We present a new algorithm based on posterior sampling for learning in Constrained Markov Decision Processes (CMDP) in the infinite-horizon undiscounted setting. The algorithm achieves near-optimal regret bounds while...
详细信息
We present a new algorithm based on posterior sampling for learning in Constrained Markov Decision Processes (CMDP) in the infinite-horizon undiscounted setting. The algorithm achieves near-optimal regret bounds while being advantageous empirically compared to the existing algorithms. Our main theoretical result is a Bayesian regret bound for each cost component of Õ(DS√AT) for any communicating CMDP with S states, A actions, and diameter D. This regret bound matches the lower bound in order of time horizon T and is the best-known regret bound for communicating CMDPs achieved by a computationally tractable algorithm. Empirical results show that our posterior sampling algorithm outperforms the existing algorithms for constrained reinforcement learning. Copyright 2024 by the author(s)
Incomplete data poses challenges in accurately assessing structural health and detecting damage. It limits the ability to capture the complete behavior and response of the structure, which may hinder the identificatio...
详细信息
Incomplete data poses challenges in accurately assessing structural health and detecting damage. It limits the ability to capture the complete behavior and response of the structure, which may hinder the identification and localization of potential damage or anomalies. Addressing the issue of incomplete data requires developing strategies and algorithms that can effectively handle missing or limited measurements. Using incomplete and noisy measurements, we propose an optimization-based damage detection method for laminated composite plates with closely-spaced eigenvalues. The proposed method consists of two stages. In the first stage, the most probable defective elements are identified by utilizing condensed mode shapes as incomplete noisy inputs for modal residual vectors. This approach significantly reduces the computational effort for damage estimation. The second stage introduces an objective function based on incomplete and noisy Condensed Frequency Response Functions (CFRFs). To optimize the damage quantification, the Improved Particle Swarm optimization (IPSO) algorithm is employed to minimize errors in the proposed objective function based on CFRFs of damaged and intact laminates. The proposed method is demonstrated on laminated composite plates with different lamination schemes, ply orientations, and multiple damaged elements in different damage scenarios. By evaluating the method on numerical results and comparing it with previous studies, its superiority is demonstrated. Furthermore, the proposed method exhibits robustness to changes in mass distribution in the system investigated by retrofitting extra masses to the plate structures that lead to worsening closely-spaced eigenvalues.
Robust optimization (RO) is one of the key paradigms for solving optimization problems affected by uncertainty. Two principal approaches for RO, the robust counterpart method and the adversarial approach, potentially ...
详细信息
In response to the problems of low search efficiency and a large number of redundant points traversed by the traditional A∗ algorithm in the path planning process of AUV, this article proposes an improved A∗ algorithm...
详细信息
For evolutionary multi-objective optimization algorithms (EMOAs), an external archive can be utilized for saving good solutions found throughout the evolutionary process. Recent studies showed that a solution set sele...
详细信息
This paper proposes a novel nature-inspired meta-heuristic optimizer, called Reptile Search Algorithm (RSA), motivated by the hunting behaviour of Crocodiles. Two main steps of Crocodile behaviour are implemented, suc...
详细信息
This paper proposes a novel nature-inspired meta-heuristic optimizer, called Reptile Search Algorithm (RSA), motivated by the hunting behaviour of Crocodiles. Two main steps of Crocodile behaviour are implemented, such as encircling, which is performed by high walking or belly walking, and hunting, which is performed by hunting coordination or hunting cooperation. The mentioned search methods of the proposed RSA are unique compared to other existing algorithms. The performance of the proposed RSA is evaluated using twenty-three classical test functions, thirty CEC2017 test functions, ten CEC2019 test functions, and seven real-world engineering problems. The obtained results of the proposed RSA are compared to various existing optimization algorithms in the literature. The results of the tested three benchmark functions revealed that the proposed RSA achieved better results than the other competitive optimization algorithms. The results of the Friedman ranking test proved that the RSA is a significantly superior method than other comparative methods. Finally, the results of the examined engineering problems showed that the RSA obtained better results compared to other various methods. Source codes of RSA are publicly available at https://***/matlabcentral/fileexchange/101385-reptile-search-algorithm-rsa-anature-inspired-optimizer
This paper presents a developed energy management system (EMS) with a hierarchical three-level distributed control approach proposed for a photovoltaic/wind turbine/diesel generator with energy storage in an islanded ...
详细信息
This paper presents a developed energy management system (EMS) with a hierarchical three-level distributed control approach proposed for a photovoltaic/wind turbine/diesel generator with energy storage in an islanded hybrid DC microgrid (MG). In the hierarchical three-level control structure, the power to be transferred by the distributed energy sources (DER) is provided by controlling the converters in the primary control layer. The secondary control layer is used to achieve DC bus voltage restoration and increase current-sharing accuracy simultaneously. In the tertiary control layer, it is aimed to minimize the production cost by performing optimum load-sharing. For the operation of this control structure, a strategy is proposed to optimally manage the energy demand of the hybrid system autonomously under variable load conditions. The EMS with three different operating modes (Mode-I, Mode-II, Mode-III) is proposed to share power between DC MG sources optimally. In Mode-I, renewable energy sources (RES) are enabled to operate with MPPT operation. Energy consumption minimization strategy (ECMS)-based linear programming (LP), genetic algorithm (GA) and pattern search (PS) algorithms are used to minimize the fuel consumption and carbon emissions of the diesel generator (DG) in Mode-II operating Mode. In Mode-III operating mode, DG is not needed, and the power flow between DERs is optimized by using the fuzzy logic controller (FLC) and state machine control strategy (SMCS). The amount of fuel consumed, the cost of fuel consumption and ESS's SOC level are used as the EMS's performance criteria for the MG. The minimum fuel consumption of the DG is calculated as 2.328 L for the operation of LP-based ECMS, and the fuel cost, in this case, is found to be $3.26. Considering the charging of the ESS, the highest SOC level is achieved with 64.87% for the PS-based ECMS-SMCS combination. By considering fuel minimization and SOC level together, it can be seen that the best result
Detecting infeasible solutions is an important challenge in black-box Multiobjective Bilevel optimization (MOBO) due to a lower-level (LL) optimization problem used as a constraint (along with equality and inequality ...
详细信息
This research endeavors to advance energy efficiency (EE) within heterogeneous networks (HetNets) through a comprehensive approach. Initially, we establish a foundational framework by implementing a two-tier network a...
详细信息
This research endeavors to advance energy efficiency (EE) within heterogeneous networks (HetNets) through a comprehensive approach. Initially, we establish a foundational framework by implementing a two-tier network architecture based on Poisson process distribution from stochastic geometry. Through this deployment, we develop a tailored EE model, meticulously analyzing the implications of random base station and user distributions on energy efficiency. We formulate joint base station and user densities that are optimized for EE while adhering to stringent quality-of-service (QoS) requirements. Subsequently, we introduce a novel dynamically distributed opportunistic sleep strategy (D-DOSS) to optimize EE. This strategy strategically clusters base stations throughout the network and dynamically adjusts their sleep patterns based on real-time traffic load thresholds. Employing Monte Carlo simulations with MATLAB, we rigorously evaluate the efficacy of the D-DOSS approach, quantifying improvements in critical QoS parameters, such as coverage probability, energy utilization efficiency (EUE), success probability, and data throughput. In conclusion, our research represents a significant step toward optimizing EE in HetNets, simultaneously addressing network architecture optimization and proposing an innovative sleep management strategy, offering practical solutions to maximize energy efficiency in future wireless networks.
In this paper, a rational cooperative foraging based grey wolf optimizer (CFGWO) is proposed to enhance the performance of conventional grey wolf optimizer (GWO). It inherits the communicating behaviou...
详细信息
暂无评论