Calculus can precisely describe the continuous changes in vehicle energy consumption with operating conditions, and multi-objective optimization algorithms can strike a balance among conflicting objectives, such as range and performance. This article focuses on the energy consumption prediction and energy-saving control of electric vehicles,elaborates on their core demands, and proposes technical paths from three aspects: dynamic prediction modeling, energy-saving control strategies, and full-scenario implementation. It points out that future energy consumption management will evolve towards "predictive energy conservation" and "personalized optimization".