基于音频特征的工程车辆工况识别研究

高效率地使用工程车辆是工程项目管理中节约成本的有效方法，无人监管环境下工程车辆的工况识别，是实现工程车辆高效率使用的有效手段。目前以GPS等技术为核心的车辆智能管理系统未对工程车辆进行工况识别，提出一种基于GRU循环神经网络的工程车辆工况识别方法，通过对工程车辆在不同工况下产生的音频信号进行分析，从中提取Mel倒谱系数作为主要特征，构建GRU循环神经网络模型进行训练和识别。实验结果表明，该方法可以实现对工程车辆工况的有效识别。     

Approximation Optimal Vibration Control with Applications to Jacket Offshore Platforms with Nonlinear Dynamic Characteristics

International Journal of Control, Automation and Systems - The vibration control problem of offshore jacket platforms is studied. The model of offshore platforms with nonlinear interactions of the...     

Real time power management strategy for fuel cell hybrid electric bus based on Lyapunov stability theorem

The fuel cell/battery durability and hybrid system stability are major considerations for the power management of fuel cell hybrid electric bus (FCHEB) operating on complicated driving conditions. In this paper, a real time nonlinear adaptive control (NAC) with stability analyze is formulated for power management of FCHEB. Firstly, the mathematical model of hybrid power system is analyzed, which is established for control-oriented design. Furthermore, the NAC-based strategy with quadratic Lyapunov function is set up to guarantee the stability of closed-loop power system, and the power split between fuel cell and battery is controlled with the durability consideration. Finally, two real-time power management strategies, state machine control (SMC) and fuzzy logic control (FLC), are implemented to evaluate the performance of NAC-based strategy, and the simulation results suggest that the guaranteed stability of NAC-based strategy can efficiently prolong fuel cell/battery lifespan and provide better fuel consumption economy for FCHEB.     

Improved Algorithm for Solving Inverse Kinematics of Biped Robots

Mobile Networks and Applications - Inverse kinematics is an important basic theory in walking control of biped robot. This study focuses on the parameter setting using the improved algorithm in...     

Cowpea resistance and bioactivity of two diatomaceous earths against Callosobruchus maculatus (F.) (Coleoptera: Chrysomelidae) on three cowpea varieties

The bruchid, Callosobruchus maculatus (F.) is the most important pest of stored cowpea in tropical climates. Experiments were conducted to evaluate the susceptibility of three cowpea varieties and the bioactivity of SilicoSec® and FossilShield® against this insect pest under ambient laboratory conditions at the contents 0, 0.5, 1, 1.5 and 2 g/kg of cowpea. The Dobie susceptibility index for cowpea genotypes was estimated to compare their relative tolerance to this insect pest. Adult mortality was assessed after 1, 2, 4 and 6 days of exposure followed by the evaluation of the F₁ progeny. Population increase, percentage seed damage and weight loss were assessed after three months of storage. The results showed that the CRPS and TN5-78 varieties were highly susceptible to C. maculatus, while Lori variety was moderately susceptible to the weevil. SilicoSec® and FossilShield® caused a significant and increasing bruchid mortality according to the dose rates and periods of exposure. Malathion 5% caused complete mortality after one day exposure time. SilicoSec® was more effective than FossilShield® on the three varieties of cowpea. The greatest dose of SilicoSec® and FossilShield® (2 g/kg) caused 100% mortality on all three varieties; 4 days of exposure in the Lori variety was the shortest exposure time leading to complete mortality. Regarding LD₅₀ and LD₉₅ values, it was observed that C. maculatus adults are more susceptible to SilicoSec® than to FossilShield®.The two formulations of diatomaceous earth (1.5 g/kg) reduced the F₁ progeny by more than 70%, and produced less than 10% of seed damage and weight loss for the three varieties of cowpea. The varietal resistance and the use of diatomaceous earth against C. maculatus in storage could represent an alternative to Malagrain® for the protection of the cowpea.     

基于改进GNG算法的燃煤锅炉数据动态特征分析与控制

数据动态特征分析与控制技术是一种重要的数据挖掘手段,燃煤锅炉数据具有明显时序性和漂移性等特点,针对目前数据跟踪与监督算法大多缺乏动态性、实时性和稳定性等问题,设计一种基于改进生长神经气模型(GNG)的自适应聚类模型,建立基于概率、范围搜寻、节点平均距离的节点生成、删除机制,实现对漂移数据实时监控.最后以燃煤锅炉动态数据为分析对象进行实验,实验结果表明该模型和算法对动态漂移数据的实时跟踪能力更强,能对燃煤锅炉动态数据进行准确、有效的监督和控制.     

Influence of fermentation with different lactic acid bacteria and in vitro digestion on the change of phenolic compounds in fermented kiwifruit pulps

The present study aimed to evaluate the effect of fermentation and gastrointestinal digestion of three kinds of fermented kiwifruit pulps with different Lactobacillus (Lactobacillus plantarum, Lactobacillus acidophilus and Lactobacillus casei). The changes in bioactive substances (total phenolic acid, total flavonoid, vitamin C and the viable count), antioxidant capacity (DPPH, ABTS, hydroxyl radical and superoxide anion radical scavenging activity) and phenolic profiles (protocatechuic acid, protocatechualdehyde, chlorogenic acid, caffeic acid and p-coumaric acid) were detected. The result showed, compared to non-fermented kiwifruit pulp, fermentation with LP and LA had higher content of TPA, TF and V_C, as well as antioxidant capacity. Fermentation with LP and the content of protocatechualdehyde, p-coumaric acid and chlorogenic acid were increased. However, after digestion, LP showed more effect in maintaining the content of antioxidants, antioxidant capacity and the viable count rather than LA. During digestion, the content of protocatechualdehyde and p-coumaric acid was increased in fermented samples compared with non-fermented samples. Overall, compared with LA and LC, LP is more suitable for the fermentation of kiwifruit pulp.     

基于DAE-BP神经网络的股票预测研究

股票指标数据种类多、维度高，且指标之间存在多重共线性。为了降低数据的维度、消除指标间的多重共线性和预测股票价格，首先构建了基于受限布尔兹曼机的深度自编码器，实现了高维数据向低维空间的压缩编码。然后基于BP神经网络建立了低维编码序列与股票价格之间的回归模型。实验结果表明，深度自编码器提取特征的能力优于主成分分析法和因子分析法；相比较使用降维前的数据，使用编码后的数据用预测股票价格，模型可以减少计算开销，并且获得更高的预测精度。     

E-voting scheme using secret sharing and K-anonymity

E-voting maybe replaces the traditional voting scheme in the future, however, the security threat must be paid enough attention. In this paper, a novel e-voting scheme is proposed using secret sharing and k-anonymity, which not only satisfies the basic security goals such as the non-cheating, the universal verifiability, the confidentiality, and the anonymity, but also achieves the addition properties including coercion-resistance and unconditional security since the security of the proposed scheme does not rely on any computational hard problem.