道岔尖轨作为铁路线路中的关键薄弱环节，其健康状况直接关系到列车的行车安全。针对现有方法无法用于道岔尖轨缺陷在线检测的问题，提出基于相控阵列式导波声场调控的尖轨缺陷在线监测方法。针对道岔尖轨复杂外形结构特点，建立非规则变截面结构超声导波无损检测理论模型，揭示局部载荷作用下在尖轨波导中激发的模态及其传播规律；基于相控阵列式导波声场调控理论与技术，实现非规则变截面结构导波聚焦及全断面扫查检测；针对常规导波检测信号在复杂结构中发生畸变、散射等问题，对基于特征空间分解的道岔尖轨导波监测信号智能识别技术展开研究，最终通过将现代无损检测技术和数据识别评价方法相结合，研究面向在役道岔尖轨在线探伤中亟待重点突破的基础理论与关键技术，为实现在役道岔尖轨的快速智能检测与监测提供新理论、新方法和新技术支撑。同时，该项目成果对飞机机翼，风电叶片等大型非规则变截面构件的无损检测和在役监测具有重要的理论和实践意义。

As the key weak part of the railway line, the health condition of switch rails is directly related to operation safety of trains. Aiming at the problem that existing methods cannot be used for on-line detection of defects in switch rails, the method for on-line monitoring of the point rail defects based on sound field controlling of the phased array guided waves is proposed. Owing to the complex shape and structure of switch rails, a nondestructive testing theory model of guided waves for structures with irregular and variable cross-sections is established. Ultrasonic guided waves in switch rails inspired by partial loadings are presented with their propagation patterns. Based on the theory and technology of sound field controlling of the phased array guided waves, guided wave focusing and whole cross-section detection of structures with irregular and variable cross-sections are realized. In order to solve the problem of aberration and scattering in complex structures of conventional guided wave detection signals, researches of intelligent identification technologies of switch rail monitoring signals based on eigenspace decomposition are carried out. Then, by combining modern nondestructive testing techniques with data recognition and evaluation methods, basic theories and kernel technologies which should be broke through prior for online detection of in-service switch rails are studied, providing new theories, methods and technologies for fast intelligent detection and monitoring of in-service switch rails. Meanwhile, achievements of this research have significant theoretical and practical meaning for nondestructive testing and in-service monitoring of giant structures with irregular variable cross-sections like aircraft wings, wind blades and so on.