Weighted Objective Function Research Articles

Model predictive control for constrained robot manipulator visual servoing tuned by reinforcement learning.

For constrained image-based visual servoing (IBVS) of robot manipulators, a model predictive control (MPC) strategy tuned by reinforcement learning (RL) is proposed in this study. First, model predictive control is used to transform the image-based visual servo task into a nonlinear optimization problem while taking system constraints into consideration. In the design of the model predictive controller, a depth-independent visual servo model is presented as the predictive model. Next, a suitable model predictive control objective function weight matrix is trained and obtained by a deep-deterministic-policy-gradient-based (DDPG) RL algorithm. Then, the proposed controller gives the sequential joint signals, so that the robot manipulator can respond to the desired state quickly. Finally, appropriate comparative simulation experiments are developed to illustrate the efficacy and stability of the suggested strategy.

Neurodynamics-driven supervised feature selection

Feature selection is an important dimensionality reduction technique in machine learning, pattern recognition, image processing, and data mining. Most existing feature selection methods are greedy in nature thus are prone to sub-optimality. Though some feature selection methods based on global optimization of unsupervised redundancy may potentiate performance improvements, they may or may not be relevant to classification as the information on pairwise features with class labels is missing. In this paper, based on a supervised similarity measure, a biconvex optimization problem is formulated for holistic feature section with a quadratically weighted objective function subject to linear equality and nonnegativity constraints. In addition, an iteratively reweighted convex quadratic program is reformulated. A two-timescale duplex neurodynamic system is applied to solve the formulated biconvex optimization problem and a projection neural network is customized to solve the iteratively reweighted convex optimization problem. Experimental results of the proposed neurodynamics-based supervised feature selection are elaborated in comparison with several existing feature selection methods based on twenty benchmark datasets to substantiate the efficacy and superiority of the neurodynamics-based method for selecting informative features in classification.

Optimization of Improved PID Control Strategy Based on Genetic Algorithm

Conventional PID control, although has the advantage that the robustness and adaptability, has certain limitations, but for nonlinear systems is difficult to achieve the desired effect. When the environment changes, control parameters cannot be adjusted in time. Aiming at these problems, this paper presents an improved PID control strategy based on genetic algorithm. Taking single-phase LCL as the research object, by analyzing the operation process of single-phase LCL grid-connected inverter and optimizing genetic algorithm, the genetic iteration process is optimized through adaptive adjustment, and the weight of objective function is adjusted, so that the system can respond quickly, better monitoring of grid-connected current improves the robustness of the system. So as to better ensure the stability of grid connection. Finally, the MATLAB model is built for simulation. Simulation endings indicate that the boosted control strategy has great stability and can effectively decrease the grid current harmonic content. The improved algorithm is global and has relatively high error tolerance, which can meet the needs of grid connection.

Multi-response optimization of process parameters in nitrogen-containing gray cast iron milling process based on application of non-dominated ranking genetic algorithm

As a new, high-strength and clean cast iron material, nitrogen-containing gray cast iron has excellent properties and a wide range of application prospects. However, the excellent material properties of the material not only make the machinability challenging, but also the high efficiency and quality of the machining process is a pressing issue. Therefore, it is necessary to study the machining characteristics of nitrogen-containing gray cast iron to obtain the optimal machining parameters to enrich the research work on nitrogen-containing gray cast iron. In this paper, the machining characteristics of nitrogen-containing gray cast iron are systematically studied, and the effects of cutting parameters on milling force, milling temperature, and surface roughness are analyzed. And, based on the machinability assessment, the objective function weights under different production requirements are determined by using hierarchical analysis trade-offs, and an integrated optimization model based on non-dominated ranking genetic algorithm and hierarchical analysis (AHP) is proposed. The model outputs the optimal combination of milling parameters by inputting the cutting speed (vc), feed rate per tooth (fz) and cutting depth (ap), surface roughness and cutting efficiency as the objective functions. The experimental results show that cutting depth has the greatest effect on the cutting force and cutting speed has the greatest effect on the cutting temperature and the surface roughness. The passivation effect of nitrogen on the graphite tip resulted in an increase in both cutting force and cutting temperature. The parameter optimization results indicated that the optimized roughing parameters significantly improve the surface quality while machining efficiently; the optimized finishing parameters improve Ra by 23.53% while ensuring higher MRR, which can achieve efficient and high-quality machining under different production requirements and provide an experimental basis for practical engineering applications of nitrogen-containing gray cast iron.

Implications of Including Tumor Presence Probabilities as Weights in Radiotherapy Optimization

<h3>Purpose/Objective(s)</h3> Recently, the fixed nature of the clinical target volume has been challenged in several studies that choose to incorporate microscopic tumor infiltration as a source of uncertainty in radiation therapy treatment planning. The approach to handle this uncertainty typically involves voxel-wise objective function weights that are proportional to each voxel's marginal probability of tumor presence. The main argument for doing so is to balance the probability of target coverage against the increased dose to organs at risk (OARs). We study the implications of basing the optimization on such probability weights, given an isotropic tumor infiltration model used for both optimization and evaluation, in which the infiltration distance is taken from a truncated gaussian probability distribution. We hypothesize that the low voxel weights far from the gross tumor volume (GTV) will cause the dose to fall off unevenly between directions that have high and low conflict with OARs, leading to a suboptimal trade-off between the probability of target coverage and OAR sparing. <h3>Materials/Methods</h3> We implement the probability weight method on a phantom geometry where suspected target voxels are weighted with probabilities from the tumor infiltration model, decreasing away from the GTV. This probability weighted plan is then compared against plans that are optimized using target volumes from an enumeration over a discrete set of scenarios from the tumor infiltration model. For both methods, plans are optimized using a range of scaling factors which scale the (uniform dose) objectives on the target in relation to those on OARs (max EUD), allowing comparison of the approximate Pareto fronts of the plans in the enumeration scheme to that of the probability weighted plan. The plans are evaluated on target samples from the tumor infiltration model which are used to compute target dose statistics. For each plan, the probability of satisfying a target coverage criterion (D98 > 0.95*D_pres.) is then plotted against the OAR objective to evaluate whether the probability weight method can produce Pareto optimal plans with respect to these criteria. In evaluation, we also include a tumor control probability (TCP) model to investigate how the methods compare on a smooth target coverage criterion. <h3>Results</h3> For a given scaling factor, the probability weighted plan is usually close to the corresponding set of plans from the enumeration scheme. However, the non-dominated subset of all generated plans consists almost entirely of plans from the enumeration scheme. Although the result holds also for TCP, there is less difference between the non-dominated subset of plans and the set of probability weighted plans. <h3>Conclusion</h3> The results indicate that finding the right combination of scaling factor and target expansion distance gives a better trade-off than the probability weight method, although efficiently doing so remains a future challenge. Meanwhile, using probability weights may serve as a practical alternative.

Multi-Task Transformer with Adaptive Cross-Entropy Loss for Multi-Dialect Speech Recognition

At present, most multi-dialect speech recognition models are based on a hard-parameter-sharing multi-task structure, which makes it difficult to reveal how one task contributes to others. In addition, in order to balance multi-task learning, the weights of the multi-task objective function need to be manually adjusted. This makes multi-task learning very difficult and costly because it requires constantly trying various combinations of weights to determine the optimal task weights. In this paper, we propose a multi-dialect acoustic model that combines soft-parameter-sharing multi-task learning with Transformer, and introduce several auxiliary cross-attentions to enable the auxiliary task (dialect ID recognition) to provide dialect information for the multi-dialect speech recognition task. Furthermore, we use the adaptive cross-entropy loss function as the multi-task objective function, which automatically balances the learning of the multi-task model according to the loss proportion of each task during the training process. Therefore, the optimal weight combination can be found without any manual intervention. Finally, for the two tasks of multi-dialect (including low-resource dialect) speech recognition and dialect ID recognition, the experimental results show that, compared with single-dialect Transformer, single-task multi-dialect Transformer, and multi-task Transformer with hard parameter sharing, our method significantly reduces the average syllable error rate of Tibetan multi-dialect speech recognition and the character error rate of Chinese multi-dialect speech recognition.

Development of Input Training Neural Networks for Multiple Sensor Fault Isolation

This paper considers the problem of inhibiting smearing effects for multiple sensor fault isolation. Although the reconstruction-based approach has received considerable attention and achieved excellent results, the investigation into nonlinear systems has been relatively limited. A new reconstruction-based input training neural network aided by the bound and branch method (BAB-RBITNN) is proposed for the nonlinear systems so as to achieve the desired performance of multiple sensor fault isolation. In BAB-RBITNN, a hypothetical faulty direction is naturally integrated into the objective testing function as residual weights for reducing the negative impacts of inaccurate sensor data, taking advantage of the non-static ITNN testing process. The weighted objective function is then considered as the fitness function to obtain the minimum predicted residual by the covariance matrix adaptation evolutionary strategy algorithm. The corresponding optimized fitness is regarded as the reconstruction-based index, which is the critical indicator of fault isolation in the proposed method. Besides, the BAB method is implemented to handle the combinatorial optimization problem instead of a brute search for intelligently pinpointing the real faulty sensors. The efficiency of the proposed method is shown through a validation example and two industrial examples. Comparisons with other methods, including reconstruction-based principal component analysis and reconstruction-based auto-associative neural network techniques, are also presented.

Clustering-based iterative heuristic framework for a non-emergency patients transportation problem

IntroductionNon-emergency patient transportation (NEPT) services are particularly important nowadays due to the aging population and contagious disease outbreaks (e.g., Covid-19 and SARS). In this work, we study a NEPT problem with a case study of patient transportation services in Hong Kong. The purpose of this work is to study the discomfort and inconvenience measures (e.g., waiting time and extra ride time) associated with the transportation of non-emergency patients while optimizing the operational costs and utilization of NEPT ambulances. MethodsA mixed-integer linear programming (MILP) formulation is developed to model the NEPT problem. This MILP model contributes to the existing literature by not only including the patient inconvenience measures in the objective function but also illustrating a better trade-off among different performance measures through its specially customized formulation and real-life characteristics. CPLEX is used to find the optimal solutions for the test instances. To overcome the computational complexity of the problem, a clustering-based iterative heuristic framework is designed to solve problems of practical sizes. The proposed framework distinctively exploits the problem-specific structure of the considered NEPT problem in a novel way to enhance and improve the clustering mechanism by repeatedly updating cluster centers. ResultsThe computational experiments on 19 realistic problem instances show the effective execution of the solution method and demonstrate the applicability of our approach. Our heuristic framework observes an optimality gap of less than 5% for all those instances where CPLEX delivered the result. The weighted objective function of the proposed model supports the analysis of different performance measures by setting different preferences for these measures. An extensive sensitivity analysis performed to observe the behavior of the MILP model shows that when operating costs are given a weightage of 0.05 in the objective function, the penalty value for user inconvenience measures is the lowest; when the weightage value for operating costs varies between 0.8 and 1.0, the penalty value for the same measures is the highest. ConclusionsThis research can assist decision-makers in improving service quality by balancing operational costs and patient discomfort during transportation.

Robust Variable Selection for Single-Index Varying-Coefficient Model with Missing Data in Covariates

As applied sciences grow by leaps and bounds, semiparametric regression analyses have broad applications in various fields, such as engineering, finance, medicine, and public health. Single-index varying-coefficient model is a common class of semiparametric models due to its flexibility and ease of interpretation. The standard single-index varying-coefficient regression models consist mainly of parametric regression and semiparametric regression, which assume that all covariates can be observed. The assumptions are relaxed by taking the models with missing covariates into consideration. To eliminate the possibility of bias due to missing data, we propose a probability weighted objective function. In this paper, we investigate the robust variable selection for a single-index varying-coefficient model with missing covariates. Using parametric and nonparametric estimates of the likelihood of observations with fully observed covariates, we examine the estimators for estimating the likelihood of observations. For variable selection, we use a weighted objective function penalized by a non-convex SCAD. Theoretical challenges include the treatment of missing data and a single-index varying-coefficient model that uses both the non-smooth loss function and the non-convex penalty function. We provide Monte Carlo simulations to evaluate the performance of our approach.

A bi-objective [formula omitted]-nearest-neighbors-based imputation method for multilevel data

We propose a bi-objective algorithm based on the k-nearest neighbors (biokNN) method to perform imputation of missing values for data with multilevel structures with continuous variables. We define the imputation method as a bi-objective minimization problem and propose a solution algorithm based on a weighted objective function. The algorithm seeks imputed values that balance the dissimilarity between the k-nearest neighbors and the observations within the same cluster. The effectiveness of the proposed method is evaluated through a simulation study, and its results are compared with those of eight benchmark imputation methods. The simulation study is based on the generation of datasets with a varying-intercept–varying-slope multilevel model, and the results are compared both by using well-known accuracy metrics and by estimating the bias of the estimates after inference has been performed. Based on the simulation, the effects of different configurations of multilevel datasets are tested, including the number of clusters, their size, their similarity, the percentage of missing values, and the effect of imbalanced clusters. The results show that the proposed method outperforms the benchmark methods, especially in cases with high intraclass correlation. A comparison of fitted linear multilevel regression models shows that our method can also reduce the bias of the estimates and the coefficient of determination. Finally, the method is tested on three commonly used machine learning datasets and shows better accuracy in most cases compared with the benchmark methods.

Online on-Road Motion Planning Based on Hybrid Potential Field Model for Car-Like Robot.

The application of Middle-sized Car-like Robots (MCRs) in indoor and outdoor road scenarios is becoming broader and broader. To achieve the goal of stable and efficient movement of the MCRs on the road, a motion planning algorithm based on the Hybrid Potential Field Model (HPFM) is proposed in this paper. Firstly, the artificial potential field model improved with the eye model is used to generate a safe and smooth initial path that meets the road constraints. Then, the path constraints such as curvatures and obstacle avoidance are converted into an unconstrained weighted objective function. The efficient least-squares & quasi-Newton fusion algorithm is used to optimize the initial path to obtain a smooth path curve suitable for the MCR. Finally, the speed constraints are converted into a weighted objective function based on the path curve to get the best speed profile. Numerical simulation and practical prototype experiments are carried out on different road scenes to verify the performance of the proposed algorithm. The results show that re-planned trajectories can satisfy the path constraints and speed constraints. The real-time re-planning period is 184 ms, which demonstrates the proposed approach’s effectiveness and feasibility.

Improvement of Airport Surface Operation at Tokyo International Airport Using Optimization Approach

Congestion and delays occur on airport surfaces as a result of a rapid increase in the demand for air transport. The aim of this study is to determine the differences between optimized and observed operations to improve airport surface operation at Tokyo International Airport by using mixed-integer linear programming to minimize the total ground movement distance and time based on real-time flight information. Receding horizon schemes are considered to adapt to dynamic environments. The model obtains results that reduce the taxi distance by 18.54% and taxi time by 29.77% compared with the observed data. A comparison of taxiway usage patterns between the optimization results and observed data provides insight into the optimization process, for example, changes in runway cross strategies and taxiway direction rules. Factors such as the objective function weights and airline–terminal relationship were found to significantly affect the optimization result. This study suggests improvements that can be made at airports to achieve a more efficient surface operation.

Multi-objective Optimal Operation of Combined Cascade Reservoir and Hydrogen System

As a kind of zero-carbon and high-efficiency energy, hydrogen has attracted more attention with clean energy technology development. However, large-scale green hydrogen application is difficult to promote because of its high production cost. China is rich in water resources, but the utilization rate is low, which causes serious water abandonment problems. This article studies the optimal operation of combined cascade reservoir and hydrogen system to improve the utilization of water resources and the system's economic benefits. In this article, a multi-objective nonlinear model is established. An expanded progressive optimization algorithm is adopted to solve the problem more efficiently and accurately. The weights of objective functions are discussed with data from the Hongru River Basin in China. The rationality and effectiveness of combining cascade reservoir and hydrogen system are verified through the optimal operation of water resources and the optimal allocation of power and hydrogen. In addition, the overall economic benefits and reservoir operation efficiency are improved.

Optimization of the Wake Oscillator for Transversal VIV

Vibrations of slender structures associated with the external flow present a design challenge for the energy production systems placed in the marine environment. The current study explores the accuracy of the semi-empirical wake oscillator models for vortex-induced vibrations (VIV) based on the optimization of (a) the damping term and (b) empirical coefficients in the fluid equation. This work investigates the effect of ten fluid damping variations, from the classic van der Pol to more sophisticated fifth-order terms, and prediction of the simplified case of the VIV of transversally oscillating rigid structures provides an opportunity for an extended, comprehensive comparison of the performance of tuned models. A constrained nonlinear minimization algorithm in MATLAB is applied to calibrate considered models using the published experimental data, and the weighted objective function is formulated for three different mass ratios. Comparison with several sources of published experimental data for cross-flow oscillations confirms the model accuracy in the mass ratio range. The study indicates the advantageous performance of the models tuned with the medium mass ratio data and highlights some advantages of the Krenk–Nielsen wake oscillator.

Tuning of Model Predictive Controllers Based on Hybrid Optimization

A tuning procedure for a model predictive controller (MPC) is presented for multi-input multi-output systems. The approach consists of two steps based on a hybrid method: the goal attainment method and a variable neighborhood search. In the first step, the weights of the MPC objective function are obtained, minimizing the square error between the closed-loop response of the internal controller model and a predefined desired reference trajectory. In the second step, the integer variables of the problem (prediction and control horizons) are obtained, minimizing the square error between the closed-loop response and an optimal trajectory, aiming a controller with low computational cost and good performance. The proposed method was tested in two benchmark processes using different MPC formulations, showing satisfactory results.

Mixed-integer linear programming, constraint programming and column generation approaches for operating room planning under block strategy

The planning of operating rooms under block strategy is addressed in this study. The decisions are about opening the operating rooms and assigning specialties and surgeons to blocks at the tactical level, and sequencing the surgeries at the operational level. This problem aims to minimize the costs of opening the operating rooms and their overtime, the waiting costs of patients, and the surgeons' idle costs. We propose two mixed-integer linear programming models, a constraint programming (CP) model and a constraint programming-based column generation (CPCG) method for handling the problem. The performance of the models is evaluated by random test instances. The results indicate that CP and CPCG models are more efficient than the linear programming models in terms of computational time, and the number of variables and constraints. The proposed method CPCG generates optimal solutions for problem instances of up to 30 surgeries in less than 4 min. The CP model finds the optimal solutions in about one minute but proving the optimality of the found solutions is time-consuming in some instances. The maximum optimality gap for the proposed two-step linear programming model is 2%, while its run time is less than 20 s. A sensitivity analysis is done on the main parameters of the problem like objectives' weights, opening cost of ORs, unit waiting cost of patients, and the maximum available time in surgery blocks. Among the three objectives, the unit waiting cost of patients has the most sensitivity to variations of the objective function weights.

Business Failure Prediction Based on a Cost-Sensitive Extreme Gradient Boosting Machine

Business failure prediction is very important for the sustainable development of enterprises. Machine learning algorithms, especially ensemble algorithms, have shown great economic benefits in enterprise financial early warning. However, the highly imbalanced class distribution of financial risk data and the inexplainable of most machine learning-based early distress warning models limit their commercial application. To address the above limitations, we enhance the business failure prediction performance by tree-ensemble in a boosting manner. Moreover, to solve the class imbalanced issue in business failure datasets, a weighted objective function, weighted cross-entropy, is embedded into the boosted tree framework, making the weighted XGBoost a cost-sensitive business failure prediction model. Besides, to tackle the second issue, we explore the intrinsic interpretability of the proposed method by visualizing the feature importance and incorporating a partial dependence plot technique to locally interpret the individual business failure event. Experimental results on business failure datasets with different predictive horizons collected from China Security Market Accounting Research (CSMAR) database show the proposed weighted XGBoost is a good solution to reduce the error on recognizing firms in business failure. Furthermore, the visualized feature importance score and partial dependence plot result both demonstrate that the cost-sensitive tree-based ensemble can be a good tool to guide the investors in making rational as well as provide interpretable business prediction results as a reference for the policy-making of the regulators.

Coupled optimization of carbon dioxide sequestration and CO2 enhanced oil recovery

CO2 injection into hydrocarbon reservoirs can simultaneously pave the path for two important processes, namely oil production increase, and CO2 storage. This combined optimization problem can be referred to as carbon capture and storage - enhanced oil recovery (CCS-EOR) problem. CCS-EOR is a challenging task with many opportunities which should be studied in a timely manner. Moreover, the attitude of the management team towards prioritizing CCS or EOR should be considered. Therefore, in the current study, CCS-EOR problem was addressed by utilizing the dynamic well flow settings (bottom-hole pressure and/or flow rate) as the optimization variables with a weighted objective function. This approach was tested on a case study. Its huge simulation time due to high dimensionality remained a serious challenge. One effective approach was to use the underlying applied knowledge of the system and extract that wisdom to reduce the simulation time in a rational manner. CO2 breakthrough time was such a knowledge. Therefore, the importance of CO2 breakthrough time was stressed, and the understanding of the response of the simulator to this event was used to considerably reduce the simulation time in a black box approach. Moreover, a concise sensitivity analysis was performed which helped to schedule the workflow based on the given priority to each process. The optimized case could improve the objective value and gas storage efficiency up to 2.86 % and 12.1 % respectively. The obtained results can be used to analyze a variety of operational matters such as the CO2 distribution map in the reservoir, and the well settings.

Optimal Designing of Fuzzy-PID Controller in the Load-Frequency Control Loop of Hydro-Thermal Power System Connected to Wind Farm by HVDC Lines

In this paper an optimized Fuzzy based controller is proposed for automatic generation control of two area hydro-thermal power system connected to the wind farm. The parameters and membership functions of the proposed controller are optimized by a modified version of Cuckoo search algorithm (CSA). Also, a weighted objective function is proposed to minimize the frequency deviation and transmission power oscillation. The suggested heuristic objective function is a weight function from maximum frequency drift and oscillations fading time. To assessing the performance of suggested controller, studies is accomplished by two different scenarios. In the first scenario, the simulations are performed without wind farm and in the second scenario, the simulation is done in the presence of a wind farm. The simulation results indicate that the wind farm presence has major effect on sustained improvement of power system and the reason is considering the load variations in the area, the demand electrical energy in the same area is provided by the wind farm and hence, the frequency oscillations are decreased in both areas.

A simple method to find optimum efficient basic solutions to bi-objective transportation problems

This paper proposes a simple method to find an optimum and efficient basic solution to bi-objective transportation problem (BOTP) using weighted goal programming approach. Preferential weights for the goals (objectives) are derived from analytic hierarchy process (AHP). Solution obtained by the proposed approach has been found to be the nearest basic feasible solution to the ideal solution in terms of Euclidean distance measure. The method is illustrated with numerical examples taken from the literature and solutions compared in terms of the number of iterations, computational complexity and proximity to the ideal solution. The modified minimum supply and demand method is used to obtain initial basic feasible solutions that are found to be optimal, and hence this method is computationally simpler compared to other linear programming methods. It has been observed that weighted objective function optimization yields optimal efficient BOTP solutions using appropriate weights for the objective functions.