Objective: To observe the applied effect of Artificial Intelligence Patient-Controlled Analgesia (Ai-PCA) in postoperative pain management of patients undergoing lower limb surgery. Method: Eighty patients undergoing lower limb fracture internal fixation surgery were randomly assigned to two groups: Ai-PCA group (Group A) and traditional patient-controlled analgesia group (Group P), with 40 patients in each group. The Group A used an intelligent pain management mode and the Group P used a traditional pain management mode. The static and dynamic NRS scores of the patients were recorded at 4, 8, 12, 24, and 48 hours after surgery as well as the Pittsburgh sleep quality index (PSQI) scores 1 day before surgery, 1 day and 2 day after surgery. The total number of button pressings, effective button pressings, and the total amount of sufentanil used in the 48 hours postoperatively were also recorded. The occurrence of adverse reactions and patient satisfaction were also noted. Results: Compared with the Group P, the static and dynamic NRS scores of the Group A were significantly lower (P0.05). The PSQI scores of the Group A were significantly lower than those of the Group P at 1 and 2 days after surgery (P0.05), and there was no statistical difference in the total number of button pressings, effective button pressings, and the total amount of sufentanil used in the 48 hours postoperatively between the two groups (P0.05). The incidence of dizziness, nausea, and vomiting in both groups had no statistically significant difference (P0.05), and the patient satisfaction of the Group A was significantly higher than that of the Group P (P0.05). Conclusion: Ai-PCA can improve the effect of postoperative pain management, improve sleep quality, and increase patient satisfaction in patients with lower limb fractures.