磷酸铁锂(LiFePO_(4),LFP)与镍钴锰酸锂(LiNi_(x)Co_(y)Mn_(2)O_(2),NCM)电池串联构建的混合动力电池系统,是突破传统单一化学体系瓶颈的关键技术。然而,混装电池包中LFP电池具有平坦的电压平台特性,导致全工作区间的荷电状态(state of ...磷酸铁锂(LiFePO_(4),LFP)与镍钴锰酸锂(LiNi_(x)Co_(y)Mn_(2)O_(2),NCM)电池串联构建的混合动力电池系统,是突破传统单一化学体系瓶颈的关键技术。然而,混装电池包中LFP电池具有平坦的电压平台特性,导致全工作区间的荷电状态(state of charge,SOC)估算精度受限,且在多算法切换时易出现SOC跳变现象。为此,本工作提出一种基于开路电压(open circuit voltage,OCV)曲线区间自适应划分的分段融合SOC估算方法。首先,考虑到LFP电池OCV斜率变化特征,设计了分段平滑策略,在高斜率区保持电压特征,在平台区增强平滑效果,并根据平滑OCV曲线的一阶差分斜率,设定自适应斜率阈值,将放电区间划分为前端高斜率区、中间平台区与后端高斜率区,为SOC算法选择提供明确依据;其次,构建分段估算框架:在高斜率区采用改进自适应扩展卡尔曼滤波进行SOC动态跟踪,在平台区则利用混合包中NCM电池的SOC进行映射估算。针对算法切换点SOC跳变问题,进一步提出梯度敏感的S型融合算法(gradient-sensitive adaptive blending,GSAB),该算法通过量化切换点邻域的SOC梯度差异,动态调整融合函数参数以生成平滑过渡权重,抑制切换点的SOC跳变。结果表明,改进自适应扩展卡尔曼滤波算法在NCM电池上的均方根误差相较于传统扩展卡尔曼滤波算法降低63.70%;GSAB策略有效消除了算法切换时的SOC突变,使过渡区波动降低72.42%。最终,在城市道路循环工况下,LFP电池全区间SOC估算的平均绝对误差与均方根误差分别降至1.08%和1.31%,验证了所提方法能有效提升LFP电池SOC全区间估算精度。展开更多
A simple graph G is induced matching extendable,shortly IM\|extendable,if every induced matching of G is included in a perfect matching of G .The degree conditions of IM\|extendable graphs are researched in...A simple graph G is induced matching extendable,shortly IM\|extendable,if every induced matching of G is included in a perfect matching of G .The degree conditions of IM\|extendable graphs are researched in this paper.The main results are as follows:\;(1) 2n/32n/32n/32n/3 is the minimum integer δ such that every graph with minimum degree at least δ is IM\|extendable,where n=|V(G)|≥6 is even.\;(2) is the minimum integer δ such that every graph with minimum degree at least δ is IM\|extendable,where n=|V(G)|≥6 is even.\;(2) (2n-2)/3 is the minimum integer k such that, for k′≥k ,every k ′ regular graph with vertex number n is IM\|extendable,where n=|V(G)| is even, n ≥8 and n ≠10.展开更多
针对工业生产场景下大型轴流风机滚动轴承受强背景噪声干扰影响导致的故障信号特征提取难题,提出一种基于延伸奇异值分解包(extended singular value decomposition packet,ESVDP)的自适应轴承故障信号提取方法。利用自回归模型,引入自...针对工业生产场景下大型轴流风机滚动轴承受强背景噪声干扰影响导致的故障信号特征提取难题,提出一种基于延伸奇异值分解包(extended singular value decomposition packet,ESVDP)的自适应轴承故障信号提取方法。利用自回归模型,引入自回归功率谱,基于信号能量分布分析实现对原算法的分解精度参数的自适应设定。引入峭度指标作为故障特征选择依据,从轴承故障信号本身的特征来更多地保留机械振动信号中的故障信息,结合峭度指标实现递推分解层数参数的设置,自适应分解出故障信号,有效保留包含故障信息的瞬态冲击成分。优化ESVDP分解结构,减少不必要的计算冗余,提升计算效率的同时对故障信号进行筛选,从而准确实现轴流风机轴承故障的诊断。仿真信号和试验结果均表明该方法在强干扰分量下可有效提取轴流风机轴承的故障特征频率,实现轴流风机轴承的故障诊断。展开更多
文摘A simple graph G is induced matching extendable,shortly IM\|extendable,if every induced matching of G is included in a perfect matching of G .The degree conditions of IM\|extendable graphs are researched in this paper.The main results are as follows:\;(1) 2n/32n/32n/32n/3 is the minimum integer δ such that every graph with minimum degree at least δ is IM\|extendable,where n=|V(G)|≥6 is even.\;(2) is the minimum integer δ such that every graph with minimum degree at least δ is IM\|extendable,where n=|V(G)|≥6 is even.\;(2) (2n-2)/3 is the minimum integer k such that, for k′≥k ,every k ′ regular graph with vertex number n is IM\|extendable,where n=|V(G)| is even, n ≥8 and n ≠10.
文摘针对工业生产场景下大型轴流风机滚动轴承受强背景噪声干扰影响导致的故障信号特征提取难题,提出一种基于延伸奇异值分解包(extended singular value decomposition packet,ESVDP)的自适应轴承故障信号提取方法。利用自回归模型,引入自回归功率谱,基于信号能量分布分析实现对原算法的分解精度参数的自适应设定。引入峭度指标作为故障特征选择依据,从轴承故障信号本身的特征来更多地保留机械振动信号中的故障信息,结合峭度指标实现递推分解层数参数的设置,自适应分解出故障信号,有效保留包含故障信息的瞬态冲击成分。优化ESVDP分解结构,减少不必要的计算冗余,提升计算效率的同时对故障信号进行筛选,从而准确实现轴流风机轴承故障的诊断。仿真信号和试验结果均表明该方法在强干扰分量下可有效提取轴流风机轴承的故障特征频率,实现轴流风机轴承的故障诊断。
