Abstract: Due to the initial difference in manufacturing process and the dynamic difference in application of a battery pack, there exists battery inconsistency of the battery pack, which has a negative impact on the overall performance of the battery pack and cause safety risks of the electric vehicles. This paper takes the battery pack of an electric vehicle as the object to study the correlation of battery pack temperature consistency and voltage consistency. Firstly, the raw data is pre-processed to solve the problems of inconsistent raw data field format, data missing, and bad points. Moreover, according to the discharge and charge state of the battery pack, effective charge and discharge segments has been divided. All the raw data are divided into 1 157 discharge and charge segments. Based on these effective segments, the characteristics of the total voltage and current under typical discharge conditions and the single battery cell voltage and temperature at different positions under the same discharge segment are then analyzed. It is found that multiple battery cells are detected with the same voltage at the same time, making it difficult to determine the single battery cell corresponding to the highest or lowest voltage. In addition, there are significant differences in the temperature at the different positions of the battery pack, and there is a certain correlation between temperature difference and discharge time. Thirdly, a method based on hierarchical clustering to analysis battery pack temperature inconsistency and voltage inconsistency under discharge condition is proposed. In the method, Euclidean distance of different temperature and voltage sampling points is used for the class division, and the average distance is used to calculate the center distance of each clustering. Fourthly, in order to quantitatively characterize the temperature consistency and voltage consistency of the battery pack, the dispersion of the whole battery pack is measured by the clustering center distance. The maximum and minimum clustering center distance are calculated, and the range of clustering distance is used as the inconsistency evaluation index to analyze the trend the temperature consistency and voltage consistency of the battery pack. It is found that seasonal changes in environmental temperature can affect the temperature distribution within the battery pack, and temperature inconsistency is inversely proportional to the average environmental temperature. When the average environmental temperature is high, the temperature consistency is better; When the environmental temperature decreases, the temperature inconsistency increases. Finally, the correlation analysis of battery temperature consistency and voltage consistency is carried out. It is found that the temperature inconsistency of No.1 module is the largest, and the temperature consistency of No.2 temperature sensor in No.1 module is the worst. The voltage consistency of the battery cell corresponding to the position with poor temperature consistency is also poor, that is, No.4 and No.69 battery cells. The voltage inconsistency is greatly affected by temperature inconsistency, and the voltage inconsistency rises in steps, which is unrecoverable.