The unfavorable factors of underground coal such as dark light, uneven illumination, band shadowing greatly make it difficult to recognize the coal rock at the mining workface accurately. To solve this problem, this paper proposes the fuse attention mechanism ’ s coal rock full-scale network (FAM-CRFSN) model. The deep extraction of coal rock semantic features is achieved by a multi-channel residual attention mechanism and a full-scale connection structure. Meanwhile, the balance between “deep” stacking and error back propagation is achieved by structures such as dilated convolution and Res2Block. Besides, a multi-dimensional loss function consisting of the cross-entropy loss, intersection over union, and multiscale structure similarity loss with pixel-level, area-level, and image-level expressions is established. Finally, the performance of the FAM-CRFSN network is tested with RGB coal rock images collected from an underground coal mining workface and superimposed with different proportions of gaussian noise and salt & pepper noise. The experimental results show that the FAM-CRFSN model can segment the coal rock regions accurately; at a noise intensity of 0.09, it achieves an MIOU of 85.77% and an MPA of 92.12%. Also, it achieves better accuracy and generalization performance than the mainstream semantic segmentation models. This study provides an important theoretical basis for promotes the unmanned and intelligent mining workface.

An experimental testing system for the two-dimensional (2D) fuze overload loading process was designed to address the loading issues of recoil overload and centrifugal overload in fuze safety and arming (S&A) device. By incorporating centrifuge rotation energy storage, impact acceleration simulation, and equivalent centrifugal rotation simulation, a block equipped with a fuze S&A device accelerated instantly upon having impact from a centrifuge-driven impact hammer, simulating recoil overload loading. The impact hammer was retracted instantaneously by adopting an electromagnetic brake, which resulted in the centrifugal rotation of the block around its track, to simulate the centrifugal overload loading. The dynamic equations of the experimental testing system and the equations of impact hammer motions were established, whereby the rotation speed of the centrifuge and the braking force of the electromagnetic brake were calculated and selected. A dynamic model of the collision between the impact hammer and block was established using ANSYS/LS-DYNA software for simulation analysis. The acceleration curves of the recoil overload and centrifugal overload with variations in the centrifuge speed, cushion material, and buffer thickness were obtained, which verified the feasibility of the proposed loading simulation method. Two-dimensional overload loading simulation tests were performed using the developed experimental testing system, and the acceleration curves of the recoil overload and centrifugal overload were measured. The test results indicated that the proposed system can accomplish 2D overload loading simulations for a recoil overload of several 10,000× g and centrifugal overload of several 1000× g.

Available hydropneumatic suspension simulation test benches have insufficient loading accuracy and limited functionality rendering them unsuitable for performance testing of heavy vehicles with this type of suspension. Therefore, a multi-functional compound simulation test bench was designed that used an electro-hydraulic proportional control technique. A mathematical model was established to describe the hydraulic loading system, and the transfer function of the system was derived. The gain and phase margins confirmed the stability of the system. A simulation model was established in the Simulink environment and step and sine signals of different frequencies were applied separately to analyze the dynamic characteristics of the system. The results showed that the system responded slowly and exhibited phase lag and signal distortion. The dynamic characteristics of the system were improved by incorporating an adaptive fuzzy PID controller. Simulation results showed that the response of the system to the step signal stabilized at the preset value within 0.3 s with no oscillation or overshoot. The improved system performed well in replicating the random vibrations of heavy vehicles operating on Class B and C roads. This confirmed that the system can satisfy the loading requirements of heavy vehicle hydropneumatic suspensions and can be used as a simulation test bench for such suspensions.

Accurately perceiving changes in water level information is key to achieving the fine control of water and flooding; however, the existing technology cannot achieve water level recognition in complex and harsh environments, such as at night; in haze, rain, or snow; or during obscuration by floating objects or shadows. Therefore, on the basis of a deep analysis of the characteristics of water level images in complex and harsh environments, in this study, we took full advantage of a deep learning network’s ability to characterise semantic features and carried out exploratory research on water level detection in no-water-ruler scenarios based on the two technical means of target detection and semantic segmentation. The related experiments illustrate that all the methods proposed in this study can effectively adapt to complex and harsh environments. The results of this study are valuable for applications in solving the difficulties of accurate water level detection and flood disaster early warnings in poor-visibility scenarios.

The fuze launch process is subjected to backseat and spin overloads. To address this issue, a loading method of a 2D dynamic acceleration environment was developed in this study for testing fuze antioverload performance on ground. The techniques of flywheel energy storage, high-speed impact, and centrifugal rotation in the track are combined in a dynamic analysis and simulation. First, the flywheel is rotated at a constant speed by a variable-frequency motor to obtain high kinetic energy. Second, an impact hammer is instantaneously released on the specimen at a high speed, loading the backseat acceleration environment. Finally, the impact hammer is retracted, and the specimen is rotated in the track instead of spinning around its axis, thereby loading the centrifugal acceleration environment. The peak value and pulse width of the 2D overload acceleration can be adjusted by changing the speed of the flywheel and buffers in the abovementioned process. The experimental and simulation results observed that the peak value of backseat acceleration could reach 34,559g, the pulse width was approximately 400μs, and the peak value of the centrifugal acceleration was 1,020g. The study results showed that the proposed approach fulfills the requirements of the 2D overload simulation test of the micro-electromechanical system (MEMS) fuze safety and arming mechanism. The proposed loading method has been successfully applied to ground simulation tests of the MEMS fuze safety and arming mechanism.