The capacity regeneration phenomenon in lithium-ion batteries is inevitable and leads to non-monotonic fluctuations in capacity degradation trajectories, significantly complicating accurate remaining useful life (RUL) prediction. To address this challenge, this paper proposes a hybrid prediction model based on CEEMDAN-SSA-SVR-BiGRU. The method first employs Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to decompose the original capacity sequence into multiple Intrinsic Mode Functions (IMFs) representing local regeneration fluctuations, and a residual component (RES) referring to the global degradation trend, thereby achieving effective signal decoupling. Subseq... More >

Non-invasive continuous glucose monitoring (CGM) systems offer the advantage of non-invasive, real-time dynamic glucose monitoring, marking a significant advancement in diabetes management. However, the complexity of their sensing principles and operational mechanisms make systems vulnerable to various factors, which may introduce measurement bias or cause system interruptions and thereby compromise patient safety and monitoring effectiveness. To address these challenges, the Design Failure Mode and Effects Analysis (DFMEA) method is employed to identify and prioritize risks by assigning expert-based scores to critical components, ultimately enabling targeted improvements for high-risk failu... More >

Remaining Useful Life (RUL) prediction is critical for ensuring equipment safety and optimizing maintenance schedules, directly impacting system reliability and maintenance efficiency. However, in real-world industrial scenarios, factors such as operating condition fluctuations and load variations lead to inconsistent data distributions, making it challenging for existing models to achieve satisfactory adaptability and accuracy. To address this issue, this paper proposes a deep learning framework based on a multi-branch serial-parallel fusion of CNN-BiLSTM-Transformer architectures. Through innovative model architecture design and optimized training strategies, the framework aims to enhance... More >

This study introduces a novel approach for enhancing production decision-making by applying Reinforcement Learning to optimize the Economic Manufacturing Quantity (EMQ) model within discrete-time production-inventory systems. By incorporating machine status, inventory levels, and production choices, a Markov Decision Process (MDP) is constructed and combined with the Q-learning algorithm to derive an adaptive control method. This method enables the dynamic adaptation of production decisions, by effectively balancing the normal operation and shutdown for rest states. Numerical simulations show that the suggested Reinforcement Learning model surpasses conventional EMQ models and steady-state p... More >

Large-scale computing systems, such as cloud data centers, grid infrastructures, and high-performance computing clusters, are the backbone of modern information technology ecosystems. These systems typically consist of numerous heterogeneous, multi-state computing nodes that exhibit varying performance levels due to component failures, degradation, or dynamic resource allocation. Performability analysis, which integrates both system reliability and performance evaluations to quantify the probability of the system operating at a specified performance level, is critical for ensuring the efficient, reliable, and cost-effective operation of these complex systems. This paper presents a comprehens... More >