以甲醇合成系统为背景建立一套数据校正方法，以流程模拟数据为基础来识别测量数据的显著误差，改进投影矩阵法来校正测量数据的随机误差；调整精馏塔的部分结构参数，用校正数据进行稳态过程精馏塔模拟优化。结果表明，本文的数据校正方法能够有效识别出测量数据的显著误差，同时校正后的数据组分摩尔流量的方差普遍下降，说明校正效果较好；调整设备参数后的稳态模拟数据与校正数据对比，验证了模拟方法的可靠性，基于此方法提出的节能措施能更加有效地减少能量的损耗。因此本文校正方法可应用于甲醇合成系统的数据校正。

A set of data reconciliation method is established on the background of methanol synthesis system.Gross errors of measured data are identified on the basis of the process simulation data,and random errors are reconciled by the improved projection matrix method.Part of the structural parameters of the distillation column are adjusted,and the reconciled data are used to simulate and optimize the steady-state process for the distillation column.Results show that this data reconciliation method can identify the gross errors of the measure data effectively,and the variance of the component molar flow rate in the reconciled data drops generally,indicating a good reconciliation effect.The reliability of the simulation method is verified by comparing the reconciliation data with the data obtained from the steady-state simulation after adjusting equipment parameters.The proposed energy saving measures based on this method can reduce the loss of energy more effectively.Therefore,this data reconciliation method can be applied to data reconciliation of methanol synthesis system.

王金凡(1995-),男,硕士生