As an important data preprocessing technique, feature selection aims to identify useful features and therefore reduce the dimensionality of the data. Recent studies have witnessed that evolutionary computation methods show significant potential in solving feature selection tasks. However, existing methods still encounter challenges due to the high computational costs, particularly when handling high-dimensional datasets. To tackle these issues, this work proposes a new evolutionary feature selection method with a divide-and-conquer strategy. The proposed method transforms a high-dimensional feature selection task into multiple low-dimensional sub-tasks. Multiple sub-populations corresponding to the multiple sub-tasks are evolved simultaneously. To further improve the quality of the generated candidate feature subsets, a set-based population update mechanism is introduced. Furthermore, diverse collaborative coevolution strategies are systematically explored and analyzed. The experiments conducted on 12 real-world classification datasets demonstrate that the proposed method achieves superior performance compared to 9 state-of-the-art feature selection methods. The results reveal that the proposed method is able to select smaller feature subsets while achieving higher classification accuracy across the majority of the used datasets with a reasonable low training time.

Sensor-equipped tool condition monitoring (TCM) is crucial for automated machining, but missing data poses a significant challenge. Existing methods struggle with the complex patterns and substantial data loss common in these dynamic processes. This paper introduces a novel duration-aware part attention mechanism for robust TCM. Unlike existing attention mechanisms, ours explicitly models time-duration dependencies within sensor signals, capturing multi-scale representations of tool degradation even with incomplete data. The part-attention operator, adapted from the Swin Transformer, can dynamically weight different time segments based on their duration and relevance. We further incorporate a cross-dimensional self-attention mechanism to fuse information across multiple sensors and time steps, capturing complex relationships indicative of tool wear. We evaluate our method on real-world machining datasets with varying levels of missing data, demonstrating its superior ability to accurately monitor tool condition compared to existing methods. The results show that the duration-aware part-attention effectively captures crucial temporal dependencies, leading to robust TCM even with substantial data loss.

Recent studies in reinforcement learning have explored brain-inspired function approximators and learning algorithms to simulate brain intelligence and adapt to neuromorphic hardware. Among these approaches, reward-modulated spike-timing-dependent plasticity (R-STDP) is biologically plausible and energy-efficient, but suffers from a gap between its local learning rules and the global learning objectives, which limits its performance and applicability. In this paper, we design a recurrent winner-take-all network and propose the spiking variational policy gradient (SVPG), a new R-STDP learning method derived theoretically from the global policy gradient. Specifically, the policy inference is derived from an energy-based policy function using mean-field inference, and the policy optimization is based on a last-step approximation of the global policy gradient. These fill the gap between the local learning rules and the global target. In experiments including a challenging ViZDoom vision-based navigation task and two realistic robot control tasks, SVPG successfully solves all the tasks. In addition, SVPG exhibits better inherent robustness to various kinds of input, network parameters, and environmental perturbations than compared methods.

Dynamic events, such as machine fault and rush order insertion, are fairly common in the job shop scheduling, which may lead to significant delay in order delivery and low production efficiency. Under such circumstance, it is urgent to consider more perspectives in the scheduling, such as delay time and equipment load rate. In this article, a dynamic many-objective flexible job shop scheduling problem (DMaFJSP) is founded to simultaneously optimize the completion time, delay time, total equipment load and energy consumption. Canonical many-objective optimization algorithms are seeing difficulties in maintaining population diversity and enduring poor adaptability in dynamic scheduling problems. The paper proposes a two-stage individual feedback non-dominated sorting genetic algorithm-III (TSIF-NSGA-III), where a new population diversity strategy and an individual feedback strategy are added to expand the global search faculty and stronger dynamic adaptability. Numerical study in many-objective problem and dynamic many-objective problem are conducted. The final results illustrate that the proposed algorithm can with effect dispose of the DMaFJSP.Note to Practitioners—This paper was motivated by the flexible job shop scheduling problem (FJSP) in practical dynamic situations. In the actual production procedure, however, FJSP is a more challenging issue. Not only operation sequencing and machine allocation matters, but also uncertain factors in the environment, such as machine fault, rush order insertion, etc., are important. In addition, the majority of current researchers formulate the FJSP simply focusing on maximum completion time. However, low carbon and high efficient manufacturing calls for more objectives. In this paper, two dynamic incidents, machine stoppage and rush order insertion, are considered. In addition, the model of DMaFJSP is established with many objectives such as total energy consumption, completion time, equipment load and delay time. To resolve foregoing problems, this article proposes a TSIF-NSGA-III algorithm, which adopts a diversity generation strategy and an individual feedback strategy to strengthen the search ability and dynamic adaptability of this algorithm. Preliminary simulation outcomes illuminate that this algorithm has certain advantages. In addition, the algorithm can also be applied to other multi-objective workshop scheduling problems, such as mixed flow workshop, distributed workshop, etc.

The potential intelligence behind advanced machining systems (AMSs) offers positive contributions toward process improvement. Imitation learning (IL) offers an appealing approach to accessing this intelligence by observing demonstrations from skilled technologists. However, existing IL algorithms that implement single policy strategies have yet to consider realistic scenarios for complex AMS tasks, where the available demonstrations may have come from various experts. Moreover, most IL assumes that the expert's policy is optimal, preventing the learning from fulfilling the previously ignored green missions. This article introduces a novel three-phase policy search algorithm based on IL, enabling the learning of heterogeneous expert policies while balancing energy preferences. The first phase equips the agent with machining basics through upper-level policy learning, generating an imitation policy distribution with various decision-making principles. The second phase enhances energy conservation capabilities by employing Pareto-improvement learning and fine-tuning the agent's policies to a Pareto-policy manifold. The third phase produces outcomes and amplifies the efficacy of human feedback by utilizing ensemble policies. The experimental results indicate that the proposed method outperforms meta-heuristics, exhibiting superior solution quality and faster computation times compared to four diverse baseline methods, each with diverse samples.

Incorporating sequence-to-sequence models into history-based Reinforcement Learning (RL) provides a general way to extend RL to partially-observable tasks. This method compresses history spaces according to the correlations between historical observations and the rewards. However, they do not adjust for the confounding correlations caused by data sampling and assign high beliefs to uninformative historical observations, leading to limited compression of history spaces. Counterfactual Inference (CI), which estimates causal effects by single-variable intervention, is a promising way to adjust for confounding. However, it is computationally infeasible to directly apply the single-variable intervention to a huge number of historical observations. This paper proposes to perform CI on observation sub-spaces instead of single observations and develop a coarse-to-fine CI algorithm, called Tree-based History Counterfactual Inference (T-HCI), to reduce the number of interventions exponentially. We show that T-HCI is computationally feasible in practice and brings significant sample efficiency gains in various challenging partially-observable tasks, including Maze, BabyAI, and robot manipulation tasks.

Abstract The heat exchanger has been widely used in the energy and chemical industry and plays an irreplaceable role in the featured applications. The design of heat exchanger is a mixed integer complex optimization problem, where the efficient design significantly improves the efficiency and reduces the cost. Many intelligent methods have been developed for heat exchanger optimal design. In this paper, a novel variant of sine and cosine algorithm named EDOLSCA is proposed, enhanced by dynamic opposite learning algorithm and the elite strategy. The proposed method is tested in CEC2014 benchmark and proved to be of significant advantages over the original algorithm. The new algorithm is then validated in the plate-fin heat exchanger (PFHE) optimal design problem. The comparison results of the proposed algorithm and other algorithms prove that EDOLSCA also has demonstrated superiority in heat exchanger optimal design.

Undiscounted return is an important setup in reinforcement learning (RL) and characterizes many real-world problems. However, optimizing an undiscounted return often causes training instability. The causes of this instability problem have not been analyzed in-depth by existing studies. In this article, this problem is analyzed from the perspective of value estimation. The analysis result indicates that the instability originates from transient traps that are caused by inconsistently selected actions. However, selecting one consistent action in the same state limits exploration. For balancing exploration effectiveness and training stability, a novel sampling method called last-visit sampling (LVS) is proposed to ensure that a part of actions is selected consistently in the same state. The LVS method decomposes the state-action value into two parts, i.e., the last-visit (LV) value and the revisit value. The decomposition ensures that the LV value is determined by consistently selected actions. We prove that the LVS method can eliminate transient traps while preserving optimality. Also, we empirically show that the method can stabilize the training processes of five typical tasks, including vision-based navigation and manipulation tasks.

Automated analysis of mouse behaviours is crucial for many applications in neuroscience. However, quantifying mouse behaviours from videos or images remains a challenging problem, where pose estimation plays an important role in describing mouse behaviours. Although deep learning based methods have made promising advances in human pose estimation, they cannot be directly applied to pose estimation of mice due to different physiological natures. Particularly, since mouse body is highly deformable, it is a challenge to accurately locate different keypoints on the mouse body. In this paper, we propose a novel Hourglass network based model, namely Graphical Model based Structured Context Enhancement Network (GM-SCENet) where two effective modules, i.e., Structured Context Mixer (SCM) and Cascaded Multi-level Supervision (CMLS) are subsequently implemented. SCM can adaptively learn and enhance the proposed structured context information of each mouse part by a novel graphical model that takes into account the motion difference between body parts. Then, the CMLS module is designed to jointly train the proposed SCM and the Hourglass network by generating multi-level information, increasing the robustness of the whole network. Using the multi-level prediction information from SCM and CMLS, we develop an inference method to ensure the accuracy of the localisation results. Finally, we evaluate our proposed approach against several baselines on our Parkinson’s Disease Mouse Behaviour (PDMB) and the standard DeepLabCut Mouse Pose datasets. The experimental results show that our method achieves better or competitive performance against the other state-of-the-art approaches.

This article investigates an issue of distributed fusion estimation under network-induced complexity and stochastic parameter uncertainties. First, a novel signal selection method based on event trigger is developed to handle network-induced packet dropouts, as well as packet disorders resulting from random transmission delays, where the${H_{2}}/{H_{\infty } }$performance of the system is analyzed in different noise environments. In addition, a linear delay compensation strategy is further employed for solving the complex network-induced problem, which may deteriorate system performance. Moreover, a weighted fusion scheme is used to integrate multiple resources through an error cross-covariance matrix. Several case studies validate the proposed algorithm and demonstrate satisfactory system performance in target tracking.

Diabetic retinopathy (DR) is one of the major causes of blindness. It is of great significance to apply deep-learning techniques for DR recognition. However, deep-learning algorithms often depend on large amounts of labeled data, which is expensive and time-consuming to obtain in the medical imaging area. In addition, the DR features are inconspicuous and spread out over high-resolution fundus images. Therefore, it is a big challenge to learn the distribution of such DR features. This article proposes a multichannel-based generative adversarial network (MGAN) with semisupervision to grade DR. The multichannel generative model is developed to generate a series of subfundus images corresponding to the scattering DR features. By minimizing the dependence on labeled data, the proposed semisupervised MGAN can identify the inconspicuous lesion features by using high-resolution fundus images without compression. Experimental results on the public Messidor data set show that the proposed model can grade DR effectively. Note to Practitioners-This article is motivated by the challenging problem due to the inadequacy of labeled data in medical image analysis and the dispersion of efficient features in high-resolution medical images. As for the inadequacy of labeled data in medical image analysis, the reasons mainly include the followings: 1) the high-quality annotation of medical imaging sample depends heavily on scarce medical expertise which is very expensive and 2) comparing with natural issues, it is more difficult to collect medical images because of privacy issues. It is of great significance to apply deep-learning techniques for diabetic retinopathy (DR) recognition. In this article, the multichannel generative adversarial network (GAN) with semisupervision is developed for DR-aided diagnosis. The proposed model can deal with DR classification problem with inadequacy of labeled data in the following ways: 1) the multichannel generative scheme is proposed to generate a series of subfundus images corresponding to the scattering DR features and 2) the proposed multichannel-based GAN (MGAN) model with semisupervision can make full use of both labeled data and unlabeled data. The experimental results demonstrate that the proposed model outperforms the other representative models in terms of accuracy, area under ROC curve (AUC), sensitivity, and specificity.

Whale optimization Algorithm (WOA), as a novel nature-inspired swarm optimization algorithm, has demonstrated superior performance in solving optimization problems. However, the performance deteriorates when applied to large-scale complex problems due to rapidly increasing running time required for huge computational tasks. Based on interactions within population, WOA is naturally amenable to parallelism, prompting an effective approach to mitigate the drawbacks of sequential WOA. Field Programmable Gate Array (FPGA) is an acceleration device of high parallelism and programmability. Meanwhile, Open Computing Language (OpenCL) provides a general architecture for heterogeneous development. In this paper, an efficient implementation of parallel WOA on FPGA is proposed named FPWOA. Experiment studies are conducted by performing WOA on CPU and FPWOA on FPGA respectively to solve ten well known benchmark functions. Numerical results show that our approach achieves a favourable speedup while maintaining optimization performance.

Flexible job shop scheduling problem (FJSP) has long been a complex problem due to the resource flexibility and strong constraints, generating a mixed-integer non-linear optimization problem. The problem becomes more complex with the increasing demand of energy reduction and the corresponding environmental impacts. Proper production scheduling is of significant potential in saving energy in the manufacturing system. In this paper, a multi-objective FJSP model is formulated with the objectives of minimizing the makespan and energy consumption considering strong transportation constraints. Two popular multi-objective optimization solver including Non-dominated Sorting Genetic Algorithm-II (NSGA-II) and A Multiobjective Evolutionary Algorithm Based on Decomposition (MOEA/D) are employed and compared in a real-world instance of the FJSP, associated with novel coding schemes. The results show that the proposed model is well solved by the two solvers and NSGA-II get the better solutions.

In sim-to-real transfer of Reinforcement Learning (RL) policies for robot tasks, Domain Randomization (DR) is a widely used technique for improving adaptability. However, in DR there is a conflict between adaptability and training stability, and heavy DR tends to result in instability or even failure in training. To relieve this conflict, we propose a new algorithm named Domain Decomposition (DD) that decomposes the randomized domain according to environments and trains a separate RL policy for each part. This decomposition stabilizes the training of each RL policy, and as we prove theoretically, the adaptability of the overall policy can be preserved. Our simulation results verify that DD really improves stability in training while preserving ideal adaptability. Further, we complete a complex real-world vision-based patrolling task using DD, which demonstrates DD’s practicality. A video is attached as supplementary material.

The teaching–learning-based optimization (TLBO) algorithm has been one of most popular bio-inspired meta-heuristic algorithms due to the competitive converging speed and high accuracy. A batch of TLBO variants has been proposed to enhance the exploitation ability and accelerate the exploration process. However, they still suffer from premature convergence in solving complex non-linear problems. In the study, a novel TLBO variant named dynamic-opposite learning TLBO (DOLTLBO) is proposed, which employs a new dynamic-opposite learning (DOL) strategy to overcome premature convergence. The search space of DOL has the characteristics of asymmetry and dynamically adjusting along with a random opposite number. The asymmetric search space significantly increase the probability for the population in obtaining the global optimum, which holistically improves the exploitation capability of DOLTLBO. Meanwhile, the dynamically changing characteristic enriches the diversity of the search space, thus enhancing the exploration ability. To validate the proposed DOL operator and DOLTLBO algorithm, shifted and rotated benchmark functions from CEC 2014, multiextremal functions and constrained engineering problems have been experimented upon. Comprehensive numerical results with the comparisons with the state-of-the-art counterparts show that DOLTLBO has significant advantages of converging to the global optimum on most benchmarks and engineering problems, which also validates the superiority of the novel DOL operator.

With the combination of multi-objective optimization problems and practical application, it is important to find as many Pareto optimal solutions as possible for solving multi-objective optimization problems. In many real world cases, different decision makers may tend to make choices from various aspects, which leads to the generation of multi-modal optimization problems. Such featured problems may include multiple Pareto solutions in the decision space corresponding to the same objective value in the objective space, remarkably challenging the conventional solvers. In this paper, a novel multi-objective multi-modal optimization algorithm named MO_Ring_CSO_SCD is proposed based on a recent proposed meta-heuristic method competitive swarm optimizer algorithm. It also combines the ring topology, non-dominated sort strategy and special crowding distance approach. Numerical experiments were conducted on 11 multi-modal test functions and the proposed optimization algorithm has been compared with other counterparts. The results show that the proposed algorithm has competitive performance in solving the multi-objective multi-modal problem.

Economic unit commitment is a mix-integer large scale optimization problem calling for powerful and efficient tools. On the other hand, environmental impact related to the power generation is attracting increasing attentions due to the global warming trend and urgent calls for sustainable energy development. In this paper, the dual objectives of economic and emission unit commitment is converted into a single objective problem. For solving this, a novel binary Jaya optimization is proposed and integrated with lambda iteration method. The proposed binary Jaya method is inspired by the Jaya evolution and generates binary bits from a v-shape transfer function. Numerical study demonstrates the significant improvement of the binary Jaya in regarding the convergence speed for solving unit commitment problem. The solution distributions of the both objectives also show the effective of the proposed methods.

As a novel family member of the redox flow batteries (RFBs), the single flow zinc-nickel battery (ZNB) without ion exchange membranes has attracted a lot of interests in recent years due to the high charging and discharging efficiencies. To understand the electrical behaviour is a key for proper battery management system. Unlike the electrochemical mechanism models and equivalent circuit models, the neural network based black-box model does not need knowledge about the electrochemical reactions and is a promising and adaptive approach for the ZNB battery modelling. In this paper, a compact radial basis function neural network is developed using a two-stage layer selection strategy to determine the network structure. While Jaya optimization is utilized to determine the non-linear parameters in the selected hidden nodes of the resultant RBF neural network (RBF-NN) model. The proposed method is implemented to model the ZNB to capture the non-linear electric behaviours through the readily measurable input signals. Experimental results manifest the accurate prediction capability of the resultant neural model and confirm the effectiveness of the proposed approach.

In this study, we present an effective hybrid algorithm based on particle swarm optimization (PSO) and teaching-learning-based optimization (TLBO), called PSO-TLBO, to solve multi-objective optimization (MOO) problems. In PSO-TLBO, teacher phase and learner phase in TLBO are incorporated to a PSO to improve the exploration and exploitation ability. Another feature of PSO-TLBO is the improved circular crowed sorting (CCS) to truncate the external archive which stores the nondominated solutions found during the search. Moreover, the polynomial mutation is used only when the velocities of all solutions are very small. The performance of PSO-TLBO algorithm for MOO is tested on ZDT and DTLZ benchmarks functions and compared with some powerful MOO algorithms in terms of both convergence and spread performance. Also, the contribution of CCS is compared with crowded sorting in NSGA-II. Comparison results reveal the powerful ability of PSO-TLBO to obtain remarkable solutions in terms of inverted generational distance (IGD) metrics.

Battery charging strategy is a key issue in battery management system to ensure good battery performance and safe operation during the charging process. In this paper, a novel battery optimal charging strategy is proposed by applying the TLBO algorithm to a LiFeP04 battery for an optimal charging based on a coupled thermoelectric model. A specific dual-objective function including battery charging time and temperature rise (both battery interior and surface) is formulated first. Then a battery optimal charging strategy is presented in detail by using the TLBO algorithm, aiming at finding a suitable constant-current-constant-voltage (CCCV) current profile to minimize the dual-objective function. Besides, the effects of different weights in dual-objective function on the optimal charging profile are analyzed. Simulation results demonstrate that the presented optimal charging strategy can provide effective and acceptable optimal charge current profile. The strategy can be also easily implemented to other battery types to effectively balance the battery charging time and battery temperature rise during charging process.

Unit commitment is a key issue in power system operation and has long been an intractable problem due to its complex mix-integer nonlinear formulation. The original unit commitment problem aims to minimize the fossil fuel cost by determining the on/off status of power units and power contribution of each online unit at the same time. However, the uncoordinated large charging power necessity of plug-in electric vehicles brings unprecedented challenges to the power system operators and further complicates the unit commitment problem. To seamless integrate the plug-in electric vehicles into the unit commitment, a new binary/real-value hybrid meta-heuristic algorithm framework is proposed in this paper, simultaneously determining the binary status and power output of units as well as the power delivered to/feedback from flexible charging and discharging of plug-in electric vehicles. A batch of binary particle swarm optimisation variants with different transfer functions are implemented and compared in solving the unit commitment problem with and without plug-in electric vehicles. Numerical studies illustrate the effectiveness of the proposed intelligent algorithm and the impact of different transfer functions is evaluated.

Electric vehicles provide an opportunity to reduce fossil fuel consumptions and to decrease the emissions of green-house gas and air pollutants from the transport sector. The adoption of a large number of plug-in electric vehicles however imposes significant impacts on the power system operation due to uncertain charging and discharging patterns. In this paper, multiple charging and discharging scenarios of electric vehicles together with the grid integration of renewable energy sources are examined and evaluated within the unit commitment problem. A quantum-inspired binary particle swarm optimization method is employed to determine the on/off status of each unit. Comparative studies show that the off-peak charging and peak discharging scenario is a viable option to significantly reduce the economic cost and to complement the renewable energy generation.

One of the main purposes of building a battery model is for monitoring and control during battery charging/discharging as well as for estimating key factors of batteries such as the state of charge for electric vehicles. However, the model based on the electrochemical reactions within the batteries is highly complex and difficult to compute using conventional approaches. Radial basis function (RBF) neural networks have been widely used to model complex systems for estimation and control purpose, while the optimization of both the linear and non-linear parameters in the RBF model remains a key issue. A recently proposed meta-heuristic algorithm named Teaching-Learning-Based Optimization (TLBO) is free of presetting algorithm parameters and performs well in non-linear optimization. In this paper, a novel self-learning TLBO based RBF model is proposed for modelling electric vehicle batteries using RBF neural networks. The modelling approach has been applied to two battery testing data sets and compared with some other RBF based battery models, the training and validation results confirm the efficacy of the proposed method.