面对日益复杂的工业过程，传统传感器无法得到有效应用，重要变量无法准确建 模，重要过程无法得到有效优化和诊断，高斯过程模型的提出和应用为工业过程建模、优 化和控制提供了一个广阔的思路，并可兼顾描述不确定信息． 文中针对工业过程的复杂特 性，不仅综述了高斯过程的基本方法以及存在的主要问题，而且归纳了其基本建模、优化、 控制及故障诊断的应用和研究成果． 最后，结合国际上发展及作者的实践经验总结并展望 了高斯过程模型在工业过程中的应用前景和发展趋势．

As industrial processes have become more and more complex,the traditional sensors are unavailable,and it is difficult to properly model the critical variables and efficiently optimize or diagnose the important parts of a process.The Gaussian process model provides an alternative to the modeling,optimization and control of industrial processes under the constraints of uncertainties.In this paper,aiming at the complexity of industrial processes,the pros and cons of the Gaussian process model are investigated,and its application to the modeling,optimization,control and fault diagnosis of industrial processes as well as the corresponding research results is generalized.Final- ly,the application prospects and development orientations of the Gaussian process model in industrial processes are summarized and forecasted by combining the international research results with the authors' practical experience.