Method for nondestructively detecting viscosity of nitro-cotton
A non-destructive testing and nitrocellulose technology, which is applied in the direction of measuring devices, flow characteristics, instruments, etc., can solve the problems of long time-consuming viscosity of nitrocellulose, inability to adapt to industrial field real-time detection, low efficiency, etc., and achieve fast measurement speed and low measurement cost , the effect of simple operation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0046] see figure 1 , a method for non-destructive testing of nitrocellulose viscosity, comprising the following steps:
[0047] a. Select the nitrocellulose sample set and collect the sample spectrum. The specific steps are as follows:
[0048] a1. According to the range of viscosity values, select representative nitrocellulose samples with viscosity to form sample sets A, B, and C corresponding to the measurement types, where the measurement types are method A, method B, and method C; A Set includes 270 samples whose viscosity is measured by method A, and the viscosity value of the sample is distributed in a gradient; set B includes 50 samples whose viscosity is measured by method B, and the viscosity value of the sample is distributed in a gradient; set C includes 60 samples passed by C Method to measure the viscosity of the sample, and the viscosity value of the sample is a gradient distribution; the viscosity range of nitrocellulose is between 0.6~2000s, and the viscosit...
Embodiment 2
[0069] A method for non-destructive detection of nitrocellulose viscosity is different from the method for detecting the viscosity of nitrocellulose described in Example 1:
[0070] In step a, optimize the spectral library formed by all sample spectra: match the spectra in the spectral library, and calculate the Mahalanobis distance between the spectra; select similar spectra with good matching according to the Mahalanobis distance between the spectra, Keep one of the similar spectra and delete the rest to update the spectral library;
[0071] Preprocess the spectra in A, B, C and the updated spectral library.
[0072] The advantage of optimizing the spectral library is that redundant information can be removed and representative spectra can be selected, so that when an unknown sample spectrum is input, a matching spectrum can be quickly found.
Embodiment 3
[0074] A method for non-destructive detection of nitrocellulose viscosity is different from the method for detecting the viscosity of nitrocellulose described in Example 2:
[0075] In step a, the spectra eliminated during the spectral library optimization are simultaneously eliminated in the corresponding A spectrum set or B spectrum set or C spectrum set, and the A spectrum set and / or B spectrum set and / or C spectrum set are updated to select the spectrum with Representative spectra to save model building time;
[0076] In step b, corresponding prediction models A model, B model and C model are established according to the updated and preprocessed A spectrum set, B spectrum set and C spectrum set.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 