基于Hamming距离和量子搜索算法的联想分类器设计

针对现有联想分类器不能存储重复样本的问题,提出一种基于Hamming距离和量子搜索算法的量子联想分类器设计方法,并给出联想分类器存储和分类的线路图.该方法需提前准备5组量子比特,分别对Hamming距离、输入样本、模式样本、类别和序号进行编码.首先,根据样本总体N,计算联想分类器所需的量子位数,再利用量子旋转门和Hadamard门将初态为|0〉的量子位旋转为恰好包含N个基态的均衡叠加态;其次,根据待存储样本的类别和值,将剩余两组初始状态为|0〉的量子位通过可控操作转换为相应的量子基态;最后,基于量子最小搜索的分类方法,计算输入样本与所有存储样本之间的Ha mming距离,再使用固定相位Grover量子搜索算法搜索这些Hamming距离的最小值,最小值对应存储样本的类别即为输入样本的类别,具体的分类结果可通过测量寄存器中的量子态得到.

基于双Hamming的NURBS曲线插补算法

针对传统非均匀有理B样条(NURBS)曲线插补中,由于理想插补步长和实际插补步长误差较大引起速度波动的问题,提出一种基于双Hamming的NURBS曲线插补算法。推导差分和导数定义分别变换Hamming方法,得到插补参数的预估、校正公式,再判断校正公式所得插补参数值是否满足速度波动率要求,满足则可继续进行插补,否则利用密勒-牛顿迭代公式对插补参数值进行迭代校正。使用四阶Runge-Kutta方法对初始两插补参数进行求解,并推导插补数值方法关于插补周期的稳定性约束条件。最后对该算法进行仿真,仿真结果表明该算法相较于二阶Taylor、四阶R-K和Adams隐显预估校正算法,平均速度波动率减少30.54%~65.57%,迭代次数减少1.23%~13.12%,证明了该算法在保证较小迭代次数的同时减少了进给速度波动率。

Dynamic changes in physicochemical property,biogenic amines content and microbial diversity during the fermentation of Sanchuan ham

Sanchuan ham is appreciated in Yunnan Province,China,for its characteristic flavor and taste,while the microbial community structure and biogenic amines content remain unclear during fermentation processes.In this study,we explored the physicochemical property,biogenic amines concentration and microbial diversity of external and internal Sanchuan ham by high-throughput sequencing during the processing of Sanchuan ham.Results showed that the nitrite remained at a stable level of 0.15 mg/kg which was significantly lower than the national health standard safety level of 20 mg/kg.In addition,compared with fresh hams,the content of total free amino acids in ripe Sanchuan ham has grown 14 folds;sour and bitter were the main tastes of Sanchuan ham.Notably,the concentration of cadaverine was the highest of all biogenic amines during the entire fermentation period.At the bacterial phyla level,Firmicutes and Actinobacteria were the two main phyla,while at the genus level,Staphylococcus was a significant strain throughout the whole fermentation.Moreover,the dry stage has a great impact on the succession change of microbial community structure.Simultaneously,the change trends and composition of bacteria in the interior have slight discrepancies with those of the exterior of Sanchuan ham.

2-Hamming图的强Menger边连通容错性

设G是连通图.若G的任一对顶点u,v之间有min{d(u),d(v)}条边不交的路,则称连通图G为强Menger边连通的.设G是强Menger边连通图,m是非负整数,若对任意满足|F|≤m的边子集F,G-F都是强Menger边连通图,则称G是m-边容错强Menger边连通图.证明了2-Hamming图H(n,k,2)是(4n-2)-边容错强Menger边连通的,其中n≥2,k≥5.

基于K-Means聚类的粒子群优化CNN-BiGRU-HAM发动机剩余使用寿命预测方法

飞机在多种工况条件下运行时,发动机退化特征复杂性不断增加,导致发动机剩余寿命预测精度低。针对此问题,提出一种基于聚类分析的端到端剩余寿命(RUL)预测方法。采用K-Means聚类方法对发动机的多种工况和运行条件进行分组;再利用卷积神经网络(CNN)提取反映剩余寿命复杂动态变化的高维特征,将结果输入到双向门控循环单元(BiGRU)中学习特征之间的变化规律,设计并引入了新的混合注意力机制(HAM),充分考虑变量之间的关系,对重要特征信息赋予更大的权重,同时抑制冗余信息的影响;然后进行非线性变换,获得RUL预测结果;最后使用粒子群优化算法对神经网络的超参数进行调优。采用美国航天局NASA研究中心提供的涡轮发动机模拟数据集验证所提网络模型的有效性。结果表明:对于多工况运行条件,所提方法的均方根误差相比于CNN、LSTM、BiLSTM、CNN-LSTM分别降低了49.2%、37.1%、33.6%、24.8%,有效提升了模型的预测精度。

Effects of potassium lactate on sensory attributes,bacterial community succession and biogenic amines formation in Rugao ham

To deepen the understanding in the effect of potassium lactate on the sensory quality and safety of Rugao ham,sensory attributes,physicochemical parameters,total volatile basic nitrogen(TVBN),microorganism community and biogenic amines of Rugao ham manufactured with different potassium lactate levels(0%,0.5%,1%,2%)were investigated;the relationship between microbial community and the formation of TVBN and biogenic amines was further evaluated.With the increase of potassium lactate from 0%to 2%,the increased sensory scores and the decreased total aerobic bacterial count and TVBN were observed;the abundance of Staphylococcus increased,while the content of Halomonas decreased.LDA effect size(LEf Se)and correlations analysis showed that Staphylococcus equorum and Lactobacillus fermentum could be the key species to improve sensory scores and decrease biogenic amines and TVBN.Metabolic pathway analysis further showed that amino acids metabolism and nitrogen metabolism were mainly involved in decreasing TVBN and biogenic amines in the treatment of 2%potassium lactate.

Stability and transepithelial transport of oligopeptide(KRQKYD)with hepatocyte-protective activity from Jinhua ham in human intestinal Caco-2 monolayer cells

The study evaluated the stability of an oligopeptide(Lys-Arg-Gln-Lys-Tyr-Asp,KRQKYD)and its transport mechanism by simulating gastrointestinal digestion and a model of human intestinal Caco-2 monolayer cells in vitro.In this study,the effects of environmental factors(temperature,pH and NaCl concentration)and simulated gastrointestinal digestion on the stability of KRQKYD were evaluated by indicators of the levels of alanine transaminase(ALT),aspartate transaminase(AST)and malondialdehyde(MDA)in an alcoholinduced hepatocyte injury model.The results showed that KRQKYD still maintained satisfactory hepatocyteprotective activity after treatment with different temperatures(20-80℃),pH(3.0-9.0),NaCl concentration(1%-7%)and simulated gastrointestinal digestion,which indicated that KRQKYD showed good stability to environmental factors and simulated gastrointestinal digestion.Furthermore,the intact KRQKYD could be absorbed in a model of Caco-2 monolayer cells with a P_(app)value of(9.70±0.53)×10^(-7)cm/s.Pretreatment with an energy inhibitor(sodium azide),a competitive peptide transporter inhibitor(Gly-Pro)and a transcytosis inhibitor wortmannin did not decrease the level of transepithelial KRQKYD transport,indicating that the transport mechanism of KRQKYD was not associated with energy dependent,vector mediated and endocytosis.The tight junction disruptor cytochalasin D significantly increased the level of transepithelial KRQKYD transport(P<0.05),suggesting that intact KRQKYD was absorbed by paracellular transport.

Variation of microbiological and small molecule metabolite profiles of Nuodeng ham during ripening by high-throughput sequencing and GC-TOF-MS

The internal microbial diversity and small molecular metabolites of Nuodeng ham in different processing years(the first,second and third year sample)were analyzed by high-throughput sequencing technology and gas chromatography-time of flight mass spectrography(GC-TOF-MS)to study the effects of microorganisms and small molecular metabolites on the quality of ham in different processing years.The results showed that the dominant bacteria phyla of Nuodeng ham in different processing years were Proteobacteria and Firmicutes,the dominant fungi phyla were Ascomycota and Basidiomycota,while Staphylococcus and Aspergillus were the dominant bacteria and fungi of Nuodeng ham,respectively.Totally,252 kinds of small molecular metabolites were identified from Nuodeng ham in different processing years,and 12 different metabolites were screened through multivariate statistical analysis.Further metabolic pathway analysis showed that 23 metabolic pathways were related to ham fermentation,of which 8 metabolic pathways had significant effects on ham fermentation(Impact>0.01,P<0.05).The content of L-proline,phenyllactic acid,L-lysine,carnosine,taurine,D-proline,betaine and creatine were significantly positively correlated with the relative abundance of Staphylococcus and Serratia,but negatively correlated with the relative abundance of Halomonas,Aspergillus and Yamadazyma.

基于一种相对Hamming距离的入侵检测方法——RHDID

首先分析了传统入侵检测方法的不足 ,即误用入侵检测方法难于检测新形式的入侵 ,异常入侵检测方法难于建立合理有效的正常行为特征和检测方法 .然后 ,通过对特权进程的系统调用和参数序列的研究 ,提出了一种相对Hamming距离入侵检测方法 (RHDID) .应用RHDID检测入侵不仅能有效降低漏报率和误报率 ,而且使实时入侵检测成为可能 .最后 ,原型系统证实了该方法的可行性 。

基于相对Hamming距离的Web聚类算法

针对Web使用挖掘中聚类结果准确性不高的问题,提出了一种改进的基于相对Hamming距离和类不一致度的聚类算法。该算法首先以Web站点的URL为行、以UserID为列建立关联矩阵,元素值为用户的访问次数;然后,对所建立关联矩阵的列向量或行向量进行相似性度量,获得相似客户群体或相关页面。实验表明,该算法具有较高的准确性。

基于Hamming距离的超立方体网络的寻径算法

本文利用Ｈａｍｍｉｎｇ距离概念，提出两种基于Ｈａｍｍｉｎｇ距离的确定性寻径算法—Ｅ１－ｃｕｂｅｒｏｕｔｉｎｇ和Ｅ２－ｃｕｂｅｒｏｕｔｉｎｇ，分析这两种算法的特点．为了克服确定性寻径算法易出现通道拥挤现象，本文综合上述两种算法，提出一种自适应寻径算法—Ｅ３－ｃｕｂｅｒｏｕｔｉｎｇ．

湖泊营养状态的Hamming贴近度评价法

本文用线性内插构造湖泊富营养化评价参数的隶属函数,在此基础上应用Ham-ming 贴近度评价湖泊的营养状态。该方法用于全国23个湖泊的实例分析,并与其它营养化评价法的评价结果比较表明该方法评价结果合乎实际,隶属函数设计简单,计算方便,具有可比性。

Hamming切趾技术提高体光栅光谱合成功率的研究

体光栅光谱合成技术是获得高功率激光输出的一种有效途径,体光栅衍射旁瓣是影响合成光束数目的主要因素。采用了Hamming切趾技术对体光栅旁瓣进行抑制,建立了Ham-ming切趾体光栅的折射率分布模型,分析了Hamming切趾体光栅的衍射特性,给出了Ham-ming切趾体光栅光谱合成效率公式,分析了切趾光栅对光谱合成效率的影响。计算结果表明:体光栅切趾后有效减小了体光栅对相邻合成光束的衍射损耗,切趾后,在20nm的带宽内,谱合成光束的数目由13束增加为20束,谱合成效率达75.3%,光谱合成功率提高为切趾前的1.5倍。

基于支持向量机与Hamming距离的虹膜识别方法

针对传统的虹膜识别方法侧重于特征提取这一现象,提出了一种侧重于模式匹配的识别算法,即基于支持向量机(Support Vector Machine,SVM)和Hamming距离的虹膜识别方法.该算法首先对采集到的虹膜图像进行预处理,准确定位出虹膜,并对其进行归一化处理;然后使用Log-Gabor滤波器提取虹膜纹理特征,在得到虹膜特征编码后,用SVM和Hamming距离方法进行模式匹配.在CASIA虹膜库上的实验结果表明:与经典的识别方法相比,该方法识别率达到了99.63%,错误接受率(FAR)和错误拒绝率(FRR)分别降到了0.02%和0.35%.

二元码的平均Hamming距离和方差

通过对二元ｎ长码Ｃ的对偶距离分布的研究，在码字数为奇数的情况下，改进了Ａｌ－ｔｈｏｆｅｒ－Ｓｉｌｌｋｅ［１］和［２］文关于Ｃ的码字间平均Ｈａｍｍｉｎｇ距离及其均方差的不等式，并在码字数为２ｎ－１或２ｎ－１－１时，确定了码Ｃ的最小平均距离及其均方差的精确值．

应用模糊Hamming网络进行企业客户信用评估

根据资信评估理论，结合我国企业实际，设计企业客户信用评估的指标体系，提出 用模糊Ｈａｍｍｉｎｇ网络进行客户信用等级评定．实证结果表明：与传统方法（ＢＰ网络、模 糊Ｃ一均值算法）相比，前者具有学习速度快，结果稳定，特别适合于具有模糊性概念的 信用评估问题．

基于Hamming神经网络的通信网可靠性指标评价模型

1.引言目前,多目标综合评价方法中计算各评价指标权重的方法根据计算权系数时原始数据的来源不同可以分为两大类:一类是主观赋权法,其原始数据主要根据评价者主观上对各指标的重视程度来决定权重的方法,如吉林法、AHP法、专家调查法等。目前,对于主观赋权法的研究比较成熟,该类方法的优点是专家可以根据实际问题,较为合理地确定各指标之间的排序,也就是说尽管主观赋权法不能准确确定各指标的权重,但在通常情况下可以在一定程度上有效地确定各指标按重要程度给定的权重的大小顺序,同时解释性好。它的主要缺点是主观随意性大,而且这一点并未因采取诸如增加专家数量、仔细挑选专家等措施而得到根本的改善,因而。

基于加Hamming窗FFT和db24小波包变换的电力系统谐波分析

目前传统谐波分析方法主要是傅里叶变换,但是它存在频谱泄露、识别定位慢等问题。结合小波包变换良好的时频局部化特性和加窗傅里叶变换良好的频域分析特性,提出一种基于加窗FFT和db24小波包变换相结合的频谱分析方法。首先利用加Hamming窗傅里叶变换对信号进行分析,得到稳态部分准确的频谱,再利用db24小波包分析对信号的奇异点进行识别和提取,定位暂态信号发生的起始和终止时间,最后对暂态谐波发生时段进行加窗傅里叶变换,求出暂态信号的幅值。MATLAB仿真结果表明,该方法可以准确分析电力系统中的稳态和暂态谐波。

Hamming码和延长Hamming码的周期分布

Hamming码是一类特殊的线性码 .该文对Hamming码和延长Hamming码的周期分布作进一步的分析 ,首次利用延长Hamming码是第 1阶R -M码的对偶码 ,给出了延长Hamming码的周期分布的表达式 .

公式真度的Hamming距离表示形式与分解定理

首先应用模糊集截集的方法,给出了多值逻辑系统L_n中广义重言式的一个等价刻画,并利用模糊集间的标准Hammin距离,定义了公式间的Hamming距离、Hamming相似度与Hamming真度,给出了计量逻辑学基本概念的Hamming距离表示方法.然后给出了计量逻辑学中公式真度的一个分解定理,这个定理指出在计量逻辑学中,任意一个公式的真度等于一些互不相容的公式的真度之和,而公式φ本身则逻辑等价于这些公式的并.最后应用所提方法定义了广义MP问题的三-Ⅰ真度解,并讨论了三-Ⅰ真度解的存在性问题.