Temperature-resistant and uneasy-to-melt freeze-dried chocolate and preparation method thereof

Abstract

The invention relates to the technical field of food processing, in particular to temperature-resistant and uneasy-to-melt freeze-dried chocolate and a making method thereof, the making method comprises a making process of the temperature-resistant and uneasy-to-melt freeze-dried chocolate, and raw materials, procedures, key parameters and storage requirements are covered: the raw materials comprise main raw materials such as cocoa butter and cocoa mass, and auxiliary agents such as a compound emulsifier and a stabilizer; the process comprises the following steps in sequence: raw material pretreatment, basic sauce preparation, three-stage refining, two-stage homogeneous emulsification, intelligent program freezing, intelligent parameter optimization and knowledge base construction, gradient freeze-drying, graded screening and sealed packaging. The key parameters relate to the temperature, pressure, rotating speed, time and the like of each step, the moisture content of the final product is less than or equal to 1.5%, and the final product is hermetically stored in an environment with the temperature of 18-22 DEG C and the humidity of less than or equal to 40%.

Description

Technical Field

[0001] This invention relates to the field of food processing technology, specifically to a freeze-dried chocolate that is heat-resistant and does not easily melt, and its manufacturing method. Background Technology

[0002] Chocolate, a globally popular snack, is favored by consumers for its flavor and taste. However, traditional chocolate is prone to softening, melting, and deformation when stored at room temperature, transported, or in high-temperature environments such as summer. This not only damages the integrity of the product's appearance but also leads to oil migration, flavor loss, and even microbial growth, seriously affecting the eating experience and product safety.

[0003] To address this pain point, existing technologies attempt to improve temperature resistance by adjusting the oil ratio, such as adding high-melting-point hydrogenated oils, adding a single stabilizer, and optimizing the cooling process. However, these methods generally have drawbacks: the introduction of high-melting-point oils can make chocolate taste greasy and hard, losing its original smooth texture. The effect of a single stabilizer is limited, making it difficult to construct a stable internal structure and resulting in insufficient improvement in temperature resistance. Traditional freezing processes lack precise control over ice crystal morphology, resulting in freeze-dried products with loose structure, high porosity, easy moisture absorption and deterioration, and poor temperature resistance.

[0004] Meanwhile, although freeze-drying technology can preserve nutrition and flavor in chocolate production, existing freeze-drying processes have problems such as low parameter matching, uneven ice crystal size, and poor connection between sublimation drying and desorption drying. In addition, parameter adjustment during production relies on manual experience and lacks a systematic optimization system, resulting in poor product stability and low pass rate, making it difficult to meet the needs of large-scale production and diversified consumption scenarios.

[0005] Therefore, developing a freeze-dried chocolate production method that can precisely control the internal structure of materials, significantly improve temperature resistance and moisture resistance, ensure taste and flavor, and is adaptable to intelligent production has become a key issue that the industry urgently needs to solve.

[0006] To address the aforementioned problems, this invention provides a freeze-dried chocolate that is heat-resistant and does not easily melt, as well as a method for producing the same. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a freeze-dried chocolate that is heat-resistant and does not easily melt, as well as its production method. It achieves precise temperature control, particle size control, and ice crystal control in multiple stages, significantly improving the product's temperature resistance and structural stability. At the same time, it uses an intelligent system to ensure production consistency, taking into account both taste and shelf life, and effectively solving problems such as poor temperature resistance and loose structure after freeze-drying in traditional chocolate.

[0008] To achieve the above objectives, one aspect of the present invention is to provide a freeze-dried chocolate that is heat-resistant and does not easily melt, comprising, It is made by the following method: take raw materials in a specific ratio according to weight, and combine emulsifier, stabilizer and thickener in a specific ratio. Cocoa butter and cocoa liquor are heated to 52±1℃ to melt and keep warm. Chocolate powder is ground three times to ≤2μm and the temperature is controlled at 45℃. Starch, compound thickener and hydroxypropyl methylcellulose are mixed in a constant temperature mixer at 35±2℃ at 300r / min for 20min to obtain premixed powder. 80%-85% melted cocoa butter, all cocoa liquor, and pretreated raw materials are fed into a fine grinding mill and ground at 50±2℃ and 800r / min for 5-6 hours until the fineness is ≤1.5μm. After three-stage refining and two-stage homogenization, an emulsion with specific indicators is obtained. The emulsion was injected into a pre-cooled mold, and after a three-stage frozen process, it was transferred to a -50℃ quick-freezing chamber for 3 hours. Parameters were optimized by comparing the three-level verification data and an intelligent knowledge base was built. The material was subjected to gradient sublimation drying and analytical drying until the moisture content was ≤1.5%. After screening qualified products, the material was vacuum-sealed with nitrogen and stored in an environment of 18-22℃ and humidity ≤40%.

[0009] The second aspect of this invention is to provide a method for producing a heat-resistant, non-melting freeze-dried chocolate, comprising: Take raw materials in a specific ratio according to weight, and combine emulsifiers, stabilizers and thickeners in a specific ratio; heat cocoa butter and cocoa liquor to 52±1℃ to melt and keep warm; grind chocolate powder three times to ≤2μm and control the temperature at 45℃; mix starch, compound thickener and hydroxypropyl methylcellulose in a constant temperature mixer at 35±2℃ at 300r / min for 20min to obtain premixed powder; Add 80%-85% melted cocoa butter, all cocoa liquor, pretreated chocolate powder, granulated sugar, and trehalose to a fine grinder, and grind for 5-6 hours at 50±2℃ and 800r / min until the fineness of the material is ≤1.5μm to obtain a homogenized basic sauce. Transfer the base sauce to a refining machine, add the remaining melted cocoa butter, refine at 58±1℃ for 1.5 hours, then cool to 47±1℃ and refine for 7-8 hours. Finally, keep at 47℃ and add various additives in batches, continue refining for 0.8 hours to form a stable oil-polysaccharide complex sauce. Add purified water at 38±2℃ to the compound sauce, stir at 500r / min for 15min, and then homogenize at 28±2MPa for the first stage and 65±3MPa for the second stage to obtain a chocolate emulsion with a particle size of 0.3-1.5μm and a viscosity of 800-1000mPa・s. The emulsion was injected into a mold pre-cooled to -10℃, and then cooled and frozen in three stages to ensure that the center temperature was ≤-48℃ and the ice crystal size was 10-30μm. After freezing, the material was separated and transferred to a -50℃ quick-freezing chamber for 3 hours. By comparing the three levels of verification data, the parameters are adjusted accordingly, the verification process is repeatedly optimized until the standard is met, the core data and models are stored, and an intelligent knowledge base mapping tasks, attributes, scenarios and solutions is built. The material is fed into a freeze dryer and first sublimated for 13-14 hours at a vacuum of 15±2Pa and a temperature gradient of 30-60℃. Then, it is desorbed for 11-12 hours at a vacuum of 110±5Pa and a temperature gradient of 32±1℃ to achieve a moisture content of ≤1.5%. Qualified products are selected, vacuumed to ≤5Pa using aluminum foil vacuum bags, sealed with nitrogen, placed in silica gel desiccant packs, and stored in an environment of 18-22℃ and ≤40% humidity.

[0010] Compared with the prior art, the present invention provides a freeze-dried chocolate that is heat-resistant and does not easily melt, and a method for making the same, which has the following beneficial effects: 1. This solution is based on raw materials with a specific ratio. It refines chocolate powder through three grinding processes, prepares premixed powder through constant temperature mixing, and then finely grinds it to obtain an ultra-fine homogeneous base sauce. A stable oil-polysaccharide complex system is constructed through three-stage refining. A narrow particle size distribution emulsion is obtained through two-stage homogenization. The ice crystal morphology is precisely controlled through three-stage programmed freezing. The core parameters are optimized by combining key data collection and verification, and an intelligent knowledge base is constructed. The moisture content is strictly controlled through gradient sublimation drying and analytical drying. Finally, the production is completed through grading and screening and "vacuum + nitrogen filling" sealed packaging. 2. This solution, through the synergistic effect of precise temperature control, particle size control, and ice crystal control in multiple stages, significantly improves the product's temperature resistance and structural stability. The intelligent system ensures production consistency while taking into account the product's taste, flavor, and shelf life, effectively solving many pain points of traditional processes.

[0011] The solution significantly improves the product's temperature resistance and structural stability through precise temperature control, particle size control, and ice crystal control at multiple stages. At the same time, it ensures production consistency through an intelligent system, taking into account both taste and shelf life. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating the entire process of making heat-resistant freeze-dried chocolate; Figure 2 Flowchart of refined pretreatment process for raw materials; Figure 3 Flowchart of intelligent program freeze forming process; Figure 4 Flowchart for precise gradient freeze-drying process; Figure 5 This is a flowchart of the grading, screening, and sealing packaging process. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0015] Traditional chocolate suffers from problems due to the melting point characteristics of cocoa butter, such as poor temperature resistance, easy softening and melting, loose structure after freeze-drying, easy moisture absorption and deterioration, and insufficient stability due to reliance on experience for production parameters.

[0016] This proposal presents a heat-resistant, non-melting freeze-dried chocolate and its production method.

[0017] This solution is based on a specific ratio of raw materials. It involves refining chocolate powder through three grinding processes, preparing premixed powder through constant temperature mixing, finely grinding to obtain an ultra-fine homogeneous base sauce, constructing a stable oil-polysaccharide complex system through three-stage refining, obtaining a narrow particle size distribution emulsion through two-stage homogenization, precisely controlling the ice crystal morphology through three-stage programmed freezing, optimizing core parameters by combining key data collection and verification, and building an intelligent knowledge base. Then, it involves gradient sublimation drying and analytical drying to strictly control the moisture content. Finally, it completes the production process through grading and screening and "vacuum + nitrogen filling" sealed packaging. This solution, through the synergistic effect of precise temperature control, particle size control, and ice crystal control in multiple stages, significantly improves the product's temperature resistance and structural stability. The intelligent system ensures production consistency while also taking into account the product's taste, flavor, and shelf life, effectively solving many pain points of traditional processes.

[0018] The solution significantly improves the product's temperature resistance and structural stability through precise temperature control, particle size control, and ice crystal control at multiple stages. At the same time, it ensures production consistency through an intelligent system, taking into account both taste and shelf life.

[0019] Example 1, as Figures 1-5 The following is an illustrative description of a method for producing a heat-resistant, non-melting freeze-dried chocolate according to an embodiment of this application.

[0020] S100: Take raw materials in a specific ratio according to weight, and combine emulsifiers, stabilizers and thickeners in the corresponding proportions; heat cocoa butter and cocoa liquor to 52±1℃ to melt and keep warm, and grind chocolate powder three times to ≤2μm and control the temperature at 45℃; mix starch, compound thickener and hydroxypropyl methylcellulose in a constant temperature mixer at 35±2℃ at 300r / min for 20min to obtain a uniform premixed powder; S200: Input 80%-85% melted cocoa butter, all cocoa liquor, pre-treated chocolate powder, white sugar, and trehalose into the fine grinder, grind at 50±2℃ and 800r / min for 5-6 hours, monitor the particle size in real time, until the fineness of the material is ≤1.5μm, and obtain homogenized basic sauce. S300: Transfer the base sauce to the refining machine, add the remaining melted cocoa butter, refine at 58±1℃ for 1.5h, then cool to 47±1℃ and refine for 7-8h, finally keep at 47℃ and add various additives in batches, continue refining for 0.8h to form a stable oil-polysaccharide complex sauce. S400: Add purified water at 38±2℃ to the compound sauce, stir at 500r / min for 15min, and then homogenize at 28±2MPa for the first stage and 65±3MPa for the second stage to obtain a chocolate emulsion with a particle size of 0.3-1.5μm and a viscosity of 800-1000mPa・s. S500: Inject the emulsion into a mold pre-cooled to -10℃, cool and freeze it according to a three-stage procedure to ensure that the core temperature is ≤-48℃ and the ice crystal size is 10-30μm. After freezing, separate the material and transfer it to a -50℃ quick-freezing chamber for 3 hours to lock in the ice crystal morphology. S600: Collect key attribute data, compare it with three-level verification data, adjust core parameters in a targeted manner, and repeatedly optimize and verify until the product meets the standards. The qualified data is associated with and stored with raw material ratio, product attributes, and application scenarios to build an intelligent knowledge base that maps "task-attribute-scenario-solution". S700: The material is fed into the freeze dryer and first sublimated at a vacuum of 15±2Pa and a gradient temperature of 30-60℃ for 13-14 hours. Then, it is adjusted to a vacuum of 110±5Pa and desorption at 32±1℃ for 11-12 hours to make the moisture content of the material ≤1.5%. S800: Select products with complete shape and qualified indicators, vacuum them to ≤5Pa using aluminum foil vacuum bags, seal them with nitrogen, place them in silica gel desiccant packs, and store them in a cool, dry environment at 18-22℃ and ≤40% humidity.

[0021] Step S100 is used for the fine pretreatment of raw materials, including: S110: Construction of Raw Material Proportioning System and Precise Weighing: S1101: The ingredient ratio is the core foundation for determining the temperature resistance, taste, and structural stability of freeze-dried chocolate. It needs to be based on the principles of fat-polysaccharide complex network construction, the stability rules of emulsification systems, and the adaptability of the freeze-drying process to determine the weight range and ratio logic of each component. First, clarify the functional positioning of the core ingredients: cocoa butter, as the main fat component, needs to balance temperature resistance and taste, with a range of 19%-21%. Below 19% will result in an incomplete fat network, while above 21% will make the product too oily; cocoa liquor provides the core flavor and color, and a ratio of 17%-19% ensures a rich flavor without affecting system stability; low-fat chocolate powder, with a total fat content of 18%-22% and an initial particle size ≤25μm, should account for 22%-24%. Its low fat content reduces the risk of melting, while its high content enhances the concentration of chocolate flavor. S1102: The formulation of functional additives needs to be optimized through synergistic effects: The compound emulsifier consists of glyceryl monostearate and polyglycerol ricinoleate in a 1:1 ratio, with an addition amount of 0.9%-1.1%. The synergistic effect of these two additives reduces the oil-water interfacial tension. Glyceryl monostearate focuses on improving oil dispersibility, while polyglycerol ricinoleate enhances the thermal stability of the emulsion system. The synergistic effect is optimal at a 1:1 ratio. Experiments have verified that at this ratio, the emulsion shows no stratification after standing for 2 hours at 60℃. The compound stabilizer consists of xanthan gum and guar gum in a 2:3 ratio, with an addition amount of 0.4%-0. Xanthan gum, at 0.5%, possesses excellent shear resistance and thermal stability, while guar gum exhibits good water solubility and significant thickening effect. A 2:3 ratio can form a dense three-dimensional network structure, effectively locking in oils and moisture and inhibiting migration. The compound thickener consists of sodium carboxymethyl starch and pectin in a 3:2 ratio, with an addition amount of 1.5%-2.0%. Sodium carboxymethyl starch has strong acid resistance and high expansion rate, while pectin has good gelling properties. The combination of the two can increase the viscosity of the material and construct an anti-melting network. Calculations show that the viscosity of the material can be stabilized at 800-1000 mPa·s under this ratio, meeting the requirements of subsequent processing. S1103: Hydroxypropyl methylcellulose, as a heat-resistant modifier, is added at a rate of 0.3%-0.5%. The hydroxyl and carboxyl groups on its molecular chain can form hydrogen bonds with oils and polysaccharides, enhancing the cross-linking degree of the system and improving its temperature resistance. The other components are: starch 4%-5%, food-grade purified water 24%-27%, moisture retention agent, glycerin 0.6%-0.8%, acidity regulator, citric acid 0.2%-0.3%, and natural cocoa flavor 0.1%-0.2%. The proportions of each component were determined through orthogonal experiments. The optimal ratio range was obtained by range analysis, using the product's temperature resistance, taste, and stability as evaluation indicators. S1104: The weighing process must use an electronic analytical balance with an accuracy of 0.01g. The balance must be calibrated before weighing. The calibration formula is: Calibration error = (mass of standard weights - mass displayed on the balance) / mass of standard weights × 100%, and the calibration error must be ≤ ±0.02%. Weighing should be carried out in the order of "solid raw materials first, liquid raw materials later" and "main raw materials first, additives later". Record the data immediately after weighing each type of raw material to ensure the accuracy of feeding. After weighing, the raw materials should be classified and stored, and labeled with their name, weight and weighing date to avoid confusion.

[0022] S120: Cocoa butter melting and constant temperature insulation treatment: S1201: The quality of cocoa butter melting directly affects the subsequent emulsification effect and the product's temperature resistance, requiring a "gradient heating + constant temperature holding" process. First, select a stainless steel constant temperature heating tank with a temperature control system. The volume is determined based on the production scale, generally 1.5 times the amount of cocoa butter added, ensuring sufficient melting space for the cocoa butter. Before use, the heating tank must be cleaned and sterilized by circulating 85℃ hot water for 30 minutes, followed by drying with sterile air to avoid residual impurities affecting product quality. S1202: Add the weighed cocoa butter to the heating tank, start the heating system, and set the initial heating rate to 5℃ / min, raising the temperature from room temperature to 45℃. This stage is the softening stage of the cocoa butter; slow heating can prevent local overheating and oxidation. Then adjust the heating rate to 2℃ / min, raising the temperature from 45℃ to 52±1℃. This temperature is the optimal melting temperature for cocoa butter. Temperatures higher than this will cause oxidation of unsaturated fatty acids in the cocoa butter, while temperatures lower will result in incomplete melting and the presence of solid particles. During the melting process, turn on the stirring device at a speed of 150 r / min to ensure even heating of the cocoa butter. The stirring paddle uses an anchor-type structure with a gap of ≤5mm between it and the tank wall to prevent unmelted cocoa butter from accumulating at the bottom. S1203: After the cocoa butter has completely melted, and no visible solid particles are observed, and the particle size is ≤0.1μm as measured by a laser particle size analyzer, it enters the constant temperature holding stage. The holding temperature is set at 50℃, and the holding time is determined according to the progress of subsequent processes, generally 1-2 hours, and no more than 4 hours, to avoid rancidity caused by prolonged holding. During the holding process, the temperature is checked every 30 minutes, and the temperature fluctuation range is controlled within ±0.5℃. If the temperature is below 49.5℃, the heating system is activated to compensate for the temperature at a rate of 1℃ / min; if the temperature is above 50.5℃, the cooling jacket is activated to cool down the temperature at a rate of 0.8℃ / min. Simultaneously, continuous stirring is required during the holding stage, with the speed adjusted to 100 rpm, to prevent the cocoa butter from separating or re-solidifying. S1204: To monitor the melting quality of cocoa butter, regular sampling and testing are required. The test indicators include acid value, peroxide value, and melting state. Acid value testing follows the method in GB5009.229-2016 "National Food Safety Standard - Determination of Acid Value in Food," requiring an acid value ≤ 1.0 mg KOH / g. Peroxide value testing follows the method in GB5009.227-2016 "National Food Safety Standard - Determination of Peroxide Value in Food," requiring a peroxide value ≤ 5.0 mmol / kg, ensuring that the cocoa butter has not undergone oxidative deterioration.

[0023] S130: Melting and constant temperature insulation of cocoa liquor: S1301: Cocoa liquor contains cocoa butter, theobromine, caffeine, and flavor compounds. Its melting process must balance thorough melting with flavor preservation. A constant-temperature heating tank of the same specifications as that used for cocoa butter melting is selected, and the cleaning and sterilization process is identical. The weighed cocoa liquor is placed into the heating tank. Since cocoa liquor is relatively hard, it needs to be broken into small pieces with a diameter ≤5cm for faster melting. A food-grade stainless steel crusher is used at a speed of 500 rpm. The crushing time is adjusted according to the initial size of the cocoa liquor, generally 3-5 minutes. After crushing, the liquor is sieved through a 10-mesh screen to remove any possible impurities. S1302: Start the heating system, using a "segmented heating" mode: The first stage raises the temperature from room temperature to 40°C at a rate of 3°C / min and holds for 30 minutes to soften the surface of the cocoa liquor, preventing scorching from direct high temperatures; the second stage raises the temperature from 40°C to 52±1°C at a rate of 2°C / min and holds for 1 hour until the cocoa liquor is completely melted. During melting, the stirring speed is 200 rpm, and the stirring paddle uses a propeller-type structure to promote material circulation, ensuring uniform temperature within the tank and preventing localized overheating areas. S1303: After melting, maintain a constant temperature of 50℃. Temperature control during this process is consistent with that of cocoa butter, with temperature fluctuations ≤ ±0.5℃. Because cocoa liquor contains small amounts of protein and sugar, prolonged heat treatment can easily lead to Maillard reactions, causing the product to darken in color and alter its flavor. Therefore, the maximum heat treatment time is 3 hours. Samples are taken every 40 minutes during heat treatment. In addition to testing acid value and peroxide value (consistent with cocoa butter), color is also tested using a colorimeter. The L value (brightness) is controlled at 25-30, the a value (red-green hue) at 10-15, and the b* value (yellow-blue hue) at 20-25 to ensure that flavor compounds are not damaged.

[0024] S140: Three-stage refining process for chocolate powder: S1401: The particle size of chocolate powder is a key factor affecting the product's taste and heat resistance. It needs to be ground in three stages: "primary grinding - fine grinding - ultrafine grinding" to refine the particle size from the initial ≤25μm to ≤2μm, ensuring uniform particle distribution. The three-stage grinding equipment used are a colloid mill, a ball mill, and an ultrafine pulverizer, respectively. Before use, the equipment must be cleaned and disinfected. The inner walls of the colloid mill and ball mill are wiped with 75% edible alcohol, and the ultrafine pulverizer is sterilized with high-temperature steam at 121℃ for 30 minutes. After sterilization, it is ventilated and dried to ensure no residual moisture or impurities remain. S1402: Initial Grinding Stage: A JM series colloid mill with a 15kW motor and a grinding gap of 0.1mm was used. Chocolate powder with an initial particle size ≤25μm was fed into the colloid mill at a feed rate of 50kg / h. During grinding, 30℃ cooling water was circulated through the cooling jacket to maintain an ambient temperature ≤45℃, preventing the melting of fats in the chocolate powder and subsequent particle agglomeration. The target particle size after initial grinding was 10μm. The grinding efficiency was calculated using the formula: η1=(d0-d1) / d0×100%, where d0 is the initial particle size (25μm) and d1 is the particle size after initial grinding (10μm). η1 was calculated to be 60%, requiring an initial grinding efficiency ≥55%. After initial grinding, the particle size was measured using a Malvern Mastersizer 3000 laser particle size analyzer. Three samples were tested per batch, and the average value was taken. If the particle size was greater than 10μm, the sample was returned to the colloid mill for re-grinding until the target was met. S1403: Fine Grinding Stage: The initially ground chocolate powder is fed into a QM series planetary ball mill. The grinding jar is made of zirconium oxide to avoid contamination. The ball-to-powder ratio is 8:1, the weight ratio of grinding balls to chocolate powder is 5mm, the grinding ball speed is 300r / min, and the grinding time is 2 hours. During the grinding process, the ambient temperature is controlled to ≤45℃, achieved through intermittent cooling, cooling for 10 minutes every 30 minutes. The target particle size after fine grinding is 5μm, and the grinding efficiency η2=(d1-d2) / d1×100%=(10-5) / 10×100%=50%, requiring a fine grinding efficiency ≥45%. After fine grinding, the particle size is checked again using a laser particle size analyzer. If the particle size does not meet the standard, the grinding time is extended by 30 minutes until the particle size is ≤5μm. S1404: Ultrafine Grinding Stage: A WFJ series ultrafine pulverizer with a motor power of 37kW is used. The grinding chamber pressure is controlled at 0.6MPa, and the feeding rate is 30kg / h. Dry nitrogen (dew point ≤ -40℃) is introduced during the grinding process to cool the grinding chamber and control the temperature to ≤45℃, and to prevent the chocolate powder from absorbing moisture and agglomerating. The target particle size after ultrafine grinding is ≤2μm, and the particle size distribution span (d90 / d10) is ≤2.5, where d90 is the particle size of 90% of the particles and d10 is the particle size of 10% of the particles. The grinding efficiency η3 = (d2-d3) / d2×100% = (5-2) / 5×100% = 60%, and the required ultrafine grinding efficiency is ≥55%. After ultrafine grinding, the particles are inspected by a laser particle size analyzer and the particle morphology is observed by a scanning electron microscope (SEM) to ensure that the particles are irregular polygons without obvious agglomerates. If agglomeration is present, 0.1% food-grade silica and anti-caking agent are added, and ultrafine grinding is repeated. S1405: After the three-stage grinding is completed, the chocolate powder is stored in a sealed sterile silo. Nitrogen gas with a purity of ≥99.9% is introduced into the silo to prevent moisture absorption and oxidation. The temperature of the silo is controlled at 20-25℃, the relative humidity is ≤40%, and the storage time does not exceed 24 hours to ensure the fluidity and dispersibility during subsequent processing.

[0025] S150: Premixed Powder Preparation and Uniformity Optimization: S1501: The premixed powder consists of starch, compound thickener, and hydroxypropyl methylcellulose. Precise mixing is required to ensure uniform dispersion of all components and avoid localized concentrations that are too high or too low, which could affect product performance. First, a V-type mixer with a 500L capacity and made of 304 stainless steel is selected. Before use, the mixer is cleaned with sterile water and then dried with sterile air. The inner wall of the mixer is smooth and free of dead corners to ensure no material residue remains during the mixing process. S1502: Add 4%-5% of the weighed starch, 1.5%-2.0% of the compound thickener, and 0.3%-0.5% of hydroxypropyl methylcellulose to the mixer in a "large to small" order. That is, first add starch (60%-70% of the total weight of the premixed powder), then add 20%-25% of the compound thickener, and finally add 5%-10% of the hydroxypropyl methylcellulose. This order of addition prevents small amounts of raw materials from adhering to the inner wall of the mixer, improving mixing uniformity. After adding the ingredients, close the mixer inlet, start the mixer, set the mixing speed to 300 r / min, and the mixing time to 20 min. The mixing coefficient K=1.2 is determined through equipment calibration. The formula for calculating mixing uniformity is: CV=(S / )×100%, where S is the standard deviation of the material composition content at each sampling point, and K is the average value of the material composition content at each sampling point. CV should be ≤3%. S1503: During the mixing process, the temperature inside the mixer is controlled at 35±2℃, achieved by introducing 35℃ constant-temperature hot water through the jacket. Excessive temperature will cause starch gelatinization, while insufficient temperature will reduce mixing efficiency. After 10 minutes of mixing, intermediate sampling is performed at three locations: the top, middle, and bottom of the mixer, with 5g sample taken from each location. The hydroxypropyl methylcellulose content is determined using high-performance liquid chromatography (HPLC) to calculate the cross-sectional area (CV) of the mixing homogeneity. If the CV > 3%, the mixing time is extended by 5 minutes, and sampling and testing are repeated until the CV ≤ 3%. S1504: After mixing, the premixed powder is sieved through a 100-mesh sieve to remove any lumps and impurities. The sieve pass rate is ≥99%. The sieved premixed powder is stored in a sterile silo of the same specifications as the chocolate powder, under the same storage conditions, and the storage time does not exceed 24 hours. Before use, the mixing uniformity is checked again to ensure that it meets the requirements before proceeding to the next process.

[0026] The S200 is used for the preparation of ultrafine homogenized basic sauces, specifically: S210: Optimization of raw material feeding sequence and precise feeding: S2101: The order in which raw materials are added directly affects the homogenization of the base sauce. A scientific order of addition must be designed based on the physicochemical properties of each raw material to avoid clumping and stratification caused by interactions between the raw materials. First, the raw materials are classified as follows: fats and oils (melted cocoa butter, cocoa liquor), powders (chocolate powder, granulated sugar, trehalose). Among them, fats and oils are the continuous phase, while powders and powders are the dispersed phase. The order of addition follows the principle of "continuous phase added first, dispersed phase added later" and "easily soluble raw materials added first, difficult-to-dissolve raw materials added later." S2102: Step 1: Add 80%-85% of the melted cocoa butter from step S100 to a chocolate grinder (model: CM-1000, 1000L capacity, made of 304 stainless steel, equipped with a temperature and speed control system and an online particle size detection device). Start the grinder's stirring device to allow the cocoa butter to form a uniform oil layer inside the machine, serving as a dispersion medium for subsequent raw materials. The amount of material added is controlled by the grinder's weighing system, with an error of ≤±0.5%. S2103: The second step is to add the completely melted cocoa liquor. Cocoa liquor and cocoa butter are both types of fats and have good compatibility. After adding, continue stirring for 10 minutes to ensure that the two are fully mixed and form a uniform mixed fat phase. The mixing temperature is maintained at 50±2℃, which is achieved through the jacket temperature control system of the fine grinding mill. S2104: The third step is to add the pretreated chocolate powder. The chocolate powder is a powdery dispersion phase. The addition rate is controlled at 20 kg / min. Stir while adding to avoid agglomeration caused by a large amount of chocolate powder being added at once. After adding, stir for 20 minutes to allow the chocolate powder to be initially dispersed in the mixed fat phase. S2105: Step 4, add granulated sugar and trehalose. Granulated sugar has high solubility (203.9g / 100g water at 20℃), while trehalose has slightly lower solubility (68.9g / 100g water at 20℃). Both are sweeteners, and trehalose can also improve the stability of the system. The addition rate is 15kg / min. After adding, stir for 15min to ensure initial dissolution and dispersion. S2106: During the feeding process, the temperature and pressure inside the grinding mill must be monitored in real time. The temperature should be controlled at 50±2℃, and the pressure ≤0.1MPa. If the temperature exceeds 52℃, the cooling system should be activated to lower the temperature; if the pressure exceeds 0.1MPa, the exhaust valve should be opened to release pressure to prevent the raw materials from deteriorating due to excessive temperature or pressure. After feeding is completed, the feed inlet should be closed and the grinding mill sealed to ensure that no air enters during subsequent grinding. At the same time, the feeding amount, feeding time, and internal temperature and pressure data of each raw material should be recorded for traceability.

[0027] S220: Setting and Calibrating Core Parameters of the Fine Grinding Mill S2201: The temperature and rotation speed of the grinding mill are core parameters affecting the grinding effect. They need to be set according to the characteristics of the raw materials and the target particle size ≤1.5μm, and precise calibration is required to ensure parameter stability. First, set the internal temperature to 50±2℃. This temperature ensures that the oily raw materials are in a molten state (cocoa butter melting point 31-35℃) while avoiding excessive temperature that could lead to oil oxidation and flavor loss. The temperature control system uses PID (Proportional-Integral-Derivative) regulation with a temperature control accuracy of ±0.5℃. The calibration method is as follows: Place three platinum resistance temperature sensors at different locations inside the grinding mill, start the temperature control system, and after stabilization, record the temperature values ​​of each sensor. Calculate the deviation between the average value and the set value. A deviation ≤±0.5℃ is considered acceptable; otherwise, adjust the PID parameters: proportional coefficient P=20, integral time I=60s, derivative time D=10s, until calibration is successful. S2202: The grinding speed is set to 800 r / min. This speed was determined experimentally: if the speed is too low (≤600 r / min), the grinding shear force is insufficient, and the particle size is difficult to reach ≤1.5 μm; if the speed is too high (≥1000 r / min), a large amount of heat will be generated, causing the internal temperature to exceed the set range, and increasing energy consumption. Speed ​​calibration uses a laser tachometer. The calibration formula is: Speed ​​error = (Actual speed - Set speed) / Set speed × 100%, requiring a speed error ≤±1%. During calibration, the fine grinding machine is started, and after stabilization, the motor speed is detected using a laser tachometer. If the error exceeds ±1%, the frequency converter frequency is adjusted until the actual speed is within the range of 792-808 r / min. S2203: In addition, the grinding time needs to be set to 5-6 hours. The formula for calculating the grinding time is: t = (d_target / d_initial) × k, where d_target = 1.5 μm, d_initial is the initial particle size of the mixture after feeding (approximately 20 μm), and k is a correction coefficient determined by the equipment characteristics, k = 1.2. The calculated t = (1.5 / 20) × 1.2 × 60 ≈ 5.4 hours. Therefore, the grinding time is set to 5-6 hours to ensure that the particle size meets the standard. Simultaneously, the online particle size detection frequency is set to once every hour, and the detection data is transmitted to the control system in real time for timely parameter adjustments.

[0028] S230: Real-time monitoring and dynamic adjustment of the grinding process: S2301: The grinding process requires real-time monitoring of key parameters such as particle size, temperature, and pressure. Adjustments are made dynamically based on the test results to ensure the homogenization of the base sauce. First, particle size monitoring is performed using an online laser particle size analyzer installed on the circulation pipe of the fine grinder. Samples are automatically taken and tested every hour, with a detection range of 0.1-100μm and an accuracy of ±1%. The test data is displayed in real-time on the control system interface. If the detected particle size is greater than 1.5μm, the cause is analyzed and adjustments are made: if insufficient rotation speed is the cause, the speed is increased by 50 r / min; if low temperature increases oil viscosity, the temperature is increased by 1-2℃; if raw material agglomeration is the cause, the grinding time is extended by 30 minutes. Particle size is tested again 30 minutes after each adjustment until the particle size is ≤1.5μm. S2302: Temperature monitoring uses a built-in platinum resistance temperature sensor, recording temperature data every 10 minutes. If the temperature is below 48℃, the heating system is activated at a heating rate of 1℃ / min; if the temperature is above 52℃, the cooling system is activated at a cooling rate of 0.8℃ / min, ensuring the temperature stabilizes at 50±2℃. Pressure monitoring uses a pressure transmitter with a range of 0-0.3MPa and an accuracy of ±0.01MPa, recording pressure data every 15 minutes. The normal pressure range is 0.05-0.1MPa. If the pressure is below 0.05MPa, it may be due to poor sealing; check the sealing of the inlet and exhaust valves. If the pressure is above 0.1MPa, it may be due to increased material viscosity during grinding; appropriately increase the temperature by 1-2℃ or reduce the rotation speed by 50r / min. After adjustment, check the pressure again after 10 minutes until it returns to normal. S2303: During the grinding process, the flowability and color of the material should also be observed. Observed through the sight glass of the fine grinder, under normal circumstances, the material should be a uniform brown color, with good flowability, and no obvious lumps or stratification. If lumps appear, it may be that the chocolate powder is not sufficiently dispersed. Grinding should be paused, high-speed stirring should be started for 5 minutes, and then grinding should continue. If stratification occurs, it may be that the oil and powder raw materials are not compatible. 0.1% of compound emulsifier should be added, the same as in S100, and after stirring evenly, grinding should continue.

[0029] S240: Base sauce quality inspection and pass / fail determination: S2401: After grinding is completed, conduct a comprehensive quality inspection on the base sauce to ensure compliance with the requirements of subsequent refining processes. The inspection indicators include particle size, viscosity, color, acid value, and peroxide value; S2402: Particle size inspection: Use an offline laser particle size analyzer to take 3 samples of 5 g each from different positions inside the fine grinder, dilute with absolute ethanol and then inspect. The requirements are that the average particle size ≤ 1.5 μm, d90 ≤ २.० μm, d10 ≥ ०.५ μm, and the particle size distribution span, d90 / d10 ≤ 3.0, to ensure uniform particle distribution; S2403: Viscosity inspection: Use a rotational viscometer. The inspection temperature is 50 °C, the rotor speed is 60 r / min, the sampling volume is 50 mL, and conduct 3 inspections and take the average value. The requirement is that the viscosity is 600 - 800 mPa・s. Too high viscosity will increase the difficulty of subsequent refining and homogenization, while too low viscosity will result in insufficient system stability. The viscosity calculation formula is: η = K × τ, where K is the rotor constant, determined by the equipment, K = 100, and τ is the torque value, directly read through the viscometer; S2404: Color inspection: Use a color difference meter. The light source is D65, the observation angle is 10°, the sampling volume is 10 g, spread it flat in a colorimetric dish, and inspect the L*, a*, and b values. The requirements are that L = 28 - 32, a* = 12 - 16, b* = 22 - 26, with uniform color and no obvious color difference; S2405: Acid value inspection: Refer to the method of GB5009.229 - 2016, take a sample of <10> g, dissolve it with neutral ethanol, use phenolphthalein as an indicator, and titrate with 0.1 mol / L potassium hydroxide standard solution. The requirement is that the acid value ≤ 1.2 mgKOH / g; S2406: Peroxide value inspection: Refer to the method of GB5009.227 - 2016, take a sample of <5> g, dissolve it with a chloroform - glacial acetic acid mixture, add potassium iodide solution, and titrate with 0.01 mol / L sodium thiosulfate standard solution. The requirement is that the peroxide value ≤ 6.0 mmol / kg.

[0030] After all inspection indicators are qualified, determine that the base sauce is qualified and transfer to the next refining process; if one indicator is unqualified, it needs to be returned to the fine grinder for re - grinding for 1 h, and then inspected again until all indicators are qualified. The inspection data needs to be recorded in detail, including sampling time, inspection results, and pass / fail determination, and establish a quality traceability file.

[0031] Specifically, S300 is used for three - stage deep refining, including: S310: Refining machine pre - heating and raw material sealed transfer: S3101: The preheating status of the refining machine directly affects the refining effect. It needs to be preheated to the set temperature in advance. At the same time, the raw material transfer must be carried out in a closed manner to avoid contamination and the entry of moisture and air. An RFM series chocolate refining machine (1200L capacity, 304 stainless steel, with segmented temperature control system and stirring device) is selected. Before use, it should be cleaned and disinfected: circulate and wash with 80℃ hot water for 30 minutes, then blown dry with sterile air, ensuring no residual moisture or impurities remain on the inner wall. S3102: Start the temperature control system of the refining mill, set the preheating temperature to 58±1℃, use electric heating with a heating power of 20kW, and preheat for 1 hour. During preheating, start the stirring device to ensure uniform temperature inside the mill. After preheating, use a platinum resistance temperature sensor to detect the temperature at different locations inside the mill. The temperature deviation should be ≤±0.5℃; otherwise, extend the preheating time by 30 minutes until the temperature stabilizes. S3103: The basic sauce is transferred from the grinding mill to the refining mill using a closed pipeline made of 304 stainless steel, with a diameter of 100mm and a length ≤10m to avoid material residue due to excessive length. Before transfer, the pipeline is purged with nitrogen for 10 minutes to remove air and moisture, at a nitrogen pressure of 0.2MPa. The grinding mill's discharge pump is started at a rate of 50L / min, and the refining mill's feed pump is started simultaneously to create negative pressure, drawing the basic sauce into the refining mill. During transfer, the material temperature is controlled at ≥48℃ to prevent solidification and pipeline blockage due to excessively low temperatures. After transfer, the pipeline is rinsed with 50℃ sterile water for 15 minutes to ensure no material residue remains. After draining the rinsing water, the pipeline is dried with nitrogen for future use. S3104: After transferring to the refining machine, add the remaining 15%-20% of melted cocoa butter from step S100. The amount added is controlled by the weighing system of the refining machine, with an error of ≤±0.5%. After adding, stir for 10 minutes to fully mix the cocoa butter with the base sauce. The mixing temperature is maintained at 58±1℃.

[0032] S320: First stage high-temperature refining: S3201: The core purpose of the first stage of refining is to remove air bubbles, odors, and trace impurities from the material. A high-temperature, short-time process is used, with the temperature set at 58±1℃ and the time at 1.5 hours. The high-temperature environment reduces the viscosity of the material, promotes the rising and bursting of air bubbles, and simultaneously volatilizes residual trace volatile odor substances in the raw materials, such as low-grade aldehydes and ketones produced during cocoa bean processing. S3202: During the refining process, the stirring speed is set to 250 r / min. The stirring paddle adopts a ribbon structure, which ensures that the materials are fully mixed and promotes the rise of bubbles. The formula for calculating the stirring rate is: v=π×D×n / 60, where D is the diameter of the stirring paddle (0.8 m) and n is the speed (250 r / min). The calculated value is v=π×0.8×250 / 60≈10.47 m / s. This rate can ensure the stirring effect without generating too much heat and causing the temperature to exceed the standard. S3203: To monitor the degassing effect, the number of bubbles on the material surface is recorded every 30 minutes using a visual counting method. A 10cm × 10cm observation area is selected, and the number of bubbles is counted. The requirement is that the number of bubbles decreases by ≥30% every 30 minutes, and the number of bubbles in the observation area is ≤5 after 1.5 hours. Simultaneously, the content of volatile odor substances in the material is detected using gas chromatography-mass spectrometry (GC-MS). The requirement is that the total volatile odor substance content is ≤0.01mg / kg, and the content of major odor substances, such as acetaldehyde and acetone, is ≤0.002mg / kg. S3204: During the impurity removal process, trace impurities in the material, such as fine fibers in chocolate powder and residual fruit peels in cocoa liquor, will settle to the bottom of the refiner and be periodically discharged through the bottom drain outlet. The draining time is every 45 minutes, with each draining session lasting 30 seconds. The amount of waste discharged is controlled to be within 0.1% of the total material volume to avoid excessive material loss. After draining, the drain outlet is closed, and refining continues to ensure thorough removal of impurities. S3205: After the first stage of refining is completed, the purity of the material is tested by gravimetric method: take 100g of material, filter it through a 0.45μm filter membrane, weigh the filter residue, and the filter residue weight is required to be ≤0.1g, that is, the impurity content is ≤0.1%. Only after passing the test can it enter the next stage of refining.

[0033] S330: Second Stage Medium Temperature Refining: S3301: The core purpose of the second-stage refining is to promote the integration of various components in the materials, enhance the smoothness and flavor richness of the sauce, and adopt a medium-temperature, long-time process, with the temperature set at 47±1℃ and the time at 7-8 hours. This temperature is the optimal temperature for the esterification reaction of fats, sugars, and proteins. Long-time refining can promote the esterification reaction of fatty acids in cocoa butter with hydroxyl and amino groups in sugars and proteins, forming stable ester compounds, enhancing the binding force between various components, and at the same time ensuring that flavor substances, such as theobromine, caffeine, and flavonoids, are evenly dispersed, thereby improving the flavor richness. S3302: Temperature adjustment employs a gradient cooling method, decreasing from 58±1℃ to 47±1℃ at a cooling rate of 0.2℃ / min to avoid rapid cooling that could lead to material stratification or crystallization. During cooling, the stirring speed is adjusted to 200 r / min to ensure uniform temperature decrease without localized temperature deviations. After cooling to 47±1℃, the temperature is maintained stable with a control accuracy of ±0.5℃, achieved through PID regulation. S3303: During the refining process, samples are taken every hour to test the smoothness and flavor of the material. Smoothness is evaluated using a sensory evaluation method, scored by 5 professional evaluators on a scale of 1-10, with 10 being the best. The average score is required to be ≥8. Flavor is evaluated using a sensory evaluation combined with electronic nose detection. The electronic nose, model: PEN3, detects the flavor profile, requiring a similarity of ≥90% to the standard flavor profile. Simultaneously, the degree of esterification reaction of the material is tested using a titration method to determine the free fatty acid content. The formula for calculating the free fatty acid content is: Free fatty acid content (%) = (V×c×M) / m×100%, where V is the volume of potassium hydroxide standard solution consumed in the titration, c is the concentration of the potassium hydroxide standard solution (0.1 mol / L), M is the average molar mass of fatty acids (282 g / mol, calculated as stearic acid), and m is the sample mass (g). The free fatty acid content is required to be ≤0.5%, indicating sufficient esterification reaction. S3304: The determination of the second-stage refining time is based on the esterification reaction kinetics. Through experiments, the reaction rate constant k = 0.08 h⁻¹ was obtained. The target free fatty acid content is 0.5%, and the initial free fatty acid content is 1.2%. According to the first-order reaction kinetic equation: ln(c0 / c) = kt, where c0 is the initial free fatty acid content and c is the target free fatty acid content, t = ln(1.2 / 0.5) / 0.08 ≈ 11 h was calculated. However, considering the influence of other factors in actual production, the refining time was determined to be 7-8 h based on experiments to ensure that the esterification reaction achieves the expected results.

[0034] S340: Third-stage additives and synergistic refining: S3401: The core purpose of the third-stage refining is to uniformly disperse compound emulsifiers, compound stabilizers, and other additives into the material, fully utilize the synergistic effect of each additive, and improve the stability of the system. The temperature is set to be maintained at 47±1℃ for 0.8 hours. The order of additive addition needs to be designed according to the solubility and mechanism of action of the additives to avoid interactions between additives affecting the effect. The order of addition is: compound emulsifier → compound stabilizer → moisture retention agent → acidity regulator → natural cocoa flavor; S3402: Addition of compound emulsifier: The compound emulsifier is oil-soluble and is dissolved in melted cocoa butter at 50℃ in advance. The mass ratio of emulsifier to cocoa butter is 1:5 to prepare an emulsifier stock solution. The addition rate is 5L / min, and the stirring speed is 220r / min while adding. After the addition is completed, stir for 15min to ensure that the emulsifier is evenly dispersed, reduce the oil-water interfacial tension, and enhance the emulsion stability. S3403: Addition of compound stabilizer: 0.4%-0.5% of the compound stabilizer is water-soluble and should be dissolved in pure water at 38℃ in advance. The mass ratio of stabilizer to water is 1:10 to prepare stabilizer stock solution. The addition rate is 3L / min. During the addition process, the stirring speed is maintained at 220r / min. After the addition is completed, stir for 15min to ensure that the stabilizer is evenly dispersed and forms a three-dimensional network structure to lock in oil and water. S3404: Addition of moisture-retaining agent glycerin: The moisture-retaining agent 0.6%-0.8% is water-soluble and should be added directly and slowly at a rate of 2L / min. The stirring speed is 220r / min. After the addition is complete, stir for 10 minutes. Glycerin can form hydrogen bonds with the moisture in the material, reducing moisture loss and improving the product's moisture retention and taste. S3405: ​​Citric acid addition: The acidity regulator 0.2%-0.3% is water-soluble and is dissolved in pure water at 38℃. The mass ratio of citric acid to water is 1:20 to prepare an acid solution. The addition rate is 1L / min. The pH value of the material is monitored in real time during the addition process using a pH meter with an accuracy of ±0.01. The target pH value is 5.5-6.0. After the addition is completed, stir for 10 minutes to ensure the pH value is uniform, adjust the flavor balance, and inhibit the growth of microorganisms. S3406: Addition of natural cocoa flavoring: 0.1%-0.2% of natural cocoa flavoring is oil-soluble, and the addition rate is 0.5L / min. Stir at 220r / min while adding. After adding, stir for 10min to ensure that the flavoring is evenly dispersed and to enhance the richness of the chocolate flavor. Adding the flavoring in the last step can avoid the volatilization of flavor substances caused by high temperature. S3407: After the additives are added, continue refining for 0.8 hours, maintaining a stirring speed of 220 r / min and a temperature of 47±1℃. During this period, take samples every 15 minutes to test the stability of the material using a centrifugal stability test: take 10 mL of material, centrifuge at 8000 r / min for 30 minutes, and observe whether it separates into layers. No separation or sedimentation is required for stability. Simultaneously test the viscosity of the material; the viscosity should be 800-1000 mPa·s to ensure it meets the requirements of subsequent homogenization processes.

[0035] S350: Quality Verification of Refined Compound Sauces: S3501: After the three-stage refining process, a comprehensive quality verification is conducted on the compound sauce to ensure that all indicators meet the requirements of the subsequent homogenization process. The verification indicators include stability, viscosity, pH value, flavor, and color. S3502: Stability Verification: Centrifugal stability test and static stability test are adopted. Centrifugal stability test: Take 10 mL of material and centrifuge at 8000 r / min for 30 min. No stratification or sedimentation is acceptable. Static stability test: Take 50 mL of material and let it stand at 50℃ for 24 h. No stratification, floating oil or sedimentation is acceptable. S3503: Viscosity verification: A rotational viscometer was used at a temperature of 50℃ and a rotor speed of 60 r / min. Three measurements were taken and the average value was recorded. The required viscosity was 800-1000 mPa·s, with a viscosity deviation of ≤±5%. pH value verification: Using a pH meter, take a 10g sample, dilute it with sterile water to 100mL, and test the pH value. The required pH value is 5.5-6.0, and the pH value deviation is ≤±0.1. Flavor Verification: Sensory evaluation was conducted by 10 professional evaluators, with evaluation indicators including cocoa flavor intensity, sweetness balance, and absence of off-flavors. A 10-point scale was used, and an average score of ≥8.5 was considered passing. Simultaneously, electronic nose detection was used, with a similarity of ≥95% to the standard flavor profile. Color verification: Use a colorimeter to measure L*, a*, and b values. The requirements are L=26-30, a*=13-17, b*=23-27, and the color should be uniform and consistent with no obvious color difference. Once all verification indicators are qualified, the compound sauce is deemed qualified and proceeds to the next homogenization process. If any indicator fails to meet the requirements, the cause must be analyzed and adjustments made: if stability is unqualified, it may be due to insufficient emulsifier or stabilizer; add 0.1% of the compound emulsifier or stabilizer and continue refining for 30 minutes. If viscosity is unqualified, add an appropriate amount of purified water if the viscosity is too high, or add an appropriate amount of compound thickener ≤0.1% if the viscosity is too low. After adjustment, re-verify until all indicators are qualified. Verification data should be recorded in detail, and a quality traceability file should be established.

[0036] Step S400 is used for two-stage high-pressure homogenization emulsification, including: S410: Pure water preheating and precise metering: S4101: The temperature and amount of purified water directly affect the emulsification effect. It needs to be preheated to approximately 38±2℃, close to the temperature of the compound sauce, to avoid low-temperature water damaging the emulsion system. Precise measurement is also crucial to ensure the water content meets requirements. A stainless steel constant-temperature water tank is used. Food-grade purified water is poured into the tank, the heating system is started, the heating temperature is set to 38±2℃, the heating power is 15kW, and the heating time is approximately 1 hour. During the heating process, the stirring device is activated to ensure uniform water temperature. S4102: After the water temperature reaches 38±2℃, maintain the temperature for 30 minutes. Use a platinum resistance temperature sensor to detect the water temperature. The temperature deviation should be ≤±0.5℃. If the temperature is too high, turn on the cooling system to lower the temperature; if the temperature is too low, continue heating until the temperature stabilizes. Use an electromagnetic flowmeter model: LDG-50, with a range of 0-100L / h and an accuracy of ±0.5% to measure the amount of pure water added. The added amount is 24%-27% of the total material weight, calculated based on the weight of the compound sauce. The calculation formula is: V=(m×w) / ρ, where m is the weight of the compound sauce (kg), w is the water content (24%-27%), and ρ is the density of pure water (1kg / L). For example, if the weight of the compound sauce is 500kg and w=25%, then V=(500×25%) / 1=125L. S4103: After metering, purified water is transported to the homogenizer's inlet through an insulated pipe. The pipe is made of 304 stainless steel with an insulation layer to ensure that the water temperature fluctuation during transportation is ≤±1℃. The transportation rate is controlled at 10L / min to avoid excessive local moisture caused by rapid addition, which could disrupt the stability of the compound sauce system.

[0037] S420: Evenly mix and stir the sauce and water: S4201: The compound sauce and purified water must be thoroughly mixed to ensure that the water is evenly distributed in the sauce and to avoid local water accumulation that could lead to emulsification failure. The mixing process is carried out in the pretreatment tank of the homogenizer. The pretreatment tank has a volume of 800L, is made of 304 stainless steel, and is equipped with a stirring device and a temperature control system. S4202: First, transfer the refined compound sauce into the pretreatment tank, control the temperature inside the tank at 50±2℃, start the stirring device, set the stirring speed to 500r / min, and use a propeller-type stirring paddle to promote material circulation. Slowly inject preheated purified water into the pretreatment tank using a metering pump at a rate of 10L / min, stirring continuously. The injection time is determined based on the amount added; for example, adding 125L of purified water would take 12.5min. S4203: After injection, continue stirring for 15 minutes. During stirring, control the temperature inside the tank at 38±2℃ to avoid excessively high or low temperatures affecting the mixing effect. The uniformity of stirring is tested using the moisture distribution test method: samples are taken from three locations—top, middle, and bottom—of the pretreatment tank, with 5g samples taken from each location. The moisture content is measured using a Karl Fischer moisture analyzer. The standard deviation of the moisture content of the three samples is calculated. A standard deviation ≤ 0.5% indicates uniform moisture dispersion. S4204: If the moisture distribution is uneven, extend the stirring time by 5 minutes and take samples again for testing until the standard deviation is ≤0.5%. Observe the state of the material during the mixing process. Under normal circumstances, the material should be a uniform emulsion without obvious water droplets or lumps. If water droplets or lumps appear, it may be due to insufficient stirring speed. Increase the speed to 600 r / min and stir for 5 minutes until the material state is normal.

[0038] S430: Homogenizer Core Parameter Calibration and Setting: S4301: The core parameters of the two-stage homogenizer are the first-stage homogenization pressure of 28±2MPa and the second-stage homogenization pressure of 65±3MPa. Precise calibration is required to ensure parameter stability. The homogenizer selected is the GJB series two-stage high-pressure homogenizer, model: GJB-50, with a maximum working pressure of 100MPa and a processing capacity of 500L / h. Before use, the equipment should be cleaned and disinfected: circulate and wash with 85℃ hot water for 30 minutes, and then dry with sterile air. S4302: Pressure Calibration: Calibrate using a standard pressure gauge with an accuracy of ±0.1MPa. Install the standard pressure gauge at the pressure detection port of the homogenizer. Start the homogenizer and set the first-stage homogenization pressure to 28MPa and the second-stage homogenization pressure to 65MPa. After stabilization, record the standard pressure gauge reading and the pressure value displayed on the equipment. Calculate the pressure error: Pressure Error = (Equipment Display Pressure - Standard Pressure Gauge Reading) / Standard Pressure Gauge Reading × 100%. The pressure error should be ≤±1%. If the error exceeds ±1%, adjust the pressure regulating valve of the homogenizer until the error is ≤±1%. After calibration, fill in the calibration record. The calibration cycle is once a day. S4303: Homogenization temperature setting: During the homogenization process, the material temperature will rise due to the pressure work. It is necessary to control the material temperature after homogenization to ≤60℃ to avoid the destruction of the emulsion system due to high temperature. The cooling medium is 20℃ cooling water with a flow rate of 5m³ / h, which is controlled by the cooling jacket of the homogenizer to ensure the cooling effect. S4304: Homogenization flow rate setting: Based on the homogenizer's processing capacity and production scale, set the homogenization flow rate to 500L / h. Flow calibration is performed using an electromagnetic flow meter, with a flow error of ≤±1%. If the flow rate does not meet the requirements, adjust the feed pump speed until the flow rate stabilizes at around 500L / h. S4305: In addition, the number of homogenization times should be set to 1, and two-stage homogenization is one process, including primary and secondary stages. Experiments have verified that one two-stage homogenization can achieve the target particle size distribution of 0.3-1.5μm. Multiple homogenizations will increase energy consumption and material temperature rise, affecting product quality.

[0039] S440: Primary homogenization process: S4401: The core purpose of primary homogenization is to break down large oil droplets and particle agglomerates after the compound sauce is mixed with water, so as to achieve coarse dispersion. The pressure is set at 28±2MPa. This pressure is determined by experiments: if the pressure is too low (≤25MPa), it is difficult to break down large oil droplets and the coarse dispersion effect is not good; if the pressure is too high (≥30MPa), it will cause the material temperature to rise too high and increase energy consumption. S4402: During the homogenization process, the material passes through the primary homogenizing valve of the homogenizer and undergoes shearing, impact, and cavitation effects under high pressure. The shearing effect refers to the high-speed flow of the material in the gap between the valve seat and the valve core, which is subjected to strong shearing force and breaks up large oil droplets. The impact effect refers to the material impacting the homogenizing ring, further breaking up the oil droplets. The cavitation effect refers to the bursting of bubbles in the material under high pressure, generating tiny shock waves that assist in breaking up the oil droplets. S4403: After primary homogenization, the target particle size of the material is 5-10 μm. The homogenization efficiency is calculated using the formula: η4 = (dbefore - dafter) / dbefore × 100%, where dbefore is the particle size of the material before homogenization (approximately 20 μm), and dafter is the particle size after primary homogenization, 5-10 μm. η4 is calculated to be 50%-75%, requiring a primary homogenization efficiency ≥ 50%. The particle size after primary homogenization is measured using an online laser particle size analyzer every 10 minutes. If the particle size is greater than 10 μm, the primary homogenization pressure is increased by 1-2 MPa, but not exceeding 30 MPa. After adjustment, the pressure is measured again after 5 minutes until the particle size is within the 5-10 μm range. S4404: During the first-stage homogenization process, the material temperature is monitored in real time and recorded every 5 minutes. The temperature rise is controlled within 5-8℃. If the temperature rise exceeds 8℃, it may be due to excessive pressure or poor cooling effect. Reduce the first-stage homogenization pressure by 1-2MPa or increase the cooling water flow rate to ensure that the material temperature after homogenization is ≤55℃. S450: Secondary homogenization process: S4501: The core purpose of secondary homogenization is to further refine the material after primary homogenization, achieve fine dispersion, and obtain a chocolate emulsion with a particle size distribution of 0.3-1.5μm. The pressure is set at 65±3MPa, which is determined according to the target particle size and based on Stokes' law: v=(d²×ρ×g) / (18×η), where v is the particle settling velocity, d is the particle diameter, ρ is the density difference between the particle and the dispersion medium, g is the gravitational acceleration, and η is the viscosity of the dispersion medium. To ensure that the particle settling velocity is small enough (to ensure the stability of the emulsion), the particle diameter d needs to be controlled within 0.3-1.5μm, thus requiring a higher homogenization pressure. S4502: In the secondary homogenization process, the material passes through the secondary homogenization valve and is subjected to stronger shearing, impact, and cavitation effects than in the primary homogenization, further breaking down oil droplets and particles, refining the particle size from 5-10μm to 0.3-1.5μm. The homogenization efficiency is calculated using the formula: η5 = (d1-d2) / d1 × 100%, where d1 is the particle size after primary homogenization (5-10μm), and d2 is the particle size after secondary homogenization (0.3-1.5μm). The calculated η5 = 85%-97%, requiring a secondary homogenization efficiency ≥ 85%. S4503: The particle size after secondary homogenization is detected by an online laser particle size analyzer every 5 minutes. If the particle size is greater than 1.5μm, the secondary homogenization pressure is increased by 2-3MPa, but not exceeding 68MPa. After adjustment, the particle size is detected again after 3 minutes until the particle size is ≤1.5μm. If the particle size is less than 0.3μm, the secondary homogenization pressure is reduced by 2-3MPa, but not lower than 62MPa, to avoid over-homogenization leading to instability of the emulsion system. S4504: During the secondary homogenization process, the material temperature will continue to rise, with the increase controlled within 3-5℃. After homogenization, the material temperature should be ≤60℃. If the temperature exceeds 60℃, increase the cooling water flow rate or reduce the secondary homogenization pressure to ensure the temperature meets the requirements. Simultaneously, observe the appearance of the emulsion. Under normal circumstances, it should be a uniform emulsion with a consistent color, without layering, sedimentation, or obvious particle texture.

[0040] S460: Emulsion Stability and Key Indicator Testing: S4601: After the two-stage homogenization is completed, the chocolate emulsion is subjected to comprehensive testing to ensure that the stability and key indicators meet the requirements of the subsequent freezing process. The testing indicators include particle size distribution, viscosity, stability and pH value. S4602: Particle size distribution detection: A laser particle size analyzer, Malvern Mastersizer 3000, is used. A 5g sample is taken, diluted with anhydrous ethanol, and then tested. The required particle size distribution is 0.3-1.5μm, d50 (median particle size) = 0.8-1.0μm, d90≤1.5μm, d10≥0.3μm, and the particle size distribution span is d90 / d10≤2.5 to ensure uniform particle distribution. S4603: Viscosity test: Use a rotational viscometer, test temperature 38℃, rotor speed 60r / min, take a 50mL sample, test 3 times and take the average value. The required viscosity is 800-1000mPa・s, and the viscosity deviation is ≤±5%. If the viscosity is too high, it will affect the subsequent casting and molding. If it is too low, the emulsion will not be stable enough. S4604: Stability detection: Centrifugal stability test, static stability test and thermal stability test are adopted. Centrifugal stability test: Take 10 mL of the emulsion and centrifuge it at 8000 r / min for 30 min. It is qualified if there is no stratification or precipitation. Static stability test: Take 50 mL of the emulsion and let it stand at 25 °C for 72 h. It is qualified if there is no stratification, no floating oil and no precipitation. Thermal stability test: Take 50 mL of the emulsion and let it stand at 60 °C for 2 h. It is qualified if there is no stratification or precipitation. S4604: pH value detection: Use a pH meter. Take 10 g of the sample, dilute it to 100 mL with sterile water, and detect the pH value. It is required that pH = 5.5 - 6.0 and the pH value deviation ≤ ±0.1 to ensure the acid-base stability of the emulsion and inhibit the growth of microorganisms.

[0041] After all the detection indicators are qualified, it is determined that the chocolate emulsion is qualified and transferred to the next freezing process; if one of the indicators is unqualified, the reason needs to be analyzed and adjusted: if the particle size distribution is unqualified, perform double-stage homogenization again and adjust the homogenization pressure; if the stability is unqualified, add 0.1% of the compound emulsifier, stir evenly and then homogenize again; if the viscosity is unqualified, add an appropriate amount of pure water or compound thickener to adjust, and re-detect after adjustment until all indicators are qualified. The detection data is recorded in detail and a quality traceability file is established.

[0042] The step S500 is used for intelligent program freezing and forming, including: S510: Mold pretreatment and aseptic preparation: S5101: The state of the mold directly affects the forming effect and demolding efficiency of the product. Pretreatment is required to ensure cleanliness, sterility and smooth surface. Select a food-grade silicone mold with a medical-grade silicone material, which is heat-resistant from -50 °C to 200 °C, has no peculiar smell and no migration. The mold specifications are designed according to product requirements, generally 10 g / block, and the surface roughness Ra of the mold cavity ≤ 0.8 μm to ensure the smooth surface of the product. S5102: The mold pretreatment process includes three steps: cleaning, disinfection and pre-cooling. Cleaning: Rinse the mold with sterile water at 50 °C to remove surface dust and impurities, then gently brush the cavity with a soft brush to avoid scratching the mold surface, and rinse it with sterile water until the rinsing water is clear. Disinfection: Put the cleaned mold into an ultraviolet disinfection box with a UV wavelength of 254 nm, an irradiation intensity of 30 μW / cm², and an irradiation time of 30 min to ensure the sterility of the mold surface with the total number of colonies ≤ 10 CFU / cm². Pre-cooling: Put the disinfected mold into a cold storage at -10 °C for 2 h to cool the mold temperature to -10 ± 1 °C. Avoid contact with normal temperature air for the pre-cooled mold to prevent surface condensation and affect the forming after the emulsion is poured. S5103: After mold pretreatment, the inner wall of the cavity is sprayed with food-grade cocoa butter, the same as that used in S100. The purpose of spraying is to reduce the adhesion between the emulsion and the inner wall of the mold, facilitating subsequent demolding. Spraying is performed using a pneumatic spray gun at a pressure of 0.3 MPa. The cocoa butter is heated to 50°C to melt, and the spraying amount is 0.1 g / cm². After spraying, the mold is placed in a -10°C cold storage for 10 minutes to allow the cocoa butter to form a uniform thin film on the inner wall of the mold. S5104: The pre-treated mold is transported to the casting station by a sterile transport vehicle. Collisions and contamination are avoided during the transport process. The mold is placed on a sterile operating table. The surface of the operating table is disinfected with 75% alcohol. Operators wear sterile work clothes, gloves and masks to ensure that the operation process is sterile.

[0043] S520: Emulsion casting and preliminary shaping: S5201: When pouring emulsion, the pouring speed and amount must be controlled to ensure that the material fills the mold cavity evenly and avoids air bubbles and insufficient filling. A pneumatic pouring machine should be selected as the pouring equipment. Before pouring, the outlet of the pouring machine should be disinfected with 75% alcohol to avoid contamination. S5202: Arrange the pre-cooled molds neatly on the worktable of the casting machine, with a mold spacing of 5cm to facilitate cold air circulation during the subsequent freezing process. Start the casting machine, set the casting volume to 10g / cavity, and adjust according to the mold specifications. The casting speed is 5g / s. During the casting process, the discharge port of the casting machine should be 5cm away from the mold cavity opening to avoid the emulsion splashing due to being too high and the material accumulating due to being too low. S5203: After pouring, use a sterile scraper to gently scrape the emulsion on the surface of the mold cavity to remove excess material and ensure that the product thickness is uniform, about 10mm. After scraping, place the mold in a pre-cooling chamber at -10℃ and let it stand for 10 minutes to allow the emulsion to initially set and avoid material flow during the subsequent cooling process, which could cause the product shape to become irregular. S5204: Observe the state of the emulsion during the initial shaping process. Under normal circumstances, it should be semi-solid and without flow. If flow occurs, it may be due to excessive pre-cooling temperature of the mold or insufficient settling time. Reduce the pre-cooling temperature of the mold to -12℃ and extend the settling time by 5 minutes to ensure the initial shaping effect.

[0044] S530: Precise setting of three-stage programmed cooling parameters: S5301: The core of the three-stage programmed cooling is precise control of the cooling rate and holding time to ensure that the material's core temperature is ≤-48℃ and the ice crystal size is controlled within 10-30μm. Ice crystals that are too small will result in excessively high porosity in the freeze-dried product, while those that are too large will damage the product structure, affecting taste and temperature resistance. The cooling equipment used is an intelligent programmed freezer, model: KX-1000, with a temperature control range of -60℃ to 20℃, a temperature control accuracy of ±0.5℃, and an adjustable cooling rate of 0.1-5℃ / min. S5302: First-stage cooling parameters: The initial temperature is approximately 38℃ after the emulsion is poured, the target temperature is -10℃, the cooling rate is 0.8℃ / min, and the holding time is 1h. The purpose of this stage is to allow the water in the emulsion to slowly precipitate out, avoiding the formation of large ice crystals due to rapid cooling. The cooling rate is set based on the water diffusion rate, determined through experiments: if the cooling rate is too fast (≥1℃ / min), the water diffusion is insufficient, resulting in uneven ice crystal size; if the cooling rate is too slow (≤0.5℃ / min), the production efficiency is too low. The cooling time calculation formula is: t1=(T0-T1) / v1, where T0=38℃, T1=-10℃, v1=0.8℃ / min, and the calculated t1=(38-(-10)) / 0.8=60min=1h, which is consistent with the set holding time; S5303: Second-stage cooling parameters: initial temperature -10℃, target temperature -35℃, cooling rate 1.5℃ / min, holding time 3h. The purpose of this stage is to promote uniform ice crystal growth and form a regular crystal structure. The cooling rate of 1.5℃ / min can balance the ice crystal growth rate and uniformity, and the holding time of 3h ensures sufficient ice crystal growth. The cooling time calculation formula is: t2=(T1-T2) / v2=(-10-(-35)) / 1.5≈16.7min. The holding time of 3h was determined experimentally to ensure that the ice crystal size reaches 20-30μm. S5304: Third-stage cooling parameters: initial temperature -35℃, target temperature -50℃, cooling rate 2.5℃ / min, holding time 2h. The purpose of this stage is to further reduce the core temperature of the material and lock in the ice crystal morphology. The cooling rate of 2.5℃ / min can quickly reduce the core temperature of the material to below -48℃, and the holding time of 2h ensures uniform temperature inside and outside the material. The cooling time calculation formula is: t3=(T2-T3) / v3=(-35-(-50)) / 2.5=6min, and the holding time of 2h was determined experimentally to ensure that the core temperature of the material is ≤-48℃; S5305: After setting the three-stage cooling parameters, input them into the intelligent cold storage control system. Before starting the program, perform parameter calibration. The calibration method is as follows: place a temperature sensor (accuracy ±0.1℃) at the center of the material in the mold, start the cooling program, record the temperature change in real time, and compare it with the set parameters. The temperature deviation should be ≤±0.5℃, and the cooling rate deviation should be ≤±0.1℃ / min. If the deviation exceeds the allowable range, adjust the control system parameters until the calibration is qualified.

[0045] S540: Real-time monitoring of freezing process temperature and ice crystal state: S5401: The freezing process requires real-time monitoring of the surface temperature, center temperature, and ice crystal state of the material to ensure that the parameters are executed correctly. Temperature monitoring adopts a multi-point temperature monitoring system, with a platinum resistance temperature sensor placed on the material surface and center of each mold, with an accuracy of ±0.1℃. The sensor data is transmitted to the control system in real time, and the temperature value is recorded every 1 minute. S5402: First stage monitoring: Monitor whether the cooling rate is stable at 0.8±0.1℃ / min. When the temperature drops to -10℃, maintain it for 1 hour and monitor whether the core temperature of the material is stable at -10±0.5℃. If the temperature fluctuation is too large, adjust the refrigeration power of the freezer to ensure temperature stability. At the same time, observe the ice crystal morphology using a low-temperature microscope (model: BX53) at a cooling temperature of -50℃. Ice crystals begin to form at this stage, with a size of about 5-10μm, and no obvious large ice crystals. S5403: Second Stage Monitoring: Monitor whether the cooling rate is stable at 1.5±0.1℃ / min. When the temperature drops to -35℃, maintain it for 3 hours. Observe the ice crystal state every 30 minutes using a cryogenic microscope. The ice crystal size should gradually increase to 20-30μm, with a regular shape and uniform distribution. If large ice crystals appear (>30μm), the cooling rate may be too fast; reduce the cooling rate by 0.2℃ / min. If the ice crystal size is too small (<20μm), the cooling rate may be too slow; increase the cooling rate by 0.2℃ / min. At the same time, monitor the core temperature of the material; it should be stable at -35±0.5℃. S5404: Third-stage monitoring: Monitor whether the cooling rate is stable at 2.5±0.1℃ / min. When the temperature drops to -50℃, maintain it for 2 hours, recording the material's center temperature every 30 minutes. The center temperature should be ≤-48℃. If the center temperature is higher than -48℃, extend the holding time by 30 minutes; if the center temperature is lower than -52℃, the holding time can be appropriately shortened, but not less than 1.5 hours. Observe the ice crystal state using a low-temperature microscope. The ice crystal size should be stable at 10-30μm with no change in morphology, ensuring successful ice crystal locking. S5405: During the freezing process, the humidity inside the freezer must also be monitored. The relative humidity should be controlled at ≤60% to avoid frost formation inside the freezer affecting the cooling effect. If the humidity exceeds 60%, the dehumidification system should be activated until the humidity drops to ≤60%. At the same time, observe the condition of the mold and the material. There should be no ice formation or deformation, and the material should fit well with the mold. S550: Demolding process optimization and quick-freezing enhancement in blast freezers: S5501: After freezing, demolding is required to separate the material from the mold. The timing and method of demolding directly affect the integrity of the product shape. Demolding should be carried out after the third stage of heat preservation, when the center temperature of the material is ≤-48℃. At this time, the material has higher hardness and the adhesion to the inner wall of the mold is minimal. S5502: Demolding is performed mechanically using a pneumatic demolding machine, model TMM-200, with a working pressure of 0.4MPa. The ejector pins of the demolding machine correspond one-to-one with the mold cavities, with a pin diameter of 1 / 3 of the cavity diameter and a pin stroke of 15mm, ensuring smooth material ejection and preventing product damage. During demolding, operators wear low-temperature gloves resistant to -50℃. The operating speed is 10 molds / min. Demolded material is immediately placed on a sterile tray made of 304 stainless steel, which is UV sterilized. The tray is lined with sterile plastic wrap to prevent material contamination and moisture absorption. S5503: After demolding, the pallet containing the material is quickly transferred to a -50℃ quick-freezing chamber, model: SD-2000, with a temperature control range of -60℃ to 0℃ and a temperature control accuracy of ±1℃, for rapid freezing enhancement. The rapid freezing time is 3 hours. The purpose of this process is to further reduce the material temperature so that the final material temperature is ≤-40℃, lock the ice crystal morphology, and prevent the ice crystals from melting or growing during subsequent transfer. S5504: Monitor the temperature of the quick-freezing chamber during the quick-freezing process, recording it every 30 minutes. The temperature should be stable at -50±1℃. The final temperature of the material should be detected by a temperature sensor and should be ≤-40℃. If the temperature is higher than -40℃, extend the quick-freezing time by 30 minutes. After quick-freezing, the material should be a hard solid with an intact shape, without cracks or deformation, and the ice crystal size should be stable at 10-30μm.

[0046] S560: Quality Inspection and Acceptance Determination of Frozen Materials S5601: The quality of the frozen material directly affects the subsequent freeze-drying effect. The test indicators include core temperature, ice crystal size, shape integrity, and moisture content. S5602: Center temperature detection: An insertion-type temperature sensor with an accuracy of ±0.1℃ is used. It is inserted into the center of the material and the temperature value is recorded. The center temperature is required to be ≤-48℃, and the qualified standard is ≤-48℃.

[0047] Ice crystal size detection: Using a low-temperature scanning electron microscope, material samples are taken, frozen and fractured with liquid nitrogen, and the ice crystal morphology on the fracture surface is observed and the ice crystal size is measured. The ice crystal size is required to be 10-30μm, uniformly distributed, without large ice crystals, and the size is >30μm.

[0048] Shape integrity inspection: Visual inspection method is used to observe whether the shape of the material is consistent with the mold cavity, with no cracks, no deformation, and no missing corners, and the pass rate is ≥98%.

[0049] Moisture content, frozen state test: A Karl Fischer moisture analyzer, KF-1A, with an accuracy of ±0.01%, was used. A material sample of about 5g was taken, rapidly pulverized, and then tested. The moisture content was required to be consistent with the moisture content of the emulsion before freezing, which was 24%-27%, with a deviation of ≤±0.5%, to ensure that there was no moisture loss during the freezing process. S5604: After all test indicators are qualified, the frozen material is deemed qualified and proceeds to the next freeze-drying process. If any indicator is unqualified, the cause must be analyzed and addressed: if the center temperature is unqualified, extend the blast freezer time; if the ice crystal size is unqualified, adjust the three-stage cooling parameters and refreeze; if the shape integrity is unqualified, check the demolding process or mold condition, adjust, and re-demold. Detailed test data should be recorded, and a quality traceability file should be established.

[0050] Step S600 is used for intelligent parameter optimization and knowledge base construction, including: S610: Construction and Calibration of Key Attribute Data Acquisition System S6101: The accuracy of key attribute data is the foundation of parameter optimization. A multi-dimensional data acquisition system needs to be built. The key data to be collected includes material attribute data, ice crystal size, core temperature, emulsion viscosity, particle size, moisture content, process parameter data, freezing rate, homogenization pressure, freeze-drying temperature and time, product performance data, temperature resistance, taste and stability. The data acquisition system consists of sensors, data transmission modules and data storage modules. S6102: Sensor Selection and Installation: Ice crystal size is acquired using a combination of low-temperature scanning electron microscopy and laser particle size analyzer. The low-temperature scanning electron microscopy is used to observe the morphology and size of ice crystals, while the laser particle size analyzer is used for quantitative analysis. A platinum resistance temperature sensor is used for the center temperature of the material, installed at the center. Emulsion viscosity is measured using an online rotational viscometer, installed at the homogenizer outlet. Particle size is measured using an online laser particle size analyzer, installed on the circulation pipes of the grinding mill and homogenizer. Moisture content is measured using an online Karl Fischer moisture analyzer, model: KFS-200, accuracy ±0.01%, installed at the pretreatment tank and freeze dryer outlet. Process parameter data, including freezing rate, homogenization pressure, freeze-drying temperature, and time, are directly acquired through the equipment control system. Product performance data is acquired through a combination of manual and instrumental testing. Temperature resistance is tested using a constant temperature chamber, model: HH-B11.500-S. Taste is evaluated using sensory evaluation, and stability is tested using centrifugal stability testing. S6103: Data Transmission Module: Employing a combination of Industrial Ethernet, Profinet, and wireless transmission, sensor data is transmitted to the local server via Industrial Ethernet, while remote data is transmitted to the cloud platform via LoRa. The data transmission rate is 10Mbps, with a transmission latency of ≤100ms, ensuring real-time data transmission. AES-256 encryption is used during data transmission to prevent data leakage and tampering. S6104: Data storage module: The local server uses a MySQL database to store real-time data for the most recent 3 months, and the cloud platform uses Alibaba Cloud OSS storage to store historical data and backup data. The data storage capacity is ≥10TB and the data retention period is ≥3 years, which is convenient for traceability and analysis. S6105: System Calibration: The data acquisition system shall be calibrated monthly. The calibration method is as follows: For sensors, calibration shall be performed using standard materials or standard equipment, such as calibrating temperature sensors with a standard thermometer and viscometers with a standard viscosity liquid, to ensure that the sensor accuracy meets the requirements; for data transmission modules, the transmission rate and delay shall be checked to ensure that they meet the design requirements; for data storage modules, the integrity and security of data storage shall be checked to ensure that there is no data loss or tampering. After calibration, a calibration record shall be filled out, and any unqualified sensors or modules shall be replaced promptly.

[0051] S620: Quantitative Definition and Classification of Product Quality Standards: S6201: Product quality standards are the basis for parameter optimization. They need to be quantitatively defined and graded to ensure that parameter optimization has clear objectives. Quality standards are divided into core indicators and secondary indicators. Core indicators include temperature resistance, moisture content, and taste. Secondary indicators include stability, color, and shape integrity. S6202: Quantitative Definition of Core Indicators: Temperature resistance: After being placed in an environment of 35℃ and 50% relative humidity for 4 hours, the product shows no softening, melting, or deformation, and there is no oil seepage on the surface. Sensory evaluation and gravimetric testing are used to determine the weight change rate, which is ≤1% (weight change rate = (weight after placement - weight before placement) / weight before placement × 100%). Temperature resistance is divided into 3 levels: Level 1 (no softening for ≥5 hours), Level 2 (no softening for 4-5 hours), and Level 3 (no softening for 3-4 hours). The target level is Level 1.

[0052] Moisture content: The moisture content of the freeze-dried product is ≤1.5%, which is tested using a Karl Fischer moisture analyzer. The moisture content is divided into 3 levels: Level 1 ≤1.0%, Level 2 1.0%-1.5%, and Level 3 1.5%-2.0%. The target level is Level 1.

[0053] Taste: Sensory evaluation was conducted by 10 professional evaluators. Evaluation indicators included smoothness, fineness, and cocoa flavor intensity. Each indicator was scored from 1 to 10. A total score of ≥85 was Level 1, 75-84 was Level 2, 65-74 was Level 3, and the target level was Level 1. S6203: Quantitative Definition of Secondary Indicators: Stability: After being sealed and stored at 25℃ for 6 months, there is no moisture absorption, no deterioration, and no odor. The change in moisture content is ≤0.3%. The stability is divided into two levels: qualified and meets the requirements, and unqualified and does not meet the requirements. The target is qualified.

[0054] Color: Detected using a CR-400 colorimeter, L*=28-32, a*=12-16, b*=22-26. The color is uniform and consistent, with a color difference ΔE≤1.0. It is divided into two levels: qualified and meets the requirements, and unqualified and does not meet the requirements. The target is qualified.

[0055] Shape integrity: No cracks, no deformation, no missing corners, pass rate ≥98%, divided into 2 levels: qualified and meets the requirements, unqualified and does not meet the requirements, with the target being qualified; S6204: After the quality standards are quantified, they are entered into the intelligent parameter optimization system as the target values ​​for parameter optimization. At the same time, a quality standard database is established to store the quality inspection data of different batches of products, which is convenient for subsequent analysis and optimization.

[0056] S630: Three-level validation dataset establishment and optimization algorithm design: S6301: The three-level validation dataset is the foundation of parameter optimization. It is constructed through orthogonal experiments and single-factor experiments and contains product performance data corresponding to different combinations of process parameters. The dataset is divided into a first-level validation dataset, single-factor experiment data, a second-level validation dataset, orthogonal experiment data, a third-level validation dataset, and actual production data. S6302: Construction of the primary verification dataset: Select key process parameters, freezing rate, primary homogenization pressure, secondary homogenization pressure, sublimation drying temperature, and analytical drying temperature. Set 5 levels for each parameter and conduct single-factor experiments. Other parameters are fixed at intermediate values. Each experimental condition is repeated 3 times. Record product performance data, temperature resistance, moisture content, and taste to construct the primary verification dataset. The sample size is 5×5×3=75 groups. S6303: Construction of the Secondary Validation Dataset: Based on the results of the Primary Validation Dataset, four parameters with significant influence were selected: freezing rate, secondary homogenization pressure, sublimation drying temperature, and desorption drying temperature. Each parameter was set to three levels, using the L9 (3...) method. 4 An orthogonal experimental design was used, with each experimental condition repeated three times. Product performance data were recorded, and a secondary verification dataset was constructed with a sample size of 9×3=27 groups.

[0057] S640: Parameter optimization process execution and iterative verification: S6401: The parameter optimization process uses "data acquisition - comparative analysis - parameter adjustment - verification and testing - iterative optimization" as its core logic. Based on a three-level verification dataset and preset quality standards, it achieves precise optimization of core process parameters through intelligent algorithms. First, key data of the current production batch is extracted from the data acquisition system, including material property data, ice crystal size d, center temperature T, emulsion viscosity η, particle size D, and process parameter data such as freezing rate vcold, primary homogenization pressure P1, secondary homogenization pressure P2, sublimation drying temperature Trise, and analytical drying temperature Tsolution. These data are then correlated and compared with the three-level verification dataset. S6402: The comparative analysis uses a weighted scoring method to construct the objective function: F = ω1 × F temperature resistance + ω2 × F moisture + ω3 × F taste + ω4 × F stability + ω5 × F color, where ω1-ω5 are weight coefficients (determined by the analytic hierarchy process: ω1=0.4, ω2=0.3, ω3=0.2, ω4=0.05, ω5=0.05), and F temperature resistance, F moisture, etc., are the scores of each quality indicator. A first-level indicator receives 10 points, a second-level indicator receives 8 points, a third-level indicator receives 6 points, and a non-conforming indicator receives 0 points. The threshold for the objective function is set to F ≥ 9 points. If the current batch's F < 9 points, the parameter adjustment process is initiated. S6403: Parameter adjustment is based on the process parameter-quality index mapping model established by the response surface methodology. The model expression is: Y=a0+a1x1+a2x2+…+anxn+b1x1²+b2x2²+…+bnxn²+c1x1x2+…+cpxqx, where Y is the quality index, x1-xn are the process parameters, and a0-an, b1-bn, and c1-cp are the regression coefficients.

[0058] For example, if the temperature resistance is substandard and the temperature resistance score is <10, and the analysis shows that the secondary homogeneous pressure P2 and the freezing rate v_cold are the main influencing factors, according to the model calculation, P2 needs to be increased by ΔP2 = (10 - temperature resistance) × kP (kP is the pressure adjustment coefficient, which was determined by experiments to be kP = 2MPa / min), and v_cold needs to be adjusted by Δv_cold = (10 - temperature resistance) × kv (kv = 0.1℃ / min / min). S6404: After parameter adjustment, start small-batch verification production. The verification batch output is 10% of the normal batch. S100-S500 processes are strictly executed according to the adjusted parameters. After production is completed, a comprehensive quality inspection is performed on the verification batch products, and the objective function F value is calculated. If F ≥ 9, the parameter optimization is deemed qualified, and this set of parameters is taken as the optimal parameters for the current production batch. If F < 9, analyze the reasons for the adjustment deviation, such as excessive parameter adjustment or failure to consider interaction effects. Substitute the parameters back into the objective function and mapping model, correct the adjustment coefficients, and conduct small-batch verification again until F ≥ 9. S6405: The termination condition for iterative verification is: the objective function F ≥ 9 points for three consecutive verification batches, and the fluctuation range of each quality indicator ≤ ±5%, ensuring the stability and repeatability of the optimized parameters. During each iteration, detailed records of parameter adjustments, verification data, and objective function scores are generated to form an optimization log, providing data support for subsequent knowledge base updates. For example, for a batch with initial parameters P2 = 65 MPa, v_cold = 1.5℃ / min, F_temperature resistance = 8 points, and a grade of 2, after calculation and adjustment, P2 = 69 MPa, v_cold = 1.7℃ / min. After verification, F_temperature resistance = 10 points, and F = 9.2 points, meeting the threshold requirements, and parameter optimization is complete.

[0059] S650: Intelligent Knowledge Base Construction and Association Mapping S6501: The intelligent knowledge base uses a "task-attribute-scenario-solution" architecture as its core to achieve multi-dimensional associated storage and rapid matching of raw material ratios, process parameters, product attributes, and application scenarios. First, the core data that meets the standards undergoes structured processing, classifying the data into three layers: a basic data layer, an association rule layer, and an application scenario layer. S6502: Basic Data Layer: Stores raw material proportioning data (weight parts of each component, proportion of compound additives), process parameter data (temperature, pressure, time, speed, etc. for each process), and product attribute data (temperature resistance, moisture content, taste score, stability, etc.). The data format adopts standardized JSON format, as shown in the example below: { Raw material ratio: {"Cocoa butter": 20%, "Cocoa liquor": 18%, ...," Compound emulsifier": 1.0%} Process parameters: {"S200 rotation speed": 800 r / min,"S400P2": 69 MPa,...,"S700T lift": 30-60℃} Product Attributes: {"Temperature Resistance": 5h,"Moisture Content": 1.2%,"Taste Score": 88 points, ...} }; S6503: Association Rule Layer: Based on machine learning algorithms, this layer mines the associations between data and calculates the influence weights of each process parameter on product attributes. The weight calculation formula is: Wij = (Cov(Xi,Yj)) / (Var(Xi) × Var(Yj))^0.5, where Wij is the weight of the i-th process parameter on the j-th product attribute, Cov(Xi,Yj) is the covariance, and Var is the variance. For example, the calculated weights for secondary homogenization pressure on temperature resistance are W = 0.85 and for freezing rate on ice crystal size are W = 0.78. These association weights are stored in a rule base for rapid parameter matching. S6504: Application Scenario Layer: This layer maps product attributes to application scenarios, which are categorized into three types: room temperature storage (18-25℃), high-temperature transportation (30-35℃), and ready-to-eat in room temperature / high-temperature environments. Each scenario corresponds to specific product attribute requirements. For example, high-temperature transportation scenarios require a temperature resistance ≥4.5h and a moisture content ≤1.3%, while room temperature storage scenarios require a temperature resistance ≥4h and a shelf life ≥12 months. A scenario-attribute mapping table is constructed, and a fuzzy matching algorithm is used to quickly associate scenarios with product attributes. S6505: The knowledge base uses a distributed database architecture for storage. Local servers store frequently accessed data and production data from the past six months, while the cloud stores all data. A data index is also established, with index keywords including raw material ratios, process parameter ranges, product attribute indicators, and application scenarios, ensuring a query response time of ≤0.5s. The knowledge base update mechanism is as follows: every 10 production batches, a data update is automatically triggered, incorporating new compliant data into the knowledge base, retraining the association rule model, and updating the weight coefficients to ensure the timeliness and accuracy of the knowledge base.

[0060] S660: Knowledge Base Validation and Usability Optimization: S6601: After the intelligent knowledge base is built, it needs to be fully verified to ensure its practicality and accuracy in actual production. The verification is divided into three stages: functional verification, matching accuracy verification, and production adaptation verification. S6602: Functional Verification: Test the query, matching, and update functions of the knowledge base. By inputting different application scenarios, such as "high-temperature transportation + taste priority," verify whether the knowledge base can quickly retrieve the corresponding raw material ratios and process parameters. The query response time should be ≤0.5s, and the matching result completeness should be ≥98%, i.e., including all relevant parameters. When testing the update function, manually input new compliant data and verify whether the knowledge base can automatically incorporate and update the association rules. The update completion time should be ≤30min. S6603: Matching Accuracy Verification: Ten different application scenarios are selected. Based on the knowledge base, corresponding process parameters are obtained through matching. Small-batch production is conducted, with three batches per scenario. The product attributes are then checked to ensure they meet the scenario requirements. The matching accuracy calculation formula is: Matching Accuracy = (Number of batches meeting scenario requirements) / (Total number of verified batches) × 100%, requiring a matching accuracy ≥ 95%. For example, for the "high-temperature transportation" scenario, the knowledge base matching parameters are P2 = 70MPa, v_cooling = 1.8℃ / min, and T_rise = 30-60℃. After production, the temperature resistance of all three batches is ≥ 4.5h, and the moisture content is ≤ 1.3%. Therefore, the matching accuracy for this scenario is determined to be 100%. S6604: Production Adaptability Verification: Apply the knowledge base to large-scale production, with a single batch output ≥1000kg, and conduct 20 consecutive batches of production, monitoring indicators such as product qualification rate, parameter adjustment frequency, and production efficiency. Requirements for large-scale production include a product qualification rate ≥95%, a parameter adjustment frequency ≤5% (i.e., ≤1 adjustment per 20 batches), and a production efficiency improvement ≥10%, compared to production efficiency without using the knowledge base.

[0061] If the matching accuracy is not up to standard or the production adaptability is poor during the verification process, analyze the reasons and optimize: if it is due to unreasonable association weights, recalculate the weight coefficients; if it is due to insufficient data coverage, supplement the test data for the corresponding scenario until all verification indicators are qualified.

[0062] Step S700 is used for precise gradient freeze-drying, including: S710: Optimization of Pre-treatment and Material Loading in Freeze Dryers: S7101: The pretreatment quality and material loading method of the freeze dryer directly affect the freeze-drying efficiency and product quality. It is necessary to ensure the freeze dryer is clean, the vacuum system is stable, and the heating plate temperature is uniform. Simultaneously, the material loading density should be optimized, and cold air circulation should be guaranteed. An LGJ series vacuum freeze dryer, model: LGJ-10000, with a chamber volume of 10m³, a vacuum range of 1-1000Pa, a heating plate temperature control range of -50℃ to 80℃, and a temperature control accuracy of ±0.5℃, should be fully pretreated before use. S7102: Freeze Dryer Cleaning and Disinfection: First, use 85℃ hot water to circulate and clean the chamber, heating plate, and condenser coil for 30 minutes to remove residual materials and impurities. Then, spray the inside of the chamber and material trays with 75% edible alcohol for disinfection. After disinfection, close the chamber door and turn on the ultraviolet disinfection lamp at a wavelength of 254nm and a power of 30W for 60 minutes to ensure sterility inside the chamber and a total bacterial count ≤10 CFU / m³. After cleaning and disinfection, dry the inside of the chamber with sterile air to avoid residual moisture affecting the vacuum level. S7103: Vacuum System Calibration and Testing: Start the vacuum system, close the chamber door, and perform a vacuum test. Set the target vacuum level to 15 Pa and record the time it takes for the vacuum level to drop from atmospheric pressure to 15 Pa. The time should be ≤30 minutes. If the time is too long, check the vacuum pump seal and any leaks in the piping, and perform maintenance and adjustments. Use a standard vacuum gauge with an accuracy of ±0.1 Pa to calibrate the freeze dryer's vacuum sensor. The calibration formula is: Vacuum error = (Freeze dryer displayed vacuum level - Standard vacuum gauge reading) / Standard vacuum gauge reading × 100%. The error should be ≤±2%; otherwise, adjust the sensor parameters. S7104: Heating plate temperature calibration and uniformity test: Set three key points for heating plate temperature: 30℃, 60℃, and 32℃. After each temperature point stabilizes for 30 minutes, use a platinum resistance temperature sensor with an accuracy of ±0.1℃ to detect the temperature at different locations on the heating plate, for a total of 10 detection points. Calculate the temperature uniformity: ΔT=Tmax-Tmin. The requirement is ΔT≤±0.5℃. If the uniformity does not meet the standard, adjust the heating power distribution of the heating plate to ensure that the temperature in each area is consistent. S7105: Material Loading Optimization: After blast freezing, the material is transferred from the blast freezer to the freeze dryer using an insulated transport vehicle, maintaining a temperature ≤-40℃ to prevent the ice crystals from melting due to material temperature rise. The material trays are made of 304 stainless steel with 5mm diameter vents spaced 20mm apart at the bottom to facilitate water vapor escape. During loading, the material is evenly spread on the trays, with a layer thickness controlled at 10-15mm and a layer density of 0.8-1.0kg / m². The spacing between trays is ≥10cm, and the spacing between the tray and the heating plate is ≥5cm to ensure uniform airflow within the chamber and prevent incomplete freeze-drying in certain areas. After loading, the trays are neatly arranged inside the freeze dryer, the door is closed, and the freeze-drying process is ready to begin.

[0063] S720: Dynamic adjustment and execution of sublimation drying parameters: S7201: The core of sublimation drying is to directly sublimate ice crystals in the material into water vapor under low-temperature vacuum conditions. This water vapor is then captured by a condenser coil to remove moisture. Gradient temperature control and dynamic vacuum regulation are necessary to ensure sublimation efficiency and product structural integrity. The sublimation drying parameters are set based on optimized data from an intelligent knowledge base. Initial parameters are: vacuum degree 15±2Pa, heating plate temperature gradually increased from 30℃ to 60℃ at a rate of 5℃ / h, sublimation drying time 13-14h, and material temperature always ≤-25℃. S7202: Gradient Heating Control: The heating process is divided into 6 stages, with each stage maintaining the temperature for 2 hours. The heating rate is 5℃ / h, specifically: 30℃ (0-2h) → 35℃ (2-4h) → 40℃ (4-6h) → 45℃ (6-8h) → 50℃ (8-10h) → 60℃ (10-14h). The purpose of gradient heating is to prevent the material surface from sublimating too quickly, causing a crust to form and hindering the escape of internal water vapor, while gradually increasing the sublimation rate. During the heating process, the temperature is precisely controlled by the PID control system of the heating plate. The temperature of the heating plate and the material temperature are recorded every 10 minutes. The material temperature is detected by an infrared temperature sensor (accuracy ±0.5℃). If the material temperature is higher than -25℃, the heating is paused, and the current temperature is maintained for 1 hour. Heating resumes only after the material temperature drops below -25℃. S7203: Dynamic Vacuum Regulation: The vacuum level is dynamically adjusted according to the sublimation rate of the material. The sublimation rate is calculated using the formula: vl = (m0 - mt) / (S × t), where m0 is the mass of the material before sublimation (kg), mt is the mass of the material at time t (kg), S is the surface area of ​​the material (m2), and t is the sublimation time (h). The sublimation rate is calculated by real-time monitoring of the material mass change using a weighing sensor. When vl > 0.5 kg / (m2·h), the vacuum level is appropriately increased to 16-18 Pa to promote rapid water vapor removal; when vl < 0.3 kg / (m2·h), the vacuum level is appropriately decreased to 12-14 Pa to prevent crust formation on the material surface. The vacuum level adjustment increment is ≤ 2 Pa each time, with an adjustment interval ≥ 30 min to ensure system stability. S7204: Condensing coil temperature control: The temperature of the condensing coil must be 10-15℃ lower than the material temperature to ensure rapid condensation of water vapor. The condensing coil temperature is set to -40℃ and precisely controlled by the refrigeration system, with a temperature fluctuation range of ≤±1℃. Check the frost condition of the condensing coil every 1 hour. If the frost thickness is ≥5mm, start the defrosting program, raise the temperature to 10℃, hold for 10 minutes, and then lower the temperature to -40℃ to avoid excessive frost affecting the condensation effect. S7205: During the sublimation drying process, the moisture content of the material should be sampled and tested every 2 hours using a Karl Fischer moisture analyzer. The moisture content of the material after sublimation drying should be ≤5%. If the moisture content is too high, the sublimation drying time should be extended by 1 hour until the moisture content meets the standard. At the same time, observe the morphology of the material. If there is no surface cracking or collapse and the structure is intact, the sublimation drying is considered qualified.

[0064] S730: Monitoring and Handling of Anomalies in the Sublimation Drying Process S7301: The sublimation drying process requires real-time monitoring of key parameters such as vacuum degree, heating plate temperature, material temperature, condenser coil temperature, and sublimation rate. An abnormality early warning mechanism should be established to handle various abnormal situations in a timely manner and ensure the stability of the freeze-drying process. S7302: Parameter Monitoring System: Employs a multi-channel data acquisition instrument (Agilent 34970A) to synchronously acquire various parameter data every 5 minutes. Data is transmitted to the control system in real time, generating trend curves for operators to observe parameter changes. Alarm thresholds are set for each parameter: vacuum ≤10Pa or ≥20Pa, heating plate temperature ≤28℃ or ≥62℃, material temperature ≥-23℃, condenser coil temperature ≥-38℃, and sublimation rate ≤0.2kg / (m²・h) or ≥0.6kg / (m²・h). When a parameter exceeds the threshold, the control system automatically issues an audible and visual alarm and records the abnormal time and value. S7303: Common exception handling solutions: Abnormal increase in vacuum: If the vacuum suddenly rises to ≥20Pa, it may be due to a faulty door seal, pipeline leaks, or a vacuum pump malfunction. First, check the door sealing ring; if it is aged or damaged, replace it. Check the pipeline connections; if there are leaks, tighten them again. If the vacuum pump pressure is insufficient, start the backup vacuum pump and simultaneously shut down the faulty pump for repair. After handling these issues, re-evacuate to 15±2Pa and continue sublimation drying. Abnormal rise in material temperature: If the material temperature is ≥-23℃, it may be due to excessively high heating plate temperature or excessively rapid sublimation rate. Immediately reduce the heating plate temperature by 5℃, simultaneously reduce the vacuum level to 12Pa, pause heating for 1 hour, and resume the gradient heating program after the material temperature drops below -25℃, adjusting the heating rate to 4℃ / h.

[0065] Excessive frost buildup on the condenser coil: If the frost thickness on the condenser coil is ≥5mm and the condensation effect decreases, activate the automatic defrosting program: Turn off the heating plate, maintain vacuum, raise the condenser coil temperature to 10℃, hold for 10 minutes, and after the frost melts, lower the condenser coil temperature to -40℃ to continue sublimation drying. Monitor the material temperature during defrosting to prevent the material from overheating. Sublimation rate too low: If the sublimation rate is ≤0.2kg / (m²・h), it may be due to insufficient vacuum or excessively low heating plate temperature. Increase the vacuum to 18Pa and simultaneously increase the heating plate temperature by 3℃. After adjustment, recalculate the sublimation rate after 30 minutes. If it still does not meet the standard, check the material layer thickness. If the layer is too thick, reduce the layer thickness appropriately to ≤12mm. S7304: After the anomaly handling is completed, record the anomaly situation, handling measures, and handling results in detail to form an anomaly handling log, providing a reference for subsequent process optimization. 30 minutes before the end of sublimation drying, conduct comprehensive parameter testing to ensure that all parameters meet the requirements and the material moisture content is ≤5% before proceeding to the desorption drying stage.

[0066] S740: Analysis of Drying Parameter Settings and Deep Dehydration: S7401: The core of analytical drying is to remove residual adsorbed water from the material. This type of water is difficult to remove by sublimation and requires deep dehydration through analytical action at a higher temperature and appropriate vacuum, so that the final moisture content of the material is ≤1.5%. Analytical drying parameters are set based on an intelligent knowledge base: vacuum degree 110±5Pa, heating plate temperature 32±1℃, analytical drying time 11-12h; S7402: Parameter Setting Basis: The temperature and vacuum level of the desorption drying process need to balance dehydration efficiency and product quality. Excessive temperature will cause the product to soften and melt, damaging its structure; insufficient temperature will result in low desorption efficiency, making it difficult to achieve the target moisture content; excessive vacuum will cause the material surface to lose water too quickly, leading to cracking; insufficient vacuum will hinder the escape of adsorbed water. Experiments have determined that 32±1℃ and 110±5Pa are the optimal parameter combination, resulting in the best desorption rate and product stability. S7403: Desorption Drying Execution Process: After sublimation drying is completed, the desorption drying program is started. First, the vacuum degree of the freeze dryer is adjusted from 15±2Pa to 110±5Pa at a rate of 10Pa / min to avoid sudden changes in vacuum degree that could damage the material structure. At the same time, the temperature of the heating plate is reduced from 60℃ to 32±1℃ at a rate of 5℃ / min. During the cooling process, the material temperature is monitored and kept ≤25℃ to avoid excessive temperature that could cause product deterioration. S7404: During the desorption drying process, the moisture content of the material is measured every 1 hour using a Karl Fischer moisture analyzer. The initial moisture content is about 5%, and the moisture content gradually decreases as the desorption drying proceeds. The formula for calculating the desorption rate is: v_desorption = (m1-m2) / (S×t), where m1 is the mass of material moisture before desorption (kg), m2 is the mass of material moisture at time t (kg), S is the surface area of ​​the material (m²), and t is the desorption time (h).

[0067] The desorption rate should be kept stable at 0.2-0.3 kg / (m²·h). If the desorption rate is too fast, the heating plate temperature should be reduced by 1°C. If the desorption rate is too slow, the vacuum level should be increased by 5 Pa. S7405: In the later stage of analytical drying, when the material moisture content drops below 2%, reduce the moisture detection frequency to once every 2 hours. When the moisture content is ≤1.5%, continue drying for 1 hour to ensure uniform moisture distribution and no localized excessive moisture. During analytical drying, observe the material morphology, ensuring it maintains structural integrity, without softening, cracking, or discoloration. If softening occurs, immediately reduce the heating plate temperature by 2°C and extend the drying time by 30 minutes.

[0068] S750: Analysis of Drying Process Monitoring and Endpoint Determination: S7501: The key monitoring points of the drying process are moisture content, material temperature, heating plate temperature, and vacuum degree. Real-time monitoring ensures stable parameters, accurately determines the drying endpoint, and avoids over-drying or insufficient drying. S7502: Moisture Content Monitoring: An online Karl Fischer moisture analyzer, model: Mettler DL38, with an accuracy of ±0.01%, was used. Moisture content was measured in real-time from inside the freeze dryer via a sampling tube, and data was recorded every hour to plot the moisture content change curve. When two consecutive moisture content readings are ≤1.5% and the difference is ≤0.1%, the drying endpoint is preliminarily determined. If the moisture content fluctuates significantly, drying continues for another 30 minutes, followed by another measurement, until the stabilization condition is met. S7503: Other parameter monitoring: The heating plate temperature is maintained at 32±1℃ and recorded every 10 minutes. The temperature fluctuation range is ≤±0.5℃. If it exceeds the range, the PID control parameters are adjusted. The vacuum degree is maintained at 110±5Pa and recorded every 15 minutes. If the vacuum degree deviates, it is adjusted through the vacuum regulating valve. The material temperature is detected by an infrared temperature sensor and maintained at ≤25℃. If the temperature is too high, the heating plate temperature is reduced or the vacuum degree is increased. S7504: Comprehensive Judgment of Drying Endpoint: In addition to moisture content, the final determination should also consider auxiliary indicators such as material appearance, weight change rate, and resistivity. Appearance requirements: Dense structure, no softening, no cracking, and uniform color; Weight change rate requirements: Material weight change rate ≤ 0.1% within 1 hour (weight change rate = (weight at time t - weight at time (t-1)) / weight at time (t-1) × 100%); Resistivity requirements: Material resistivity is measured using a resistance meter; when resistivity ≥ 1 × 10¹ 0 When the value reaches Ω・m, it indicates that the material is sufficiently dried and the adsorbed water has been largely removed.

[0069] When the moisture content is ≤1.5%, and the appearance is acceptable, the weight change rate is ≤0.1%, and the resistivity is ≥1×10¹, the product is suitable for use. 0When the three auxiliary indicators Ω・m are reached, the desorption drying is officially considered complete. The freeze dryer's heating and vacuum systems are then shut down, and the material is ready for discharge. After the desorption drying is completed, the drying time, final moisture content, and parameter change curves are recorded to establish a freeze-drying process quality archive.

[0070] S760: Quality Inspection and Acceptance Criteria for Freeze-dried Materials S7601: Freeze-dried materials must undergo comprehensive quality testing to ensure that all indicators meet the preset standards. The testing indicators include moisture content, temperature resistance, structural integrity, taste, color, and microbiological indicators. S7602: Moisture content detection: Using a Karl Fischer moisture analyzer, take 5 samples (5g each) from different trays and locations to test the moisture content. The average moisture content should be ≤1.5%, and the moisture content deviation of a single sample should be ≤±0.2%.

[0071] Temperature resistance test: Place the sample in a constant temperature and humidity chamber at 35℃ and 50% relative humidity for 4 hours and observe its condition. It should show no softening, melting, deformation, or oil seepage on the surface, and the weight change rate should be ≤1%. At the same time, observe it again after 5 hours. Grade 1 products should still show no softening.

[0072] Structural integrity testing: The hardness and brittleness of the product are tested using a texture analyzer, model: TA.XTPlus. The hardness requirement is ≥20N and the brittleness requirement is ≥15N. The internal structure of the product is observed by scanning electron microscopy (SEM). The pores are required to be uniform, with a porosity ≤30%, no obvious large pores, and a pore diameter >100μm.

[0073] Taste test: Sensory evaluation was conducted by 10 professional evaluators. Evaluation indicators included smoothness, fineness, and cocoa flavor intensity. Each indicator was scored from 1 to 10. A total score of ≥85 was considered qualified, and ≥90 was considered excellent.

[0074] Color detection: Use a CR-400 colorimeter to detect L*, a*, and b values. The requirements are L=28-32, a*=12-16, b*=22-26, color difference ΔE≤1.0, and uniform color.

[0075] Microbiological index testing: Refer to GB17399-2016 "National Food Safety Standard for Chocolate, Cocoa Butter Substitute Chocolate and its Products" to test for total bacterial count, coliform bacteria, mold, and yeast. The requirements are: total bacterial count ≤100 CFU / g, coliform bacteria ≤10 CFU / g, mold ≤5 CFU / g, yeast ≤5 CFU / g, and pathogenic bacteria, Salmonella, Staphylococcus aureus, and Bacillus cereus must not be detected. S7603: After all test indicators are qualified, the freeze-dried material is deemed qualified and proceeds to the next step of grading and screening. If any indicator fails to meet the standard, the cause must be analyzed and addressed: if the moisture content exceeds the standard, perform desorption and drying again for 1 hour; if the temperature resistance does not meet the standard, return to the parameter optimization process to adjust the process parameters; if the microbiological indicators fail to meet the standard, investigate the contamination links in the production process, perform cleaning and disinfection again, and then conduct production verification. Detailed test data should be recorded, and a quality traceability file should be established.

[0076] Step S800 is used for flow calculation, dynamic adjustment, and pollution trend prediction, including: S810: Grading, Screening, and Sealing Packaging: S8101: The purpose of grading and screening is to remove unqualified products and ensure that the quality of finished products is uniform. The screening equipment needs to be prepared and the parameters set according to the product characteristics. Multi-stage screening production lines should be selected, including vibrating screening machines, visual inspection machines, and weight sorting machines. The equipment should be fully debugged and calibrated before use. S8102: Preparation and Parameter Setting of Vibrating Screen: A ZS series linear vibrating screen, model ZS-1200, is selected. It has 3 screen layers with mesh sizes of 12mm, 10mm, and 8mm, used for screening products of different particle sizes. Before use, clean the screen to remove residual impurities and check for damage; replace any damaged screens immediately. Set the vibration frequency to 50Hz, amplitude to 3mm, and screening speed to 100kg / h. During screening, adjust the vibration angle by 15-20° to control the material's movement speed on the screen, ensuring thorough screening. The upper 12mm screen is used to remove oversized or agglomerated products; the middle 10mm screen filters out qualified particle sizes of 8-12mm; and the lower 8mm screen is used to remove undersized products. S8103: Visual Inspection Machine Preparation and Parameter Setting: A CCD visual inspection machine, model: MV-1000, with an inspection accuracy of 0.1mm and an inspection rate of 50 pieces / minute, is selected. It is used to inspect the shape integrity, color uniformity, surface defects, cracks, missing corners, and contamination of products. Before using the equipment, calibrate the camera focal length and light source intensity, and set the inspection parameters: allowable shape deviation ≤ 0.5mm, difference between actual product size and standard size, color deviation ΔE ≤ 1.0, and surface defect area ≤ 0.5mm². Import image templates of standard products to establish an image library of qualified products. During inspection, an image comparison algorithm automatically identifies and removes unqualified products. S8104: Preparation and Parameter Setting of the Weight Sorting Machine: A WS series weight sorting machine, model WS-200, is selected. Its weighing range is 5-20g, with an accuracy of ±0.01g. It is used to screen products that meet weight standards. The standard weight is 10g, and the acceptable weight range is set to 9.8-10.2g. Products exceeding this range are automatically rejected. Before use, calibrate the sorting machine with a standard 10g weight. The calibration formula is: Weight Error = (Sorting Machine Display Weight - Standard Weight Mass) / Standard Weight Mass × 100%. The error should be ≤ ±0.1%; otherwise, adjust the weighing sensor parameters. Set the sorting rate to 50 pieces / minute to match the speed of the vision inspection machine, ensuring continuous operation of the production line.

[0077] After the screening equipment is debugged, a trial run is conducted. 100 known qualified products are selected for testing. The requirements are that the screening efficiency of the vibrating screening machine is ≥99%, the recognition accuracy of the vision inspection machine is ≥99.5%, and the sorting accuracy of the weight sorter is ≥99.8%. If the requirements are not met, the equipment parameters are readjusted until the trial run is qualified.

[0078] S820: Multi-stage screening process execution and handling of non-conforming products: S8201: The multi-stage screening process is executed in the order of "vibration screening → visual inspection → weight sorting" to ensure that each screening process can effectively remove specific types of unqualified products. The process parameters are strictly controlled during the screening process to ensure screening quality and efficiency. S8202: First Stage: Vibration Screening: The freeze-dried material is evenly fed into the feed hopper of the vibration screen, with the feeding rate controlled at 100 kg / h to avoid insufficient screening due to excessive feeding speed. The material moves along the screen under vibration. The upper screen traps oversized or agglomerated products, the middle screen collects products of acceptable particle size, and the lower screen collects undersized products. The screen is checked every 30 minutes during screening to remove trapped non-compliant products and prevent screen blockage. Qualified products after vibration screening are conveyed to a vision inspection machine via a chute. Non-compliant products, including those that are too large, too small, or agglomerated, are collected in a dedicated container, labeled "Screening Failed," and subsequently centrally processed, either crushed and recycled as raw material or treated as industrial waste. S8203: Level 2: Visual Inspection: Qualified particle size products enter the conveyor line of the visual inspection machine at a speed of 50 pieces / minute. Each product passes through the camera's shooting area, and the camera captures three images from different angles: front, side, and top. These images are transmitted to the image processing system for comparison with a standard template. If the product's shape deviation is >0.5mm, color deviation ΔE >1.0, or surface defect area >0.5mm², the system sends a signal, and a pneumatic rejection device removes the unqualified product to a dedicated container. Qualified products continue to be conveyed to the weight sorting machine. During the visual inspection process, 10 products are randomly selected every hour, and the inspection results are manually verified to ensure an accuracy rate ≥99.5%. If a misjudgment is found, the image template and inspection parameters are recalibrated.

[0079] S8204: Level 3: Weight Sorting: Products that pass visual inspection enter the weight sorting machine. The sorting machine accurately weighs each product. If the weight is within the range of 9.8-10.2g, it is considered to be of acceptable weight and is transported to the acceptable product collection area; if the weight exceeds the range, it is automatically rejected and placed in a dedicated container. During the weight sorting process, the weighing sensor is calibrated with standard weights every 30 minutes to ensure weighing accuracy. After the grading and screening are completed, the types and proportions of non-conforming products are statistically analyzed. The proportion of oversized / undersized products is ≤1%, the proportion of shape / color defects is ≤0.5%, the proportion of weight defects is ≤0.2%, and the total defect rate is ≤2%. If the defect rate exceeds the range, the cause is analyzed and the screening parameters or the previous process are adjusted. Non-conforming products must be handled in accordance with the principles of "collection by category, clear labeling, and compliant disposal". Different types of non-conforming products should be collected separately, and the reasons for non-conformity, production date, and batch number should be labeled. After review by the quality department, recyclable products, such as those that are only non-conforming in weight but meet other indicators, should be crushed and mixed into new raw materials at a ratio of ≤10% for reprocessing. Non-recyclable products should be destroyed in accordance with environmental protection requirements. The handling process should be recorded in detail, and a non-conforming product disposal file should be established.

[0080] S830: Pre-treatment of qualified products and preparation of packaging materials: S8301: Qualified products must undergo pretreatment before packaging to remove surface impurities and static electricity. At the same time, packaging materials that meet the requirements must be prepared to ensure that the packaged products are moisture-resistant, oxidation-resistant, and have a long shelf life. S8302: Pre-treatment of qualified products: First, qualified products after grading and screening are sent to a cleanroom with a cleanliness level ≥100,000, temperature 18-22℃, and relative humidity ≤30% to avoid moisture absorption and contamination. Static electricity is removed from the product surface using an electrostatic eliminator (model: SL-001), with an ion flow rate ≥10m³ / min and an electrostatic elimination time ≤0.5s, to prevent the adsorption of dust and impurities. Then, the product surface is gently cleaned with a sterile brush to remove residual fine impurities. During the cleaning process, operators wear sterile gloves and masks to avoid human contamination. The pre-treated products are then transported to the packaging station via a sterile conveyor line made of 304 stainless steel with a smooth surface and no dead corners. The conveyor line is disinfected with 75% alcohol every 2 hours. S8303: Packaging Material Selection and Pretreatment: The packaging material is a high-barrier aluminum foil vacuum bag made of PET / AL / PE composite film with a thickness ≥120μm, oxygen permeability ≤2cm³ / (m²・24h・0.1MPa), and water vapor permeability ≤1g / (m²・24h), ensuring effective barrier against oxygen and moisture. The bag size is designed according to product specifications, with a single bag capacity of 100g, 10 bags × 10g, and a zipper and tear opening for easy use and sealing. S8304: Packaging Material Pretreatment Process: ① Cleaning: Wipe the surface of the packaging bag with a sterile cloth to remove dust and impurities; ② Disinfection: Place the packaging bag in a UV disinfection chamber and irradiate for 20 minutes at an intensity of 30 μW / cm² to ensure sterility inside the packaging bag; ③ Pre-drying: Place the disinfected packaging bag in a drying chamber at a temperature of 50℃ and a relative humidity of ≤20% for 30 minutes to remove residual moisture inside the packaging bag and prevent the product from absorbing moisture after packaging. The pretreated packaging bags should be stored in sealed containers in a clean room to prevent contamination and moisture absorption, and the storage time should not exceed 24 hours. S8305: In addition, prepare food-grade silica gel desiccant, type A, moisture absorption rate ≥30%, particle size 1-3mm, desiccant packaging is breathable non-woven fabric bag, each bag weighs 3g, 3% of the total product weight, the desiccant must be sterilized at high temperature 121℃ for 30min to ensure sterility.

[0081] S840: Optimization and Implementation of Vacuum Nitrogen Filling Sealing Process: S8401: Vacuum nitrogen-filling sealing is key to extending product shelf life. It involves removing air from the bag through vacuuming and then filling it with inert nitrogen gas to isolate oxygen and moisture, preventing oxidation and moisture absorption. A DZ series vacuum nitrogen-filling packaging machine, model DZ-600, is selected. Its vacuum range is 0-100Pa, nitrogen filling pressure range is 0.1-0.5MPa, and packaging speed is 20 bags / min. The equipment must be debugged and calibrated before use. S8402: Process Parameter Setting and Optimization: The optimal process parameters were determined through experiments: ① Vacuuming time: 30s, after vacuuming the bag the vacuum degree ≤5Pa, to ensure complete removal of air and moisture; ② Nitrogen filling time: 20s, nitrogen purity ≥99.9%, after nitrogen filling the bag the pressure is 0.02-0.03MPa (slight positive pressure), to prevent external air and moisture from seeping in; ③ Heat sealing temperature: 180±5℃, heat sealing time: 3s, heat sealing pressure: 0.3MPa, to ensure the bag opening is firmly sealed and there is no air leakage; S8403: Parameter Optimization Verification: 100 packaging bags were selected for testing. The packaging effect was measured under different vacuuming times (25s, 30s, 35s), nitrogen filling times (15s, 20s, 25s), and heat sealing temperatures (175℃, 180℃, 185℃). The test indicators included sealing strength, residual oxygen rate inside the bag, and residual water content inside the bag. The test results showed that when the vacuuming time was 30s, the nitrogen filling time was 20s, and the heat sealing temperature was 180℃, the sealing strength was 65N / 15mm, the residual oxygen rate was 0.3%, and the residual water content was 0.08g, resulting in the best packaging effect. Therefore, this set of parameters was determined to be the optimal process parameters. S8404: Sealing Process Execution Flow: ① Bagging: In a clean room, the operator places 10 qualified products (100g each) and 1 bag of silica gel desiccant (3g each) into an aluminum foil vacuum bag. The desiccant is placed in a corner of the bag to avoid direct contact with the product. ② Sealing Pre-treatment: The bag opening is placed flat into the heat sealing strip of the packaging machine, ensuring that there are no wrinkles or product residues at the bag opening, otherwise the sealing effect will be affected. ③ Vacuuming: The packaging machine is started, and vacuuming is performed for 30 seconds to reduce the vacuum degree inside the bag to ≤5Pa. ④ Nitrogen Filling: After vacuuming, food-grade nitrogen with a purity ≥99.9% is automatically filled in for 20 seconds, and the pressure inside the bag reaches 0.02-0.03MPa. ⑤ Heat Sealing: After nitrogen filling, the heat sealing strip is heated to 180±5℃, pressurized at 0.3MPa, and heat-sealed for 3 seconds to form a sealing edge with a width ≥10mm. ⑥ Cooling: After heat sealing, cooling air is used for 5 seconds to quickly solidify the sealing edge and improve the sealing strength. S8405: During the packaging process, one bag out of every 50 bags is randomly selected for sealing performance testing using the vacuum decay method (model: Lighthouse900) to test the seal strength, which must be ≥50N / 15mm. A residual oxygen meter (model: OX-100, accuracy ±0.01%) is used to test the residual oxygen rate inside the bag, which must be ≤0.5%. A Karl Fischer moisture analyzer is used to test the residual moisture content inside the bag, which must be ≤0.1g. If the tests fail, the heat sealing temperature, pressure, or time is adjusted, and the packaging is repeated.

[0082] S850: Packaging Integrity Inspection and Finished Product Labelling S8501: Packaging Integrity Inspection: A combination of "appearance inspection + sealing inspection + pressure resistance inspection" is adopted: ① Appearance inspection: The appearance of the packaging bag is manually observed. It is required that there is no damage, no wrinkles, no pinholes, the sealing edge is flat and there are no bubbles, and the printed pattern is clear. If the appearance is found to be unqualified, it is rejected immediately; ② Sealing inspection: The packaged finished product is immersed in water at 25℃. The packaging bag is pressed to slightly increase the internal pressure. It is observed whether bubbles emerge. No bubbles emerge, indicating that the seal is qualified. If bubbles are present, it indicates that there is a leak, and the finished product is rejected; ③ Pressure resistance inspection: The finished product is placed in a pressure testing machine, model: NYL-200, pressure range 0-1MPa. A pressure of 0.5MPa is applied and held for 30 seconds. If the packaging bag is not broken or deformed, it is qualified for pressure resistance, ensuring that the packaging will not be damaged due to compression during storage and transportation; S8502: Randomly select 1% of finished products from each batch for packaging integrity testing, with a minimum of 30 pieces. If the failure rate is ≥2%, the sampling ratio will be increased to 5%. If the failure rate is still ≥2%, the entire batch will be tested, and the non-conforming products will be discarded. After the testing is completed, qualified finished products will be transferred to the finished product warehouse, and non-conforming finished products will be repackaged or destroyed. S8503: Finished Product Labeling: In accordance with the requirements of GB7718-2011 "National Food Safety Standard General Rules for Labeling of Prepackaged Foods" and GB28050-2011 "National Food Safety Standard General Rules for Nutrition Labeling of Prepackaged Foods," clear labeling information shall be printed on the packaging bag, including: product name, heat-resistant freeze-dried chocolate, ingredient list (listed in order of ingredient ratio), net weight 100g, production date, shelf life 12 months, storage conditions (18-22℃, relative humidity ≤40%, store in a cool, dry place in a sealed container), producer's name, address, contact information, product standard number, food production license number, and nutrition facts table (energy, protein, fat, carbohydrates, sodium, etc.). Food-grade ink shall be used for labeling, and the printing shall be on both the front and back of the packaging bag to ensure clear readability, without blurring or misprints.

[0083] S860: Finished Product Storage Condition Control and Quality Monitoring S8601: Finished Goods Warehouse Preparation and Environmental Control: The finished goods warehouse will be a cool, dry facility constructed with thermal insulation materials and equipped with a temperature control system, a humidity control system, and a ventilation system. The storage temperature will be set at 18-22℃, with a fluctuation range of ≤±1℃. This will be controlled in real-time via central air conditioning and temperature sensors, with temperature data recorded every hour. Relative humidity will be controlled at ≤40%, controlled in real-time via dehumidifiers and humidity sensors, with humidity data recorded every hour. Shelves will be installed in the warehouse, with a shelf height ≥30cm. Finished goods will be stacked on the shelves, away from walls and the ground, at least 50cm away from walls and at least 10cm away from the ground, to prevent moisture and contamination. The warehouse will be kept well-ventilated, with ventilation for 10 minutes every 2 hours. The dehumidifier will be turned off during ventilation and restarted afterward to ensure fresh air in the warehouse. S8602: Finished Product Stacking and Batch Management: Finished products should be neatly stacked according to production batches, with an aisle of ≥30cm between each batch for easy management and monitoring. The stacking height should not exceed 3 layers to prevent deformation of the bottom products. Each batch of finished products should be labeled with a batch number, production date, quantity, and inspection status. For qualified products, a "first-in, first-out" (FIFO) management principle should be implemented to ensure that the first produced products are sold first, avoiding expiration. S8603: Quality Monitoring Mechanism: Establish a quality monitoring mechanism of "regular sampling inspection + near-expiration warning": ① Regular sampling inspection: After each batch of finished products is put into storage, 10 pieces will be randomly selected for quality testing at the 1st, 3rd, 6th, 9th and 12th months. The test indicators include moisture content, temperature resistance, taste, color and microbial indicators. The test method is the same as that for freeze-dried materials. All indicators are required to meet the standards; ② Near-expiration warning: A warning will be issued for finished products with a shelf life of less than 3 months. They will be stored separately and given priority for sale. If the quality indicators of near-expiration products are found to be substandard, sales will be stopped immediately and they will be destroyed. S8604: During storage, if the warehouse temperature or humidity exceeds the set range, immediately activate the emergency plan: if the temperature is too high, turn on the backup refrigeration equipment; if the humidity is too high, increase the dehumidifier's operating time; if the finished product shows signs of moisture absorption, softening, or deterioration, immediately isolate the batch of products, conduct comprehensive testing, analyze the cause, and take remedial measures. Detailed records of temperature and humidity during storage, along with quality testing data, must be archived to establish a finished product storage quality file, ensuring product quality traceability.

[0084] Experimental example: I. Experimental Objective: The feasibility of the technical solution of "refined raw material processing + multi-stage structural stabilization + intelligent parameter optimization + precise freeze-drying" was verified, and the effects of each process parameter on the product's temperature resistance, structural stability, taste and shelf life were clarified. Finally, freeze-dried chocolate products that meet the requirements of "stable storage at 35℃ for ≥4h, moisture content ≤1.5%, and smooth taste without graininess" were obtained.

[0085] II. Experimental Materials and Equipment Experimental materials: 20% cocoa butter, 18% cocoa liquor, low-fat chocolate powder (total fat content 20%, initial particle size 22μm), 23% white sugar, 13% trehalose, 1.0% compound emulsifier, glyceryl monostearate: polyglycerol ricinoleate = 1:1, 0.45% compound stabilizer, xanthan gum: guar gum = 2:3, 1.8% compound thickener, sodium carboxymethyl starch: pectin = 3:2, 0.4% hydroxypropyl methylcellulose, 4.5% starch, 25% food-grade purified water, 0.7% humectant, 0.25% acidity regulator, 0.15% natural cocoa flavor; food-grade silica gel desiccant (moisture absorption rate 32%), high-barrier aluminum foil vacuum bag.

[0086] Experimental equipment: Constant temperature heating tanks, colloid mills, ball mills, ultrafine pulverizers, V-type constant temperature mixers, chocolate grinding mills, chocolate refining machines, high-pressure homogenizers, intelligent programmable freezers, quick-freezing warehouses, vacuum freeze dryers, multi-stage screening production lines, vibrating screens, CCD vision inspection machines, weight sorting machines, vacuum nitrogen-filling packaging machines, laser particle size analyzers, rotational viscometers, low-temperature scanning electron microscopes, Karl Fischer moisture analyzers, constant temperature and humidity chambers, texture analyzers, colorimeters, and microbial testing equipment.

[0087] III. Experimental Procedure Refined pretreatment of raw materials: Each raw material was accurately weighed according to weight proportions, and the electronic analytical balance was calibrated with an error ≤ ±0.02%. Cocoa butter and cocoa liquor were separately placed in a constant-temperature heating tank, melted at 52℃, and then kept at 50℃. The acid value was measured to be ≤0.8 mg KOH / g and the peroxide value ≤4.2 mmol / kg. Chocolate powder was initially ground to 10 μm using a colloid mill, then finely ground to 5 μm using a ball mill, and finally ultra-fine ground to 1.8 μm using an ultra-micro pulverizer, with the temperature controlled at 45℃ throughout the process. Starch, compound thickener, and hydroxypropyl methylcellulose were added to a constant-temperature mixer at 36℃ and mixed at 300 rpm for 20 minutes. The mixing uniformity (CV) was measured to be 2.8%, yielding a clump-free premixed powder.

[0088] Preparation of ultrafine homogenized base sauce: 82% melted cocoa butter, all cocoa liquor, pretreated chocolate powder, granulated sugar, and trehalose were fed into a fine grinder. The temperature was set to 51℃ and the rotation speed to 800r / min. Grinding was continued for 5.5 hours, with particle size checked every hour. The final material fineness was 1.2μm, viscosity was 720mPa・s, and color was L*=30, a*=14, and b*=24.

[0089] Three-stage deep refining: The base sauce was transferred to a refining machine, and the remaining 18% melted cocoa butter was added. The mixture was refined at 58°C for 1.5 hours to remove bubbles and off-flavors. The temperature was then lowered to 47°C and refined for 7.5 hours. The free fatty acid content was measured to be 0.45%. Subsequently, compound emulsifiers, compound stabilizers, and other additives were added in batches, and the mixture was refined at 47°C for another 0.8 hours. The final compound sauce had a viscosity of 920 mPa·s, a pH of 5.8, and showed no stratification in the centrifugal stability test.

[0090] Two-stage high-pressure homogenization emulsification: Add purified water at 39℃ to the compound sauce and stir at 500 rpm for 15 min. The standard deviation of the moisture distribution is 0.3%. After primary homogenization at 29 MPa and secondary homogenization at 66 MPa, a chocolate emulsion is finally obtained with a particle size distribution of 0.5-1.3 μm, a viscosity of 900 mPa·s, and no stratification after standing at 60℃ for 2 h in a thermal stability test.

[0091] Intelligent program freeze forming: The emulsion was injected into a silicone mold pre-cooled to -10℃, and a three-stage cooling process was initiated: the temperature was reduced from 38℃ to -10℃ at 0.8℃ / min, to -35℃ at 1.5℃ / min, and to -50℃ at 2.5℃ / min. The core temperature of the material was measured to be -49℃, and the ice crystal size was observed to be 22μm using a cryogenic scanning electron microscope. Subsequently, the material was transferred to a -50℃ quick-freezing chamber for 3 hours, and the final temperature of the material was -42℃.

[0092] Intelligent parameter optimization and knowledge base construction: Key data such as ice crystal size and emulsion viscosity were collected and compared with the three-stage validation dataset. Under the initial parameters, the product remained stable at 35°C for 3.5 hours. Using a response surface methodology model, the secondary homogenization pressure was adjusted to 68 MPa, and the third-stage freezing and heat preservation time was adjusted to 2.5 hours. After small-batch validation production, the product remained stable at 35°C for 4.2 hours with a moisture content of 1.3%, and the objective function F=9.2, meeting the threshold requirements. These parameters were then associated with product attributes and high-temperature transportation scenarios and stored in an intelligent knowledge base, with a query response time of 0.3 seconds.

[0093] Precision gradient freeze-drying process: Frozen materials were loaded into a freeze dryer. Sublimation drying was performed with an initial vacuum of 16 Pa, and the heating plate temperature was increased from 30°C to 60°C in a gradient of 5°C / h, maintained for 13.5 hours. The material temperature was monitored in real-time to be ≤-25°C, and the vacuum was dynamically adjusted to 15 Pa. The moisture content after sublimation was 4.8%. Desorption drying was performed with an adjusted vacuum of 112 Pa and a heating plate temperature of 32°C, dried for 11.5 hours. The final moisture content was 1.3%, and the resistivity was 1.2 × 10¹. 0 Ω・m is used to determine the drying endpoint.

[0094] Grading, screening, and sealing packaging: After grading and screening using a vibration screening machine, CCD vision inspection machine, and weight sorting machine, the overall failure rate was 1.8%. Qualified products were placed in aluminum foil vacuum bags with 3g of silica gel desiccant, then vacuum-packed with nitrogen. The sealing strength was tested at 65N / 15mm, and the residual oxygen rate inside the bag was 0.3%. Finished products were stored in a warehouse at 20℃ and 35% humidity, implementing batch management and the "first-in, first-out" principle.

[0095] IV. Experimental Data Key process data: Raw material processing: The final particle size of the chocolate powder is 1.8μm, and the uniformity of the premixed powder is CV=2.8%; Basic sauce: fineness after grinding is 1.2μm, viscosity is 720mPa・s; Compound sauce: after refining, viscosity is 920 mPa·s, pH=5.8, and free fatty acid content is 0.45%; Emulsion: After two-stage homogenization, the particle size is 0.5-1.3 μm and the viscosity is 900 mPa·s; Freezing process: core temperature -49℃, ice crystal size 22μm, temperature after quick-freezing -42℃; Freeze-drying process: sublimation drying for 13.5 h, desorption drying for 11.5 h, final moisture content 1.3%; Packaging inspection: sealing strength 65N / 15mm, residual oxygen rate 0.3%, residual water content 0.08g.

[0096] Finished product quality inspection data: Physicochemical properties: Moisture content 1.3%, temperature resistance 35℃ and humidity 50% environment for 4.5h without softening or melting, weight change rate 0.8%; hardness 23N, brittleness 17N, porosity 28%; color L*=31, a*=15, b*=25, color difference ΔE=0.8; Sensory evaluation: Score by 10 professional evaluators: Smoothness 9.2, Fineness 9.0, Cocoa flavor richness 9.3, Total score 88.5; Microbiological indicators: total bacterial count 82 CFU / g, coliform bacteria 6 CFU / g, mold 3 CFU / g, yeast 2 CFU / g, pathogenic bacteria not detected; Stability test: After 12 months of sealed storage at 25℃, the moisture content was 1.4%. The temperature resistance was no abnormality after being placed at 35℃ for 4 hours, and there was no moisture absorption or deterioration.

[0097] V. Experimental Conclusions This experiment strictly followed the technical plan throughout the entire process, ensuring controllable process parameters and stable data. Three-stage grinding refined the chocolate powder to 1.8μm, two-stage homogenization yielded a narrow particle size distribution emulsion, and three-stage programmed freezing precisely controlled the ice crystal size to 22μm. Combined with gradient freeze-drying and vacuum nitrogen-filled packaging, the final product met all preset standards. The finished product had a moisture content of 1.3% ≤ 1.5%, remained stable at 35℃ for 4.5h ≥ 4h, achieved a taste score of 88.5 points ≥ 85 points, met the microbiological standards of GB17399-2016, and maintained stable quality throughout a 12-month storage period.

[0098] Experiments verified the effectiveness of the "refined raw material processing - multi-stage structural stabilization - intelligent parameter optimization - precision freeze-drying" technical system. Specifically, three-stage grinding, three-stage refining, and two-stage homogenization synergistically improved the product's structural density; intelligent programmed freezing and gradient freeze-drying ensured temperature resistance and moisture control; intelligent parameter optimization significantly improved production stability; and vacuum nitrogen-filled packaging extended shelf life. This technical solution enables large-scale production, and the product outperforms traditional processes in terms of temperature resistance, taste, and shelf life, making it suitable for diverse scenarios such as room temperature storage and high-temperature transportation.

[0099] Example 2: A freeze-dried chocolate that is heat-resistant and does not easily melt, a method for making a freeze-dried chocolate that is heat-resistant and does not easily melt, comprising the following modules; It is made by the following method: take raw materials in a specific ratio according to weight, and combine emulsifier, stabilizer and thickener in a specific ratio. Cocoa butter and cocoa liquor are heated to 52±1℃ to melt and keep warm. Chocolate powder is ground three times to ≤2μm and the temperature is controlled at 45℃. Starch, compound thickener and hydroxypropyl methylcellulose are mixed in a constant temperature mixer at 35±2℃ at 300r / min for 20min to obtain premixed powder. 80%-85% melted cocoa butter, all cocoa liquor, and pretreated raw materials are fed into a fine grinding mill and ground at 50±2℃ and 800r / min for 5-6 hours until the fineness is ≤1.5μm. After three-stage refining and two-stage homogenization, an emulsion with specific indicators is obtained. The emulsion was injected into a pre-cooled mold, and after a three-stage frozen process, it was transferred to a -50℃ quick-freezing chamber for 3 hours. Parameters were optimized by comparing the three-level verification data and an intelligent knowledge base was built. The material was subjected to gradient sublimation drying and analytical drying until the moisture content was ≤1.5%. After screening qualified products, the material was vacuum-sealed with nitrogen and stored in an environment of 18-22℃ and humidity ≤40%.

[0100] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A freeze-dried chocolate that is heat-resistant and does not easily melt, characterized in that, include, It is made by the following method: take raw materials in a specific ratio according to weight, and combine emulsifier, stabilizer and thickener in a specific ratio. Cocoa butter and cocoa liquor are heated to 52±1℃ to melt and keep warm. Chocolate powder is ground three times to ≤2μm and the temperature is controlled at 45℃. Starch, compound thickener and hydroxypropyl methylcellulose are mixed in a constant temperature mixer at 35±2℃ at 300r / min for 20min to obtain premixed powder. 80%-85% melted cocoa butter, all cocoa liquor, and pretreated raw materials are fed into a fine grinding mill and ground at 50±2℃ and 800r / min for 5-6 hours until the fineness is ≤1.5μm. After three-stage refining and two-stage homogenization, an emulsion with specific indicators is obtained. The emulsion was injected into a pre-cooled mold, and after a three-stage frozen process, it was transferred to a -50℃ quick-freezing chamber for 3 hours. Parameters were optimized by comparing the three-level verification data and an intelligent knowledge base was built. The material was subjected to gradient sublimation drying and analytical drying until the moisture content was ≤1.5%. After screening qualified products, the material was vacuum-sealed with nitrogen and stored in an environment of 18-22℃ and humidity ≤40%.

2. The freeze-dried chocolate with heat resistance and minimal melting according to claim 1, characterized in that: The compound emulsifier is composed of glyceryl monostearate and polyglycerol ricinoleate in a 1:1 ratio, the compound stabilizer is composed of xanthan gum and guar gum in a 2:3 ratio, and the compound thickener is composed of sodium carboxymethyl starch and pectin in a 3:2 ratio. The weight percentages of each compound additive are 0.9%-1.1%, 0.4%-0.5%, and 1.5%-2.0%, respectively.

3. The freeze-dried chocolate with heat resistance and minimal melting according to claim 1, characterized in that: The two-stage homogenization involves primary homogenization at 28±2 MPa followed by secondary homogenization at 65±3 MPa, resulting in an emulsion with a particle size distribution of 0.3-1.5 μm and a viscosity of 800-1000 mPa・s. The three-stage programmed freezing involves lowering the temperature from 38℃ to -10℃ at 0.8℃ / min and holding for 1 hour, then lowering it to -35℃ at 1.5℃ / min and holding for 3 hours, and finally lowering it to -50℃ at 2.5℃ / min and holding for 2 hours, with the material's core temperature ≤ -48℃.

4. The freeze-dried chocolate that is heat-resistant and does not easily melt according to claim 1, characterized in that: The gradient sublimation drying is carried out under a vacuum of 15±2Pa, with the heating plate increasing the temperature from 30℃ to 60℃ at a rate of 5℃ / h for 13-14 hours; the desorption drying is carried out under a vacuum of 110±5Pa and a temperature of 32±1℃ for 11-12 hours; the vacuum nitrogen-filled sealing is carried out by evacuating to ≤5Pa and then filling with nitrogen gas with a purity of ≥99.9%, and the packaging contains 3% silica gel desiccant by weight of the product.

5. A method for producing a heat-resistant, non-melting freeze-dried chocolate, used to achieve the heat-resistant, non-melting freeze-dried chocolate as described in any one of claims 1-4, characterized in that, Includes the following features: S100: Take raw materials in a specific ratio according to weight, and combine emulsifiers, stabilizers and thickeners in a specific ratio; heat cocoa butter and cocoa liquor to 52±1℃ to melt and keep warm, grind chocolate powder three times to ≤2μm and control the temperature at 45℃, and mix starch, compound thickener and hydroxypropyl methylcellulose in a constant temperature mixer at 35±2℃ at 300r / min for 20min to obtain premixed powder; S200: Add 80%-85% melted cocoa butter, all cocoa liquor, pre-treated chocolate powder, white sugar, and trehalose to a fine grinder and grind for 5-6 hours at 50±2℃ and 800r / min until the fineness of the material is ≤1.5μm to obtain a homogenized basic sauce. S300: Transfer the base sauce to the refining machine, add the remaining melted cocoa butter, refine at 58±1℃ for 1.5h, then cool to 47±1℃ and refine for 7-8h, finally keep at 47℃ and add various additives in batches, continue refining for 0.8h to form a stable oil-polysaccharide complex sauce. S400: Add purified water at 38±2℃ to the compound sauce, stir at 500r / min for 15min, and then homogenize at 28±2MPa for the first stage and 65±3MPa for the second stage to obtain a chocolate emulsion with a particle size of 0.3-1.5μm and a viscosity of 800-1000mPa・s. S500: Inject the emulsion into a mold pre-cooled to -10℃, and cool and freeze it according to a three-stage procedure to ensure that the center temperature is ≤-48℃ and the ice crystal size is 10-30μm. After freezing, separate the material and transfer it to a -50℃ quick-freezing chamber for 3 hours. S600: Compare the three-level verification data, adjust the parameters accordingly, repeatedly optimize the verification process until the standard is met, store core data and models, and build an intelligent knowledge base that maps tasks, attributes, scenarios and solutions. S700: The material is fed into the freeze dryer and first sublimated and dried at a vacuum of 15±2Pa and a temperature gradient of 30-60℃ for 13-14 hours. Then, it is adjusted to a vacuum of 110±5Pa and a temperature gradient of 32±1℃ for 11-12 hours of desorption and drying to achieve a moisture content of ≤1.5%. S800: Select qualified products, vacuum them to ≤5Pa using aluminum foil vacuum bags, seal them with nitrogen, place them in silica gel desiccant packs, and store them in an environment of 18-22℃ and ≤40% humidity.

Images