To address the problems existing in nuclear reactor accident emergency training, a design scheme and system prototype of radiation detection training equipment for penetrating radiation fields in enclosed spaces, based on inertial sensors and wireless Bluetooth communication is proposed. First, the penetrating radiation field is modeled. On this basis, a calculation model of the neutron/γ dose equivalent rate is established. This model is based on the motion path of simulated radiation detection equipment. Second, the MPU6050 inertial sensor is designed and developed. It monitors the three-axis acceleration and three-axis angular acceleration values in real time. This enables the indoor positioning function of the simulated detection training equipment. The Digital Motion Processor (DMP) filtering algorithm is used to process the measured data. This improves the detection accuracy. Finally, a Bluetooth communication module is designed and developed. It transmits the detected position data to the main control computer in real time. The main control computer performs calculation and analysis to obtain the radiation intensity value. This value is sent to the Arduino controller. The Arduino controller controls the display of the value on the liquid crystal screen. Experimental verification is carried out. Experimental verification indicates that the maximum error of the system’s three-dimensional spatial positioning is 0.08 m, the mean relative error of the radiation dose rate simulation is 4.81%, and the maximum relative error is 7.8%. The system relatively accurately achieves radiation dose simulation and radiation source localization according to different working modes, providing a high cost-effectiveness training method for radiation detection training with high safety and good economy.

Based on actual topography measurement data, this study adopts a paraboloid of revolution to characterize asperity geometry and establishes a lateral contact model that accounts for substrate deformation and covers the full elastic–plastic process. The proposed model achieves smooth transitions in contact behavior, effectively avoiding non-physical oscillations and discontinuities in the elastoplastic regime exhibited by existing models. Comparative results with multiple classical models demonstrate that the proposed model maintains strict monotonicity and continuity in predicting contact load, average contact pressure, and contact stiffness. This significantly improves prediction accuracy and physical consistency, providing a more reliable theoretical tool for modeling precision mechanical joint interfaces.

Bolted flange joints are critical load-bearing connection components of complex mechanical equipment, whose service performance is dominated by the coupled evolution of bolt preload relaxation and interfacial stiffness degradation under long-term cyclic combined loads. Under cyclic transverse excitation, the contact interfaces experience successive full-stick, partial microslip and macroslip states, accompanied by embedding, creep and friction-induced wear, resulting in time-varying preload attenuation and the continuous degradation of joint mechanical properties. Distinguishing preload relaxation without nut rotation from rotational self-loosening is essential for revealing multi-bolt flange degradation mechanisms. This paper presents a systematic review of physics-driven constitutive modeling and degradation characterization of bolted flange connections. A unified classification framework for joint interface models is established, covering static friction models; velocity-dependent dynamic friction models; hysteresis stick–slip models, represented by the four-parameter Iwan model and Valanis endochronic model; and reduced-order equivalent joint models. The inherent differences between the Dahl model and LuGre model regarding the Stribeck velocity–friction characteristics are clarified. The representative models are comprehensively evaluated from the perspectives of physical interpretability, hysteresis reproduction accuracy, preload–degradation correlation and engineering applicability. A further comparative analysis is conducted of parameter identification bottlenecks of the Iwan, LuGre and Valanis models, focusing on multi-solution risk, anti-noise robustness and cross-working-condition transferability. Compared with single-bolt lap specimens, multi-bolt annular flanges exhibit prominent inter-bolt elastic interaction, circumferential contact pressure non-uniformity and local-to-global chain-reaction degradation behaviors. The common experimental test-beds and sensing techniques for joint degradation monitoring are summarized, and the measurement limitations, including sensor drift, synchronization error and installation constraints, and their influences on model parameter identification, are discussed. The existing machine-learning, digital-twin and physics-informed modeling applications for bolted joints are briefly outlined. Finally, this review summarizes the existing research consensus, unresolved contradictions and open research challenges. Special attention is paid to the limitations of existing time-variant Iwan-type models for multi-bolt flange degradation. Potential future research directions include multi-channel synchronous sensing matching time-varying constitutive models, quantification of inter-bolt load redistribution, and robust identification strategies under noisy experimental data.

Bolt self-loosening under transverse vibration is governed by frictional contact and accumulated interfacial slip. In this study, transverse vibration tests were performed on M8 bolted joints under different initial preloads, displacement amplitudes, and excitation frequencies, with three tests per condition. Preload histories were analyzed using an operational four-stage framework comprising an initial abrupt decrease, lower-rate attenuation, a sustained acceleration interval, and a decelerating tail. N₂ denotes the operationally identified onset of sustained acceleration, rather than an instantaneous physical discontinuity. Stage-II/III rate contrasts vary between conditions and are modest for several 0.3 mm cases. A model with an independently estimated N₂ was favored over the specified alternative with acceleration starting at N₁, subject to the comparison assumptions and conditional uncertainty analysis. Quantitative comparisons use repeat-test statistics normalized to each test’s own initial preload. At 5 Hz and 0.3 mm, the mean cycle counts to 50% preload are 206.20 ± 25.35 and 116.25 ± 17.75 cycles for nominal initial preloads of 10 and 12 kN, respectively. At 10 kN and 5 Hz, reducing amplitude from 0.3 to 0.2 mm increases this milestone to 922.25 ± 97.15 cycles. For nominal 12 kN and 0.3 mm, the values at 5, 7 and 10 Hz are 116.25 ± 17.75, 137.34 ± 17.36 and 267.70 ± 32.00 cycles, respectively. Values are means ± sample SDs. In the illustrated records, an independently identifiable later acceleration interval was not observed within 1110 cycles at 15 kN (Group 5) or within 1405 cycles at 0.1 mm (Group 7). The finite element model reproduces the overall staged form but shows substantial transition-time mismatch and is used to interpret representative contact-state evolution. The four-stage framework is a conditional description of the tested histories; it does not establish universally discrete mechanisms or a stage sequence applicable to every condition.

Silicone rubber foam is widely used in multi-field engineering protection due to its excellent cushioning and thermal insulation properties. However, its performance degradation caused by long-term service aging seriously affects equipment reliability. Establishing a constitutive model that can accurately characterize the mechanical response during aging is crucial for studying performance degradation and finite element simulation. Traditional multi-parameter aging constitutive models suffer from problems such as easy convergence to local optimal solutions and poor physical interpretability of parameters. To address these issues, this study systematically characterizes the evolution laws of the stress–strain response, compression set, and stress relaxation of silicone rubber foam over an aging period of 0–768 h through accelerated thermal aging and uniaxial compression tests and proposes a second-order Ogden aging constitutive model with a single time-varying parameter. This model fixes α₁, α₂, and μ₂ as constants and only sets μ₁ as the time-varying parameter, reducing the number of parameters to be fitted from four to one. The coefficient of determination (R²) of the full-cycle stress–strain curve fitting is ≥0.9966. Meanwhile, a quantitative physical correlation between μ₁ and macroscopic aging performance indicators is established, enabling the direct prediction of the mechanical response of aged materials using measurable macroscopic indicators. This work provides an efficient and reliable modeling method for the aging performance evaluation and structural simulation of silicone rubber foam.