Set Of Tuning Parameters Research Articles

Artificial bee colony with predator effect algorithm for proportional integral derivative controller tuning

This paper proposes a novel artificial bee colony with predation effect (ABCPE) algorithm for tuning a proportional integral derivative (PID) controller. The mathematical model of ABCPE algorithm to introduce predator effect in the foraging behavior of artificial bees colony algorithm has been formulated. The proposed algorithm has been tested on tuning problems of different process models. The simulation results reveal that the closed-loop responses are relatively fast and non-oscillatory as compared to the frequency response analysis method for reference tracking. Further, the results obtained using ABCPE are also compared with other evolutionary algorithms. The exhaustive analysis shows that the ABCPE-based solution approach leads to a set of tuning parameters having smaller overshoot, less setting time, and rise time compared to other solution approaches. The stability analysis using Nichols plot reveals that the phase margin of proposed algorithm is higher as compared to other tuning methods. Finally, the convergence behavior and robustness analysis reveals the effectiveness of the proposed approach to solve engineering design problems.

The causes of bank customer churn based on XGBoost and LightGBMmodels: the evidence from the Kaggle dataset

As the digital economy continues to grow, the expansion of Internet finance introduces new challenges for the conventional banking industry. Banks must deal with multiple pressures, such as digital transformation, declining customer loyalty, and fintech competition. Analyzing the potential factors of bank customer churn from multiple perspectives and constructing models for predicting churn can help bank managers understand the causes of churn, identify problems, detect potential churn customers promptly, and develop efficient retention strategies based on customer characteristics and preferences. In this paper, we used a combination of visualization, data mining, and machine learning methods to analyze the factors used to predict bank customer churn from multiple perspectives, such as feature selection (Random Forest Feature Importance Ranking), feature extraction (PCA), visualization, etc. We also constructed two churn prediction models based on the gradient boosting tree algorithms, XGBoost and LightGBM, compared the evaluation measures before and after feature selection and before and after tuning parameters, and interpreted the model through SHAP methods. After the paper, the following conclusions were drawn: (1) Total Trans Amt, Total Trans Ct, and Total Revolving Bal are pivotal in analyzing and predicting customer churn; (2) the SHAP Summary Plot can react to the visual analysis of predictors of customer churn to a certain extent; (3) the effect of feature selection on the assessment of the results is sometimes insignificant; (4) tuning parameter settings can enhance model performance to a certain extent, but the optimal parameters may vary based on the preprocessing method employed. These conclusions will assist banks in comprehending customer churn factors more deeply, constructing a higher performance churn prediction model, and conducting a comprehensive result synthesis analysis.

Integrated multiple criteria decision-making framework for ranking Pareto optimal solutions of the multiobjective optimization problem of tuned mass dampers

Tuned mass dampers (TMD) have demonstrated effectiveness and robustness in controlling vibrations in marine structures. Solutions to the TMD's Multi-objectives optimization problem (MOOP) always yield a set of Pareto compromise solutions for different conflicting objectives. However, because all the Pareto solutions are non-dominated, the solution of the MOOP still needs to be completed since a single Pareto solution must be selected from a large set of Pareto solutions. Researchers in the literature concentrated on obtaining the Pareto front by solving the MOOP of the TMD for different objective functions. Nonetheless, no effort has been put into selecting the best solution from the Pareto set. Therefore, this paper presents the novel application of the multiple-criteria decision-making (MCDM) techniques in the solution of the MOOP of the TMD applied to a marine structure. The presented framework ensures that the decision-makers (DMs) obtain the most optimal solution from the Pareto set of solutions of the TMD. In this paper, the problem of the TMD applied to a jacket platform is formulated as a MOOP, which comprises conflicting objectives related to TMD cost and performance. NSGA-II is utilized to find a large set of non-dominated TMD tuning parameters from which the DMs must select only one solution. Then, the DMs' preferences are prioritized using the Analytical Hierarchy Process (AHP). Meanwhile, the reasonableness of the DMs' preferences is assessed. The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is utilized for ranking the non-dominated set of solutions by assessing the relative closeness of each solution to the ideal solution and farness away from the negative ideal solution. A Finite Element (FE) model of a fixed offshore platform under the dynamic disturbance of irregular waves is numerically solved using COMSOL multiphysics to assess the goals mentioned earlier. A TMD is applied to the marine structure to mitigate wave-induced vibrations. The hydrodynamic forces affecting the structure are computed using Morison's equation, and the irregular waves are characterized using the Joint North Sea Wave Project (JONSWAP) wave spectrum. The presented framework will be used for ranking the Pareto solutions of the TMD applied to the deck of the FE model. The TMD tuning parameters considered in this regard are the TMD mass ratio μ, damping ratio ξopt, and frequency ratio fopt. The findings indicate that the framework presented provides an efficient decision-making methodology for selecting the best solution in various design scenarios. These scenarios can be presented in 2D, 3D, or high-dimensional objective space. Meanwhile, the DMs will be able to select the best compromise among the objectives without the need for numerical judgments. As a result, an efficient TMD that provides valuable response mitigation in a low-cost and effective space was presented.

EEG-Based Driver Fatigue Detection Using FAWT and Multiboosting Approaches

Globally, 14%–20% of road accidents are mainly due to driver fatigue, the causes of which are instance sickness, travelling for long distance, boredom as a result of driving along the same route consistently, lack of enough sleep, etc. This article presents a flexible analytic wavelet transform (FAWT)-based advanced machine learning method using single modality neurophysiological brain electroencephalogram signals to detect the driver fatigues (i.e., FATIGUE and REST) and to alarm the driver at the earliest to prevent the risks during driving. First, signals of undertaking study groups are subjected to the FAWT that separates the signals into LP and HP channels. Subsequently, relevant subband frequency components with proper setting of tuning parameters are extracted. Then, comprehensive low order features which are statistically significant for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p&lt; 0.05$</tex-math></inline-formula> , are evaluated from the input subband searched space and embedded them to various ensemble methods under multiboost strategy. Results are evaluated in terms of various parameters including accuracy, F-score, AUC, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\kappa$</tex-math></inline-formula> . Results show that the proposed approach is promising in classification and it achieves optimum individual accuracies of 97.10% and 97.90% in categorizing FATIGUE and REST states with F-score of 97.50%, AUC of 0.975, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\kappa$</tex-math></inline-formula> of 0.950. Comparison of the proposed method with the prior methods in the context of feature, accuracy, and modality profiles undertaken, indicates the effectiveness and reliability of the proposed method for real-world applications.

Type-II progressive censoring with GLM-based random removal mechanism dependent on the experimental conditions

This article presents a novel stochastic removal mechanism under Type-II progressive random censoring in which removal probabilities are allowed to be dependent on the lifetime conditions through Generalized Linear Models (GLM). These conditions potentially include failure distances (the time required to observe the next failure) or other covariate information available in the experiment. The proposed GLM-based random removal mechanism includes a set of tuning parameters that are determined by the researcher according to the possible failure distance category. These parameters allow flexible determination of the removal probabilities leading to necessary experimental cost and time reductions. To establish the proposed mechanism, the Proportional Hazard Rate (PHR) family of distributions is considered. Also, the maximum likelihood estimators of parameters and their asymptotic variances are derived for the Weibull distributed lifetime data. A simple simulation algorithm for generating Type-II progressive censoring samples with GLM-based dependent removal probabilities is also presented. The expected experiment time required to complete the life test under this censoring scheme is also investigated using the Monte Carlo integration method. Several simulation studies are conducted to evaluate and compare the performance of the proposed mechanism. A sensitivity analysis is also considered to study the effect of misspecification of removal mechanism coefficients. Finally, two real data sets are analyzed for illustrative purposes.

The unpopular Package: A Data-driven Approach to Detrending TESS Full-frame Image Light Curves

The majority of observed pixels on the Transiting Exoplanet Survey Satellite (TESS) are delivered in the form of full-frame images (FFIs). However, the FFIs contain systematic effects such as pointing jitter and scattered light from the Earth and Moon that must be removed (i.e., “detrended”) before downstream analysis. We present unpopular, an open-source Python package to obtain detrended TESS FFI light curves optimized for variable sources. The unpopular package implements a variant of the causal pixel model to remove systematics and allows for simultaneous fitting with a polynomial component to capture nontransit astrophysical variations, such as supernova signals or stellar variability, that tend to be removed in techniques optimized for exoplanet detection. We validate our method by detrending different sources (e.g., supernovae, tidal disruption events (TDEs), exoplanet-hosting stars, fast-rotating stars) and comparing our light curves to those obtained by other pipelines when appropriate. Our supernova and TDE light curves are visually similar to those obtained by others using the ISIS image subtraction package, indicating that unpopular can be used to extract multisector light curves by preserving astrophysical signals on timescales of a TESS sector (∼27 days). We note that our method contains tuning parameters that are currently set heuristically, and that the optimal set of tuning parameters will likely depend on the particular signal the user is interested in obtaining. The unpopular source code and tutorials are freely available online.

Altering the tuning parameter settings of a commercial powered prosthetic foot to increase power during push-off may not reduce collisional work in the intact limb during gait.

Increased knee osteoarthritis risk in patients with unilateral lower extremity limb loss is attributed to increased intact limb loading. Modulating powered ankle prosthesis push-off power may be an effective way to modulate intact limb loading. We examined how changes in the parameter settings of a commercial prosthetic ankle affect power delivery during push-off and the resulting collisional work experienced by the intact limb in persons with unilateral lower extremity limb loss. Five subjects with unilateral transtibial amputation were fitted with a commercially available powered ankle prosthesis (Ottobock Empower). Subjects walked on a treadmill in seven conditions, where ankle power delivery settings were adjusted using methods accessible to clinicians. Kinetics and kinematics data were collected. Standard adjustment of parameter settings within the prosthetic foot did not alter timing of peak prosthesis power or intact limb collisional work but did have a significant effect on the magnitude of positive prosthesis ankle work. Increased prosthesis work did not decrease intact limb collisional work as predicted. Altering the parameter settings on a commercial powered ankle prosthesis affected the magnitude, but not the timing, of power delivered. Increased prosthesis push-off power did not decrease intact limb loading.

SERA: An ISIP analysis approach to estimate fracture height and correlate stress escalation and relaxation in multifrac horizontal wells

In hydraulic fracturing of horizontal wells in unconventional reservoirs, every fracturing stage by increasing the minimum horizontal stress modifies the vertical stress contrast as well as reduces the stress anisotropy. The induced stress gradually declines thereafter due to fracture closure caused by fluid leak-off. The time-dependent stress behavior with localized spatial effects generated by fixed but dynamic sources offers an incentive to fracturing optimization studies. Here, drawing upon the stress superposition principle, an analytical model is derived for Stress Escalation and Relaxation Analysis (SERA) to correlate the stress state at the wellbore. The SERA's model possesses a set of tuning parameters that are adaptively adjusted based on perforation design, waiting time between stage treatments, and ISIP of all completion stages. As a computationally efficient ISIP analysis technique, SERA estimates the average height of wet fracture and sets an upper limit to the critical stress state. In case of a dominant vertical stress contrast, SERA allows to determine explicitly the conditions under which the onset of wet fractures into the bounding layers of lower stress takes place, and when the stress re-orientation controls the fracturing regime, it narrows down the range of estimates of the horizontal stress anisotropy. Furthermore, it provides a tool to design optimal timing and sequencing strategies for alternating fracturing by correlating post-treatment stress relaxation parameters. We validate the model with numerical simulations and perform a sensitivity analysis on the influential parameters that can be carefully designed to give more control over stress shadowing.

Out-of-Sample Predictability of Economic Efficiency Measures of U.S. Banks: Evidence of Capital Adequacy Requirements

Abstract This paper contributes to the sparse debate on the effect of capital adequacy requirements on banks’ economic efficiency measures. Precisely, we evaluate the out-of-sample predictability of capital adequacy requirements on banks’ economic efficiency measures using Support Vector Regression (SVR) model with Linear, Polynomial and Radial Basis Function kernels and ordinary least squares (OLS) model. This analysis is important because a prediction of economic efficiency measures allows for an untangle view of bank’s progress that is useful for management as it gains a high degree of transparency in the evaluation of future events. Our framework adapts optimization of h-block cross-validation to account for serial correlation of economic variables to produce robust sets of tuning parameters for SVR model. Using a total of 10,380 December quarterly observations of U.S. Commercial and Domestic banks spanning from 2008 through 2019, empirical results show that SVR model provides better benchmarking insights in the evaluation of economic efficiency measures compared to the OLS model. Furthermore, in contrast to previous approaches identifying a single “best” model among competing models, the results of Model Confidence Test suggests that the out-of-sample forecasting confidently identifies superior predictive accuracy of SVR model-based forecasts over OLS model.

Performance evaluation of an optimized PID controller for propofol and remifentanil coadministration in general anesthesia

In this paper we present the preliminary clinical results obtained with a PID control scheme for propofol and remifentanil coadministration during general anesthesia. The bispectral index scale is the only process variable, and the extra degree of freedom in the control architecture is handled by introducing an appropriate ratio between the infusion rates of the two drugs. The parameters of the PID controller are selected by means of an optimization procedure. The proposed control system automatically handles both the induction and the maintenance phase by using two different set of tuning parameters, specifically optimized for each phase, and a gain scheduling approach. The preliminary clinical evaluation of our controller has been obtained on ten patients undergoing general anesthesia during plastic surgery. The system has been able to satisfactorily induce and maintain anesthesia in the entire population of patients, without requiring any manual intervention by the anesthesiologist, thus confirming the effectiveness of the overall design approach.

Data-Driven Scenario Optimization for Automated Controller Tuning With Probabilistic Performance Guarantees

Systematic design and verification of advanced control strategies for complex systems under uncertainty largely remains an open problem. Despite the promise of black-box optimization methods for automated controller tuning, they generally lack formal guarantees on the solution quality, which is especially important in the control of safety-critical systems. This letter focuses on obtaining closed-loop performance guarantees for automated controller tuning, which can be formulated as a black-box optimization problem under uncertainty. We use recent advances in non-convex scenario theory to provide a distribution-free bound on the probability of the closed-loop performance measures. To mitigate the computational complexity of the data-driven scenario optimization method, we restrict ourselves to a discrete set of candidate tuning parameters. We propose to generate these candidates using constrained Bayesian optimization run multiple times from different random seed points. We apply the proposed method for tuning an economic nonlinear model predictive controller for a semibatch reactor modeled by seven highly nonlinear differential equations.

Time Series Prediction of Electricity Demand Using Adaptive Neuro-Fuzzy Inference Systems

This paper is concerned with the reliable prediction of electricity demands using the Adaptive Neuro-Fuzzy Inference System (ANFIS). The need for electricity demand prediction is fundamental and vital for power resource planning and monitoring. A dataset of electricity demands covering the period of 2003 to 2018 was collected from the Electricity Distribution Company of Ghana, covering three urban areas namely Mallam, Achimota, and Ga East, all in Ghana. The dataset was divided into two parts: one part covering a period of 0 to 500 hours was used for training of the ANFIS algorithm while the second part was used for validation. Three scenarios were considered for the simulation exercise that was done with the MATLAB software. Scenario one considered four inputs sampled data, scenario two considered an additional input making it 5, and scenario 3 was similar to scenario 1 with the exception of the number of membership functions that increased from 2 to 3. The performance of the ANFIS algorithm was assessed by comparing its predictions with other three forecast models namely Support Vector Regression (SVR), Least Square Support Vector Machine (LS-SVM), and Auto-Regressive Integrated Moving Average (ARIMA). Findings revealed that the ANFIS algorithm can perform the prediction accurately, the ANFIS algorithm converges faster with an increase in the data used for training, and increasing the membership function resulted in overfitting of data which adversely affected the RMSE values. Comparison of the ANFIS results to other previously used methods of predicting electricity demands including SVR, LS-SVM, and ARIMA revealed that there is merit to the potentials of the ANFIS algorithm for improved predictive accuracy while relying on a quality data for training and reliable setting of tuning parameters.

A novel meta-heuristic search algorithm for solving optimization problems: capuchin search algorithm

Meta-heuristic search algorithms were successfully used to solve a variety of problems in engineering, science, business, and finance. Meta-heuristic algorithms share common features since they are population-based approaches that use a set of tuning parameters to evolve new solutions based on the natural behavior of creatures. In this paper, we present a novel nature-inspired search optimization algorithm called the capuchin search algorithm (CapSA) for solving constrained and global optimization problems. The key inspiration of CapSA is the dynamic behavior of capuchin monkeys. The basic optimization characteristics of this new algorithm are designed by modeling the social actions of capuchins during wandering and foraging over trees and riverbanks in forests while searching for food sources. Some of the common behaviors of capuchins during foraging that are implemented in this algorithm are leaping, swinging, and climbing. Jumping is an effective mechanism used by capuchins to jump from tree to tree. The other foraging mechanisms exercised by capuchins, known as swinging and climbing, allow the capuchins to move small distances over trees, tree branches, and the extremities of the tree branches. These locomotion mechanisms eventually lead to feasible solutions of global optimization problems. The proposed algorithm is benchmarked on 23 well-known benchmark functions, as well as solving several challenging and computationally costly engineering problems. A broad comparative study is conducted to demonstrate the efficacy of CapSA over several prominent meta-heuristic algorithms in terms of optimization precision and statistical test analysis. Overall results show that CapSA renders more precise solutions with a high convergence rate compared to competitive meta-heuristic methods.

Fast tensorial JADE.

We propose a novel method for tensorial‐independent component analysis. Our approach is based on TJADE and k‐JADE, two recently proposed generalizations of the classical JADE algorithm. Our novel method achieves the consistency and the limiting distribution of TJADE under mild assumptions and at the same time offers notable improvement in computational speed. Detailed mathematical proofs of the statistical properties of our method are given and, as a special case, a conjecture on the properties of k‐JADE is resolved. Simulations and timing comparisons demonstrate remarkable gain in speed. Moreover, the desired efficiency is obtained approximately for finite samples. The method is applied successfully to large‐scale video data, for which neither TJADE nor k‐JADE is feasible. Finally, an experimental procedure is proposed to select the values of a set of tuning parameters. Supplementary material including the R‐code for running the examples and the proofs of the theoretical results is available online.

First experiments of anesthesia control with optimized PID tuning

In this paper we present and discuss the first experimental results obtained with a recently devised PID control scheme for the propofol and remifentanil coadministration in general anesthesia. In particular, the depth of hypnosis is controlled by considering only the bispectral index scale as the process variable and the extra degree of freedom in the controller is handled by selecting an appropriate ratio between the infusion rates of the two drugs. The parameters of the PID controllers are selected by using a tuning rule obtained through an optimization procedure that exploits the PK/PD model of a set of patients. A gain scheduling approach is used to switch between two sets of tuning parameters, one for the induction phase and the other for the maintenance phase. The experimental results confirm the effectiveness of the overall design approach.

Simple tuning of modified Smith predictor for unstable FOPTD processes

A simple tuning for modified Smith predictor is reported here to control unstable first-order plus time delay (FOPTD) processes. In practice, a good number of chemical processes (like combustion, evaporation, distillation, etc.) are unstable as well as delay dominated in nature. Modified Smith predictor technique is a well-accepted methodology for its simplicity and efficacy to regulate delay dominated processes to ascertain desirable closed-loop response. Two-degree-of-freedom (2-DOF) control scheme of the proposed multi-loop controller provides performance enhancement during both set point tracking and load recovery phases. To mitigate tuning complexity of more than one controller involved in modified Smith predictor, the proposed simple guideline involves the same set of tuning parameters for all three controllers. Distinctive feature of the proposed tuning scheme is that there is no overshoot during set point tracking as well as smooth and improved recovery is observed subsequent to uncertain load changes. Superiority of the proposed methodology is established through performance evaluation and stability analysis with well-accepted modified Smith predictor-based tuning relations.

A Coherent Framework for Predicting Emerging Market Credit Spreads with Support Vector Regression

We propose a coherent framework using support vector regression (SRV) for generating and ranking a set of high quality models for predicting emerging market sovereign credit spreads. Our framework adapts a global optimization algorithm employing an hv-block cross-validation metric, pertinent for models with serially correlated economic variables, to produce robust sets of tuning parameters for SRV kernel functions. In contrast to previous approaches identifying a single "best" tuning parameter setting, a task that is pragmatically improbable to achieve in many applications, we proceed with a collection of tuning parameter candidates, employing the Model Confidence Set test to select the most accurate models from the collection of promising candidates. Using bond credit spread data for three large emerging market economies and an array of input variables motivated by economic theory, we apply our framework to identify relatively small sets of SVR models with su perior out-of-sample forecasting performance. Benchmarking our SRV forecasts against random walk and conventional linear model forecasts provides evidence for the notably superior forecasting accuracy of SRV-based models. In contrast to routinely used linear model benchmarks, the SRV-based models can generate accurate forecasts using only a small set of input variables limited to the country-specific credit-spread-curve factors, lending some support to the rational expectation theory of the term structure in the context of emerging market credit spreads. Consequently, our evidence indicates a better ability of highly flexible SVR to capture investor expectations about future spreads reflected in today's credit spread curve.

A Binary Distillation Column Top and Bottom Compositions Control Using Two PI Controllers Tuned with SIMC

The opportunity analysis of using two Proportional-Integral (PI) controllers, tuned with Skogestad�s variant of Simple Internal Model Control based (SIMC) method, for controlling the top and bottom compositions of a propylene/propane distillation column represents the paper aim. The SIMC method for PID controllers is the simplest and extremely good model in tuning procedures. Because the distillation process is nonlinear, the process model is described by a reunion of different linear models, for the different process operating ranges and, for each operating range, different sets of controllers� tuning parameters are computed using SIMC tuning method. In order to validate the experimental results a comparison is made with the case when the PI controllers are tuned using Ziegler-Nichols (ZN) method that uses the step response and process approximation exactly as Skogestad�s variant of SIMC.

INTEGRATING REAL TIME OPTIMIZATION AND MODEL PREDICTIVE CONTROL OF A CRUDE DISTILLATION UNIT

This work reports the integration of Real Time Optimization and Model Predictive Control in the multi-layer control structure of an existing Crude Distillation Unit (CDU) of an oil refinery. The MPC considers output control zones and targets for the inputs or outputs. Both the infinite horizon and the finite output horizon controllers were tested. The plant results show that the infinite horizon controller tends to perform similarly or better then the finite horizon MPC when the CDU system needs to operate at quite different conditions. Although the dynamic layer based on the infinite horizon controller is nominally stable for any set of tuning parameters, in practice, it is observed that the interaction between the layers of the control structure associated to model uncertainty may result in oscillations in some variables that fail to converge to the optimum operation point. This problem can be solved with the retuning of the intermediary layer (target calculation layer), which indicates that the frequent tuning of the MPC is recommended and should be performed in conjunction with tuning of the intermediary layer.

Performance of biologically inspired algorithms tuned on TiO2 nanoparticle benchmark system

Biologically inspired optimization algorithms (BIOA), such as particle swarm optimization (PSO) and differential evolution (DE), are often applied to a specific problem with the tacit assumptions that each algorithm will have a unique set of “ideal” tuning parameters that provide “best” performance, and that, once tuned, two different optimization algorithms ought to exhibit distinct performance. This work tests those assumptions by systematically comparing PSO and DE applied to identify stable and metastable configurations of a benchmark system, a titania dimer Ti2O4, whose energy is described by a modified Matsui-Akaogi pairwise potential. The hypotheses (H) are formally expressed as: (H1) for both PSO and DE there exists a unique, optimal set of tuning parameters that maximize its performance on a given problem; (H2) when running two different BIOAs, one algorithm must demonstrate superior performance over the other. By identifying and studying an approximate Pareto-optimal set of tuning parameters for DE and PSO, H1 is shown to be strictly false for DE, and tentatively false for PSO, while H2 is shown to be false with respect to certain statistics. We further investigate the correlations between key tuning parameters and algorithm performance. Our conclusion may be implemented to further optimize and develop BIOAs applied to materials systems.