State estimation is a statistical method used to obtain accurate estimates of system operating states. It has been used to provide essential operating data for analysis, control and optimization in various energy supply systems. In a combined heat and power system, it is impractical to perform a state estimation procedure in a pool due to the information gap between the different energy sectors. To handle this problem, a decentralized state estimation algorithm based on the asynchronous alternating direction method of multipliers for a whole combined heat and power system is proposed. A dynamic state estimation model for a combined heat and power system is formulated considering the spatiotemporal quasidynamics of the network of heating pipelines. A framework based on the alternating direction method of multipliers is then proposed to solve the state estimation model in a decentralized way. First, an alternating estimation strategy is employed to solve district heating network subproblems to address the nondifferentiable and nonlinear constraints involved. Second, an asynchronous framework is employed to implement the alternating direction method of multipliers to reduce the idle time in the calculation of the subproblems and improve the overall computational efficiency. Case studies show that the proposed algorithm can still provide accurate estimates of the states of the integrated system even though it is decentralized. Compared to the standard implementation of the alternating direction method of multipliers, the proposed framework offers improved computational efficiency.