As the number of electric vehicles （EVs） increases， reinforcement learning （RL） in EV charging scheduling faces challenges， particularly uncertainties and the curse of dimensionality from large‑scale applications. A microgrid model for residential areas， considering the vehicle‑to‑grid （V2G） mode and various nonlinear charging models is developed. The problem is formulated as a constrained Markov decision process （CMDP）， with a model‑free RL framework to handle uncertainties. To address the curse of dimensionality， a strategy is designed where EVs are grouped by states， and agents send control signals to these sets， thus reducing the dimensionality of the action space. A Lagrangian deep deterministic policy gradient （LDDPG） algorithm is employed to solve the charging scheduling problem， with a safety filter ensuring constraint compliance. Numerical simulations validate the strategy’s effectiveness.

To address the performance degradation and lack of robustness in existing forgery localization models when dealing with small region facial manipulations in multi-person scene images， a FMG-L model based on facial mask guidance for forgery localization is proposed. Firstly， to mitigate interference from background information in multi-person scene images， a facial mask guidance module is designed to encourage the model to focus on critical facial regions. Secondly， to enhance the robustness against image degradations， a three-channel feature extraction module is developed to extract multi-dimensional features， and a feature fusion module based on a dual attention network is also designed to enhance the forgery clues. Finally， a forgery localization module is used for forgery localization. Experimental results on the OpenForensics， ManulFake， FFIW， and DiffSwap datasets demonstrate that the FMG-L effectively localizes forgery regions and shows strong robustness against various image degradations and different online social platforms.

In order to maximize the measurement performance of Coriolis flowmeter in practical application， a quadrature demodulation algorithm based on correlation operation and Hilbert transform （CH-QD algorithm） is proposed. The proposed algorithm uses the same-frequency cosine frequency characteristic of autocorrelation and the phase-shift characteristic of Hilbert transform to generate the reference signals required by quadrature demodulation， which effectively solves the problem that the traditional quadrature demodulation algorithm needs to predict the signal frequency and is sensitive to the performance of the frequency detection algorithm， which improves the detection accuracy of the algorithm， and realizes the joint estimation of signal frequency and phase difference. Simulation results show that the proposed algorithm has high detection accuracy， noise resistance and dynamic tracking performance. Based on TMS320F28335DSP， the algorithm is implemented on Coriolis flowmeter transmitter， and the performance analysis and water flow calibration experiments are carried out. The calibration results show that the measurement error of the system is less than 0.1% and the repeatability is less than 0.05% in the range of range ratio of 20∶1.

Machine vision-based environmental perception technology is one of the key tasks in the field of intelligent transportation. Traditional deep learning algorithms typically meet the detection needs of individual targets in simple scenarios. However， they are not capable of addressing the intelligent perception requirements in complex traffic environment. To improve the intelligent perception capability of vehicles in such environment， this paper proposes an improved YOLOv8 object detection network model， integrating attention mechanisms， optimizers， and deformable convolutional layers to achieve multi-target detection in complex urban traffic environment. To verify the effectiveness of the algorithm， comparative experiment were conducted using YOLOv4， YOLOv8， and the improved YOLOv8 algorithm on sample images from complex traffic environments. The results show that， compared to YOLOv4 and YOLOv8， the improved YOLOv8 algorithm increased the average accuracy by 40.76% and 16.92%， respectively. The detection accuracy and real-time performance of the improved YOLOv8 algorithm meet the practical application requirements， and through multi-sensor information fusion， it can realize intelligent perception in complex urban traffic environment.

To address the issue of false positives caused by multiple hypothesis testing in big data mining， as well as the extremely time-consuming nature of calculating theoretical results for controlling the false discovery rate （FDR）. Aiming at the computational efficiency of theoretical FDR values， a distributed false-positive control algorithm based on DPFDR（distributed permutation testing-based false discovery rate） is proposed. The algorithm firstly mining the representative patterns based on the conditional frequent pattern tree （CFP） method， and using the representative patterns to compress the pattern space. Then， the workload of the corresponding task is estimated according to the representative mode， the data is divided according to the workload， and the task is allocated to each compute node through the load balancing policy. Finally， the effective FDR false-positive control threshold is obtained by merging and sorting the calculation results of each node. A series of experimental results on real data sets show that the proposed DPFDR algorithm can greatly improve the computational efficiency of FDR false positive control threshold.

The dynamic self-balancing mechanism of a dual-ball automatic balancing device based on a single-disc rotor model is investigated using the variation amplitude method. The system control equations are established through Lagrange equations， and the amplitude-frequency characteristic equations under four working conditions are derived. The Jacobian matrix is directly constructed based on the solution form of the variable amplitude method， avoiding the case-by-case discussion of steady-state equations in the traditional methods， and the system stability conditions are obtained using the Routh-Hurwitz criterion. Numerical analysis reveals that： system vibration intensifies in the lower resonance region； in the upper resonance region， zero-amplitude motion occurs only when parameters satisfy the complete balancing condition， at which point the balls jump to the opposite side of the mass center， achieving complete or partial balance. The research results are validated through time-domain simulations， providing a theoretical basis for the design and optimization of ball-type automatic balancing devices.

The surface of 20CrMnTi steel with high abrasion resistance has great demands in the manufacturing of core components such as shafts and gears. However， it is hard for the existing surface finishing and strengthening methods to balance productivity and energy consumption. Combining laser micro-carburizing and abrasive processing technology， an efficient and controllable integrated manufacturing method for surface machining and strengthening of 20CrMnTi was proposed： laser heat-assisted carburizing grinding. A series of laser heat-assisted carburizing grinding tests were carried out on an automatic surface grinding machine， and the creation mechanism of enhanced surfaces under laser heat-assistance was analyzed by the metallographic microscope and X-ray diffractometer， etc. At the same time， friction and wear tests were carried out and the mechanical properties of the strengthened surface were tested. The results showed that the surface hardness of laser heat-assisted carburizing 20CrMnTi can reach 550 HV and above， and the wear loss rate is reduced to 50% of the matrix. The grinding heat tempering with phase difference tends to homogenize the strengthening phase， precipitate granular carbides， and achieve diffuse strengthening. The proposed method can help and realize the high-performance manufacturing of low-carbon steel surfaces.

Aiming at the problems of complex bearing operating conditions， weak generalization ability and low accuracy of model recognition of traditional deep learning fault diagnosis methods， a rolling bearing fault diagnosis method based on the GRM-IConvNeXt model is established. Firstly， a coding method of global relationship matrix （GRM） is proposed， which can transform one-dimensional vibration signals into two-dimensional images by taking the advantage of preserving the original signal features. Then， an improved ConvNeXt （IConvNeXt） model for small sample classification of bearing fault diagnosis is constructed， and a convolution kernel with a size of 5×5， multiple BN layers and Hardswish activation function are selected to enhance the feature extraction performance. At the same time， weights are adaptively generated according to the GRM image features through the CBAM（convolutional block attention module） mechanism. The experimental results show that the GRM-IConvNeXt model has good feature extraction ability and generalization under off-design conditions and small samples.

Aiming at the deformation of carbon fiber reinforced polymer （CFRP） bolted joints subjected to transverse load as well as the loading situation， a model is established to predict the tensile strength of a titanium alloy bolted joint with CFRP plates based on the three-dimensional Hashin failure criterion， in which the load-displacement curve of the CFRP bolted joint from the beginning of loading to complete destruction is divided into five stages， i.e. the stationary stage， the slip stage， the complete slip stage， the bolted rod load-bearing stage and the failure stage. Then， the contact state as well as the mode of load transfer in each stage are analyzed， the magnitude of bending moment and shear force as well as the distribution of the bolted joint at different stages are investigated， and the effects of CFRP thickness and bolt preload on the strength of the joint， bending moment and shear force at the joint are analyzed. The results show that the maximum error between theoretical calculations and simulation results for bending moment and shear force at the bolted joint does not exceed 20% in the slip stage and the complete slip stage， and the maximum error does not exceed 15% in the bolted rod load-bearing stage and the failure stage.

Taking the thin-walled carbon fiber tubes in the support structure of satellite antennas as the research object， the damage resistance of thin-walled carbon fiber tubes with damage was investigated with respect to the invisible delamination damage generated by low-energy impacts that are prone to occur in the process of tube transporting and assembling. A finite element model of a thin-walled carbon fiber tube with delamination damage was developed using ABAQUS solid units， and the finite element model was validated by low-energy transverse impact experiments. Considering the actual working conditions of the tube， the effects of fiber orientation and delamination position on the compressive properties of the tube and the effect of delamination damage on the secondary transverse impact were investigated. The results showed that the closer the fiber orientation is to the axial direction， the smaller the effect of delamination damage on the compressive properties of the tube. The resistance of the tube to transverse impact damage decreases only in the position of the damage.

To ensure the stiffness of the rotary table while minimizing its weight and enhancing its reliability， a reliability-based design optimization （RBDO） method is proposed for the rotary table. This method takes into account the invariance of the assembly relationships between the internal structures of the rotary table， and also incorporates the random uncertainty factors. Through sensitivity analysis， size parameters that significantly affect the performance of the rotary table are identified， and optimization design is conducted on these parameters. Unlike the traditional optimization methods， a Kriging surrogate model is applied to replace time-consuming finite element analysis. Furthermore， an efficient meta-heuristic algorithm is introduced during the optimization process to address the proposed RBDO problem. Subsequently， the impact of various parameter configurations on the deformation reliability of the rotary table is investigated. Finally， the effectiveness and robustness of the proposed optimization method are validated through practical examples.

Based on the finite element method， the effect of osteoporosis on lumbar fusion surgery under vibration conditions was explained from a clinical perspective using such indicators as stress， segmental lordosis angle， and intervertebral disc height. It was found that when the human body is exposed to a vibrating environment， the segments undergoing lumbar fusion surgery may cause other vertebral segments to become unstable. This instability significantly increases the risk of other segments developing diseases， failure of the fusion device， failure of the fixation device， and reduced lumbar stability. From the perspective of the effectiveness of lumbar fusion surgery， osteoporosis leads to bone cells in the fusion segments being exposed to a more adverse vibrational growth environment， resulting in poorer fusion outcomes.

The thermal quantity released/absorbed during gas adsorption/desorption significantly influences the evolution of permeability rebound and recovery in coal seams. Based on the thermo-hydro-mechanical （THM） coupling theory， this study investigates the temporal evolution of permeability rebound and recovery under single/multiple gas adsorption thermal effects and further assesses their impact on coalbed methane （CBM） extraction. Results indicate that under a single-factor influence， permeability rebound and recovery time exhibit a positive correlation with adsorption heat intensity， and its response to Langmuir’s constant is characterized similarly to the former and tends to stabilize when the adsorption performance parameter exceeds the critical value. Under multi-factor coupling， adsorption heat enhances permeability rebound and recovery on the temporal scale while the spatial scale is less susceptible to adsorption heat. Furthermore， neglecting adsorption thermal effects may lead to an over-estimation of the gas production increase induced by permeability rebound and recovery.

To solve the problem of difficulty in accurately measuring ventilation parameters at non-ideal test sections of ventilation systems， based on the numerical simulation method， the effects of pipe diameter， air velocity， curvature-diameter ratio， particulate matter mass concentration， particle size， density and other factors on the airflow and particle concentration distribution in the 90° circular bend are explored. The measurement errors and rules of the equal-area ring method and the center point method on the non-ideal test section are analyzed. The results show that the air velocity distribution in the bend is only affected by the curvature-diameter ratio， and the particulate matter mass concentration is affected by the curvature-diameter ratio， particle size and density. The maximum error of air velocity at the elbow outlet measured by the equal area ring method is 7.8%. When the curvature-diameter ratio is fixed， there is a functional relationship between the error of each measured section and the pipe diameter， and there is a linear relationship between the wind speed of the pipe center and the average air velocity of the section. The particulate matter mass concentration errors of the elbow inlet and outlet measured by the equal-area ring method are all below 10%. This study can provide a reference for the monitoring of air velocity and particulate matter mass concentration in non-ideal measurement section.

At present， many hyperelastic constitutive models are widely used in the finite element analysis of rubber isolation bearings. The selection of proper constitutive models and corresponding parameter determination are crucial in this process. This study aims to investigate the applicability of various hyperelastic constitutive models for different brands of tire rubber and the variation law of rubber parameters after aging. Three different brands of waste tires were selected for uniaxial tensile tests， while a batch of tires were selected for accelerated aging tests using hot air at 77， 154， 231， and 308 h. Employing the determined model parameters， the scrap tire rubber pads （STP） were modeled and compared with vertical experiments after aging using ABAQUS software. The results show that the stress-strain curves of the rubber in all three tire brands exhibit typical reverse “S” shapes， and the Yeoh model is determined to be optimal after comparing multiple models. The Yeoh model parameters approximately show monotonic changes with increasing aging time. The errors between the STP vertical stiffness test and the finite element simulation values before and after aging are within an acceptable range， which verifies the accuracy of the time-varying law of the aging parameters. This research provides theoretical support for the whole life design of STP and other waste tire products.

In order to clarify the dynamic response of prestressed anchor cable in rock slope under earthquake， a set of anchor cable dynamic calculation models based on the synergistic action of dangerous rock mass， free section of anchor cable， slope structural plane， anchorage section of anchor cable and bedrock is proposed， and the analytical solution of dynamic response of anchor cable is obtained. The accuracy of the anchor cable dynamic calculation model is verified by the centrifuge model test， and the key parameters are analyzed according to an engineering case. The results show that the increase of seismic circular frequency can trigger the slope resonance effect， and the peak value of axial force increment in the free segment of anchor cable increases significantly. With the increase of bedrock shear modulus， the spring stiffness and damping coefficient of the anchorage interface increase， the axial transmission of the dynamic response is limited， and the top of the anchorage segment undertakes a larger shear stress increment. The damping ratio of the structural plane and the Poisson’s ratio of the bedrock have little influence on the axial force increment of the free segment and the internal force increment distribution in anchorage segment of the anchor cable， so these parameters are not taken as the key object in seismic design of the anchor cable. The research results can provide theoretical basis for dynamic analysis and seismic design of prestressed anchor cable.

In order to explore the influence of side spray position on thermal runaway propagation of lithium-ion batteries （LIBs）， a nitrogen and water two-phase flow ultra-fine water mist （NWM） test platform was set up. By changing the position of NWM applied on the side of LIB module， the cooling， thermal radiation attenuation and asphyxiation effects of NWM at different side injection locations， and the dominant mechanism of NWM inhibiting runaway thermal propagation of lithium-ion batteries at different locations were compared and analyzed. The results show that flame can promote the thermal runaway propagation of LIB module. The effect of NWM applied at the pole position on LIB module thermal runaway propagation is more significant， and the smothering and attenuating flame thermal radiation play a leading role. When NWM is applied in the middle position， the dominant mechanism to inhibit runaway thermal propagation is to cool the surface temperature of the LIB module.

With the increasing depletion of easily sortable zinc sulphide ores， it is significant important to develop zinc oxide ores. The flotation behavior and adsorption mechanism of smithsonite was systematically investigated using sodium lauryl glutamate （SLG） as a collector to provide a reference for the development of high-efficiency flotation collectors for zinc oxide ores. The flotation test results show that SLG has a better ability to collect smithsonite compared with sodium laurate （SL）， with a flotation recovery of smithsonite reaching to 94.8% at a weakly alkaline pH and collector concentration of 80 mg/L. The findings from contact angle， Zeta potential and FT-IR analysis demonstrate that the primary chemical adsorption between SLG and smithsonite surfaces， as evidenced by XPS analysis， is predominantly attributable to the bonding between C==O in amide group of the SLG molecule and the mineral surface sites.