为得到可靠的小麦粉中面筋含量定量分析模型,基于光谱预处理及模拟退火算法(simulated annealing algorithm,SAA)对近红外光谱(near infrared spectroscopy,NIR)进行优化处理。偏最小二乘(partial least squares,PLS)回归用于建立预测模...为得到可靠的小麦粉中面筋含量定量分析模型,基于光谱预处理及模拟退火算法(simulated annealing algorithm,SAA)对近红外光谱(near infrared spectroscopy,NIR)进行优化处理。偏最小二乘(partial least squares,PLS)回归用于建立预测模型,以决定系数R2、校正均方根误差(root mean square error of calibration,RMSEC)、预测均方根误差(root mean square error of prediction,RMSEP)为指标,对比在不同光谱预处理条件下建立的回归模型与光谱预处理结合模拟退火算法优化处理条件下的回归模型。结果表明光谱预处理结合SAA-PLS模型能够有效提高模型的稳定性和预测能力,将R2从0.763 7提高到0.949 1、RMSEC从1.371 2降低到0.589 8、RMSEP从1.450 2降低到0.534 1。结果说明,光谱预处理结合模拟退火算法对光谱进行优化处理是可行的,模型预测能力和稳定性均优于未处理模型和仅进行光谱预处理的模型。展开更多
On the basis of computational of transonic viscid flow, the simulated annealing algorithm, used in statistical mechanics to study solid cooling process, is adaptedfor soving multi-objective cascades design problem. Th...On the basis of computational of transonic viscid flow, the simulated annealing algorithm, used in statistical mechanics to study solid cooling process, is adaptedfor soving multi-objective cascades design problem. The simulated annealing algorithmdraws an analogy between the energy minimization in physical system and the objectivefunction in the real design problem, To model the multi-objective functions, a minimumdeviation method is used. In this paper, the loss and work are considered as the objectivefunctions. Simulated results indicate that this algorithm can be effectively applied to theimprovement of the design for transonic turbine cascades.展开更多
文摘为得到可靠的小麦粉中面筋含量定量分析模型,基于光谱预处理及模拟退火算法(simulated annealing algorithm,SAA)对近红外光谱(near infrared spectroscopy,NIR)进行优化处理。偏最小二乘(partial least squares,PLS)回归用于建立预测模型,以决定系数R2、校正均方根误差(root mean square error of calibration,RMSEC)、预测均方根误差(root mean square error of prediction,RMSEP)为指标,对比在不同光谱预处理条件下建立的回归模型与光谱预处理结合模拟退火算法优化处理条件下的回归模型。结果表明光谱预处理结合SAA-PLS模型能够有效提高模型的稳定性和预测能力,将R2从0.763 7提高到0.949 1、RMSEC从1.371 2降低到0.589 8、RMSEP从1.450 2降低到0.534 1。结果说明,光谱预处理结合模拟退火算法对光谱进行优化处理是可行的,模型预测能力和稳定性均优于未处理模型和仅进行光谱预处理的模型。
文摘On the basis of computational of transonic viscid flow, the simulated annealing algorithm, used in statistical mechanics to study solid cooling process, is adaptedfor soving multi-objective cascades design problem. The simulated annealing algorithmdraws an analogy between the energy minimization in physical system and the objectivefunction in the real design problem, To model the multi-objective functions, a minimumdeviation method is used. In this paper, the loss and work are considered as the objectivefunctions. Simulated results indicate that this algorithm can be effectively applied to theimprovement of the design for transonic turbine cascades.
