Abstract: To overcome the problems of uneven energy consumption and premature death caused by the restrictions of sensor node position and energy in the wireless temperature measuring system of solid state fermentation, a novel optimization algorithm, named as new uneven clustering algorithm, which is based on differential evolution method is presented in this paper. In the new algorithm, considering the system stability, a fixed number of cluster heads are selected at once and the node coverage rate is optimized by the differential evolution algorithm instead of the mechanism of frequent selected cluster each round in the LEACH protocol. Also, to extend the network’s life and distribute the initial energy of cluster head nodes and ordinary temperature nodes reasonably, the energy matching differentiation strategy is introduced. The simulation results show that this algorithm can effectively improve the problem of “hot spot” and balance the energy consumption, and the algorithm can be able to prolong the network lifetime in a certain extent.

Key words: wireless temperature measuring system of solid state fermentation, Differential Evolution（DE） algorithm, matching strategy of energy difference, uneven clustering, network lifetime

摘要： 针对固态发酵无线测温系统中传感器节点位置、能量等受限因素导致能量消耗不均而过早死亡的问题，提出了一种基于差分进化改进的非均匀分簇算法。该算法省去了LEACH协议中每轮频繁选簇的机制，而是从系统的稳定性出发，一次性选择固定数目的簇首，采用差分进化算法优化节点覆盖率。同时，采用能量差异化匹配策略，合理地分配簇首节点与普通测温节点的初始能量，延长网络寿命。仿真结果表明，该算法有效改善了“热区”问题，均衡了能耗，在一定程度上延长了网络的生命周期。

关键词: 固态发酵无线测温系统, 差分进化算法, 能量差异化匹配, 非均匀分簇, 网络生存周期