An available-flow neural network for solving the dynamic groundwater network maximum flow problem

The normal operation of infrastructure networks such as groundwater maintains people’s life and work. Therefore, it is great significance to estimate the residual flow when these are damaged evaluate their anti-risk ability. This paper abstracts problems damage-network time-varying maximum problem (DTMFP), where arc capacity in network set a function. Since water subject electric-driven periodic changes, damage defined by some arcs or nodes doesn’t work at all. Although existing algorithms Cai-Sha can solve problem, uncertainty topology that suffers from random makes difficult for problems. which suffer algorithms. key DTMFPs find network. In this paper, we propose an available neural (AFNN) algorithm solving DTMFPs. idea AFNN determine through back-information (BINN) first, then obtain single-path (SPNN) algorithm. BINN algorithm, departure node continuously activated given time period sends waves along destination node. wave contains be activated. Utah channel impact response wireless sensor network, American north marin district New York road used confirm effectiveness proposed