination can lead to unnecessary costs and design intricacies. Enter M. O. Sheykin, a researcher from the National Research University MPEI, who has pioneered a novel approach to optimize pneumatic actuators using discrete valves, potentially revolutionizing the energy sector and beyond.
Sheykin’s research, recently published in the journal “Advanced Engineering Research” (translated from Russian as “Advanced Engineering Research”), focuses on addressing the trade-off between switching frequency and positioning accuracy in pneumatic actuators. Traditional methods of optimization often prioritize individual performance indicators, leaving a gap in finding a balanced solution. “Existing studies mainly focus on optimizing individual performance indicators of pneumatic actuators and do not offer effective methods for finding a compromise between conflicting criteria,” Sheykin explains.
To tackle this challenge, Sheykin developed a methodology for multicriteria optimization using surrogate models, which are simplified representations of complex systems. By employing neural networks to build these models, the research significantly reduces computational costs while maintaining high accuracy. The Latin hypercube method was used to ensure a uniform distribution of calculation points, further enhancing the precision of the surrogate models.
One of the key findings of the study is the effectiveness of sliding control as a control algorithm. This approach compensates for external disturbances and system uncertainties, enabling high positioning accuracy with minimal distributor switching frequency. “The optimization of control parameters has shown the possibility of reaching high positioning accuracy with a minimum frequency of distributor switching,” Sheykin notes.
The implications of this research are substantial for the energy sector and other industries relying on pneumatic systems. By reducing the frequency of distributor switching, the proposed approach can extend the service life of equipment and increase the reliability of automated systems. This translates to lower maintenance costs and improved operational efficiency, which are critical factors in the energy sector.
The use of surrogate models and neural network technology opens up new avenues for the faster design of complex systems. As Sheykin points out, “The use of surrogate models and neural network technology opens up new prospects for faster design of complex systems.” This innovation could accelerate the development of advanced automation solutions, making them more accessible and affordable for a wider range of applications.
In the broader context, this research highlights the potential of integrating advanced computational techniques with traditional engineering practices. By leveraging the power of neural networks and surrogate models, engineers can achieve more efficient and cost-effective solutions, paving the way for future advancements in industrial automation. As the energy sector continues to evolve, the insights gained from this study could play a pivotal role in shaping the next generation of pneumatic systems and beyond.