In view of the shortcomings of traditional water supply network fault monitoring system such as the number of false positives,poor anti-interference ability,low reliability and so on,the paper used multi sensors to monitor the pressure,current,flow rate and liquid level parameters in the process of water supply of,and used the multi sensor data fusion technology based on D-S evidence theory,and improved the D-S evidence theory by fusion method in time domain and spatial domain so as to achieve the early detection of water supply pipe network.The results show that the method can effectively avoid the errors of monitoring data,and solve the problem of evidence conflict caused by multi sensor robustness,and improve the reliability of fault alarm of water supply pipe network.