Water-cooling wall tube leakage accident is an important factor affecting the safe and economic operation of power station boilers. In this article, the acoustic method is applied to locate the leakage from a simulated water-cooling wall tube, and several key technologies are studied in order to identify the precise location of leakage source. Sound wave propagation path in ‘nonuniform temperature field is obtained to overcome the shortcoming that the traditional location technology always assumed that sound waves propagate along the straight line. Considering the refraction effect of sound waves, the reconstruction method based on radial basis function approximation with polynomial reproduction (RBF-PR) is used to reconstruct the temperature field. A generalized cross correlation with second correlation is applied to estimate the time delay of arrival (TDOA). According to the temperature field measured by acoustic method and the propagate time of sound waves, the location fingerprint method including the statistical correlation characteristics between the coordinate of fingerprint points and TDOA are constructed. Radial basis function artificial neural network (RBF ANN) is used to estimate the coordinate of leakage point. Numerical simulations and experimental study are implemented to evaluate the effectiveness of the proposed location system. An acoustic location test platform is constructed to study the location accuracy in different working conditions. The results indicate that considering the refraction effect, the reconstruction performance of temperature field is superior than that without the refraction effect, and acoustic method with RBF ANN algorithm can gain the leakage location with high accuracy and good antinoise ability in the above high-quality reconstructed nonuniform temperature distribution.

中文翻译：

---

基于声学方法和人工神经网络的模拟水冷壁管泄漏定位

水冷壁管泄漏事故是影响电站锅炉安全经济运行的重要因素。本文采用声学方法对模拟水冷壁管的泄漏进行定位，并研究了几项关键技术，以实现泄漏源的精确定位。获得了非均匀温度场中的声波传播路径，克服了传统定位技术一直假设声波沿直线传播的缺点。考虑到声波的折射效应，采用基于径向基函数近似多项式再现（RBF-PR）的重构方法来重构温度场。应用具有二次相关的广义互相关来估计到达时间延迟 (TDOA)。根据声学法测得的温度场和声波的传播时间，构建了包含指纹点坐标与TDOA统计相关特性的位置指纹法。径向基函数人工神经网络（RBF ANN）用于估计泄漏点的坐标。实施数值模拟和实验研究来评估所提出的定位系统的有效性。搭建声学定位测试平台，研究不同工况下的定位精度。结果表明，考虑到折射效应，