Fuel cells are considered as one of the most promising candidates for future power source due to its high energy density and environmentally friendly properties, whereas short lifespan blocks large-scale commercialization. In order enhance reliability durability proton exchange membrane fuel cell, a fusion prognostic approach based on particle filter (model-based) long-short term memory recurrent neural network (data-driven) is proposed in this paper. Both remaining useful life estimation...

Prognostic of the proton-exchange membrane fuel cell can effectively extend lifespan, which contribute to its large-scale commercialization. In this article, a hybrid prognostic approach is proposed predict output voltage and other aging parameters that reflect stack's internal degradation. During training stage, are obtained by using extended Kalman filter (EKF). Besides, used train long short-term memory (LSTM) recurrent neural network. prediction EKF LSTM method will parameters,...

In recent years, machine learning, especially deep has developed rapidly and shown remarkable performance in many tasks of the smart grid field. The representation ability learning algorithms is greatly improved, but with increase model complexity, interpretability worse. a critical infrastructure area, so models involving it must be interpretable order to user trust improve system reliability. Unfortunately, black-box nature most remains unresolved, decisions intelligent systems still lack...

This article proposes a fault-tolerant control method for an input-parallel–output-series (IPOS) converter under open-circuit switch failure, which mainly focuses on two parts: fault diagnosis (fault detection and identification) remedial action. The is realized based immersion invariant observer (I&IO), has strong robustness to parameter uncertainty external disturbances, therefore, it can be designed using only the crude model with nominal parameters. Moreover, sampling frequency required...

To optimize the voltage regulation performance and enhance robustness, an adaptive model predictive controller is proposed in this article for interleaved dc–dc boost converter. The constructed by linearizing nonlinear equations of converter at current operating point. A novel parameter update mechanism developed based on static model, which enables fast accurate identification parameters, thus control law can guarantee desired wide range. Based a disturbance observer to correct predictions...

The fuel cell output power depends highly on the random load profile, and converters play a key role in system. Robust control design of converter can help to online optimize diagnostic method for fuel-cell-based applications. In this article, an advanced algorithm dc–dc proton exchange membrane is realized through robustness based flatness active disturbance rejection (ADRC). Flatness track demand, ADRC estimate total required real time. effectiveness proposed verified two-phase interleaved...

Fault diagnosis is essential for the stable and efficient operation of proton exchange membrane fuel cell (PEMFC) system. However, manifold balance plant (BOP) components coupling phenomenon involving multiple physical fields will significantly increase probability system fault, which makes it difficult to realize a timely effective diagnosis. In this study, novel online method an open-cathode PEMFC proposed, only based on output voltage measurements, both normal state fault states caused by...

This article proposes an easily implementable and computationally effective model predictive control (MPC) scheme for fuel cell dc–dc boost converter. The proposed contains explicit MPC method just with a single-step horizon. Thus, the computational burden can be greatly reduced. To further enhance its antidisturbance ability, observer technique is adopted to, respectively, estimate load impedance input voltage models. Since estimated variables directly sent to cost function, satisfactory...

Fuel cells are considered as the preferred future power source due to their environmental friendliness and high efficiency, whereas short lifespan cost hinder large-scale commercialization. cell prognostic can contribute prolonging fuel life reducing overall cost, it has attracted research attention recently. However, most of methods treat measured voltage health indicator thus only be applied that works under constant current condition. To handle frequent load change condition, this article...

The increasing environmental issues such as air pollution and energy scarcity of fossil fuels require the acceleration electrification in various fields, especially for transportation. According to this development trend, fuel cell systems gradually become a potential alternative traditional power high efficiency, density, zero pollution, while relatively short lifetime has tremendously limited its large-scale commercial application. To extend lifespan, degradation predictive methods are...

The single-bus dc electrical power system is a promising architecture for more electric aircraft. However, the high penetration of constant loads (CPLs) will seriously threaten stability. To this end, in article, novel model predictive control (MPC) proposed to eliminate destabilizing effects CPLs and optimize transient performance systems over wide operating range. First, state coordinate introduced obtain canonical form model, newly defined output disturbed by load uncertainties. Then,...

Many machine learning models are naturally multitask, which may involve regression and classification tasks, in they can be trained by the multitask network to yield a more generalized model with aid of correlated features. When these deployed on Internet-of-Things devices, computation efficiency privacy data pose significant challenge developing federated (FL) algorithm for both higher performance better protection. In this article, new FL framework is proposed class problems hard...

Real-time License Plate Recognition plays a significant role in traffic congestion control and road safety monitoring. Practically, network camera may capture multiple license plates one frame while the data collected by different cameras cannot be shared due to privacy concern. In this paper, federated learning framework is introduced simultaneously detect over through semantic communication. Specifically, achieve high efficiency of recognition real time, segmentation model applied locally...

The power converters can play an important role in fuel cell systems, and robust control design of the improve both system efficiency stability. In this paper, advanced strategy a two-phase interleaved boost converter for applications is proposed implemented, which consists outer voltage regulation loop inner current tracking loop. designed based on Super-Twisting sliding mode to track its reference tightly. active disturbance rejection (ADRC) applied regulate output generate stability loops...

The development of the DC microgrid system has promoted circuit breaker. However, traditional breaker exists many problems such as long period fault interruption, complex structure, existing arc, low reliability and anti-interference. Aiming to solve these problems, an improved solid-state was proposed in this paper. This uses a transformer instead LC legs which exist Z-source breakers realize function turning off thyristor. working principle designing process are discussed paper according...

This paper presents an observer based power switch open-circuit fault diagnosis method for a two-phase interleaved boost converter (IBC). Motivated by the simplicity and robustness of immersion invariant (I&I) theory, I&I observers are adopted residual generation. The inductor current output voltage used in controllers selected input estimation, which avoids use extra sensors. Furtherly, to eliminate unexpected effects, caused parameter uncertainty, adaptive threshold is designed. simulation...

To improve the durability and efficiency, a novel energy management strategy (EMS) considering internal loss of fuel cell unmanned aircraft vehicle (UAV) hybrid power system is proposed. First, equivalent circuit model established to calculate its loss. The hydrogen consumption cells discharge behavior battery can be regarded as consumption. Then, total used first cost function, average efficiency considered second function. And Pontryagin's minimum principle (PMP) control applied achieve...

The power converter plays a crucial role in energy conversion systems, while advanced control algorithms can help stabilize the bus voltage and match demand. In this work, novel fast terminal sliding mode is proposed for boost converters to strengthen disturbance rejection ability. First, full-order manifold guarantee finite-time convergence of system states, an integral-type switching law deal with large disturbances. Second, estimation method used identify uncertainties theoretical...

The proton exchange membrane fuel cell(PEMFC) is a complex nonlinear dynamic system coupled with multiple physical fields. To analyze the external characteristics of cell system, it regarded as black box and neural network structure used to identify model. This paper uses long short-term memory(LSTM) model PEMFC model's input current, output voltage. First, fit voltage current characteristic curve under certain operation conditions according empirical formula, train LSTM through data...

Distribution system state estimation (DSSE) plays a crucial role in the real-time monitoring, control, and operation of distribution networks. Besides intensive computational requirements, conventional DSSE methods need high-quality measurements to obtain accurate states, whereas missing values often occur due sensor failures or communication delays. To address these challenging issues, forecast-then-estimate framework edge learning is proposed for DSSE, leveraging large language models...