The current conventional distribution network alert state adjustable potential assessment method mainly obtains the assessment matrix by constructing the affiliation function to achieve the adjustable potential assessment, which leads to poor assessment accuracy due to the lack of analysis of line loss abnormal state. In this regard, a new distribution network alert state adjustable potential assessment method based on the improved isolated forest algorithm is proposed. Combined with the isolated forest algorithm, a random hyperplane cutting space is constructed to calculate the line loss anomaly fraction, and the value is used as one of the evaluation indexes to construct a potential evaluation index system, and the actual values of different indexes are calculated to realize the adjustable potential evaluation. In the experiments, the proposed method is verified for the evaluation accuracy. The experimental results show that the error value between the evaluation score and the actual result is small when the proposed method is used to evaluate the adjustable potential of the alert state of the distribution network, which has a more desirable evaluation accuracy.