Search

The advancement of new energy vehicles has led to more demanding standards for detecting defects in cylindrical coated lithium batteries. The current research lacks robustness and has low performance. This paper seeks to provide real-time defect identification in cylindrical coated lithium batteries and improve the object detection method of the YOLOv5s model. This paper presents an MGSEC3 module with multi-scale feature extraction and integration of the SENet network in the YOLOv5 Backbone network. This module aims to reduce computational burden and enhance feature extraction efficiency as much as possible. At the same time, the CARAFE operator is utilized to enhance the up-sampling operator in order to reduce the loss of feature information. In addition, enhancements to the loss function enhance both the detection performance and convergence speed. The enhanced YOLOv5s model achieved an average detection rate of 82.4% on the custom cylindrical coated lithium battery dataset, 2.3% higher than the original YOLOv5s. This paper significantly enhances the precision and efficiency of flaw identification in cylindrical coated lithium batteries. [ABSTRACT FROM AUTHOR]

Positioning technology is one of the key steps of automatic docking. At present, vehicle docking technology is widely used in airport baggage cars, towing trailers and other fields, and the docking process is mostly done manually. In order to realize the automatic docking of vehicles in mixed scenes, this paper proposes a fast positioning method for docking vehicles in mixed traffic scenes. First, using the regression relationship between the template size in the camera's field of view and the actual distance, FFT template matching and threshold judgment methods are employed to locate, track, and match the template, as well as verify the threshold results, in order to achieve template long-distance positioning and ROI extraction. Then, based on the regression relationship between AprilTag pixels in the camera's field of view and actual distance, the experiment of AprilTag minimum recognition pixel is designed. Finally, the real-time pixels of AprilTag are calculated from the real-time ranging results, and compared with the minimum recognition pixel of AprilTag, it is judged whether the recognition conditions are met, so as to carry out the subsequent AprilTag positioning. The experimental results demonstrate that the method proposed in this paper can achieve rapid and accurate positioning of a target docking vehicle at a long distance, while reducing the mismatch rate. On this basis, by designing the identification conditions of AprilTag, the rapid identification and positioning of AprilTag is realized, which reduces the traversal time of the algorithm. The feasibility of this positioning method is verified by the attitude angle error experiment, which provides supporting conditions for the subsequent docking path planning and vehicle control. [ABSTRACT FROM AUTHOR]

This paper proposes a predictive control algorithm for the speed and displacement tracking of urban rail trains. Firstly, the train dynamics model is constructed considering the resistance existing in the actual operational scenarios. Secondly, based on the model predictive control (MPC) framework, a control objective function for tracking the desired speed and displacement and the stable change of control quantities is designed. Finally, combined with constraints of the train operation, the proposed MPC algorithm for the train tracking control problem is transformed into quadratic programming with inequality constraints, thereby facilitating a solution with the commonly-used solvers. Experimental results demonstrate that the proposed algorithm can effectively enhance the speed and displacement control performance while improving energy efficiency, ensuring the safety, stability, and riding comfort of the train. [ABSTRACT FROM AUTHOR]

Abstract-In recent years, electric vehicles have garnered significant attention, with lithium-ion batteries (LIBs) being central to their operation. Researchers and scholars have prioritized the accurate estimation of the state of charge (SOC) within the battery management system (BMS) as a key area of study. In this paper, by analyzing different equivalent circuit models, we choose to use the second-order RC model, elaborate the Kalman filter (KF) principle, and propose the adaptive extended Kalman filter (AEKF) to construct the estimation model of SOC. MATLAB validates the AEKF estimation model under two different operating conditions, UDDS and LA92, and the results show that the designed model can efficiently and accurately estimate the battery charge state with high competitiveness and accurately predict the real SOC direction regardless of the initial state, AEKF is more competitive than KF in terms of SOC prediction accuracy, Despite the different initial values of SOC, the roof-mean-square error of prediction was able to be controlled around one percent. [ABSTRACT FROM AUTHOR]

The progressive iterative approximation (PIA) is a versatile method for interpolating or fitting a given data set. The convergence behavior of PIA plays a pivotal role in determining the computational efficiency of data interpolation or fitting. This paper introduces an accelerated iterative approach, GS-PIA, derived from the Gauss-Seidel splitting, specifically designed for interpolating data points. The convergence and computational cost of the proposed iterative method are discussed. Numerical results underscore the superiority of GS-PIA in data interpolation when compared to other existing methods. [ABSTRACT FROM AUTHOR]

This paper presents a decision-making approach grounded in rough set theory and evidential reasoning to address the demand for expert decision-making in greenhouse environmental control systems. Furthermore, a decision-making model is developed by integrating the D-S evidence theory with an expert knowledge table for greenhouse environmental control systems. The model's reasoning process encompasses continuous attribute discretization, expert decision table formation, attribute reduction, and evidence combination reasoning. Firstly, the fuzzy C-means clustering algorithm is employed to discretize the original environmental data and cluster it. Subsequently, an attribute reduction algorithm based on information entropy is utilized to optimize the decision table by eliminating unnecessary conditional attributes in expert knowledge. The reduced indicators are then combined using evidential theory. Finally, suitable greenhouse control methods are determined by the confidence decision proposed by the D-S evidence theory. To assess the efficacy of this intelligent decision-making algorithm based on rough set and D-S evidence theory, its performance is compared with traditional SVM algorithms and small-shot learning algorithms. The results indicate that this proposed method significantly enhances the credibility of control decision-making processes, with an average running time of 0.002378s for the fusion decision algorithm and 0.017939s for the support vector machine (SVM) algorithm, respectively. The SVM accuracy rate after testing and training stands at 90.34%. Moreover, retraining based on information entropy attribute reduction leads to a correct decision rate increase of up to 100%. This method notably improves confidence levels in decision-making processes while reducing uncertainty and demonstrates reliability when applied in making decisions regarding greenhouse environments. [ABSTRACT FROM AUTHOR]

Person Re-Identification falls within the scope of computer vision, acting a technique to ascertain the presence of a specified pedestrian within a video or image library. The related research is of great significance in real-world environments such as criminal investigation and statistical analysis of commercial foot traffic and has received extensive attention from the academic community. However, traditional methods such as manual extraction cannot adapt to largescale data volumes, and deep learning-based methods at this stage suffer from interference in complex environments such as similar costumes, perspective changes, and occlusion. Therefore, in this paper, we investigate the above problems. Firstly, we expand the dataset by introducing random erasure-based preprocessing of pedestrian images to enhancing the robustness and generalization capability of neural networks. Secondly, a composite attention mechanism is introduced after the network residual layer to enhance the spatial information capability and feature expression. Finally, the union loss composed of Circle Loss, Ternary Loss, and Cross Entropy Loss was chosen for network training in the loss optimization phase. Findings from the experiments reveal that the improved method proposed in this experiment achieves 96.0%Rank-1 and 88.3%mAP in Market1501, which reflects the validity of the approach proposed in this manuscript, and provides valuable reference suggestions for Person Re-Identification related research. [ABSTRACT FROM AUTHOR]

Aiming at the problems of low accuracy, false detection, missed detection, and low real-time detection of current traffic sign detection, this paper proposes an improved traffic sign detection algorithm based on the YOLOv8s algorithm. Firstly, this paper proposes a double-layer semicomposite backbone network structure (DSCB), which uses the auxiliary backbone network to extract features, and then transmits the extracted features to the backbone network to enhance the ability of the backbone network to extract target features. At the same time, the deformable convolution is integrated into the DC2f structure of the auxiliary backbone network to enhance the generalization performance of the network. Secondly, the coordinate attention mechanism is used after the SPPF layer. The coordinate attention mechanism can better retain the coordinate position information of small targets, reduce the miss rate of the model, and increase detection accuracy. Finally, this paper introduces a new CAB module to learn and aggregate the output of each layer of the feature pyramid for global spatial context to enhance the feature representation ability further. The experimental results show that the improved algorithm achieves 90.51% detection accuracy, 82.00% recall rate, 89.51% mAP@0.5 on the TT100K dataset, and the FPS reaches 106. Compared with the original algorithm model, the detection accuracy is increased by 2.27%, and the recall rate is increased by 2.48%. mAP@0.5 is increased by 2.01%, and FPS is increased by 1. The improved traffic sign detection algorithm meets the requirements in detection accuracy and real-time detection. [ABSTRACT FROM AUTHOR]

Deep learning is proliferating within the field of computer vision. Road damage detection technology, as its offshoot, already plays vital role in road maintenance and traffic safety. With road damage such as potholes and cracks, accurate and efficient detection results are essential for timely road safety repair and maintenance. Therefore, road damage detection algorithms based on deep learning have attracted wide attention. YOLOv5 is an advanced target detection algorithm known for its efficient detection speed and good accuracy. However, there is still room for further improvement in its performance for road damage detection. The ability of multi-scale damage detection and spatial structure capture is not perfect. Therefore, this paper proposes three improvement points to improve the accuracy of road damage detection based on YOLOv5. The first introduced module is the Non-linear Spatial Pyramidal Pooling-Fast (NSPPF) module. This module allows for better capture of detailed features of road damage areas. Non-linear transformation and fast pyramid operation improve the sensing ability and multi-scale damage detection ability. Secondly, a combination of the CoordConv and SK attention modules is constructed. The CoordConv module fuses coordinate information with features to provide a more spatially informed representation. The SK attention module also learns correlations between global and local features, enhancing the model's ability to detect damages at different scales. This paper can better capture road injuries' spatial structure and context information by combining these two modules. Finally, experimental results on the RDD2020 dataset demonstrate the effectiveness of our model. Compared to the baseline model, the proposed improvement algorithm increases the accuracy by 2.2%, resulting in a mAP of 58.2%. This demonstrates its effectiveness and feasibility. [ABSTRACT FROM AUTHOR]

This paper intends to develop a new integral filled function method with four properties. First, the proposed filled function is non-exponential and non-logarithmic. This property is aimed at avoiding the overflow effect during the computational stage. Second, our method does not contain parameters, so the possibility of non-converging iterations can be prevented. Third, the filled function proposed in this paper is continuously differentiable. By this property, any local minimization procedures can be applied. Finally, the suggested filled function is dependent on the objective function, and thus it can provide any information for locating a minimum of the filled function. To show that the proposed method is competitive, a comparison has been made with one of the deterministic approaches, namely the DIRECT method. [ABSTRACT FROM AUTHOR]

In recent years, there is massive development of new metaheuristics as stochastic methods. Meanwhile, there is not any metaheuristics is powerful to handle all problems as stated in the no-free-lunch (NFL) theory. Based on this circumstance, this paper introduces a new swarm-based metaheuristics with the main strategy moving toward the resultant of better swarm members and avoiding the resultant of worse swarm members called group better-worse algorithm (GBWA). It consists of five searches: moving toward the best swarm member, moving toward the resultant of better swarm members, moving away from the resultant of worse swarm members, searching locally, and jumping to the opposite area. GBWA is then evaluated in three ways. The first evaluation is a comparative evaluation where GBWA is compared to five recent metaheuristics: coati optimization algorithm (COA), average and subtraction-based optimization (ASBO), clouded leopard optimization (CLO), total interaction algorithm (TIA), and osprey optimization algorithm (OOA). The second evaluation is the individual search evaluation. The third evaluation is hyperparameter test. The collection of 23 classic functions is chosen as the use case in all evaluations. The result of the first evaluation shows that GBWA is better than COA, ASBO, CLO, TIA, and OOA in 20, 21, 20, 21, and 21 functions consecutively. Meanwhile, the result of the second evaluation shows the equal contribution between the motion toward the best swarm member and the motion toward the resultant of better swarm members. [ABSTRACT FROM AUTHOR]

Nowadays, web applications play central roles in information systems using the Internet. Then, client-side web programming using HTML, CSS, and JavaScript should be mastered first by novice students. Previously, we have presented the element fill-in-blank problem (EFP) for its self-study. An EFP instance requests to fill in the blank elements in the given source code by referring to the screenshots of the corresponding web page. The correctness of any answer is marked through string matching. However, these blanks were manually selected by considering the importance of elements and the uniqueness of their correct answers. In this paper, we propose a blank element selection algorithm to automatically generate a new EFP instance from a given source code for client-side web programming. We define the seven rules on blank element selections from the code, and implement the procedure in Python using the open source BeautifulSoup and regular expressions' ' For evaluations, we applied the algorithm to the 47 source codes used for manual generations and obtained the better EFP instances with more blanks. Besides, we verified the effectiveness by generating 10 new instances with the algorithm and assigning them to 40 students. In addition, we extended its application to three source codes for games and verified the effectiveness by assigning them to 20 students, to further validate the applicability of the algorithm in EFP instance generations. We also evaluated the relationships between the number of blanks, the number of lines in source codes, the submission times and answer rates of students to further assess the adaptability of the algorithm. These results allow us to measure the algorithm's versatility in generating a wide range of EFP instances and contributes to comprehensive understanding of instance difficulties and learning outcomes. [ABSTRACT FROM AUTHOR]

To overcome the challenges of obstacle avoidance for Unmanned Aerial Vehicles (UAVs) in autonomous flights, this paper proposes the Dual Experience Attention Convolution Soft Actor-Critic (DAC-SAC) algorithm. This algorithm integrates a dual experience buffer pool, a self-attention mechanism, and the Soft-Actor-Critic algorithm with a convolutional network. The dual experience buffer pools are used to solve the problem of ineffective UAV training due to the scarcity of successful training data. To overcome the drawbacks of the original Soft Actor-Critic (SAC) algorithm in handling image data, a Convolutional Neural Network (CNN) is applied to reconstruct the actor and critic network, allowing for better image feature extraction and classification. Furthermore, a self-attention mechanism is employed by adding a convolutional self-attention layer to the network. This modification enables dynamic adjustments for the attention weights based on varying input image features, effectively addressing focus-related challenges. Two simulation experiments are performed and the DAC-SAC algorithm achieves a 99.5% success rate in a known environment and an 84.8% success rate when dealing with an unknown environment. These results confirm that the proposed algorithm enables autonomous obstacle avoidance for UAVs even when considering depth images as input. [ABSTRACT FROM AUTHOR]

This paper proposes an explainable recommendation algorithm based on deep learning for developing a transparent and explainable recommendation system. The proposed algorithm combines a multi-feature fusion model for text sentiment analysis and the innovative DeepxDeepFM recommendation model to provide accurate and interpretable recommendations. First, Bi-LSTM and MCNN are employed to extract multi-dimensional features from comment data. Then, the DeepxDeepFM recommendation model is used explicitly for comment data with numerous feature vectors and significant linear changes. Finally, experimental results demonstrate that our proposed explainable recommendation algorithm increases the accuracy by 1.57% and decreases the root mean square error by 2.69%, contributing to higher model performance. Compared to other models, the improved interpretable recommendation model is smaller in size and more accurate, so it can maximize the click-through rate of e-commerce recommendation systems, which is crucial for achieving precise recommendations in the field of e-commerce. [ABSTRACT FROM AUTHOR]

A new swarm-based metaheuristic, namely the enriched coati osprey algorithm (ECOA), is proposed in this paper. As its name suggests, ECOA hybridizes two new metaheuristics, the coati optimization algorithm (COA) and the osprey optimization algorithm (OOA). ECOA is constructed by five searches performed sequentially by the swarm members. The first three are directed searches, while the last two are neighborhood searches. All three directed searches are adopted from COA and OOA. Meanwhile, the four-bordered neighborhood search is developed based on a new approach. During the assessment, ECOA was challenged to overcome the set of 23 functions and contended with five new metaheuristics: total interaction algorithm (TIA), golden search optimization (GSO), average and subtraction-based optimization (ASBO), COA, and OOA. The result shows that ECOA outperforms TIA, GSO, ASBO, COA, and OOA in 16, 23, 18, 21, and 21 functions. Meanwhile, the individual search test result shows that the directed searches perform better than the neighborhood searches. Moreover, the directed search toward the best member becomes the most dominant search. [ABSTRACT FROM AUTHOR]

The Honey Badger Algorithm (HBA) represents a novel swarm intelligence optimization algorithm introduced in recent years. However, its predominant constraints are linked to inadequate convergence accuracy and a vulnerability to entrapment in local optima. In an effort to mitigate these challenges, this paper introduces an Improved Honey Badger Algorithm Based on a Hybrid Strategy (OHBA). Firstly, during the population initialization phase, a method involving the utilization of a good point set is introduced to enhance the diversity and introduce more randomness into the population. Secondly, in the position update phase, the Beta distribution is employed as an alternative to the Uniform distribution, aiming to strike a balance between global exploration and local exploitation capabilities. Thirdly, an improved adaptive density factor strategy is incorporated into both global and local position updates to enhance the algorithm's convergence precision and speed. Lastly, within the global exploration stage, a Cauchy mutation strategy based on the Sine chaotic mapping is introduced to facilitate the algorithm in overcoming local optima and reinforcing its optimization capabilities. The improved algorithm's performance has been evaluated through a comprehensive set of assessments, including CEC-2017 functions, CEC-2022 functions, Wilcoxon rank-sum tests, and practical engineering optimization problems. These evaluations were undertaken to assess the algorithm in comparison to classical intelligent optimization algorithms. The experimental results show that OHBA possesses significant advantages in terms of convergence speed, optimization accuracy, robustness and its practical utility and effectiveness in addressing complex optimization challenges. This establishes OHBA as a highly competitive option in these critical aspects of optimization. [ABSTRACT FROM AUTHOR]

Let f: V (G) → {1, 2, . . ., n} be a bijection for a graph G of order n. The weight of a vertex v of G denoted by w(v) is defined as sum of labels of all vertices adjacent to vertex v in G. If w(u) ̸= w(v) for every pair of distinct vertices u, v ∈ V (G) then labeling f is called distance antimagic. Any graph G which admits such a labeling is called a distance antimagic graph. In this paper, we determine the distance antimagic labeling of Kneser graph K(2n, n), bipartite Kneser graph H(2n+1, n) using computer-aided algorithm. We investigate the existence of distance antimagic labeling for cycle related graphs having application in surveillance systems of civil engineering and urban planning and verify that vertex weights are distinct using computer algorithm. Also, we present some families of disconnected graphs that admit distance antimagic labeling and later show that Kneser graph K(n, 2) does not admit (a, d)-distance antimagic labeling. [ABSTRACT FROM AUTHOR]

We improve the approximated solution for the inventory model with a linear trend in demand such that there is no necessary to adjust the last or the first time interval. Our formula is easier to compute. We estimate our approximated result to the optimal solution of the equal time interval constraint, then our result is measured with an error of less than two-thirds. For the second inventory system examined in this paper, we consider an inventory system with two retailers and a manufacturer. Through the algebraic process, we obtain the optimal profit that is missing in the previously published paper. We show the elegant character of the algebraic process. [ABSTRACT FROM AUTHOR]

In radial distribution networks, the unreasonable use of distributed generators (DGs) will enhance the effects of voltage quality reduction and network loss. Consequently, the optimal allocation of distribution groups (OADG) has far-reaching implications for the economical, secure, and stable operation of radial distribution networks. The purpose of the study on the optimal allocation of distributed generators is to allocate the locations and capacities of DGs scientifically and rationally, enhance voltage quality, and reduce power loss. This paper proposes an enhanced ISC-SSA algorithm with constraint strategy for the OADG problems, which addresses the disadvantages of traditional salp swarm algorithm (SSA), such as low solution accuracy and slow convergence speed, and introduces a sine cosine algorithm (SCA) to optimise the salp swarm followers. To reduce the limitation of only the leader guiding the followers for position updates, a symbiotic strategy with multiple subpopulations is introduced to simulate the sharing of information of the salp swarm population, relying on the decision-maker and the leader to jointly guide the population movement, thereby increasing the global optimization-seeking capability of the algorithm, and the method is implemented in IEEE-33, IEEE-69, and IEEE-119 systems. In addition, the paper extends the ISC-SSA algorithm to the MOISC-SSA algorithm to solve in IEEE-33, IEEE-69 test systems the multi-objective OADG problems. Multiple sets of simulation results demonstrate that the proposed method solves radial distribution system problems of varying sizes with greater search precision and greater significance. [ABSTRACT FROM AUTHOR]

Aiming at the problems of low recognition accuracy and large calculation amount of age estimation, this paper proposes a staged single aggregation backbone network VoVNetv4. In order to improve the feature extraction ability of the network, the Regional One-Shot Aggregation (ROSA) module is proposed in the network for regional encapsulation of the feature module. The classifier adopts the COnsistent RAnk Logits (CORAL) framework, which can solve the problem of inconsistency of classifiers in ordered regression CNNs for multiple binary classification tasks. This paper conducts experimental verification on the Asian Face Age Dataset (AFAD) and compares it with other research methods. The experimental results show that VoVNetv4 can effectively improve the feature extraction and detection ability of the network and improve the detection efficiency of age estimation. [ABSTRACT FROM AUTHOR]

This paper proposes a robust identification and control-based on a neural network method for a Twin Rotor Multivariable System (TRMS) using a recursive adaptive training algorithm. The algorithm is based on a recursive least squares approach with an additional steepest descent stage. An Adaline neural network is used for modeling the system, and a robust structure is selected based on a linear auto-regressive structure with exogenous inputs (ARX) related to the estimation error. The identification is performed online and the system is controlled under a polynomial structure by pole placement with a dead-beat strategy. The method is evaluated in terms of estimation and tracking error in the presence of external additive disturbances, parametric disturbances, and sinusoidal reference signals. The Root-Mean Square Error (RMSE) is used to evaluate the estimation performance and the Integral-Time Absolute Error (ITAE) is used to evaluate the tracking performance. As a result, a novel robust controller based on a neural network is designed where the best results are obtained for a training recursive least squares algorithm with an additional steepest descent stage. [ABSTRACT FROM AUTHOR]

The Internet of Vehicles (IoV) necessitates high data throughput and minimal latency for effective and practical implementation. In addition, the rapid movement of vehicles can lead to considerable load imbalances in IoV base stations (BSs). This paper proposes an adaptive clustering algorithm for 3D IoV coordinated multipoint (CoMP) transmission and reception communication within the 5th-generation (5G) communication environment to address these challenges. The IoV communication nodes are divided into BSs and vehicle users (hereafter called "users"). Among the users, there are two subcategories: Cell-center users (CCUs) and cell-edge users (CEUs). To select the Cluster Head (CH), the user with the highest count is chosen based on the number of distances to similar users that meet a specific distance criterion. The remaining associated users form a user cluster with the CH. The process continues in this manner until all users have been clustered. The CHs of CCUs and CEUs communicate with the primary service BS situated at the minimum distance to the CH and with coordinated BSs, respectively. The simulation results show that the novel algorithm can improve the average information rate of the IoV by 108.11%, reduce the delay by 43.88%, and decrease the load of IoV BSs by 73.56%. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is evolving rapidly and has made many theoretical breakthroughs while widely applied in various fields. ML allows systems the ability to access data and use it to enable computers to execute cognitive processes such as learning and improving from previous experiences and solving complicated issues. Many first-order stochastic optimization methods have been used to solve the optimization model of ML. These algorithms adopt Barzilai-Borwein (BB) step size instead of fixed or diminishing step size to improve performance. However, the BB step size format involves fractional calculation, which inevitably leads to a zero denominator, especially when the objective function is non-convex. The BB technique will be violated if the denominator is near 0 or even negative. To improve the computation of the step size, a Positive Defined Stabilized Barzilai-Borwein (PDSBB) approach is introduced in this paper. Integrating PDSBB with the stochastic variance reduced gradient (SVRG) approach, a new method SVRG-PDSBB is proposed. Numerical experiments have shown that the new algorithm has stabilized the performance of the new step size, which successfully avoiding zero denominators and effectively solving the common problems in machine learning. The convergence of SVRG-PDSBB is theoretically and numerically proven, and the effectiveness of the new algorithm is shown by comparison with other algorithms. [ABSTRACT FROM AUTHOR]

The Lotka-Volterra competition model consisting of two equations is studied. The existence and uniqueness of solutions on an infinite interval are proved by using the Schauder fixed point theorem, Gronwall’s inequality and some special analytical techniques. Some conditions of existence for positive solutions are obtained. Iterative algorithms and error estimations for solving this model are established. The results of this paper can be generalized to the cases consisting of more than two equations. [ABSTRACT FROM AUTHOR]

Multi-objective problems have gained much attention during the last decade. To balance the diversity and the convergence of the multi-objective differential evolution algorithm (MODE), an improved MODE is proposed based on the affinity propagation clustering (APC) and the non-dominated count approach in this paper. The proposed algorithm is referred to as AP-MODE, which improves the search efficiency by utilizing the affinity propagation approach to find out the population distribution structure for guiding search. In addition, mating restriction probability is used to select parent individuals for recombination from the neighborhoods or the whole population. Meanwhile, the mating restriction probability is updated according to the non-dominated count approach at each generation. This proposed algorithm is verified by comparing it with some state-of-the-art multi-objective evolutionary algorithms, and the simulation results on DTLZ test problems indicate that AP-MODE can efficiently achieve two goals of multi-objective optimization, i.e., the convergence to actual Pareto front and uniform spread of individuals along Pareto front. [ABSTRACT FROM AUTHOR]

In foggy days, the color saturation and contrast of the image are reduced due to factors such as scattering and refraction of light. Aiming at the problems of fog residue and loss of detailed features after processed by the existing dehazing methods, a defogging model network APSA-DehazeNet (Adaptive Pyramid Split Attention-DehazeNet) based on split convolution is proposed in this paper. Firstly, the adaptive multi-scale feature fusion module is used to extract the features of the foggy image, capture the features of different scales, then perform weighted fusion and obtain the shallow features of the image. Secondly, the deep features of the image are further obtained by the split convolutional network (PSANet) based on attention mechanism, and a more thorough dehazing effect is obtained. Finally, to effectively solve the problem of detail loss caused by a single loss, a joint loss function of perceived loss and structural similarity loss is proposed. Compared with other algorithms, experimental results demonstrate that the PSNR and SSIM indexes on the synthetic fog map dataset are improved by an average of 4.7dB and 7.4%, respectively. [ABSTRACT FROM AUTHOR]

A clique is a maximal complete subgraph of a graph. The cc-degree (clique-clique degree) of a clique K (dcc(K)) is the number of cliques adjacent to K. A clique C strongly clique-dominates a clique K if C is adjacent to K and dcc(C) ≥ dcc(K). Let C(G) be the set of all cliques in a graph G. A set S ⊆ C(G) is a strong clique-clique dominating set (SCCDset) of G if every clique in C(G)-S is strongly clique-dominated by at least one clique in S. The strong clique-clique domination number γscc(G) is the cardinality of a smallest SCCD-set of G. Similarly, the weak clique-clique domination number γwcc(G) is defined. In this paper, we study some properties of these strong (weak) clique-clique domination parameters and obtain Gallaitype results. We present an algorithm to find γscc(G) (γwcc(G)) and obtain some bounds for the newly defined parameters. Further, we define and study clique-clique domination balanced graphs and clique-posets. [ABSTRACT FROM AUTHOR]

Motivated by some well-known conjugate gradient methods, in this paper, we propose a new hybrid conjugate gradient method for unconstrained optimization. Without any dependence on line search, the new method generates sufficient descent property. Under some mild conditions, we prove that the new method with MWWP line search is globally convergent. Our numerical experiments show the effectiveness and robustness of the new modified method in comparison with three exiting algorithms. [ABSTRACT FROM AUTHOR]

The optimal configuration problem of distributed generators (DGs) and shunt capacitors (SCs) in a radial distributed network (RDN) is to find the best installation locations and optimal capacities of DGs and SCs for optimizing a certain performance indicator. The discontinuity characteristic and huge computation of optimizing DGs and SCs in RDN make it no longer applicable by traditional methods. In this paper, a new bus processing method and an improved butterfly algorithm are proposed to solve the optimal configuration of DGs and SCs. The method of processing nodes considers not only the power loss index (PLI) but also the voltage amplitude of the original system, forming a sequence of candidate nodes to guide algorithm to optimizing, which can simplify the algorithm search space and improve search efficiency. Meanwhile, the butterfly algorithm with constriction factor (BF-CF) combines the inertia coefficient to introduce a constriction factor, improves the speed update model and the local search pattern, and overcomes the disadvantages of the original butterfly algorithm which is easy to fall into the local optimum and the precision of result is not high. To verify the performance of the proposed method, minimizing the active power loss and voltage deviation are selected as the objective function and reactive power loss and the worst voltage are taken as reference targets, which are performed in three standard test systems of 33-bus, 69-bus and 119-bus, respectively. The simulation results illustrate that, compared with the original butterfly algorithm and other intelligent algorithms, the method proposed in this paper has more obvious advantages and performance in solving the configuration problem of DGs and SCs in systems of various scales. [ABSTRACT FROM AUTHOR]

G(V, E) represents the basic chart without circle, if existing a one-to-one mapping f:E(G) → {1,2,...,|E|}, for any two vertices in the diagram, in the event that d(u)= d(v). S(u)=S(v), where S(u)= Σuw∈E(G)f(uw), d(u) represents the degree of the vertex u, then call the mapping f: Adjacent Vertex Reducible Edge Labeling (alluded as AVREL). In graph theory, graph coloring and graph labeling are two research directions of graph theory, and there is little correlation between the two in previous research results. In the process of researching the concept of Adjacent Reducible Edge Coloring proposed by Professor Zhang Zhongfu, we found that there are several graph classes whose coloring number reaches the sum of the number of vertices and edges, so we propose a new concept of Adjacent Reducible Edge Labeling. In the transportation network, the edge weight represents the transportation capacity, and the node transportation capacity is represented by the sum of its associated edges. Two nodes with the same degree of adjacency require the transportation capacity to be as equal as possible, which can be described by the Adjacent Vertex Reducible Edge Coloring model. when the road diversity reaches the extreme value, it can be described by the Adjacent Vertex Reducible Edge Labeling model. In this paper, designing and using Adjacent Vertex Reducible Edge Labeling algorithm (abbreviation: AVREL algorithm). The algorithm recursively looks through the arrangement space of the Adjacent Reducible Edge Label through the underlying label of the edge, lastly sifts through the graph book fulfilling the edge label and results as a label matrix. In the wake of examining the algorithm results, some special graphs such as Petersen-pyramid graphs, Möbius ladder graphs, bicyclic graphs, and some joint graphs in various situations are summed up, the proofs and conjectures are given. [ABSTRACT FROM AUTHOR]

Unmanned system research has attracted increasing academic and corporate attention in recent years. Traffic-sign and road detection systems play an important role in unmanned systems. However, current algorithms perform poorly in reading traffic signs from a long distance and cannot satisfy the real-time requirements of accurate and rapid detection. Therefore, Yolov5s-Swin model is proposed in this paper. First, the Swin Transformer module and convolution module are fused, the ResUnit module in the Yolov5s backbone network is improved, the Crswin module is proposed, and the PAFPN in the network is modified to strengthen the ability of the model to capture local feature information and improve network detection accuracy. The Swin Transformer module uses the moved-window attention function to segment the image into different windows of a certain size but cannot establish a connection between neighboring windows. Consequently, we designed a method for hyperbolic window attention. This approach updates the window partition method such that the window block for attention calculation at each pixel changes, which can improve the receptive field, increase the information extraction ability of the target, and solve the information loss caused by decreased resolution during image training. Experimental results show that the proposed improved Yolov5s-Swin model exhibits significantly better detection accuracy than existing models. [ABSTRACT FROM AUTHOR]

The traditional sparse subspace clustering algorithms are easily affected by the similarity matrix, which may lead to different clustering results by different similarity matrix. That is to say, constructing a reasonable similarity matrix is the key to sparse subspace clustering. Based on reweighting subspace clustering, a semi-supervised sparse subspace clustering algorithm based on re-weighting is proposed in the paper. Firstly, the global similarity structure of the data can be better captured by constraining the coefficient between the cannot-link labels to be 0. Secondly, with the help of reweighted l1-norm minimization sparse optimization framework, the adaptive similarity matrix can be obtained. Furthermore, the above similarity matrix can be adjusted by using prior information. Experimental results indicate that the proposed clustering algorithm is more efficient than other clustering algorithms on benchmark data sets. [ABSTRACT FROM AUTHOR]

In this paper, a modified and hybrid flower pollination algorithms (MHFPA) is proposed for dealing with the multi-objective optimal power flow (MOOPF) problem with conflictive objectives. The algorithm combines the mutation and crossover process in the differential evolution (DE) algorithm, introduces the sinusoidal nonlinear dynamic switching probability (SNDSP) and the elite strategy of elder generation (ESEG), which can improve the shortcomings of the original pollen algorithm that it is easy to fall into the local optimum and the diversity is insufficient. A screening approach with Pareto-dominant rule (SAPR) is proposed to ensure that the state variable can satisfy the inequality constraints of the power system. A uniformly distributed Pareto optimal set (POS) is obtained by the non-dominant sorting with elite strategy (NSES) based on Rank and Density estimation, and the best trade-off solution (BTS) is determined from the POS obtained by the fuzzy affiliation theory. For practicality, the total fuel cost, active power loss, emissions and voltage deviation are selected as objective functions. Due to the limitations of the actual power system, the valve point effect is also considered. The IEEE30-, 57- and IEEE118-bus test systems are used to verify the performance of the proposed MHFPA. In addition, two performance indicators, Hypervolume (HV) and Spacing (SP), quantitatively evaluate the diversity and uniformity of the POS obtained by MHFPA. The simulation results show that, compared with the classic MOPSO and NSGA-II algorithms, the method proposed in this paper shows a greater competitive advantage in dealing with different scales and non-convex optimization problems. [ABSTRACT FROM AUTHOR]

Multi-objective reactive power optimization (MORPO) is a high-dimensional, nonlinear, multi-constraint problem. To solve this problem, an improved multi-objective honey badger algorithm (MOIHBA) is proposed. To address the shortcomings of the original algorithm, such as easy falling into local optimum and insufficient population diversity, the improved algorithm introduces a sine chaotic mapping strategy to expand the population diversity, a backward learning mechanism to narrow the range of high-quality solution sets, and a cross-learning mechanism to improve the precision of the algorithm optimization process. In addition, in order to obtain the pareto optimal set (POS), a method based on calculating individual rank and crowding distance is proposed to sort the non-inferior solution, and the best compromise solution (BCS) is obtained by using a fuzzy theory strategy. By introducing three objective functions of active power loss, voltage stability index, and voltage deviation, the multi-objective reactive power optimization is established. To investigate the robustness of the introduced improved algorithm and its ability to solve the MORPO problem, this paper uses IEEE30, IEEE57, and IEEE118 as test systems that optimize the dual objective and triple objective simultaneously. In order to study the comprehensive performance of the improved algorithm, the algorithm time complexity, GD index, and HV index are adopted for evaluation. The simulation results and performance index results show that compared with other algorithms, MOIHBA has better BCS and pareto fronts (PFs) with the best uniformity and convergence. Therefore, the MOIHBA algorithm has a greater competitive advantage in solving the MORPO problem. [ABSTRACT FROM AUTHOR]

In recent years, the Corona Virus Disease 2019 (Covid-19) epidemic has raged around the world, with more than 500 million people diagnosed. Relevant medical research and analysis results on Covid-19 indicate that wearing masks is an effective method to prevent and restrain virus transmission. Mask detection stations have been set up in hospitals, railway stations, schools, where there is large crowd flow, but results are not as good as expected. In order to ameliorate pandemic preventing and control measures, a mask wearing detection algorithm YOLOv3-M3 was designed and proposed in this paper. The algorithm can effectively detect people without mask, while consequently reminding them. Firstly, we substituted the feature extraction network of YOLOv3 with MobileNetv3, a lightweight convolutional neural network. Secondly, we utilized K-Means++ to substitute the original ground truth clustering algorithm to improve prediction precision. In addition, the bounding box regression loss function was revised as CIoU loss function. This loss function solves the issues of overlapping between the ground truth and the anchor box, which has increased the training speed. After experiments, the precision of YOLOv3 algorithm on mAP 0.5 and mAP 0.75 is 93.5% and 71.9%, respectively. Elevating 3.1% and 2.6%, respectively, higher than that of YOLOv3 algorithm, and it was superior to SSD, SSD Lite, YOLOv3-Tiny and other one-stage object detection algorithms. The detection speed can reach 13.6 frame/s, which has met the requirements of pandemic prevention and control in most places and can be deployed on terminal devices for object detection. [ABSTRACT FROM AUTHOR]

It is significant to detect the bridge from LiDAR automatically and quickly, which is regarded as an important artificial building and transport hub. In this paper, a skinny triangle-based detection algorithm of bridge over water is proposed. First, Triangulated Irregular Network(TIN) is generated in LiDAR data, during which a skinny triangle is used to detect points on the river edge and bridge, and the elevation difference between them is used to separate those points. Second, a nonlinear classifier designed with TIN is used to separate the LiDAR points on the river edge. Points of bridge borders can be separated automatically and adaptively according to the collinear characteristics of the bridge border. Finally, creating fitted curves to obtain river bank and bridges borders respectively. The coordinate of bridge angular points can be obtained by solving the equations set, which is consists of the river bank curve and bridges borders curve simultaneous. The experimental results demonstrate that the proposed algorithm can perform well to extract bridge over water from LiDAR data. [ABSTRACT FROM AUTHOR]

Trapdoor functions contributed since their announcement in the evolvement cryptography as we know it, especially the lossy mode, by helping reduce the residual leakage for an optimal rate, but to make it more resilient cryptographically: generic constructions were made based on graph isomorphism, or other NP-hard problems defended by the zero-knowledge proof, such were used in Indistinguishability under Chosen-Plaintext Attack (IND-CPA), Computationnal Diffie-Hellman (CDH), or Decisional Diffie-Hellman (DDH). Once schemes like Indistinguishability under Chosen- Ciphertext Attack (IND-CCA) were adopted it became clear it cannot simulate a decryption using Lossy Trapdoor Functions (LTF); the problem with existing trapdoor functions in general is partial information leakage, lack of randomness and multiple messages insecurity. In the light of the following issues came the idea to present through this paper a simple but important fix, in the note of randomness a new Variate of the Engel expansion (VEE) is chosen, providing a pseudo-random bit sequence as an output, the reason being to recover the seed of the algorithm for an attacker, it is considered a hard number theory problem, and surely after the new construction in this paper, another NP-complete problem emerging from tensors the scheme is more secure. As for the strenghtening evidence of how it can be trusted, it seems more robust to supply a proof of its ergodicity as being done in this article, instead of semantic security analysis, to prove the efficiency of the new construction resolving the issues surrounding multi-challenge using a lossy trapdoor function. [ABSTRACT FROM AUTHOR]

Distributed generators (DGs) are recognized as an effective method for controlling the power loss, voltage stability, etc.. In this paper, a novel hybrid algorithm of beetle antennae search (BAS) and particle swarm optimization (PSO) is presented for optimal allocation of DGs in radial distribution network. The BAS describes the beetle's individual search by smell, PSO describes the group search of birds by location. The proposed algorithm combines their advantages, proceeds with individual optimization while conducting group optimization. Therefore, the proposed algorithm searches widely, and converges fast. In this paper, a series of improvement measures are proposed to deal with the shortcoming of PSO-BAS which is easy to fall into local optimum. These methods include equal interval initialization, cross mutation, and non-linear learning factor. This paper will show the comparison results of PSO-BAS and IPSO-BAS in the six confessed test functions to prove the necessity of the improved methods. Simultaneously, in order to verify the feasibility and effectiveness of this proposed algorithm in terms of practical application, it is tested on the standard IEEE 33-bus, IEEE 69-bus and IEEE 119-bus systems. The results of active power loss and voltage stability show that proposed algorithm is more effective and more suitable for the power distribution system than other algorithms. At the same time, this article also explores the impact of new energy sources on the annual network loss. Besides, a method for optimizing the annual network loss is proposed. Here, the IPSO-BAS algorithm is used to adjust the size of multiple biomass energy sources within 24 hours to optimize the network loss, and a comparison plan is designed to verify the feasibility of the proposed method. According to the final result, this proposed method can greatly reduce the annual network loss. [ABSTRACT FROM AUTHOR]

This paper proposes a hybrid algorithm for enhancing the performance of visible light communication (VLC). VLC-based system utilizes light-emitting diodes (LEDs) to illuminate the surroundings and transmit the data simultaneously. The key motivation behind this study was to design a system that can reduce the bit error rate (BER). Consequently, it is important to evaluate different modulation strategies for VLC-based systems and compare their performances. This paper focuses on optimizing various parameters of VLC systems by proposing a novel hybrid scheme. The hybrid strategy combines the On-off keying (OOK) method with the Pulse Width Modulation (PWM) to optimize performance parameters. The study designs the circuit for both modulation techniques including the transmitter and receiver circuits. The key parameters to be studied are BER, bandwidth and flickering to determine which modulation technique is better and finds the best trade-offs among these modulation schemes. MATLAB based simulations compare the modified algorithms with the original algorithms and prove the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

To effectively handle the many-objective optimal power flow (MOOPF) problems considering the simultaneous reduction of power loss, emission and fuel cost, an improved NSGA-Ⅲ (INSGA-Ⅲ) algorithm is put forward in this paper. In detail, the proposed INSGA-Ⅲ algorithm adopts the competitive solutions preliminarily optimized by traditional NSGA-Ⅲ method as the initial population and integrates the novel adaptive dominant (NAD) strategy. Comparing with the original NSGA-Ⅲ algorithm, INSGA-Ⅲ obtains the more preferable Pareto front (PF) with uniform distribution. More significantly, an entirely new BP fuel-cost prediction network is proposed to explore the potential elite power flow (EPL) solutions. These EPL solutions determined around the best compromise solution (BCS) of INSGA-Ⅲ algorithm provide decision-makers with more and better scheduling schemes. The effectiveness and superiorities of proposed INSGA-Ⅲ algorithm and BP fuel-cost prediction model are verified by both dual-objective and triple-objective MOOPF simulation experiments. In general, this paper presents an innovative way to solve the complex engineering problems by computer technologies represented by intelligent algorithms and neural networks. [ABSTRACT FROM AUTHOR]

The thinning algorithm is one of the fastest approaches to extract skeletons from an object, especially when adopting the parallel strategy. Skeletons are very useful descriptors and can be applied in many recognition fields. However, one of the drawbacks that limits the use of these techniques is that thinning algorithms are not robust against inner noise and outer noise, which may produce many unwanted branches. To alleviate the influence of noise and increase the robustness, pruning methods and scale-space methods have been proposed in the past, in which pruning methods are aimed at suppressing the outer noise (boundary noise) and scale-space methods are aimed at suppressing the inner noise (such as scratch noise and dithering noise). In this paper, we proposed an improved framework that can deal with both inner noise and outer noise. The experiment proved that the proposed framework has better visual effects than the existing pruning method and existing scale-space method. In addition, the proposed framework is an adaptive framework that does not require manual tuning of parameters. [ABSTRACT FROM AUTHOR]

In this paper, we present a Crank-Nicolson leapfrog (CNLF) time stepping decoupled scheme for the fluid-fluid interaction problems. The scheme is based on the modular grad-div stabilized (MGDS) method which is very efficient for the fluid-fluid interaction problems. The algorithm is very simple to implement and retain the benefits of grad-div stabilization. The unconditional stability of the proposed scheme is proven rigorously. Numerical tests are given to verify the efficiency and robustness of the proposed scheme. [ABSTRACT FROM AUTHOR]

In this paper, we mainly propose a new parameterfree filled function, which derives from two inverse trigonometric functions. First, we accomplish the analytical studies, to prove that filled function proposed in this paper possess all the properties of filled function. Secondly, based on this parameter-free filled function, a comparatively new algorithm was built. Finally, we implemented the algorithm to solve unconstrained global optimization problems. The computational results demonstrated in this paper revealed the effectiveness of the proposed filled function compared to some related results of the filled functions in the literature. [ABSTRACT FROM AUTHOR]

This paper proposes an improved novel bionic intelligent Moth-Flame Optimization (MFO) algorithm optimized linear active disturbance rejection control (LADRC) strategy for automatic generation control (AGC) problems. LADRC is a control strategy that combines linear extended state observer (LESO) and traditional proportional-derivative (PD) control, with the advantages of its resistance to disturbance and simplicity. The three-order LADRC controller is designed in load frequency control (LFC) system. The Improved MFO (IMFO) algorithm is proposed by Cauchy mutation strategy and self-adaptive weight to overcome the weaknesses in easily falling into local optima and poor optimization search accuracy, and the IMFO algorithm is employed for tuning the parameters of LADRC. Besides, an objective function takes account of the performance index integral-time-multiplied-AE (ITAE) with dynamic indicators, which is used to improve the effectiveness of LADRC. The proposed control strategy is tested for robustness in a two-area non-reheat thermal system and compared with recent control methods published in the literature. As a result, its superiority is demonstrated for the lowest cost function values of ITAE = 0.0036, ITSE = 1.24E-05, ISE = 6.81E-05 and IAE= 0.0086. To further explore the potential of the LADRC controller based on the IMFO algorithm, it is also extended to a two-area nonlinear system and an unequal three-area reheat system. The performance indicator functions and dynamic performance indicators are analyzed qualitatively in the meantime. [ABSTRACT FROM AUTHOR]

This paper deals with multi-objective school bus routing problem, which includes route balance, total number of school buses and total travel distance optimization objectives. An improved non-dominated sorting genetic algorithm (NSGAII) is proposed to solve this problem. First, the definition of measurement indicator that denotes the degree of route balance is given based on the analysis of the solved problem. And then, the multi-objective optimization function is provided. In the proposed algorithm, the individuals are obtained by using the tournament selection, sequential crossover and inverse mutation. The 2-opt neighborhood operator is adopted to improve the best individuals obtained in each iteration. At the same time, the route selection rule based on the degree of route balance is applied to select the final optimal solution set. The solution with better balance degree will be taken as the best solution. Finally, some benchmark instances are used to test the effectiveness of proposed algorithm. The results reveal that the proposed algorithm outperforms the standard NSGAII and Multi-objective Evolutionary Algorithm(MOEA). The experimental results also show that our algorithm has good stability. [ABSTRACT FROM AUTHOR]

Let graph G(p,q) be a simple graph, if there is a positive integer k(1 ≤ k≤ |E|) and the mapping f: E(G) → {1, 2, . . ., k}, such that for any edge 𝐮 v ∈ E(G), when d(u) = d(v), S(u) = S(v), where S(v) = E𝒖v∈E(G) f(𝒖 v) . d(u) represents the degree of point u, then f is the coloring of Adjacent Vertex Sum Reducible Edge of G. The maximum value k is called the Adjacent Vertex Sum Reducible Edge chromatic number of graph G, denoted as XA𝒖𝒗sr ' (G). Based on the existing graph coloring concepts and reducible concepts, this paper proposes a new concept of adjacent vertex sum reducible edge coloring combined with practical problems and designs a new type of adjacent vertex sum reducible edge coloring algorithm. [ABSTRACT FROM AUTHOR]

For traditional data mining, all data shall be loaded into memory for analysis and calculation. It belongs to a stand-alone computing mode, which has low calculation efficiency, and a high mining failure rate during the work process. As the data storage and computer technology develop rapidly, how to store and process big data effectively has become an important problem to be solved. Cloud computing can quickly obtain resources from the computing resource pool, and implement parallel improvement of data mining algorithms, which can achieve an efficient combination of cloud computing platform and data mining, and effectively make up for the bottlenecks faced by traditional data mining processes. Therefore, based on the Hadoop cloud computing platform, this paper makes full use of the characteristics of the MapReduce programming framework, and proposes a parallel design of decision tree nodes, node attribute metrics, and Gini index ranking for the SPRINT decision tree algorithm. The performance of the parallelized SPRINT algorithm on classification accuracy, scalability, and speedup ratio is tested. The results indicate that the parallel design of the SPRINT algorithm can obtain good scalability and parallel speedup under the premise of ensuring classification accuracy, which verifies the feasibility of the parallel design of data mining algorithms on the basis of cloud computing. [ABSTRACT FROM AUTHOR]

In the event of an emergency, the efficient and reasonable scheduling of emergency rescue workers in the initial stage of the emergency rescue process can considerably reduce the associated loss. In this regard, this paper proposes a bi-level game scheduling model for emergency rescue workers under the condition of limited rationality, primarily considering the demand game for limited emergency rescue workers in multiple disaster areas and the limited rational behavior of victims in the game process. Based on the proposed model, a Shuffled Frog Leaping Algorithm (SFLA) is designed to solve the problem. Finally, the rationality of the model is verified through a case study. The results indicate that the model of the emergency rescue workers scheduling (ERWS) proposed in this paper can take into account both the fairness of the scheduling while ensuring a satisfactory basic rescue effect, thereby considerably improving the satisfaction of the victims and reducing the loss resulting from the emergency. The findings can provide further emergency decision-making reference for decision-makers. [ABSTRACT FROM AUTHOR]

In this paper, a algorithm of the partial derivatives of the generalized hypergeometric function pFq ( ã; ˜b; z ) is obtained, where ã = {a1, a2, . . ., ap}, ˜b = {b1, b2, . . ., bq}. Moreover, we compare some algorithms of calculating the partial derivatives of pFq ( ã; ˜b; z ) . Numerical examples show the algorithm given in this paper improves the precision and accelerates the calculation of partial derivatives of the generalized hypergeometric function. Furthermore, we obtain some applications of the partial derivatives of the generalized hypergeometric function in calculating improper integrals and the derivative of the special function, such as the derivatives with respect to the order of Bessel function etc. The accuracy and the speed of calculating the improper integrals are improved by numerical examples. [ABSTRACT FROM AUTHOR]

At present, artificial bee colony algorithm (ABC) is one of the hot issues in swarm intelligence algorithm. Since it was proposed, people have done a lot of improvement work for ABC algorithm. To address the shortcomings of ABC, an improved ABC guided by experience (named as EABC) is proposed in this paper. In EABC, it collects the experience of individual improvement caused by dimension change in the iterative process, and selects the dimensions to change according to a ratio when a new position needs to be generated. In this way, the individual can choose a good direction to improve its quality. Numerical experiments show that EABC has a better performance. [ABSTRACT FROM AUTHOR]