Abstract Displacement processes accompany every start-up or shut-down of spiral-wound membrane module plants used for the production of food concentrates. In this process, large amounts of product are usually discarded due to dilution with the displacing fluid. In order to reduce product losses by improved process control, the relationships between the key process conditions and the duration of the displacement of non-Newtonian products from spiral-wound membrane modules need to be known. While previous studies assess the impact of single factors on the duration of the displacement, we reveal that considering one factor at a time is not sufficient for describing the displacement duration because the factors interact. Our conclusions are based on statistical designs of experiments and response surface models. As expected, the models suggest that the flow rate, the spacer dimensions, and the concentrate viscosity affect the displacement duration. Moreover, the models reveal interactions of the mentioned factors. Surprisingly, neither the temperature nor the viscosity of the displacing fluid turned out to have a significant effect on the displacement duration. Finally, the Reynolds number was not identified as a predictor for the displacement duration. The practical relevance of the study is that displacement durations become more predictable allowing improved process control and thus an option for the reduction of product losses during displacement processes.

中文翻译：

---

从螺旋缠绕膜组件中置换蛋白质浓缩物的工艺条件的重要性

摘要 用于生产食品浓缩物的螺旋缠绕膜组件设备的每次启动或关闭都会伴随置换过程。在此过程中，由于被置换液稀释，通常会丢弃大量产品。为了通过改进过程控制来减少产品损失，需要了解关键工艺条件与非牛顿产品从螺旋缠绕膜组件中置换的持续时间之间的关系。虽然之前的研究评估了单个因素对位移持续时间的影响，但我们发现一次考虑一个因素不足以描述位移持续时间，因为这些因素相互作用。我们的结论基于实验的统计设计和响应面模型。正如预期的那样，模型表明流速、隔板尺寸和浓缩液粘度会影响置换持续时间。此外，模型揭示了上述因素的相互作用。令人惊讶的是，驱替流体的温度和粘度都对驱替持续时间没有显着影响。最后，雷诺数未被确定为位移持续时间的预测因子。该研究的实际意义在于置换持续时间变得更加可预测，从而允许改进过程控制，从而减少置换过程中的产品损失。结果表明，驱替液的温度和粘度均未对驱替持续时间产生显着影响。最后，雷诺数未被确定为位移持续时间的预测因子。该研究的实际意义在于置换持续时间变得更加可预测，从而允许改进过程控制，从而减少置换过程中的产品损失。结果表明，驱替液的温度和粘度均未对驱替持续时间产生显着影响。最后，雷诺数未被确定为位移持续时间的预测因子。该研究的实际意义在于置换持续时间变得更加可预测，从而允许改进过程控制，从而减少置换过程中的产品损失。