Component detection system for unmanned automatic production of extracting solution

Abstract

The invention discloses a component detection system for unmanned automatic production of an extracting solution, and relates to the technical field of automatic production and quality control. The system integrates a plurality of detection technologies, including infrared spectroscopic analysis, OD value detection, pH value monitoring and absorbance detection, so as to realize real-time monitoring and automatic production management of the components of the extracting solution. By establishing a standard product fingerprint spectrum database, the system can automatically compare real-time detection data with a standard spectrum, and immediately triggers an alarm or stops production once an abnormality is found, so that the stability of product quality and the high efficiency of a production process are ensured. The system not only improves the automation degree of a production line, but also predicts and optimizes bioactive components through a whole genome nutrition group analysis technology in combination with species genome, transcriptome data and metabolome information, realizes precise formula and efficient production, and provides powerful support for an unattended automatic production system.

Description

Technical Field

[0001] The present invention relates to a component detection system for unmanned automatic production of extracts, which is particularly suitable for detection and quality control of bioactive components based on genome and metabolome integrated analysis, and belongs to the field of biotechnology and automated production. Background Art

[0002] In whole-genome nutrient group studies, component analysis of extracts is crucial for understanding the metabolic pathways and biological activities of nutrients. Traditional methods for extract analysis usually rely on offline laboratory analysis, which is time-consuming and difficult to achieve real-time monitoring, resulting in challenges in quality control during production. Therefore, developing a method that can monitor extract composition in real time and automate production is of great significance for improving production efficiency and product quality.

[0003] Traditional quality inspection methods usually rely on manual operation, which is inefficient and prone to errors. In recent years, with the development of automation technology and detection technology, such as infrared detection, OD value detection, pH value detection and absorbance detection, it has become possible to realize the automated production and quality control of soybean extract. Summary of the invention

[0004] The purpose of the present invention is to provide a component detection system for unmanned automatic production of extracts, so as to solve the problem that traditional quality detection methods usually rely on manual operation, are inefficient and prone to errors.

[0005] Technical solution: A component detection system for unmanned automatic production of extracts, including: A transparent pipeline module is used to transmit the extraction liquid; AI-driven multi-spectral fusion detection module uses AI-driven detection algorithms to detect the spectral components of the extract; AI controls the fluid property sensor array analysis module, and uses AI-driven detection algorithms to detect and analyze the temperature, conductivity, turbidity, pH, and OD value of the extract; Data processing module, used to analyze the detection data in real time and compare it with the standard fingerprint database; Multi-level alarm module, used to trigger sound and light alarms according to the detection results; Industrial Internet of Things integration module, which integrates the system with the Industrial Internet of Things platform to achieve remote monitoring and automated operation based on the test results; Database and model building modules for constructing whole genome-based nutrient group analysis equipment.

[0006] Preferably, the pipeline module adopts a polytetrafluoroethylene transparent pipeline, and the specifications of the polytetrafluoroethylene transparent pipeline are: inner diameter 32mm, outer diameter 26mm, and the outer side of the pipeline is fixed and guided by an iron frame.

[0007] Preferably, the AI-driven multi-spectral fusion detection module includes a high-performance computing device, a spectral acquisition device, a data processing software, a spectral data block, and a cloud service platform, wherein the spectral acquisition device includes an online infrared detection unit, which is a built-in ATR-FTIR probe that performs component analysis on the extract to generate fingerprint spectrum data, with zinc selenide crystal as the sampling element, an infrared light incident angle of 45°, and a wave number range of 4000-600cm -1 The multi-wavelength absorbance detection component includes a three-channel LED light source with wavelengths of 385nm / 460nm / 550nm, a photodiode array detector with a response time of <10ms, and a flow-through cuvette with an optical path of 10mm and a pressure resistance of 0.5MPa. Combined with a spectrophotometer of a specific wavelength, it detects the absorbance of specific components in the extract.

[0008] Preferably, the AI-controlled fluid property sensor array analysis module includes a high-performance computing device, a fluid property sensor array, data processing software, a data management membrane block, and a cloud service platform, wherein the fluid property sensor array includes a temperature detection unit, a conductivity detection unit, a turbidity detection unit, a high-precision pH monitoring unit, and an OD value detection device, wherein the high-precision pH monitoring unit uses a pH sensor to monitor the pH value of the extract in real time, and uses a composite electrode with a measurement range of 0-14 and a resolution of 0.01pH, and an automatic temperature compensation module with an NTC 10K sensor as a temperature sensor, so as to effectively cope with the impact of temperature changes through real-time monitoring and adjustment; the OD value detection device monitors the OD value of the extract in real time through a spectrophotometer for evaluating the concentration of the extract, adopts a 90° scattered light measurement structure, and the scattered light adopts a double-beam compensation design, and effectively improves the measurement accuracy and stability through comparative measurement and real-time correction.

[0009] Preferably, the data processing module processes the spectral data and numerical data acquired by the detection module, including spectral denoising, baseline correction, characteristic peak extraction, etc., and judges in real time whether the components of the extract meet the preset standards by comparing with the standard fingerprint database. By collecting 200 sets of normal batch spectral data, the spectral data is subjected to SNV preprocessing and second-order derivative transformation processing, the interference factors in the spectrum are eliminated to establish a reference database, and the characteristic wavelengths with VIP values ​​> 1.5 are screened by developing a PLS-DA discriminant model.

[0010] Preferably, the multi-level alarm module adopts a sliding window DTW (dynamic time warping) algorithm and sets three levels of warning thresholds (yellow: 2σ, orange: 3σ, red: 5σ).

[0011] Preferably, the industrial Internet of Things integration module integrates the system with the industrial Internet of Things platform to realize remote monitoring and automatic operation according to the detection results. When an abnormal component is detected, the system can automatically issue an alarm or suspend production. The automatic control task is realized by combining the parameter adjustment module of the PID algorithm and the multi-parameter association model, wherein the key parameters of the PDI controller are Kp=0.8, Ki=0.2, Kd=0.05; the multi-parameter association model is realized by the linear model ΔOD 600 =0.15×ΔpH+0.32×ΔAbs 254 to set up.

[0012] Preferably, the database and model building module constructs a bioactive component map by combining species genome, transcriptome and metabolome data based on whole genome nutrient group analysis equipment.

[0013] The beneficial effects of the present invention are: 1. Realize real-time online detection of extract components to avoid errors caused by manual sampling.

[0014] 2. Comprehensively evaluate the quality of the extract through multi-dimensional testing (infrared spectrum, OD value, pH value, absorbance, turbidity, temperature, conductivity).

[0015] 3. Combine artificial intelligence and industrial Internet of Things integration modules to achieve unmanned production and quality monitoring.

[0016] 4. The bioactive component map constructed based on genomic and metabolomic data provides a scientific basis for the optimization of extract formula. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Infrared spectrum of soybean active ingredient extract generated by the instrument. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0020] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0021] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0022] An embodiment of the present invention provides a component detection system for unmanned automatic production of extracts, which is specifically applied as follows: The soybean extraction equipment is connected to the transparent pipeline, and the multi-spectral fusion detection module and the fluid property sensor array module are connected. The extract is monitored in real time through infrared detection equipment, OD value detector, pH value detector and absorbance detector, and the real-time collected data is transmitted to the data processing module; the data processing module analyzes the detection data and compares it with the standard fingerprint database to determine whether the ingredients meet the standards; the multi-level alarm module decides whether to alarm and the alarm level based on the analysis results; the automation control module adjusts the production process according to the analysis results to realize unmanned production.

[0023] Example: Take the soybean isoflavone extraction production line as an example: 1. Hardware configuration: High-performance computing equipment to process large amounts of spectral and monitoring data as well as complex AI algorithms; Spectral acquisition equipment, including spectrometers, spectral sensors, etc., used to obtain spectral information of substances; Fluid property sensing array equipment, including OD value detector, pH value detector, spectrophotometer, online turbidity analyzer, electronic thermometer, conductivity meter; Multi-level alarm module, used to trigger sound and light alarms according to the detection results; A detection unit (flow rate 1.5m³ / h, pressure 0.3MPa) is installed on the extraction tank outlet pipeline module and equipped with an industrial-grade PLC controller.

[0024] 2. Software configuration: The general AI development platform uses multimodal fusion algorithms to effectively combine data from different modalities such as images, text, and data to achieve online detection and decision analysis; Data processing software, which performs preprocessing on the original spectra, such as denoising and baseline correction; Machine learning framework for building and training AI models; Data analytics platform to provide visualization, model evaluation, and results analysis; Cloud service platform, which provides online analysis and AI model services, and enables remote access and sharing of data; Database and model building module for constructing equipment based on whole genome nutrient group analysis; Industrial Internet of Things platform, used to achieve remote monitoring and automated operations based on test results.

[0025] Parameter settings: Normal range: pH 5.2-6.8, OD 600 0.7-1.1, Abs 254 0.3-0.6, T20-30℃, A0-4, NTU 0-2, σ55-100; Control response: When pH>6.8 for 5 minutes, citric acid adjustment solution is automatically injected (flow rate 50mL / min).

[0026] 3. Workflow: a) Start the self-check procedure: perform NIST standard material calibration (SRM 2036); b) Real-time data collection stream: FTIR spectrum (4cm -1 resolution), three-wavelength absorbance, pH / OD value, temperature, absorbance, turbidity and conductivity; c) Data preprocessing: Savitzky-Golay smoothing (window 21 points, cubic polynomial); d) Feature extraction: using moving window PCA algorithm (window width 30s); e) Decision-making: If three consecutive sampling points exceed the 3σ range, the secondary alarm will be triggered and the backup circulation pipeline will be started.

[0027] Specifically, the embodiment of the present invention limits the types of soybean active ingredient extract data. In actual application, there are many factors that affect the extract. The embodiment of the present invention limits the soybean active ingredient extract data to achieve processing of limited data. It can be understood that the data that can be monitored and collected is not limited to the listed data, and other extract data can be added, which are not listed here one by one.

[0028] The above-mentioned embodiments only express the implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A component detection system for unmanned automatic production of extracts, characterized in that: include: A transparent pipeline module is used to transmit the extraction liquid; AI-driven multi-spectral fusion detection module uses AI-driven detection algorithms to detect the spectral components of the extract; AI controls the fluid property sensor array analysis module, and uses AI-driven detection algorithms to detect and analyze the temperature, conductivity, turbidity, pH, and OD value of the extract; Data processing module, used to analyze the detection data in real time and compare it with the standard fingerprint database; Multi-level alarm module, used to trigger sound and light alarms according to the detection results; Industrial Internet of Things integration module, which integrates the system with the Industrial Internet of Things platform to achieve remote monitoring and automated operation based on the test results; Database and model building modules for constructing whole genome-based nutrient group analysis equipment.

2. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The transparent pipe module adopts a polytetrafluoroethylene transparent pipe, and the specifications of the polytetrafluoroethylene transparent pipe are: inner diameter 32mm, outer diameter 26mm, and the outer side of the pipe is fixed and guided by an iron frame.

3. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The AI-driven multi-spectral fusion detection module includes a high-performance computing device, a spectral acquisition device, data processing software, a spectral data block, and a cloud service platform. The spectral acquisition device includes an online infrared detection unit, which is a built-in ATR-FTIR probe that analyzes the components of the extract and generates fingerprint spectrum data. The zinc selenide crystal is used as the sampling element, the infrared light incident angle is 45°, and the wave number range is 4000-600cm -1 The multi-wavelength absorbance detection component includes a three-channel LED light source with wavelengths of 385nm / 460nm / 550nm, a photodiode array detector with a response time of <10ms, and a flow-through cuvette with an optical path of 10mm and a pressure resistance of 0.5MPa. Combined with a spectrophotometer of a specific wavelength, it detects the absorbance of specific components in the extract.

4. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The AI-controlled fluid property sensor array analysis module includes a high-performance computing device, a fluid property sensor array, data processing software, a data management membrane block, and a cloud service platform, wherein the fluid property sensor array includes a temperature detection unit, a conductivity detection unit, a turbidity detection unit, a high-precision pH monitoring unit, and an OD value detection device, wherein the high-precision pH monitoring unit uses a pH sensor to monitor the pH value of the extract in real time, and uses a composite electrode with a measurement range of 0-14 and a resolution of 0.01pH, and an automatic temperature compensation module with an NTC10K sensor as a temperature sensor, so as to effectively cope with the impact of temperature changes through real-time monitoring and adjustment; the OD value detection device monitors the OD value of the extract in real time through a spectrophotometer to evaluate the concentration of the extract, adopts a 90° scattered light measurement structure, and the scattered light adopts a double-beam compensation design, and effectively improves the measurement accuracy and stability through comparative measurement and real-time correction.

5. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The data processing module processes the spectral data and numerical data obtained by the detection module, including spectral denoising, baseline correction, characteristic peak extraction, etc., and judges in real time whether the components of the extract meet the preset standards by comparing with the standard fingerprint database. By collecting 200 groups of normal batch spectral data, the spectral data is subjected to SNV preprocessing and second-order derivative transformation processing, the interference factors in the spectrum are eliminated to establish a reference database, and the characteristic wavelengths with VIP values ​​>1.5 are screened by developing a PLS-DA discrimination model.

6. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The multi-level alarm module adopts a sliding window DTW (dynamic time warping) algorithm and sets three levels of warning thresholds (yellow: 2σ, orange: 3σ, red: 5σ).

7. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The industrial Internet of Things integration module integrates the system with the industrial Internet of Things platform to achieve remote monitoring and automated operation according to the detection results. When an abnormal component is detected, the system can automatically issue an alarm or suspend production. The parameter adjustment module of the PID algorithm and the multi-parameter association model are combined to achieve automated control tasks. The key parameters of the PDI controller are Kp=0.8, Ki=0.2, and Kd=0.05; the multi-parameter association model is based on the linear model ΔOD 600 =0.15×ΔpH+0.32×ΔAbs 254 to set up.

8. The component detection system for unmanned automatic production of extract according to claim 1, characterized in that: The database and model building module constructs a bioactive component map by combining species genome, transcriptome and metabolome data based on whole genome nutrient group analysis equipment.

Images