Heat source reconstruction from noisy temperature fields using an optimised derivative Gaussian filter

Infrared image filtering applied to the restoration of the convective heat transfer coefficient distribution in coiled tubes

“…To this aim, the temperature distribution on the external sur− face of a coiled wall was obtained using a high−precision infrared camera and used as input data to the inverse heat conduction problem in the wall with a filtering approach. To reconstruct the second derivative of the signal, a Gaussian filter was used because it was proven to be the most effec tive filter for this type of application [9][10][11]. The choice of the cut−off frequency, which is fundamental to the success of the estimation procedure, was based on the discrepancy principle.…”

“…Thus, it is widely used to enhance image quality in graphics software. To smooth the noisy temperature data, the averaging property of the Gaussian kernel was experimented by Murio [11], Bozzoli et al [13] and Delpueyo et al [10]. The transfer function of the Gaussian filter in the 1−D frequency domain is expressed as follows where u is the frequency and u c is the cutoff frequency value as shown in Fig.…”

“…To bypass these difficulties, which are particularly critical when infrared thermography is used as the temperature−measuring technique, many methods based on experimental−data processing have been suggested and validated in the literature. Among these techniques are the conjugate gradient iterative method; Laplace transform method; sequential function specification method; regulari− sation methods, such as Tikhonov regularisation; mollifica− tion method; truncated singular value decomposition method; and filtering technique method [6][7][8][9][10][11][12].…”

“…Noise reduction techniques of infrared images have been the object of several studies (for example, [10,15,16]), al− though the lack of practical applications of these approaches to engineering problems remains.…”

High Cycle Fatigue Behavior Evaluation of Q235 Steel Based on Energy Dissipation

“…效机理的研究得到快速发展 [4] 。材料损伤过程中的 温度场信号多用于分析损伤演化的物理过程 [5][6] 、验 证与损伤萌生相关的自生热规律 [7][8] 、预测材料的宏 观疲劳参数 [9][10] 和构建热力学能耗控制方程 [11][12][13] 。 但是，考虑到温度场信号并不是材料的固有特性， 易受材料本身的热扩散性质和热边界条件影响，因 此通过疲劳温升信息直接研究材料的疲劳性能往往 充满争议。于是，基于热力学理论框架来确定与损 伤演化过程密切相关的能量耗散成为完善疲劳性能 快速评估方法的一种途径。然而利用热力学方程来 求解能量耗散源是典型的病态热力学反问题，必须 通过多种合理假设将空间三维问题简化为二维、一 维甚至零维情况。CHRYSOCHOOS 等 [14] 通过对薄 板试件的研究，建立了二维和一维热传导方程，并 结合热像数据的拟合来计算固有耗散源，该方法得 到了广泛的应用。DELPUEYO 等 [15] 利用偏微分高 斯滤波器来处理热像数据中的噪声以重构热源场的 分布，并通过误差分析和试验验证了该种方法，丰 富了基于热像数据来反算热源的思路。EZANNO 等 [16] 融合了疲劳热像法、宏微观力学理论和概率统 计学方法来研究材料的分散性特征，建立了一种预 测材料概率疲劳寿命的方法。李源等 [17] 建立了高周 疲劳过程中薄板试样的热扩散方程，对单个循环周 期内的固有耗散进行计算，提出了一种新的高周疲 劳性能评估理论。樊俊铃等 [18][19] [16] ( ) [16] ( ) , , , , , , …”