欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
基于间隔策略的信息波长选择是近红外光谱分析中广泛应用的一种方法。针对传统算法忽略非线性因素的缺点，首次考虑将最小二乘支持向量机（1east—squaressupportvectormachine，LSSVM）方法应用于间隔选择策略，进而提出了一种新的波长选择方法iLSSVM（intervalLSSVM）。该算法克服了传统间隔选择算法依赖于线性模型的缺陷，对存在较强非线性的光谱数据能够准确地选择最优信息区间，极大地减少建模变量并显著改善模型预测精度。应用两组业界标准的光谱数据来验证该算法的性能，并和传统方法进行了比较。实验结果表明，在两组数据集上该算法取得的标准预测偏差（rootmeansquareerrorofprediction，RMSEP）分别比全谱PLS建模降低了20％和4％，比传统的间隔偏最小二乘算法（intervalpartialleast～squares，iPI．s）降低了28％和2％。
英文摘要：
Strategy based Inspired by interval partia method combining interva on interval selection is widely used in the near-infrared spectroscopy analysis. 1 least squares method （iPLS）, the present paper proposed a new wavelength 1 selection strategy with least-squares support vector machine （I.SSVM）. By overcoming the shortcomings of traditional interval selection methods whose predictive ability totally depend on the linear model, this new algorithm, named as iLSSVM （interval LSSVM）, can select the optimal informative interval more reasonably to significantly improve the model prediction accuracy with less modeling variables. Two real near-infrared datasets were applied to this new approach and the prediction performance was compared to the other interval selection methods. The experimental results demonstrated that the root mean square error of prediction （RMSEP） of this new method is 20% and 4% smaller than that of full-spectrum PLS modeling method respectively, and is 28% and 2% smaller than that of the traditional iPLS （interval partial least-squares） method respectively.
同期刊论文项目
同项目期刊论文
A hybrid dehumidifier model for real-time performance monitoring, control and optimization in liquid
An adaptive multiple-modality sensor network data compression algorithm based on lifting wavelet and
Classification of washing powder brands using near-infrared spectroscopy combined with chemometric c
期刊信息
位置：