Abstract: In recent years, global warming has led to the frequent occurrence of extremely high-temperature events, which not only significantly increases the power demand but also drastically reduces the power generation capacity of distribution grids with a high proportion of renewable energy sources, which puts great pressure on power supply preservation. Based on this, this paper proposes a dynamic reconfiguration strategy of the distribution network that considers demand response to address the problem of reliable power consumption by users under extremely high-temperature scenarios. Firstly, the PV and wind turbine outputs and loads under extremely high-temperature scenarios are modeled, and their uncertainties are simulated by using scenario sets and scenario reduction methods. At the same time, the uncertainty model of the declared amount is constructed by considering the psychological factors of the users, and the proportional demand response incentive mechanism is proposed to address the problems of low motivation of the users to respond to the demand and large deviation of the actual response amount from the amount declared in the previous day, and the satisfaction model is established to measure the satisfaction level of the users to the demand response. Next, load types are divided, and based on the reliability needs of different types of users, the impact of outage loss is analysed, and a node outage loss coefficient matrix is established to measure the priority of node power supply. Then, a dynamic reconfiguration model of the distribution network for extreme high temperature scenarios taking into account demand response is established with the objective of minimizing the cost of lost load, demand response subsidy cost and switch action cost. Finally, the effectiveness of the proposed method is verified on a 148-node system, and the results show that the proposed model and method are conducive to guaranteeing reliable power supply to customers under extremely high-temperature scenarios.