Due to resource constraints in Wireless Sensor Networks (WSNs), this paper contributes a distributed clustering algorithm suitable for a large scale Voronoi cell-based WSNs with sensors randomly deployed according to homogenous spatial Poisson process and each sensor becomes a cluster head (CH) with a possibility p while non-CH sensors join the cluster of the closest CH to form a Voronoi tessellation. We explore a new sensor node deployment and generate stochastic geometry for the proposed algorithm being capable of showing how the critical parameters give significant influences on minimizing energy cost. Without loss of generality, the highly creditable simulation results prove that the proposed algorithm outperformance the Max-Min D-Cluster algorithm in terms of energy efficiency under certain network specifications. Moreover, scalability and robustness of the algorithm are also verified over extensive experiments.