Published: July 6, 2024
Language: Английский
Biomimetics, Journal Year: 2023, Volume and Issue: 8(5), P. 396 - 396
Published: Aug. 28, 2023
The slime mold algorithm (SMA) and the arithmetic optimization (AOA) are two novel meta-heuristic algorithms. Among them, has a strong global search ability. Still, oscillation effect in later iteration stage is weak, making it difficult to find optimal position complex functions. utilizes multiplication division operators for updates, which have randomness good convergence For above, this paper integrates algorithms adds random central solution strategy, mutation restart strategy. A hybrid with center learning (RCLSMAOA) proposed. improved retains update formula of exploration section. It replaces local exploitation stage. At same time, stochastic strategy adopted improve efficiency diversity population. In addition, also used accuracy enhance comparison experiments, different kinds test functions specific performance improvement algorithm. We determine final by analyzing experimental data images, using Wilcoxon rank sum Friedman test. results show that algorithm, combines effective. Finally, on practical engineering problems was evaluated.
Language: Английский
Citations
6
IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 136219 - 136268
Published: Jan. 1, 2023
Proportional-Integral-Derivative (PID) is a simple and intuitive feedback-based control mechanism being useful to track set points reject disturbances. A key question in gradient-free optimization ascertain whether the class of algorithms based on difference vectors generalize reasonably well tackle large PID problems. For generalization practical purposes, it would be desirable render able tune controllers over diverse problems/tasks with minimal human intervention (self-adaptation features), under tight computational budgets. In this paper, aiming fill above-mentioned gap, we propose investigate effectiveness new algorithm self-adaptation mechanisms for tuning. As such, introduce Differential Evolution success-based Particle Adaptations (DEPA), which unifies principles vectors, particle schemes trial/parameter adaptation through archive (memory) mechanisms. Our simulations using large/relevant 25 problem instances (tracking linear, nonlinear, continuous, discontinuous trajectories motor position control, velocity magnetic levitation, inverted pendulum, crane stabilization), comparisons closely related algorithms, their extended adaptive variants (23 total) has shown outperforming benefits proposed approach convergence performance function evaluation budgets (1000 evaluations). Also, experiments real-world pendulum device show potential transferability learned gains unseen situations during training. Furthermore, evaluated algorithmic extension towards fitness landscapes CEC 2017 benchmark suite, showing attractive/outperforming overall instances. particular, framework performs better 358 compared other tasks, general than 182, 215 235 suite 10, 30 50 dimensions, respectively. obtained results have further advance developing efficient self-adaptive may find use wider
Language: Английский
Citations
5
International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Journal Year: 2024, Volume and Issue: 14(2), P. 1473 - 1473
Published: Jan. 26, 2024
Fast pyrolysis in auger reactor gains attention for efficient bio-oil production. Due to the quick nature of process, precise temperature control using proportional-integral-derivative (PID) algorithm is paramount. This study harnesses various PID tuning approaches through modelling and experimental validation optimize continuous temperature. System identification was done investigate process dynamic with fit accuracy above 93% design a suitable control. Comparison experiment data shows favorable result rise time settling match 75%. Ziegler-Nichols (ZN) Cohen-Coon (CC) methods were implemented system undistinguished results, yielding steady-state error (SSE) below 1% around 4,300 4,800 seconds. The heuristic fine-tuning method improved by stabilizing before 3,600 Furthermore, robustness controllers verified disturbance rejection test, keeping SSE deviation inside boundary 2%. Finally, setup could support maximum pyrolytic oil production 69.6% at 500 °C. implies that controller provide stable rugged response productive sustainable plant operation.
Language: Английский
Citations
1
Fractal and Fractional, Journal Year: 2024, Volume and Issue: 8(4), P. 223 - 223
Published: April 12, 2024
This work introduces an alternative approach for developing a customized Metaheuristic (MH) tailored tuning Fractional-Order Proportional-Integral-Derivative (FOPID) controller within Automatic Voltage Regulator (AVR) system. Leveraging Automated Algorithm Design (AAD) methodology, our strategy generates MHs by utilizing population-based Search Operator (SO) domain, thus minimizing human-induced bias. eliminates the need manual coding or daunting task of selecting optimal algorithm from vast collection current literature. The devised MH consists two distinct SOs: dynamic swarm perturbator succeeded Metropolis-type selector and genetic crossover perturbator, followed another selector. fine-tunes FOPID controller’s parameters, aiming to enhance control performance reducing overshoot, rise time, settling time. Our research includes comparative analysis with similar studies, revealing that significantly improves speed 1.69 times while virtually eliminating overshoot. Plus, we assess tuned resilience against internal disturbances AVR subsystems. study also explores facets performance: impact fractional orders on conventional PID efficiency delineating confidence region stable satisfactory operation. work’s main contributions are introducing innovative method deriving efficient in electrical engineering systems demonstrating substantial benefits precise tuning, as evidenced superior compared existing solutions.
Language: Английский
Citations
1
Published: July 6, 2024
Language: Английский
Citations
1