NFC fruit juice non-thermal processing system based on intelligent regulation and control

Abstract

The invention discloses an NFC fruit juice non-thermal processing system based on intelligent regulation and control, and relates to the technical field of food processing, and the system comprises an integrated physical module for realizing whole-process sterile processing; the data acquisition and preprocessing module is responsible for parameter acquisition and processing; the digital twin modeling module constructs a virtual-real mapping model; the intelligent regulation and control early warning module generates a regulation and control instruction and fault early warning; the visual operation and maintenance module realizes interactive operation and full-process parameter optimization management; through an integrated physical module and an intelligent regulation and control framework, NFC fruit juice whole-process sterile intelligent processing is achieved, nutrition damage caused by hot processing is avoided, and sterilization safety is guaranteed. By means of data preprocessing and digital twinning technologies, the problem of traditional parameter lag is solved, and the product freshness and the production continuity are improved; through virtual-real coupling regulation and control and residual life prediction, precise machining and prospective operation and maintenance are realized, the shutdown maintenance probability is reduced, the efficiency is improved, and the cost is saved.

Description

Technical Field

[0001] This invention relates to the field of food processing technology, specifically to an NFC non-thermal fruit juice processing system based on intelligent control. Background Technology

[0002] NFC (Not From Concentrate) juice boasts natural nutritional components and flavor, meeting consumers' demands for healthy food and thus driving increasing market demand. While traditional heat sterilization methods can achieve sterilization, they can easily destroy heat-sensitive substances such as vitamins and active enzymes in the juice, reducing product quality. Therefore, non-thermal processing technologies have gradually become a research hotspot. With the development of intelligent manufacturing technology, sensors, digital twins, and intelligent control technologies are being applied in food processing, gradually achieving precision and efficiency in non-thermal processing. However, due to technological limitations, current non-thermal processing is only suitable for single-equipment systems, has poor adaptability to raw materials, and cannot be adjusted online in real time. To meet the automation, precision, speed, and intelligence requirements of NFC juice production, an integrated, sterile, and intelligent non-thermal processing system needs to be established to further improve product quality.

[0003] Traditional NFC (Not From Concentrate) non-thermal processing technology for juice has several shortcomings: First, the processing mode involves the assembly of various devices, lacking a unified design. Online detection and rejection of defective products rely on manual intervention, resulting in weak sealing at pipeline connections, a risk of secondary contamination, and the inability to achieve seamless online connection. Second, data acquisition lacks targeted preprocessing, making sensor signals susceptible to processing pressure and environmental interference, leading to insufficient data accuracy. Third, there is no production control and maintenance based on a virtual-to-real mapping model of the processing process. Various production conditions are difficult to adjust according to actual production conditions, making it impossible to correctly judge and adjust based on different raw material properties and processing load conditions; parameter adjustment is lagging and lacks precision. Fourth, repairs are mostly carried out after malfunctions occur, making it impossible to predict and analyze the equipment's operating status. Failure of critical components may cause the equipment to stop working, and maintenance is generally performed manually, which is extremely labor-intensive. Product quality issues, high energy consumption, and instability are also serious problems. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an NFC (Not From Concentrate) juice non-thermal processing system based on intelligent control. This system achieves aseptic and intelligent management of the entire process from raw material processing to finished product bottling through the coordinated operation of five modules: integrated physical processing, data acquisition and preprocessing, digital twin modeling, intelligent control and early warning, and visualized operation and maintenance. The system uses ultra-high pressure-cold plasma-coordinated non-thermal processing as its core, combined with proprietary formulas such as adaptive weighted filtering based on raw material characteristics and multi-field coordinated control through virtual-real coupling. This allows for precise data filtering and dynamic optimization of processing parameters, avoiding problems such as nutrient loss and parameter adaptation lag in traditional thermal processing. Relying on digital twin virtual-real synchronization and load-related remaining life prediction technology, it achieves visualized processing and proactive equipment operation and maintenance, improving the nutrient retention rate and quality stability of NFC juice while reducing energy consumption and operation and maintenance costs, providing efficient technical support for large-scale production and small-batch customization in the industry.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an NFC non-thermal juice processing system based on intelligent control, the system comprising: Integrated physical module: Through conveying pipelines and quick-release clamp sealing structure, it connects intelligent sorting equipment, cleaning equipment, pressing equipment, ultra-high pressure-cold plasma synergistic non-thermal processing equipment, aseptic filling equipment and online detection equipment to form a whole-process aseptic processing structure covering raw material processing, non-thermal processing to finished product filling; Data acquisition and preprocessing module: Collects various parameters such as raw material characteristics, equipment operation, material status and energy consumption through sensors, calls built-in data resources, completes data processing and transmission through the raw material characteristic adaptive weighted filtering formula; Digital twin modeling module: Based on the processed data transmitted by the above preprocessing module, construct a virtual-real mapping model that corresponds to the physical processing process in a 1:1 ratio, and maintain real-time synchronization between the model and the physical processing process; Intelligent control and early warning module: Receives model data from the digital twin modeling module, generates control commands for the processing equipment through a virtual-real coupled multi-field collaborative control formula, and generates fault early warnings for key components through a load-related remaining life prediction formula. Visualized Operation and Maintenance Module: Receives control commands and fault warning information from the intelligent control and early warning module, performs interactive operations on multiple terminals, and uses closed-loop optimization weight iteration formula to complete parameter optimization and operation and maintenance management throughout the entire process.

[0006] Furthermore, the online detection equipment of the integrated physical module includes a liquid level sensor, a sealing sensor, and a metal foreign object detector, etc., wherein different types of sensors each complete the detection tasks of different indicators after the finished product is filled; the online detection equipment is equipped with a defective product rejection device, which reacts very quickly and will not affect the normal production process.

[0007] Furthermore, in the integrated physical module, the conveying pipes are made of materials that meet food contact safety requirements, and the aseptic treatment at the pipe joints is ensured by the quick-release clamp sealing structure. The ultra-high pressure-cold plasma synergistic non-thermal processing equipment includes an ultra-high pressure chamber and a cold plasma generating unit. After the ultra-high pressure chamber is strengthened, it can meet the processing pressure requirements. In the cold plasma generating unit, dielectric barrier discharge technology is used, and the distance between the electrodes is fixed and the discharge power can be adjusted.

[0008] Furthermore, the formula for calculating the adaptive weighted filtering of raw material characteristics is as follows: ; in, For the first Preprocessed data from each sensor; Weights based on raw material characteristics; For the first The raw data collected by each sensor; The neighboring sensor collaborative weights are such that the sum of all neighboring sensor collaborative weights is 1; This refers to the number of neighboring sensors within the same unit. For the first Raw data collected by the neighborhood sensors; This is the load sensitivity coefficient; For the first The processing pressure corresponding to each sensor; For the first The processing pressure corresponding to each neighboring sensor.

[0009] Furthermore, the digital twin modeling module is constructed using 3D modeling technology combined with a dynamic simulation engine, integrating fluid dynamics models, structural mechanics models, and multi-field coupling models; it has built-in databases of various fruit juice raw material characteristics and equipment parameters, and stores real-time data on site in a time-series database. When errors or mismatches occur in the digital twin, the model can be adjusted by real-time incremental updates and periodic full calibration to ensure that the error between the digital twin and the physical processing process is at a low level.

[0010] Furthermore, the calculation formula for the virtual-real coupled multi-field coordinated control is as follows: ; in, For the first Output value of each control parameter; Standard parameter weights; For the first Standard process values ​​for each control parameter; The virtual-to-real deviation response coefficient; This is a discrepancy between the real and virtual values; This is the synergistic effect coefficient; The rate of change of multi-field synergistic effect; This is a raw material characteristic correction factor.

[0011] Furthermore, the formula for calculating the load-related remaining life prediction is as follows: ; in, For the remaining life of the component, For the rated service life of the component, The loss acceleration factor, To average processing pressure, For rated pressure, The pressure load sensitivity index, The average vibration amplitude, Vibration threshold The vibration sensitivity index For the current runtime, This indicates the duration of time the component has been in use.

[0012] Furthermore, the closed-loop optimization weight iteration calculation formula includes: ; in, For the next moment Weights of raw material characteristics for each sensor; For the current moment Weights of raw material characteristics for each sensor; This is the quality feedback coefficient; This represents the overall value of actual product quality. This is the standard quality composite value; For the current moment Data from each sensor after preprocessing; For the next moment Each control parameter has a standard parameter weight; For the current moment Each control parameter has a standard parameter weight; This is the lifespan feedback coefficient; This represents the remaining lifespan of the component at the current moment. The remaining safe lifespan threshold for the component; For the current moment Each control parameter value.

[0013] Furthermore, the intelligent control and early warning module's fault warning is determined based on the result calculated by the load-related remaining life prediction formula. In the process of implementing this function, graded warnings are issued according to the difference between the component's remaining life and the safe remaining life threshold. Specifically, a level one warning is issued when the component's remaining life is more than 50% higher than the safe threshold; a level two warning is issued when the component's remaining life is between 20% and 50% of the safe threshold; and a level three warning is issued when the component's remaining life is less than 20% of the safe threshold. The warning information includes the component name, installation location, maintenance suggestions, etc., and is pushed to multiple terminals by the visual operation and maintenance module. At the same time, the location information of the faulty component is highlighted on the digital twin modeling module.

[0014] Compared with existing technologies, this NFC non-thermal juice processing system based on intelligent control has the following advantages: I. This invention constructs an intelligent control architecture for the entire NFC juice production process, based on an integrated physical module and combined with multi-module collaboration. The goal is to achieve sterilization and intelligent control throughout the entire NFC juice production process, from raw materials to finished product. It establishes an integrated physical module, a multi-module intelligent collaborative system, and a real-time online monitoring and control system, ensuring precise and reliable connection of processing equipment. It features food-grade conveying pipelines and quick-connect clamp sealing structures, and utilizes ultra-high pressure cold plasma in conjunction with non-thermal processing equipment. This reduces the impact of heat treatment on the nutritional components of the juice and ensures sterilization effectiveness and processing safety through dielectric barrier discharge technology and enhanced ultra-high pressure chamber design. Furthermore, considering the dielectric characteristics of this type of food, an adaptive weighted filtering formula based on raw material characteristics is used to preprocess multi-sensor sampling data, which is then applied to the 1:1 virtual-real mapping and real-time synchronization mechanism of the digital twin modeling module. This makes the processing process visible and traceable, solving the drawbacks of lagging traditional processing parameter matching, reducing large fluctuations in product quality, and achieving the goals of high freshness, high nutrient retention, and strong production continuity for NFC juice.

[0015] II. This invention employs a virtual-real coupled multi-field collaborative control mechanism and load-related remaining life prediction technology to achieve precise control of the processing process and proactive equipment maintenance. Utilizing a proprietary formula within the intelligent control and early warning module, dynamic control commands are generated based on standard process parameters, virtual-real deviations, and multi-field synergistic effects, ensuring that all raw materials with different characteristics can be matched with the most suitable processing scheme. Through calculation and analysis using the load-related remaining life prediction formula, parameters such as average processing pressure and vibration amplitude are obtained, and early warnings of potential failures in critical equipment components are provided. With parameter adjustments using a closed-loop optimization weight iteration formula, the goal of significantly reducing equipment downtime and energy consumption is achieved. A visual maintenance module enables a simple operating interface on all terminals. This not only facilitates operation for personnel, improves production and maintenance efficiency, reduces reliance on human intervention, and lowers operating costs, but also provides strong technical support for improving the quality and increasing the quantity of NFC juice industrial production.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 A flowchart of an NFC non-thermal juice processing system based on intelligent control; Figure 2 This is a framework diagram of an NFC (Not From Concentrate) non-thermal juice processing system based on intelligent control. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] Example 1: Scaled-up continuous production scenario of fresh orange NFC juice This embodiment is applied to the large-scale continuous production of fresh orange NFC juice. By having various modules in the system work collaboratively, the entire process from raw material processing and non-thermal processing to finished product bottling is conducted in a sterile environment. Through intelligent methods, a stable production process and continuous, efficient production are achieved. The implementation steps are as follows: Figure 1 As shown.

[0021] In the raw material pretreatment stage, freshly picked oranges are transported to an intelligent sorting device with an integrated physical module. This device can automatically identify and remove rotten, damaged, and unripe oranges, ensuring the consistency of the raw materials and reducing quality deviations caused by the raw materials in subsequent processing. The remaining qualified oranges after screening enter the cleaning equipment, where they are cleaned with high-pressure spraying to remove dirt and residual contaminants from the surface of the oranges, meeting the cleanliness requirements in the food processing process and facilitating subsequent aseptic processing.

[0022] After cleaning, the fresh oranges enter the pressing equipment and are juiced using a low-temperature pressing process in a sterile environment. Low-temperature pressing can maximize the retention of heat-sensitive nutrients such as vitamin C in the fresh oranges, while avoiding the deterioration of juice flavor caused by high temperatures. The extracted orange juice is transported through conveying pipes that meet food contact safety standards. The pipe connections use a quick-release clamp sealing structure, which can effectively prevent external microorganisms from entering the pipe, ensuring the sterility of the juice during transportation and avoiding secondary contamination.

[0023] During the juice delivery process, the data acquisition and preprocessing module activates sensors deployed at key nodes to collect real-time data on raw material characteristics, equipment operation, material status, and energy consumption. Raw material characteristic data includes orange juice acidity and pulp particle size; equipment operation data includes pressing pressure and conveying speed; material status data includes juice temperature and viscosity; and energy consumption data includes the real-time power consumption of each piece of equipment. After data collection, the module calls upon its built-in orange juice raw material characteristic database and equipment parameter database, and processes the raw data using an adaptive weighted filtering formula based on raw material characteristics. The adaptive weighted filtering formula for raw material characteristics is as follows: ; in, For the first Preprocessed data from each sensor; Weights based on raw material characteristics; For the first The raw data collected by each sensor; The neighboring sensor collaborative weights are such that the sum of all neighboring sensor collaborative weights is 1; This refers to the number of neighboring sensors within the same unit. For the first Raw data collected by the neighborhood sensors; This is the load sensitivity coefficient; For the first The processing pressure corresponding to each sensor; For the first The formula can filter out interference information during the data acquisition process and improve data accuracy by reasonably allocating the weights of raw material characteristics and the collaborative weights of neighboring sensors. The processed effective data will be transmitted to the digital twin modeling module.

[0024] After receiving the pre-processed available data, the digital twin modeling module uses 3D modeling technology and a dynamic simulation engine to construct a virtual-real mapping model that corresponds to the actual production process in a 1:1 ratio. It integrates simulation models such as fluid dynamics, structural mechanics, and multi-field coupling, which can accurately simulate the flow of raw materials in the pipeline, the structural stress of equipment operation, and non-thermal processing. Based on a time-series database, it stores real-time data and improves it by combining real-time incremental updates and periodic full calibration. In this process, real-time incremental updates ensure that the model can quickly adapt to the physical process, while periodic full calibration prevents the accumulation of deviations caused by long-term application of the model. This allows the model to better synchronize with the physical processing process, minimizes errors, and provides excellent simulation data support for the subsequent intelligent control.

[0025] Fresh orange juice enters the ultra-high pressure-cold plasma synergistic non-thermal processing device within the integrated physical module. The ultra-high pressure chamber undergoes enhanced treatment to meet the higher processing pressure required for high-volume processing, while preventing chamber deformation. The cold plasma generation unit employs dielectric barrier discharge technology, with a fixed electrode distance and adjustable discharge power. Different discharge powers are selected for different juice processing scenarios to achieve better plasma sterilization effects. At this point, the intelligent control and early warning module acquires model data output from the digital twin modeling module and derives control commands for the processing equipment through a virtual-real coupled multi-field synergistic control formula. The virtual-real coupled multi-field synergistic control calculation formula is as follows: ; in, For the first Output value of each control parameter; Standard parameter weights; For the first Standard process values ​​for each control parameter; The virtual-to-real deviation response coefficient; This is a discrepancy between the real and virtual values; This is the synergistic effect coefficient; The rate of change of multi-field synergistic effect; The formula is a raw material characteristic correction factor. It can comprehensively consider the standard parameter weight, virtual and real deviation, multi-field synergistic effect change rate and raw material characteristic correction factor. It can enable the control command to meet the sterilization requirements and avoid the loss of juice nutrients and changes in flavor caused by over-processing, so as to achieve precise control of the processing process.

[0026] Fresh orange juice that has undergone non-thermal processing flows directly through an integrated physical module for online inspection. This type of equipment typically includes components such as a level sensor, a sealing sensor, and a metal foreign object detector. The level sensor detects whether the juice level is stable and whether normal filling can proceed, ensuring that the filling quantity is equal across all lines. The sealing sensor detects whether there are leaks in the outlet pipeline that could lead to the intrusion of external microorganisms. The metal foreign object detector checks for any foreign objects mixed into the juice that could endanger human health and safety. Once a defective product is detected, it is rejected by the defective product rejection device on the online inspection equipment without causing downtime, enabling continuous production without interruption and improving the product qualification rate.

[0027] Qualified fresh orange juice enters the aseptic filling equipment of the integrated physical module. The filling process is carried out in a sterile environment, which can avoid microbial contamination during the filling process. The equipment completes the filling and capping operations according to the preset specifications to form the finished fresh orange NFC juice. The aseptic filling process can significantly extend the shelf life of the finished product while preserving the fresh flavor of the orange.

[0028] During production, the intelligent control and early warning module calculates the remaining lifespan of critical components such as the ultra-high pressure cavity and cold plasma electrodes in real time based on the load-related remaining lifespan prediction formula. The load-related remaining lifespan prediction calculation formula is as follows: ; in, For the remaining life of the component, For the rated service life of the component, The loss acceleration factor, To average processing pressure, For rated pressure, The pressure load sensitivity index, The average vibration amplitude, Vibration threshold The vibration sensitivity index For the current runtime, This refers to the usage time of the component. The formula can accurately assess the component's wear and tear by combining parameters such as average processing pressure, average vibration amplitude, and current operating time. The module divides the warning range into three levels based on the difference between the component's remaining lifespan and the safe remaining lifespan threshold. If the remaining lifespan of a critical component is between 20% and 50% of the safe threshold, a level two warning will be triggered. The warning information includes the name of the critical component, its installation location, and maintenance recommendations, and will be simultaneously pushed to multiple ports such as production management and equipment maintenance for relevant personnel to view its status in real time. Simultaneously, the location of the fault point will be marked in the digital twin modeling module, highlighting the fault risk point in the digital twin scenario and providing operators with more visual alerts. The visualized operation and maintenance module achieves full-process parameter optimization and operation and maintenance management through the closed-loop optimization weight iteration formula. The closed-loop optimization weight iteration calculation formula includes: ; in, For the next moment Weights of raw material characteristics for each sensor; For the current moment Weights of raw material characteristics for each sensor; This is the quality feedback coefficient; This represents the overall value of actual product quality. This is the standard quality composite value; For the current moment Data from each sensor after preprocessing; For the next moment Each control parameter has a standard parameter weight; For the current moment Each control parameter has a standard parameter weight; This is the lifespan feedback coefficient; This represents the remaining lifespan of the component at the current moment. The remaining safe lifespan threshold for the component; For the current moment The formula can iteratively update the weights of raw material characteristics and standard parameters by combining the differences between the actual product quality comprehensive value and the standard quality comprehensive value, as well as the remaining life of the components. This will continuously improve the stability and efficiency of the production process and provide better parameter support for subsequent large-scale production.

[0029] In summary, the intelligent control-based non-thermal processing system for large-scale continuous production of fresh orange NFC juice comprises five modules that achieve end-to-end aseptic and intelligent control. The integrated physical module enables aseptic transport, aseptic filling, non-thermal processing of NFC juice, and low-temperature juicing to preserve the heat-sensitive nutrients of fresh oranges, from raw material receiving and processing to finished product bottling. The data acquisition and preprocessing module uses an adaptive weighted filtering formula based on raw material characteristics to filter and remove interfering data for later model building. The digital twin modeling module utilizes a low-synchronization-error virtual-real mapping model to reflect the current production status in real time. The intelligent control and early warning module optimizes processing parameters based on a virtual-real coupled multi-field collaborative control formula and uses a load-related remaining life prediction formula to provide tiered early warnings for key components. The visualized operation and maintenance module uses a closed-loop optimization weighted iterative formula to continuously improve relevant processes, which can largely ensure product quality stability under large-scale production conditions and significantly improve production efficiency and continuity.

[0030] Example 2: Small-batch customized production scenario of blueberry NFC juice This embodiment is applied to the small-batch customized production of blueberry NFC juice. Addressing the characteristics of large batch variations in raw materials and flexible product specifications in small-batch production, it achieves precise control over customized production through personalized adaptation of various system modules, ensuring that each batch of products meets customized quality requirements. The implementation steps are as follows: Figure 2 As shown.

[0031] In the raw material pretreatment stage, this stage focuses on small-batch customization needs and uses high-quality fresh blueberries as raw materials. The blueberries are transported to an intelligent sorting device with an integrated physical module. This device can accurately sort the blueberries according to their size, ripeness, and other characteristics, removing unripe, rotten, and damaged blueberries to ensure the uniformity of raw materials and reduce batch quality fluctuations caused by raw material differences. After sorting, the blueberries are rinsed and sterilized by a gentle, slow spray of water. During this process, high-pressure water should not be used to directly wash the surface of the blueberries to prevent damage to the skin, resulting in pulp loss and anthocyanin loss. At the same time, it can also remove impurities and contaminants from the surface of the raw materials to ensure that the hygiene of the raw materials meets the requirements.

[0032] The washed blueberries are directly put into the press, where they are juiced using a low-pressure, low-temperature pressing method. This low-pressure method, instead of the previously used crushing and pressing, reduces the amount of bitter substances squeezed out of the blueberry seeds and avoids raising the temperature of the pulp due to high pressure, thus maximizing the retention of anthocyanins and other active substances. The extracted blueberry juice is then transported to the appropriate location via food-grade conveying pipes. The quick-release clamps at the pipe connections prevent the intrusion of external bacteria, keeping the juice in a sterile state during transmission and preventing secondary contamination.

[0033] During the raw juice transportation process, the data acquisition and preprocessing module will be activated to collect various types of data such as the characteristics of blueberry raw materials, equipment operation parameters, material status, and energy consumption. After the data acquisition is completed, the module will call the exclusive blueberry raw material characteristic database and use the raw material characteristic adaptive weighted filtering formula to process the data. The key point highlighted in this formula is the weight value of the blueberry raw material characteristics, making the obtained data better match the data characteristics of the blueberry processing process, effectively filtering out interference information, and ensuring that the processed data can more accurately represent the actual blueberry processing situation. The effective processed data is transmitted to the digital twin modeling module.

[0034] Based on the effective data collected, the digital twin modeling module establishes a customized virtual-real mapping model suitable for small-batch production, integrating fluid dynamics models, structural mechanics models, and multi-field coupling models to accurately simulate the flow and processing state of blueberry raw juice in the small-batch production process. The real-time production data is saved using a time series database. Considering the characteristic that small-batch production is easily affected by raw material batches, a more frequent real-time incremental update and regular full-scale calibration mechanism are adopted in the model. The real-time incremental update can promptly reflect the fluctuations in the production process caused by the changes in the characteristics of each batch of raw materials, and the regular full-scale calibration can ensure the accuracy of the long-term operation of the model, ultimately achieving the real-time synchronization of the model and the physical processing process, providing precise support for the intelligent control of small-batch customized production.

[0035] The blueberry raw juice enters the ultra-high pressure-cold plasma collaborative non-thermal processing equipment in the integrated physical module. The intelligent control and early warning module calculates the corresponding personalized control instructions through the virtual-real coupling multi-field collaborative control formula based on the real-time model data fed back by the digital twin modeling module to the intelligent control and early warning module. When calculating, the raw material characteristic correction coefficient is comprehensively considered, and the parameters such as the pressure in the ultra-high pressure chamber and the discharge power of the cold plasma generation unit are adjusted at any time according to the virtual-real deviation and the change rate of the multi-field collaborative effect, achieving the purpose of ensuring that the raw juice can kill germs and meet food safety requirements, and at the same time achieving the purpose of not destroying the nutritional components and unique flavor of blueberries, thus achieving the purpose of precise processing control in the customized production process.

[0036] The non-thermally processed blueberry raw juice flows through the on-line detection equipment, which comprehensively detects indicators such as the retention rate of nutritional components, microbial content, and metal foreign object content in the raw juice. The detection of the retention rate of nutritional components can ensure that the active components in blueberries are not damaged due to processing, the detection of microbial content can verify whether the sterilization effect meets the standard, and the detection of metal foreign objects can ensure product safety. The detection data is real-time fed back to the digital twin modeling module for further calibration and optimization of the model. If unqualified products are detected, the rejection device配套 with the on-line detection equipment will be immediately activated to quickly reject the unqualified products and prevent unqualified products from entering the subsequent links.

[0037] After passing inspection, the blueberry juice enters the aseptic filling equipment, where it undergoes standardized aseptic filling and capping operations according to customized requirements. The entire process is carried out in a sterile environment, ensuring that the filling process is not contaminated by external microorganisms, thus guaranteeing the stability of the finished product quality and meeting the customized specifications of different customers.

[0038] During production, the intelligent control and early warning module uses a load-related remaining life prediction formula to monitor the remaining life of critical components such as the pressing rod and ultra-high pressure chamber seals in real time. It calculates parameters such as average processing pressure, average vibration amplitude, and usage time of key components using formulas to determine the degree of component wear. When the remaining life of a critical component exceeds 50% of the safety threshold, a level one early warning is triggered. This warning information is pushed to relevant terminals through the visual operation and maintenance module to remind maintenance personnel to pay attention to the component's status and make preparations in advance to avoid production stoppages due to sudden component failure. The visual operation and maintenance module has multiple terminal interface functions, allowing production management personnel to view production progress, equipment operation status, and early warning messages in real time on computers and mobile phones. Based on the output results of this batch of blueberry NFC juice and the corresponding remaining life data of the components, the system uses the specific characteristics of the closed-loop optimization weight iteration formula. Building upon previous work, it iteratively adjusts the weights of raw material characteristics and standard parameters of control parameters, saving the adjusted parameters to the database. This ensures better quality stability and work efficiency for subsequent small-batch customized production batches of blueberry NFC juice.

[0039] In summary, in the small-batch customized production of blueberry NFC juice, this system provides personalized control from the perspectives of raw material batch differentiation and individual customization. The integrated physical module employs precise sorting, gentle cleaning, and low-pressure, low-temperature pressing to process NFC juice, while minimizing anthocyanin loss and flavor deterioration. The collected data is processed using a blueberry-specific database and its data acquisition and preprocessing algorithms, utilizing an adaptive weighted filtering formula based on raw material characteristics to obtain suitable data values. A digital twin modeling method is used to establish a high-frequency incremental update real-time modeling mechanism for small-batch production. Customized processing instructions are obtained through a virtual-real coupled multi-field collaborative control formula, and component early warning information is obtained based on a load-related remaining life prediction formula. The visualized operation and maintenance module supports multi-terminal interaction and optimizes parameters using a closed-loop optimization weight iteration formula, ensuring that each batch of products meets quality standards and providing accurate parameter accumulation for subsequent customized production, adapting to the flexibility requirements of small-batch production.

[0040] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. NFC juice non-thermal processing system based on intelligent control, characterized by, The system comprises: An integrated physical module: through the conveying pipeline and the quick-mounting clamp sealing structure, the intelligent sorting device, the cleaning device, the pressing device, the ultra-high pressure-cold plasma collaborative non-thermal processing device, the sterile filling device and the online detection device are connected in series to form a full-process sterile processing structure covering raw material processing, non-thermal processing to finished product filling; A data acquisition and preprocessing module: through sensors, raw material characteristics, equipment operation, material state and energy consumption parameters are collected, after calling built-in data resources, data processing is completed through a raw material characteristic self-adaptive weighted filtering formula and then data are transmitted; A digital twin modeling module: based on the processed data transmitted by the above-mentioned preprocessing module, a virtual-real mapping model corresponding to the physical processing process 1:1 is constructed, and the model is kept in real-time synchronization with the physical processing process; An intelligent control and early warning module: receiving the model data fed back by the digital twin modeling module, generating the control instructions of the processing equipment through the virtual-real coupling multi-field collaborative control formula, and generating the fault warning of the key components through the load-related residual life prediction formula; A visual operation and maintenance module: receiving the control instructions and fault warning information output by the intelligent control and early warning module, realizing multi-terminal interactive operation, and completing the optimization and operation and maintenance management of the full-process parameters through the closed-loop optimization weight iteration formula.

2. The NFC juice non-thermal processing system based on intelligent regulation of claim 1, wherein, The online detection device of the integrated physical module comprises a liquid level sensor, a sealing sensor and a metal foreign matter detector, and various sensors correspond to the detection of related indexes after finished product filling respectively; the online detection device is matched with an unqualified product rejection device, and the rejection device responds quickly to ensure the continuity of production.

3. The NFC juice non-thermal processing system based on intelligent regulation of claim 1, wherein, The conveying pipeline of the integrated physical module meets the food contact safety requirements, and the quick-mounting clamp sealing structure is adopted to keep the pipeline connection sterile; the ultra-high pressure-cold plasma collaborative non-thermal processing device comprises an ultra-high pressure cavity and a cold plasma generating unit, the reinforced ultra-high pressure cavity is used to adjust the processing pressure, the cold plasma generating unit uses the dielectric barrier discharge method, the electrode spacing is fixed, and the discharge power can be adjusted.

4. The smartly regulated NFC juice non-thermal processing system according to claim 1, wherein, The material characteristic self-adaptive weighted filtering calculation formula is: ; in, For the first Preprocessed data from each sensor; Weights based on raw material characteristics; For the first The raw data collected by each sensor; The neighboring sensor collaborative weights are such that the sum of all neighboring sensor collaborative weights is 1; This refers to the number of neighboring sensors within the same unit. For the first Raw data collected by the neighborhood sensors; This is the load sensitivity coefficient; For the first The processing pressure corresponding to each sensor; For the first The processing pressure corresponding to each neighboring sensor.

5. The smartly regulated NFC juice non-thermal processing system according to claim 1, wherein, The digital twin modeling module is constructed by using three-dimensional modeling technology combined with a dynamic simulation engine, integrates fluid dynamics models, structural mechanics models and multi-field coupling models, has built-in various fruit juice raw material characteristic databases and equipment parameter databases, uses a time series database to store real-time data, and adopts a mechanism combining real-time incremental updating and periodic full-amount calibration to keep the digital twin and the physical processing process with low synchronization error.

6. The smartly regulated NFC juice non-thermal processing system according to claim 1, wherein, The virtual-real coupling multi-field synergistic regulation calculation formula is: ; wherein, is the value of the first control parameter; is the value of the first control parameter; is the standard parameter weight; is the value of the first control parameter; is the value of the first control parameter; is the virtual-real deviation response coefficient; is the virtual-real deviation; is the synergy coefficient; is the multi-field synergy change rate; is the raw material characteristic correction factor.

7. The smartly regulated NFC juice non-thermal processing system according to claim 1, wherein, The load-related residual life prediction calculation formula is: ; wherein, is the remaining life of the component, is the rated service life of the component, is the wear acceleration factor, is the average processing pressure, is the rated pressure, is the pressure load sensitivity index, is the average vibration amplitude, is the vibration threshold, is the vibration sensitivity index, is the current runtime, is the component used time.

8. The smartly regulated NFC juice non-thermal processing system according to claim 1, wherein, The closed loop optimization weight iterative calculation formula comprises: and ; wherein, is a next time first sensor raw material characteristic weight; is a current time first sensor raw material characteristic weight; is a quality feedback coefficient; is an actual product quality comprehensive value; is a standard quality comprehensive value; is a current time first sensor pre-processed data; is a next time first control parameter standard parameter weight; is a current time first control parameter standard parameter weight; is a life feedback coefficient; is a current time component remaining life; is a component safe remaining life threshold value; is a current time first control parameter value.

9. The smartly regulated NFC juice non-thermal processing system according to claim 1, wherein, The fault warning of the intelligent control and early warning module is triggered based on the calculation result of the load-related residual life prediction formula, divides three warning ranges according to the difference between the residual life of the component and the safety residual life threshold, the residual life above 50% of the safety threshold is the first warning, the residual life between 20% and 50% of the safety threshold is the second warning, and the residual life below 20% of the safety threshold is the third warning; the warning information includes the component name, the installation position and the maintenance suggestion, is synchronously pushed through the multi-terminal of the visual operation and maintenance module, and the corresponding fault component position is highlighted in the digital twin modeling module.

Images