Battery Energy Storage Systems (BESS) are essential for advancing the transition to sustainable power systems, with lithium-ion batteries (LIB) widely adopted for their high efficiency and scalability. However, the electrochemical complexity of LIBs presents significant challenges in maintaining long-term safety and performance. This study introduces battery reconfiguration as an innovative system-level maintenance strategy within a benefit-oriented predictive maintenance (PdM) framework, aiming to improve both operational reliability and energy efficiency. A binary-state characterization model is proposed, where safety is inferred from internal resistance using a Hidden Markov Model (HMM), and performance is determined by capacity—together constraining the system’s energy output. To evaluate economic value, a comprehensive generation-side benefit model is developed, incorporating both peak shaving services and electricity market participation. The reconfiguration task is formulated as a large-scale combinatorial optimization problem (COP) and is efficiently solved using a tailored degradation-aware improved genetic algorithm (DA-IGA). Numerical experiments on a 10×20 BESS system demonstrate that the proposed strategy achieves a 14.73% increase in energy output, a 22.47% improvement in net benefit, and an 84.58% reduction in renewable energy curtailment compared to a non-reconfigured baseline.

Predicting power consumption for the Mars Express (MEX) mission is essential for optimizing its operational lifespan and mission assignments. However, the complexity of the Martian environment and the extended solar cycle obscure the periodicity of power consumption, making it difficult for existing methods to capture both intraperiodic and interperiodic features. This study introduces the bionic hierarchical learning network (BHL-Net) to enhance power consumption predictions. Leveraging 2-D frequency preprocessing and brain visual modeling techniques, BHL-Net mimics natural image encoding in the prefrontal cortex (PFC) to improve predictive performance. It incorporates a temporal oscillation activation module and a stripe intensity attention module to focus on local features, while a multihead attention adaptive aggregation module identifies key global features. Comparative experiments show that BHL-Net outperforms existing transformer-based models for MEX power consumption prediction. Ablation studies further validate the effectiveness of the FFT-based 2-D transformation and bionic attention framework. By emulating human brain response coding mechanisms, BHL-Net captures variations within and between complex cycles, providing a competitive solution for time series prediction in industrial applications.

Forecasting the power consumption of the spacecraft is critical for optimizing its lifespan and task allocation. However, the complex electromagnetic environment of outer space introduces unavoidable noise into the collected electrical signals. Moreover, the various subsystems of a multipower spacecraft are affected differently by internal and external noise, making it challenging for the existing methods to effectively capture the features of long-term power consumption sequences. We propose adaptive frequency-pruning-enhanced (AFPE)-iTransformer, a robust time-series forecasting model designed for spacecraft telemetry forecasting under noise and long-range dependency conditions. The model combines three key components: Legendre memory projection for historical compression, adaptive top-kfrequency pruning for per-channel denoising, and an improved inverted transformer for cross-subsystem attention. Evaluated on three years of Mars Express (MEX) data, our method consistently outperforms the state-of-the-art baselines in both within-year and cross-year forecasting. It also achieves competitive efficiency, with fast model load time and moderate parameter size. While focused on power forecasting, the model’s modular design supports broader applications in telemetry and industrial forecasting. Model code and configurations are open-sourced for reproducibility.

Accurate degradation modeling is a prerequisite for reliable prognostics. When historical data are scarce and operating conditions vary over time, conventional approaches struggle to balance accuracy, adaptability, and interpretability, and lack robustness to changing environments, especially when encoding the effects of operating conditions directly in the model. To overcome these limitations, this paper proposes a novel generalized degradation model based on a semi-physics-informed neural stochastic differential equation, where neural stochastic differential equation (NSDE) is utilized to describe the degradation dynamics. In contrast, the effect of operating conditions on degradation rate is injected in a plug-and-play prior without being locked into the NSDE structure. A variational inference-based generative training procedure jointly estimates the parameters of the NSDE and the prior, mitigating the adverse effect of imperfect physics and requiring only modest historical data. Then, an approximate closed-form distribution for the remaining useful lifetime (RUL) is derived. Thus, an approach for RUL prognostics of in-service products under dynamic operating conditions is established, leveraging the knowledge of degradation from historical data. Comprehensive studies on simulated and battery degradation data demonstrate the robustness and effectiveness of the proposed model.

Nowadays, the rise of operating leases has promoted the popularity of two-stage leasing in the aviation industry. That is, after the first lease expires, the aircraft will be subleased after system reconditioning. Due to the differentiated system reliability and customer requirements at two leasing stages, this leasing mode has posed new challenges to the two-stage maintenance service design. Especially, for service-oriented aviation manufacturers, the competition with independent maintenance providers (IMPs), the influence of component sales, and the maintenance capacity shortage have aggravated this difficulty. To address these challenges, this article proposes a novel approach for manufacturers to determine the optimal maintenance service strategy for two-stage leased systems under capacity constraints. Particularly, the reconditioning action after the first lease and the capacity investment are introduced into the collaborative strategy. To ensure both cost efficiency and competitiveness of the proposed strategy, a maintenance service competition mechanism is established considering two-stage differentiated customer utilities. Then, the time-varying maintenance demand accumulated from the two stages is dynamically predicted. Further, aiming at maximizing the profit from both maintenance services and component selling, two patterns addressing capacity shortage: independent investment and cooperation with an IMP are modeled. Finally, case studies validate the effectiveness of the proposed models and provide important managerial insights into pattern selection.

Degradation modeling and prognostics serve as the basis for system health management. Recently, various sensors provide plentiful monitoring data that can reflect the system status. A multitude of feature fusion techniques based on multisensor data have been proposed to generate a composite health index (HI) for prognostics, which can represent the underlying degradation mechanism. Most existing methods have used linear fusion models and neglected the practical requirements for HI construction, which are insufficient to reveal the nonlinear relations among features and difficult to obtain accurate HIs for complicated systems. This study proposes a novel feature fusion-based HI construction method with deep learning and multiobjective optimization. Multiple degradation features are fused by a deep neutral network (DNN). Several desired properties that the HIs should have for prognostics are adopted to formulate the objective functions of DNN training. To balance the spatial complexity and performance of the fusion model, a multiobjective optimization model is generated for training the DNN. Then, a generalized nonlinear Wiener process model is used to predict the remaining useful life with the resulted HIs. Finally, two cases are analyzed to illustrate the effectiveness and robustness of the proposed method.

The degradation data of highly reliable products are usually analyzed by stochastic process models, such as Wiener process, gamma process and inverse Gaussian process models. If such a specific degradation model is wrongly assumed, then poor analysis results of reliability assessment would be obtained. Therefore, a class of exponential-dispersion processes, named Tweedie exponential-dispersion process (TEDP), is proposed to describe the products’ degradation paths. The TEDP model which comprises the aforementioned stochastic processes as its special cases, is more flexible and applicable for degradation modeling. Considering the nonlinear characteristics of degradation paths and the unit-to-unit variability among the product units, random-effect models are established based on the nonlinear TEDP models with random drift and dispersion parameters. To improve the mathematical tractability of these models, the variational inference, expectation maximization algorithm and differential evolution algorithm are used to estimate the unknown model parameters. Furthermore, two nonlinear TEDP models with accelerated factors and random effects are developed for accelerated degradation analysis. Finally, a simulation study and three real applications are presented to show the effectiveness and superiority of the proposed models and methods.

Degradation modeling is an important method of reliability analysis for highly reliable products. The common degradation models are based on specific stochastic processes. This limits the widespread application of the modeling methods. A unified approach toward general degradation models is lacked. To address this issue, this article uses Tweedie exponential dispersion processes (TEDP) to establish degradation models. In such a way, the common stochastic processes turn out to be the special cases of TEDP. Then, the TEDP models can provide us more suitable models to describe the degradation paths and thereby improve the accuracy of reliability analysis. To develop the mathematical tractability of TEDP, we use the saddle-point approximation method to approximate the probability density function. Considering the unit-to-unit variability and imperfect observation, the TEDP model incorporated random effects and measurement errors are discussed. To illustrate the applicability and advantages of the TEDP models, we propose a Bayesian framework for the prognostic. A component-wise Metropolis-Hastings algorithm is developed to update the distributions of remaining useful life. Additionally, we also construct an optimization model under the constraint of budget for the accelerated degradation test planning by using TEDP. Finally, two case studies are presented to illustrate the proposed methods.