Source Term Estimation with Unknown Number of Sources using Improved Cuckoo Search Algorithm

This paper is focused on the source term estimation (STE) with unknown number of sources. The dispersion model of multiple point sources is constructed firstly. Then the cost function is established by using the outputs of dispersion model and the fixed sensor network's measurements. To ensure finding the global minimum of the cost function, the Improved Cuckoo Search (ICS) algorithm is applied. Simulations with different sensor and noise modes are designed to evaluate the estimating performance. Finally, comparison of ICS, Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) proves the superiority of ICS in this problem.