近年来，神经网络在从金融到医学的许多领域得到了越来越多的应用。饮用水行业也见证了其在这一领域的发展份额，为水管理过程的各个方面创建了计算模型。本文介绍了神经计算理论，以及在饮用水领域的潜在应用。讨论了反向传播神经网络的两个例子。第一个模拟每日氯化后所需的剂量；另一个预测是，在未来的一个或多个步骤中，氯剂量产生的残余消毒剂。研究结果表明，神经网络在水处理过程中的应用，以及在配水系统水质控制中的应用前景广阔。

Neural networks have been applied increasingly in recent years in a number of fields, from finance to medicine. The drinking water industry has also witnessed its share of developments in this area, with computing models being created for various aspects of the water management process. This paper presents an introduction to neural computing theory, as well as the potential for applications in the drinking water field. Two examples of back-propagation neural networks are discussed. The first simulates required daily post-chlorination dosage; the other forecasts, at one or more steps in the future, the residual disinfectant resulting from chlorine dosage. The results suggest promising possibilities for the application of neural networks in the water treatment process, as well as in the control of water quality in distribution systems.