Learning-based fault localization has been intensively studied recently. Prior studies have shown that traditional Learning-to-Rank techniques can help precisely diagnose locations using various dimensions of fault-diagnosis features, such as suspiciousness values computed by off-the-shelf techniques. However, with the increasing features considered advanced techniques, it be quite challenging for algorithms to automatically identify effective existing/latent features. In this work, we...

Online reinforcement learning often suffers from slow convergence and faults on early stages. In this paper, we propose a multi-grid method of Q-learning to handle these problems. It adopts coarse model fast converge good policy stages then fine further improve the optimization result. This approach is applied an optimal control problem energy conservation comfort HVAC in buildings. The algorithm implemented simulation using Matlab EnergyPlus. results show improvement our both performance...

The intermittency and uncertainty of the renewable energy deteriorate stability microgrids. In this article, we focus on a grid-connected microgrid with wind power battery storage system (BESS). electricity load is satisfied by from turbine, BESS, grid, together. purpose to reduce fluctuation grid charging or discharging BESS dynamically. A Markov chain used depict randomness measured variance power. Since measurement quadratic nonadditive, dynamic optimization problem does not fit standard...

Multi-attribute group decision making (MAGDM) is one of the most important research hotspots in field sciences. Many practical problems are often characterized by MAGDM. The aim this paper to develop a new approach for MAGDM problems, which attribute values take form picture fuzzy information, and information about weights attributes makers unknown. Firstly, some interaction operators presented, such as weighted averaging (PFWIA) operator, ordered (PFOWIA) operator hybrid (PFHOWIA) operator....

Unmanned aircraft systems (UAS) are essential components in the future air-combat. Due to high dynamics and randomness of aircrafts, traditional methods difficult solve optimal control strategy. The characteristics reinforcement learning (RL) match difficulty this problem. In paper, we build an air-combat game environment train agent with deep Q-learning (DQN). Despite increasing probability loses slightly, our method performs much better than other algorithms simulations. Compared searching...

Organizational culture serves not only as a set of values and norms that characterize the way business is conducted within company but also critical determinant its success competitiveness. This article explores importance organizational basis for successful management. It examines role values, norms, traditions, practices shape company's in creating conducive environment development. Additionally, it investigates impact on efficiency performance enterprise, well employee motivation...

Unmanned SystemsAccepted Papers No AccessActive Anti-Disturbance Control Using Compensation Function Observer for QUAV Flight amidst Wind and PayloadsKuo Li, Guoyuan Qi, Kun Wang, Jianbing Hu, Xia Zengqiang ChenKuo Qihttps://orcid.org/0000-0002-0059-4879 Search more papers by this author , Wang Hu Li Chen https://doi.org/10.1142/S2301385026500275Cited by:0 (Source: Crossref) Next AboutFiguresReferencesRelatedDetailsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsRecommend...

This study introduces a non-contact, single-target method for real-time monitoring of dairy cow calving posture and behavior using the YOLOv8 model. In total, 600 videos were collected, from which 10,544 image samples extracted through frame-by-frame processing. Complete video recordings 86 cows (30 primiparous 56 multiparous) utilized to investigate changes in behavior. The model achieved excellent performance with precision (P), recall (R), mean average (mAP) 96.72%, 96.53%, 97.41%,...

In this technical note, we discuss the optimality properties of service rate control in closed Jackson networks. We prove that when cost function is linear to a particular rate, system performance monotonic w.r.t. (with respect to) and optimal value can be either maximum or minimum (we call it Max-Min optimality); When second-order derivative always positive (negative), which makes strictly convex (concave), such for maximization (minimization) problem minimum. To best our knowledge, most...

In this paper, we study the optimal control of service rates in a queueing system with Markovian arrival process (MAP) and exponential times. The rate is allowed to be state dependent, i.e., can adjust according queue length phase MAP. cost function consists holding operating cost. goal find that minimize long-run average total To achieve that, use matrix-analytic methods (MAM) together sensitivity-based optimization (SBO) theory. A performance difference formula derived, which quantify...

With the fast development of Chinese high speed railway, demand for maintenance railway equipment has increased rapidly. This paper studies an intelligent diagnosis method high-speed turnout based on support vector machine. We focus study about methods generating feature and propose a rule-based that utilizes experts' experience knowledge. By experimenting current curve collected real turnouts, we compare advantages different generation demonstrate our performs best in fault diagnosis. work...

Abstract Energy efficiency of data centers (DCs) is great concern due to their large amount energy consumption and the foreseeable growth in demand digital services future. The past decade witnessed improvements DCs from an extensive margin—a shift small large, more efficient DCs. Improvements intensive margin, that is, operation, would be critical limiting environmental impact upcoming period. Machine learning algorithms have advantages optimizing operation improve as they shown potential...

This study investigates the optimization problem of an infinite stage discrete time Markov decision process (MDP) with a long‐run average metric considering both mean and variance rewards together. Such performance is important since indicates returns risk or fairness. However, couples at all stages, traditional dynamic programming inapplicable as principle consistency fails. We this from new perspective called sensitivity‐based theory. A difference formula derived it can quantify...

The maintenance problem with safety-critical components is significant for the economical benefit of companies. Motivated by a practical asset project, new joint replacement introduced in this paper. dynamics are modelled as Markov decision process, whose action space increases exponentially number asset. To deal curse dimensionality, we identify key property optimal solution: performance can always be achieved class policies which satisfy so-called shortest-remaining-lifetime-first (SRLF)...