Physical-based electrochemical models have shown great potential in battery designed optimization and applications. However, the enormous parameter number and complex identification make it difficult to achieve the expected accuracy. Thus, it is essential to categorize and simplify the parameter identification process. In this paper, we classified electrochemical parameters into three groups based on their physical mechanism in the manufacturing process: geometry parameters, electrode properties parameters, and electrochemical kinetic parameters. The effect of eighteen factors on the electrochemical model performance is first chosen to study the terminal voltage fluctuation at various temperatures and charging rates based on the Doyle-Fuller-Newman model of a commercial LiNixCoyMn1-x-yO2 battery. The terminal voltage of complete process under various operating circumstance was selected as the comparative benchmark. By the Sobol global sensitivity analysis method, the influence degree was analysed for the parameter changes in the referenceable range. Not only the parameters with sensitive terminal voltage were analysed, but also the trend of the parameter influence changes under different conditions were analysed and eight parameters with high electrochemical sensitivity were found. Finally, by omitting the examination of the curve bundle and relying solely on the measurable terminal voltage data to accomplish parameter identifiability, this classification approach can streamline the process of identifying electrochemical parameters. As a result, this work provides through guidelines for battery design and simulation that are based on experimental data.