Electrochemical modeling is of great significance for estimating the operational state, performing full-life-cycle fault diagnosis, and enabling multi-condition safety management of lithium-ion power batteries. However, early proposed pseudo-two-dimensional electrochemical models are difficult to apply in engineering practice due to issues such as numerous parameters and challenging identification. To address the problem that current liquid simplified pseudo-two-dimensions (LSP2D) models based on approximate reconstruction of solid-liquid phase diffusion processes struggle to meet high-rate conditions, this paper proposes a reduced-order reconstruction method for lithium-ion battery electrochemical models based on an improved LSP2D model. By refining the third-order parabolic approximation of the liquid-phase lithium-ion concentration and innovatively introducing a liquid-phase diffusion characterization term during the approximation process, the method achieves accurate characterization of the lithium-ion diffusion process dominated by concentration gradients under high-rate conditions. Simulation results show that while computational efficiency remains comparable, the improved LSP2D model outperforms the traditional LSP2D model in prediction accuracy across various rates. For liquid-phase lithium-ion concentration, the prediction accuracy of the improved LSP2D model is increased by over 86.96% compared to the traditional model at discharge rates of 1C–7C. For terminal voltage, the improvement exceeds 97.12% at 1C–3C discharge rates, and remains above 29.56% at 4C–7C discharge rates. Furthermore, the theoretical reasons for the significantly lower accuracy improvement at 4C–7C compared to 1C–3C are discussed, providing direction for subsequent research on reduced-order reconstruction of electrochemical models under high-rate conditions. The proposed method offers a new approach for high-accuracy reduced-order reconstruction of lithium-ion battery electrochemical models and contributes to enhancing the engineering practicality of electrochemical models for lithium-ion power batteries.