Abstract: Based on 400 days-period of operating data of the electrical desalination system, the big data analysis method was used to establish a mapping relationship between the operating parameters and the after-desalination salt content. The model was then applied in intelligent prediction of the after-desalination salt content for a total of 1052 operating samples acquired form Apr. 2019 to Sept. 2020. An average error of 0.42mg/L was found by comparing the prediction salt content with actual value. In order to analysis the influence of operating parameters on salt removing efficiency, a model-based optimization method was proposed. The method was used to study several key factors that affect the salt removing efficiency for electric desalination process. Based on the screening results, several key operating parameters that role the salt removing efficiency for electric desalination system were analyzed and corresponding optimal operation suggestions provided. The after-desalination salt content can be significantly reduced by following suggestions provided with the optimization method.

摘要： 基于电脱盐系统400余天的运行数据，采用大数据分析方法建立了运行参数与脱后盐含量之间的映射关系模型，模型对2019.4~2020.9期间共1052个电脱盐系统的实际运行样本进行了智能预测，与实际监测结果比对预测的平均误差在0.42mg/L。提出了基于模型的电脱盐系统影响因素分析及系统优化方法，利用该模型对影响电脱盐的关键因素进行筛选。依据筛选结果，分析了当前电脱盐系统的关键运行参数，并提出了优化运行建议。依据运行建议可以使电脱盐系统的脱后盐含量显著下降。

关键词: 大数据, 电脱盐, 运行优化, 智能监测, 影响因素, big data, electric desalination, operation optimization, intelligent monitoring, influence factors