: Increasing environmental variability and operational complexity in reservoir systems necessitate advanced optimization frameworks for flood control. This study proposes the ε-constrained Improved Zebra Optimization Algorithm (ε-IZOA), a novel metaheuristic algorithm integrating an enhanced (ZOA) with adaptive ε-constraint handling, to address multi-reservoir control optimization. Three strategic modifications advance standard ZOA: (1) Bernoulli chaotic mapping diversified population...

Unlike traditional omni-directional sensors that always have an omni-angle of sensing range, directional may be able to switch several directions and each direction has a limited angle range. Area coverage is still essential issue in sensor network which consists number sensors. In this paper, we study novel "area by sensors" problem. We propose the maximum area (MDAC) problem maximize covered scheduling working network. prove MDAC NP-complete one distributed algorithm for MDAC. Simulation...

In this paper, based on the existing sample data, give normal and lognormal distribution group drivers reaction time separately, obtained majority drivers' perceived within 1.5s. Utilize research results of cognitive psychology, set up linear model under hypothesis that driver's vision receive external emergent stimulation (interference), does not consider difference situation, think interference degree driver have something to do with angle stimulators previous car, slope is 0.025.

In this paper, the technique of image recognition algorithm is used to conduct an in-depth study and analysis intelligent classification recycling system solid waste optimize design its system. The network structure detection principle YOLO target based on convolutional neural nets are analysed, images construction collected as a dataset, dataset expanded using data enhancement techniques, objects in labelled train their own models. To facilitate testing algorithm-based Using for...

Off-grid algorithms for direction of arrival (DOA) estimation have become attractive because their advantages in resolution and efficiency over conventional ones. In this paper, we propose a grid reconfiguration (GRDOA) method based on sparse Bayesian learning. Unlike other off-grid methods, the points GRDOA are treated as dynamic parameters. The number position varied iteratively via root fission process. Then, gets reconfigured through some criteria. By updating grid, DOAs estimated...

As the core foundation for naval strategic transformation and mission accomplishment, command control system is a complex electronic information involving multiple elements, such as control, intelligence reconnaissance, early warning detection, communication navigation, countermeasures, integrated support, combat personnel. With of tasks, has gradually evolved from traditional single to multi-functional system, which raised number problems challenges reliability design system. This paper...

In this paper, we present a family of circular or square optimal polynomial filters for prefiltering two-dimensional polygons and images. The criterion designing is to maximize the energy concentration within period spectra filters. are nonnegative can have arbitrary radius order. For given radius, converge very fast when order increases, making low-order suffice high-quality prefiltering. With filters, it convenient evaluate integral over parts filter mask with closed-form solutions,...

Given the huge capacity of solar photovoltaic installed, operation and maintenance plants became a complicated issue. An automatic defects detection method with aerial images for is proposed. The solved two technical problems: (1) an adapted generative adversarial network designed to augment size limited visible images; (2) Considering quality augmented images, algorithm presented remove poor quality; (3) deep learning-based model proposed classify modules images. system can significantly...

In order to provide scientific maintenance basis for the state of relay protection device, an improved method that PSO optimizes SVM (Support Vector Machines) parameters is proposed, and this applied evaluation device. Firstly, according operational data collected by acquisition unit, preprocessed form a sample database, at same time, used optimize kernel function parameter error penalty factor. And PSO-SVM model established evaluate problem. By calculating distance from point hyperplane, it...

Status evaluation of transformer insulation aging is an important issue. The key to solving the problem find electrical or physicochemical characteristic features method which can effectively reflect status aging. Usually, experimental tests are conducted achieve this target. Although testing directly indicate loss life, it expensive and restricted by actual working conditions. This mainly used for power transformers in main networks, but difficult be widely distribution (DT) due larger...

In the cigarette production process, detecting defects in outer packaging is a crucial step. To address issue of high false recognition rates factory environments, we propose an improved YOLOv7-tiny target detection algorithm. Our algorithm based on network structure and enhances feature extraction location by adding SA module at end Backbone. Additionally, integrate GSConv into Neck part to improve information flow fusion ability. experimental results show that increases average accuracy...

A power-aware job scheduling algorithm is proposed to increase the energy efficiency of high performance computer by selectively hibernating some idle nodes during periods low load. The records switching logs nodes' power states, dynamic setting sleeping status. Detailed experimentation using traces from Parallel Workloads Archive indicates that can achieve effective savings with good control on frequency nodes'power states.

According to the site layout of cigarette industry and requirement AGV soft time window for waste tobacco recycling, change demand recycling based on ant colony algorithm is proposed. The method smoke recovery centered unit station, Considering influence actual production rhythm complex Workshop time, membership function when arrives at station used characterize satisfaction timeliness business. Therefore, with average arrival as constraint condition minimum cost objective, an optimization...

The station is the basic production unit of railway transportation. Sorting capacity an important part transport capacity. Based on field data, sorting south Xichang district during peak period (February), general (July), and idle (June) were calculated respectively in this paper by using utilization rate method. And basis quantitative calculation, qualitative analysis was also completed. In addition, considering real throughput demand, three kinds strategies proposed. Obviously, each...

This paper proposes DSHA, a dynamic self-healing approach to handle exceptions and failures in composite services processes. DSHA will choose the optimal exception handling strategy when arise during process execution. meets functional requirements of for service failures, it also better satisfies time constraint critical applications. Simulated experiments show our is feasible.

In order to improve the precision and reliability of prediction deformation monitoring data, radial basis function artificial neural network is used in data processing. The result this method compared with BP methods, it concluded that through better can be obtained.

The construction of transportation infrastructure develops rapidly in China. However, the number traffic accidents is rising continuously. On basis more than one million road accident records China from 2000 to 2001, C-mean fuzzy clustering method adopted explore time distribution accidents, direct economic losses and death, this paper. So has a close relationship with time.

Abstract AGV is an important material handling equipment for flexible, lean and intelligent production of smart factory. Because the high cost car congestion vehicles in system, this chapter studies waste smoke recovery path planning task scheduling strategy multi system to improve efficiency AGV.

Generative Adversarial Network (GAN) has an excellent performance in " Image-to-Image Translation" field, this paper proposes effective algorithm to solve the problem of image colored-pencil-style transformation based on Pix2Pix GAN. The improved generator and discriminator are designed be suitable for high resolution generation. In loss function, TV Loss is using improving smoothness images removing noise overlapping shadows. order adapt practical application requirement transformation, our...