Tuning Algorithm Research Articles

Optimal and Quasi-Optimal Automatic Tuning of Vibration Neutralizers

Vibration neutralizers are single-degree-of-freedom devices affixed to vibrating structures in order to reduce the response at a specific troublesome harmonic excitation frequency. As this frequency may vary over time, it becomes imperative to track and adjust the neutralizer to maintain the optimal performance. Recent years have witnessed the emergence of adaptive tunable vibration neutralizers, offering real-time adjustment capabilities through external actions. Thanks to real-time control algorithms, these devices enable the automatic mitigation of vibration levels in mechanical structures. A particularly successful algorithm for the automatic tuning of these devices leverages the phase angle between the base acceleration and the neutralizer’s mass. This study critically examines the justification for employing such an algorithm and scrutinizes its optimal applicability limits, particularly in the context of viscous and structurally damped systems. The findings reveal that this algorithm accurately approximates optimum tuning for systems with low damping. Moreover, from an engineering perspective, the algorithm remains acceptable even for heavily damped structures. Through a focused and comprehensive analysis, this paper provides valuable insights into the efficacy and limitations of the phase-angle-based tuning algorithm, contributing to the advancement of adaptive vibration control strategies in smart structures.

Federated Adaptive Prompt Tuning for Multi-Domain Collaborative Learning

Federated learning (FL) enables multiple clients to collaboratively train a global model without disclosing their data. Previous researches often require training the complete model parameters. However, the emergence of powerful pre-trained models makes it possible to achieve higher performance with fewer learnable parameters in FL. In this paper, we propose a federated adaptive prompt tuning algorithm, FedAPT, for multi-domain collaborative image classification with powerful foundation models, like CLIP. Compared with direct federated prompt tuning, our core idea is to adaptively unlock specific domain knowledge for each test sample in order to provide them with personalized prompts. To implement this idea, we design an adaptive prompt tuning module, which consists of a meta prompt, an adaptive network, and some keys. The server randomly generates a set of keys and assigns a unique key to each client. Then all clients cooperatively train the global adaptive network and meta prompt with the local datasets and the frozen keys. Ultimately, the global aggregation model can assign a personalized prompt to CLIP based on the domain features of each test sample. We perform extensive experiments on two multi-domain image classification datasets across two different settings -- supervised and unsupervised. The results show that FedAPT can achieve better performance with less than 10% of the number of parameters of the fully trained model, and the global model can perform well in diverse client domains simultaneously.

Grey wolf optimization tuned drivetrain vibration controller with backlash compensation strategy using time-dependent-switched Kalman filter

To improve the performance and durability of vehicle components, efforts have been made to reduce driveline oscillations using advanced active control algorithms. However, existing methods often rely on subjective parameter adjustments, which can be burdensome for designers. This study introduces an effective tuning algorithm for a driveline vibration controller that accounts for nonlinear backlash effects. Initially, a driveline dynamics model is developed to focus on transient oscillations resulting from changes in driving force and the presence of nonlinear backlash. The backlash impact is incorporated into the model through a discontinuous dead-zone region. Two operational dynamics, which are the contact mode and the backlash mode, are considered. A dynamic output feedback [Formula: see text] controller is designed as a baseline controller to mitigate low-frequency resonance in the driveline. A solution for managing the nonlinear backlash challenges is introduced, involving the use of a simple control mode switching algorithm in conjunction with the controller. This algorithm relies on a time-dependent-switched Kalman filter. Additionally, the optimal settings for the parameters needed by the mode-switching algorithm are autonomously determined using the grey wolf optimizer (GWO). The proposed active controller can be implemented in real vehicles by using an on-vehicle acceleration sensor and electronic control unit (ECU). In a simulation environment, the vehicle body vibration is online fed back to the resultant controller, and an actuator is supposed to apply control commands to the driveline. The effectiveness of this newly proposed active controller is confirmed through comparative tests, revealing the superior vibration control.

Australian Fire Danger Rating System Research Prototype: a climatology†

Background Historical records of fire weather phenomena provide valuable insights into spatial and temporal trends which can inform further research and are important tools for planning. Aims We outline a 19-year climatology of a Research Prototype (AFDRSRP), of the new Australian Fire Danger Rating System, documenting its spatial and temporal characteristics. Methods The analysis utilises the Australian Bureau of Meteorology’s high-resolution reanalysis suite (BARRA), together with fuel data provided by Australian fire agencies. We examine the spatial and temporal distribution of the AFDRSRP. Distributions are categorised by fuel type, analysing relative variability across time and space. Key results The results validate the broad behaviour of the new system and provide insight into the variation of fire danger throughout Australia, adding detail to the understanding of timing of peak fire danger both diurnally and annually. Conclusions While the AFDRSRP differs from the operational system in its fire danger rating categories and tuning of algorithms, it nonetheless provides useful insights into the operational implementation. Implications These results will be essential for planning during fire seasons.

Adaptive Bayesian contextual hyperband: A novel hyperparameter optimization approach

<p>Hyperparameter tuning plays a significant role when building a machine learning or a deep learning model. The tuning process aims to find the optimal hyperparameter setting for a model or algorithm from a pre-defined search space of the hyperparameters configurations. Several tuning algorithms have been proposed in recent years and there is scope for improvement in achieving a better exploration-exploitation tradeoff of the search space. In this paper, we present a novel hyperparameter tuning algorithm named adaptive Bayesian contextual hyperband (Adaptive BCHB) that incorporates a new sampling approach to identify best regions of the search space and exploit those configurations that produce minimum validation loss by dynamically updating the threshold in every iteration. The proposed algorithm is assessed using benchmark models and datasets on traditional machine learning tasks. The proposed Adaptive BCHB algorithm shows a significant improvement in terms of accuracy and computational time for different types of hyperparameters when compared with state-of-the-art tuning algorithms.</p>

Data-Driven Indirect Iterative Learning Control.

In this work, a data-driven indirect iterative learning control (DD-iILC) is presented for a repetitive nonlinear system by taking a proportional-integral-derivative (PID) feedback control in the inner loop. A linear parametric iterative tuning algorithm for the set-point is developed from an ideal nonlinear learning function that exists in theory by utilizing an iterative dynamic linearization (IDL) technique. Then, an adaptive iterative updating strategy of the parameter in the linear parametric set-point iterative tuning law is presented by optimizing an objective function for the controlled system. Since the system considered is nonlinear and nonaffine with no available model information, the IDL technique is also used along with a strategy similar to the parameter adaptive iterative learning law. Finally, the entire DD-iILC scheme is completed by incorporating the local PID controller. The convergence is proved by applying contraction mapping and mathematical induction. The theoretical results are verified by simulations on a numerical example and a permanent magnet linear motor example.

Learning-Based Faulty State Estimation Using SOH-Coupled Model Under Internal Thermal Faults in Lithium-Ion Batteries

Estimating the state of charge (SOC), state of health (SOH), and core temperature under internal faults will significantly improve the battery management system’s (BMS’s) autonomy and accuracy in range prediction. This paper presents a neural network (NN) based state estimation scheme that can estimate the SOC, core temperature, and SOH under internal faults in lithium-ion batteries (LIBs). First, we propose a model-based internal fault detection scheme by employing a SOH-coupled electro-thermal-aging model (ETA) of the LIB. Then, a nonlinear observer is used to estimate the proposed SOH-coupled model’s healthy states for a residual generation. The fault diagnosis scheme compares the output voltage and surface temperature residuals against the designed adaptive threshold to detect thermal faults. The adaptive threshold effectively alleviates the false positives due to degradation and model uncertainties of the battery under no-fault conditions. Upon fault detection, we employ an additional NN-based observer in the second step to learn the faulty dynamics. A novel NN weight tuning algorithm is proposed using the measured voltage, surface temperature, and estimated healthy states. The convergence of the nonlinear and NN-based observer state estimation errors is proven using the Lyapunov theory. Finally, numerical simulation results are presented.

Data-driven crack image-based seismic failure mode identification for damaged RC columns

The seismic failure mode identification of reinforced concrete columns is crucial for post-earthquake stability and safety assessment of the damaged concrete buildings. In this paper, a multi-feature crack image-based methodology is proposed for non-destructive non-contact failure mode prediction in reinforced concrete columns damaged in an earthquake. A databank of 455 surface crack patterns from quasi-static experiments on 111 reinforced concrete columns with a variety of structural and geometric properties is collected by the authors aimed at the development and validation of the methodology. For the intricacy retrieval of the surface crack maps, multi-feature fractal-based procedures are taken into consideration. Various machine learning algorithms are employed to train the seismic failure mode predictive models using the characteristics of the damaged RC columns. Ten scenarios are defined based on the image-extracted indices, and structural and geometric characteristics of rectangular reinforced concrete columns. Hyperparameter tuning of the machine learning algorithms using GridsearchCV function has optimized the performance of the models. A k-fold cross-validation procedure is also executed to examine the robustness of the results. Results reveal that the predictive model in which the image-extracted indices and structural parameters are jointly used as input features of supervised learning provides 98% accuracy for the testing dataset with Extreme gradient boost algorithm. In addition, among the scenarios that only image-derived parameters are considered as input features, an accuracy of 83% is obtained with Extreme gradient boost algorithm for the testing dataset.

Unsupervised damage assessment under varying ambient temperature based on an adjusted artificial neural network and new multivariate covariance-based distances

Temperature variability is one of the critical environmental conditions that causes confusing changes in structural properties and dynamic responses of bridges similar to damage. In this case, false alarms and mis-detection are among the major errors in health monitoring of such civil structures. High damage detectability is another significant challenge in bridge health monitoring. To deal with these issues, this article proposes an unsupervised damage assessment technique comprising two steps of data normalization and novelty detection. For the first step, an adjusted artificial neural network is considered to remove the effects of temperature variability from dynamic features (modal frequencies). This process is carried out by an auto-associative neural network by adjusting its hidden layer neurons through a new hyperparameter selection algorithm. Using normalized features obtained from the first step, this article proposes three multivariate covariance-based distances called linear dissimilarity analysis, multivariate Kullback–Leibler divergence, and multivariate Bregman distance to compute damage indices or novelty scores for damage assessment. The fundamental principles of these distances lie in three aspects: dividing the normalized features into segments, estimating the covariances of segmented feature sets, and incorporating the estimated covariances into the proposed distance measures. The major contributions of this article include proposing three non-parametric distance measures and developing an unsupervised data normalization framework via a new hyperparameter tuning algorithm for adjusting an artificial neural network. A concrete box-girder bridge is considered to verify the proposed approach, along with several comparative studies. Results show that the method presented here can mitigate severe temperature variability and increase damage detectability with superiority over some traditional and state-of-the-art damage assessment techniques.

Crow search algorithm with deep transfer learning driven skin lesion detection on dermoscopic images

The early diagnosis of skin cancer is of paramount importance for effective patient treatment. Dermoscopy, a non-surgical technique, utilizes precise equipment to examine the skin and plays a crucial role in identifying specific features and patterns that may indicate the presence of skin carcinoma. In recent times, machine learning (ML) methods have been developed to recognize and classify dermoscopic images as either malignant or benign. Deep learning (DL) systems, including Convolutional Neural Networks (CNNs), as well as various ML models like Random Forest (RF) classifiers and Support Vector Machine (SVM), are employed to extract relevant features from these images. This study introduces the Crow Search Algorithm with Deep Transfer Learning Driven Skin Lesion Detection on Dermoscopic Images (CSADTL-SLD) technique. The CSADTL-SLD method starts with the application of a median filter (MF) to remove noise from the images and utilizes the GoogleNet model for feature extraction. GoogleNet is well-regarded for its capacity to capture intricate and meaningful patterns within the data, which are essential for accurate lesion characterization. Furthermore, the CSADTL-SLD technique applies the Crow Search Algorithm (CSA) for parameter tuning of the GoogleNet model. After feature selection, the system employs the MLP classification model for precise lesion categorization. The comprehensive results of this research demonstrate the superiority of the CSADTL-SLD algorithm, showing significant enhancements in skin lesion detection accuracy and robustness when compared to existing methods. This approach holds promise as an effective solution for automating the detection and classification of skin lesions in dermoscopic images.

A low-cost wireless extension for object detection and data logging for educational robotics using the ESP-NOW protocol.

In recent years, inexpensive and easy to use robotics platforms have been incorporated into middle school, high school, and college educational curricula and competitions all over the world. Students have access to advanced microprocessors and sensor systems that engage, educate, and encourage their creativity. In this study, the capabilities of the widely available VEX Robotics System are extended using the wireless ESP-NOW protocol to allow for real-time data logging and to extend the computational capabilities of the system. Specifically, this study presents an open source system that interfaces a VEX V5 microprocessor, an OpenMV camera, and a computer. Images from OpenMV are sent to a computer where object detection algorithms can be run and instructions sent to the VEX V5 microprocessor while system data and sensor readings are sent from the VEX V5 microprocessor to the computer. System performance was evaluated as a function of distance between transmitter and receiver, data packet round trip timing, and object detection using YoloV8. Three sample applications are detailed including the evaluation of a vision-based object sorting machine, a drivetrain trajectory analysis, and a proportional-integral-derivative (PID) control algorithm tuning experiment. It was concluded that the system is well suited for real time object detection tasks and could play an important role in improving robotics education.

Stabilization and tracking control of underactuated ball and beam system using metaheuristic optimization based TID-F and PIDD2-PI control schemes.

In this paper, we propose two different control strategies for the position control of the ball of the ball and beam system (BBS). The first control strategy uses the proportional integral derivative-second derivative with a proportional integrator PIDD2-PI. The second control strategy uses the tilt integral derivative with filter (TID-F). The designed controllers employ two distinct metaheuristic computation techniques: grey wolf optimization (GWO) and whale optimization algorithm (WOA) for the parameter tuning. We evaluated the dynamic and steady-state performance of the proposed control strategies using four performance indices. In addition, to analyze the robustness of proposed control strategies, a comprehensive comparison has been performed with a variety of controllers, including tilt integral-derivative (TID), fractional order proportional integral derivative (FOPID), integral-proportional derivative (I-PD), proportional integral-derivative (PI-D), and proportional integral proportional derivative (PI-PD). By comparing different test cases, including the variation in the parameters of the BBS with disturbance, we examine step response, set point tracking, disturbance rejection analysis, and robustness of proposed control strategies. The comprehensive comparison of results shows that WOA-PIDD2-PI-ISE and GWO-TID-F- ISE perform superior. Moreover, the proposed control strategies yield oscillation-free, stable, and quick response, which confirms the robustness of the proposed control strategies to the disturbance, parameter variation of BBS, and tracking performance. The practical implementation of the proposed controllers can be in the field of under actuated mechanical systems (UMS), robotics and industrial automation. The proposed control strategies are successfully tested in MATLAB simulation.

Efficient data-driven machine learning models for scour depth predictions at sloping sea defences

Seawalls are critical defence infrastructures in coastal zones that protect hinterland areas from storm surges, wave overtopping and soil erosion hazards. Scouring at the toe of sea defences, caused by wave-induced accretion and erosion of bed material imposes a significant threat to the structural integrity of coastal infrastructures. Accurate prediction of scour depths is essential for appropriate and efficient design and maintenance of coastal structures, which serve to mitigate risks of structural failure through toe scouring. However, limited guidance and predictive tools are available for estimating toe scouring at sloping structures. In recent years, Artificial Intelligence and Machine Learning (ML) algorithms have gained interest, and although they underpin robust predictive models for many coastal engineering applications, such models have yet to be applied to scour prediction. Here we develop and present ML-based models for predicting toe scour depths at sloping seawall. Four ML algorithms, namely, Random Forest (RF), Gradient Boosted Decision Trees (GBDT), Artificial Neural Networks (ANNs), and Support Vector Machine Regression (SVMR) are utilised. Comprehensive physical modelling measurement data is utilised to develop and validate the predictive models. A Novel framework for feature selection, feature importance, and hyperparameter tuning algorithms are adopted for pre- and post-processing steps of ML-based models. In-depth statistical analyses are proposed to evaluate the predictive performance of the proposed models. The results indicate a minimum of 80% prediction accuracy across all the algorithms tested in this study and overall, the SVMR produced the most accurate predictions with a Coefficient of Determination (r2) of 0.74 and a Mean Absolute Error (MAE) value of 0.17. The SVMR algorithm also offered most computationally efficient performance among the algorithms tested. The methodological framework proposed in this study can be applied to scouring datasets for rapid assessment of scour at coastal defence structures, facilitating model-informed decision-making.

Convergence and Performance Analysis of a Particle Swarm Optimization Algorithm for Optical Tuning of Gold Nanohole Arrays.

Gold nanohole arrays, hybrid metal/dielectric metasurfaces composed of periodically arranged air holes in a thick gold film, exhibit versatile support for both localized and propagating surface plasmons. Leveraging their capabilities, particularly in surface plasmon resonance-oriented applications, demands precise optical tuning. In this study, a customized particle swarm optimization algorithm, implemented in Ansys Lumerical FDTD, was employed to optically tune gold nanohole arrays treated as bidimensional gratings following the Bragg condition. Both square and triangular array dispositions were considered. Convergence and evolution of the particle swarm optimization algorithm were studied, and a mathematical model was developed to interpret its outcomes.

Learning-Based Parameter Optimization for a Class of Orbital Tracking Control Laws

This paper presents a learning algorithm for tuning the parameters of a family of stabilizing nonlinear controllers for orbital tracking, in order to minimize a cost function which combines convergence time and fuel consumption. The main feature of the proposed approach is that it achieves performance optimization while guaranteeing closed-loop stability of the resulting controller. This property is exploited also to restrict the class of admissible controllers and hence to expedite the training process. The learning algorithm is tested on three case studies: two different orbital transfers and a rendezvous mission. Numerical simulations show that the learned control parameters lead to a significant improvement of the considered performance measure.

Cultivating Ensemble Diversity through Targeted Injection of Synthetic Data: Path Loss Prediction Examples

Machine Learning (ML)-based models are steadily gaining popularity. Their performance is determined from the amount and the quality of data used at their inputs, as well as from the competence and proper tuning of the ML algorithm used. However, collecting high-quality real data is time-consuming and expensive. Synthetic Data Generation (SDG) is therefore employed in order to augment the limited real data. Moreover, Ensemble Learning (EL) provides the framework to optimally combine a set of standalone ML algorithms (base learners), capitalizing on their individual strengths. Base learner diversity is essential to build a strong ensemble. The proposed method of Targeted Injection of Synthetic Data (TIoSD) combines the EL and SDG concepts in order to further diversify the base learners’ predictions, thus giving rise to an even stronger ensemble model. We have applied TIoSD in two different Path Loss (PL) datasets, using two well-established SDG methods (namely SMOGN and CTGAN). While the conventional ensemble model reached a Minimum Absolute Error (MAE) value of 3.25 dB, the TIoSD-triggered ensemble provided a MAE value of 3.16 dB. It is therefore concluded that targeted synthetic data injection, due to its diversity-triggering characteristics, enhances the ensemble’s performance. Moreover, the ratio between synthetic and real data has been investigated. The results showed that a proportion of 0.1 is optimal.

Fuzzy system of automatic tuning of electromagnetic excitation system of vibration-frequency liquid density meter

A fuzzy system for setting the location of electromagnetic exciter of vibration-frequency density meter is presented, taking into account the possible nonlinearities of the object. This paper proposes a system and algorithm for tuning the location of electromagnetic exciter of vibration-frequency density meter using fuzzy set theory. The use of fuzzy control methods allowed us to obtain a system for tuning the location of the electromagnetic exciter of the vibration-frequency density meter without the use of cumbersome computational procedures characteristic of the classical control method.

Better trees: an empirical study on hyperparameter tuning of classification decision tree induction algorithms

Better trees: an empirical study on hyperparameter tuning of classification decision tree induction algorithms

Resolution enhanced sea ice concentration: a new algorithm applied to AMSR2 microwave radiometry data

Abstract Passive-microwave sea ice concentration (SIC) algorithms employ different frequencies and polarisations in their operational implementations. Commonly, these algorithms utilise combinations such as 19/37 GHz, yielding reduced measurement uncertainties but at a coarse spatial resolution. Alternatively, these algorithms can solely use 89 GHz, producing a higher spatial resolution but with increased measurement uncertainties. This study evaluates the application of a resolution-enhancing SIC algorithm (reSICCI3LF), initially developed for the coarser Special Sensor Microwave Imager / Sounder, on the Advanced Microwave Scanning Radiometer. By applying reSICCI3LF, we aim to produce a 5 km SIC for 2013–2020 in the Fram Strait and the Barents and Kara Sea region that gains the benefits of both types of algorithms, high spatial resolution and low measurement uncertainty. We present the algorithm tuning, spectral analysis of spatial resolutions, and validation against the Round Robin Data Package of 0% and 100% SIC points and SIC derived from Landsat-8. The findings demonstrate that the reSICCI3LF algorithm produces a SIC field with fine details, achieving a balance between high spatial resolution and lower measurement uncertainties compared to a 89 GHz based SIC. Consequently, this resolution-enhanced SIC technique can potentially initialise higher-resolution coupled ocean and sea ice forecasting systems through data assimilation.

Optimizing 3D printed diamond lattice structure and investigating the influence of process parameters on their mechanical integrity using nature-inspired machine learning algorithms

This research investigates the effect of process parameters on the mechanical integrity of 3D-printed Polylactic Acid (PLA+) lattice structures, with a focus on diamond lattice configurations. Adhering to ASTM D695 standards, the study employs compression testing to analyze the impact of various factors, including layer height, infill pattern, cell size, and infill density. The printing process utilizes the Ultimaker Cura Fused Deposition Modelling method, ensuring accuracy and uniformity in sample production. Mechanical properties were assessed using a Universal Testing Machine, highlighting the influence of the studied parameters on compressive strength and specific energy absorption (SEA). Innovatively, this study integrates nature-inspired machine-learning algorithms for hyperparameter tuning. Particle Swarm Optimization (PSO) is employed to optimize the settings of Random Forest and XGBoost models, enhancing their predictive capabilities. Results indicate that XGBoost, fine-tuned with PSO, slightly outperforms its counterpart in predicting key mechanical properties. Furthermore, feature significance assessments reveal cell size as a critical factor for compressive strength. This research not only provides a comprehensive analysis of mechanical properties in 3D-printed structures but also introduces a novel approach to applying nature-inspired algorithms for machine learning optimization in additive manufacturing.