A method and device for judging and regulating the fermentation state of Maotai-flavor liquor

Abstract

This application discloses a method and apparatus for judging and controlling the fermentation state of Maotai-flavor liquor, belonging to the field of liquor brewing technology. The method includes: calibrating the mean μ and standard deviation σ of each core parameter, and defining a safe zone, a warning zone, and a danger zone; collecting the measured values ​​of each core parameter in the current fermentation cycle, calculating the standardized offset of each core parameter, and performing a weighted summation to calculate the fermentation progress index (FPI); judging the current fermentation trajectory type based on the FPI value and the consistency constraint of the offset direction of at least three core parameters; predicting the offset trend of downstream parameters in subsequent cycles using the offset of upstream parameters in the current cycle based on the causal chain cascade relationship; and performing feedforward control before the downstream parameters actually shift based on the judged trajectory type and the FPI value. This technical solution can construct a full-cycle fermentation evaluation benchmark, predict deviations using feedforward, and achieve precise pre-fermentation control of Maotai-flavor liquor.

Description

Technical Field

[0001] This application belongs to the field of liquor brewing technology, specifically relating to a method and apparatus for judging and controlling the fermentation state of sauce-flavored liquor. Background Technology

[0002] The Maotai-flavor baijiu is produced using the traditional 12987 brewing process, which includes nine fermentation stages: the initial fermentation, the rough fermentation, and stages one through seven. Significant time lags exist between the fermentation parameters of each stage; flavor deviations in stage N can often be traced back to the operating parameters of stage N-1 or N-2. Therefore, systematically identifying and dynamically controlling the fermentation state across stages is crucial for ensuring the quality of the base liquor.

[0003] In existing technologies, fermentation status monitoring mainly focuses on feedback control for single rounds and single processes. Specifically, operators typically collect static physicochemical indicators such as moisture and reducing sugars in the mash as a basis for judgment, and then passively adjust process parameters only after deviations in the product-end indicators are detected. Furthermore, existing methods lack mathematical tools to integrate multi-dimensional monitoring parameters into a unified quantitative evaluation benchmark. The data for each parameter are scattered and independent, making it difficult to form a comprehensive evaluation of the overall fermentation progress; and without establishing a standard parameter trajectory as a reference system, deviation judgments lack objective basis.

[0004] The aforementioned existing technologies have the following prominent problems: they only focus on the local state of a single round, and cannot answer the question of "what kind of operating trajectory the entire 9 rounds of fermentation is on" from a system level; they adopt feedback control, and only respond after the deviation actually occurs, failing to utilize the cross-round delay effect for advance prediction; they rely on static physicochemical indicators, and cannot reflect the metabolic activity and evolution of the microbial community in real time; moreover, abnormalities of a single parameter are prone to misjudgment; and they lack quantitative integration methods for multi-dimensional parameters and supporting automated discrimination and control devices, resulting in insufficient timeliness and accuracy of fermentation process control, affecting the stability of the base wine production quality. Summary of the Invention

[0005] This application provides a method and apparatus for judging and controlling the fermentation state of Maotai-flavor liquor, addressing the technical problems in existing Maotai-flavor liquor fermentation control, such as single-round feedback lag, static indicators failing to reflect metabolic activity, multiple parameters being scattered without a unified evaluation benchmark, and a lack of automated feedforward control.

[0006] In a first aspect, embodiments of this application provide a method for determining and controlling the fermentation state of Maotai-flavor liquor, the method comprising: Step 1: Based on historical batch data, the mean μ and standard deviation σ of each core parameter are calibrated for the nine fermentation cycles of the Maotai-flavor liquor process, and safe zone, warning zone and danger zone are defined; the core parameters include: degree of gelatinization, comprehensive enzyme activity index, respiration intensity, acidity and microbial diversity index; Step 2: Collect the measured values ​​of each core parameter in the current round, calculate the standardized offset of each core parameter based on the measured values ​​and the mean μ and standard deviation σ of each core parameter as pre-calibrated, and perform a weighted summation of each standardized offset according to the preset causal chain weights to calculate the fermentation progress index FPI, wherein the weights are assigned in descending order from upstream to downstream according to the position of the parameter in the causal chain. Step 3: Based on the value of FPI and the consistency constraint of the offset direction of at least three core parameters, determine the current orbit type of fermentation. The orbit type includes typical orbit, forward offset orbit, backward offset orbit, and fluctuating orbit. Step 4: Based on the causal chain cascade relationship between the five core parameters, use the offset of the upstream parameter in the current round to predict the offset trend of the downstream parameter in subsequent rounds. Step 5: Based on the identified track type and the value of FPI, perform feedforward control before the downstream parameters actually shift; wherein, when shifting the track forward, the control amount decreases as FPI increases; when shifting the track backward, the control amount increases as the absolute value of FPI increases.

[0007] Furthermore, in step 2, the standardized offset Δ of each core parameter i Calculated using the formula: ; The fermentation progress index (FPI) is calculated using the following formula: ; Where w i is the causal chain weight corresponding to the i-th core parameter.

[0008] Furthermore, the specific weights assigned to the causal chain are as follows: weight w1 for degree of gelatinization = 0.30, weight w2 for comprehensive enzyme activity index = 0.25, weight w3 for respiration intensity = 0.20, weight w4 for acidity = 0.15, and weight w5 for microbial diversity index = 0.10.

[0009] Furthermore, in step 3, the rule for determining the orbit type is as follows: When |FPI| < 0.5, it is determined to be a typical orbit; When FPI ≥ 0.5 and there are at least one consecutive rounds, and at least three parameters Δi > 0, it is determined to be a forward offset track; When FPI ≤ -0.5 and there are at least one consecutive rounds, and at least three parameters Δi < 0, it is determined to be a backward deviation track; When |FPI|≥0.5 and the FPI signs of adjacent rounds are reversed, and the parameter offset directions are inconsistent, it is determined to be a fluctuating trajectory.

[0010] Furthermore, in step 4, the causal chain cascade relationship includes: Gelatinization degree shift predicts the next round of same-direction shift in the comprehensive enzyme activity index; The overall enzyme activity index shift predicts the next round of respiratory intensity shift in the same direction; Respiratory intensity shift predicts acidity shift in the same direction in the next round; Acidity shift predicts two rounds of inverse shifts in the microbial diversity index; The prediction formula is △ i+1 (downstream) = β × △ i (Upstream), where β is the causal transmission coefficient and 0 < β < 1.

[0011] Furthermore, in step 5: Adjustment amount of steaming time when shifting the track forward Adjust the amount of moisture in the fermented mash. Where k1 ranges from 5 to 8 minutes per unit of FPI, and k2 ranges from 0.5 to 1.0 percentage points per unit of FPI; Adjustment amount of steaming time when shifting the track backward Adjust the amount of moisture in the fermented mash. .

[0012] Furthermore, step 5 also includes: When the trajectory fluctuates, reduce the amplitude of a single operation by 30% to 50%, check the consistency of the process, and do not adjust the steaming time and moisture for the time being.

[0013] Furthermore, in step 1, the safe zone is defined as μ±1.0σ, the warning zone is defined as μ±1.5σ, and the danger zone is defined as the range beyond μ±1.5σ.

[0014] Secondly, this embodiment also provides a device for judging and controlling the fermentation state of Maotai-flavor liquor, the device comprising: The parameter acquisition module is used to collect the measured values ​​of each core parameter in the current round. The core parameters include: degree of gelatinization, comprehensive enzyme activity index, respiration intensity, acidity, and microbial diversity index. The calculation module, connected to the parameter acquisition module, is used to obtain the mean μ and standard deviation σ of each core parameter pre-calibrated based on historical batch data, calculate the standardized offset of each core parameter according to the measured value, mean μ and standard deviation σ, and perform weighted summation of each standardized offset according to the preset causal chain weights to calculate the fermentation progress index FPI, wherein the weights are assigned in descending order from upstream to downstream according to the position of the parameter in the causal chain. The discrimination module, connected to the calculation module, is used to determine the current fermentation trajectory type based on the value of the FPI and the consistency constraints of the offset direction of at least three core parameters. The trajectory type includes typical trajectory, forward offset trajectory, backward offset trajectory, and fluctuating trajectory. The prediction module, connected to the discrimination module, is used to predict the offset trend of the downstream parameter in subsequent rounds based on the causal chain cascade relationship between the five core parameters and the offset of the upstream parameter in the current round. The control module, connected to the prediction module, is used to output a feedforward control command to the grain steaming equipment and / or moisture control equipment before the downstream parameters actually shift, based on the identified track type and the value of FPI. When shifting the track forward, the control amount decreases as FPI increases; when shifting the track backward, the control amount increases as the absolute value of FPI increases.

[0015] Furthermore, the parameter acquisition module specifically includes: The gelatinization degree detection unit, comprehensive enzyme activity detection unit, CO2 release rate online monitoring unit, acidity detection unit, and microbial diversity detection unit are used to detect the corresponding core parameters.

[0016] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0017] Fourthly, embodiments of this application provide a readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0018] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method as described in the first aspect.

[0019] The technical solution provided in this application realizes a quantitative comprehensive evaluation of multidimensional deviations by constructing a five-dimensional parameter system covering the substrate to the community and calculating the fermentation progress index. It effectively reduces the risk of misjudgment of single-parameter anomalies by utilizing the consistency constraint of multi-parameter directions. Based on the causal chain cascade relationship, it completes feedforward prediction and early intervention before the deviation occurs at the product end, breaking through the lag limitation of traditional feedback control. At the same time, it combines quantitative control formulas to form a complete processing logic of discrimination, prediction and control, which significantly improves the accuracy and timeliness of the control of the multi-round fermentation process of Maotai-flavor liquor. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the method for determining and controlling the fermentation state of Maotai-flavor liquor provided in the embodiments of this application; Figure 2 This is a schematic diagram of the fermentation state discrimination and control device for Maotai-flavor liquor provided in the embodiments of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, specific embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining this application and not for limiting it. It should also be noted that, for ease of description, only the parts relevant to this application are shown in the drawings, not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but may also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0022] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0023] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0024] The following description, in conjunction with the accompanying drawings, details the apparatus, device, equipment, and medium for determining and controlling the fermentation state of Maotai-flavor liquor provided in this application, through specific embodiments and application scenarios.

[0025] Example 1 Figure 1This is a flowchart illustrating the method for determining and controlling the fermentation state of Maotai-flavor liquor provided in this application embodiment. Figure 1 As shown, the method includes: Step 1: Based on historical batch data, the mean μ and standard deviation σ of each core parameter are calibrated for the nine fermentation cycles of the Maotai-flavor liquor process, and safe zone, warning zone and danger zone are defined; the core parameters include: degree of gelatinization, comprehensive enzyme activity index, respiration intensity, acidity and microbial diversity index; The brewing process of Maotai-flavor liquor refers to the 12987 brewing process, which includes multiple rounds of cyclical fermentation. Historical batch data refers to all original parameter test data retained from the past nine complete fermentation processes on the same production line. The nine fermentation rounds refer to nine consecutive fermentation cycles: Xia Sha, Cao Sha, first round, second round, third round, fourth round, fifth round, sixth round, and seventh round.

[0026] The mean μ is the arithmetic mean of all historical samples of the same core parameter in a single round. The standard deviation σ is the discrete fluctuation of all historical samples of the same core parameter in a single round.

[0027] The core parameters refer to five categories of monitoring indicators that cover the entire biological pathway, including substrate, catalysis, metabolism, products, and microbial community.

[0028] Gelatinization degree refers to the percentage substrate index characterizing the degree of thermal denaturation of starch in fermented mash. Comprehensive enzyme activity index refers to a comprehensive index of catalytic capacity integrating the activities of amylase, protease, and cellulase. Respiration intensity refers to a monitoring index characterizing the real-time overall metabolic activity of microorganisms in fermented mash, expressed as the CO2 release rate. Acidity is a physicochemical indicator characterizing the accumulation of organic acids in fermented mash. Microbial diversity index is an evaluation index quantifying the complexity of the microbial community structure in fermented mash.

[0029] The safe zone refers to the range where core parameters fluctuate normally, and there is no quality risk during fermentation. The warning zone refers to the range where core parameters deviate slightly from the standard benchmark, indicating a potential risk of abnormal fermentation. The danger zone refers to the range where core parameters deviate significantly from the standard benchmark, which will inhibit microbial activity and reduce the quality of the resulting wine.

[0030] This solution can retrieve historical batch data stored locally, statistically analyze samples of each type of core parameter in rounds, calculate the mean μ and standard deviation σ respectively, store the calculation results in the system database, and then use the calculated mean μ and standard deviation σ to define the range of three types of intervals, and synchronously store the interval determination criteria in the system for subsequent use.

[0031] Step 2: Collect the measured values ​​of each core parameter in the current round, calculate the standardized offset of each core parameter based on the measured values ​​and the mean μ and standard deviation σ of each core parameter as pre-calibrated, and perform a weighted summation of each standardized offset according to the preset causal chain weights to calculate the fermentation progress index FPI, wherein the weights are assigned in descending order from upstream to downstream according to the position of the parameter in the causal chain. The current fermentation cycle refers to the fermentation cycle that is currently in progress.

[0032] The measured value refers to the original measured value of the core parameters output in real time by on-site testing equipment, online sensors, and laboratory testing equipment.

[0033] Standardized offset refers to a dimensionless value that eliminates the differences in the dimensions of different parameters and uniformly measures the deviation of a parameter from a standard reference.

[0034] Causal chain weights refer to the weighting coefficients assigned to upstream and downstream positions of parameters in a biological pathway.

[0035] The Fermentation Progress Index (FPI) is a comprehensive evaluation index that integrates the deviation of five core parameters and quantifies the degree of overall fermentation progress being ahead or behind schedule.

[0036] A causal chain refers to a biological transmission pathway in which the degree of gelatinization, the comprehensive enzyme activity index, the respiration intensity, the acidity, and the microbial diversity index have a unidirectional, stepwise influence.

[0037] Specifically, various detection units can collect real-time data from fermentation mash samples, and the measured values ​​of each core parameter can be uploaded to the central control system via data transmission lines. Then, the stored mean μ and standard deviation σ can be substituted into the corresponding formulas to obtain standardized offsets. Preset causal chain weights are then retrieved, and all standardized offsets are summed using weighted averages to output a unique fermentation progress index (FPI). Simultaneously, weight values ​​can be set for parameters at their upstream and downstream positions in the causal chain, with higher weights for upstream substrate and catalytic parameters, and lower weights for downstream product and community parameters.

[0038] Step 3: Based on the value of FPI and the consistency constraint of the offset direction of at least three core parameters, determine the current orbit type of fermentation. The orbit type includes typical orbit, forward offset orbit, backward offset orbit, and fluctuating orbit. The offset direction consistency constraint refers to the judgment condition that determines whether most parameters deviate from the standard reference synchronously and in the same direction by statistically analyzing the positive and negative signs of the standardized offsets of multiple core parameters.

[0039] Track type refers to the four types of fermentation operation states classified based on the fermentation progress index value and multi-parameter offset characteristics.

[0040] A typical trajectory refers to a normal fermentation state where the fermentation progress matches the standard process progress of the corresponding round.

[0041] A forward deviation from the track refers to an abnormal fermentation state where the overall fermentation progress is earlier than the standard process progress of the corresponding round.

[0042] A backward deviation from the track refers to an abnormal fermentation state in which the overall fermentation progress lags behind the standard process progress of the corresponding round.

[0043] A fluctuating trajectory refers to an oscillating fermentation state in which the fermentation progress is unstable and the direction of deviation changes repeatedly.

[0044] Specifically, the central control system can retrieve the real-time calculated FPI value and all standardized offsets, compare them with preset judgment conditions, and output the track type corresponding to the current fermentation.

[0045] Step 4: Based on the causal chain cascade relationship between the five core parameters, use the offset of the upstream parameter in the current round to predict the offset trend of the downstream parameter in subsequent rounds. The cascading causal chain relationship refers to the hierarchical transmission pattern in which changes in upstream indicators of five core parameters have a delayed impact on downstream indicators.

[0046] Upstream parameters refer to the degree of gelatinization, comprehensive enzyme activity index, and respiration intensity at the earlier positions in the causal chain.

[0047] Downstream parameters refer to acidity and microbial diversity indices located later in the causal chain.

[0048] Deviation trend refers to the direction and magnitude of the deviation of the corresponding parameters from the standard baseline in subsequent fermentation cycles.

[0049] The central control system can retrieve the standardized offset of upstream parameters in the current round, and based on the preset cascade transmission law, deduce and calculate the offset change trend of downstream parameters in subsequent rounds, and generate risk warning information.

[0050] Step 5: Based on the identified track type and the value of FPI, perform feedforward control before the downstream parameters actually shift; wherein, when shifting the track forward, the control amount decreases as FPI increases; when shifting the track backward, the control amount increases as the absolute value of FPI increases.

[0051] Feedforward control refers to an active intervention mode that adjusts process parameters in advance before fermentation abnormalities or product indicators exceed standards.

[0052] The control amount refers to the specific values ​​that need to be adjusted for the steaming time and the moisture content of the mash.

[0053] Negative correlation means that the larger the FPI value, the greater the reduction in process adjustment.

[0054] Absolute value refers to the pure numerical value of FPI after removing the positive and negative signs.

[0055] Positive correlation means that the larger the absolute value of FPI, the greater the improvement in process adjustment.

[0056] The central control system can combine track type and FPI value to calculate the steaming time and mash moisture adjustment amount, and issue adjustment instructions to the steaming equipment and moisture control equipment to correct the process operation in advance.

[0057] This technical solution obtains a single FPI evaluation index through multi-parameter integration, determines the fermentation trajectory by combining multi-parameter constraints, and predicts fermentation risks in advance and realizes pre-intervention based on the causal transmission law of parameters. It solves the defects of existing technologies that only provide single-round feedback control, lack unified evaluation standards, and cannot predict fermentation deviations in advance, and realizes systematic and predictive precise control of the entire fermentation cycle of Maotai-flavor liquor.

[0058] In one embodiment, optionally, the standardized offset Δ of each core parameter i Calculated using the formula: ; The fermentation progress index (FPI) is calculated using the following formula: ; Where w i is the causal chain weight corresponding to the i-th core parameter.

[0059] Standardized offset This refers to the deviation from the benchmark value of a single core parameter after eliminating its dimensions.

[0060] , refers to the real-time detection value corresponding to the i-th core parameter.

[0061] It refers to the standard mean of the round corresponding to the i-th core parameter.

[0062] It refers to the standard deviation of the i-th core parameter corresponding to the round.

[0063] This technical solution clarifies the standardized calculation formulas for standardized offset and FPI, unifies the calculation logic of all parameters and indices, eliminates human calculation errors, ensures the consistency of FPI calculation results under different batches and different equipment, and improves the repeatability and industrial applicability of quantitative evaluation of fermentation status.

[0064] In one embodiment, optionally, the specific values ​​assigned to the causal chain weights are: weight w1 for degree of gelatinization = 0.30, weight w2 for comprehensive enzyme activity index = 0.25, weight w3 for respiration intensity = 0.20, weight w4 for acidity = 0.15, and weight w5 for microbial diversity index = 0.10.

[0065] This solution can combine the causal transmission value of parameters, set fixed weight values ​​for five types of core parameters and write them into the calculation program of the central control system. When the system calculates FPI, it can directly call the fixed weights.

[0066] This technical solution limits the gradient weight values ​​to highlight the dominant influence of upstream substrates and catalytic parameters on the overall fermentation state, allowing the FPI index to better reflect the deviation of front-end parameters with feedforward predictive value, and further strengthening the method's ability to provide early warning of fermentation risks.

[0067] In one embodiment, optionally, in step 3, the rule for determining the orbit type is: When |FPI| < 0.5, it is determined to be a typical orbit; When FPI ≥ 0.5 and there are at least one consecutive rounds, and at least three parameters Δi > 0, it is determined to be a forward offset track; When FPI ≤ -0.5 and there are at least one consecutive rounds, and at least three parameters Δi < 0, it is determined to be a backward deviation track; When |FPI|≥0.5 and the FPI signs of adjacent rounds are reversed, and the parameter offset directions are inconsistent, it is determined to be a fluctuating trajectory.

[0068] "At least one consecutive cycle" means that the same offset feature persists for one or more complete fermentation cycles.

[0069] △i>0 means that the measured value of the corresponding core parameter is higher than the standard mean of this round, and the parameter deviates from the benchmark in a positive direction.

[0070] △i<0 means that the measured value of the corresponding core parameter is lower than the standard mean of that round, and the parameter deviates negatively from the benchmark.

[0071] The reversal of the sign of FPI in adjacent rounds means that the FPI of two consecutive fermentation rounds is one positive and one negative, with the offset direction changing alternately.

[0072] The central control system can synchronously match the real-time FPI, the positive and negative values ​​of each parameter Δi, and the historical FPI records of each round with the multi-dimensional judgment rules defined in this article, and automatically match and output a unique track type, so that multiple judgment results will not occur.

[0073] This technical solution quantifies the clear judgment thresholds for four types of fermentation trajectories and constrains the consistency of multiple parameters. Occasional anomalies in a single parameter will not trigger misjudgment of the trajectory, greatly reducing the probability of misidentification of fermentation status and ensuring that subsequent control strategies match the actual fermentation conditions.

[0074] In one embodiment, optionally, the causal chain cascade relationship includes: Gelatinization degree shift predicts the next round of same-direction shift in the comprehensive enzyme activity index; The overall enzyme activity index shift predicts the next round of respiratory intensity shift in the same direction; Respiratory intensity shift predicts acidity shift in the same direction in the next round; Acidity shift predicts two rounds of inverse shifts in the microbial diversity index; The prediction formula is △ i+1 (downstream) = β × △ i (Upstream), where β is the causal transmission coefficient and 0 < β < 1.

[0075] The next round of same-direction offset refers to the situation where, when the upstream parameter deviates positively or negatively, the downstream parameter in the next round will deviate from the standard reference in the same direction.

[0076] The next two rounds of reverse shift refer to the microbial diversity index deviating from the baseline in the opposite direction after the acidity deviates, with an interval of two fermentation cycles.

[0077] △ i+1 (Downstream) refers to the standardized offset of the predicted downstream parameters.

[0078] △ i (Upstream) refers to the standardized offset of upstream parameters obtained from the actual measurement in the current round.

[0079] β refers to the causal transmission coefficient, which represents the attenuation ratio when upstream deviations are transmitted downstream.

[0080] The central control system reads the current upstream parameter offset, substitutes it into the transmission formula, multiplies it by the calibrated β coefficient, calculates the downstream parameter offset magnitude and direction for the corresponding subsequent rounds, and generates a fermentation exceedance risk warning.

[0081] This technical solution clarifies the precise lag transmission law and standardized prediction formula of five types of parameters, which can predict the risk of exceeding the standard of product and microbial community indicators at least 1 to 2 rounds in advance. Compared with traditional post-event feedback control, it significantly extends the process intervention window period.

[0082] In one embodiment, optionally, in step 5: Adjustment amount of steaming time when shifting the track forward Adjust the amount of moisture in the fermented mash. Where k1 ranges from 5 to 8 minutes per unit of FPI, and k2 ranges from 0.5 to 1.0 percentage points per unit of FPI; Adjustment amount of steaming time when shifting the track backward Adjust the amount of moisture in the fermented mash. .

[0083] The adjustment amount of steaming time △t refers to the amount of time that needs to be increased or decreased in the steaming process.

[0084] k1 refers to the steaming time control ratio coefficient. k2 refers to the mash moisture control ratio coefficient.

[0085] The moisture adjustment amount △w of the fermented mash refers to the percentage of moisture that needs to be adjusted upwards or downwards.

[0086] The unit FPI refers to the process adjustment range corresponding to a change of 1 unit in the FPI value.

[0087] The central control system can read the track type and real-time FPI value, match the corresponding calculation formula, substitute the preset coefficients in the k1 and k2 intervals, output the precise adjustment values ​​for steaming grain and moisture, and send them to the corresponding production equipment for execution.

[0088] This technical solution provides a quantitative and directly applicable calculation formula for process adjustments. The control range changes synchronously with the degree of fermentation deviation, achieving graded and precise control with stronger intervention as the deviation increases, thus avoiding fermentation fluctuations caused by extensive manual adjustments.

[0089] In one embodiment, step 5 may optionally further include: When the trajectory fluctuates, reduce the amplitude of a single operation by 30% to 50%, check the consistency of the process, and do not adjust the steaming time and moisture for the time being.

[0090] The range of a single operation refers to the extent of process changes made in a single process such as steaming grain, mixing feed, or adding water.

[0091] Process consistency refers to the degree of uniformity in the execution of process operations by different work groups and different equipment.

[0092] The central control system issues instructions to the production terminal, limiting the scope of a single process operation change to 50% to 70% of the original range. If a checklist is generated, it is pushed to the on-site operators to verify the team's operating habits and differences in equipment operating parameters.

[0093] This technical solution employs differentiated control strategies to address unstable oscillating trajectories. It does not directly alter the core fermentation substrate conditions but prioritizes eliminating the source of operational fluctuations, avoiding blind adjustments that could exacerbate fermentation oscillations and stabilizing the fermentation ecosystem.

[0094] In one embodiment, optionally, in step 1, the safe zone is defined as μ±1.0σ, the warning zone is defined as μ±1.5σ, and the danger zone is defined as the range beyond μ±1.5σ.

[0095] The central control system can write the three-segment interval judgment thresholds into the program, and then substitute the subsequent detection parameters for comparison, automatically marking the parameters as being in the safe zone, warning zone, or danger zone, and synchronously matching the corresponding risk level.

[0096] This technical solution quantifies the boundary standards of three risk intervals, clearly classifies the risk levels of parameter deviation, facilitates the system's graded early warning and graded adjustment of control intensity, and achieves differentiated management based on the severity of risk.

[0097] To help technical personnel better understand this solution, a specific implementation method is also provided.

[0098] This invention provides a fermentation state discrimination and control method and device based on fermentation progress index (FPI). By quantifying the fermentation progress, the fermentation trajectory type is determined by comparing the matching relationship between the fermentation degree and the cycle progress. Furthermore, the causal relationship between parameters is used to implement advance control before the deviation is transmitted to the product end.

[0099] The core idea of ​​this invention is to calibrate the standard parameter trajectory, calculate the fermentation progress index (FPI), determine the fermentation trajectory type, predict the parameter trend of the next round, and implement regulation.

[0100] The technical improvements of this invention include, but are not limited to: (1) Five parameters across biological levels are integrated into a single fermentation progress index (FPI) to achieve a quantitative comprehensive evaluation of multidimensional deviations; (2) The track discrimination requires multi-parameter directional consistency constraints, and a single parameter abnormality will not trigger track switching; (3) Feedforward prediction is achieved based on the causal relationship between parameters, and intervention is completed before the deviation at the product end occurs.

[0101] Step 1: Parameter calibration; Five core parameters were selected to cover the complete biological pathway from substrate to catalysis to metabolism to product to community. (1) Degree of gelatinization (%): Substrate-end parameter, reflecting the degree of starch denaturation and determining the total amount of available carbon source.

[0102] (2) Comprehensive enzyme activity index: a catalytic end parameter, which is a weighted comprehensive value of the activities of amylase, protease and cellulase. It directly reflects the catalytic ability of microorganisms and has more feedforward prediction value than product indicators such as reducing sugar.

[0103] (3) Respiratory intensity (CO2 release rate, mg·kg) -1 ·h-1 Metabolic parameters reflect the metabolic activity of the microbial community in fermented mash, enabling non-destructive online monitoring.

[0104] (4) Acidity (°T): Product end parameter, reflecting the degree of organic acid accumulation. Acidity > 1.5°T is the threshold for inhibiting amylase activity.

[0105] (5) Microbial diversity index (Shannon index): a community-end parameter that quantifies the complexity of the microbial community structure and reflects the stability of the fermentation ecosystem.

[0106] The 5-parameter causal chain is as follows: degree of gelatinization (substrate end) → comprehensive enzyme activity index (catalytic end) → respiration intensity (metabolic end) → acidity (product end) → Shannon index (community end). It has a unidirectional, layer-by-layer influence and amplification. The deviation of upstream parameters can predict the trend of downstream parameters along the causal chain.

[0107] For each of the nine fermentations, the mean μ and standard deviation σ of each parameter were calibrated (based on historical batch data), and the safe zone (μ±1.0σ), the warning zone (μ±1.5σ), and the danger zone (beyond μ±1.5σ) were defined.

[0108] The summary table of classic trajectory calibration values ​​(μ±σ) for 9 rounds × 5 parameters is as follows:

[0109] Step 2: Calculation of Fermentation Progress Index (FPI); Collect the measured values ​​of 5 parameters for the current round and calculate the standardized offset of each parameter: Δ i = (Measured value) i μ i ) / σ i Calculate the Fermentation Progress Index (FPI) by assigning weights based on the causal chain position: FPI = Σ(w i × Δ i = 0.30×Δ_gelatinization degree + 0.25×Δ_enzyme activity + 0.20×Δ_respiration + 0.15×Δ_acidity + 0.10×Δ_Shannon; Weighting design principle: Upstream parameter (substrate end, catalyst end) shifts have causal predictive value for downstream processes and are assigned higher weights; downstream parameter (product end, community end) shifts reflect the already occurred state more and are assigned lower weights. FPI>0 indicates that the fermentation stage is ahead of the cycle progress, FPI<0 indicates lag, and the larger the |FPI|, the more severe the deviation.

[0110] The FPI parameter weights and causal chain positions are shown in the table below:

[0111] Step 3: Track determination; The core logic of track classification is to compare the matching relationship between "fermentation degree" and "round progress", using FPI as the quantitative benchmark, and requiring consistency constraints in multiple parameter directions: (1) Typical track (standard): |FPI|<0.5, fermentation degree matches the round progress.

[0112] (2) Forward offset track (forward movement): FPI ≥ 0.5 and for ≥ 1 consecutive rounds, with at least 3 / 5 of the parameters Δ i >0 (positive direction) indicates that the fermentation process is ahead of the previous cycle.

[0113] (3) Backward offset track (shift): FPI≤ 0.5 and for at least one consecutive round, with at least 3 / 5 of the parameters Δ i <0 (same direction, slightly negative) indicates that the fermentation process is lagging behind the progress of the next round.

[0114] (4) Fluctuation track (oscillation): |FPI|≥0.5 but the FPI sign reverses in adjacent rounds and the parameter offset direction is inconsistent, indicating that the relationship between the fermentation degree and the round progress is unstable.

[0115] The significance of multi-parameter consistency constraints: a single parameter anomaly (such as an occasional high degree of gelatinization in a certain round) does not trigger track switching. Multiple parameters are required to confirm the offset direction, reduce the risk of misjudgment, and at the same time make the 5-parameter system not simplistic to single-parameter monitoring.

[0116] The four orbital classifications and criteria are shown in the table below:

[0117] Step 4: Causal chain prediction; Based on the cascading relationship of the 5-parameter causal chain, the shift trend of downstream parameters in subsequent rounds can be predicted from the upstream parameter shift in the current round: (1) Gelatinization degree shift → predict the next round of the comprehensive enzyme activity index shift in the same direction (substrate release change → microbial enzyme secretion response, lag of about 1 round).

[0118] (2) Comprehensive enzyme activity index shift → predict the next round of respiratory intensity shift in the same direction (catalytic capacity change → metabolic rate response, lags by about 1 round).

[0119] (3) Respiratory intensity shift → predicted acidity shift in the same direction in the next round (metabolic rate change → organic acid accumulation response, lags by about 1 round).

[0120] (4) Acidity shift → predicted Shannon index two rounds of reverse shift (acidity accumulation → community selection pressure, lag of about 2 rounds).

[0121] Prediction formula: Δ i+1 (Downstream) ≈ β × Δ i (Upstream), β is the causal transmission coefficient (0<β<1, indicating cascade attenuation), which is calibrated by regression from historical data.

[0122] The value of feedforward prediction: One to two cycles before the actual occurrence of product-side (acidity) deviation, the risk can be predicted based on the deviations at the substrate and catalyst ends, thus gaining a window of time for regulation.

[0123] Step 5: Feedforward regulation; Based on the track type and FPI amplitude, control is performed before downstream parameter deviation occurs, and the control amplitude is proportional to the FPI: Forward trajectory control: Steaming time adjustment Δt = k1 × FPI (minutes), where k1 is 5~8; Moisture adjustment amount of fermented mash Δw = k2 × FPI (percentage points), where k2 ranges from 0.5 to 1.0; Reverse trajectory control: Take the opposite direction, Δt = +k1×|FPI|, Δw = +k2×|FPI|.

[0124] Fluctuation track control: Reduce the amplitude of a single operation by 30-50%, check for consistency in processes (differences between work groups and equipment), and temporarily refrain from adjusting the steaming time and moisture content to eliminate the source of vibration.

[0125] Intervention timing principle: When the feedforward prediction of the causal chain indicates that downstream parameters are at risk of exceeding the threshold, regulation should be implemented in the next round after the upstream deviation round, rather than waiting until the downstream parameters actually exceed the standard. Acidity <1.0°T is the conventional intervention window, >1.0°T requires increased intensity, and >1.5°T is an unconventional intervention (requiring manual assessment).

[0126] This invention also provides a device for judging and controlling the fermentation state of Maotai-flavor liquor, comprising the following functional modules: (1) Parameter acquisition module: used to collect the measured data of 5 core parameters in the current round, including the gelatinization degree detection unit, the comprehensive enzyme activity detection unit, the CO2 release rate online monitoring unit, the acidity detection unit and the microbial diversity detection unit.

[0127] (2) FPI calculation module: connected to the parameter acquisition module, used to calculate FPI according to the formula Δ i = (Measured value) i μ i ) / σ i Calculate the standardized offset of each parameter, and calculate the fermentation progress index according to the causal chain weight FPI = 0.30×Δ_gelatinization degree + 0.25×Δ_enzyme activity + 0.20×Δ_respiration + 0.15×Δ_acidity + 0.10×Δ_Shannon.

[0128] (3) Orbit discrimination module: connected to the FPI calculation module, used to determine the fermentation orbit type based on the FPI value and the multi-parameter direction consistency constraint (≥3 / 5 parameters in the same direction), and output the discrimination results of typical orbit, forward offset, backward offset or fluctuating orbit.

[0129] (4) Causal chain prediction module: connected to the orbit discrimination module, used to predict the causal chain based on the 5-parameter causal chain cascade relationship, according to the prediction formula Δ i+1 (Downstream) ≈ β × Δ i (Upstream) Predict the shift trend of downstream parameters in subsequent rounds and output risk warning information.

[0130] (5) Control Execution Module: Connected to the causal chain prediction module, used to control according to the track type and FPI amplitude using the control formula Δt = k1×FPI、Δw = k2×FPI calculates the adjustment amount of the control parameters and outputs execution instructions to the grain steaming equipment and moisture control equipment.

[0131] The device can be integrated into the central control system of a liquor production line to achieve an automated closed loop from parameter acquisition to control execution.

[0132] The technical solution provided by this invention can achieve the following beneficial effects: 1. Fermentation Progress Index (FPI) achieves comprehensive quantification of multidimensional biases: For the first time, five parameters across biological levels are integrated into a single index through causal chain weighting, solving the problem of scattered multi-parameter monitoring data and the inability to conduct unified evaluation. The FPI weights are designed to decrease in order of position in the causal chain, so that the feedforward predictive value of upstream parameters is reflected in the index.

[0133] 2. Multi-parameter joint orbit discrimination reduces the risk of misjudgment: orbit switching requires FPI to exceed the threshold and ≥3 / 5 parameters to be in the same direction. An abnormality of a single parameter will not trigger orbit switching, making the 5-parameter system irreplaceable and impossible to simplify into single-parameter monitoring.

[0134] 3. Causal chain feedforward prediction achieves "prevention of disease": Based on the cascade lag relationship between parameters, the risk is predicted and intervened in advance 1 to 2 rounds before the deviation occurs at the product end, breaking through the lag limitation of existing feedback control.

[0135] 4. Quantitative regulation driven by FPI: The regulation range is proportional to the FPI, achieving precise regulation of "the greater the deviation, the stronger the intervention", rather than the general "lower if too high, higher if too low".

[0136] 5. Complete Closed-Loop and Device-Based Implementation: From parameter calibration → FPI calculation → track discrimination → causal chain prediction → feedforward control, a complete closed loop from discrimination to control is formed. The supporting device can realize the automated process from parameter acquisition to control execution, suitable for industrial online applications.

[0137] Practical examples: forward trajectory identification, causal chain feedforward prediction and regulation; The measured data of 5 parameters and FPI calculation for four consecutive cycles (Xia Sha, Cao Sha, First Cycle, and Second Cycle) of a certain Maotai-flavor liquor production line are as follows. Feedforward control is implemented starting from the First Cycle.

[0138] The measured values ​​of the parameters, the standardized offset Δ, and the calculated FPI are shown in the table below:

[0139] The data from the first and second rounds are the data after the implementation of feedforward regulation.

[0140] Track identification (rough sand wheel): FPI = 0.30×1.20 + 0.25×1.07 + 0.20×1.00 + 0.15×0.50 + 0.10×0 =0.90, ≥0.5 threshold.

[0141] Of the five parameters, three have Δ>0 (gelatinization degree +1.20, comprehensive enzyme activity +1.07, and respiration intensity +1.00), satisfying the ≥3 / 5 directional consistency constraint.

[0142] Decision: Deflect forward on the track (move forward).

[0143] Causal chain feedforward prediction (rough sand → prediction round 2): The gelatinization degree of the rough sand substrate Δ=+1.20 (high) and the comprehensive enzyme activity at the catalytic end Δ=+1.07 (high). The causal chain deduction is as follows: excessive release of substrate + excess catalytic capacity → the first round of respiration intensity will continue to be high (predicted Δ≈+1.0) → the acidity of the second round will exceed the 1.5°T inhibition threshold (predicted acidity≈1.7~1.8°T).

[0144] Without intervention, the acidity in the second round is expected to reach 1.7~1.8°T, exceeding the 1.5°T amylase inhibition threshold, which will lead to a decrease in enzyme activity and hinder fermentation in subsequent rounds.

[0145] Feedforward regulation (executed in one round): Based on a coarse sand FPI of 0.90, the steaming time adjustment Δt is calculated using the control formula: 6 × 0.90 ≈ 5 minutes; adjust the moisture content of the fermented mash Δw = 0.6 × 0.90 ≈ 0.5 percentage points.

[0146] Control timing: After the rough sand is determined to be on the forward track, control is implemented in the first round (at which time the acidity in the second round has not yet exceeded the standard), realizing feedforward intervention.

[0147] Verification of the regulatory effect: The verification of the causal chain feedforward prediction and regulation effect is shown in the table below:

[0148] Product quality results: The yield of the second round of liquor was 8.28% (up 0.28 percentage points year-on-year), and the proportion of first-grade liquor was 90.5% (up 1.8 percentage points year-on-year).

[0149] in conclusion: (1) For the rough sand wheel, FPI=0.90, the 3 / 5 parameters are too high and the direction is consistent, so it is determined that the track has moved forward. The joint judgment of the 5 parameters avoids the misjudgment of occasional high single parameters.

[0150] (2) The causal chain feedforward prediction accurately predicted the risk of acidity exceeding the threshold in the second round. If no intervention is needed, the acidity is expected to be 1.7~1.8°T.

[0151] (3) Feedforward regulation was performed in the first round (one round before the deviation of the second round). The measured acidity in the second round was 1.4°T, which did not exceed the 1.5°T inhibition threshold, and the FPI regressed to 0.17 (typical trajectory).

[0152] (4) This method realizes a feedforward control closed loop of "upstream deviation → causal chain prediction → early intervention in one round → downstream parameters not exceeding the standard".

[0153] Example 2 Figure 2 This is a schematic diagram of the fermentation state determination and control device for Maotai-flavor liquor provided in an embodiment of this application. Figure 2 As shown, the device includes: The parameter acquisition module 201 is used to acquire the measured values ​​of each core parameter in the current round. The core parameters include: degree of gelatinization, comprehensive enzyme activity index, respiration intensity, acidity, and microbial diversity index. The calculation module 202, connected to the parameter acquisition module, is used to acquire the mean μ and standard deviation σ of each core parameter pre-calibrated based on historical batch data, calculate the standardized offset of each core parameter according to the measured value, mean μ and standard deviation σ, and perform weighted summation of each standardized offset according to the preset causal chain weights to calculate the fermentation progress index FPI, wherein the weights are assigned in descending order from upstream to downstream according to the position of the parameter in the causal chain. The discrimination module 203, connected to the calculation module, is used to determine the current fermentation trajectory type based on the value of the FPI and the consistency constraints of the offset direction of at least three core parameters. The trajectory type includes typical trajectory, forward offset trajectory, backward offset trajectory and undulating trajectory. The prediction module 204, connected to the discrimination module, is used to predict the offset trend of the downstream parameter in subsequent rounds based on the causal chain cascade relationship between the five core parameters and the offset of the upstream parameter in the current round. The control module 205, connected to the prediction module, is used to output a feedforward control command to the grain steaming equipment and / or moisture control equipment before the downstream parameters actually shift, based on the identified track type and the value of FPI. When shifting the track forward, the control amount decreases as FPI increases; when shifting the track backward, the control amount increases as the absolute value of FPI increases.

[0154] Furthermore, the parameter acquisition module specifically includes: The gelatinization degree detection unit, comprehensive enzyme activity detection unit, CO2 release rate online monitoring unit, acidity detection unit, and microbial diversity detection unit are used to detect the corresponding core parameters.

[0155] The fermentation state discrimination and control device for Maotai-flavor liquor provided in this application embodiment can realize the various processes of the above method embodiment and has corresponding beneficial effects. To avoid repetition, it will not be described again here.

[0156] Example 3 This application also provides an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the various processes of the above-described embodiment of the fermentation state discrimination and control device for soy sauce-flavored liquor and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0157] It should be noted that the electronic devices in the embodiments of this application include mobile electronic devices and non-mobile electronic devices as described above.

[0158] Example 4 This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described embodiment of the fermentation state discrimination and control device for soy sauce-flavored liquor and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0159] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0160] Example 5 This application also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described embodiment of the fermentation state discrimination and control device for Maotai-flavor liquor, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0161] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0162] It should be noted that, in this document, the term "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element. Furthermore, it should be noted that the scope of the methods and systems in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0163] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0164] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms fall within the scope of protection of this application.

[0165] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. A method for determining and controlling the fermentation state of soy sauce-flavored Baijiu, characterized in that, The method includes: Step 1: Based on historical batch data, the mean μ and standard deviation σ of each core parameter are calibrated for the nine fermentation cycles of the Maotai-flavor liquor process, and safe zone, warning zone and danger zone are defined; the core parameters include: degree of gelatinization, comprehensive enzyme activity index, respiration intensity, acidity and microbial diversity index; Step 2: Collect the measured values ​​of each core parameter in the current round, calculate the standardized offset of each core parameter based on the measured values ​​and the mean μ and standard deviation σ of each core parameter as pre-calibrated, and perform a weighted summation of each standardized offset according to the preset causal chain weights to calculate the fermentation progress index FPI, wherein the weights are assigned in descending order from upstream to downstream according to the position of the parameter in the causal chain. Step 3: Based on the value of FPI and the consistency constraint of the offset direction of at least three core parameters, determine the current orbit type of fermentation. The orbit type includes typical orbit, forward offset orbit, backward offset orbit, and fluctuating orbit. Step 4: Based on the causal chain cascade relationship between the five core parameters, use the offset of the upstream parameter in the current round to predict the offset trend of the downstream parameter in subsequent rounds. Step 5: Based on the identified track type and the value of FPI, perform feedforward control before the downstream parameters actually shift; wherein, when shifting the track forward, the control amount decreases as FPI increases; when shifting the track backward, the control amount increases as the absolute value of FPI increases.

2. The method according to claim 1, characterized in that, In step 2, the standardized offset Δ of each core parameter i Calculated using the formula: ； The fermentation progress index (FPI) is calculated using the following formula: ； Where w i is the causal chain weight corresponding to the i-th core parameter.

3. The method according to claim 2, characterized in that, The specific weights assigned to the causal chain are as follows: weight w1 for degree of gelatinization = 0.30, weight w2 for comprehensive enzyme activity index = 0.25, weight w3 for respiration intensity = 0.20, weight w4 for acidity = 0.15, and weight w5 for microbial diversity index = 0.

10.

4. The method according to claim 1, characterized in that, In step 3, the rule for determining the orbit type is: When |FPI| < 0.5, it is determined to be a typical orbit; When FPI ≥ 0.5 and there are at least one consecutive rounds, and at least three parameters Δi > 0, it is determined to be a forward offset track; When FPI ≤ -0.5 and there are at least one consecutive rounds, and at least three parameters Δi < 0, it is determined to be a backward deviation trajectory; When |FPI|≥0.5 and the FPI signs of adjacent rounds are reversed, and the parameter offset directions are inconsistent, it is determined to be a fluctuating trajectory.

5. The method according to claim 1, characterized in that, In step 4, the causal chain cascade relationship includes: Gelatinization degree shift predicts the next round of same-direction shift in the comprehensive enzyme activity index; The overall enzyme activity index shift predicts the next round of respiratory intensity shift in the same direction; Respiratory intensity shift predicts acidity shift in the same direction in the next round; Acidity shift predicts two rounds of inverse shifts in the microbial diversity index; The prediction formula is △ i+1 (downstream) = β × △ i (Upstream), where β is the causal transmission coefficient and 0 < β < 1.

6. The method according to claim 1, characterized in that, In step 5: Adjustment amount of steaming time when shifting the track forward Adjust the amount of moisture in the fermented mash. Where k1 ranges from 5 to 8 minutes per unit of FPI, and k2 ranges from 0.5 to 1.0 percentage points per unit of FPI; Adjustment amount of steaming time when shifting the track backward Adjust the amount of moisture in the fermented mash. .

7. The method according to claim 6, characterized in that, Step 5 further includes: When the trajectory fluctuates, reduce the amplitude of a single operation by 30% to 50%, check the consistency of the process, and do not adjust the steaming time and moisture for the time being.

8. The method according to claim 1, characterized in that, In step 1, the safe zone is defined as μ±1.0σ, the warning zone is defined as μ±1.5σ, and the danger zone is defined as the range beyond μ±1.5σ.

9. A device for determining and controlling the fermentation state of soy sauce-flavored Baijiu, characterized in that, include: The parameter acquisition module is used to collect the measured values ​​of each core parameter in the current round. The core parameters include: degree of gelatinization, comprehensive enzyme activity index, respiration intensity, acidity, and microbial diversity index. The calculation module, connected to the parameter acquisition module, is used to obtain the mean μ and standard deviation σ of each core parameter pre-calibrated based on historical batch data, calculate the standardized offset of each core parameter according to the measured value, mean μ and standard deviation σ, and perform weighted summation of each standardized offset according to the preset causal chain weights to calculate the fermentation progress index FPI, wherein the weights are assigned in descending order from upstream to downstream according to the position of the parameter in the causal chain. The discrimination module, connected to the calculation module, is used to determine the current fermentation trajectory type based on the value of the FPI and the consistency constraints of the offset direction of at least three core parameters. The trajectory type includes typical trajectory, forward offset trajectory, backward offset trajectory, and fluctuating trajectory. The prediction module, connected to the discrimination module, is used to predict the offset trend of the downstream parameter in subsequent rounds based on the causal chain cascade relationship between the five core parameters and the offset of the upstream parameter in the current round. The control module, connected to the prediction module, is used to output a feedforward control command to the grain steaming equipment and / or moisture control equipment before the downstream parameters actually shift, based on the identified track type and the value of FPI. When shifting the track forward, the control amount decreases as FPI increases; when shifting the track backward, the control amount increases as the absolute value of FPI increases.

10. The apparatus according to claim 9, characterized in that, The parameter acquisition module specifically includes: The gelatinization degree detection unit, comprehensive enzyme activity detection unit, CO2 release rate online monitoring unit, acidity detection unit, and microbial diversity detection unit are used to detect the corresponding core parameters.

Images