上海佳實dh-9900蛋白儀的近紅外分析算法解析
上海佳實電子科技有限公司研制的dh-9900蛋白檢測系統(tǒng)的軟件分析系統(tǒng)具有自學習的能力。
近紅外（nir）檢測的優(yōu)點：樣品無需預處理，近紅外區(qū)內光穿透深度大，使得近紅外光譜技術可以用漫反射技術對樣品直接測定，同時還具有分析具有非破壞性、分析速度快、遠距離測定和實時分析、測定重現(xiàn)性好、適用的樣品范圍廣、分析成本較低、對操作人員的要求較低。簡而言之，使用方便，無需培訓。
nir用于物質成份的定量測定時，譜帶較寬并且容易重疊，需要有化學計量學方法進行分析（主成分分析pca、主成分回歸pcr、多元線性回歸、偏小二乘法pls、ann），應用多的為pca、pls。
dh-9900蛋白測量儀采用的算法是pca和pls，目前，上海佳實研發(fā)部的軟件工程師正在設計多的測量算法如pcr, ann等以升儀器，提。
測定固體樣品的nir光譜時，一般要測定樣品不同面的光譜以減少測定誤差并獲得可靠的信息，有時可以進行光譜的重復測定以提光譜圖的信噪比。在得到nir光譜圖后，一般還需要對光譜進行預處理（基線校正、一階導數(shù)、二階導數(shù)和正交信號校正）。
感謝廣大客戶的信賴和支持！
analysis of near infrared analysis algorithm of shanghai jiashi dh-9900 protein meter
the software analysis system of dh-9900 protein detection system developed by shanghai jiashi electronic technology co., ltd. has self-learning ability.
advantages of near infrared (nir) detection: the sample does not require pretreatment, and the depth of light penetration in the near-infrared region is large, so that near-infrared spectroscopy can directly measure the sample by diffuse reflection technology, and also has non-destructive analysis and analysis. fast, long-distance measurement and real-time analysis, good reproducibility, wide range of applicable samples, low cost of analysis, and low operator requirements.
when nir is used for the quantitative determination of material components, the bands are wide and easy to overlap, and chemometric methods are needed for analysis (principal component analysis pca, principal component regression pcr, multiple linear regression, partial least squares pls, ann). the most widely used are pca and pls.
the algorithms used in the dh-9900 protein meter are pca and pls. currently, software engineers in the r&d department are developing more measurement algorithms such as pcr, ann, etc. to upgrade the instrument and improve accuracy.
when measuring the nir spectrum of a solid sample, it is common to measure the spectrum of different faces of the sample to reduce the measurement error and obtain reliable information. sometimes, the repeated measurement of the spectrum can be performed to improve the signal-to-noise ratio of the spectrum. after obtaining the nir spectrum, it is generally necessary to preprocess the spectrum (baseline correction, first derivative, second derivative and quadrature signal correction).
thank you for your trust and support!