Search

The posture detection of underground personnel can provide effective information for disaster warning and accident rescue. The postures of the underground personnel are complex and diverse and they are time series data. The existing human posture detection methods are difficult to process continuous related posture data. And the real-time performance is poor due to the complex algorithm and the need to configure an independent computer. In order to solve the above problems, a human posture detection method in coal mine based on improved long short term memory network(LSTM) is proposed. The pressure sensor and angle sensor are used to obtain the posture data of underground personnel, such as foot pressure, waist and leg angle, etc. The portable edge operation decision unit carried by the personnel can discriminate the posture. It can realize the real-time detection of five postures of underground personnel, such as standing, walking, bending, squatting (sitting) and lying down. In order to reduce the dimension of the original sampled data of human posture and improve the compute efficiency, LSTM is improved. The long short term memory sparse autoencoder(LSTMSA) is designed. The characteristics of the original sampled data is extracted by the sparse autoencoder(SA) to reduce the dimension, and then the human posture is detected by the LSTM. Human posture data are collected in the laboratory environment, and LSTMSA, LSTM and recursive neural network(RNN) are trained and tested respectively. The results show that under the same experimental settings and sampling data, the accuracy of LSTMSA for five kinds of human posture detection reaches more than 90%, which is close to LSTM and greater than RNN. The computing time of LSTMSA is shortened by more than 50% compared with LSTM, which meets the real-time requirements of human posture detection in coal mine.

对高酸值米糠油的活性白土与硅胶联合脱色工艺进行研究。对酸值（KOH）为27 mg/g的米糠油进行先活性白土后硅胶脱色，以脱色率、谷维素含量和精炼率为考察指标，在单因素试验的基础上采用响应面试验优化活性白土与硅胶联合脱色条件。结果表明：活性白土脱色最优工艺条件为活性白土添加量25%、脱色时间30 min、脱色温度90 ℃；硅胶脱色最优工艺条件为硅胶添加量2.0%、脱色时间30 min、脱色温度82 ℃。在最优工艺条件下，活性白土脱色后，米糠油脱色率为79.15%，谷维素含量为2.14%，精炼率为90.05%；再经硅胶脱色后，米糠油脱色率为47%，谷维素含量为2.10%，精炼率为88.36%。经过活性白土与硅胶联合脱色后米糠油的色泽明显改善，色泽（25.4 mm罗维朋比色槽）由原油的Y35、R15降至Y5、R0.3。 The bleaching of high acid value rice bran oil through activated clay combined with silica gel was studied.The rice bran oil with acid value 27mgKOH/g was firstly bleached by activated clay then by silica gel. With the bleaching rate, oryzanol content and refining rate as indexes, response surface methodology was used to optimize the combined bleaching conditions based on single factor experiment. The results showed that the optimal bleaching conditions for activated clay were obtained as follows: activated clay dosage 2.5%,bleaching time 30 min and bleaching temperature 90 ℃. The optimal bleaching conditions for silica gel were obtained as follows: silica gel dosage 2.0%, bleaching time 30 min and bleaching temperature 82 ℃. Under the optimal conditions, the bleaching rate of rice bran oil was 79.15%, the content of oryzanol was 2.14%, and the refining rate was 90.05% after bleaching by activated clay,and the bleaching rate of rice bran oil was 47%, the content of oryzanol was 2.10%, and the refining rate was 88.36% after bleaching by silica gel. After combined bleaching of activated clay and silica gel, the color of the rice bran oil improved significantly from Y35,R15 of crude oil to Y5, R0.3 (254 mm Rovipon colourimetric bath).

M‐N‐C single‐atom catalysts (SACs) are promising electrode materials for many electro‐reduction reactions. However, their stability is far from practical applications, and their deactivation mechanism has been rarely investigated. Herein, we demonstrate the structural degradation of M‐N‐C (M=Co, Ni, and Fe) at industrial‐grade current density for long‐term electro‐reduction. Both M‐N and N‐C bonds are broken, resulting in the gradual hydrogenation and dissolution of N in the form of ammonia. The residual M is finally converted to M‐containing core–shell nanoparticles after sequential dissolution, redeposition, and electro‐reduction. The destruction of the M‐N‐C structure and the formation of nanoparticles greatly affect the electrocatalytic performance. Our work highlights the structural degradation and deactivation mechanism of M‐N‐C‐type SACs under strong reductive conditions and provides useful information for inspiring researchers to develop new strategies to improve the electrocatalytic stability of similar types of materials. [ABSTRACT FROM AUTHOR]

Two‐dimensional (2D) non‐layered materials, along with their unique surface properties, offer intriguing prospects for sensing applications. Introducing mechanical degrees of freedom is expected to enrich the sensing performances of 2D non‐layered devices, such as high frequency, high tunability, and large dynamic range, which could lead to new types of high performance nanosensors. Here, we demonstrate 2D non‐layered nanomechanical resonant sensors based on β‐In2S3, where the devices exhibit robust nanomechanical vibrations up to the very high frequency (VHF) band. We show that such device can operate as pressure sensor with broad range (from 10−3 Torr to atmospheric pressure), high linearity (with a nonlinearity factor as low as 0.0071), and fast response (with an intrinsic response time less than 1 μs). We further unveil the frequency scaling law in these β‐In2S3 nanomechanical sensors and successfully extract both the Young's modulus and pretension for the crystal. Our work paves the way towards future wafer‐scale design and integrated sensors based on 2D non‐layered materials. [ABSTRACT FROM AUTHOR]

The combination of satellite laser ranging (SLR) observations to various low earth orbit (LEO) satellites can enhance the accuracy and robustness of SLR-derived geodetic parameters, benefiting the realization of the International terrestrial reference frames. Observation stochastic models play a critical role in the integrated processing of SLR observations to multiple LEO satellites. The consideration of precision in heterogeneous SLR observations from various satellites is essential. In this study, we aim to improve the combination of multi-LEO SLR observations for geodetic parameters determination by optimizing the stochastic model using variance component estimation (VCE). We perform weekly estimates of the geodetic parameters, including station coordinates, Earth rotation parameters, and geocenter coordinates (GCC), using three years of SLR observations to seven LEO satellites at different orbits. The satellite-dependent, station-dependent, and satellite–station-dependent variance components are separately estimated through VCE processing to refine the stochastic models. Given the fact that the precision of SLR observations significantly differs in satellites and stations, the multiple LEO combination can be significantly improved with the implementation of VCE. Satellite–station-pair-dependent variance components are more suitable to the SLR VCE and the accuracy of station coordinates, pole coordinates, and length of day can be averagely improved by 8.4, 22.6, and 21.9%, respectively, compared to the equal-weight solution. Our result also indicates that the observation insufficiency for some stations may result in an unreliable VCE estimation, and eventually leads to an accuracy degradation for station coordinates. To overcome this deficiency, we adopt the variance components derived from the monthly solutions to build the stochastic model in the weekly solutions. The application of monthly weights can effectively mitigate the accuracy deterioration of station coordinates, improving the repeatability of the station coordinates by 15.9, 14.6, and 9.2% with respect to the equal-weight solution in E, N, and U components. The global geodetic parameters also benefit from this processing. The import of monthly weight decreases the outliers in the GCC series, especially in the X and Y components. [ABSTRACT FROM AUTHOR]

Conventional coalbed methane (CBM) reservoir models for injection fall-off testing often disregard the quadratic pressure gradient's impact. This omission leads to discrepancies in simulating the transient behavior of formation fluids and extracting critical reservoir properties. Accurate determination of permeability, storability, and other properties is crucial for effective reservoir characterization and production forecasting. Inaccurate estimations can lead to suboptimal well placement, ineffective production strategies, and ultimately, missed economic opportunities. To address this shortcoming, we present a novel analytical model that explicitly incorporates the complexities of the quadratic pressure gradient and dual-permeability flow mechanisms, prevalent in many CBM formations where nanopores are rich, presenting a kind of natural nanomaterial. This model offers significant advantages over traditional approaches. By leveraging variable substitution, it facilitates the derivation of analytical solutions in the Laplace domain, subsequently converted to real-space solutions for practical application. These solutions empower reservoir engineers to generate novel type curves, a valuable tool for analyzing wellbore pressure responses during injection fall-off tests. By identifying distinct flow regimes within the reservoir based on these type curves, engineers gain valuable insights into the dynamic behavior of formation fluids. This model goes beyond traditional approaches by investigating the influence of the quadratic pressure gradient coefficient, inter-porosity flow coefficient, and storability ratio on the pressure response. A quantitative comparison with traditional models further elucidates the key discrepancies caused by neglecting the quadratic pressure gradient. The results demonstrate the proposed model's ability to accurately depict the non-linear flow behavior observed in CBM wells. This translates to more reliable pressure and pressure derivative curves that account for the impact of the quadratic pressure gradient. [ABSTRACT FROM AUTHOR]

The Tree-Seed Algorithm (TSA) has been effective in addressing a multitude of optimization issues. However, it has faced challenges with early convergence and difficulties in managing high-dimensional, intricate optimization problems. To tackle these shortcomings, this paper introduces a TSA variant (DTSA). DTSA incorporates a suite of methodological enhancements that significantly bolster TSA's capabilities. It introduces the PSO-inspired seed generation mechanism, which draws inspiration from Particle Swarm Optimization (PSO) to integrate velocity vectors, thereby enhancing the algorithm's ability to explore and exploit solution spaces. Moreover, DTSA's adaptive velocity adaptation mechanism based on count parameters employs a counter to dynamically adjust these velocity vectors, effectively curbing the risk of premature convergence and strategically reversing vectors to evade local optima. DTSA also integrates the trees population integrated evolutionary strategy, which leverages arithmetic crossover and natural selection to bolster population diversity, accelerate convergence, and improve solution accuracy. Through experimental validation on the IEEE CEC 2014 benchmark functions, DTSA has demonstrated its enhanced performance, outperforming recent TSA variants like STSA, EST-TSA, fb-TSA, and MTSA, as well as established benchmark algorithms such as GWO, PSO, BOA, GA, and RSA. In addition, the study analyzed the best value, mean, and standard deviation to demonstrate the algorithm's efficiency and stability in handling complex optimization issues, and DTSA's robustness and efficiency are proven through its successful application in five complex, constrained engineering scenarios, demonstrating its superiority over the traditional TSA by dynamically optimizing solutions and overcoming inherent limitations. [ABSTRACT FROM AUTHOR]

In order to solve the problems of small field of view and unclear detail features of monitoring images in coal mines, a multi-view image stitching method in mine is proposed. Firstly, an improved adaptive histogram equalization method with contrast limitation is used to pre-process the images to highlight the image details and improve the contrast. Secondly, the ORB algorithm is used to extract the image feature points, and the improved Brief algorithm is used to calculate the feature descriptors. Thirdly, the K-nearest neighbor(KNN) algorithm is used to achieve rough matching of feature point pairs, and the random sample consensus(RANSAC) algorithm is used to filter and eliminate the mismatched feature point pairs, and the optimal perspective transformation matrix is solved to transform the coordinates of the pixel points of the image to be matched. Finally, the hat function weighted average fusion algorithm is used to stitch and fuse the fixed image and the image to be matched. The experimental results show that compared with the speeded up robust features(SIFT) and KAZE algorithms, the ORB algorithm reduces the number of feature points extracted for a single image by 48% and 33% respectively, which improves the effective feature point extraction capability. The feature point extraction time is reduced by 17% and 34% respectively, which improves the calculation efficiency. The images stitched by this method avoid the phenomenon of cracks and black lines at the joints, the image transition is natural, and the definition is high.