目的:探讨基于改进神经网络PID算法的医用电子直线加速器温控系统的实现。方法:针对医用电子直线加速器温度控制难题,提出了基于共轭梯度算法改进的BP神经网路PID控制算法,并在温度控制系统中进行了仿真。结果:该控制器能够实现较好控制效果。结论:利用共轭梯度算法改进的BP神经网路PID控制算法对医用电子直线加速器温度控制系统进行控制方法切实有效,简单可行。 Objective: To discuss the realization of temperature control system of medical linear accelerator based on improved neural network PID algorithm. Methods: Aiming at the problem of temperature control in medical linear accelerator, a BP neural network PID control algorithm based on conjugate gradient algorithm was proposed and simulated in the temperature control system. Result: This controller achieves better control. Conclusion: The PID neural network PID control algorithm improved by the conjugate gradient algorithm is effective, simple and feasible to control the temperature of medical linear accelerator.