Preparation method of additive-free broussonetia papyrifera concentrated paste based on AI assistance

Abstract

The invention discloses a preparation method of additive-free mulberry fruit concentrated paste based on AI assistance, and belongs to the field of traditional Chinese medicine processing. According to the method, a form-nutrition association database is constructed through an AI system to screen high-quality paper mulberry germplasm, and the accurate harvesting time is predicted in combination with growth environment data; the method comprises the following steps: detecting the cleanliness of harvested paper mulberry fruits through AI, carrying out AI pressure-controlled squeezing and ginseng-controlled centrifugation to obtain pure fruit juice, filtering, concentrating at the constant temperature of 80 DEG C for 6 hours through AI, judging the concentration end point of dripping into beads by combining an online refractometer and visual inspection, and finally, carrying out AI sealing detection, bottling and sealing. According to the method, no substance is added, AI assists in achieving accurate control over technological parameters, nutritional active ingredients such as amino acid and flavone of the broussonetia papyrifera are reserved, it is guaranteed that the product quality is stable, the requirement for medical digital intelligence transformation is met, and the prepared concentrated paste is dense in nutrition, easy to absorb, outstanding in health care effect and suitable for natural health preserving scenes.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine processing technology, specifically to an artificial intelligence (AI)-assisted method for preparing additive-free concentrated extract of mulberry fruit, and particularly to a technology for preparing concentrated traditional Chinese medicine products by combining AI technology to optimize mulberry fruit germplasm resources, intelligently monitor the preparation process, and precisely control process parameters. Background Technology

[0002] The mature fruit of the paper mulberry tree (also known as paper mulberry fruit) is a traditional Chinese medicine. It has medicinal value in tonifying the kidneys and clearing the liver, improving eyesight and promoting diuresis. It is also rich in amino acids, vitamins, minerals and active ingredients such as flavonoids and polysaccharides, and has outstanding nutritional and health benefits. The following are the core shortcomings currently existing in the paper mulberry processing industry: Blindness in germplasm and harvesting: The lack of accurate identification technology for high-quality germplasm of fruit trees and the reliance on manual experience for harvesting time lead to inconsistent quality of raw materials. The non-natural and complex nature of the process: Some processing techniques add hydrolytic enzymes, preservatives and other additives, which destroy the natural components of the fruit; traditional processes (high temperature drying and fermentation) are complicated and costly, and high temperatures can degrade heat-sensitive nutrients such as flavonoids and vitamins. The preparation process is not intelligent: temperature control and concentration endpoint judgment rely on manual operation, which has large errors and is prone to problems such as "insufficient concentration (easy to spoil)" or "over-concentration (poor taste)". In addition, it is impossible to dynamically adjust parameters according to raw material batches, resulting in poor product stability. Insufficient digital empowerment: It has not responded to the requirement of "AI empowering the entire pharmaceutical industry chain" in the "Implementation Plan for Digital Transformation of Pharmaceutical Industry (2025-2030)", and the integration of traditional Chinese medicine processing with AI technology is low. Summary of the Invention

[0003] This invention aims to solve: The precision of screening and harvesting high-quality germplasm of paper mulberry is insufficient; The use of additives and complex processes during processing result in a "lack of naturalness + excessively high cost"; Manual control leads to large process errors, significant nutrient loss, and unstable quality. The lack of AI technology in the field of traditional Chinese medicine processing. To achieve the above objectives, the present invention provides the following technical solution: A method for preparing an AI-assisted, additive-free concentrated extract of mulberry fruit includes the following steps: (1) AI-assisted screening of high-quality fruit germplasm and determination of optimal harvest period: An AI-powered intelligent system integrating image recognition, genomics analysis, and growth environment monitoring modules was used to screen high-quality mulberry fruit germplasm. The image recognition module collected data on fruit size, color, and texture characteristics, comparing them with a pre-built morphology-nutrient association database to select high-quality germplasm with flavonoid content ≥2.0% and vitamin C content ≥15mg / 100g. The genomics analysis module performed gene sequencing on leaf samples of high-quality germplasm, comparing the sequencing sequences with a gene bank of authentic medicinal herb germplasm to select gene sequences with a correlation of ≥85% with highly active ingredients and constructing a mulberry fruit germplasm resource database. The growth environment monitoring module collected data every 2 hours using temperature and humidity sensors and soil nutrient sensors, monitoring a temperature range of 20-30℃, humidity of 50-70%, and soil moisture content of 20-30%. After fitting the data, the AI ​​model predicted a 2-3 day harvest period corresponding to mulberry fruit with flavonoid content ≥2.3% and soluble solids ≥18%. (2) Raw material pretreatment: Harvest fresh mulberry fruit according to the optimal harvest period, rinse with running water for 2-3 minutes, and then collect dual-angle surface images using an AI vision inspection system to remove fruits with impurities accounting for ≥0.5% to obtain pre-treated fresh mulberry fruit. (3) Pressing and centrifugal separation: The pretreated fresh mulberry fruit is fed into a spiral low-temperature press. The AI ​​adjusts the pressing pressure to 0.3-0.4 MPa according to the fruit's firmness to obtain the virgin juice. The virgin juice is then placed in a high-speed centrifuge. The AI ​​reads the turbidity data of the virgin juice, sets the centrifugation speed to 3000-4000 r / min and the centrifugation time to 10-15 min. If the turbidity is >15 NTU, the centrifugation time is extended by 2-3 min to separate and obtain pure mulberry fruit juice. (4) Filtration treatment: The pure mulberry juice is filtered under normal pressure through a 100-mesh stainless steel filter screen. The turbidity of the filtrate is determined to be ≤5 NTU by an AI vision detection system combined with an online refractometer, thus obtaining clear mulberry juice. (5) AI-assisted isothermal concentration: The clarified fruit juice was placed in an electrically heated automatic temperature-controlled pot. The AI ​​temperature control module maintained the temperature at 80℃ with a fluctuation range of ≤±0.5℃ using a PID algorithm, and the concentration was continuously carried out for 6 hours. At the same time, the soluble solids content was monitored in real time by an online refractometer. When the solids content reached 45%, the AI ​​visual detection system collected images of the concentrated liquid droplets every 10 seconds. The concentration was stopped after the droplets were determined to be in a hemispherical droplet state of 0.7-0.9cm, and the concentrated fruit juice was obtained. (6) Bottling and sealing: The concentrated fruit extract was placed into a brown glass sealed bottle. An AI sealing test system was used to fill the bottle with 0.1 MPa of inert gas and maintain it for 30 seconds. After the pressure drop was detected to be <0.005 MPa, the sealing was completed. Furthermore, the morphology-nutrient association database mentioned in step (1) contains morphological characteristic data of fruits from different production areas and different varieties, as well as corresponding nutritional component detection data of flavonoids, vitamin C, and soluble solids, and the database will be continuously updated as new germplasm is screened. Furthermore, the sensor arrangement density of the growth environment monitoring module in step (1) is one group per 10 square meters to ensure that the collected data covers the main area of ​​fruit growth and that the data is transmitted to the AI ​​system in real time for dynamic analysis. Furthermore, the pressing temperature of the spiral low-temperature press in step (3) is controlled at 25-30℃ to avoid the loss of heat-sensitive nutrients in the fruit due to high temperature. Furthermore, the electric heating automatic temperature control pot mentioned in step (5) is made of food-grade stainless steel and is equipped with a stirring device. The AI ​​system controls the stirring rate to be 30-50 r / min to avoid local overheating of the concentrate. Furthermore, the light transmittance of the brown glass sealed bottle mentioned in step (6) is ≤10%, which is used to reduce the impact of light on the degradation of active ingredients in the fruit extract concentrate. Beneficial effects of this invention: (1) AI empowers raw material end: precise screening of high-quality germplasm + prediction of harvest period to ensure high nutritional activity of raw materials; (2) Natural and additive-free: No substances are added throughout the process, preserving the true flavor and nutrition of the fruit to the greatest extent, which meets the consumer demand for "natural food"; (3) Simplified intelligent process: only 4 core processes, AI controls equipment parameters, ordinary producers can operate, while improving juice utilization and reducing costs; (4) Double guarantee of nutrition and quality: Low temperature concentration at 80℃ reduces the loss of heat-sensitive components, and AI visual detection accurately controls the concentration endpoint to avoid spoilage / poor taste; (5) Align with digital transformation: Respond to the national policy of digital transformation of medicine and achieve deep integration of AI and traditional Chinese medicine processing. Attached Figure Description Figure 1 This is a schematic diagram illustrating the implementation process of an AI-assisted, additive-free mulberry fruit concentrate preparation method according to the present invention. Detailed Implementation To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. This invention proposes an AI-assisted method for preparing additive-free mulberry fruit concentrate, characterized by the following steps: (1) AI-assisted screening of high-quality fruit germplasm and determination of optimal harvest period An AI-powered intelligent system integrating image recognition, genomics analysis, and growth environment monitoring modules: Image recognition module: Pre-collect morphological images (including fruit size, color, and texture) of fruit from different production areas and varieties, and associate them with corresponding nutritional component data (flavonoid and vitamin C content) to establish a "morphology-nutrition" association database; during screening, extract the morphological features of the candidate fruit and compare them with the "high-nutrition germplasm" threshold in the database to identify high-quality germplasm; Genomics analysis module: Gene sequencing is performed on high-quality germplasm, and the sequences are compared with the germplasm gene bank of traditional Chinese medicinal materials. Gene sequences with a correlation of ≥85% with highly active ingredients are screened out and included in the fruit germplasm resource database. The optimization parameters of the breeding system are updated simultaneously. The growth environment monitoring module (temperature and humidity / soil nutrient sensor) collects environmental temperature and humidity, soil moisture content and organic matter data every 2 hours. The AI ​​model fits the correlation between the data and the accumulation of nutrients in the fruit. When the predicted flavonoid content reaches 2.3% and the soluble solids reach 18%, it outputs a precise harvest window of 2-3 days to ensure the quality of raw materials. (2) Raw material pretreatment Fresh mulberry fruit is harvested according to the harvest time predicted by AI. After rinsing with running water for 2 minutes, the surface image is collected by an AI vision inspection system (dual-angle high-definition camera). Fruits with more than 0.5% of damage, insect infestation, or impurities are removed to obtain pre-treated mulberry fruit (e.g., 10 catties of fresh mulberry fruit). (3) Pressing and centrifugal separation The pre-treated fruit is fed into a spiral low-temperature press. The AI ​​adjusts the pressing pressure according to the fruit's firmness (visually detected firmness value) (0.3 MPa for high maturity, 0.4 MPa for medium maturity) to obtain the virgin juice. The virgin juice is then placed in a high-speed centrifuge. The AI ​​reads the turbidity data of the virgin juice and sets the centrifugation speed to 3000-4000 r / min and the time to 10-15 min (extend the time by 2-3 min if the turbidity exceeds 15 NTU) to separate and obtain pure fruit juice (4 jin of juice from 10 jin of fresh fruit). (4) Filtration treatment The pure fruit juice is filtered through a 100-mesh stainless steel filter at normal pressure. The AI ​​vision inspection system uses an online refractometer to determine that the turbidity of the filtrate is ≤5 NTU. After confirming that there is no residual fruit residue, the clarified fruit juice is obtained. (5) AI-assisted isothermal concentration The clarified juice is poured into an electrically heated automatic temperature-controlled pot. The AI ​​temperature module maintains a constant temperature of 80℃ (temperature fluctuation ±0.5℃) through a PID algorithm, and the juice is continuously concentrated for 6 hours. At the same time, the solid content is monitored in real time by an online refractometer. When the solid content reaches 45%, the "drop-to-bead" high-frequency visual detection is activated (once every 10 seconds). Heating is stopped after the result is consistent with the morphology database, yielding 1 jin of concentrated fruit extract. (6) Bottling and sealing The concentrated paste is placed into a brown glass sealed bottle. The AI ​​sealing test system completes the sealing by filling it with 0.1MPa inert gas and detecting the pressure change (a drop of <0.005MPa is considered acceptable). Example: Preparation of Additive-Free Morus Fruit Concentrate Based on AI Assistance This embodiment selects an experimental plot in the authentic producing area of ​​Broussonetia papyrifera in the middle and lower reaches of the Yangtze River (longitude 116°E, latitude 30°N). Three batches of high-quality Broussonetia papyrifera germplasm have been entered into the archive in this area in the early stage. The goal of this batch is to prepare concentrated Broussonetia papyrifera extract with high flavonoid content. (1) AI germplasm screening and harvesting The AI ​​image recognition module first collected images of the fruits of 10 paper mulberry trees within the plot, extracting the color parameters (L value 52±2, a value 18±1) and diameter (1.2±0.1cm) of each fruit. These parameters were compared with the characteristic thresholds of "high flavonoid germplasm" in the "Paper Mulberry Fruit High-Quality Germplasm Image Database," and three paper mulberry trees meeting the criteria were selected. Subsequently, the genomics analysis module performed gene sequencing on leaf samples from these three paper mulberry trees. The obtained sequences showed a 92% match with the "key genes for flavonoid synthesis" in the database, confirming them as the target germplasm. The growth environment monitoring module collected data every 2 hours starting from the fruit enlargement period: daytime temperature 25-27℃, nighttime temperature 18-20℃, air humidity 60-65%, soil moisture content 22-25%, and soil organic matter content 1.8%. The AI ​​model input this data into the "Sorbus nutritional accumulation model," predicting that the flavonoid content would peak at 2.4% and the soluble solids content at 18.5% on the third day after fruit ripening. Therefore, the harvest period was determined to be "2-3 days after ripening." During harvesting, damaged and insect-infested fruits were removed using AI visual inspection, ultimately yielding 10 jin (5 catties) of qualified fresh Sorbus nut. (2) Raw material pretreatment The harvested paper mulberry fruits were poured into a cleaning tank and rinsed with running water for 2 minutes. After rinsing, they were sent to an AI visual inspection station (equipped with two high-definition cameras, one for overhead views and one for side views) to capture surface images of each fruit and identify the percentage of pixels with residual impurities. When the percentage was less than 0.5%, the fruit was considered clean and qualified. In this batch, 0.2 catties of fruit with excessive impurities were removed, resulting in 9.8 catties of pre-treated fresh paper mulberry fruits. (3) Pressing and centrifugal separation 9.8 catties of pre-treated fruit were fed into a spiral low-temperature press. The AI ​​(AI system) adjusted the pressing pressure to 0.35 MPa based on the fruit's firmness (hardness value 45±3N) obtained through visual detection. During pressing, the AI ​​monitored the juice extraction rate in real time (maintaining it at around 200 g / min) to prevent excessively fast extraction from causing pulp contamination, ultimately yielding 4.2 catties of virgin juice. The virgin juice was then transferred to a high-speed centrifuge. The AI ​​read the turbidity value of the virgin juice as 12 NTU and set the centrifugation speed to 3500 r / min and the centrifugation time to 12 minutes. During centrifugation, the AI ​​monitored the centrifuge's vibration amplitude (controlled within 0.5 mm) to prevent equipment malfunctions from affecting the separation effect. After centrifugation, the pulp and juice were clearly separated. 4 catties of pure juice was collected through an automatic separating valve (the juice yield was slightly lower than the theoretical value due to the removal of a small amount of substandard fruit, but within the error range). (4) Filtration treatment Four kilograms of pure fruit juice were filtered under normal pressure through a 100-mesh stainless steel filter. After filtration, the AI ​​vision inspection station collected images of the clarity of the filtrate. The "turbidity-pixel" correlation model determined that the turbidity of the filtrate was 3 NTU, which meets the standard of "clarified fruit juice" (turbidity ≤ 5 NTU), thus obtaining four kilograms of clear fruit juice. (5) AI isothermal concentration The clarified juice was poured into a 5L electrically heated automatic temperature-controlled pot. The AI ​​temperature module first preheated the pot to 60℃, then increased the temperature to 80℃ at a rate of 5℃ / min. Subsequently, the heating power was adjusted through a PID algorithm (maintained at around 800W) to ensure that the temperature fluctuation was controlled within ±0.5℃. During the concentration process, the soluble solids content was measured every 30 minutes by an online refractometer: 28% in the 3rd hour, 35% in the 4th hour, 42% in the 5th hour, and 45% in the 6th hour. At this time, the AI ​​started the "droplet formation" high-frequency monitoring (collecting a drop image every 10 seconds). At 6 hours and 15 minutes, the collected droplets formed a hemispherical shape with a diameter of 0.8cm on the glass plate, which perfectly matched the characteristics of the "droplet formation morphology database". The AI ​​immediately issued a command to stop heating. After natural cooling for 5 minutes, 1 jin of fruit concentrate with a deep purplish-red color was obtained. (6) Bottling and sealing The concentrated paste was dispensed into five 200g brown glass sealed bottles. After dispensing, the bottles were sent to the AI ​​(Inert Gas) sealing test station: 0.1MPa inert gas was introduced into the bottles, and the pressure change was measured after 30 seconds. The pressure drop of all five bottles in this batch was less than 0.005MPa, indicating that the seal was qualified. Finally, the batch information and key parameters recorded by the AI ​​(flavonoid content 2.35%, solids content 46%) were labeled on the bottles, and the sealing was completed. The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for preparing an AI-assisted, additive-free concentrated extract of mulberry fruit, characterized in that, Includes the following steps: (1) AI-assisted screening of high-quality fruit germplasm and determination of optimal harvest period: An AI-powered intelligent system integrating image recognition, genomics analysis, and growth environment monitoring modules was used to screen high-quality mulberry fruit germplasm. The image recognition module collected data on fruit size, color, and texture characteristics, comparing them with a pre-built morphology-nutrient association database to select high-quality germplasm with flavonoid content ≥2.0% and vitamin C content ≥15mg / 100g. The genomics analysis module performed gene sequencing on leaf samples of high-quality germplasm, comparing the sequencing sequences with a gene bank of authentic medicinal herb germplasm to select gene sequences with a correlation of ≥85% with highly active ingredients and constructing a mulberry fruit germplasm resource database. The growth environment monitoring module collected data every 2 hours using temperature and humidity sensors and soil nutrient sensors, monitoring a temperature range of 20-30℃, humidity of 50-70%, and soil moisture content of 20-30%. After fitting the data, the AI ​​model predicted a 2-3 day harvest period corresponding to mulberry fruit with flavonoid content ≥2.3% and soluble solids ≥18%. (2) Raw material pretreatment: Harvest fresh mulberry fruit according to the optimal harvest period, rinse with running water for 2-3 minutes, and then collect dual-angle surface images using an AI vision inspection system to remove fruits with impurities accounting for ≥0.5% to obtain pre-treated fresh mulberry fruit. (3) Pressing and centrifugal separation: The pretreated fresh mulberry fruit is fed into a spiral low-temperature press. The AI ​​adjusts the pressing pressure to 0.3-0.4 MPa according to the fruit's firmness to obtain the virgin juice. The virgin juice is then placed in a high-speed centrifuge. The AI ​​reads the turbidity data of the virgin juice, sets the centrifugation speed to 3000-4000 r / min and the centrifugation time to 10-15 min. If the turbidity is >15 NTU, the centrifugation time is extended by 2-3 min to separate and obtain pure mulberry fruit juice. (4) Filtration treatment: The pure mulberry juice is filtered under normal pressure through a 100-mesh stainless steel filter screen. The turbidity of the filtrate is determined to be ≤5 NTU by an AI vision detection system combined with an online refractometer, thus obtaining clear mulberry juice. (5) AI-assisted isothermal concentration: The clarified fruit juice was placed in an electrically heated automatic temperature-controlled pot. The AI ​​temperature control module maintained the temperature at 80℃ with a fluctuation range of ≤±0.5℃ using a PID algorithm, and the concentration was continuously carried out for 6 hours. At the same time, the soluble solids content was monitored in real time by an online refractometer. When the solids content reached 45%, the AI ​​visual detection system collected images of the concentrated liquid droplets every 10 seconds. The concentration was stopped after the droplets were determined to be in a hemispherical droplet state of 0.7-0.9cm, and the concentrated fruit juice was obtained. (6) Bottling and sealing: The concentrated fruit extract was placed into a brown glass sealed bottle. An AI sealing test system was used to fill the bottle with 0.1 MPa of inert gas and maintain it for 30 seconds. After the pressure drop was detected to be <0.005 MPa, the sealing was completed.

2. The method for preparing an AI-assisted, additive-free mulberry fruit concentrate according to claim 1, characterized in that: The morphology-nutrient association database mentioned in step (1) contains morphological characteristic data of fruit from different production areas and different varieties, as well as corresponding nutritional component detection data of flavonoids, vitamin C and soluble solids, and the database will be continuously updated as new germplasm is screened.

3. The method for preparing an AI-assisted, additive-free mulberry fruit concentrate according to claim 1, characterized in that: The sensor arrangement density of the growth environment monitoring module in step (1) is 1 group per 10 square meters to ensure that the collected data covers the main area of ​​fruit growth and that the data is transmitted to the AI ​​system in real time for dynamic analysis.

4. The method for preparing an AI-assisted, additive-free mulberry fruit concentrate according to claim 1, characterized in that: The pressing temperature of the spiral low-temperature press in step (3) is controlled at 25-30℃ to avoid the loss of heat-sensitive nutrients in the fruit due to high temperature.

5. The method for preparing an AI-assisted, additive-free mulberry fruit concentrate according to claim 1, characterized in that: The electric heating automatic temperature control pot mentioned in step (5) is made of food-grade stainless steel and is equipped with a stirring device. The AI ​​system controls the stirring rate to be 30-50 r / min to avoid local overheating of the concentrate.

6. The method for preparing an AI-assisted, additive-free mulberry fruit concentrate according to claim 1, characterized in that: The light transmittance of the brown glass sealed bottle mentioned in step (6) is ≤10%, which is used to reduce the impact of light on the degradation of active ingredients in the concentrated fruit extract.

Images