从DCS系统得到的实时数据的完整性和准确性是汽轮发电机组优化操作专家系统的基础。针对预测—验证模型的关键问题--计算量和存储量,本文推导了一个实时参数在线预测—验证模型,能很好地预测—验证未来数据。测试结果和实际应用表明该模型适用于汽轮发电机组优化运行专家系统中实时数据的在线动态预测—验证。 The integrity and accuracy of the real-time data obtained from the DCS system is the basis for an expert system for optimizing the operation of the turbine generator set. Aiming at the key problem of prediction-verification model-computation and storage, a real-time parameter online prediction-verification model is derived in this paper, which can predict and verify future data well. The test results and practical application show that the model is suitable for the on-line dynamic prediction-verification of real-time data in the expert system of Turbogenerator optimization operation.