基于容量增量变化量曲线的锂电池健康状态估计

提出了基于容量增量变化量曲线的电池健康状态估计新方法,该曲线具有随电池老化平移不明显的特点,并有利于特征参数的确定和提取。利用随机森林方法分析该曲线上各个函数值的重要性,根据重要性筛选出曲线上可以作为表征参数的函数值,并使用神经网络建立起表征参数与健康状态的映射关系。利用马里兰大学老化数据验证该方法的平均绝对百分比误差为0.77%,健康状态估计值误差基本控制在2%以内,具有较高的精度。     

Effects of roughness on hydrodynamic characteristics of a submerged floating tunnel subject to steady currents

The effects of surface roughness as induced by marine fouling on the hydrodynamic forces on a submerged floating tunnel (SFT) are experimentally and numerically investigated in detail at Reynolds numbers Re = 8.125 × 10³–5.25 × 10⁴. A sensitivity analysis to different roughness parameters including roughness height, skewness, coverage ratio, and spatial arrangement is performed. In addition, an optimized parametric cross-section for an SFT is proposed, and the hydrodynamic performance of the parametric shape and circular SFT cross-section shape with roughness elements is compared. The pressure distribution along the SFT, flow separation and wake characteristics are analyzed to provide a systematic insight into the fundamental mechanism relating the roughness parameters and flow around an SFT. In order to better understand the nonlinear relationships among structural geometry, roughness parameters, flow states, and structural response, an artificial intelligence method using Random Forest (RF) for feature importance ranking is applied. The results show that with the parametric shape, the hydrodynamic forces on the fouled SFT can be effectively mitigated. The roughness height and coverage ratio affect the equivalent blockage and hence, change flow separation and recirculation length in the wake. Lower skewness of the roughness elements can increase the critical Re by changing the relative roughness parameter. Horizontal arrangement of the roughness elements on an SFT generally results in the largest hydrodynamic forces, compared to staggered and vertical distributions. Throughout the feature importance ranking, the flow regime is found to be the most important feature of the hydrodynamics of the SFT. In addition, the SFT cross-section shape and roughness coverage ratio play a dominant role.     

基于SVD的地铁盾构洞门钢环空间形态参数计算

为解决现有地铁盾构隧道洞门钢环空间形态参数计算方法步骤繁琐、需要初始值等问题，在分析空间平面方程和球面方程求解系数特点的基础上，提出利用奇异值分解（SVD）法一次性解算出球面方程参数和洞门钢环所在空间平面单位法向量，再利用上述参数直接计算洞门钢环的中心位置、半径、平整度、圆度等空间形态参数。该过程无需初始值和迭代计算，更为便捷高效。为消除观测点中可能存在的粗差，进一步结合方差比值检验法实现粗差的探测。应用结果表明： SVD法计算结果与商业软件一致，即使存在粗差，利用方差比值检验法也可以准确定位粗差，二者结合具有较好的实用价值。