The invention relates to a method for diagnosing the parameter inconsistency of power battery packs, and belongs to the technical field of battery management. The method comprises the following steps:S1: selecting power batteries having different initial performance and similar initial performance, forming two types of battery packs in a series parallel manner, and collecting technical parametersthereof; S2: simulating real vehicle working conditions under different roads, controlling the temperature of each monomer in the battery packs, performing a charge-discharge experiment on the powerbattery packs, collecting the voltage, current and temperature data of each battery cell, and establishing a real vehicle working condition test database; S3: performing data processing and feature extraction on time domain data of the collected voltage, current and temperature by using a feature extraction method; and S4, evaluating the consistency of the battery packs for the extracted featuresby using a weight method, and achieving parameter inconsistency diagnosis by combining a multi-scale entropy with an artificial neural network. By adoption of the method provided by the invention, faulty batteries can be diagnosed in real time, the diagnosis accuracy of the parameter inconsistency of the battery packs is improved, and timely maintenance is facilitated.