Multi-link quality control production process of algae stream lard stearin product

Abstract

The invention relates to the technical field of food processing, in particular to a multi-link quality control production process of a Jingxi lard stearin product, which comprises the following steps: checking and accepting raw materials, pretreating, frying, burdening, stirring, briquetting and slicing, vacuum packaging, sterilizing, cooling, boxing, warehousing and the like. A multi-light-source colorimeter and an electronic nose system are adopted for fine detection of raw materials, and intelligent control over production links is achieved through intelligent temperature and time double-control frying, internet-of-things ingredient tracing, microcomputer precise ingredient, a molecular imprinting technology and the like; the product quality is improved through a customized hydraulic briquetting, intelligent vacuum packaging and sterilization process intelligent optimization system; a quality tracing two-dimensional code and an artificial intelligence visual detection system are deployed, and full-life-cycle quality tracing and real-time online detection are achieved. The process is integrated with a multi-dimensional innovative technology, the quality of raw materials can be accurately controlled, the production efficiency and the product percent of pass can be improved, the quality guarantee period of products can be prolonged, and the quality control transparency can be enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of food processing technology, in particular to a multi-link quality control production process for algae creek lard residue products. BACKGROUND

[0002] Under the background of the vigorous development of the food industry today, consumers pay more and more attention to the quality, safety and characteristics of food. As a traditional delicacy carrying regional characteristics and profound cultural heritage, algae creek lard residue is facing the urgent needs of quality improvement and production process innovation. The inherent shortcomings of traditional production process in raw material control, production process control, packaging preservation and quality traceability have made it difficult to meet the stringent requirements of modern consumers for high-quality, safe and reliable food.

[0003] Patent No. CN106616429A discloses a lard residue production process, which has significant defects in many key links:

[0004] First, in the raw material acceptance link, only relying on conventional manual visual inspection and basic physical and chemical detection methods has great limitations. For the judgment of the color of the raw material, it is difficult to capture the subtle color change by relying solely on the human eye observation, and some color abnormalities caused by the initial deterioration of the raw material are easily ignored. In terms of odor detection, the human olfactory sensitivity is limited, and manual detection is often helpless for the odor of trace harmful gases that may exist in the raw material. This allows a large number of unqualified raw materials to flow into the subsequent production links, posing a serious risk to product quality and may lead to problems such as deterioration and odor in the final product, greatly affecting product quality and consumer experience.

[0005] Second, in the frying link, the process uses ordinary fryers with fixed frying temperature and time parameters. However, in actual production, different batches of raw materials have obvious differences in physical properties such as thickness and moisture content, and fixed frying parameters cannot be dynamically adjusted according to the changes in these raw material properties. This will cause some raw materials to be severely scorched due to excessive oil temperature during frying, resulting in a large loss of nutrients and a poor taste. Or because the oil temperature is not enough, the oil in the raw material cannot be fully penetrated, resulting in products that are too greasy and cannot be fully cooked, making it difficult to ensure food safety. This "one-size-fits-all" frying method cannot ensure the stability and uniformity of the quality of each batch of products, seriously affecting the market competitiveness of the products.

[0006] III. In the batching step, the prior art uses manual weighing and batching, which is not only inefficient, but also difficult to ensure accuracy. In the process of adding food additives, there is a lack of precise quantitative technology, and only manual experience is used for addition, which cannot strictly meet the requirements of GB 2760. Deviation in the amount of food additives may lead to unstable product taste, which cannot meet the expectations of consumers for product taste consistency, and may even bring food safety risks due to improper use of additives, posing a potential threat to the health of consumers, which does not meet the strict standards of modern food production for batching accuracy and compliance.

[0007] IV. In the packaging step, the process selects a conventional vacuum packaging machine without a residual oxygen detection module, which makes it impossible to accurately control the vacuum degree and residual oxygen content in the package. During product storage, the product is prone to oxidation and deterioration due to the presence of oxygen, greatly shortening the shelf life. At the same time, the ordinary packaging film used has poor barrier properties, making it difficult to effectively block the intrusion of external moisture and microorganisms, further reducing the preservation effect and quality stability of the product, increasing the probability of quality problems in the circulation link.

[0008] In summary, the pig fat residue production process disclosed (announced) in CN106616429A has obvious defects in each key link, and cannot achieve comprehensive and accurate control of product quality. SUMMARY

[0009] The purpose of the present application is to provide a multi-link quality control production process for algae creek pig fat residue products to solve the problems raised in the above background technology.

[0010] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0011] A multi-link quality control production process for algae creek pig fat residue products, specifically comprising the following steps:

[0012] S10 Raw material inspection: strictly inspect the raw materials according to the inspection specifications to ensure that the raw materials are not discolored, deteriorated, odorless, impurity-free, and undamaged;

[0013] S20 Pretreatment: pretreat the raw materials that have passed the inspection;

[0014] S30 Frying: fry the pretreated raw materials, with a frying temperature of 180±2℃ and a frying time of 25±2min, using a fryer;

[0015] S40 Batching: strictly batch according to the formula, with the amount of food additives meeting the requirements of GB2760, using an electronic scale and a workbench for batching;

[0016] S50 mixing: the fried raw materials are mixed with the prepared materials;

[0017] S60 briquetting and slicing: the mixed materials are briquetted, and then the briquetted materials are sliced;

[0018] S70 vacuum packaging: the sliced materials are vacuum packaged;

[0019] S80 sterilization: the vacuum packaged materials are sterilized by using a double-tank sterilization pot, and for algae pig oil residue products, the constant temperature sterilization time is 20±2 min, and the sterilization temperature is 90±2°C;

[0020] S90 cooling, boxing and product storage: the sterilized materials are cooled, boxed and then the boxed products are stored.

[0021] As preferred, in the raw material acceptance link, a special inspection specification is formulated, and the sensory and physicochemical index inspection methods of the raw materials are refined, such as using a multi-light source color difference instrument to accurately detect the color of the raw materials, and using an electronic nose system to determine the odor. Compared with conventional raw material acceptance, more accurate and comprehensive quality control of the raw materials is achieved. The special inspection specification cooperates with the multi-light source color difference instrument and the electronic nose system to realize fine detection of the raw materials. The multi-light source color difference instrument can analyze the color of the raw materials from different spectral angles, avoid misjudgment caused by lighting conditions, and accurately identify the discoloration problem of the raw materials; the electronic nose system can detect subtle odors that human olfactory cannot detect through analysis of characteristic gas components, such as trace harmful gases generated in the early stage of deterioration. Compared with conventional manual visual inspection and simple physicochemical detection, this method improves the raw material quality control accuracy to a molecular level, effectively reduces the subsequent production loss caused by raw material problems, ensures the uniformity and stability of the raw materials for production, and lays a solid foundation for product quality.

[0022] As preferred, in the frying link, the frying machine is configured with an intelligent temperature and time double control system, which is dynamically adjusted in real time. When the oil temperature fluctuation or the thickness and water content difference of the raw materials are detected, the system automatically adjusts the temperature and time, so that each batch of raw materials can be fried under the best conditions. Compared with traditional fixed parameter frying, this technology avoids the problems of burning, nutrient loss caused by excessively high oil temperature, or oil residue, poor taste caused by insufficient oil temperature, etc. The fried pig oil residue has uniform color and crisp taste, and the product qualified rate is improved by more than 20%, while reducing energy waste and frying time, and improving production efficiency.

[0023] As preferred, in the ingredient link, an ingredient traceability system based on Internet of Things is constructed, and each batch of ingredient data is uploaded to the cloud in real time, which can be accurately traced. At the same time, a microcomputer precision ingredient machine is used, and the precision is improved to 0.01 g level, which ensures the accurate execution of the formula, and the food additives are accurately added by using molecular imprinting technology, which strictly conforms to the provisions of GB2760. The ingredient traceability system based on Internet of Things realizes real-time cloud storage and accurate tracing of each batch of ingredient data. Once quality problems occur, the source of raw materials, the ingredient process and the operator can be quickly located, which is convenient for responsibility tracing and problem rectification. The microcomputer precision ingredient machine improves the precision to 0.01 g level, ensuring that the amount of each raw material added is accurate, and avoiding taste deviation caused by manual weighing error. The accurate quantitative addition of food additives by molecular imprinting technology strictly conforms to the provisions of GB2760, ensuring food safety while realizing accurate flavor matching of products to meet the taste needs of different consumer groups.

[0024] As preferred, in the briquetting link, a customized hydraulic briquetting equipment is used, which has precise control functions of pressure and molding time. According to the characteristics of algae pig oil residue materials, the briquetting parameters are optimized to improve the uniformity of briquetting density by more than 30%, which provides guarantee for the stability of subsequent slicing and product form. The customized hydraulic briquetting equipment accurately controls the pressure and molding time parameters according to the characteristics of algae pig oil residue materials, such as easy adhesion and high oil content. Through the optimized briquetting process, the internal density uniformity of the material is improved by more than 30%, effectively solving the problems of loose edges and over-dense centers in the traditional briquetting process. Uniform briquetting density provides a stable basis for the subsequent slicing process, and the slicing process is not prone to cracking, uneven thickness and other phenomena, improving the slicing qualification rate and product form aesthetics, and also helping to maintain the integrity of the product during packaging and transportation, reducing losses.

[0025] As preferred, in the vacuum packaging link, an intelligent vacuum packaging machine is used, which is equipped with a residual oxygen detection module. The vacuum degree can reach below 10 Pa, and the residual oxygen is controlled within 0.5%. Compared with conventional vacuum packaging, the product shelf life is effectively extended, and the packaging film uses degradable and high-barrier composite film, which takes into account environmental protection and quality preservation, solving the problems of poor environmental protection and insufficient barrier of traditional packaging film. The intelligent vacuum packaging machine monitors the residual oxygen content in the package in real time, controls the vacuum degree below 10 Pa and the residual oxygen rate within 0.5%, almost isolates the oxidation effect of oxygen on the product, and inhibits the growth of microorganisms, making the product shelf life extended by more than 1 times compared with conventional vacuum packaging. The use of degradable and high-barrier composite film ensures the barrier performance of the packaging to water vapor and oxygen while solving the environmental pollution problem caused by the difficulty of degrading traditional PE film, which meets the green production trend and improves the product's environmental image and market competitiveness.

[0026] As preferred, in the sterilization section, the double-tank sterilization pot is equipped with an intelligent sterilization process optimization system, which automatically adjusts the sterilization temperature and time curve according to the initial microbial load of algae pig fat residue products, packaging specifications, and other parameters. The intelligent sterilization process optimization system equipped in the double-tank sterilization pot automatically generates the optimal sterilization temperature-time curve through big data analysis and algorithm model according to the initial microbial load of algae pig fat residue products, packaging specifications (such as packaging materials, packaging volume), etc. Compared with fixed parameter sterilization, this system can not only ensure the complete killing of harmful microorganisms, but also avoid the problems of nutrient loss and poor taste caused by excessive sterilization, maximize the retention of flavor substances and nutrients of pig fat residue, and improve product quality and eating value on the premise of ensuring food safety.

[0027] As preferred, in the cooling and boxing section, a segmented gradient cooling device is designed, and the cooling rate is controllable. From the sterilization temperature to the normal temperature, it is divided into rapid cooling and slow uniform temperature stages. Cooling and boxing section-rapid cooling stage: In this stage, the segmented gradient cooling device rapidly reduces the material temperature from the sterilization temperature to a certain temperature (such as about 40°C) at a faster cooling rate, greatly shortens the residence time of the material in high temperature state, effectively inhibits the continuous influence of high temperature on product quality, such as preventing further oxidation of oil, avoiding rapid reproduction of microorganisms at suitable temperature, and reducing the excessive evaporation of product moisture caused by long time high temperature, maintaining the stability of product taste and weight. Cooling and boxing section-slow uniform temperature stage: The slow uniform temperature stage reduces the material temperature to normal temperature at a lower cooling rate, avoiding the stress concentration caused by sudden temperature drop, preventing problems such as oil separation, product cracking and deformation. This stage makes the internal and external temperatures of the material uniform, ensures the stability of product quality, creates good conditions for subsequent boxing and storage, improves the consistency of product appearance and internal quality, and reduces the rate of defective products.

[0028] As preferred, set up a quality traceability two-dimensional code generation system at each key link of raw material acceptance, frying, ingredient mixing, and sterilization, associate the finished product two-dimensional code with the whole chain information of raw material batch, production link parameters, and equipment operation data, consumers can query by scanning the code, realize the whole life cycle quality traceability from farmland to dining table, build a differentiated quality control and consumer trust system, which is different from the conventional food production limited traceability mode, the quality traceability two-dimensional code generation system adopts high-definition industrial cameras (resolution ≥8K), ring shadowless light source, multi-axis motion control platform, the camera is installed above / side of the key workstations of each production equipment to ensure the collection of full-view images of the materials; train a deep learning model (such as based on CNN convolutional neural network architecture) with a large number of qualified / unqualified algae pig oil residue sample images (covering raw material color difference, impurities, damage, burnt after frying, uneven color, size deviation after slicing, missing corners, loose sealing after packaging, film surface damage, etc.), so that the system has the ability to accurately identify defects at each link, the detection algorithm analyzes the collected images in real time, when it identifies unqualified products, it links the production line actuators (such as pneumatic rejection machine, production line pause module) to automatically reject / mark abnormal products; at the same time, upload the detection results (including defect type, occurrence workstation, treatment) to the production database in real time as the core data source of the quality traceability two-dimensional code association information, consumers can view the key link visual detection result summary by scanning the code, realize the quality transparency from production detection to the consumer end, and build an intelligent quality control and trust system that is different from conventional food production.

[0029] As preferred, introduce an artificial intelligence visual detection system in the production process to conduct real-time online detection on the appearance after raw material acceptance, the color and shape after frying, the size uniformity after slicing, and the integrity of finished product packaging, etc., with a detection accuracy of 0.1mm level, which can automatically identify and reject unqualified products, the specific process of the artificial intelligence visual detection system is as follows:

[0030] 1. Hardware deployment:

[0031] At the raw material acceptance workstation, set up the first group of visual detection units at 30-50cm above the conveyor belt, including 2 high-definition industrial cameras (frame rate ≥60fps) at 90° to each other, matched with a diffuse ring light source, covering the full width of the raw material conveying to ensure the collection of images of the surface color, texture, shape, and impurities of the raw materials;

[0032] At the post-frying cooling temporary storage area, set up the second group of detection units, the camera vertically downward collects the overall color and shape images of the materials, cooperates with the linear light source to eliminate the reflection interference of the frying oil film, and identifies defects such as burnt and deformation;

[0033] After the slicing station, a third group of detection units is arranged at an interval of 10-20 cm along the slicing output direction, a linear array camera (resolution >=12K) is used for continuous scanning of the slices to identify size deviation (accuracy +0.1mm), missing corners, continuous knife and other problems;

[0034] At the vacuum packaging station, a fourth group of detection units is arranged at the discharge end of the packaging machine, a surface array camera is used to collect the packaging product image to identify defects such as tight sealing (bubble >=0.5mm 2 ), film surface damage, label misplacement and the like.

[0035] 2. Software and algorithm:

[0036] Based on the Python+TensorFlow framework, a deep learning model is built to pre-process the images collected at each station (including greying, noise reduction and enhancement), and then a trained CNN model (trained by at least 100,000 labeled samples with an identification accuracy of >=99.5%) is used to quickly determine whether the material / product is qualified.

[0037] The algorithm has a built-in defect learning library that can automatically record new defect types (such as rare impurity forms of raw materials and color deviation caused by special processes) that appear during production. After manual review and labeling, the model training set is updated to achieve dynamic iteration of detection capability.

[0038] 3. Production line linkage and data application:

[0039] The system is connected with the PLC control module of the production line. When an unqualified product is identified, the pneumatic rejection device of the corresponding station is triggered (response time <=0.2s) to sort the abnormal product to a special recycling channel.

[0040] The detection data is synchronized in real time to the production MES system to generate quality reports of each link (including defect rate, defect distribution and trend analysis) to provide data support for process parameter optimization (such as frying temperature adjustment and slicing tool maintenance). Compared with manual detection, the efficiency is improved by more than 5 times, the missed detection rate is <=0.1%, and the quality control is upgraded from "after-the-fact sampling" to "in-process full detection" and "intelligent early warning".

[0041] Compared with the prior art, the beneficial effects of the present application are:

[0042] The multi-link quality control production process of the algae creek lard residue product of the present application realizes deep integration and collaborative operation of multi-dimensional innovative technologies, and exhibits significant effects in terms of raw material quality, production efficiency, product quality, environmental performance and consumer trust, etc.

[0043] 1. Accurate control of raw material quality: exclusive inspection specifications combined with multi-light source color difference meter and electronic nose system for comprehensive detection of raw materials from color, odor, and physicochemical indicators. Compared with traditional manual visual inspection, it can accurately identify subtle defects and potential problems, effectively prevent unqualified raw materials from entering the production process, ensure product quality stability from the source, and reduce production risks and cost losses caused by raw material problems.

[0044] 2. Intelligent and efficient production process: intelligent equipment and innovative technology are introduced in each key production link. The intelligent temperature control system in the frying link can dynamically adjust the temperature and time according to the characteristics of the raw materials, ensuring uniform frying and avoiding local burning or undercooking. The microcomputer precision batching machine combined with molecular imprinting technology realizes 0.01g-level high-precision batching, strictly follows GB2760 standard to accurately add food additives, and ensures formula consistency. The intelligent optimization system of double-tank sterilization pot automatically plans the sterilization curve based on product parameters, ensuring sterilization effect while minimizing damage to product nutrition and flavor. These technologies greatly improve production automation level, reduce human error, and improve production efficiency and product pass rate.

[0045] 3. Comprehensive improvement of product quality: Customized hydraulic briquetting equipment optimizes briquetting parameters, making material density uniformity improve by more than 30%, providing support for slice and subsequent product form stability; Intelligent vacuum packaging machine controls vacuum degree below 10Pa and residual oxygen rate within 0.5%, combined with degradable high-barrier composite film, effectively isolating external oxygen, moisture and microorganisms, prolonging product shelf life and improving packaging environmental friendliness; Artificial intelligence vision detection system covers the whole production process, with 0.1mm-level detection accuracy for real-time online monitoring, automatic identification and rejection of unqualified products, with a detection rate as low as 0.1%, which is 5 times more efficient than manual detection, ensuring that product appearance, specifications, and packaging integrity meet high standards, significantly improving product overall quality.

[0046] 4. Transparent and reliable quality control: Whole-chain quality traceability QR code generation system links raw material batches, production parameters, equipment operation data, etc. Consumers can scan the code to obtain the whole life cycle quality data from raw materials to finished products, including key link visual detection result summary, realizing production process transparency. This open sharing of quality information enhances consumer trust in products, helps establish brand image and build differentiated competitive advantage, and also facilitates quality problem tracing and improvement for enterprises, forming a virtuous quality control cycle.

[0047] 5. Flexible and Versatile Process Adaptability: The process can be adapted to small-batch, highly customized production scenarios by customizing raw material acceptance standards and optimizing briquetting parameters, demonstrating excellent adaptability to different production needs. Whether for large-scale standardized production or personalized customized production, this process can ensure stable product quality, meet diverse market demands, and enhance the company's market responsiveness and competitiveness by flexibly adjusting parameters and technology applications at each stage. Detailed Implementation

[0048] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0049] Example 1

[0050] I. Raw Material Acceptance Process

[0051] We have established exclusive inspection standards and adopted a multi-light source colorimeter to accurately detect the color of raw materials. By analyzing from different spectral perspectives, we ensure that the color deviation of raw materials is within a very small range. We also use an electronic nose system to identify odors by recognizing characteristic gas components. We conduct comprehensive sensory and physicochemical tests on raw materials and strictly screen out raw materials that do not change color, do not deteriorate, have no odor, no impurities, and are not damaged.

[0052] II. Pre-processing stage

[0053] For raw materials that have passed the raw material inspection, routine cleaning operations are carried out to remove dust, impurities, etc. attached to the surface. At the same time, appropriate trimming is performed according to the shape of the raw materials to prepare for subsequent processes.

[0054] III. Deep-frying process

[0055] The fryer is equipped with an intelligent temperature and time dual control system. This system can collect data such as the temperature inside the frying chamber and the heating status of the raw materials in real time, and dynamically adjust the frying temperature within the range of 180±2℃ and the time within 25±2min to ensure that the raw materials are heated evenly and the quality of the fried materials is stable.

[0056] IV. Ingredient Preparation

[0057] An IoT-based ingredient traceability system is constructed, with each batch of ingredient data (including raw material type, amount added, and time of addition) uploaded to the cloud in real time for accurate traceability. A microcomputer precision dispensing machine is adopted, improving the dispensing accuracy to the 0.01g level, and strictly following the formula for dispensing operations. Molecular imprinting technology is used for precise quantitative addition of food additives. In accordance with GB2760, the target additive molecules are accurately identified and combined to achieve quantitative addition, ensuring accurate execution of the formula.

[0058] V. Mixing Process

[0059] The fried ingredients and prepared seasonings are put into a conventional mixing equipment. The mixing paddle rotates to fully mix the ingredients and make the materials evenly coated with the seasonings.

[0060] VI. Pressing and Slicing Process

[0061] A customized hydraulic briquetting device is used, which has precise control over pressure and molding time. Based on the material characteristics of Zaoxi pig lard residue, the briquetting pressure, time and other parameters are optimized to improve the uniformity of briquetting density by more than 30%, which ensures the stability of subsequent slicing and product shape. After briquetting, conventional slicing equipment is used to cut the briquetting material into slices of specified thickness and size.

[0062] VII. Vacuum Packaging Process

[0063] The intelligent vacuum packaging machine is equipped with a residual oxygen detection module, which can reduce the vacuum level inside the packaging to below 10Pa and control the residual oxygen to within 0.5%, effectively extending the product's shelf life. The packaging film is made of biodegradable, high-barrier composite film, which ensures product quality and freshness while also being environmentally friendly, solving the problems of poor environmental performance and insufficient barrier properties of traditional packaging films.

[0064] 8. Sterilization process

[0065] The dual-tank sterilizer is equipped with an intelligent optimization system for sterilization process. It first detects parameters such as the initial microbial load and packaging specifications of the Zaoxi pork lard residue product, and then automatically adjusts the sterilization temperature and time curve based on these parameters to sterilize the vacuum-packed material. The constant temperature sterilization time is controlled at 20±2 min and the sterilization temperature is controlled at 90±2℃.

[0066] 9. Cooling, Packing, and Finished Product Warehousing

[0067] A segmented gradient cooling device is designed to cool the sterilized material. The cooling process is divided into a rapid cooling stage and a slow temperature equalization stage. The cooling rate is controllable, so that the material can be steadily cooled from the sterilization temperature to the room temperature. The cooled material is then boxed and the finished product is sent to the warehouse for storage.

[0068] 10. Quality Traceability and Visual Inspection

[0069] A quality traceability two-dimensional code generation system is arranged at each key link such as raw material acceptance, frying, ingredient mixing, sterilization, briquetting, slicing, and vacuum packaging, and the product two-dimensional code is associated with raw material batches, production link parameters (such as frying temperature and time, ingredient data, sterilization curve, etc.), and full-chain information of equipment operation data; at the same time, an artificial intelligence visual inspection system is deployed, a high-definition industrial camera (resolution ≥ 8K), a ring shadowless light source, and a multi-axis motion control platform are adopted, the camera is installed above or on the side of the key workstations of each link production equipment, and full-view images of the materials are collected; a large number of qualified / unqualified sample images of algae pig oil residue (covering raw material color difference, impurities, damage, burnt and uneven color after frying, size deviation and angle defect after slicing, and tightness and film damage after packaging, etc.) are used to train a deep learning model (based on CNN convolutional neural network architecture), the system has the ability to accurately identify defects at each link, and the detection algorithm analyzes the collected images in real time, when unqualified products are identified, the production line executive mechanism (such as a pneumatic rejection machine and a production line pause module) is linked to automatically reject or mark abnormal products; and the detection results (including defect type, occurrence workstation, and treatment) are uploaded to the production database in real time as the core data source of the quality traceability two-dimensional code associated information, and consumers can view the visual inspection result summary of the key links by scanning the code, realizing quality transparency from production detection to the consumer end.

[0070] Example 2: “Intelligent equipment + environmentally friendly packaging”

[0071] I. Raw material acceptance link

[0072] The same as example 1, a special inspection specification is developed, and a multi-light source color difference meter and an electronic nose system are used for raw material inspection.

[0073] II. Pretreatment link

[0074] The raw materials are cleaned and trimmed as usual, and the operation is the same as example 1.

[0075] III. Frying link

[0076] A frying machine with intelligent temperature and time double control system is used to dynamically adjust the frying temperature to 180±2℃ and the time to 25±2min in real time, to ensure the quality of the fried materials, and the operation is the same as the intelligent control part of the frying link in example 1.

[0077] IV. Ingredient mixing link

[0078] An ingredient traceability system based on the Internet of Things is constructed, and each batch of ingredient data is uploaded to the cloud in real time for accurate traceability; a microcomputer precision ingredient mixer is used, with an ingredient accuracy of 0.01g, but without using molecular imprinting technology to quantitatively add food additives, only adding them by conventional accurate weighing, to ensure the basic accurate execution of the formula.

[0079] V. Mixing link

[0080] Conventional mixing equipment is used for stirring to mix the post-frying raw materials and ingredients evenly, as in Example 1.

[0081] Six, briquetting and slicing

[0082] A general briquetting machine is used for briquetting without customized and optimized parameters; a conventional equipment is used for slicing, and the operation is the same as the conventional slicing part in Example 1.

[0083] Seven, vacuum packaging

[0084] An intelligent vacuum packaging machine is used, equipped with a residual oxygen detection module, with a vacuum degree of below 10 Pa and residual oxygen controlled within 0.5%, and a degradable and high-barrier composite film is used for packaging, effectively extending the shelf life of the product while considering environmental protection. The operation is the same as the vacuum packaging part in Example 1.

[0085] Eight, sterilization

[0086] A double-tank sterilization pot is used for sterilization with fixed parameters (constant temperature sterilization time of 20 min, sterilization temperature of 90°C), without the intelligent optimization system of sterilization process, and the sterilization curve is not automatically adjusted according to product parameters.

[0087] Nine, cooling, boxing, and finished product storage

[0088] A segmented gradient cooling device is designed to cool the sterilized materials, divided into rapid cooling and slow uniform heating stages, and then boxed and stored after cooling. The operation is the same as the corresponding part in Example 1.

[0089] Ten, quality traceability and visual inspection

[0090] A two-dimensional code generation system is used for quality traceability, associated with raw material batches and some production process parameters (such as frying and packaging parameters); visual inspection is only deployed at the raw material inspection station and the finished product vacuum packaging station. The raw material inspection station uses a visual inspection unit similar to Example 1 to collect raw material images and identify defects; the finished product vacuum packaging station uses a face array camera to collect packaging product images, identify defects such as poor sealing, film damage, and label misalignment, and remove unqualified products when identified. The detection data is only partially uploaded to the production database, and consumers can view the raw material and finished product visual inspection result summary by scanning the code.

[0091] Example 3: "AI quality detection + full-chain traceability"

[0092] One, raw material inspection

[0093] A special inspection specification is developed, and a multi-light source color difference meter and an electronic nose system are used to inspect the raw materials, with the same operation as in Example 1.

[0094] Two, pretreatment

[0095] Routine cleaning, trimming raw materials, same as example 1.

[0096] III. Frying section

[0097] Using a conventional temperature control fryer, set the fixed frying temperature to 180°C, time 25 min, no dynamic adjustment, only according to the fixed parameters to fry the raw materials.

[0098] IV. Ingredient section

[0099] Same as example 1, build an Internet of Things ingredient traceability system, use a microcomputer precision ingredient machine (precision 0.01 g level), use molecular imprinting technology to accurately add food additives, and ensure accurate execution of the formula.

[0100] V. Mixing section

[0101] Routine mixing equipment stirring and mixing, same as example 1.

[0102] VI. Block, slicing section

[0103] Using a general block press, no customized optimized parameters; using a conventional slicing device, same as the conventional slicing operation in example 1.

[0104] VII. Vacuum packaging section

[0105] Same as example 1, using an intelligent vacuum packaging machine (residual oxygen detection, degradable high-barrier film) for packaging, consistent operation.

[0106] VIII. Sterilization section

[0107] Using a double-tank sterilization pot, setting a fixed constant temperature sterilization time of 20 min and a sterilization temperature of 90°C, without loading a sterilization process intelligent optimization system, sterilizing according to fixed parameters.

[0108] IX. Cooling, boxing, and finished product storage section

[0109] Same as example 1, after segmented gradient cooling, boxing and storage, consistent operation.

[0110] X. Quality traceability and visual inspection section

[0111] At each key link of raw material inspection, frying, ingredient mixing, sterilization, briquetting, slicing, and vacuum packaging, a quality traceability two-dimensional code generation system and an artificial intelligence vision detection system are deployed. The quality traceability two-dimensional code is associated with raw material batches, production parameters throughout the entire process (including fixed frying parameters and fixed sterilization parameters), and equipment operation data; the vision detection system hardware is deployed as in Embodiment 1, images are collected at each station, processed by a deep learning model (trained with at least 100,000 labeled samples, with an identification accuracy of ≥99.5%) based on the Python+TensorFlow framework, to determine whether the materials / products are qualified, and the algorithm has a built-in “defect learning library” for dynamic iteration; when unqualified products are identified, the line execution mechanism is linked to remove them, the detection data is synchronized in real time to the production MES system, quality reports are generated, quality transparency is achieved from production detection to the consumer end, and an intelligent quality control and trust system is established.

[0112] Embodiment 4: Customized process adaptability verification

[0113] I. Raw material inspection link

[0114] According to the characteristics of local raw materials, develop exclusive inspection specifications and strengthen impurity detection processes. Use a multi-light source color difference meter to detect raw material color more carefully, adapt to the color characteristics of local raw materials; optimize the odor recognition model of the electronic nose system to focus on identifying special odors that may occur in local raw materials; and increase detection means for fine impurities in raw materials, such as high-magnification microscopic imaging for auxiliary detection, to ensure that high-quality raw materials are selected.

[0115] II. Pretreatment link

[0116] Conventional cleaning and trimming of raw materials, appropriate adjustment of cleaning intensity and trimming method according to the shape of customized raw materials to ensure that the state of the pretreated raw materials is suitable for subsequent processes.

[0117] III. Frying link

[0118] Use a frying machine with an intelligent temperature and time double control system to dynamically adjust the frying temperature to 180±2℃ and the time to 25±2min, as in Embodiment 1, to ensure the quality of the fried materials.

[0119] IV. Ingredient mixing link

[0120] Use a microcomputer precision ingredient mixer with an accuracy of 0.01g to precisely adjust the amount of each raw material and food additive for small-batch customized formulations, and use molecular imprinting technology to accurately add food additives, strictly in accordance with GB2760, to ensure the precise execution of customized formulations.

[0121] V. Mixing link

[0122] Use conventional mixing equipment to stir the fried raw materials and the customized ingredients to ensure that they are thoroughly mixed, as in Embodiment 1.

[0123] Six, briquetting, slicing link

[0124] Customized hydraulic briquetting equipment is used for small batch production scenarios. Adjust the pressure, molding time and other parameters to make the briquetting meet the shape and density requirements of small batch production. After briquetting, use conventional slicing equipment to slice according to customized size and thickness.

[0125] Seven, vacuum packaging link

[0126] Same as Example 1, use intelligent vacuum packaging machine (residual oxygen detection, degradable high barrier film) for packaging, consistent operation.

[0127] Eight, sterilization link

[0128] The double-tank sterilization pot is equipped with an intelligent sterilization process optimization system. According to the initial microbial load of small batch algae pig oil residue products, packaging specifications and other parameters, the sterilization temperature and time curve are automatically adjusted. The constant temperature sterilization time is controlled within 20±2 min, and the sterilization temperature is controlled within 90±2℃.

[0129] Nine, cooling, boxing and product storage link

[0130] Same as Example 1, segmented gradient cooling, boxing and storage, consistent operation.

[0131] Ten, quality traceability and visual inspection link

[0132] Same as Example 1, set up a quality traceability two-dimensional code generation system and an artificial intelligence visual inspection system at each key link. Correlate the information throughout the chain to realize transparent quality control and ensure that the customized production process and product quality are traceable and controllable.

[0133] Comparative Example 1: Conventional production process (no quality control)

[0134] One, raw material acceptance link

[0135] Use manual visual inspection method to observe the color, shape and odor of raw materials by experience, and manually pick out obvious impurities and damaged raw materials. Only simple physicochemical detection (such as rapid moisture detection) is performed, which cannot accurately control the quality of raw materials and is prone to miss potential problem raw materials.

[0136] Two, pretreatment link

[0137] Conventional cleaning of raw materials only removes obvious surface stains and does not target trimming. The shape of the raw materials is not uniform.

[0138] Three, frying link

[0139] Using a general fryer, set the fixed frying temperature to 180°C, timing 25 minutes, which cannot be dynamically adjusted according to the actual situation of the raw materials, and the raw materials are not evenly heated, which is easy to appear local scorched or not cooked through.

[0140] Four, ingredient section

[0141] Manual use of ordinary platform scale weighing ingredients, low precision, adding food additives by experience, unable to guarantee the precise amount of addition to meet GB2760 regulations, large formula execution error.

[0142] Five, mixing section

[0143] Manual mixing of raw materials and ingredients, uneven mixing, poor material wrapping effect.

[0144] Six, briquetting and slicing section

[0145] Using a general briquetting machine, there is no precise control function, and the briquetting density is uneven; using ordinary slicing equipment, the size and thickness deviation is large.

[0146] Seven, vacuum packaging section

[0147] Using a conventional vacuum machine, there is no residual oxygen detection function, the vacuum degree and residual oxygen content are unstable, using ordinary PE film packaging, poor environmental protection, insufficient barrier property, short product shelf life and easy to be affected by the outside world. Modified.

[0148] Eight, sterilization section

[0149] Using a fixed parameter sterilization pot, set the constant temperature sterilization time to 20 minutes and the sterilization temperature to 90°C, without considering the differences in product initial microbial load, packaging specifications, etc., the sterilization effect is unstable.

[0150] Nine, cooling and boxing, finished product storage section

[0151] Natural cooling, slow and uneven cooling speed, easy to cause material deterioration and shape deformation; manual random boxing, no standard management of finished product storage.

[0152] Ten, quality traceability and visual inspection section

[0153] No quality traceability two-dimensional code generation system, only manual recording of part of the production information, unable to realize full-chain traceability; no artificial intelligence visual inspection system, completely dependent on manual sampling inspection, low sampling ratio (5% of raw materials and finished products), high missed detection rate, and extremely low product control efficiency.

[0154] Comparative Example 2: missing "AI visual inspection + full-chain traceability"

[0155] One, raw material inspection section

[0156] Same as example 1, make exclusive inspection specification, use multi-light source colorimeter and electronic nose system to inspect raw materials, and guarantee the quality of raw materials.

[0157] II. Pretreatment link

[0158] Conventional cleaning and trimming of raw materials, same as example 1.

[0159] III. Frying link

[0160] Same as example 1, use frying machine with intelligent temperature and time double control system to dynamically adjust frying parameters.

[0161] IV. Ingredient link

[0162] Same as example 1, build Internet of Things ingredient traceability system and microcomputer precise ingredient (molecular imprinting additive) to guarantee formula execution.

[0163] V. Mixing link

[0164] Conventional mixing equipment stirring, same as example 1.

[0165] VI. Block pressing and slicing link

[0166] Same as example 1, customized hydraulic block pressing (optimized parameters) + conventional slicing to guarantee material form.

[0167] VII. Vacuum packaging link

[0168] Same as example 1, intelligent vacuum packaging machine (residual oxygen detection + degradable film) to extend shelf life and be environmentally friendly.

[0169] VIII. Sterilization link

[0170] Same as example 1, double-tank sterilization pot with sterilization process intelligent optimization system to automatically adjust sterilization curve.

[0171] IX. Cooling, boxing and product storage link

[0172] Same as example 1, segmented gradient cooling, then boxing and storage to guarantee material quality.

[0173] X. Quality traceability and visual detection link

[0174] The artificial intelligence visual detection system, quality traceability only relies on manual recording of production parameters in each link (such as frying temperature and time, ingredient data, sterilization curve, etc.), and regularly enters the database to generate a simple batch traceability code; 10% of raw materials and finished products are randomly inspected by manual sampling, and unqualified products are identified by manual judgment, with a high rate of missed detection, and the detection data cannot be synchronized in real time with the production system, consumers cannot scan the code to check the production detection information, and the quality transparency from production detection to the consumer end cannot be realized, and the product control efficiency and quality control precision are far lower than the process including AI visual detection + full-chain traceability.

[0175] The key data of the above four groups of examples and two groups of comparative examples are compared in the following table, and the process parameters and product performance indicators are extracted for quantitative analysis as shown in the following table:

[0176]

[0177] According to the quantitative analysis of the above table data

[0178] From the comparison of the key data of the examples and the comparative examples, the algal pig fat residue multi-link quality control production process proposed by the present application has significant advantages in many aspects:

[0179] Raw material inspection: Examples 1-4 use a multi-light source color difference instrument combined with an electronic nose system and a special inspection specification, which can more accurately and comprehensively control the quality of raw materials compared to the manual visual inspection and simple physicochemical detection of Comparative Example 1. The manual detection method of Comparative Example 1 is prone to miss potential problem raw materials, while the detection means of the examples analyzes from multiple dimensions such as spectrum and gas composition, greatly reducing the risk of unqualified raw materials entering subsequent production links, laying a foundation for product quality. Example 4 strengthens impurity detection for local raw materials, further demonstrating the flexibility and adaptability of the process.

[0180] Production equipment and process control: In key production links such as frying, ingredient mixing, briquetting, vacuum packaging, and sterilization, the examples all use advanced equipment or intelligent control systems. For example, the intelligent temperature control fryer used in Examples 1, 2, and 4 dynamically adjusts the temperature, which is more effective than the fixed parameters of the ordinary fryer in Comparative Example 1 in avoiding local overcooking or undercooking of the raw materials and ensuring the stability of the fried product quality; In the ingredient mixing process, Examples 1, 3, and 4 use Internet of Things traceability, microcomputer precise ingredient mixing, and molecular imprinting technology, which far surpass the manual weighing method of Comparative Example 1 in terms of ingredient mixing accuracy and additive compliance; In the vacuum packaging process, the intelligent vacuum machine of Examples 1-4 is equipped with a degradable high-barrier film, which is superior to the conventional vacuum machine and ordinary PE film of Comparative Example 1 in terms of residual oxygen control and environmental performance; In the sterilization process, the intelligent optimization system of Examples 1 and 4 automatically adjusts the sterilization curve according to product parameters, which is more effective than the fixed parameter sterilization in ensuring sterilization effect while reducing damage to product nutrition and flavor.

[0181] Quality control system: Examples 1, 3 and 4 deploy full-link AI visual inspection and full-chain two-dimensional code traceability system, Comparative Example 2 only uses manual sampling inspection and manual record traceability, and Comparative Example 1 completely lacks these two technologies. AI visual inspection controls the missed detection rate of Examples 1, 3 and 4 to be ≤0.1%, while Comparative Examples 1 and 2 rely on manual detection, and the missed detection rates are ≥5% and ≥3% respectively, indicating that the intelligent detection system greatly improves the product control efficiency; the full-chain two-dimensional code traceability makes the consumer code scanning query rate of Examples 1, 3 and 4 reach ≥80%, which reflects the high recognition of consumers to product quality transparency, while the low query rate of Comparative Examples 1 and 2 even without query function, which reflects that the traditional traceability method is difficult to meet the needs of consumers for product quality information.

[0182] Product effect: In terms of shelf life, Examples 1-4 are better than Comparative Example 1 under accelerated experiment, and Examples 1 and 4 even reach ≥90 days, which is due to the synergistic effect of intelligent vacuum packaging, precise sterilization and other multi-link innovative technologies on product preservation; Comparative Example 1 lacks advanced packaging and sterilization technology, and the shelf life is only ≤30 days. Consumer trust data shows that Examples 1, 3 and 4 with full-chain traceability function have higher code scanning query rate, which shows that a complete quality control and transparent traceability system can effectively improve consumer trust in products, and thus promote the market competitiveness of products.

[0183] In summary, the present application achieves a comprehensive breakthrough over traditional production processes in raw material control, production process, quality control and product effect through the synergistic application of multi-link innovative technologies, not only improves product quality and production efficiency, but also enhances consumer trust, and has significant innovation and practicality.

[0184] The multi-link quality control production process of the algae creek lard residue product of the present application realizes the deep integration and synergistic operation of multi-dimensional innovative technologies in raw material quality, production efficiency, product quality, environmental performance and consumer trust, and exhibits significant beneficial effects:

[0185] 1. Accurate control of raw material quality: The exclusive inspection specification combined with multi-light source color difference meter and electronic nose system detects the raw material from color, odor, physical and chemical indicators and other dimensions, which can accurately identify subtle defects and potential problems compared with traditional manual visual inspection, effectively avoiding unqualified raw materials into the production link, ensuring product quality stability from the source, and reducing production risks and cost losses caused by raw material problems.

[0186] 2. Intelligent and efficient production process: Intelligent equipment and innovative technology are introduced in each key production link. The intelligent temperature control system in the frying link can dynamically adjust the temperature and time according to the characteristics of the raw materials, ensuring uniform frying and avoiding local burning or undercooking. The microcomputer precision batching machine combined with molecular imprinting technology realizes high-precision batching of 0.01g, strictly follows the GB2760 standard to accurately add food additives, and ensures the consistency of the formula. The intelligent optimization system of the double-tank sterilization pot automatically plans the sterilization curve based on product parameters, ensuring sterilization effect while minimizing the damage to product nutrition and flavor. These technologies greatly improve production automation level, reduce manual operation errors, and improve production efficiency and product pass rate.

[0187] 3. Overall improvement of product quality: Customized hydraulic briquetting equipment optimizes briquetting parameters, making the uniformity of material density improve by more than 30%, providing stability for slice and subsequent product form; intelligent vacuum packaging machine controls the vacuum degree below 10Pa and the residual oxygen rate within 0.5%, combined with degradable high-barrier composite film, effectively isolating external oxygen, moisture and microorganisms, prolonging product shelf life and improving packaging environmental friendliness; artificial intelligence vision detection system covers the whole production process, with 0.1mm level detection accuracy for real-time online monitoring, automatic identification and rejection of unqualified products, with a detection rate as low as 0.1%, more than 5 times the efficiency of manual detection, ensuring that product appearance, specifications, packaging integrity and other indicators meet high standards, significantly improving product overall quality.

[0188] 4. Transparent and reliable quality control: The whole-chain quality traceability QR code generation system links raw material batches, production parameters, equipment operation data, etc. Consumers can obtain the whole life cycle quality data from raw materials to finished products by scanning the code, including key link visual detection result summary, realizing production process transparency. This open sharing of quality information enhances consumers' trust in products, helps to establish brand image and build differentiated competitive advantage, and also facilitates enterprises to trace and improve quality problems, forming a virtuous quality control cycle.

[0189] 5. Flexible and diverse process adaptation: For small batch and high customization production scenarios, customized raw material acceptance standards and optimized briquetting parameters can be used to demonstrate the good adaptability of the process under different production demands. Whether it is large-scale standardized production or personalized customized production, the process can be adjusted by adjusting the parameters and technology applications of each link to ensure product quality stability, meet diversified market demand, and improve enterprise market response ability and competitiveness.

[0190] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Various changes and improvements can be made to the present application without departing from the spirit and scope of the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A multi-stage quality control production process for Zaoxi pork lard residue products, characterized by: Specifically, the steps include the following: S10 Raw Material Acceptance: Raw materials shall be accepted in strict accordance with inspection specifications to ensure that they do not change color, deteriorate, have no odor, are free of impurities, or are undamaged; S20 Pretreatment: Pretreatment operation is performed on raw materials that have passed the acceptance test; S30 Frying: The pre-treated raw materials are fried at a temperature of 180±2℃ for 25±2 minutes using a fryer. S40 Ingredients: Ingredients are prepared strictly according to the formula, and the amount of food additives used complies with the provisions of GB2760. Electronic scales and workbenches are used for ingredient preparation. S50 Mixing: Mixing the fried ingredients with the prepared seasonings; S60 briquetting and slicing: The mixed material is briquetting, and then the briquetting material is sliced. S70 vacuum packaging: Vacuum packaging of sliced ​​materials; S80 sterilization: A double-tank sterilizer is used to sterilize the vacuum-packed materials. For Zaoxi pork lard residue products, the constant temperature sterilization time is 20±2 min and the sterilization temperature is 90±2℃. S90 Cooling and Packing, Finished Product Warehousing: After sterilization, the materials are cooled and packed into boxes, and the cooled and packed finished products are then stored in the warehouse.

2. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the raw material acceptance process, exclusive inspection standards are formulated, and the inspection methods for sensory and physicochemical indicators of raw materials are refined. For example, a multi-light source colorimeter is used to accurately detect the color of raw materials, and an electronic nose system is used to determine odors. Compared with conventional raw material acceptance, more accurate and comprehensive raw material quality control is achieved.

3. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the frying process, the fryer is equipped with an intelligent temperature and time dual control system that dynamically adjusts the temperature and time in real time.

4. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the ingredient preparation process, an IoT-based ingredient traceability system is constructed, with each batch of ingredient data uploaded to the cloud in real time for accurate backtracking. At the same time, a microcomputer precision ingredient dispensing machine is used, improving the accuracy to the 0.01g level, ensuring accurate execution of the formula. Furthermore, molecular imprinting technology is used to precisely add food additives in quantitative quantities, strictly complying with GB2760 regulations.

5. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the briquetting process, a customized hydraulic briquetting device is used, which has the function of precise control of pressure and molding time. Based on the material characteristics of Zaoxi lard residue, the briquetting parameters are optimized to improve the uniformity of briquetting density by more than 30%, which provides a guarantee for the stability of subsequent slicing and product shape.

6. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the vacuum packaging process, an intelligent vacuum packaging machine is used, equipped with a residual oxygen detection module. The vacuum degree can reach below 10Pa, and the residual oxygen is controlled within 0.5%. Compared with conventional vacuum packaging, it effectively extends the product shelf life. In addition, the packaging film uses a biodegradable, high-barrier composite film, which takes into account both environmental protection and quality preservation, and solves the problems of poor environmental protection and insufficient barrier properties of traditional packaging films.

7. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the sterilization process, the dual-tank sterilizer is equipped with an intelligent optimization system for the sterilization process, which automatically adjusts the sterilization temperature and time curve based on parameters such as the initial microbial load and packaging specifications of the Zaoxi pork lard residue product.

8. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, In the cooling and packing process, a segmented gradient cooling device is designed with a controllable cooling rate, which divides the temperature drop from the sterilization temperature to room temperature into rapid cooling and slow temperature equalization stages.

9. The multi-stage quality control production process for Zaoxi pork lard residue products according to claim 1, characterized in that, A quality traceability QR code generation system is set up in key stages such as raw material acceptance, frying, ingredient preparation, and sterilization. The finished product QR code is linked to the entire chain of information, including raw material batches, parameters of each production stage, and equipment operation data. Consumers can scan the code to check the information, realizing full life cycle quality traceability from farm to table. This builds a differentiated quality control and consumer trust system, which is different from the limited traceability model of conventional food production.

10. The multi-stage quality control production process for Zaoxi pork lard residue products according to any one of claims 1-8, characterized in that, An artificial intelligence visual inspection system is introduced into the production process to conduct real-time online inspections of the appearance of raw materials after acceptance, the color and shape after frying, the uniformity of the size after slicing, and the integrity of the finished product packaging. The inspection accuracy reaches 0.1mm level, and it can automatically identify and reject unqualified products.

Images