本研究的目的是确定是否有新的颗粒物检测方法 该系统具有检测膜系统内单纤维破裂和 评估是否可以确定致病性大小范围内是否存在颗粒。 此外，目标是使用至少两种方法来模拟劈开的 并比较使用这种新型传感器的每种方法的结果 技术 本研究中使用的两种测试包括切割实际的膜纤维和 新的光纤旁路方法。劈裂试验过去已经使用过，并且有很多优点 变量，例如破裂的位置、破裂的性质、颗粒物 过滤进水的特性、纤维的年龄等。第二次测试，纤维 旁路法被设计成切割纤维的替代方法。此过滤器旁路测试 包括在污水处理后立即将进水输送至污水流 膜盒。执行上述任一完整性损失 方法，在监测监测系统基线的有效变化时， 将指示微粒检测传感器是否能够检测到相应的 失败 测试是在严格的条件下进行的。海水的浑浊 进水量非常低，测量值在0.8到1.0 NTU之间。温度 水温也很低，只有4摄氏度。这两个因素会减缓水流通过 纤维断裂，使检测更加困难。即使在这些困难的条件下 传感器检测到纤维旁路和实际膜纤维断裂的变化 测验。当应用过滤器旁路测试时，0.5到3之间的影响被破坏 对纤维进行了模拟。微粒检测传感器可以轻松地确定每一个 变化。膜切割试验没有显示出相同的趋势响应，这表明切割纤维的位置、通过纤维的流速、 通过纤维的压降和水的粘度都会影响结果。 然而，最终，浊度传感器确实看到了纤维的变化。此外，还对粒径范围为 0.028至5 um，以评估完整性损失是否符合粒度要求。 该测试的初步结果表明，可以确定 完整性损失中的粒度范围。 测试还表明，当使用 灵敏度高的浊度计。当检测灵敏度为 为了避免污染环境，采取了越来越多的取样方法 系统 到目前为止，使用直接膜工艺取代传统过滤已经取得了很大进展 在几乎所有情况下，我们都非常成功地生产出高质量的饮用水。 结果表明，新的颗粒物监测系统 本文讨论的传感器对膜中的单纤维断裂非常敏感 同时为整个膜提供连续监测方法 系统此外，该系统还提供了一种直接、快速的方法来隔离病毒 膜盒存在完整性问题。 包括8个参考文献、表格、图表。

The objectives of this study were to determine if a new particulate detection system has the sensitivity to detect a single fiber breach within a membrane system and to assess whether the presence of particles in the pathogenic size range could be determined. In addition, the objectives were to use at least two methodologies for simulating a cleaved membrane fiber and compare the results of each method using this new sensor technology. Two tests used in this study include cleaving an actual membrane fiber or fibers and a new fiber bypass method. The cleaving test has been used in the past and has many variables such as the location of the break, the nature of the break, the particulate characteristics in the filter influent, the age of the fibers, etc. A second test, the fiber bypass method, was designed to be a substitute for cleaving a fiber. This filter bypass test involves transporting influent water to the effluent stream immediately after the membrane cartridge. The execution of either of these integrity loss methods, while monitoring the effective change in the baseline of the monitoring system, will indicate if the particulate detection sensor is capable of detecting the respective failure. The testing was conducted under stringent conditions. The turbidity of the influent was very low, with measured values between 0.8 and 1.0 NTU. The temperature of the water was also very low, at 4 C. These two factors slow the flow through a broken fiber and make the detection more difficult. Even in these difficult conditions, the sensors detected changes from the fiber bypass and the actual membrane fiber cleaving tests. When the filter bypass testing was applied, the effects of between 0.5 and 3 broken fibers were simulated. The particulate detection sensor easily determined each of these changes. The membrane cleaving tests did not show the same trending response, indicating that the location of the cleaved fibers, the flow rate through the fiber, the pressure drop through the fiber, and the viscosity of the water all affected the results. However, in the end, the turbidity sensor did see the changes to the fibers. Additional testing was also performed with particle surrogates, in the size range of 0.028 to 5 um, to assess if an integrity loss could be qualified in regard to particle size. Preliminary results from this testing indicate that it may be possible to determine a particle size range in an integrity loss. The testing also demonstrated that sampling methods were critical when using a turbidimeter with highly enhanced sensitivity. When the detection sensitivity was increased, sampling methods were carefully applied to avoid contaminating the system. To date, the use of direct membrane processes to replace traditional filtration has been very successful in producing high quality drinking water in nearly all occurrences. It was demonstrated that the new particulate monitoring system discussed in this paper has the sensitivity to see a single fiber break in a membrane cartridge while providing a continuous monitoring approach for an entire membrane system. In addition, this system provides a direct and rapid means of isolating a membrane cartridge with an integrity problem. Includes 8 references, tables, figures.