Applied Soft Computing最新文献

{"title":"Online Q-learning for stochastic linear systems with state and control dependent noise","authors":"Hongxu Zhu ,&nbsp;Wei Wang ,&nbsp;Xiaoliang Wang ,&nbsp;Shufan Wu ,&nbsp;Ran Sun","doi":"10.1016/j.asoc.2024.112417","DOIUrl":"10.1016/j.asoc.2024.112417","url":null,"abstract":"<div><div>For continuous-time (CT) systems that are characterized by stochastic differential equations (SDEs) with completely unknown dynamics parameters, a reinforcement learning (RL)-based optimal control framework is presented in this paper. To obtain the near-optimal control policy, an online Q-learning algorithm is proposed by learning the data sampled from the system state trajectory, while an integral reinforcement learning (IRL) approach is developed in stochastic situation so as to formulate the Q-learning iterative algorithm. In particular, an actor/critic neural network (NN) structure is applied in iteration, where a critic approximator is used for estimating the designed Q-function while an actor approximator is for estimating the optimal control policy online. To ensure the convergence of iteration, the tuning laws of two neural networks are designed, respectively, by a gradient descent scheme. Moreover, the mean-square stability of the closed-loop system is proved through rigorous analysis, and the convergence to optimal solution is guaranteed as well.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112417"},"PeriodicalIF":7.2,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A strengthened constrained-dominance based evolutionary algorithm for constrained many-objective optimization","authors":"Wei Zhang ,&nbsp;Jianchang Liu ,&nbsp;Junhua Liu ,&nbsp;Yuanchao Liu ,&nbsp;Shubin Tan","doi":"10.1016/j.asoc.2024.112428","DOIUrl":"10.1016/j.asoc.2024.112428","url":null,"abstract":"<div><div>Solving constrained multi-objective optimization problems have received increasing attention. However, there are few researches based on constrained many-objective optimization problems that widely exist in real life. Given the above fact, we propose a strengthened constrained-dominance based evolutionary algorithm for constrained many-objective optimization (SCEA). The proposed SCEA includes the following main components. First, a dual-assistance mating selection is developed to select elite parents for variation, and further accelerate the generation of feasible solutions. Second, a strengthened constrained-dominance relation is proposed, which favors feasible solutions but still leaves the room for selecting infeasible solutions. This is achieved by simultaneously considering the objective optimization and constraint satisfaction. Third, the designed unconstrained aggregation (UA) indicator and crowded detector cooperate reference points to promote the convergence and diversity of population. Finally, a cooperation mechanism based on the constrained aggregation (CA) indicator and hierarchical clustering is designed to drive individuals toward different feasible regions, and further balance the objective optimization and constraint satisfaction. Extensive experimental studies are conducted on three benchmark test suites and two real-world applications to validate the performance of SCEA. The corresponding experiment results have demonstrated that SCEA is more competitive than its peer competitors.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112428"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654124","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-objective optimization and multi-attribute decision-making support for optimal operation of multi stakeholder integrated energy systems","authors":"J.H. Zheng ,&nbsp;L.X. Zhai ,&nbsp;Fang Li ,&nbsp;Dandan Wang ,&nbsp;Yalou Li ,&nbsp;Zhigang Li ,&nbsp;Q.H. Wu","doi":"10.1016/j.asoc.2024.112426","DOIUrl":"10.1016/j.asoc.2024.112426","url":null,"abstract":"<div><div>To efficiently tackle the optimal operation problem of multi-stakeholder integrated energy systems (IESs), this paper develops a multi-objective optimization and multi-attribute decision-making support method. Mathematically, The optimal operation of IESs interconnected with distributed district heating and cooling units (DHCs) via the power grid and gas network, can be formulated as a multi-objective optimization problem considering both economic, reliability and environment-friendly objectives with numerous constraints of each energy stakeholder. Firstly, a multi-objective group search optimizer with probabilistic operator and chaotic local search (MPGSO) is proposed to balance global and local optimality during the random search iteration. The MPGSO utilizes a crowding probabilistic operator to select producers to explore areas with higher potential but less crowding and reduce the number of fitness function calculations. Moreover, a new parameter selection strategy based on chaotic sequences with limited computational complexity is adopted to escape the local optimal solutions. Consequently, a set of superior Pareto-optimal fronts could be obtained by the MPGSO. Subsequently, a multi-attribute decision-making support method based on the interval evidential reasoning (IER) approach is used to determine a final optimal solution from the Pareto-optimal solutions, taking multiple attributes of each stakeholder into consideration. To verify the effectiveness of the MPGSO, the DTLZ suite of benchmark problems are tested compared with the original GSOMP, NSGA-II and SPEA2. Additionally, simulation studies are conducted on a modified IEEE 30-bus system connected with distributed DHCs and a 15-node gas network to verify the proposed approach. The quality of the obtained Pareto-optimal solutions is assessed using a set of criteria, including hypervolume (HV), generational distance (GD), and Spacing index, among others. Simulation results show that the number of Pareto-optimal solutions (NPS) of MPGSO are higher by about 32.6 %-62.1%, computation time (CT) are lower by about 2.94 %-46.1 % compared with other algorithms. Besides, to further evaluate the performance of the proposed approach in addressing larger-scale issues, the study employs the modified IEEE 118-bus system of greater magnitude. The proposed MPGSO algorithm effectively handles multi-objective and non-convex optimization problems with Pareto sets in terms of better convergence and distributivity.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112426"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142656803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radial basis function neural network based data-driven iterative learning consensus tracking for unknown multi-agent systems","authors":"Kechao Xu,&nbsp;Bo Meng,&nbsp;Zhen Wang","doi":"10.1016/j.asoc.2024.112425","DOIUrl":"10.1016/j.asoc.2024.112425","url":null,"abstract":"<div><div>This paper provides a novel data-driven-distributed-consensus control protocol for unknown nonlinear non-affine discrete-time multi-agent systems (MAS) with repetitive properties. The leader’s commands are directed to the followers in the topological graph. The dynamic linearization technology (DLT) is used to build the distributed iterative learning (IL) controller along the iteration axis. In the iterative process, the control gain is automatically adjusted by updating the weight matrix of the high-order radial basis function neural network (RBFNN, HORBFNN). In global control, the higher order parameter (HOP) Newton method is used to achieve global convergence and stability of the control process. All the above processes do not require the understanding of dynamical equations or physical models for each agent, and only use local communication information of multi-agent to achieve consistent tracking of MAS leaders and followers. Based on the strong connection, the convergence performance, stability and boundedness properties of the proposed control protocol in the fixed topology as well as in the iterative topology are validated by a rigorous theoretical analysis. Simulation experiments are conducted to verify the effectiveness of the control protocol.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112425"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142656731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CSKINet: A multimodal network model integrating conceptual semantic knowledge injection for relation extraction of Chinese corporate reports","authors":"Shun Luo ,&nbsp;Juan Yu ,&nbsp;Yunjiang Xi","doi":"10.1016/j.asoc.2024.112401","DOIUrl":"10.1016/j.asoc.2024.112401","url":null,"abstract":"<div><div>Recognizing the associations among entities in corporate reports accurately is crucial for market regulation and policy development. Nevertheless, confronted with massive corporate information, the traditional manual screening approach is cumbersome, struggling to match the demand. Consequently, we propose a multimodal network model incorporating conceptual semantic knowledge injection, CSKINet, for accurately extracting relations from Chinese corporate reports. The essential highlights in the design of the CSKINet model are the following: (1) Integrate the conceptual descriptions of corporations from external resources to construct the semantic knowledge repository of corporate concepts, which provides a solid semantic foundation for the model. (2) Multimodal features are extracted from the documents by various means and corporate conceptual knowledge is integrated into the model representation to enhance the representation capability of the model. (3) The multimodal self-attention mechanism that captures cross-modal associations and the biaffine classifier with Taylor polynomial loss function that optimizes training iterations further improve the learning efficiency and prediction accuracy. The results on the real corporate report dataset show that our proposed model can more accurately extract the relations from Chinese corporate reports compared to other baseline models, where the F1 score reaches 85.76%.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112401"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A dynamical teaching-learning-based optimization algorithm for fuzzy energy-efficient parallel batch processing machines scheduling in fabric dyeing process","authors":"Jing Wang ,&nbsp;Debiao Li ,&nbsp;Hongtao Tang ,&nbsp;Xixing Li ,&nbsp;Deming Lei","doi":"10.1016/j.asoc.2024.112413","DOIUrl":"10.1016/j.asoc.2024.112413","url":null,"abstract":"<div><div>Fabric dyeing is the most time-consuming and energy-intensive process in textile production with some batch processing machines (BPMs) and uncertainty. In this study, a fuzzy energy-efficient parallel BPMs scheduling problem (FEPBSP) with machine eligibility and sequence-dependent setup time (SDST) in fabric dyeing process is investigated, and a dynamical teaching-learning-based optimization algorithm (DTLBO) is proposed to simultaneously optimize the total agreement index, fuzzy makespan, and total fuzzy energy consumption. In DTLBO, multiple classes are constructed by non-dominated sorting. Dynamical class evolution is designed, which incorporates diversified search among students and adaptive self-learning of teachers. The former is implemented using various combinations of the teacher phase and the learner phase, and the latter is achieved through teacher quality and an adaptive threshold. Additionally, a reinforcement local search based on neighborhood structure dynamic selection is also applied. Extensive experiments are conducted, and the computational results demonstrated that the new strategies of DTLBO are effective, and it is highly competitive in solving the considered problem.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112413"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive fuzzy neighborhood decision tree","authors":"Xinyu Cui ,&nbsp;Changzhong Wang ,&nbsp;Shuang An ,&nbsp;Yuhua Qian","doi":"10.1016/j.asoc.2024.112435","DOIUrl":"10.1016/j.asoc.2024.112435","url":null,"abstract":"<div><div>Decision tree algorithms have gained widespread acceptance in machine learning, with the central challenge lying in devising an optimal splitting strategy for node sample subspaces. In the context of continuous data, conventional approaches typically involve fuzzifying data or adopting a dichotomous scheme akin to the CART tree. Nevertheless, fuzzifying continuous features often entails information loss, whereas the dichotomous approach can generate an excessive number of classification rules, potentially leading to overfitting. To address these limitations, this study introduces an adaptive growth decision tree framework, termed the fuzzy neighborhood decision tree (FNDT). Initially, we establish a fuzzy neighborhood decision model by leveraging the concept of fuzzy inclusion degree. Furthermore, we delve into the topological structure of misclassified samples under the proposed decision model, providing a theoretical foundation for the construction of FNDT. Subsequently, we utilize conditional information entropy to sift through original features, prioritizing those that offer the maximum information gain for decision tree nodes. By leveraging the conditional decision partitions derived from the fuzzy neighborhood decision model, we achieve an adaptive splitting method for optimal features, culminating in an adaptive growth decision tree algorithm that relies solely on the inherent structure of real-valued data. Experimental evaluations reveal that, compared with advanced decision tree algorithms, FNDT exhibits a simple tree structure, stronger generalization capabilities, and superior performance in classifying continuous data.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112435"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142656730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CoaddNet: Enhancing signal-to-noise ratio in single-shot images using convolutional neural networks with coadded image effect","authors":"Zhi-Ren Pan ,&nbsp;Bo Qiu ,&nbsp;A-Li Luo ,&nbsp;Qi Li ,&nbsp;Zhi-Jun Liu ,&nbsp;Fu-Ji Ren","doi":"10.1016/j.asoc.2024.112395","DOIUrl":"10.1016/j.asoc.2024.112395","url":null,"abstract":"<div><div>Noise in astronomical images significantly impacts observations and analyses. Traditional denoising methods, such as increasing exposure time and image stacking, are limited when dealing with single-shot images or studying rapidly changing astronomical objects. To address this, we developed a novel deep-learning denoising model, CoaddNet, designed to improve the image quality of single-shot images and enhance the detection of faint sources. To train and validate the model, we constructed a dataset containing high and low signal-to-noise ratio (SNR) images, comprising coadded and single-shot types. CoaddNet combines the efficiency of convolutional operations with the advantages of the Transformer architecture, enhancing spatial feature extraction through a multi-branch structure and reparameterization techniques. Performance evaluation shows that CoaddNet surpasses the baseline model, NAFNet, by increasing the Peak Signal-to-Noise Ratio (PSNR) by 0.03 dB and the Structural Similarity Index (SSIM) by 0.005 while also improving throughput by 35.18%. The model significantly improves the SNR of single-shot images, with an average increase of 22.8, surpassing the noise reduction achieved by stacking 70-90 images. By boosting the SNR, CoaddNet significantly enhances the detection of faint sources, enabling SExtractor to detect an additional 22.88% of faint sources. Meanwhile, CoaddNet reduced the Mean Absolute Percentage Error (MAPE) of flux measurements for detected sources by at least 27.74%.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112395"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142656806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of ant colony optimization algorithm based on farthest point optimization and multi-objective strategy in robot path planning","authors":"Shuai Wu ,&nbsp;Ani Dong ,&nbsp;Qingxia Li ,&nbsp;Wenhong Wei ,&nbsp;Yuhui Zhang ,&nbsp;Zijing Ye","doi":"10.1016/j.asoc.2024.112433","DOIUrl":"10.1016/j.asoc.2024.112433","url":null,"abstract":"<div><div>With the continuous development of high technology and the continuous progress of intelligent industry, mobile robots are gradually widely used in various fields. In the field of mobile robot research, path planning is crucial. However, the current ant colony optimization algorithm applied to mobile robot path planning still has some limitations, such as early blind search, slower convergence speed, and lower path smoothness. To overcome these problems, this paper proposes an ant colony optimization algorithm based on farthest point optimization and multi-objective strategy. The algorithm introduces new heuristic information such as the normal distribution model, triangle inequality principle, smoothness function, safety value function, etc. It adopts multi-objective comprehensive evaluation indexes to judge the quality of paths. For the high-quality and poor-quality paths, the algorithm takes additional pheromone increments and decrements in pheromone concentration to speed up the algorithm’s convergence. Besides, the farthest point optimization strategy is used to improve the quality of the paths further. Finally, to verify the algorithm’s effectiveness, the algorithm is compared with 20 existing methods for solving the robot path planning problem, and the experimental results show that the algorithm exhibits better results in terms of convergence, optimal path length, and smoothness. Specifically, the algorithm can produce the shortest path in four different environments while realizing the least number of turns with faster convergence, further proving the effectiveness of the improved algorithm in this paper.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112433"},"PeriodicalIF":7.2,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Segmentation of the customers based on customer value: A three-way decision perspective","authors":"Xiang Li ,&nbsp;Zeshui Xu","doi":"10.1016/j.asoc.2024.112415","DOIUrl":"10.1016/j.asoc.2024.112415","url":null,"abstract":"<div><div>This paper establishes an innovative value evaluation framework based on the criterion-oriented three-way decision (3WD) in the double hierarchy linguistic term (DHLT) environment to help the customer manager finish customer segmentation. Customer relationship management is the key to the success of enterprises in the information economy era. The segmentation of customers based on their relative criteria can identify the customers who are high-value customers for enterprises. However, multi-criteria decision-making can only display the value ranking of customers, rather than the value segmentation of customers. The employment of 3WD solves this problem. Then we classify the customers based on the 3WD method. First, the criteria are evaluated by using DHLTs, while the weights of criteria are acquired according to the maximum deviation method. Second, the conditional probabilities are estimated by the improved TOPSIS method combined with gray relation analysis, while the threshold values are calculated by the relative utilities which are constructed on the basis of the criterion information. Subsequently, the segmentation of customers is obtained according to the maximum-utility principle. Lastly, case research about the segmentation of customers based on value is used to demonstrate the practicality of our method, while some strategies about customer relationship management are given based on customer segmentation for obtaining maximum returns with minimum investment.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112415"},"PeriodicalIF":7.2,"publicationDate":"2024-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142656737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}