Tuning Algorithm Research Articles

Intelligent short term traffic forecasting using deep learning models with Bayesian contextual hyperband tuning

AbstractAn intelligent transport system (ITS) is fully valuable only if it can dynamically and aptly integrate all the latest cutting‐edge technologies. An ITS focuses on providing services like promptly offering real‐time road traffic information to interested parties, finding ways to reduce the average waiting time and offer secure and reliable services for commuters using past statistics. Short‐term traffic prediction is one such area in which the research community has focused in the past decade. Existing models developed for prediction has scope to improve in terms of accuracy and training time. There is a necessity to develop a best‐performing model that is computationally affluent to train with the optimal hyperparameter configuration as input which leads to improved performance. This article proposes a model that captures the traffic flow trend present in the past data to predict the flow for a future time interval. This model is an amalgamation of seasonal global trend (SGT) model and long short‐term memory (LSTM) model with attention mechanism. A novel hyperparameter tuning algorithm is also proposed which is based on multi‐armed bandit strategy with context, incorporating the right trade‐off between exploitation and exploration of the hyperparameter space using successive halving. Experimental results conducted proves that our forecast model in combination with the proposed hyperparameter tuning algorithm outperforms the existing models like SGT, LSTM models in terms of accuracy and time.

An innovation-based actuator/surface fault detection, isolation and filter tuning

PurposeThe purpose of the paper is to present an innovation-based new actuator/surface fault detection and isolation (FDI) method, which is sensitive to the changes in the innovation mean of the Kalman filter (KF) and the KF tuning method for the case of actuator/surface failure.Design/methodology/approachThe multiple system noise scale factors (MSNSFs) are used in this method as the monitoring statistics. MSNSFs are determined to make it possible to perform the actuator/surface FDI operations simultaneously.FindingsThe introduced FDI algorithm can detect and isolate the loss of effectiveness type actuator/surface faults in real time. The proposed KF tuning method works effectively against actuator/surface fault. The actuator/surface fault detection, isolation and filter tuning are achieved by just using a simple modification over the conventional KF.Originality/valueThe MSNSF-based actuator/surface fault detection, isolation and filter tuning algorithms are investigated together for the first time. The actuator/surface FDI operations are performed simultaneously.

Enhanced design of PID controller and noise filter for second order stable and unstable processes with time delay

Abstract In this work, a PID controller along with a noise filter is designed using direct synthesis method for second order stable and unstable processes with time delay. Proposed method involves a single design parameter λ whose value need to be adjusted such that a desired balance between performance and robustness of the system is achieved. Guidelines to select suitable values of λ based on the maximum sensitivity values are provided. Various computer simulations are performed in presence of measurement noise on stable, unstable, integrating and double integrating dynamic systems to illustrate the advantages of suggested control method over some contemporary techniques. Performance measures (integral absolute error and integral squared error) and robustness measures such as maximum sensitivity (M s), complimentary sensitivity (M t) are calculated for the quantitative performance evaluation of the proposed tuning strategy. The suggested tuning algorithm is simple and it yields improved closed loop response compared to some reported methods, satisfactory robustness and smooth control action in presence of noise.

Online tuning of generalized predictive controllers using fuzzy logic

The generalized predictive controller (GPC) is an efficient strategy for controlling processes with time-varying parameters, as long as the GPC tuning parameters are chosen correctly. This study aims to present a new online tuning algorithm for the parameters of the GPC. The controllers are initially tuned by a model simulation (offline), via genetic algorithm, seeking quick answers and a small error. After variations in the setpoint, injection of disturbances in the output of the plant, and variations in the gains of the system operating in closed loop, the algorithm performs an online adjustment of these parameters using Fuzzy Logic. Based on the error information between the setpoint and the controlled variable and the variation of this error, the algorithm readjusts the tuning parameters of the GPC, so the performance of the control system response is not degraded. The algorithm is validated via model simulations representing the main characteristics of industrial plants. In the simulations, tests are presented by applying disturbances in the output of the plant, changing the dynamics of the model, and changing the setpoint. It is shown that the performance indexes of each plant are presented as being at least similar to those presented in [1], because it is still widely used in recent applications, and in some cases of variation of the dynamics of the plant, the proposed algorithm remained with a satisfactory result, while the presented by [1] became unstable.

Optimasi PID Kontroller Pada Sistem Pengaturan Irigasi Menggunakan Metode Bat Algorithm (BA)

Provision of irrigation water on agricultural land aims to meet crop water needs. In its utilization, irrigation water must be used optimally. An automatic irrigation system is needed that is able to provide water for plants with the expected conditions. PID (Proportional, Integral and Derivative) is widely used for optimization of control systems. In irrigation control system and sensor connection with software program (Matlab 2013b). PID simulation was applied to the irrigation prototype used to increase efficiency and determine the amount of irrigation water to regulate the provision of irrigation water as needed. As a comparison controller, conventional PID (PID-Konv) was used, Matlab auto tuning PID (PID-Auto), and Bat Algorithm tuning PID (PID-BA). The simulation results show that PID-BA is the best method in the simulation with the smallest overshot (0.043) and the fastest settlingtime (84 seconds).

A Generalized Framework of Multifidelity Max-Value Entropy Search Through Joint Entropy.

Bayesian optimization (BO) is a popular method for expensive black-box optimization problems; however, querying the objective function at every iteration can be a bottleneck that hinders efficient search capabilities. In this regard, multifidelity Bayesian optimization (MFBO) aims to accelerate BO by incorporating lower-fidelity observations available with a lower sampling cost. In our previous work, we proposed an information-theoretic approach to MFBO, referred to as multifidelity max-value entropy search (MF-MES), which inherits practical effectiveness and computational simplicity of the well-known max-value entropy search (MES) for the single-fidelity BO. However, the applicability of MF-MES is still limited to the case that a single observation is sequentially obtained. In this letter, we generalize MF-MES so that information gain can be evaluated even when multiple observations are simultaneously obtained. This generalization enables MF-MES to address two practical problem settings: synchronous parallelization and trace-aware querying. We show that the acquisition functions for these extensions inherit the simplicity of MF-MES without introducing additional assumptions. We also provide computational techniques for entropy evaluation and posterior sampling in the acquisition functions, which can be commonly used for all variants of MF-MES. The effectiveness of MF-MES is demonstrated using benchmark functions and real-world applications such as materials science data and hyperparameter tuning of machine-learning algorithms.

On-line method for optimal tuning of PID controllers using standard OPC interface

Introduction− The controlled PID is the most widely used mathematical algorithm as a regulatory control strategy in industrial environments. The applications are varied; however, its answer depends on the proper calculation of its three parameters: the proportional, the derivative, and the integral. Analytical tuning and experimental methods solve the problem, but new tuning possibilities are now enabled within the digital and process integration context. Objective− Automatically and remotely obtain the optimal parameters of the PID controller, taking advantage of an online connection via the OPC communication protocol to analyze the transient response of the system. Methodology− The study is carried out in three main phases; it begins with a PD3 SMAR thermal process with connection via OPC; in this phase, the mathematical model of the process is built analytically based on fundamental laws. In the second phase, using an analytical tuning method, the PID control architecture is created on which the online experimentation is carried out. In the third phase, the genetic algorithms for automatic tuning are implemented, extracting performance measures from the PID controller through the transient response of the process and optimally determining the values for the proportional, derivative, and integral parameters. Results− The automatic tuning method was tested with two properly instrumented industrial processes. The potential for application can be seen due to its good result and because it does not require specific mathematical knowledge compared to conventional tuning methods. Conclusions− The automatic tuning method can be used remotely to calculate the optimal parameters of a PID controller. The parameters are calculated from the transient response and the definition of design criteria adaptable to any need for control, response, and process.

Optimized speed control with torque ripple reductions of BLDC motor based on SMC approach using LFD algorithm

Brushless DC motors (BLDCM) are utilized in various applications, including electric cars, medical and industrial equipment, where highefficiency speed control is necessary to meet load and tracking reference fluctuations. In this study, the optimal parameters of proportional-integral (PI) and sliding mode controller (SMC) for BLDC motor speed control are determined using Lévy flight distribution (LFD) technique. The integral time absolute error (ITAE) is used as a fitness function with the LFD algorithm for tuning the PI and SMC parameters. The optimization algorithms' performance is presented statistically and graphically. The simulation results show the SMC based on the LFD technique has superiority over SMC without optimization and PI controller in fast-tracking to the desired value with zero overshoot with rising time (6 ms) and low-speed ripple up to (± 9 RPM) under non-uniform conditions.

Mitigation of transformers’ loss of life in power distribution networks with high penetration of electric vehicles

This paper presents a fuzzy-logic-based framework for aging mitigation of distribution transformers. At the first stage, a model for loading estimation of a transformer installed in the head of a power distribution network was developed in the MATLAB-Simulink environment. The model considers several factors that negatively affect the transformer's operation and lead to overheating of its windings, including high ambient temperatures, harmonic voltage distortions generated by nonlinear loads, reverse power flows, overloads due to high penetration of plug-in electric vehicles. At the second stage, a tuning algorithm was developed and measures aimed at optimizing the transformer's load level and power factor were elaborated. The developed model analyzes the parameters and factors that affect the normal operation of the transformer, and warns of the presence of dangerous factors that pose a threat to reliability and can lead to malfunction. Moreover, the effectiveness of photovoltaic generation units, shunt capacitor banks and battery energy storage systems, installed at the secondary voltage side, for distribution transformers' lifetime preserving were analyzed and discussed. The proposed measures allowed to reduce the transformer's loss of life more than 7 times, compared to the initial scenario.

Development and Assessment of a Movement Disorder Simulator Based on Inertial Data.

The detection analysis of neurodegenerative diseases by means of low-cost sensors and suitable classification algorithms is a key part of the widely spreading telemedicine techniques. The choice of suitable sensors and the tuning of analysis algorithms require a large amount of data, which could be derived from a large experimental measurement campaign involving voluntary patients. This process requires a prior approval phase for the processing and the use of sensitive data in order to respect patient privacy and ethical aspects. To obtain clearance from an ethics committee, it is necessary to submit a protocol describing tests and wait for approval, which can take place after a typical period of six months. An alternative consists of structuring, implementing, validating, and adopting a software simulator at most for the initial stage of the research. To this end, the paper proposes the development, validation, and usage of a software simulator able to generate movement disorders-related data, for both healthy and pathological conditions, based on raw inertial measurement data, and give tri-axial acceleration and angular velocity as output. To present a possible operating scenario of the developed software, this work focuses on a specific case study, i.e., the Parkinson’s disease-related tremor, one of the main disorders of the homonym pathology. The full framework is reported, from raw data availability to pathological data generation, along with a common machine learning method implementation to evaluate data suitability to be distinguished and classified. Due to the development of a flexible and easy-to-use simulator, the paper also analyses and discusses the data quality, described with typical measurement features, as a metric to allow accurate classification under a low-performance sensing device. The simulator’s validation results show a correlation coefficient greater than 0.94 for angular velocity and 0.93 regarding acceleration data. Classification performance on Parkinson’s disease tremor was greater than 98% in the best test conditions.

Adaptive reachability algorithms for nonlinear systems using abstraction error analysis

In many reachability algorithms for nonlinear ordinary differential equations (ODEs), the tightness of the computed reachable sets mainly depends on abstraction errors and the choice of the set representation. One has to mitigate the resulting wrapping effects by suitable tuning of internally-used algorithm parameters since there exists no wrapping-free algorithm to date. In this work, we investigate the fundamentals governing the abstraction error in reachability algorithms – which we also refer to as set-based solvers – and its dependence on the time step size, leading to the introduction of a gain order. This order is related to measures for local and global abstraction errors and thus relates the well-known concept of convergence order from classical ODE solvers to set-based solvers. Furthermore, the simplification of the set representation is tackled by limiting the Hausdorff distance between the original and reduced sets; we demonstrate this for zonotopes. Both these theoretical advancements are incorporated in a modular adaptive parameter tuning algorithm suited for multiple classes of nonlinear ODEs whose efficiency is demonstrated on a wide range of benchmarks.

Automated insulin delivery systems: from early research to routine care of type 1 diabetes.

Automated insulin delivery (AID) systems, so-called closed-loop systems or artificial pancreas, are based upon the concept of insulin supply driven by blood glucose levels and their variations according to body glucose needs, glucose intakes and insulin action. They include a continuous glucose monitoring device which provides a signal to a control algorithm tuning insulin delivery from an infusion pump. The control algorithm is the key of the system since it commands insulin administration in order to maintain blood glucose in a predefined target range and close to a near-normal glucose level. The last two decades have shown dramatic advances toward the use in free life of AID systems for routine care of type 1 diabetes through step-by-step demonstrations of feasibility, safety and efficacy in successive hospital, transitional and outpatient trials. Because of the constraints of pharmacokinetics and dynamics of subcutaneous insulin delivery, the currently available AID systems are all 'hybrid' or 'semi-automated' insulin delivery systems with a need of meal and exercise announcements in order to anticipate rapid glucose variations through pre-meal bolus or pre-exercise reduction of infusion rate. Nevertheless, these AID systems significantly improve time spent in a near-normal range with a reduction of the risk of hypoglycemia and the mental load of managing diabetes in everyday life, representing a milestone in insulin therapy. Expected progression toward fully automated, further miniaturized and integrated, possibly implantable on long-term and more physiological closed-loop systems paves the way for a functional cure of type 1 diabetes.

An innovative algorithm for tuning of composite nonlinear feedback control in continuous‐time systems

AbstractThe transient response improvement is the most crucial feature of the composite nonlinear feedback (CNF) controllers. This study proposes a novel CNF tuning method for systems subject to saturation. To achieve the goals, the linear part is optimally obtained via a performance index and the linear matrix inequality (LMI). Then, employing the passivity property, a heuristic way is suggested to adjust the nonlinear term. Thus, the automatic and online features would be the key novelties of this algorithm compared with the offline, trial and error, or nonautomatic ones. Some systems are simulated to evaluate the technique and compare with the existing methods. The outcomes confirm that the proposed tuning yields outstanding improvement concerning the aforementioned systems.

Time series-based groundwater level forecasting using gated recurrent unit deep neural networks

ABSTRACT In this research, the mean monthly groundwater level with a range of 3.78 m in Qoşaçay plain, Iran, is forecast. Regarding three different layers of gated recurrent unit (GRU) structures and a hybrid of variational mode decomposition with gated recurrent unit (VMD-GRU), deep learning-based neural network models are developed. As the base model for performance comparison, the general single-long short-term memory-layer network model is developed. In all models, the module of sequence-to-one is used because of the lack of meteorological variables recorded in the study area. For modeling, 216 monthly datasets of the mean monthly water table depth of 33 different monitoring piezometers in the period April 2002–March 2020 are utilized. To boost the performance of the models and reduce the overfitting problem, an algorithm tuning process using different types of hyperparameter accompanied by a trial-and-error procedure is applied. Based on performance evaluation metrics, the total learnable parameters value and especially the model grading process, the new double-GRU model coupled with multiplication layer (×) (GRU2× model) is chosen as the best model. Under the optimal hyperparameters, the GRU2× model results in an R 2 of 0.86, a root mean square error (RMSE) of 0.18 m, a corrected Akaike’s information criterion (AICc) of −280.75, a running time for model training of 87 s and a total grade (TG) of 6.21 in the validation stage; and the hybrid VMD-GRU model yields an RMSE of 0.16 m, an R 2 of 0.92, an AICc of −310.52, a running time of 185 s and a TG of 3.34.

Solution path algorithm for twin multi-class support vector machine

The twin support vector machine and its extensions have made great achievements in dealing with binary classification problems. However, it suffers from difficulties in effective solution of multi-classification and fast model selection. This work devotes to the fast regularization parameter tuning algorithm for the twin multi-class support vector machine. Specifically, a novel sample data set partition strategy is first adopted, which is the basis for the model construction. Then, combining the linear equations and block matrix theory, the Lagrangian multipliers are proved to be piecewise linear w.r.t. the regularization parameters, so that the regularization parameters are continuously updated by only solving the break points. Next, Lagrangian multipliers are proved to be 1 as the regularization parameter approaches infinity, thus, a simple yet effective initialization algorithm is devised. Finally, eight kinds of events are defined to seek for the starting event for the next iteration. Extensive experimental results on nine UCI data sets show that the proposed method can achieve comparable classification performance without solving any quadratic programming problem.

TunaOil: A tuning algorithm strategy for reservoir simulation workloads

Reservoir simulations for petroleum fields and seismic imaging are known as the most demanding workloads for high-performance computing (HPC) in the oil and gas (O&G) industry. The optimization of the simulator numerical parameters plays a vital role as it could save considerable computational efforts. State-of-the-art optimization techniques are based on running numerous simulations, specific for that purpose, to find good parameter candidates. However, using such an approach is highly costly in terms of time and computing resources. This work presents TunaOil, a new methodology to enhance the search for optimal numerical parameters of reservoir flow simulations using a performance model. In the O&G industry, it is common to use ensembles of models in different workflows to reduce the uncertainty associated with forecasting O&G production. We leverage the runs of those ensembles in such workflows to extract information from each simulation and optimize the numerical parameters in their subsequent runs.To validate the methodology, we implemented it in a history matching (HM) process that uses a Kalman filter algorithm to adjust an ensemble of reservoir models to match the observed data from the real field. We mine past execution logs from many simulations with different numerical configurations and build a machine learning model based on extracted features from the data. These features include properties of the reservoir models themselves, such as the number of active cells, to statistics of the simulation’s behavior, such as the number of iterations of the linear solver. A sampling technique is used to query the oracle to find the numerical parameters that can reduce the elapsed time without significantly impacting the quality of the results. Our experiments show that the predictions can improve the overall HM workflow runtime on average by 31%.

Principled, practical, flexible, fast: a new approach to phylogenetic factor analysis.

Biological phenotypes are products of complex evolutionary processes in which selective forces influence multiple biological trait measurements in unknown ways. Phylogenetic comparative methods seek to disentangle these relationships across the evolutionary history of a group of organisms. Unfortunately, most existing methods fail to accommodate high-dimensional data with dozens or even thousands of observations per taxon. Phylogenetic factor analysis offers a solution to the challenge of dimensionality. However, scientists seeking to employ this modeling framework confront numerous modeling and implementation decisions, the details of which pose computational and replicability challenges.We develop new inference techniques that increase both the computational efficiency and modeling flexibility of phylogenetic factor analysis. To facilitate adoption of these new methods, we present a practical analysis plan that guides researchers through the web of complex modeling decisions. We codify this analysis plan in an automated pipeline that distills the potentially overwhelming array of decisions into a small handful of (typically binary) choices.We demonstrate the utility of these methods and analysis plan in four real-world problems of varying scales. Specifically, we study floral phenotype and pollination in columbines, domestication in industrial yeast, life history in mammals, and brain morphology in New World monkeys.General and impactful community employment of these methods requires a data scientific analysis plan that balances flexibility, speed and ease of use, while minimizing model and algorithm tuning. Even in the presence of non-trivial phylogenetic model constraints, we show that one may analytically address latent factor uncertainty in a way that (a) aids model flexibility, (b) accelerates computation (by as much as 500-fold) and (c) decreases required tuning. These efforts coalesce to create an accessible Bayesian approach to high-dimensional phylogenetic comparative methods on large trees.

УСВОЕНИЕ ДАННЫХ ПРИЗЕМНЫХ НАБЛЮДЕНИЙ ДЛЯ АНАЛИЗА ВЛАЖНОСТИ ПОЧВЫ МНОГОСЛОЙНОЙ МОДЕЛИ ДЕЯТЕЛЬНОГО СЛОЯ СУШИ ИВМ РАН - МГУ В СОСТАВЕ ГЛОБАЛЬНОЙ СИСТЕМЫ МОДЕЛИРОВАНИЯ АТМОСФЕРЫ ПЛАВ

The implementation of the soil moisture analysis algorithm based on screen-level observations of air temperature and relative humidity for the INM RAS-MSU multilayer active soil layer model is presented. This soil model is a part of the global atmospheric model SL-AV. Methodical experiments on tuning of the simplified extended Kalman filter algorithm for considered atmospheric modeling system are carried out. It is found that the use of soil moisture analysis as initial data allows increasing the accuracy of short- and medium-range forecasts for screen-level temperature and relative humidity during summer season at the part of Asia, the European part of Russia, and the southeastern part of North America. The recommendations obtained in this work can be used for tuning the simplified extended Kalman filter algorithm to prepare initial data for long-range weather forecasts.

Analysis of Thermal Crosstalk in Photonic Integrated Circuit Using Dynamic Compact Models

Thermal modeling of photonic integrated circuits (PICs) is required for circuit design and the development of thermal control algorithms. In this work, we present a methodology for obtaining a compact dynamic thermal <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$RC$ </tex-math></inline-formula> model of a PIC. PICs are sensitive to thermal coupling, and in this article, it is shown that traditional resistive coupling does not suffice, and a new coupling method is derived. The case study to which the new methodology is applied is a dense wavelength-division multiplexing ring filter with eight channels. With the obtained <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$RC$ </tex-math></inline-formula> model, the computational time is reduced from 5.5 h for finite element simulation to < 1 s for a single-step response. Using the compact model, circuit-level analysis of frequency response is carried out. Because of its high computational efficiency, the compact model opens possibilities for simulating thermal tuning algorithms.

Hyperparameter optimization strategies for machine learning-based stochastic energy efficient scheduling in cyber-physical production systems

One of the major challenges faced by system operators in an integrated energy system (IES) is to come up with an optimal schedule of the energy generation units in the presence of uncertainties at both the demand and supply sides. Employing efficient forecasts of the uncertain parameters while ensuring the least possible prediction error is crucial to the overall scheduling of IESs. Long Short-Term Memory (LSTM), a Recurrent Neural Network (RNN) based architecture is widely used in natural language processing and time series forecasting and has reported to be a viable and powerful forecasting technique. The aim of this work is to enhance the autonomy of energy scheduling in cyber-physical production systems (CPPS) through the development of data-driven LSTM based forecast models for the uncertain parameters. However, one of the demanding tasks in time series forecasting methods such as LSTM and the subject that received little attention is the judicious choice/fine-tuning of the hyperparameter values, which significantly impacts the performance of the developed forecasting models. This paper mainly focuses on the performance analysis of various hyperparameter tuning techniques and algorithms used by LSTM networks in forecasting uncertain parameters for stochastic energy scheduling in the context of smart manufacturing. The plant under consideration is the highly energy intensive Model Factory (MF) at the Singapore Institute of Manufacturing Technology (SIMTech), that features an actual production environment supporting both experimentation and learning of digitalisation technologies for Industry 4.0. Three main energy sources installed in an IES framework consisting of solar photo voltaic (PV) system, waste-to-energy (WTE) unit and the main utility grid serves as the energy supply module to meet the requested energy by the industrial facility. Various strategies to tune the LSTM hyperparameters in forecasting the uncertain parameters are examined. Among the four strategies, the first one deals with the traditional manual tuning approach and the second one focuses on automating the traditional manual approach using a FOR loop. While the third one focuses on a tuning strategy using Optuna with a grid search algorithm, the fourth strategy concentrates on adopting Optuna with a Bayesian optimization framework. The overall objective entails generating an optimal day-ahead forecast of the uncertain parameters (in this case solar PV irradiance, temperature, main utility grid energy price and the energy demand of the MF) using LSTM based on the best performing hyperparameter tuning techniques. These forecasts are then utilized to solve a day-ahead stochastic energy scheduling problem that generates an optimal energy dispatch schedule with the overall objectives to minimize the total operational cost and carbon emission.