中文摘要：

溶氧浓度软测量模型是用于预测水产养殖水质中溶氧变化情况的重要手段，对于控制鱼类生长存活至关重要。阐述了建立溶氧浓度软测量模型的目的和意义，综述了溶氧浓度软测量模型的研究现状。溶氧浓度软测量模型分为神经网络和优化算法组合、支持向量机和优化算法组合，总结和分析了这两类建模方法存在的问题，提出了采用混合软测量方法建立溶氧浓度模型的改进方案；最后对溶氧浓度软测量模型未来研究方向进行了简要展望，指出融合多种信号并结合专家知识、机器学习等智能方法，基于机理和数据驱动的混合软测量是建立溶氧浓度模型的发展方向。

英文摘要：

The soft sensing model for dissolved oxygen is an important tool which is used to predict the variations of dissolved oxygen concentration in aquaculture process. It is essential for the growth and survival of fish. To promote the soft sensing research for dissolved oxygen concentration and industrial applications, itsobjectives and significance were introduced, and the present situation of monitoring and detection of dissolved oxygen was also illustrated. According to the related research papers, the current characteristics of dissolved oxygen concentration detection system and problems of soft sensing modelestablishment were summarized and analyzed. Finally, improved strategy using mixed soft sensing method to establish dissolved oxygen concentration model were put forward, and the future research orientations and priorities in this field were briefly discussed.