Dynamic Parameter Tuning Research Articles

AI-based hybrid power quality control system for electrical railway using single phase PV-UPQC with Lyapunov optimization

This research paper presents an advanced AI-driven hybrid power quality management system for electrical railways that addresses critical challenges in 25 kV AC traction networks through a novel integration of single-phase PV-UPQC with ANN-Lyapunov control architecture. The system effectively manages voltage unbalance exceeding 2%, high THD, voltage variations of ± 10%, and poor power factor through a dual-approach methodology combining ANN-based reference signal generation with Lyapunov optimization, enabling dynamic parameter tuning and real-time load adaptation. MATLAB/Simulink simulations validate the system’s superior performance, demonstrating significant improvements, including voltage unbalance reduction from 1.5 to 0.8%, THD reduction below 1%, unity power factor correction, 40% faster dynamic response, and DC link voltage regulation within ± 2%, while maintaining 95% overall system efficiency. Integrating ANN-based shunt and series APF control, Lyapunov optimization, and PV integration establishes a robust framework for enhanced energy efficiency and power quality management in modern railway systems.

Machine Learning-Based Reconnaissance Bee Colony with Custom AES & AE for Robust IoT Data Security and Transmission

Objectives: To propose a robust method to enhance data security and transmission in IoT to address the challenges in real-time data transmission and create a comprehensive security solution. The new method takes credit for the machine learning-based bioinspired method to minimize data loss and maximize end-to-end data transmission rate. Methods: Custom-AES with Avalanche Effect (AE) and Machine Learning-based Reconnaissance Bee Colony are employed to secure the data with enhanced encryption by creating a multi-layered IoT security solution. Key stretching is used to provide high resistance to cryptographic attacks. In order to optimize threat detection, Pareto-based selection and dynamic parameter tuning with ML-RBC are used. To improve attack resilience, rank, and tournament selection methods are employed to target DDoS and MMIT attacks with a minimized encryption time. The NSL-KDD IoT network dataset is used to test the proposed CAES-AE with the ML-RBC model using a network simulator. To assess the performance of the proposed work, the results are compared with prevailing IoT data transmission techniques such as ERSA, QoS-BFT, AES, Custom AES, and RBC. Findings: The suggested CAES-AE with ML-RBC data security and transmission method attains the promising results of 97% authentication success rate, 6 seconds response time to security actions, 96% vulnerability detection speed, 97% data encryption strength, 5 seconds incident response time, and 97.8% network performance (energy efficiency & lifetime), which is higher than earlier methods. Novelty: This research provides a secured data transmission method in an IoT environment to address the real-time data transmission challenges faced in sensor-based networks. This will function as a multi-layered IoT security model that optimizes real-time attack detection and resilience. The performance metrics are evaluated with multiple packet transfer rates to boost the network performance. Keywords: Machine Learning, Advanced Networking, Data Security, Secured Data Transmission, Bio­Inspired Algorithm

Enhancing Hybrid Flow Shop Scheduling Problem with a Hybrid Metaheuristic and Machine Learning Approach for Dynamic Parameter Tuning

This paper addresses the Hybrid Flow Shop Scheduling Problem (HFSSP) by integrating metaheuristic (MHs) and machine learning (ML) approaches. Specifically, we propose a hybrid algorithm by combining Ant Colony Optimization (ACO) and Iterated Local Search (ILS) to form ACOILS. To further enhance the performance of this hybrid approach, we employ Proximal Policy Optimization (PPO), which is used for dynamic tuning of key parameters within the hybrid algorithm. The introduction of PPO allows real-time adjustment of key parameters, such as pheromone evaporation rates and local search intensity, to balance exploration and exploitation more effectively. Comparative experiments against the non-learning version of ACOILS and Simulated Annealing (SA) show that the learning based LACOILS significantly reduces the percentage deviation from the lower bound while maintaining stable performance through dynamic tuning. In terms of numerical results, LACOILS consistently outperforms SA and ACOILS. For smaller instances (N=20), it achieves up to 56.52% improvement over ACOILS and 12.5% over SA. For larger instances (N=150), LACOILS shows up to 29.82% improvement over ACOILS and 9.09% over SA, demonstrating its superior solution quality and efficiency.

Enhanced permeability prediction in porous media using particle swarm optimization with multi-source integration

Accurately and efficiently predicting the permeability of porous media is essential for addressing a wide range of hydrogeological issues. However, the complexity of porous media often limits the effectiveness of individual prediction methods. This study introduces a novel Particle Swarm Optimization-based Permeability Integrated Prediction model (PSO-PIP), which incorporates a particle swarm optimization algorithm enhanced with dynamic clustering and adaptive parameter tuning (KGPSO). The model integrates multi-source data from the Lattice Boltzmann Method (LBM), Pore Network Modeling (PNM), and Finite Difference Method (FDM). By assigning optimal weight coefficients to the outputs of these methods, the model minimizes deviations from actual values and enhances permeability prediction performance. Initially, the computational performances of the LBM, PNM, and FDM are comparatively analyzed on datasets consisting of sphere packings and real rock samples. It is observed that these methods exhibit computational biases in certain permeability ranges. The PSO-PIP model is proposed to combine the strengths of each computational approach and mitigate their limitations. The PSO-PIP model consistently produces predictions that are highly congruent with actual permeability values across all prediction intervals, significantly enhancing prediction accuracy. The outcomes of this study provide a new tool and perspective for the comprehensive, rapid, and accurate prediction of permeability in porous media.

Iterative Tomographic Image Reconstruction Algorithm Based on Extended Power Divergence by Dynamic Parameter Tuning.

Computed tomography (CT) imaging plays a crucial role in various medical applications, but noise in projection data can significantly degrade image quality and hinder diagnosis accuracy. Iterative algorithms for tomographic image reconstruction outperform transform methods, especially in scenarios with severe noise in projections. In this paper, we propose a method to dynamically adjust two parameters included in the iterative rules during the reconstruction process. The algorithm, named the parameter-extended expectation-maximization based on power divergence (PXEM), aims to minimize the weighted extended power divergence between the measured and forward projections at each iteration. Our numerical and physical experiments showed that PXEM surpassed conventional methods such as maximum-likelihood expectation-maximization (MLEM), particularly in noisy scenarios. PXEM combines the noise suppression capabilities of power divergence-based expectation-maximization with static parameters at every iteration and the edge preservation properties of MLEM. The experimental results demonstrated significant improvements in image quality in metrics such as the structural similarity index measure and peak signal-to-noise ratio. PXEM improves CT image reconstruction quality under high noise conditions through enhanced optimization techniques.

Research on dynamic PI parameter tuning of double-fed wind turbine based on ant colony algorithm

Considering the drawbacks of flexible adjustment of PI controller control parameters in the constant pressure control system of doubly fed fans, a PI controller control parameter flexible adjustment method considering ant colony optimization algorithm (ACO) is proposed. This method uses the proportional coefficient and integration time parameters in the PI controller as the ants in the ant colony optimization algorithm, and the ultimate optimization goal is to control the absolute error integration function. During the control process, the parameters of the PI controller are dynamically adjusted. When the electrified wire netting IR-drop and wind velocity changes, a comparative analysis of the dynamic tuning PI controller parameters and the non-dynamic tuning method was conducted. At last, the simulation studies have shown that the constant voltage control system using the PI parameter Flexible adjustment control method found on ACO can rapidly improve the grid voltage stability of double-fed wind turbines and reduce response time.

A dynamic parameter tuning method for SpMM parallel execution

SummarySparse matrix‐matrix multiplication (SpMM) is a basic kernel that is used by many algorithms. Several researches focus on various optimizations for SpMM parallel execution. However, a division of a task for parallelization is not well considered yet. Generally, a matrix is equally divided into blocks for processes even though the sparsities of input matrices are different. The parameter that divides a task into multiple processes for parallelization is fixed. As a result, load imbalance among the processes occurs. To balance the loads among the processes, this article proposes a dynamic parameter tuning method by analyzing the sparsities of input matrices. The experimental results show that the proposed method improves the performance of SpMM for examined matrices by up to 39.5% on a single vector engine and 3.49 on a single CPU.

TradeBot: Bandit learning for hyper-parameters optimization of high frequency trading strategy

Quantitative trading takes advantage of mathematical functions for automatically making stock or futures trading decisions. Specifically, various trading strategies that proposed by human-experts are associated with weight hyper-parameters to determine the probability of selecting a specific strategy according to market conditions. Prior work manually adjusting the weight hyper-parameters is error-prone, because the essential advantage of quantitative trading, i.e., automation, is lost. In this paper, we propose a dynamic parameter tuning algorithm, i.e., TradeBot, based on bandit learning for quantitative trading. We consider sequentially selecting hyper-parameters of rules for trading as a bandit game, where a set of hyper-parameters of trading rule is considered as an action. A novel reward-agnostic Upper Confidence Bound bandit method is proposed to solve the automatically trading problem with a reward function estimated by inverse reinforcement learning. Experimental results on China Commodity Futures Market Data show state-of-the-art performance. To our best knowledge, this is one of the first work deployed in the online trading system via reinforcement learning, in published literature.

Mitigation of subsynchronous control interaction in DFIGs using active disturbance rejection control

Recently, a new type of subsynchronous oscillation (SSO) has occurred in doubly-fed induction generator (DFIG)-based wind farms. This oscillation is also called the subsynchronous control interaction (SSCI), which is mainly caused by the interaction between the transmission series compensation and the DFIG controller. Therefore, the oscillation frequency of the SSCI is susceptibly impacted by DFIG operating conditions, which presents major challenges to conventional mitigation strategies. To solve this problem, a grid-connected series-compensated DFIG-based wind farm is established, and the formation process of the SSCI is analysed. The strategy of active disturbance rejection control (ADRC), which can automatically estimate and compensate for the total disturbance of the system in real-time, is introduced into the rotor-side converter to mitigate the SSCI in the DFIG. ADRC is designed based on dynamic parameter tuning and chaos optimization. Finally, a detailed model-based time-domain simulation is conducted to evaluate the performance of the proposed optimized ADRC (O-ADRC) compared to conventional virtual impedance and ADRC. The simulation results show that O-ADRC can effectively mitigate the SSCI under different operating conditions and provide sufficient damping for the DFIG-based wind farms.

A simple way to achieve black-and-white designs in topology optimization

In order to obtain crisp black-and-white designs, it is vital to use filtering techniques in density-based topology optimization. However, the applying of filters result in a large number of gray transition zones between the solid and void phases in the final designs. To solve this problem, a dynamic parameter tuning method is proposed in this paper. With the approach, the filter radius can be dynamically adjusted during the optimization process, thereby suppressing the gray elements in the final designs. The proposed method is tested by the classic MBB beam problems. The experimental results show that the proposed method can effectively inhibit the appearance of gray transition areas in the optimization results.

Output control of wireless power transmission system based on dynamic parameter tuning

Flexible output power control is very important for wireless power transmission system. In the wireless charging application of electric vehicles, the load is usually the battery. For the battery, both constant current charging and constant voltage charging are needed. In the constant voltage stage, the current should continue to decrease, so we need to adjust the output current of the system in the later stage of charging. Nowadays, the main methods used in WPT system are phase shift and pulse width modulation, adding a first stage DC-DC circuit before the inverter, and controlling the energy injection of the transmitting coil. In view of its shortcomings, this paper uses a method based on adjusting capacitor parameters to achieve output control. Firstly, the influence of capacitance parameters on output current is analyzed, and the relationship between capacitance parameters and output current is deduced; secondly, its advantages are analyzed, and the working principle is analyzed through simulation; finally, the feasibility and correctness of the control method are verified by experiment and simulation.

Speed-Sensorless Control of Linear Induction Motor Based on the SSLKF-PLL Speed Estimation Scheme

In most motor drive applications, the speed sensor is of concern because it potentially deteriorates the system reliability and increases the system cost. To mitigate the adverse effects of speed sensors, various speed estimation schemes are employed to achieve speed-sensorless control of motor drives. With this, the concept of the steady-state linear Kalman filter-based phase-locked loop (SSLKF-PLL) is adopted in this article to enable speed estimation for speed-sensorless control of linear induction motor drives. However, the SSLKF-PLL scheme may be challenged by disturbances, e.g., harmonics and parameter variations in practice. To address this issue, the disturbance rejection schemes, including the prefilters and amplitude normalization unit, are developed for the SSLKF-PLL scheme in this article. Additionally, the dynamic performance analysis and parameter tuning of the SSLKF-PLL scheme are detailed by using the derived small-signal model. Experimental tests have validated the analysis and discussions of the SSLKF-PLL scheme.

A sigmoid attractiveness based improved firefly algorithm and its applications in IIR filter design

Recently, a novel population-based optimisation algorithm, namely firefly algorithm (FA), which mimics the flashing and attraction behaviour of fireflies, has shown promising performance in solving global optimisation problems. However, the preliminary studies have shown that FA often gets stuck in local optima. In this paper, we investigate the reasons why the FA suffers from getting stuck in local optima; and then propose an improved firefly algorithm (IFA). These improvements are twofold: first, a sigmoid-based attractiveness is employed to reformulate its definition and strengthen its local refinement ability; second, a dynamic step parameter tuning strategy is designed to adjust the random search intensity and narrow the search space iteratively to strengthen its global search ability. The empirical results indicate IFA can well balance between the global exploration and the local exploitation, and provides the best solutions, at least the competitive results, for most of 12 global optimisation problems over other FA variants. Besides, by employing IFA to solve well-known infinite impulse response filter design problems, we evaluate the effectiveness and efficiency of IFA. The experimental results and comparisons show that IFA performs better than, at least as competent again, other meta-heuristics in terms of the solution accuracy, solution robustness, and convergence rate.

High-Speed Transfer Optimization Based on Historical Analysis and Real-Time Tuning

Data-intensive scientific and commercial applications increasingly require frequent movement of large datasets from one site to the other(s). Despite growing network capacities, these data movements rarely achieve the promised data transfer rates of the underlying physical network due to poorly tuned data transfer protocols. Accurately and efficiently tuning the data transfer protocol parameters in a dynamically changing network environment is a major challenge and remains as an open research problem. In this paper, we present a novel dynamic parameter tuning algorithm based on historical data analysis and real-time background traffic probing, dubbed HARP. Most of the previous work in this area are solely based on real-time network probing or static parameter tuning, which either result in an excessive sampling overhead or fail to accurately predict the optimal transfer parameters. Combining historical data analysis with real-time sampling lets HARP tune the application-layer data transfer parameters accurately and efficiently to achieve close-to-optimal end-to-end data transfer throughput with very low overhead. Instead of one-time parameter estimation, HARP uses a feedback loop to adjust the parameter values to changing network conditions in real-time. Our experimental analyses over a variety of network settings show that HARP outperforms existing solutions by up to 50 percent in terms of the achieved data transfer throughput.

Learning Change from Synthetic Aperture Radar Images: Performance Evaluation of a Support Vector Machine to Detect Earthquake and Tsunami-Induced Changes

This study evaluates the performance of a Support Vector Machine (SVM) classifier to learn and detect changes in single- and multi-temporal X- and L-band Synthetic Aperture Radar (SAR) images under varying conditions. The purpose is to provide guidance on how to train a powerful learning machine for change detection in SAR images and to contribute to a better understanding of potentials and limitations of supervised change detection approaches. This becomes particularly important on the background of a rapidly growing demand for SAR change detection to support rapid situation awareness in case of natural disasters. The application environment of this study thus focuses on detecting changes caused by the 2011 Tohoku earthquake and tsunami disaster, where single polarized TerraSAR-X and ALOS PALSAR intensity images are used as input. An unprecedented reference dataset of more than 18,000 buildings that have been visually inspected by local authorities for damages after the disaster forms a solid statistical population for the performance experiments. Several critical choices commonly made during the training stage of a learning machine are being assessed for their influence on the change detection performance, including sampling approach, location and number of training samples, classification scheme, change feature space and the acquisition dates of the satellite images. Furthermore, the proposed machine learning approach is compared with the widely used change image thresholding. The study concludes that a well-trained and tuned SVM can provide highly accurate change detections that outperform change image thresholding. While good performance is achieved in the binary change detection case, a distinction between multiple change classes in terms of damage grades leads to poor performance in the tested experimental setting. The major drawback of a machine learning approach is related to the high costs of training. The outcomes of this study, however, indicate that given dynamic parameter tuning, feature selection and an appropriate sampling approach, already small training samples (100 samples per class) are sufficient to produce high change detection rates. Moreover, the experiments show a good generalization ability of SVM which allows transfer and reuse of trained learning machines.

Modeling time series data with deep Fourier neural networks

We present a method for training a deep neural network containing sinusoidal activation functions to fit to time-series data. Weights are initialized using a fast Fourier transform, then trained with regularization to improve generalization. A simple dynamic parameter tuning method is employed to adjust both the learning rate and the regularization term, such that both stability and efficient training are achieved. We show how deeper layers can be utilized to model the observed sequence using a sparser set of sinusoid units, and how non-uniform regularization can improve generalization by promoting the shifting of weight toward simpler units. The method is demonstrated with time-series problems to show that it leads to effective extrapolation of nonlinear trends.

Design of SCADA water resource management control center by a bi-objective redundancy allocation problem and particle swarm optimization

SCADA11Supervisory Control And Data Acquisition: SCADA. is an essential system to control critical facilities in big cities. SCADA is utilized in several sectors such as water resource management, power plants, electricity distribution centers, traffic control centers, and gas deputy. The failure of SCADA results in crisis. Hence, the design of SCADA system in order to serve a high reliability considering limited budget and other constraints is essential. In this paper, a bi-objective redundancy allocation problem (RAP) is proposed to design Tehran׳s SCADA water resource management control center. Reliability maximization and cost minimization are concurrently considered. Since the proposed RAP is a non-linear multi-objective mathematical programming so the exact methods cannot efficiently handle it. A multi-objective particle swarm optimization (MOPSO) algorithm is designed to solve it. Several features such as dynamic parameter tuning, efficient constraint handling and Pareto gridding are inserted in proposed MOPSO. The results of proposed MOPSO are compared with an efficient ε-constraint method. Several non-dominated designs of SCADA system are generated using both methods. Comparison metrics based on accuracy and diversity of Pareto front are calculated for both methods. The proposed MOPSO algorithm reports better performance. Finally, in order to choose the practical design, the TOPSIS algorithm is used to prune the Pareto front.

Study on PI Parameters Dynamic Tuning Based on Ant Colony Algorithm for Doubly-fed Wind Turbines

For the shortcoming that the PI controller parameters can’t been dynamic tuning in constant voltage control system of doubly-fed wind turbines, a PI controller parameters dynamic tuning strategy based on the ant colony optimization (ACO) algorithm is presented. This strategy makes the two parameters in PI controller as the ant of the ant colony, the controlled absolute error integral function that between terminal voltage actual value and its reference value for doubly-fed wind turbines is selected as the optimization objective, the PI controller parameters dynamically is adjusted in the control process. When the wind speed changes and the grid voltage drops, the dynamic and non-dynamic parameter tuning methods are compared and analyzed. The simulation results show that the constant voltage control system using the PI controller parameters tuning strategy based on the ant colony optimization algorithm is superior to non-dynamic tuning method on the stability and response time of system.

An improved PSO-based approach with dynamic parameter tuning for cooperative multi-robot target searching in complex unknown environments

Target searching in complex unknown environments is a challenging aspect of multi-robot cooperation. In this paper, an improved particle swarm optimisation (PSO) based approach is proposed for a team of mobile robots to cooperatively search for targets in complex unknown environments. The improved cooperation rules for a multi-robot system are applied in the potential field function, which acts as the fitness function of the PSO. The main improvements are the district-difference degree and dynamic parameter tuning. In the simulation studies, various complex situations are investigated and compared to the previous research results. The results demonstrate that the proposed approach can enable the multi-robot system to accomplish the target searching tasks in complex unknown environments.

Solving a generalised precedence multi-objective multi-mode time-cost-quality trade-off project scheduling problem using a modified NSGA-II algorithm

In this paper, a new mathematical formulation is proposed to model a generalised precedence multi-objective multi-mode time-cost-quality trade-off project scheduling problem (GPDTCQTP). Afterwards, a modified NSGA-II algorithm is developed to solve the proposed GPDTCQTP. The modified NSGA-II utilises a dynamic parameter tuning and a heuristic self-adaptive constraint handling strategy. These properties result in proper performance in regenerating the Pareto front of the GPDTCQTP. Investigating the efficiency of proposed algorithm several benchmark instances are systematically generated and solved. The proposed procedure is straightforward and results are promising.