A distributed congestion management scheme based on iterative distribution locational marginal price (iDLMP) to integrate distributed energy resources into the distribution network has been studied recently. However, this scheme has a requirement on the optimisation formulation for prosumers, i.e. the suboptimisation problem should be a convex one, which in certain cases might face challenges. A typical example is the non-linear constraints of the energy storage system (ESS), which ensures that charging and discharging of ESS are mutually exclusive. In this study, the authors consider the prosumer as a selforganising unit to integrate diverse resource flexibilities. Two roles exist in prosumer's energy operation: as a producer and as a consumer, which are also mutually exclusive for one prosumer at one time. When applying the iDLMP method to integrate prosumers energy operation, one meets a similar challenge as before. Thus, they make some reasonable assumptions about the transformer capacity, line capacity, and the price sensitivity parameters to convexify the optimisation problem. With these assumptions, a distributed convex optimisation framework is developed for the prosumers that can handle the non-convexity issue in the optimisation problem. The numerical results indicate that the proposed method can achieve local energy sharing in the distribution network, respecting the line congestion alleviation.