Manufacturing graphene-reinforced aluminum composites by laser powder bed fusion (LPBF) attracts expanding interests for the prospective significant strengthening of aluminum alloys allowed by the superior mechanical properties of graphene and the advantages of LPBF in rapid net-forming. However, the achievable strengthening contribution of graphene in LPBF Al/graphene composites remains obscure, as vastly varying performances of as-built Al/graphene composites have been reported for the processing complexities in mixing graphene platelets into Al alloy powders and subsequent LPBF. To unravel the contribution of graphene, AlSi7Mg alloy and AlSi7Mg/graphene (0.1 wt%) composite were prepared by LPBF and further treated by solution-and-ageing (SA) to dissolve the fine Al-Si eutectic networks along boundaries of cellular α-Al grains and the Si nanoparticles inside cellular α-Al grains which are the main strengthening sources in as-built AlSi7Mg alloy and AlSi7Mg/graphene composite. Microstructure characterization indicates no significant change induced by graphene addition in AlSi7Mg alloy both before and after SA treatment. Tensile tests on as-built AlSi7Mg alloy and AlSi7Mg/graphene composite show almost identical yield strength (260–270 MPa), ultimate tensile strength (440–445 MPa), and total elongation (14%-16%), while the results of SA-treated AlSi7Mg alloy (335 MPa) and AlSi7Mg/graphene composite (363 MPa) show an increment of ∼28 MPa in ultimate tensile strength but similar yield strength (260 MPa) along with slightly increased total elongation (from 13.4% to 16%), which demonstrates the strengthening contribution of graphene (0.1% wt.) in the strain-hardening of SA-treated AlSi7Mg/graphene composite. These observations would help to understand the strengthening effect of graphene in aluminum/graphene composites.

To address the high nonlinearity, strong coupling, parameter uncertainty, and external disturbances in attitude tracking control of carrier-based aircraft under complex marine environments, this study proposes a prescribed performancebased backstepping control method.A 6-DOF nonlinear dynamic model is adopted and transformed into affine form. Prescribed performance control is introduced to convert constrained tracking errors into an unconstrained form via performance functions and error transformation, thereby quantitatively bounding convergence rate and overshoot to ensure both transient and steadystate performance. Nonlinear disturbance observers are designed separately for the outer-loop attitude and inner-loop angular velocity subsystems to estimate and compensate composite disturbances in real time, improving robustness. Lyapunov stability analysis proves that all closed-loop signals remain bounded and the tracking error converges within the prescribed performance bounds.

Laser powder bed fusion (LPBF) enables the fabrication of complex aluminum components for thermal management applications. Graphene-reinforced Al-Si alloys are promising candidates due to the exceptional thermal conductivity of graphene, although their effective thermal conductivity is strongly limited by graphene agglomeration and associated metallurgical defects that disrupt heat-transfer pathways. In this study, an integrated processing strategy combining ultrasonic powder mixing and laser remelting was developed to simultaneously regulate graphene dispersion and suppress defects in LPBF-fabricated AlSi7Mg alloys. Ultrasonic mixing enhanced the initial dispersion of graphene, while remelting reduced porosity, achieving a relative density exceeding 99.8%, and promoted graphene redistribution. The remelted sample exhibited a thermal conductivity of 168 W/(m·K), representing a ∼29% increase, while maintaining excellent mechanical properties with an ultimate tensile strength of 445 MPa and an elongation of 14.9%. Microstructural analysis reveals that, beyond conventional densification effects, thermal transport is governed by the coupled roles of defect suppression and graphene redistribution, which together reconstruct effective heat-transfer pathways and facilitate both electron and phonon mediated transport. A semi-empirical thermal conductivity model incorporating both porosity and graphene dispersion is established, demonstrating that graphene connectivity provides an additional governing contribution beyond porosity alone. This work establishes a generalizable process-structure-property framework linking graphene dispersion to thermal transport in LPBF-fabricated metal matrix composites.

The spatial dependence of heterogeneous microstructure is crucial for additive manufacturing (AM) of high-temperature titanium (Ti) alloys rich in β-stabilizing element. However, the mechanism of phase and microstructure evolution remains unclear during the non-isothermal process, especially in the heating stage of AM technology. The present study focuses on the phase transition and microstructure evolution of a near-β Ti alloy fabricated by direct energy deposition during continuous heating, while the effects of different initial microstructures and heating rates on them are also investigated, systematically. The results demonstrate that the phase transition sequence of Ti alloy is β → α″ + α′ + β → α + β → β, starting from the initial single-β phase. The characteristic of nanoscale martensite phase varies with heating rate, leading to the formation of homogeneously distributed fine α lath. For the microstructure of fully intragranular Widmannstatten α phase (αWI), its phase transition sequence during the heating process is α + β → β. The extent of dissolution for α phase lags behind the equilibrium value, since its transition is diffusion-driven, and the secondary α phase transforms into the β phase followed by the αWI lath. This work is of great significance for understanding the microstructure evolution of near-β Ti alloy fabricated by the AM technology, and thus guiding the design of new materials.

Cement grinding serves as the final process in cement production, where its grinding effectiveness is critical to the quality and yield of the finished cement product. To address the challenges of multi-parameter coupling and modeling difficulties in the grinding process of ball mills within cement combined grinding systems, this study proposes a digital twin modeling method for cement grinding ball mills based on EDEM-RecurDyn coupling. The geometric model of the ball mill is established using SolidWorks software. The collision processes between cement particles and grinding media are simulated via EDEM discrete element software, while the dynamic response characteristics of the cylinder are analyzed using RecurDyn multibody dynamics software. The complete construction process of the ball mill digital twin model is systematically described. Coupled simulation analyses using RecurDyn-EDEM are conducted, and a virtual-real interaction framework integrating physical equipment, coupled models, and control decisions is established. This provides technical support for subsequent intelligent operation, maintenance, and decision-making control in cement combined grinding systems.

The combined grinding process exhibits strong nonlinear and coupled dynamics, making stable control of particle size distribution critical for product quality. This paper proposes a particle size distribution control strategy based on improved unscented Kalman filter assisted heuristic dynamic programming (IUKF-HDP). A differential evolution optimized echo state network (DE-ESN) is employed to model the dynamic behavior of the cement combined grinding process, with the separator speed and the rear main exhaust fan speed as control inputs. An improved unscented Kalman filter, incorporating an adaptive covariance inflation factor, is integrated into the HDP framework to perform online weight estimation, enhancing control stability and robustness. The closed-loop stability is analyzed using Lyapunov theory, confirming Uniformly Ultimately Bounded stability. Simulation results show that the DE-ESN model achieves prediction accuracy with $\mathbf{R}^{2}$ values exceeding 0.96. Compared with the conventional HDP method, the proposed IUKF-HDP strategy exhibits faster convergence, higher steady-state accuracy, and stronger disturbance rejection capability.

To enhance the mechanical properties of Inconel 625 alloy, this study incorporated nano-TiC particles at varying weight percentages (1 %, 2 %, 3 %, and 5 %) as reinforcements to fabricate metal matrix composites through Laser Directed Energy Deposition (DED). The microstructure evolution and mechanical properties were systematically investigated. The results indicate that the microstructure of the as-deposited Inconel 625 alloy was primarily dominated by columnar dendrites. With the increment of nano-TiC content, these dendrites were progressively refined, transitioning from a columnar to an equiaxed morphology. Moreover, the introduction of nano-TiC particles effectively reduced the space available for the growth of the Laves phase between dendrites. At a nano-TiC content of 5 %, the microhardness, yield strength, and tensile strength of the TiC/Inconel 625 composite were notably increased by 70 HV, 115.54 MPa, and 130.31 MPa, respectively. The tensile properties of the composites demonstrated higher strength but reduced ductility. The fracture surface analysis revealed typical dimple characteristics, indicating that the fracture mode was through microvoid coalescence. The comprehensive mechanical performance of the composite was optimized at a nano-TiC content of 3 %.

In response to the complex operating conditions, strong nonlinearity, and difficult control of the outlet temperature in cement calciners, this paper proposes a model-free adaptive integral sliding mode constrained control strategy for the calciner outlet temperature. First, an adaptive observer is designed to obtain pseudo-partial derivative estimates, which linearize the dynamic nonlinear system. Then, an integral sliding surface is constructed, and both the equivalent controller and feedback controller are designed. To address actuator saturation during the coal injection process, an anti-saturation compensator is introduced to compensate for the system. Stability analysis of the system is then performed. Finally, simulation experiments are conducted on a cement pre-decomposition mathematical model, confirming that the proposed control method achieves good control performance in regulating the calciner outlet temperature.

In the cement combined grinding production process, particle size is crucial to product quality, especially the particle size content less than 45 microns directly affects product performance. This paper presents a cement particle size prediction model integrating FFRLS, an adaptive observer, and a onestep prediction algorithm to enhance accuracy by compensating for disturbances. First, the average influence value method is used to screen key variables, and the data is processed by mean filtering. Then, the cement particle size prediction model is constructed using FFRLS. Finally, the adaptive observer and the one-step prediction algorithm are introduced to estimate the disturbance in real time and perform dynamic compensation to improve the prediction effect of the model. The experimental results show that the mean square error (MSE) of the test set of the FFRLS model is 0.1663, the mean absolute error (MAE) is 0.3056, and the R2 is 0.8361; the MSE of the test set of the model combined with the adaptive observer is significantly reduced to 0.0021, the MAE is reduced to 0.0355, and the R2 is increased to 0.9979, and the error is greatly reduced, which verifies the effectiveness of the method. This method significantly improves the accuracy of cement particle size prediction and provides an effective prediction tool for particle size control in cement production.

The ball mill is an important equipment in the combined grinding process. Accurately predicting the mill load is conducive to improving production efficiency and ensuring product quality. Traditional prediction methods have difficulty capturing the complex relationships between mill load and multiple influencing factors. Therefore, a cement combined grinding mill load prediction model based on the CNN-LSTM is proposed. First, the Mean Impact Value (MIV) algorithm is used to reduce the dimension of input variables and the mean filtering method is employed to preprocess the data. Then, the circulating fan speed and the feeding rate are sent as inputs to the convolutional layer. Finally, the feature information extracted by the convolutional layer is sent to the LSTM layer for data fitting and prediction. The experimental results show that, compared with traditional machine learning (SVM) and single deep learning models (CNN and LSTM), the proposed CNN-LSTM hybrid model has higher prediction accuracy. Among them, the $R^{2}$ is 0.9994, the MAE is 0.0286, the MBE is 0.0021 and the RMSE is 0.0359.

In order to solve the problem that the load of cement ball mill is difficult to judge, based on particle swarm optimization(PSO) algorithm, a stacked convolutional autoencoder(SCAE) is proposed to judge the load. Firstly, an infinite impulse response(IIR) digital band-pass filler is designed to filler the vibration signals from the mill's wall, extracting vibration signals within an appropriate frequency range. Subsequently, two convolutional autoencoders are established to capture deep features from the vibration signals. Finally, the encoders from these two convolutional autoencoders are slacked together, followed by the addition of a classification layer to form the stacked convolutional autoencoder. This model is then utilized to classify the load state. PSO is employed to optimize both the filtering frequency band and the hyperparameters of the neural network. The results demonstrate that the stacked convolutional autoencoder achieves high accuracy, with an accuracy rate exceeding 97.53% on the test set. This provides effective assistance in judging the load state of cement ball mills and improving cement grinding quality.

The coal-fired power plant (CFPP) coupled with the molten salt thermal energy storage system is a potential way to improve its flexibility and peak-shaving ability. The steam generation system (SGS) is a suitable choice to convert the feed water into hot steam by using the heat from the molten salt. In this paper, the schematic of an SGS coupled with the CFPP is presented. A dynamic model of the SGS basing lump parameter method is established and validated to investigate its dynamic response characteristics. Five different disturbance experiments including feed water inlet parameters, molten salt inlet parameters and medium pressure steam valve, are conducted and analyzed. The dynamic response curves of molten salt and steam are obtained. Moreover, the load adjustment process of SGS with three different load changing rates (3 %, 6 %, 10 % Pe/min) under the rated conditions is compared and its influence on the SGS is analyzed. The temperature changing rates in the thick-wall components of SGS are within 2 degrees C/min which meets the safe operating standard. Based on the above results, the three elements control strategy with excellent robustness for the SGS safety operation is proposed and demonstrated. The results show that the response time for the SGS load regulation process and the water level fluctuation have been significantly improved as the three elements control strategy is imposed. The water level can be stabilized in 200 s with tiny fluctuation when the SGS loads down 10 % thermal load with 10 % Pe/min load changing rate. These results could provide useful references for the design and control strategy making of the CFPP coupled with the molten salt thermal energy storage system.

In recent years, transition metal dichalcogenides (TMDs) have been widely used as saturable absorbers (SAs) in ultrafast fiber lasers. Among them, we have chosen manganese sulfide (MnS), also a type of TMD, for its small bandgap and excellent nonlinear optical properties. In this paper, we have fabricated a MnS thin film as a saturable absorber and demonstrated a passively Q-switched erbium-doped fiber laser based on MnS. It has achieved stable short-pulsed output. With a pump power of 29.8 mW from a 980 nm semiconductor laser, we obtained stable Q-switched pulses at a center wavelength of 1563.3 nm with a repetition rate of 29 kHz. The 3-dB spectral bandwidth is 1.25 nm, and the pulse width is 7.15 μs. The average output power is 0.95 mW, and the pulse energy is 32.8 nJ. The signal-to-noise ratio of the RF spectrum at the fundamental frequency is 54 dB, indicating the high stability of the Q-switched pulses. These results demonstrate that MnS, as a promising new SA material, provides a new research approach for generating stable short-pulsed output. Additionally, this is the first proposal of using MnS as a saturable absorber in Q-switched fiber lasers.

The preparation of Al-Li alloy by laser powder bed fusion (LPBF) technology, especially Al-Li alloy with high Li content, is of great significance for lightweight of aerospace equipment. However, the significant susceptibility of Al-Li alloys to hot cracking during the process limits their advancement. This study starts from the two aspects of process control and composition modification, to achieve the production of crack-free and high-quality 1460 Al-Li alloys. Crack-free 1460 specimens can only be prepared at extremely low scanning velocity. The process window is markedly expanded by Sc/Zr modification. The heterogeneous nucleation of Al3(Li,Sc,Zr) results in significant grain refinement and effectively suppresses crack initiation and propagation. The unique bimodal heterogeneous microstructure endows the alloy with notable mechanical properties. After the addition of 0.6Sc-0.3Zr, the ultimate tensile strength (UTS), yield strength (YS), and elongation (δ) of 1460 alloy are increased by 104%, 158%, and 43.9%, respectively. The strength-plasticity synergism is primarily attributed to grain refinement strengthening, precipitation strengthening, and hetero-deformation induced strain hardening resulting from its unique bimodal heterogeneous microstructure. The UTS and YS of 1.2Sc-0.6Zr modified 1460 alloy increased by 156% and 279%, respectively. However, this further increase in the Sc/Zr content significantly deteriorates the plasticity of the alloy. This work establishes a foundation for advancing the use of Al-Li alloys in high-performance, lightweight, and complex aerospace structures.

Timely and efficient corrosion detection in steel is beneficial for improving material quality monitoring and maintenance, extending the lifespan of steel structures, reducing safety risks, and minimizing resource wastage. In comparison to traditional methods, existing intelligent approaches exhibit exceptional performance under the assumption of independently and identically distributed data conditions. However, as steel corrosion typically does not conform to standard data distributions, there is a need to develop a new method with the capability to identify unknown corrosion conditions to meet practical requirements. To address this, this paper proposes an evidence-based deep learning approach for identifying steel corrosion in an open testing dataset. Specifically, we approach the task of steel corrosion identification from an out-of-distribution perspective, defining the states of steel as non-corroded, normal, and unknown corrosion states. This means treating corroded steel as an unknown class, introducing the uncertainty inference strategy of evidence-based deep learning. This allows the model to effectively recognize steel as corroded even when encountering data on corrosion states that it has not seen before. Experimental results demonstrate that, whether steel has corroded to an unknown state due to salt mist or acid mist, the model can classify it as an unknown corrosion state. This helps avoid the problem of recognition failure when the degree of corrosion in steel in practical scenarios does not conform to the distribution of training data.

In this paper, the microstructure evolution of TC17 titanium alloy formed by laser solid forming were analyzed, and the influence of solution and two-stage aging heat treatment on the microstructure homogenization and room temperature tensile properties were investigated. The extension of heat treatment temperature and time will lead to the size and volume fraction of primary lath α phase increasing, but decreasing for secondary α phase. Aging has no effect on the change of grain morphology but leading to the primary α lath coarsening. The hardness is the lowest at the secondary solution temperature of 750 °C. The strength and plasticity of heat-treated samples suffering 920 °C/0.5 h/ water quenching + 750 °C/1 h/water quenching + 630 °C/4 h/air cooling match well. The anisotropy is basically eliminated at the secondary solution temperature of 830 °C.

AlSi7Mg alloy manufactured by selective laser melting (SLM) was subjected to different heat treatments to satisfy the need for excellent mechanical and thermal properties. Effect of direct aging (DA), annealing (AN) and solution + aging (SA) treatment on microstructure, mechanical and thermal properties of SLMed AlSi7Mg alloy was investigated. The results showed that the molten pool boundaries which can be clearly observed in the as-built sample fade away with the increasing heat treatment temperature. The heat treatments have slight effects on the morphology and size of alpha-Al grains. However, significant differences can be observed in the characteristics of substructure. For the DA treatment, the Al-Si network becomes discontinuous and simultaneously promotes the precipitation of nano-sized Si particles from the super-saturated Al matrix. The original Al-Si network becomes discontinuous and coarsening in the AN sample. For the SA sample, the original cellular structure and eutectic Al-Si network completely disappeared. Coarsened and irregularly shaped Si particles, fine equiaxed Si particles as well as the needle shaped Fe-rich precipitations were observed. The room temperature tensile test showed that the strength of DA sample is the highest, followed by as-built, AN, and SA samples in a descending order. The thermal diffusivity of SA sample is the highest, followed by AN, DA and as-built sample in a descending order. (c) 2023 Elsevier B.V. All rights reserved.

In this work, the microstructural evolution of a directed energy deposited near beta titanium alloy, Ti5Al2Sn2Zr4Mo4Cr, during post-heat treatment was investigated, and its effect on the mechanical prop-erties was evaluated. The results showed that the equilibrium volume fraction of the alpha phase was reached in the first 30 min over a temperature range of 720-840 degrees C. Fine basketweave alpha laths (alpha BW) were homo-genously distributed within the prior beta grains at 840 degrees C. Both fine alpha BW and alpha colonies developed from the grain boundary alpha layers (alpha GBW) were obtained in the sample solutions treated at 720, 760, and 800 degrees C. Water-quenching and air-cooling were employed to retain the beta phase at room temperature. The alpha phase volume fraction increased significantly and reached the equilibrium volume fraction during the furnace -cooling process. In addition, the manner in which the samples were subjected to isothermal temperatures can strongly affect the evolution of the alpha phase. The homogenized specimen containing a single beta phase exhibited the lowest strength and the highest plasticity. The strength increased and the ductility decreased after subtransus solution treatment. The difference in the mechanical properties can be attributed to the differences in the alpha and beta phases caused by the various heat treatments. The modified rule of mixtures appropriately described the relationship between the yield strength and microstructure characteristics. (c) 2022 Elsevier B.V. All rights reserved.

In this study, TiC/CM247LC nickel-based composite was successfully prepared by selective laser melting, then was heat treated at a solid solution temperature of 1260 °C and different aging temperature of 840 °C, 870 °C, 900 °C and 930 °C respectively. Effects of aging temperatures on the microstructures and mechanical properties were systematically studied. The results show that the microstructures of all the heat treated samples are composed of γ matrix, carbides and γ′ phase. The γ grains remain a columnar shape after treatments, but the size of γ′ phase grows up gradually with the increasing aging temperature. The composite treated at an aging temperature of 870 °C exhibits the best mechanical properties with the tensile strength of 1073 MPa, yield strength of 1004 MPa and elongation of 7.57