As an important sensing unit, the evaluation of the health status of distribution terminal units is conducive to detecting potential security risks of the terminal, arranging a reasonable maintenance plan, and ensuring the safe operation of the power grid. On the basis of building a comprehensive and practical hierarchical evaluation index system, this paper introduces a cloud model integrating the randomness and fuzziness of objective things to solve the index membership degree, and improves the Interval Analytic Hierarchy Process (IAHP) to give consideration to the subjectivity and objectivity of the index weight. Then, based on D-S evidence theory, the membership degree of the basic layer index is transformed into the trust degree of the target layer, and the terminal health status is determined by the maximum trust degree, and the distribution of the conflict coefficient is further analyzed to give early warning of the indicators of the bad state of the basic layer. Finally, taking the feeder terminal of a 20 kV line as an example, the feasibility of the qualitative index membership degree solution, combination weight acquisition and evaluation process is verified. This work is supported by the National Key Research and Development Program of China (No. 2019YFB1505404).