To overcome the premature problem of the traditional flower pollination algorithm

a novel niche flower pollination algorithm is proposed by combining the niche strategy with flower pollination algorithm. It is designed for the parameter inversion of the space fractional order diffusion equation

so as to provide some theoretical basis for the pollutants source identification and air pollution prevention. Firstly

twenty multimodal functions were selected to verify the performance of the flower pollination algorithm and its improved algorithm. Then

we carried out direct simulation with implicit finite difference scheme. Based on the forward simulation results

the flower pollination algorithm and the improved algorithm were applied to invert the source term and the diffusion coefficients of the space fractional differential equation. The sensitivity analysis of the proposed algorithms regarding initial interval

perspectives of population and transition probability has also been completed. Furthermore

the anti-noise properties of the proposed algorithms were discussed. The numerical results demonstrate that the improved flower pollination algorithm has achieved a higher precision and accuracy.