Method for controlling quality of highland barley biscuits based on extension factor

Abstract

The application discloses a method for controlling the quality of highland barley biscuits based on an extensibility factor, and belongs to the technical field of food processing. The method is used for solving the technical problems that the extensibility of highland barley biscuits is unstable and the product consistency is poor due to the quality fluctuation of raw materials. The method is characterized in that the quality indexes of raw highland barley powder are rapidly detected and input into a multiple regression model to predict a theoretical extensibility factor. The addition amount of emulsifiers or dietary fibers is dynamically adjusted according to the deviation of the predicted value from a target range. The best baking parameters are matched from a pre-set process parameter database according to the peak viscosity and total starch content of the raw materials. The actual extensibility factor is determined after dough preparation and baking. The accumulated training data are continuously optimized and adjusted to make the deviation of the actual extensibility factor tend to be minimum. The method is mainly used for accurate quality control of highland barley biscuits and realizes standardized production.

Description

Technical Field

[0001] This invention relates to the field of food processing technology. More specifically, this invention relates to a method for controlling the quality of highland barley biscuits based on extensibility factors. Background Technology

[0002] Barley, also known as naked barley, is the most distinctive and advantageous crop in the Qinghai-Tibet Plateau region and a staple food for the Tibetan people. It not only possesses biological characteristics such as cold resistance and tolerance to poor soil, but is also rich in protein, dietary fiber, vitamins, and unique functional components such as β-glucan, earning it the reputation of being a model of a nutritious grain that is high in protein, fiber, and vitamins, and low in fat and sugar. With the increasing demand for healthy diets, barley has transformed from a regional staple food to a widely popular raw material for nutritious and health-promoting foods. Its processing and utilization, especially the development of mass-market food products, has become key to driving industrial upgrading and increasing farmers' income.

[0003] Among various barley food products, biscuits have become an important consumer form due to their easy consumption, long shelf life, and portability. One of the core quality indicators of biscuits is their extensibility, usually quantified by the spread factor, which is the ratio of the diameter to the thickness of the biscuit after baking. This indicator directly determines the sensory appearance, crispness, and texture of the biscuit. However, the industrial production of barley biscuits faces a fundamental technical bottleneck: barley protein lacks the key components (glutenin and prolamins) needed to form the gluten network. This means that barley flour cannot form a dough with excellent viscoelasticity and extensibility when mixed with water, unlike wheat flour. Consequently, barley biscuits commonly suffer from insufficient extensibility, brittle texture, and irregular shape, severely affecting the sensory quality and commercial value of the product.

[0004] To improve the quality of barley biscuits, existing technologies primarily attempt solutions at two levels. First, at the formulation level, exogenous ingredients such as wheat flour, various hydrocolloids (e.g., guar gum, xanthan gum), and emulsifiers are added to simulate the gluten network or improve dough properties. However, this method is typically based on fixed formulation ratios and cannot adapt to the natural fluctuations in the quality (e.g., starch characteristics, protein composition, particle hardness) of barley raw materials from different origins and varieties. When raw material batches change, biscuits produced with a fixed formulation exhibit significant quality differences and poor stability. Second, at the processing level, relatively fixed baking temperatures and times are commonly used. This ignores the sensitivity of the processing to differences in raw material quality. For example, barley starches with different peak viscosities require different temperatures for gelatinization and setting. A fixed process cannot achieve optimal treatment for all raw materials, potentially leading to products that are too dry, over-scorched, or insufficiently extensible.

[0005] A deeper problem lies in the current lack of a scientific and quantitative quality prediction and process control system for the highland barley processing industry, tailored to the specific characteristics of its raw materials. The quality evaluation of raw materials largely follows or references wheat standards, failing to establish a key quality indicator system directly related to the final edible quality of highland barley biscuits (such as extensibility). Production decisions heavily rely on the personal experience of technical personnel, making it impossible to effectively predict the quality of the finished product before raw material input, or to make precise formula and process interventions based on real-time prediction results during production. The existing model of post-production testing and experience-based adjustments leads to significant uncertainty in the production process, high defect rates, and difficulty in ensuring quality consistency, severely hindering the large-scale, standardized, and high-quality development of highland barley biscuits and other highland barley flour products.

[0006] In summary, the processing of highland barley food products still faces the following shortcomings: First, there is a lack of rapid detection methods and mathematical models capable of accurately predicting the processing suitability of highland barley raw materials (especially their extensibility potential); second, there is a lack of scientific rules for dynamically and quantitatively adjusting formulas based on raw material prediction results, leading to an over-reliance on qualitative experience; third, there is a lack of a decision database that links key raw material properties with optimal processing parameters, resulting in rigid process settings; and fourth, the entire production process fails to form a closed loop of prediction-adjustment-verification-optimization, hindering the system's self-learning and continuous improvement. These shortcomings collectively result in highland barley biscuit production remaining in a state of unstable quality, unclear processing adaptability, and heavy reliance on manual experience, urgently requiring a precise quality control method that intelligently couples raw material characteristics, formula engineering, and processing technology. Summary of the Invention

[0007] One object of the present invention is to solve at least the above-mentioned problems and to provide at least the advantages that will be described later.

[0008] Another objective of this invention is to provide a method for controlling the quality of highland barley biscuits based on extensibility factors. This method rapidly detects and predicts the extensibility potential of raw materials, and dynamically adjusts the formula and process accordingly. It also continuously optimizes and adjusts the rules by combining production feedback data, thereby achieving precise and stable control of the extensibility of highland barley biscuits and effectively solving the problem of inconsistent product quality caused by raw material fluctuations.

[0009] To achieve these objectives and other advantages according to the present invention, a method for controlling the quality of barley biscuits based on the extensibility factor is provided, comprising: S1. Based on the mapping relationship between the quality indicators of raw highland barley flour and the extensibility factor of biscuits made from raw highland barley flour under the basic formula, a multiple regression model is established. The quality indicators of raw highland barley flour are obtained through rapid detection methods. The data are input into the multiple regression model to predict the theoretical extensibility factor of biscuits made from raw highland barley flour under the standard basic formula. The quality indicators include the grain brightness value L*, total starch content, globulin content, peak viscosity and ash content of raw highland barley flour. S2. Adjust the basic formula based on the deviation between the theoretical spread factor prediction value of the raw material barley flour and the preset spread factor target range; if the theoretical spread factor prediction value is lower than the lower limit of the preset spread factor target range, determine the increase in emulsifier addition according to the first pre-adjustment rule; if the theoretical spread factor prediction value is higher than the upper limit of the preset spread factor target range, determine the increase in dietary fiber addition according to the second pre-adjustment rule. S3. Based on the peak viscosity and total starch content of the raw material barley flour, match the corresponding optimal baking parameter combination from the preset process parameter database, and use the formula pre-adjusted in step S2 to prepare and bake the dough to obtain barley biscuits. The optimal baking parameter combination includes at least the top heat temperature, bottom heat temperature and baking time. S4. Measure the actual extensibility factor of the barley biscuits obtained in step S3, and calculate the deviation between it and the median of the preset extensibility factor target range as the extensibility factor deviation value M. Take the theoretical extensibility factor prediction value of the current batch, the pre-adjustment scheme adopted, and the resulting extensibility factor deviation value as a set of training data. When the accumulated amount of the training data reaches a preset threshold, use multiple sets of the training data to optimize and update the first pre-adjustment rule and / or the second pre-adjustment rule in step S2, so that the adjustment scheme determined based on the optimized rule can minimize the actual extensibility factor deviation.

[0010] Preferably, the rapid detection method described in step S1 includes: The grain brightness value L* of the raw barley flour was determined using a colorimeter; the total starch content and globulin content of the raw barley flour were determined using a near-infrared spectroscopy analyzer; the peak viscosity of the raw barley flour was determined using a viscosity analyzer; and the ash content of the raw barley flour was determined using the high-temperature ignition gravimetric method.

[0011] Preferably, the standard basic formula in step S1 is based on 100 parts by weight of raw barley flour and also includes: 15-30 parts water, 15-25 parts oil, 10-30 parts sugar, 5-15 parts eggs, 0.5-2 parts leavening agent, 0.5-1.5 parts salt, D parts emulsifier and W parts dietary fiber, wherein D is 5-12 parts and W is 3-6 parts.

[0012] Preferably, in step S2, the first pre-adjustment rule is: the incremental addition of emulsifier is D×k1×ΔE; the second pre-adjustment rule is: the incremental addition of dietary fiber is W×k2×ΔF; where ΔE is the difference between the lower limit of the preset target range of the stretching factor and the predicted value of the theoretical stretching factor; ΔF is the difference between the predicted value of the theoretical stretching factor and the upper limit of the preset target range of the stretching factor; k1 takes the value of 0.05~0.15, and k2 takes the value of 0.05~0.15.

[0013] Preferably, the pre-set process parameter database in step S3 is constructed through the following steps: A1. Select various raw barley flour samples with different peak viscosities and total starch contents; A2. Using the standard basic formula, each raw material barley flour sample was tested for biscuit preparation under multiple different combinations of baking parameters. Each combination of baking parameters included a set of top heat temperature, bottom heat temperature and baking time. A3. Determine the extensibility of the cookies produced under each combination of baking parameters; A4. For each raw material barley flour sample, select the baking parameter combination from the multiple different baking parameter combinations that makes the extensibility factor of the biscuits made closest to the midpoint of the preset extensibility factor target range, and take it as the optimal baking parameter combination corresponding to the raw material barley flour sample. A5. Establish a mapping relationship between the peak viscosity and total starch content of the raw material highland barley flour sample and its corresponding optimal baking parameter combination, and store this mapping relationship as the preset process parameter database.

[0014] Preferably, in step S4, the first pre-adjustment rule and / or the second pre-adjustment rule are optimized and updated using multiple sets of training data, specifically including: The least squares method is used to perform linear regression on (ΔE, ΔE') in multiple sets of training data, and the value of k1 is updated according to its linear regression coefficient a1; the linear regression on (ΔF, ΔF') in multiple sets of training data is performed, and the value of k2 is updated according to its linear regression coefficient b1. Wherein, ΔE' is the theoretical equivalent value of emulsifier addition increment required to achieve the extension factor target, which is derived from the deviation value of the extension factor; ΔF' is the theoretical equivalent value of dietary fiber addition increment required to achieve the extension factor target, which is derived from the deviation value of the extension factor.

[0015] Preferably, ΔE'=M / P1; ΔF'=M / P2; P1 is the emulsifier efficacy coefficient, which represents the average impact of a unit amount of emulsifier added on the spreadability factor based on the standard basic formula; P2 is the dietary fiber efficacy coefficient, which represents the average impact of a unit amount of dietary fiber added on the spreadability factor based on the standard basic formula.

[0016] Preferably, based on the standard basic formula, the amount of emulsifier or dietary fiber added is changed in a gradient manner to prepare and measure the extensibility factor of the corresponding biscuits. The average change in extensibility factor caused by a unit amount of addition is obtained by linear fitting, and the values ​​of P1 and P2 are obtained.

[0017] Preferably, the final updated k1* is determined using the linear regression coefficient a1 according to the following rules: First, determine the reference value k1' based on the linear regression coefficient a1: If 0.05 ≤ a1 ≤ 0.2, then k1' = a1; If a1 < 0.05, then k1' = 0.05 + λ(0.05 - a1); If a1 > 0.2, then k1' = 0.2 + λ(0.2 - a1); Then, based on the reference value k1', determine the updated k1*: k1*=η1·k1+(1-η1)·k1'; The final updated k2* is determined using the linear regression coefficient b1 according to the following rules: First, determine the reference value k2' based on the linear regression coefficient b1: If 0.05 ≤ b1 ≤ 0.2, then k2' = b1; If b1 < 0.05, then k2' = 0.05 + μ(0.05 - b1); If b1 > 0.2, then k2' = 0.2 + μ(0.2 - b1); Then, based on the reference value k2', determine the updated k2*: k2* = η2·k2 + (1-η2)·k2'; Where λ and μ are both boundary attenuation coefficients between 0.1 and 0.5; η1 and η2 are both smoothing coefficients between 0 and 1.

[0018] The present invention has at least the following beneficial effects: Firstly, this invention fundamentally solves the problem of unstable biscuit extensibility caused by fluctuations in barley raw materials by constructing a closed-loop control system that integrates rapid prediction, dynamic adjustment, precise process, and feedback learning. It achieves a leap from relying on human experience to data-driven intelligent control, significantly improving the consistency of product quality. Secondly, by quantifying adjustment rules and scientifically constructing a process database, this invention provides clear and repeatable calculation basis for both formula adjustment and process matching, overcoming the shortcomings of blind adjustment and rigid processes in traditional methods, and greatly improving the accuracy and reliability of production control. Thirdly, this invention can continuously optimize and adjust rules using production data, especially the coefficient update mechanism with boundary constraints and smoothing processing, which ensures that the self-learning process can not only make the control strategy continuously approach the optimal, but also effectively prevent over-adjustment, thus guaranteeing the stability and robustness of long-term operation.

[0019] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Detailed Implementation

[0020] The present invention will now be described in further detail with reference to specific embodiments, so that those skilled in the art can implement it based on the description.

[0021] It should be understood that terms such as “having,” “comprising,” and “including” as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0022] Establishing a multiple regression model: Fifty samples of highland barley flour of different qualities were prepared. The powder brightness value L* of each sample was measured using a colorimeter. The total starch content and globulin content of each sample were measured using a near-infrared spectrometer. The peak viscosity of each sample was measured using a rapid viscosity analyzer. The ash content of each sample was determined using the high-temperature ignition gravimetric method specified in GB5009.4-2016.

[0023] Using a standard basic formula (based on 100 parts by weight of barley flour: 22 parts water, 20 parts oil, 20 parts sugar, 10 parts eggs, 1 part leavening agent, 1 part salt, 8 parts emulsifier, and 4 parts dietary fiber), each barley flour sample was made into cookies under the optimal baking parameters corresponding to the standard basic formula, and their extensibility was measured.

[0024] Using five quality indicators (L*, total starch content, globulin content, peak viscosity, and ash content) measured for each sample as independent variables and the corresponding biscuit spread factor Y as the dependent variable, multiple linear regression analysis was performed using SPSS software to establish the following multiple regression model: Y = 5.12 + 0.08 × L* - 0.05 × total starch content + 0.15 × globulin content + 0.002 × peak viscosity - 1.2 × ash content. The coefficient of determination R of the regression model was [value missing]. 2 =0.96.

[0025] Construction of the process parameter database: The aforementioned 50 barley flour samples covered different peak viscosities (200–800 mPa·s) and total starch contents (65%–75%). Each sample was baked using the aforementioned standard base formula at various temperatures (top / bottom heat combinations: top heat 170–190℃, bottom heat 170–190℃) and baking times (12–18 min), and the extensibility factor of the biscuits was measured under each condition. For each sample, the process parameters that brought its extensibility factor closest to 9.0 were selected as its optimal parameter combination. Finally, a database of correspondences was established, with (peak viscosity, total starch content) as input and (optimal top heat temperature, optimal bottom heat temperature, optimal baking time) as output.

[0026] Example 1 For the new batch of raw barley flour, the above-mentioned rapid detection method was used to obtain the values ​​of its five quality indicators. The values ​​were then substituted into the above-mentioned multiple regression model to obtain the theoretical extensibility factor prediction value of the new batch of raw barley flour under the standard basic formula of 8.2 (the preset target range of the extensibility factor of barley biscuits is 9.0±0.5, that is, the lower limit is 8.5, the upper limit is 9.5, and the median is 9.0).

[0027] The predicted value of the theoretical extension factor is lower than the lower limit of the target range of the extension factor, and the deviation between the two is ΔE=0.3. According to the first pre-adjustment rule, an emulsifier needs to be added, and the incremental amount of emulsifier is 0.24 parts (k1 is 0.1). The adjusted formula is: 100 parts highland barley flour, 22 parts water, 20 parts oil, 20 parts sugar, 10 parts egg, 1 part leavening agent, 1 part salt, 8.24 parts emulsifier, and 4 parts dietary fiber.

[0028] For this batch of raw material, highland barley flour, the peak viscosity was measured to be 450 mPa·s, and the total starch content was 70%. A database search revealed the closest pre-defined optimal process parameters: top heat 185℃, bottom heat 180℃, and baking time 15 min. The dough was prepared, shaped, and baked according to these parameters to produce highland barley biscuits. The actual extensibility factor of this batch of biscuits was measured to be 8.9, which meets the pre-defined target range for highland barley biscuits, indicating the product is qualified.

[0029] Example 2 Based on Example 1, for this batch of raw material, highland barley flour, after baking 8 groups, the emulsifier efficacy coefficient P1 was measured to be 0.5 through preliminary experiments (i.e., for every 1 part of emulsifier added, the spread factor increased by an average of 0.5). Therefore, the spread factor deviation value M of Example 1 is -0.1. ΔE' = M / P1 = (-0.1) / 0.5 = -0.2. This value means that, based on the actual deviation, theoretically, if the emulsifier increment is reduced by 0.2 parts (i.e., the equivalent ΔE decreases by 0.2), the result may be closer to the target. Least squares linear regression was performed on (ΔE, ΔE') in the 30 datasets, yielding a regression coefficient a1 = -0.18. Since a1 = -0.18 < 0.05, λ = 0.3 was taken, and k1' = 0.05 + 0.3 * (0.05 - (-0.18)) = 0.119. With a smoothing coefficient η1 = 0.8 and the current k1 = 0.10, the updated k1* = 0.8 * 0.10 + (1 - 0.8) * 0.119 = 0.1038. The k1 value is then updated to 0.104 for subsequent batch adjustments, thus achieving rule self-optimization. Similarly, k2 can be periodically updated.

[0030] Comparative Example Production is carried out using a fixed formula and experience-based adjustment methods commonly used in the industry.

[0031] The industry-standard barley biscuit recipe is as follows: 100 parts barley flour, 25 parts water, 22 parts oil, 25 parts sugar, 8 parts eggs, 1.5 parts leavening agent, 1 part salt, 9 parts emulsifier, and 3 parts dietary fiber. The baking parameters are: top heat 190℃, bottom heat 190℃, and baking time 16 minutes. The resulting biscuits had an actual extensibility factor of 7.8, were noticeably hard, and had a rough texture, failing to meet the target range (9.0±0.5), and were therefore substandard. Based on the previous batch results and experience, the technicians decided to add more oil and emulsifier. The oil content was adjusted to 24 parts, and the emulsifier to 10.5 parts, and production was restarted. After the adjustment, the extensibility factor of the biscuits was 9.4, falling within the target range.

[0032] Using the methods of Example 1 and the comparative example, 10 batches of barley raw materials from different sources were continuously produced to investigate the pass rate of the extension factor and the coefficient of variation of the extension factor. The results are shown in Table 1.

[0033] Table 1. Pass rate and coefficient of variation of extensibility factor for highland barley biscuits Evaluation indicators Example 1 Comparative Example Expansion factor pass rate 100% (all 10 batches fell within the target range) 40% (only 4 batches fell within the target range) Extension factor coefficient of variation 2.1% 15.7% As shown in Table 1, the method of this invention achieved a 100% pass rate for the extensibility factor in 10 consecutive batches of highland barley flour production, with all batches meeting the preset quality targets. In contrast, the compliance rate of the comparative method was only 40%, with most batches failing to meet the extensibility standards. Furthermore, the coefficient of variation for the extensibility factor of the method of this invention was only 2.1%, significantly lower than the 15.7% of the comparative method, indicating higher batch-to-batch stability of the products obtained by this method. This is because the invention uses a multiple regression model to predict the extensibility potential of raw materials and combines pre-adjustment rules with a process database to achieve precise control of extensibility, solving the quality fluctuation problem caused by batch-to-batch differences in raw materials. Moreover, the data-driven closed-loop control of this invention makes the formulation and process adjustments repeatable and scientific, significantly reducing the risk of human error.

[0034] Twenty batches of highland barley flour samples (with significant quality differences) were selected and tested in groups. Group A used the method of Example 1, with initial k1=0.1 and k2=0.1, and the formula was adjusted according to a fixed rule without updating the coefficients. Group B used the method of Example 2, with initial k1=0.1 and k2=0.1. After producing 8 batches, k1 and k2 were updated using accumulated data. The absolute deviation of the extensibility factor of each batch of biscuits from the target median, the average value of the extensibility factor of all batches, and the coefficient of variation were evaluated, as well as the number of batches required for the system to reach stability (deviation ≤0.2 for 3 consecutive batches). The results showed that Group B reached a stable state after 16 batches, and the coefficient of variation of the extensibility factor of all batches was lower than that of Group A. The average deviation of Group B was significantly smaller than that of Group A. This indicates that the dynamic optimization and adjustment method of Example 2 not only improved the accuracy of single-batch control but also achieved higher stability and applicability in long-term production.

[0035] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and the specific embodiments shown and described herein.

Claims

1. A method for controlling the quality of highland barley biscuits based on extensibility factor, characterized in that, include: S1. Based on the mapping relationship between the quality indicators of raw highland barley flour and the extensibility factor of biscuits made from raw highland barley flour under the basic formula, a multiple regression model is established. The quality indicators of raw highland barley flour are obtained through rapid detection methods. The data are input into the multiple regression model to predict the theoretical extensibility factor of biscuits made from raw highland barley flour under the standard basic formula. The quality indicators include the grain brightness value L*, total starch content, globulin content, peak viscosity and ash content of raw highland barley flour. S2. Adjust the basic formula based on the deviation between the theoretical extensibility factor prediction value of the raw material barley flour and the preset extensibility factor target range; if the theoretical extensibility factor prediction value is lower than the lower limit of the preset extensibility factor target range, determine the incremental amount of emulsifier added according to the first pre-adjustment rule. If the theoretical expansion factor prediction value is higher than the upper limit of the preset expansion factor target range, then the incremental amount of dietary fiber added is determined according to the second pre-adjustment rule. S3. Based on the peak viscosity and total starch content of the raw material barley flour, match the corresponding optimal baking parameter combination from the preset process parameter database, and use the formula pre-adjusted in step S2 to prepare and bake the dough to obtain barley biscuits. The optimal baking parameter combination includes at least the top heat temperature, bottom heat temperature and baking time. S4. Measure the actual extensibility factor of the highland barley biscuit obtained in step S3, and calculate the deviation between it and the median value of the preset extensibility factor target range as the extensibility factor deviation value M. The theoretical extension factor prediction value of the current batch, the pre-adjustment scheme adopted, and the resulting extension factor deviation value are used as a set of training data. When the accumulated amount of the training data reaches a preset threshold, the first pre-adjustment rule and / or the second pre-adjustment rule in step S2 are optimized and updated using multiple sets of the training data, so that the adjustment scheme determined based on the optimized rule can minimize the actual extension factor deviation.

2. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 1, characterized in that, The rapid detection method described in step S1 includes: The grain brightness value L* of the raw barley flour was determined using a colorimeter; the total starch content and globulin content of the raw barley flour were determined using a near-infrared spectroscopy analyzer; the peak viscosity of the raw barley flour was determined using a viscosity analyzer; and the ash content of the raw barley flour was determined using the high-temperature ignition gravimetric method.

3. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 2, characterized in that, The standard basic formula in step S1 is based on 100 parts by weight of raw barley flour, and also includes: 15-30 parts water, 15-25 parts oil, 10-30 parts sugar, 5-15 parts eggs, 0.5-2 parts leavening agent, 0.5-1.5 parts salt, D parts emulsifier, and W parts dietary fiber, where D is 5-12 parts and W is 3-6 parts.

4. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 3, characterized in that, In step S2, the first pre-adjustment rule is: the incremental addition of emulsifier is D×k1×ΔE; the second pre-adjustment rule is: the incremental addition of dietary fiber is W×k2×ΔF; where ΔE is the difference between the lower limit of the preset target range of the stretching factor and the predicted value of the theoretical stretching factor; ΔF is the difference between the predicted value of the theoretical stretching factor and the upper limit of the preset target range of the stretching factor; k1 takes a value of 0.05~0.15, and k2 takes a value of 0.05~0.

15.

5. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 4, characterized in that, The pre-set process parameter database in step S3 is constructed through the following steps: A1. Select various raw barley flour samples with different peak viscosities and total starch contents; A2. Using the standard basic formula, each raw material barley flour sample was tested for biscuit preparation under multiple different combinations of baking parameters. Each combination of baking parameters included a set of top heat temperature, bottom heat temperature and baking time. A3. Determine the extensibility of the cookies produced under each combination of baking parameters; A4. For each raw material barley flour sample, select the baking parameter combination from the multiple different baking parameter combinations that makes the extensibility factor of the biscuits made closest to the midpoint of the preset extensibility factor target range, and take it as the optimal baking parameter combination corresponding to the raw material barley flour sample. A5. Establish a mapping relationship between the peak viscosity and total starch content of the raw material highland barley flour sample and its corresponding optimal baking parameter combination, and store this mapping relationship as the preset process parameter database.

6. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 5, characterized in that, In step S4, the first pre-adjustment rule and / or the second pre-adjustment rule are optimized and updated using multiple sets of training data, specifically including: The least squares method is used to perform linear regression on (ΔE, ΔE') in multiple sets of training data, and the value of k1 is updated according to its linear regression coefficient a1; the linear regression on (ΔF, ΔF') in multiple sets of training data is performed, and the value of k2 is updated according to its linear regression coefficient b1. Wherein, ΔE' is the theoretical equivalent value of emulsifier addition increment required to achieve the extension factor target, which is derived from the deviation value of the extension factor; ΔF' is the theoretical equivalent value of dietary fiber addition increment required to achieve the extension factor target, which is derived from the deviation value of the extension factor.

7. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 6, characterized in that, ΔE'=M / P1; ΔF'=M / P2; P1 is the emulsifier efficacy coefficient, which represents the average impact of a unit amount of emulsifier added on the spreadability factor based on the standard basic formula; P2 is the dietary fiber efficacy coefficient, which represents the average impact of a unit amount of dietary fiber added on the spreadability factor based on the standard basic formula.

8. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in claim 7, characterized in that, Based on the standard basic formula, the amount of emulsifier or dietary fiber added was changed in a gradient manner to prepare and measure the extensibility factor of the corresponding biscuits. The average change in extensibility factor caused by a unit amount of addition was obtained by linear fitting, and the values ​​of P1 and P2 were obtained.

9. The method for controlling the quality of highland barley biscuits based on the extensibility factor as described in any one of claims 6 to 8, characterized in that, The final updated k1* is determined using the linear regression coefficient a1 according to the following rules: First, determine the reference value k1' based on the linear regression coefficient a1: If 0.05 ≤ a1 ≤ 0.2, then k1' = a1; If a1 < 0.05, then k1' = 0.05 + λ(0.05 - a1); If a1 > 0.2, then k1' = 0.2 + λ(0.2 - a1); Then, based on the reference value k1', determine the updated k1*: k1* = η1·k1 + (1-η1)·k1'; The final updated k2* is determined using the linear regression coefficient b1 according to the following rules: First, determine the reference value k2' based on the linear regression coefficient b1: If 0.05 ≤ b1 ≤ 0.2, then k2' = b1; If b1 < 0.05, then k2' = 0.05 + μ(0.05 - b1); If b1 > 0.2, then k2' = 0.2 + μ(0.2 - b1); Then, based on the reference value k2', determine the updated k2*: k2* = η2·k2 + (1-η2)·k2'; Where λ and μ are both boundary attenuation coefficients between 0.1 and 0.5; η1 and η2 are both smoothing coefficients between 0 and 1.