基于数据质量评价的目标估计

针对目标估计过程需要大量人工参与、自动化程度低的问题，提出了基于数据质量评价的目标估计方法。利用目标数据质量评价方法，对不同传感器得到的目标数据质量进行科学、有效的测度和评价，并根据质量得分动态调整各数据源在目标估计过程中所占的权重，从而减少人工干预，提高目标估计效能。仿真试验结果证明了该方法的有效性。     

基于单光子计数激光雷达的Bayesian检测研究

在单光子计数激光雷达检测领域,目前的检测方法在低信噪比情况下虚警概率会增加,同时也无法适应噪声变化的问题。针对这些问题,提出了一种基于Bayesian的检测方法,该方法首先通过雷达方程估计回波信号光子数的范围,将其作为先验信息,而后结合二项分布建立了累计概率模型,基于Bayesian判决准则计算得到检测阈值,此阈值能够在检测概率与虚警概率中间择其平衡。这种方法不仅克服了低信噪比检测困难的情况,还减少了先验信息的获取难度。实验结果表明,对比固定阈值其虚警概率降低了10倍。对比“恒虚警”其检测概率提高了约20。验证了方法具有良好的检测效果,具备一定的可操作性。     

基于改进MobileFaceNet的人脸识别方法北大核心

针对目前大多数人脸识别算法参数多、计算量大,难以部署到移动端和嵌入式设备中的问题,提出了一种基于改进MobileFaceNet的人脸识别方法。通过对MobileFaceNet模型结构的调整,将bottleneck模块优化为sandglass模块,改良深度卷积和逐点卷积的相对位置,适当增大sandglass模块的输出通道数,从而减少特征压缩时的信息丢失,增强人脸空间特征的提取。实验结果表明:改进后的方法在LFW测试数据集上准确率达99.15%,模型大小和计算量分别仅为原算法的61%和45%,验证了所提方法的有效性。     

碳中和背景下配电网差异化节能降耗潜力评估分

为了提高配电网差异化节能降耗效果，解决现有潜力评估方法存在的应用性能差的问题，提出碳中和背景下配电网差异化节能降耗潜力优化评估方法。根据配电网的空间结构，构建相应的等值电路模型。在该模型下，从设备损耗和运行附加损耗2个方面计算配电网的损耗量。根据损耗量计算结果，确定配电网差异化碳中和节能降耗方式。从静态和动态2个角度设置潜力评估指标，通过指标数据处理、指标权重求解等步骤，得出配电网差异化节能降耗潜力的综合量化评估结果。将设计潜力评估方法应用到配电网的差异化节能降耗改造工作中，能够有效降低配电网的实际线损量、降低区域损耗费用，并具有较高的应用价值。     

电子雷管技术的推广应用助力爆破器材安全管理工作

电子雷管技术提高了爆破器材的本质安全。通过对电子雷管技术发展历程的介绍,对其推广应用的可行性进行了深入分析,论述了电子雷管技术的推广应用对爆破器材安全管理工作的促进作用,并提出了今后推广应用中还需加强的工作。     

Natural hydrogen emanations in Namibia: Field acquisition and vegetation indexes from multispectral satellite image analysis

Natural hydrogen exploration is now active in various places of the world. Onshore, correlation between natural H₂ generation and the presence of iron rich rocks especially from Archean and Neoproterozoic cratons have been observed. Emanations and accumulations of H₂ have already been confirmed in such geological settings in Australia, South Africa and Brazil. The geological similitude and the presence of numerous sub circular depressions that are a good proxy for hydrogen emanations suggest that hydrogen resources may also exist in Namibia. We present here the results of a data acquisition campaign which allowed us to confirm the presence of natural hydrogen in this country in the vicinity of Neoproterozoic Banded Iron Formation. The H₂ content in the soil, as in Brazil, is variable within the depressions in time and space and is particularly time sensitive across the day. Comparison of the H₂ signal versus time within these two regions shows a similar behavior of the soils with an increase of the H₂ flow at the middle of the day. In addition, these new data allow us to better constrain the morphological characteristics of such H₂-emiting depressions. By using satellite images and digital elevation model we propose a new proxy to differentiate potentially H₂-emiting features from other type of depressions such as Salt Pan. The Landsat multispectral images and their processing through NDVI and SAVI indexes, that highlight a ring of healthy vegetation around the sub circular area with scarce vegetation already observed appear able to discriminate between H₂ emitting structures and other soft depressions.     

Effects of different metal corrosion inhibitors on hydrogen suppression and film formation with Mg–Zn alloy dust in NaCl solutions

To improve the safety of wet dust removal systems for processing magnesium-based alloys, a new method is proposed for preventing hydrogen generation. In this paper, hydrogen generation by Mg–Zn alloy dust was inhibited with six common metal corrosion inhibitors. The results showed that sodium dodecylbenzene sulfonate was the best hydrogen inhibitor, while CeCl₃ enhanced hydrogen precipitation. The film-forming stability of sodium dodecylbenzene sulfonate was tested with different contents, temperatures, Cl^? concentrations and perturbation rates. The results showed that this inhibitor formed stable protective films on the surfaces of Mg–Zn alloy particles, and adsorption followed the Langmuir adsorption model.     

Heterofullerene C48B12-impregnated MOF-5 and IRMOF-10 for hydrogen storage: A combined DFT and GCMC simulations study

The H₂ storage properties of isoreticular metal-organic framework materials (IRMOFs), MOF-5 and IRMOF-10, impregnated with different numbers and types of heterogeneous C₄₈B₁₂ molecules were investigated using density functional theory and grand canonical Monte Carlo (GCMC) calculations. The excess hydrogen adsorption isotherms of IRMOFs at 77 K within 20 bar indicate that suitable number and type of C₄₈B₁₂ molecules play a crucial role in improving the H₂ storage properties of IRMOFs. Among the studied pure and nC₄₈B₁₂ (n = 1, 2, 4, 8) in Ci symmetry impregnating into MOF-5, at 77 K under 6 bar, MOF-5-4C₄₈B₁₂ with a 3.5 wt% and 29.9 g/L hydrogen storage density, and at 77 K under 12 bar, the pure MOF-5 with a 4.9 wt% and 31.0 g/L hydrogen storage density has the best hydrogen storage properties. Whereas, among the studied pure and nC₄₈B₁₂ (n = 1, 2, 4, 8) in S6 symmetry impregnating into IRMOF-10, IRMOF-10-8C₄₈B₁₂ always shows the best hydrogen storage properties among the pure and C₄₈B₁₂-impregnated IRMOF-10 at 77 K within 20 bar. IRMOF-10-8C₄₈B₁₂ has a 6.0 wt% and 34.6 g/L hydrogen storage density at 77 K under 6 bar, and has a 7.1 wt% and 41.4 g/L hydrogen storage density at 77 K under 12 bar. The confinement effect of IRMOFs on C₄₈B₁₂ molecules, and steric hindrance effect of C₄₈B₁₂ molecules on IRMOFs mainly affects the H₂ uptake capacity by comparing the absolute H₂ molecules in individual IRMOFs units, C₄₈B₁₂ molecules, and IRMOFs-nC₄₈B₁₂ compounds. The absolute hydrogen adsorption profiles show that eight C₄₈B₁₂ molecules impregnating into MOF-5 can exert obvious steric effects for H₂ adsorption. The saturated gravimetric and volumetric H₂ densities of IRMOF-10-8C₄₈B₁₂ higher than those of MOF-5-8C₄₈B₁₂ due to with larger free volume.     

Influencer-defaulter mutation-based optimization algorithms for predicting electricity prices

Efficient electricity price forecasting plays a significant role in our society. In this paper, a novel influencer-defaulter mutation (IDM) mutation operator has been proposed. The IDM operator has been combined with six well-known optimization algorithms to create mutated optimization algorithms whose performance has been tested on twenty-four standard benchmark functions. Further, the artificial neural network is integrated with mutated optimization algorithms to solve the electricity price prediction problem. The policymakers can identify appropriate variables based on the predicted prices to help future market planning. The statistical results prove the efficacy of the IDM operator on the recent optimization algorithms.