论文部分内容阅读
极限学习机(Extreme Learning Machine,ELM)具有学习速度快、算法简单易实现等优点,但是其泛化性却相对较差。针对这些缺点,提出了一种新的优化算法:泛优化算法(Wide Optimization Algorithm,WOA),将其应用于极限学习机输入权重和阈值的优化,利用泛优化算法的全局寻优能力,寻找训练误差较小时极限学习机的输入权重和阈值,从而提高极限学习机的泛化性,使其可以用较少的隐含层神经元获得较高的精度。将优化后的极限学习机应用于球磨机料位测量。实验结果表明优化后的极限学习机与传统极限学习机相比具有更高的测量精度和泛化性。
Extreme Learning Machine (ELM) has the advantages of fast learning speed and simple algorithm, but its generalization is relatively poor. Aiming at these shortcomings, a new optimization algorithm is proposed: Wide Optimization Algorithm (WOA), which is applied to the optimization of input weights and thresholds of extreme learning machines. By using the global optimization ability of pan optimization algorithm, When the error is small, the input weights and thresholds of the limit learning machine increase the generalization of the limit learning machine so that it can obtain higher precision with fewer hidden layer neurons. The optimized limit learning machine applied to ball mill level measurement. The experimental results show that the optimized extreme learning machine has higher accuracy and generalization compared with the traditional extreme learning machine.