To improve the accuracy and economy of fault diagnosis, the layout method of water pressure monitoring points is proposed. Firstly, a fault diagnosis model of water supply network is designed by using Modified Fruit Fly Optimization Algorithm (MFFOA) and Kernel Extreme Learning Machine (KELM). The experimental results show that the fault diagnosis model has the advantages of fast training speed and high accuracy. Then, based on the Fruit Fly Optimization Algorithm (FFOA) and the fault diagnosis model which is proposed, a water pressure monitoring point layout method is designed. In the method, firstly a set of water pressure detection points is formed by FFOA. Then, the diagnostic accuracy of each scheme is calculated by the fault diagnosis model and the accurate and economic scheme is chosen to be the candidate optimal program. After that, FFOA algorithm is used to iterate and find the optimal layout of water pressure monitoring solution which has the best accuracy and best economic performance. Finally, the proposed method which is realized is tested by the actual pipe network data. The results show that the method presented in this paper is an effective method to optimize the water pressure monitoring points. Comparing with other methods, the accuracy and economy of fault diagnosis are improved obviously.