基于改进粒子群优化算法nMOSFET寿命预测

对于电子器件寿命预测问题，文章提出了基于改进粒子群优化算法的BP神经网络电子器件寿命预测方法。首先对nMOSFET元件在不同应力条件下进行寿命试验，根据试验测试获得的寿命数据，得出对应的可靠性。文章通过结合改进粒子群优化算法和BP神经网络结合，建立电子器件寿命预测模型，应用该模型对相同应力条件的电子器件寿命进行预测，同时对应力加速条件下寿命的预测。通过试验证明，该算法具有更强的非线性拟合能力和更高的准确率。

Life Prediction of nMOSFET Based on MPSO BP neural network

For the prediction of life expectancy of electronic components, a life prediction method based on modified particle swarm optimization and BP neural network is proposed. Firstly the corresponding in various conditions of life data was obtained according to the life test of nMOSFET in a variety of stress conditions, which can draw the corresponding reliability of electronic components. By combining the modified particle swarm optimization algorithm and BP neural network combined forecasting model of electronic components life established, the electronic components life in same stress condition can be predicted. The experimental results show it has higher nonlinear fitting prediction accuracy.