Restoration of phase aberrations is crucial for addressing atmospheric turbulence in light propagation.Traditional restoration algorithms based on Zernike polynomials(ZPs)often encounter challenges related to high com...Restoration of phase aberrations is crucial for addressing atmospheric turbulence in light propagation.Traditional restoration algorithms based on Zernike polynomials(ZPs)often encounter challenges related to high computational complexity and insufficient capture of high-frequency phase aberration components,so we proposed a Principal-Component-Analysis-based method for representing phase aberrations.This paper discusses the factors influencing the accuracy of restoration,mainly including the sample space size and the sampling interval of D/r_(0),on the basis of characterizing phase aberrations by Principal Components(PCs).The experimental results show that a larger D/r_(0)sampling interval can ensure the generalization ability and robustness of the principal components in the case of a limited amount of original data,which can help to achieve high-precision deployment of the model in practical applications quickly.In the environment with relatively strong turbulence in the test set of D/r_(0)=24,the use of 34 terms of PCs can improve the corrected Strehl ratio(SR)from 0.007 to 0.1585,while the Strehl ratio of the light spot after restoration using 34 terms of ZPs is only 0.0215,demonstrating almost no correction effect.The results indicate that PCs can serve as a better alternative in representing and restoring the characteristics of atmospheric turbulence induced phase aberrations.These findings pave the way to use PCs of phase aberrations with fewer terms than traditional ZPs to achieve data dimensionality reduction,and offer a reference to accelerate and stabilize the model and deep learning based adaptive optics correction.展开更多
保护测量回路是电力系统继电保护的基石,其误差评估对电网安稳运维举足轻重。针对保护测量回路静态隐藏误差可能诱发保护误动/拒动的风险且难以在线监测问题,提出了一种基于递推主元分析和改进灰狼算法优化极限学习机(recursive princip...保护测量回路是电力系统继电保护的基石,其误差评估对电网安稳运维举足轻重。针对保护测量回路静态隐藏误差可能诱发保护误动/拒动的风险且难以在线监测问题,提出了一种基于递推主元分析和改进灰狼算法优化极限学习机(recursive principal component analysis and extreme learning machine optimized by grey wolf optimization,RPCA-GELM)数据驱动的保护测量回路误差评估方法。首先基于电力系统正常运行下历史数据与实时数据,应用RPCA技术在线更新主元特征模型以缩短评估时间,进一步引入4种统计算法生成4类误差监测特征量,构建误差综合评判方法进行特征优选,提升误差评估准确率。然后针对模型评估精度取决于关键参数C、σ,引入国际无限折叠混沌映射策略对灰狼算法进行优化,以提升参数寻优精度和收敛速度,在此基础上结合ELM算法提出了基于GELM的保护测量回路误差评估方法。最后通过多组对比实验验证了所提方法能实现模型性能优化,且相对其他方法有效提升了保护测量回路误差评估准确率与精度。展开更多
A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the mai...A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the main influence on day-ahead price, avoiding the strong correlation between the input factors that might influence electricity price, such as the load of the forecasting hour, other history loads and prices, weather and temperature; then GRNN was employed to forecast electricity price according to the main information extracted by PCA. To prove the efficiency of the combined model, a case from PJM (Pennsylvania-New Jersey-Maryland) day-ahead electricity market was evaluated. Compared to back-propagation (BP) neural network and standard GRNN, the combined method reduces the mean absolute percentage error about 3%.展开更多
文摘Restoration of phase aberrations is crucial for addressing atmospheric turbulence in light propagation.Traditional restoration algorithms based on Zernike polynomials(ZPs)often encounter challenges related to high computational complexity and insufficient capture of high-frequency phase aberration components,so we proposed a Principal-Component-Analysis-based method for representing phase aberrations.This paper discusses the factors influencing the accuracy of restoration,mainly including the sample space size and the sampling interval of D/r_(0),on the basis of characterizing phase aberrations by Principal Components(PCs).The experimental results show that a larger D/r_(0)sampling interval can ensure the generalization ability and robustness of the principal components in the case of a limited amount of original data,which can help to achieve high-precision deployment of the model in practical applications quickly.In the environment with relatively strong turbulence in the test set of D/r_(0)=24,the use of 34 terms of PCs can improve the corrected Strehl ratio(SR)from 0.007 to 0.1585,while the Strehl ratio of the light spot after restoration using 34 terms of ZPs is only 0.0215,demonstrating almost no correction effect.The results indicate that PCs can serve as a better alternative in representing and restoring the characteristics of atmospheric turbulence induced phase aberrations.These findings pave the way to use PCs of phase aberrations with fewer terms than traditional ZPs to achieve data dimensionality reduction,and offer a reference to accelerate and stabilize the model and deep learning based adaptive optics correction.
文摘保护测量回路是电力系统继电保护的基石,其误差评估对电网安稳运维举足轻重。针对保护测量回路静态隐藏误差可能诱发保护误动/拒动的风险且难以在线监测问题,提出了一种基于递推主元分析和改进灰狼算法优化极限学习机(recursive principal component analysis and extreme learning machine optimized by grey wolf optimization,RPCA-GELM)数据驱动的保护测量回路误差评估方法。首先基于电力系统正常运行下历史数据与实时数据,应用RPCA技术在线更新主元特征模型以缩短评估时间,进一步引入4种统计算法生成4类误差监测特征量,构建误差综合评判方法进行特征优选,提升误差评估准确率。然后针对模型评估精度取决于关键参数C、σ,引入国际无限折叠混沌映射策略对灰狼算法进行优化,以提升参数寻优精度和收敛速度,在此基础上结合ELM算法提出了基于GELM的保护测量回路误差评估方法。最后通过多组对比实验验证了所提方法能实现模型性能优化,且相对其他方法有效提升了保护测量回路误差评估准确率与精度。
基金Project(70671039) supported by the National Natural Science Foundation of China
文摘A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the main influence on day-ahead price, avoiding the strong correlation between the input factors that might influence electricity price, such as the load of the forecasting hour, other history loads and prices, weather and temperature; then GRNN was employed to forecast electricity price according to the main information extracted by PCA. To prove the efficiency of the combined model, a case from PJM (Pennsylvania-New Jersey-Maryland) day-ahead electricity market was evaluated. Compared to back-propagation (BP) neural network and standard GRNN, the combined method reduces the mean absolute percentage error about 3%.
