Minimal Performance Degradation Research Articles

Understanding Mechanisms and Reversibility of Anode Degradation in PEM Water Electrolysis

Water electrolysis is an important route toward sustainable industrial production of hydrogen, as well as mitigation of the CO2 emissions associated with current hydrogen production from natural gas. The industrial deployment of clean electrolytic hydrogen will depend on long-term electrolyzer durability under the intermittent operating conditions associated with coupling with renewable sources of electricity. Thus, understanding cell degradation mechanisms is required to develop mitigation strategies and minimize performance degradation over extended periods of operation.Here, we investigate the effect of temperature and intermittent conditions on proton exchange membrane (PEM) water electrolyzer performance over a variety of timescales. Performance losses consisting of both reversible losses and irreversible degradation accelerate quickly during holds at high current densities and low temperatures, reaching high cell voltages within several hours. However, the cell can be returned to near beginning-of-test performance by interrupting these current holds with a variety of intermittent states of operation, such as interruption of the applied current or operation at higher temperatures. Reversible losses can harm the overall cell efficiency without being captured in short-term polarization curves, and may also contribute to stressors causing long-term degradation. The impact of different shut-down and idle states on the catalyst layers and cell degradation will also be discussed. The presented insights could contribute to the development of accelerated stress test protocols as well as performance recovery strategies for long-term operation of PEM water electrolyzers.

Multifunctional daytime radiative cooler resistant to UV aging

Passive radiative cooling presents a promising approach to mitigate global warming and reduce energy consumption. Existing coolers, however, face challenges in practical prolonged adoption due to their unappealing design, susceptibility to environmental pollutants, as well as degradation of polymers by ultraviolet (UV) irradiation from sunlight. This study introduces a multifunctional daytime radiative cooler (MDRC) with a dual-layered structure, combining practical functionality with aesthetic value. Its base layer consists of an Ag–TiO2–Ag thin film stack, imparting vibrant cyan, magenta, and yellow colors, even maintaining visually appealing from various incident angles, up to 75°, to ensure consistent color display. The top layer, comprising S–SiO2 nanospheres with UV resistance rather than polymers, not only forms a superhydrophobic surface with a contact angle of 155.502°, enhancing the self-cleaning capability of the MDRC against contamination and dust, but also renders the MDRC with extraordinary anti-aging capacity. Outdoor experiments has shown that the MDRC maintains operation at temperatures above ambient under direct solar irradiation, achieving a substantial temperature reduction of 6.82 °C compared to the same colored commercial coating. After three months of outdoor exposure, the MDRC demonstrated minimal degradation in performance, with decreases of only 0.29 °C during the day and 0.12 °C at night, underscoring its superior UV-resistant aging properties. This MDRC represents a robust solution for diverse durable outdoor raidative cooling applications, advancing toward real-world utilization.

Unveiling the enhanced hydrogen evolution performance of ternary heterostructures Mo5O14-Y(OH)3-CoMoO4/CF electrocatalysis

The challenge in electrocatalysis lies in finding stable and efficient catalysts for hydrogen evolution, especially those based on non-noble metals. Herein, the Mo5O14-Y(OH)3-CoMoO4/CF (CF stands for cobalt foam) catalyst rich in ternary heterostructures have been successfully synthesized. The catalyst is composed of cumin-like nanoparticles randomly dispersed on microcolumns. This unique structure ensures both its electrocatalytic activity and structural stability. Electrochemical tests demonstrate exceptional water splitting capabilities, with overpotentials of just 40 mV at 10 mA cm−2 (j10) and 219 mV at 1000 mA cm−2 (j1000). Furthermore, the catalyst shows minimal performance degradation after long-term testing, maintaining stable electrochemical performance for 60 h at j10 and 18 h at j800. Surface energy calculations and scanning electron microscopy (SEM) indicate that yttrium(Y) promotes the formation of cumin-like nanoparticles. Density functional theory (DFT) analysis, along with in situ Raman spectroscopy, suggests that the interaction among the three metals forms Mo5O14-Y(OH)3-CoMoO4 ternary heterostructures, effectively reducing the energy barrier for the hydrogen evolution reaction (HER). Notably, the Y sites in the CoMoO4-Y(OH)3 heterostructure exhibit the lowest hydrogen adsorption free energy (ΔGH⁎: −0.31 eV), making them the primary active sites for Mo5O14-Y(OH)3-CoMoO4/CF.

Advanced Nix/MoSx/MOF-2@g-C3N4 carbon nanostructures for the effective eradication of the Methylene blue dye

In this research, we investigated different hybrid photocatalysts: Ni/Mo.S2/MOF-2@g-C3N4, Mo.S2/MOF-2@g-C3N4, Ni.S2/MOF-2@g-C3N4 and Ni/Mo.S2/MOF-2. These photocatalysts were synthesized using a solvothermal method by incorporating Ni.x/Mo.Sx, the heterojunction showed improved separation of photoinduced charges due to disparity in charge transfer. This resulted in a narrower energy band gap and excellent catalytic activity in photodegradation. The Ni/Mo.S2/MOF-2@g-C3N4 heterostructure demonstrated outstanding photocatalytic efficiency, with up to 91% degradation of methylene blue (MB). This significant degradation was driven by the generation of superoxide (O2 •) and hydroxyl (•OH) radicals. The process adhered to pseudo-first-order kinetics and was effectively completed within 90 min of light exposure. Remarkably, the Ni/Mo.S2/MOF-2@g-C3N4 heterostructure displayed excellent stability, with only minimal performance degradation observed after five consecutive cycles. These results underscore the superior stability of the heterostructure. In summary, our research confirms that the Ni/Mo.S2/MOF-2@g-C3N4 heterojunction photocatalyst is a highly efficient material for the effective removal of methylene blue dye.

Healable and Conductive Two-Dimensional Sulfur Iodide for High-Rate Sodium Batteries.

Self-healing functional materials can repair cracks and damage inside the battery, ensuring the stability of the battery material structure. This feature minimizes performance degradation during the charging and discharging processes, improving the efficiency and stability of the battery. Here, we have developed a novel healing conductive two-dimensional sulfur iodide (SI4) composite cathode. This process integrates both sulfur and iodine compounds into carbon nanocages, forming a SI4@C core-shell structure. This cathode design improves electrical conductivity and repairability, facilitates rapid activation, and ensures structural integrity, resulting in a typical Na-SI4 battery with high capacity and an exceptionally long cycle life. At 10.0 A g-1, the capacity of the Na-SI4 battery can still reach 217.4 mAh g-1 after more than 500 cycles, and the capacity decay rate per cycle is only 0.06%. In addition, the cathode exhibits a cascade redox reaction involving S and I, contributing to its high capacity. The in situ growth of a carbon shell further enhances the conductivity and structural robustness of the entire cathode. The flexibility and bendability of SI4@C-carbon cloth make it applicable for flexible electronic devices, providing more possibilities for battery design. The strategy of engineering a two-dimensional self-healing structure to construct a superior cathode is expected to be widely applied to other electrode materials. This study provides a new pathway for designing novel binary-conversion-type sodium-ion batteries with excellent long-term cycling performance.

Development of a high-capacitance flexible supercapacitor with enhanced cycling stability based on nanotubular polyaniline-modified titanium sheet

Nanotubular polyaniline modified titanium (PANI/Ti) sheet was synthesized via one-step chemical polymerization of aniline on a Ti sheet. The structural and morphological investigations of the PANI/Ti sheet were performed using X-ray diffraction, field emission scanning electron microscopy, and X-ray photoelectron spectroscopy. These investigations revealed that a nanotubular PANI network with an average diameter of 37.5 nm was vertically aligned on the Ti sheet. Electrochemical properties of PANI/Ti were investigated via cyclic voltammetry, galvanostatic charging/discharging, and electrochemical impedance spectroscopy in a three-electrode system. The PANI/Ti sheet exhibited an exceptional specific capacitance (Csp) of 1174 mF/cm2 at 2 mA/cm2, which was significantly higher than that of the PANI/GC (199.53 mF/cm2) and the bare Ti sheet (0.87 mF/cm2). The prepared PANI/Ti sheet exhibited a power density of 21.99 mW/cm2 at an energy density of 3.05 mWh/cm2. Moreover, the electrode demonstrated outstanding cycle stability, maintaining 97.62% efficiency over 3000 cycles and exhibited impressive flexibility, with minimal performance degradation even when bent at angles up to 90°. Furthermore, PANI/Ti sheet significantly reduces the solution and charge transfer resistance compared to bare Ti sheet. So PANI-modified Ti sheet will be a good choice for the fabrication of new flexible energy storage as well as electronic devices in the future.

Anti-fouling rotating polymer-based heat exchanger for zero liquid discharge humidification-dehumidification desalination

The objective of this investigation is to assess the performance of various heat exchangers for application in a novel solar-powered zero liquid discharge humidification-dehumidification desalination system. In this study four heat exchangers (HX) having two different flow configurations namely counter flow (cf), cross flow (cr) made up of three different materials namely high-density polyethylene (HDPE), aluminum (Al), and Polypropelyne (PP) were compared in terms of their effectiveness and overall heat transfer coefficient under varying salinity levels (up to 10%) and mixing ratios (0.22–0.45). At a mixing ratio of 0.22 and 0% salinity, PP-HXcr and Al-HXcf exhibited similar effectiveness (∼85%), surpassing that of HDPE-HXcf (∼65%). Despite PP-HXcr's lower thermal conductivity in comparison to Al-HXcf, comparable effectiveness was achieved due to the superior flow distribution in PP-HXcr. Further investigations focused on the impact of salinity on heat exchanger performance. At 3.5% salinity, all heat exchangers experienced a decline in effectiveness and heat transfer coefficient (HTC), with Al-HXcf experiencing a more pronounced decrease compared to PP-HXcr. The higher thermal conductivity of Al-HXcf led to greater salt accumulation, while PP-HXcr demonstrated minimal fouling. As the experiment progressed, fouling increased for all heat exchangers, with the Al-HXcf being practically ineffective at 10% salinity with an effectiveness below 10%. To address the issue of fouling, a rotating cross-flow heat exchanger (RPP-HXcr) was introduced. While the effectiveness of the PP-HXcr drops from 85% to approximately 60% with increasing salinity from 0% to 10%, the RPP-HXcr demonstrates only a marginal decline in effectiveness with increasing salinity. For instance, at mixing ratio of 0.22 when the salinity is increased from 0% to 10%, the effectiveness of RPP-HXcr only drops from 83% to 77%. This exceptional performance was attributed to the continuous contact between the rotating tubes and the incoming feed, effectively preventing fouling and ensuring sustained efficiency. Rotating cross-flow heat exchanger (RPP-HXcr) is introduced and validated as a potentially reliable solution for mitigating fouling, as it demonstrates sustained efficiency and minimal performance degradation across different salinity conditions.

Soft independence guided filter pruning

Filter pruning (FP) is an effective method for reducing the computational costs of convolutional neural networks, and herein, the most critical task involves evaluating the significance of each convolutional filter and eliminating the less important ones while minimizing performance degradation. Most existing FP methods consider only local information, which may prevent them from accurately recognizing the most important filters. To address this limitation, we propose the soft filter independence (SFI) method, which leverages global information to identify the most important filters using their magnitude and correlation information in different functional layers. The SFI criterion measures the replaceability of filters from a global perspective in a network. Filters with low independence can be represented effectively by others, so their information can be accurately conveyed by other filters. In addition, we introduce a novel SFI-based asymptotic pruning ratio, which improves training and pruning stability. Compared to the most advanced FP methods, our method enables CNNs to achieve higher pruning rates and better classification performance.

An Agile Super-Resolution Network via Intelligent Path Selection

In edge computing environments, limited storage and computational resources pose significant challenges to complex super-resolution network models. To address these challenges, we propose an agile super-resolution network via intelligent path selection (ASRN) that utilizes a policy network for dynamic path selection, thereby optimizing the inference process of super-resolution network models. Its primary objective is to substantially reduce the computational burden while maximally maintaining the super-resolution quality. To achieve this goal, a unique reward function is proposed to guide the policy network towards identifying optimal policies. The proposed ASRN not only streamlines the inference process but also significantly boosts inference speed on edge devices without compromising the quality of super-resolution images. Extensive experiments across multiple datasets confirm ASRN’s remarkable ability to accelerate inference speeds while maintaining minimal performance degradation. Additionally, we explore the broad applicability and practical value of ASRN in various edge computing scenarios, indicating its widespread potential in this rapidly evolving domain.

SIM-FED: Secure IoT malware detection model with federated learning

Many IoT devices are presently in use without sufficient security measures. The vulnerability of these devices to malware highlights the necessity for effective methods to identify malicious activity while protecting data privacy. Federated Learning is a collaborative learning technique that can create a fair model while keeping the local data private. To this end, this paper presents the SIM-FED, a novel model for detecting malware in IoT devices. This model utilizes both deep learning and federated learning algorithms. The SIM-FED offers a privacy-preserving solution for IoT malware detection, without sharing data, enhancing security in distributed environments. The model utilizes a lightweight one-dimensional CNN with optimized hyperparameters, reducing preprocessing, and computational overhead. Various federated aggregation strategies are evaluated, and the FedAvg strategy is selected to integrate the results of local models in the proposed model. Moreover, the model's resilience against white-box and black box cyber-attacks is assessed, revealing minimal performance degradation and highlighting its robustness, which is often overlooked in previous studies. The performance of the SIM-FED is assessed by conducting a series of experiments using the IoT-23 dataset. The results show that the SIM-FED outperforms other deep learning and federated learning models in terms of all evaluation metrics, achieving a remarkable 99.52 % accuracy.

What Makes Quantization for Large Language Model Hard? An Empirical Study from the Lens of Perturbation

Quantization has emerged as a promising technique for improving the memory and computational efficiency of large language models (LLMs). Though the trade-off between performance and efficiency is well-known, there is still much to be learned about the relationship between quantization and LLM performance. To shed light on this relationship, we propose a new perspective on quantization, viewing it as perturbations added to the weights and activations of LLMs. We call this approach ``the lens of perturbation". Using this lens, we conduct experiments with various artificial perturbations to explore their impact on LLM performance. Our findings reveal several connections between the properties of perturbations and LLM performance, providing insights into the failure cases of uniform quantization and suggesting potential solutions to improve the robustness of LLM quantization. To demonstrate the significance of our findings, we implement a simple non-uniform quantization approach based on our insights. Our experiments show that this approach achieves minimal performance degradation on both 4-bit weight quantization and 8-bit quantization for weights and activations. These results validate the correctness of our approach and highlight its potential to improve the efficiency of LLMs without sacrificing performance.

A lightweight multi-task learning network based on key area guidance for counterfeit detection

Counterfeit detection traditionally relies on manual efforts, but manual detection efficiency is notably low. The accuracy of deep learning methods is challenging because of the insufficient samples, so it is crucial to allow the model to learn effective representation at a lower training cost. Given the above problems, we proposed a lightweight multi-task learning method that employs an uncomplicated auxiliary task to enhance the main task’s attention and reduce the training sample requirements. A key area guidance algorithm is designed to construct the auxiliary task, disturbing key image areas to generate new samples and training the auxiliary task to recognize the disturbance. This guides the main task in discerning authenticity from these key areas. Additionally, a tailored data preprocessing strategy was designed to improve the method’s performance further. Achieving an impressive 98.8% accuracy in identifying various counterfeiting points, our method outperforms existing advanced methods. Importantly, the method significantly reduces training costs. Even with an 80% reduction in the sample size, the method maintains a 92.1% accuracy, demonstrating minimal performance degradation compared to alternative methods.

Structural effects of asymmetric magnet shape on performance of surface permanent magnet synchronous motors

This study proposes a novel surface permanent magnet synchronous motor (N-SPMSM) structure, which features asymmetric magnets attached to the rotor surface. The N-SPMSM structure exhibits reduced structural complexity and minimal electromagnetic performance degradation. The properties of N-SPMSM are analyzed by comparing its structural complexity (in terms of the shape) and manufacturing complexity and electromagnetic performance [in terms of the cogging–mutual torque ratio and back-electromotive force (EMF) values], with the corresponding values of a ring-type SPMSM (R-SPMSM) and step-skew-type SPMSM (T-SPMSM). The analysis results demonstrate that N-SPMSM has lower shape complexity than T-SPMSM and lower manufacturing complexity than both R-SPMSM and T-SPMSM. The cogging torque reduction and back-EMF performances of N-SPMSM are similar to that of R-SPMSM and T-SPMSM.

Composite aerogel incorporating low temperature phase change microcapsules for enhanced thermal insulation

Cryogenic transportation and storage confront significant challenges from harsh weather conditions, heightened energy consumption, and epidemic situations, compelling the need for the creation of exceptionally efficient thermal insulation materials. To address this demand, a composite phase change aerogel was designed in this study through incorporating low-temperature microencapsulated phase change microcapsules (MPCM) into a cellulose nanofiber/polyvinyl alcohol (CNF/PVA) system. The MPCM, consisting of a polyurethane-acrylate (PUA) shell and an n-tetradecane core, exhibited excellent encapsulation performance with leak-proof capability. Remarkable low-temperature phase change energy storage properties were observed, including a phase change temperature of approximately 6 °C and an impressive phase change enthalpy of 112 J/g. The MPCM also demonstrated stability during successive heating–cooling repetitions, maintaining its heat storage capacity and morphology for at least 300 cycles. These exceptional thermal characteristics endowed the resulting aerogel with effective thermal insulation and temperature retardation abilities. Meanwhile, the integration of CNF/PVA as the matrix in the composite aerogel led to minimal degradation of thermal storage performance compared to pure MPCM. Moreover, the addition of MPCM significantly enhanced the compressive strength, reaching 5.6 times that of the neat CNF/PVA aerogel. The composite aerogel showed a notably low density of 0.165 g/cm3 and could be reshaped through heating. This work provides a simple yet effective idea for designing bulk materials with low-temperature phase change capabilities, offering promising prospects in the field of thermal insulation.

Improved chaos grasshopper optimizer and its application to HRES techno-economic evaluation

Current political and economic trends are moving more and more toward the use of renewable and clean energy as a result of rising energy use and diminishing fossil fuel supplies. In this paper, an improved chaos-based grasshopper optimizer used for techno-economic evaluation in integrated green power systems is investigated. The integrated system consists of a fuel cell system, a wind farm, and solar energy. The integrated solar, wind, and hydrogen fuel cell architectures increase the effectiveness and electrical output of the system while needing less energy storage in structures that are unconnected from the grid. The grasshopper optimization technique and chaos theory have been combined to create the suggested chaotic grasshopper optimizer in this study. The performance, precision, and robustness of this optimization were then assessed, using four benchmark tasks. The ICGO model is utilized to assign suitable ratings to all system devices, thereby guaranteeing the attainment of optimal performance and efficiency. The Net Present Cost (NPC) analysis revealed that the ICGO algorithm attained the lowest minimum NPC value of 274.541E4 USD and the highest maximum NPC value of 311.94E4 USD. The average NPC value of the ICGO algorithm (289.176E4 USD) was found to be comparable to the other algorithms examined in the study. These findings indicate that the ICGO algorithm outperformed other optimization algorithms in minimizing the cost of the renewable energy system. The chaotic grasshopper optimizer can handle several targets, restrictions, and variables with ease, and the results demonstrate that it is substantially more efficient and precise than standard optimization techniques. It is also quite durable, with minimal performance degradation as compared to the benchmark solutions. This study demonstrates the effectiveness of the chaos grasshopper optimizer as an HRES technique.

Amine-functionalized Schiff base covalent organic frameworks supported PdAuIr nanoparticles as high-performance catalysts for formic acid dehydrogenation and hexavalent chromium reduction

Formic acid (FA) holds significant potential as a liquid hydrogen storage medium. However, it is important to improve the reaction rates and extend the practical applications of FA dehydrogenation and Cr(VI) reduction through the development of efficient heterogeneous catalysts. This study reports the synthesis of a uniformly dispersed PdAuIr nanoparticles (NPs) catalyst loaded with amine groups covalent organic frameworks (COFs). The alloyed NPs demonstrated exceptional effectiveness in FA dehydrogenation rate and Cr(VI) reduction. The initial turnover of frequency (TOF) value for FA dehydrogenation without additives was 9970 h−1 at 298 K, the apparent activation energy (Ea) was 30.3 kJ/mol and the rate constant (k) for Cr(VI) reduction was 0.742 min−1. Additionally, it showcased the ability to undergo recycling up to six times with minimal degradation in performance. The results indicate that its remarkable catalytic performance can be attributed primarily to the favorable mass transfer attributes of the aminated COFs supports, the strong metal-support interaction (SMSI), and the synergistic effects among the metals. This study offers a novel perspective on the advancement of efficient and durable heterogeneous catalysts with diverse capabilities, thereby making significant contributions to the fields of energy and environmental preservation.

Numerical roll motion control by using fins based on the linear quadratic regulator and dynamic mode decomposition

Reducing the roll motion and acceleration of a ship is a critical concern for the safety and comfort requirement in a seaway. In this paper, the uncontrolled roll motion time series is obtained under the harsh beam sea simulation for a naval vessel using a single degree of freedom nonlinear equation of motion. Then the linear quadratic controller (LQR) controller is applied by utilizing one pair of fins (port and starboard) that were positioned amidship. The amplitude and rate saturations of the fins are taken into consideration in the simulations. The LQR controller is designed based on an analytical mathematical model, with the selection of appropriate controller parameters. Then, an estimation is made based on the dynamic mode decomposition method for the ship model by taking measurements from online states as if those are derived from the experimental measurements. Finally, the ‘estimated mathematical model’ is used for the LQR controller design. The results demonstrate the successfulness of the estimation process, with only minimal degradation in performance observed.

Impact of Oxygen Storage Components in Prototype Pd-Based Three-Way Catalysts under Exhaust Conditions Relevant to Propane Engines

With increasing concerns about global warming, the push for sustainable and eco-friendly fuels is accelerating. Propane, recognized as liquefied petroleum gas or LPG, has garnered research interest as an alternative fuel due to its notable advantages, including a high-octane rating, reduced greenhouse gas emissions, and potential cost-effectiveness. However, to realize its full potential as an alternative fuel it is essential to develop catalysts that efficiently handle emissions at low temperatures. In our research, we investigated three distinct palladium (Pd)-based three-way catalyst (TWC) formulations (PdRh, Pd-only, and Pd-OSC) to investigate the influence of typical TWC components rhodium (Rh) and oxygen storage components (OSC) in exhaust scenarios relevant to propane-fueled engines. Among these, the formulation containing oxygen storage components (Pd-OSC) showed the highest reactivity for both NO and C3H8 while minimizing performance degradation from hydrothermal aging (HTA). Notably, the temperature of 50% conversion (T50) for propane in the Pd-OSC fresh and HTA sample was lower by 30 °C and 13 °C, respectively, compared to the Pd-only sample, highlighting the role of oxygen storage materials in enhancing catalyst performance, even without dithering. Additionally, N2 physisorption showed that the Pd-OSC sample has a higher surface area and increased pore volume. This underscores the idea that OSC materials not only augment the catalyst’s porosity but also optimize reactant accessibility to active sites, thus elevating catalytic efficiency. In addition to evaluating performance, we further explored the performance and characteristics of the catalysts using catalytic probe reactions, such as water–gas shift and steam reforming reactions.

An implicit barotropic mode solver for MPAS-ocean using a modern Fortran solver interface

We demonstrate use of a modern Fortran solver interface to manage solver algorithms for an implicit barotropic mode solver in the Model for Predictions Across Scales-Ocean (MPAS-O). ForTrilinos, a Fortran interface to Trilinos that contains a large collection of solver capabilities written in C++, has been implemented in MPAS-O to provide access to a suite of linear solver options. By virtue of the simplified wrapper and interface generator (SWIG) automation tool that generates modern Fortran interfaces to C++ code, we were able to implement the Fortran solver interface in MPAS-O using a familiar Fortran coding style while minimizing performance degradation. The ForTrilinos solver interface is written within MPAS-O’s time stepping modules as a subroutine in conjunction with MPAS-O code. Applied to an idealized ocean and a high-resolution realistic ocean test case, parallel performance of ForTrilinos solvers is examined. It is found that parallel scalability of the ForTrilinos solvers is highly dependent on the number of global synchronization points per solver iteration in each iterative solver algorithm. ForTrilinos solvers perform best compared to the Fortran hand-crafted (FHC) solver when the amount of work per processor is large enough. However, parallel scalability is better with the FHC solver and so when the work per core is modest FHC outperforms ForTrilinos. The intercomparison between the ForTrilinos and FHC solvers reveals that this performance hit in the ForTrilinos solver mostly comes from the global synchronization process, while suggesting that the matrix-vector multiplication process in the FHC solver needs to be optimized for better performance.

Enhance membership inference attacks in federated learning

In Federated learning, models in training will unintentionally memorize detailed information about private data, and the aggregation process on the central server requires users to upload their model parameters, making the models still susceptible to membership inference attacks. However, the existing membership inference attacks in federated learning have been less effective. This paper proposes a new membership inference attack method in federated learning that utilizes data poisoning and sequence prediction confidence. By injecting toxic data, the model can remember the detailed information of specific classes in the target private dataset to the maximum extent. The privacy data detailed information of the target clients contained in the model will be represented through the output confidence vector. Afterward, we aggregate the confidence information obtained from multiple epochs in federated learning and utilize the AdaBoost classifier to learn the details from it. Finally, we use different thresholds to partition the predicted confidence scores output by the AdaBoost classifier obtaining the membership information. We conducted experiments on multiple datasets and models to validate the effectiveness of our attack method. The results showed a high attack effectiveness with minimal overall model performance degradation.