Wireless Sensor Network (WSN) based on IEEE 802.15.4 technology continues to gain considerable importance. The present work is based on this standard and a cluster tree network topology. This network is divided into many subgroups each of which is charac- terized by its special IEEE parameters (Beacon Order (BO) and Superframe Order (SO)). A mathematical model for computing energy consumption was proposed to enable the Personal Area Network (PAN) coordinator to detect the current energy level in the bat- tery. The proposed adaptive intervention is based on a comparison between the remaining energy and the diverse thresholds which lead to an auto-management of energy consump- tion. The proposed approach was evaluated using Inetmanet/Omnet++ simulator.