A kind of crude drug powder online detection device and detection method

Abstract

The present invention provides an online detection device and detection method for crude drug powder. The detection device includes a mixer, a sampling pipeline, a sampling switch, a sampling platform, an optical fiber probe, a near-infrared spectrometer, and a control system. The online detection method includes the following steps: (a) The establishment of an online detection device; (b) Collection of crude drug powders of different production batches; (c) Determination of quality control indicators in crude drug powders; (d) Collection of near-infrared spectral data of crude drug powders; (e) Near-infrared Spectral modeling band selection and spectral preprocessing; (f) establishment and evaluation of near-infrared spectral models; (g) fast online detection with the built model; (h) judgment and control of detection results. The on-line detection device and detection method provided by the invention are beneficial to rapid and accurate on-line monitoring of the quality of crude drug powder.

Description

Technical field [0001] The present invention belongs to the field of near-infrared detection, and more particularly to an online detecting device and detection method of halody powder. Background technique [0002] In the medium-completed pharmaceutical preparation, the mixing of the biotechnical powder is a key step, the homogeneity and content reaches the target value is two important assessment indicators for the mixing of the biopsy powder. During the actual production process, sampling trouble, traditional detection and analysis methods, and seriously affect production efficiency. And in the pulverization process, the active ingredient of some of the pharmaceutical powder is susceptible to the temperature of the volatile or structural change. In the pulverization process, the temperature increases the temperature of the machine in the pulverization process so that its component loss has no target value, resulting in unqualified drug powder. The timely operation of material w...

AI Technical Summary

Problems solved by technology

However, most of these literatures are aimed at a single or a few index components in the finished products of Liuwei Dihuang Pills Concentrated Pills, and most of them use the partial least squares regression algorithm as the modeling method of the quantitative model, and use the reverse artificial neural network based on principal component analysis for Liuwei There are no relevant reports on the online detection of multiple quality control indicators in the crude drug powder of Dihuang Pill Xiaomi Pill

Images