Search

An enhanced algorithm, YOLOv5s-FCW, is put forward in this study to tackle the problems that exist in the current helmet detection (HD) methods. These issues include having too many parameters, a complex network, and large computation requirements, making it unsuitable for deployment on embedded and other devices. Additionally, existing algorithms struggle with detecting small targets and do not achieve high enough recognition accuracy. Firstly, the YOLOv5s backbone network is replaced by FasterNet for feature extraction (FE), which reduces the number of parameters and computational effort in the network. Secondly, a convolutional block attention module (CBAM) is added to the YOLOv5 model to improve the detection model’s ability to detect small objects such as helmets by increasing its attention to them. Finally, to enhance model convergence, the WIoU_Loss loss function is adopted instead of the GIoU_Loss loss function. As reported by the experimental results, the YOLOv5s-FCW algorithm proposed in this study has improved accuracy by 4.6% compared to the baseline algorithm. The proposed approach not only enhances detection concerning small and obscured targets but also reduces computation for the YOLOv5s model by 20%, thereby decreasing the hardware cost while maintaining a higher average accuracy regarding detection.

Autonomous aerial refueling technology can significantly extend the operational endurance of unmanned aerial vehicles (UAVs), enhancing their ability to perform long-duration missions efficiently. In this paper, we address the identification of refueling drogues in the close docking phase of autonomous aerial refueling. We propose a high-precision real-time drogue recognition network called DREP-Net. The backbone of this network employs the DGST module for efficient feature extraction and improved representation of multi-scale information. For occlusion and complex background problems, we designed the RGConv module, which combines the re-parameterization module with the GhostNet idea to improve the detection of an occluded drogue. Meanwhile, we introduced the efficient local attention mechanism into the neck network to enhance the overall attention to the target region. Then, we designed Phead, a lightweight detection head that combines the advantages of decoupling and coupling heads to improve the detection speed. Finally, we compared our network with mainstream algorithms on a real drogue dataset, and the results show that DREP-Net has 2.7% higher mean average precision (mAP) compared to the YOLOv8n model, and the detection speed is improved by 31.4 frames per second.

Abstract This paper presents an improved Fast-Segmentation Convolutional Neural Network (Fast-SCNN) and U-Net networks based on the channel attention mechanism. While ensuring the speed of network detection, the accuracy of image segmentation is also considered. The experimental results show that the accuracy rate of improved Fast-SCNN based on the channel attention mechanism is greatly improved compared with the original Fast-SCNN, reaching 88.056%, and the mean intersection over union is also improved to a certain extent, reaching 81.087%, and the detection speed is better than the original Fast-SCNN network. The accuracy of improved U-Net network based on the channel attention mechanism is 0.91805, which is better than the original U-Net network. In terms of detection speed, the improved U-Net network based on channel attention mechanism has greatly improved compared with the original U-Net network, reaching 24.02 frames per second. In addition, a method of preventing clogging of ore conveyor belts based on static image detection is proposed in this paper. By judging and predicting the blockage of the ore conveyor belt. When the conveyor belt is about to be blocked or has been blocked, the fuzzy algorithm is used to control the ore conveyor belt to slow down and stop, to improve the safety and efficiency of the conveyor belt.

Manganese is widely used in steel, chemical, non-ferrous metallurgy, batteries and other fields. Most of the manganese oxide ore in China are lean ore with high iron and low manganese, and most of them need beneficiation pretreatment before smelting. Commonly used processes include blast furnace smelting method, reduction roasting-leaching method, and direct reduction-leaching method. Among them, the reduction roasting-leaching process can realize the selective separation of iron and manganese minerals, and the comprehensive resource utilization rate is high, but the energy consumption of the roasting process is large. The thermodynamic conditions and kinetics of the simultaneous reduction of iron and manganese minerals must be studied, and the interaction between the magnetization reduction of iron oxide minerals and the pre-reduction of manganese oxide and the law of simultaneous reduction, the phase transformation and crystal form transformation of manganese oxide and iron oxide ore must be clarified, the mineral composition and bonding method of the bond during the reduction process. Provide a theoretical basis for realizing the low-temperature reduction of iron minerals and manganese minerals and saving energy consumption.

A large number of studies have shown that oolitic hematite is an iron ore that is extremely difficult to utilize because of its fine disseminated particle size, high harmful impurity content and oolitic structure. To recover iron from oolitic hematite, we developed a novel multistage dynamic magnetizing roasting technology. Compared with traditional magnetizing roasting technologies, this novel technology has the following advantages: firstly, the oolitic hematite is dynamically reduced in a multi-stage roasting furnace, which shortens the reduction time and avoids ringing and over-reduction; secondly, the novel dynamic magnetizing roasting technology has strong raw material adaptability, and the size range of raw materials can be as wide as 0–15 mm; thirdly, the roasting furnace adopts a preheating-heating process, and the low-calorific value blast furnace gas can be used as the fuel and reductant, which greatly reduces the cost. The actual industrial production data showed that the energy consumption in the roasting process can be less than 35 kg of standard coal per ton of raw ore. The iron grade of the concentrate and iron recovery reached 65% and 90%, respectively.

Aiming at the low efficiency of vector control system caused by unreasonable current allocation, an improved energy saving control strategy based on weighted average current method is proposed in this paper, and the research is carried out on the background of interior permanent magnet synchronous motor vector control system. This method does not add new parameters, but achieves the goal of reducing motor loss and improving current utilization rate. Firstly, according to the mathematical model of permanent magnet synchronous motor considering iron loss resistance, the relationship between motor torque and d-q axis current is deduced. Then, by introducing the weighted average current factor, the minimum copper loss current and the minimum iron loss current are calculated analytically. Finally, the simulation analysis shows that the energy saving strategy proposed in this paper can further reduce the electrical loss of permanent magnet motor compared with the traditional energy saving control method, which proves the feasibility of this method in improving the energy efficiency of the vector control system of permanent magnet motor.

The data synchronous acquisition is crucial to the seafloor subsidence monitoring system for gas hydrate mining areas based on microelectromechanical sensors (MEMS). Because the independent and high-precision time reference sources on land cannot be used on the seafloor, especially in the deep sea, a relative time synchronization method based on input/output (I/O) and controller area network (CAN) bus was proposed to realize the internal time synchronization of the system. To demonstrate the feasibility of the proposed method, tests including the deformation test of the MEMS sensor array under high pressure, synchronous accuracy test, and landslide and collapse simulation tests were carried out. The synchronization method was performed once every 24 h, and the time drift was reduced to 0.38 ms from more than 30 ms, demonstrating that method can achieve consistent internal time of the system. The method does not require additional hardware devices and has adjustable accuracy.

For the detection of deep-sea natural gas hydrates, it is very important to accurately detect the 13CO2/12CO2 isotope ratio of dissolved gas in seawater. In this paper, a 13CO2/12CO2 isotope ratio sensor is investigated, which uses a tunable diode laser absorption spectroscopy (TDLAS) technique at 4.3 μm. The proposed sensor consists of a mid-infrared interband cascade laser (ICL) operating in continuous wave mode, a long optical path multi-pass gas cell (MPGC) of 24 m, and a mid-infrared mercury cadmium telluride (MCT) detector. Aiming at the problem of the strong absorption intensity of the two absorption lines of 13CO2 and 12CO2 being affected by temperature, a high-precision temperature control system for the MPGC was fabricated. Five different concentrations of CO2 gas were configured to calibrate the sensor, and the response linearity could reach 0.9992 for 12CO2 and 0.9996 for 13CO2. The data show that the carbon isotope measurement precision was assessed to be 0.0139‱ when the integration time was 92 s and the optical path length was 24 m. The sensor is combined with a gas−liquid separator to detect the 13CO2/12CO2 isotope ratio of CO2 gas extracted from water. Results validate the reported sensor system’s potential application in deep-sea natural gas hydrate exploration.

An apparatus for the analysis of pressure cores containing gas hydrates at in situ pressures was designed, and a series of experiments to determine the compressional wave response of hydrate-bearing sands were performed systematically in the laboratory. Considering the difficulties encountered in performing valid laboratory tests and in recovering intact hydrate bearing sediment samples, the laboratory approach enabled closer study than the marine environment due to sample recovery problems. The apparatus was designed to achieve in situ hydrate formation in bearing sediments and synchronous ultrasonic detection. The P-wave velocity measurements enabled quick and successive ultrasonic analysis of pressure cores. The factors influencing P-wave velocity (Vp), including hydrate saturation and formation methodology, were investigated. By controlling the initial water saturation and gas pressure, we conducted separate experiments for different hydrate saturation values ranging from 2% to 60%. The measured P-wave velocity varied from less than 1700 m/s to more than 3100 m/s in this saturation range. The hydrate saturation can be successfully predicted by a linear fitting of the attenuation (Q−1) to the hydrate saturation. This approach provided a new method for acoustic measurement of the hydrate saturation when the arrival time of the first wave cannot be directly distinguished. Our results demonstrated that the specially designed non-embedded ultrasonic detection apparatus could determine the hydrate saturation and occurrence patterns in pressure cores, which could assist further hydrate resource exploration and detailed core analyses.

In this study, the effects of various characteristics of artificial and natural magnetite concentrates on their pelletizing behaviors were elucidated through batch pelletizing experiments. The results showed that artificial magnetite concentrates exhibit higher hydrophilicity and lower electronegativity compared with natural magnetite concentrates. Additionally, scanning electron microscopy of magnetite particles demonstrated that the surface of artificial magnetite particles was rougher than that of natural magnetite particles, which causes the artificial magnetite to have a larger specific surface area compared with natural magnetite. Pelletizing experiments indicated that the compressive strength of artificial magnetite green pellets was higher than that of natural magnetite green pellets. However, the shock temperature of artificial magnetite green pellet was only 255°C, which is far lower than the requirements of industrial production. It was found that when the pelletizing raw materials were mixed to consist of 20% artificial magnetite and 80% natural magnetite, the ballability increased significantly. [ABSTRACT FROM AUTHOR]

This paper presents a systematic approach for robust performance analysis of active suspension system. Firstly, the background information on H - μ control methods is briefly described. Secondly, the formula of the structured uncertainties of active suspension are given, and the difference between the maximum singular values and structured singular values of a transfer function matrix is discussed. Then the robust performance of an active suspension system in the presence of assumed parameter variation is analyzed. All the steps are illustrated by means of a numerical example of a railway car with lateral active suspension system. [ABSTRACT FROM AUTHOR]