A method and system for fault monitoring of the microbial fermentation process in food production

Abstract

The present invention relates to the technical field of food production, and specifically discloses a method and system for fault monitoring of the microbial fermentation process in food production. The method uses an optical fiber dissolved oxygen sensor to collect the dissolved oxygen amounts in different regions, calculates through the coefficient of variation and the Euclidean distance model to evaluate the uniformity of yeast growth, and determines whether the growth is stable and uniform. When the yeast growth is uneven, it screens the growth-limited and normal regions according to the dissolved oxygen amount, analyzes the influence of the oxygen supply amount on yeast growth through the Pearson distance model. If the oxygen supply amount causes yeast growth limitation, a two-dimensional space model of the region is constructed, grouped according to the central point coordinates of the growth-limited region, and the oxygen supply adjustment amounts are calculated respectively, solving the problems of how to accurately evaluate the uniformity of yeast growth, determine the reasons for growth limitation, and targetedly adjust the oxygen supply, etc., so as to be able to timely detect fermentation abnormalities, accurately locate the root cause of the problem, effectively adjust the oxygen supply to optimize the yeast growth environment, and improve the fermentation efficiency and product quality.

Description

Technical Field

[0001] The present invention relates to the technical field of food production, and in particular to a method and system for monitoring faults in a microbial fermentation process of food production. Background Art

[0002] In the microbial fermentation process of food production, such as bread fermentation, wine fermentation and soy sauce fermentation, the aerobic fermentation stage plays a key role in the growth of yeast and the formation of metabolites. However, in the actual fermentation process, different fermentation areas may have problems such as uneven dissolved oxygen and inconsistent oxygen flow, which in turn limits yeast growth and affects fermentation efficiency and product quality.

[0003] Existing monitoring methods are often unable to accurately evaluate the growth status of yeast, determine the cause of growth restriction and make effective adjustments. Therefore, the present application uses a fiber optic dissolved oxygen sensor to collect dissolved oxygen in different areas, calculates the coefficient of variation and the Euclidean distance model, evaluates the uniformity of yeast growth, and determines whether the growth is stable and uniform. When the yeast grows unevenly, the growth restricted and normal areas are screened according to the dissolved oxygen content, and the effect of oxygen flow on yeast growth is analyzed by the Pearson distance model. If the oxygen flow causes yeast growth restriction, a two-dimensional spatial model of the region is constructed, and the growth restricted areas are grouped according to the coordinates of the center points of the growth restricted areas, and the oxygen flow adjustment amount is calculated separately. This solves the problems of how to accurately evaluate the uniformity of yeast growth, determine the cause of growth restriction and make targeted adjustments to oxygen flow, so that fermentation abnormalities can be discovered in a timely manner, the root cause of the problem can be accurately located, and the oxygen flow can be effectively adjusted to optimize the yeast growth environment and improve the fermentation efficiency and product quality. Summary of the invention

[0004] The object of the present invention is to provide a method and system for monitoring faults in a microbial fermentation process of food production, so as to solve the above-mentioned problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] In a first aspect, a method for monitoring a microbial fermentation process failure in food production comprises the following steps:

[0007] During the aerobic fermentation cycle, the dissolved oxygen content of the fermentation area was monitored during the divided monitoring period, and the coefficient of variation model and Euclidean distance model were used for comprehensive analysis to evaluate the uniformity of yeast growth;

[0008] If the yeast grows unevenly, the growth-restricted area and the normal growth area are screened out according to the dissolved oxygen content in the fermentation area, and the oxygen permeability of the growth-restricted area and the normal growth area are obtained respectively, and input into the Pearson distance model to determine whether the oxygen permeability affects the yeast growth;

[0009] If it affects yeast growth, a two-dimensional spatial model of the construction area is built. According to the coordinate positions of the center points of different growth-limited areas, it is judged whether the growth-limited areas can be adjusted centrally. If so, the centralized oxygen supply adjustment amount is obtained. If not, the single oxygen supply adjustment amount is obtained.

[0010] As a further solution of the present invention: the evaluation of yeast growth uniformity is carried out as follows:

[0011] According to the time series of the monitoring period, the dissolved oxygen amounts in the fermentation area are sorted and integrated to obtain the regional dissolution sequence, which is successively input into the coefficient of variation model and the Euclidean distance model, and the uniformity determination value is output;

[0012] If the uniformity determination value is greater than the uniformity determination threshold, a growth non-uniform signal is generated.

[0013] As a further solution of the present invention: the screening process of the growth-limited area and the growth-normal area is as follows:

[0014] The average value of the dissolved oxygen amounts in the fermentation area at different monitoring periods is calculated, and the dissolved oxygen average value of the period is output;

[0015] If the dissolved oxygen average value of the period is greater than or equal to the preset dissolved oxygen amount, it is recorded as the growth-normal area;

[0016] If the dissolved oxygen average value of the period is less than or equal to the preset dissolved oxygen amount, it is recorded as the growth-limited area.

[0017] As a further solution of the present invention: obtaining the oxygen supply amount sequences of the growth-limited and growth-normal areas and inputting them into the Pearson distance model, the execution process is as follows:

[0018] The oxygen supply amounts in the growth-limited area and the growth-normal area at different monitoring periods are respectively obtained and sorted according to the monitoring period order to obtain the limited oxygen supply sequence and the normal oxygen supply sequence;

[0019] Based on the same monitoring period as the benchmark, the oxygen supply amounts are respectively selected from the limited oxygen supply sequence and the normal oxygen supply sequence and input into the Pearson distance model, and the regional deviation value is output.

[0020] As a further solution of the present invention: judging whether the oxygen supply amount affects yeast growth, the judgment process is as follows:

[0021] If the regional deviation value is greater than the regional deviation threshold, an oxygen supply influence signal is generated.

[0022] As a further solution of the present invention: obtaining the adjacent distance group and the separated distance group according to the coordinate positions of the center points of different growth-limited areas, the execution process is as follows:

[0023] Within the regional two-dimensional space model, randomly combine the central point coordinates within two growth-restricted regions to obtain multiple sets of distance analysis groups, and input them into the coordinate distance calculation model respectively to obtain the restricted region spacing.

[0024] If the restricted region spacing is greater than the preset restricted spacing, it is a separated distance group.

[0025] If the restricted region spacing is less than or equal to the preset restricted spacing, it is an adjacent distance group.

[0026] As a further solution of the present invention: perform the following operations on the adjacent distance group:

[0027] Respectively obtain the oxygen ventilation amounts of the growth-restricted regions within the adjacent distance group during the same monitoring period, and input them into the Euclidean distance calculation model, and output to obtain the adjacent oxygen ventilation deviation value.

[0028] If the adjacent oxygen ventilation deviation value is greater than the preset adjacent oxygen ventilation deviation value, a single adjustment signal.

[0029] If the adjacent oxygen ventilation deviation value is less than or equal to the preset adjacent oxygen ventilation deviation value, a centralized adjustment signal.

[0030] As a further solution of the present invention: for the adjacent distance group, obtain the oxygen ventilation adjustment amount, and the execution process is as follows:

[0031] If a single adjustment signal is generated, respectively perform mean value calculation on the minimum oxygen ventilation amount and the maximum oxygen ventilation amount of the growth-restricted regions within the adjacent distance group, then subtract the preset oxygen ventilation amount, take the absolute value, and obtain the centralized oxygen ventilation adjustment amount.

[0032] If a centralized adjustment signal is generated, perform mean value calculation on the oxygen ventilation amounts of all growth-restricted regions within the adjacent distance group, then subtract the preset oxygen ventilation amount, take the absolute value, and obtain the centralized oxygen ventilation adjustment amount.

[0033] As a further solution of the present invention: perform the following operations on the separated distance group:

[0034] Within the separated distance group, respectively obtain the oxygen ventilation amounts of the growth-restricted regions during different monitoring periods, perform mean value calculation, then subtract the preset oxygen ventilation amount, take the absolute value, and obtain the single oxygen ventilation adjustment amount.

[0035] In a second aspect, a fault monitoring system for the microbial fermentation process of food production includes the following modules:

[0036] Uniform growth evaluation module: During the aerobic fermentation cycle, monitor the dissolved oxygen amount in the fermentation region during the divided monitoring period, and perform comprehensive analysis through the coefficient of variation model and the Euclidean distance model to evaluate the yeast growth uniformity.

[0037] Limited growth analysis module: If the yeast growth is uneven, screen out the growth-limited area and the normal growth area according to the dissolved oxygen content in the fermentation area, obtain the oxygen ventilation amounts of the growth-limited area and the normal growth area respectively, input them into the Pearson distance model, and judge whether the oxygen ventilation amount affects the yeast growth;

[0038] Oxygen ventilation limitation adjustment module: If it affects the yeast growth, construct a two-dimensional regional space model, judge whether the growth-limited areas can be adjusted centrally according to the central point coordinates of different growth-limited areas. If yes, obtain the centralized oxygen ventilation adjustment amount. If not, obtain the single oxygen ventilation adjustment amount.

[0039] Advantages of the present invention:

[0040] (1) During the aerobic fermentation cycle, the present invention uses an optical fiber dissolved oxygen sensor to monitor the dissolved oxygen content in different areas, divides the fermentation cycle into equally spaced monitoring time periods, integrates the dissolved oxygen contents in different areas of each time period to obtain a regional dissolution sequence, calculates the time period coefficient through a coefficient of variation model, quantifies the dispersion degree of the dissolved oxygen content in each time period to reflect the yeast growth situation, and then uses the Euclidean distance calculation model to obtain a uniformity determination value to reflect the difference in yeast growth uniformity between adjacent time periods, solving the problem of how to evaluate whether the yeast growth in different fermentation areas in the fermentation tank is uniform;

[0041] (2) The present invention screens out the growth-limited and normal areas according to the dissolved oxygen content in different monitoring time periods of the fermentation area, obtains and integrates the oxygen ventilation amount sequences of the two, inputs the oxygen ventilation amounts in the same monitoring time period into the Pearson distance model, and obtains the regional deviation value by calculating the covariance and standard deviation, so as to obtain the difference degree of the oxygen ventilation amounts between the growth-limited area and the normal growth area in the same monitoring time period through the regional deviation value, thereby indirectly reflecting that the yeast growth limitation in the growth-limited area is caused by the oxygen ventilation amount, further determining the cause of the growth-limited area, and providing data support for the subsequent adjustment and optimization of the growth-limited area;

[0042] (3) Based on the oxygen ventilation influence signal, the present invention constructs a two-dimensional regional space model for different fermentation areas in the fermentation tank, obtains the central point coordinates of the growth-limited areas and combines them into a distance analysis group, obtains the distance between the limited areas through the coordinate distance calculation model, and divides them into an adjacent distance group and a separated distance group accordingly. For the adjacent distance group, use the Euclidean distance calculation model to quantify the difference in oxygen ventilation amounts in different time periods to obtain the adjacent oxygen ventilation deviation value, and then calculate the adjustment sub-value of the limited area by calculating the average values of the minimum and maximum oxygen ventilation amounts in the group, and further calculate the oxygen ventilation adjustment amount; for the separated distance group, directly average the oxygen ventilation amounts in different time periods of each growth-limited area in the group to calculate the oxygen ventilation adjustment amount, and use these oxygen ventilation adjustment amounts to accurately adjust the oxygen ventilation flow rate of the growth-limited areas, solve the problem of yeast growth limitation caused by uneven oxygen ventilation amount in the aerobic fermentation stage, improve the adjustment efficiency of the yeast growth limitation problem, improve the poor fermentation condition caused by uneven oxygen ventilation amount, and optimize the yeast growth environment. Brief Description of the Drawings

[0043] The present invention will be further described below in conjunction with the accompanying drawings.

[0044] Figure 1 is a flowchart of the steps of a method for monitoring faults in the microbial fermentation process of food production according to the present invention;

[0045] Figure 2 is a schematic diagram of a system for monitoring faults in the microbial fermentation process of food production according to the present invention. Detailed Embodiments

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0047] Embodiment 1

[0048] Please refer to Figure 1 As shown, the present invention is a method for monitoring faults in the microbial fermentation process of food production. The microbial fermentation in food production includes: bread fermentation, wine fermentation, soy sauce fermentation, etc. In the microbial fermentation process of food production, there are also aerobic fermentation stages and anaerobic fermentation stages. However, for the entire wine fermentation process, the aerobic fermentation stage in the early stage is crucial for the growth of yeast and the formation of metabolites. Therefore, the dissolved oxygen content in different fermentation regions during the aerobic fermentation stage is compared and analyzed to monitor whether the growth of yeast in the aerobic fermentation stage is restricted, ensuring that the yeast in different fermentation regions can grow normally. The steps are as follows:

[0049] Step 1: During the aerobic fermentation cycle, monitor the dissolved oxygen content in different fermentation regions of the fermentation tank through an optical fiber dissolved oxygen sensor to evaluate whether the growth of yeast in different fermentation regions of the fermentation tank is uniform;

[0050] In a preferred embodiment, the aerobic fermentation cycle is divided into several monitoring periods with equal time intervals;

[0051] Obtain the dissolved oxygen content in different fermentation regions during the monitoring period in real time through the optical fiber dissolved oxygen sensor, and sort and integrate the dissolved oxygen content in different fermentation regions according to the time series to obtain a regional dissolution sequence;

[0052] It should be noted that the fiber optic dissolved oxygen sensors are arranged in different areas of the fermentation tank. The optical fiber can transmit optical signals to the sensor probe to measure the dissolved oxygen in each area. And based on the fluorescence quenching principle, the fluorescent substance in the sensor emits fluorescence under the action of the excitation light. After the oxygen molecules interact with the fluorescent substance, the fluorescence intensity decreases. The dissolved oxygen amount is obtained by measuring the change in fluorescence intensity;

[0053] Arbitrarily select the regional dissolution sequence corresponding to the monitoring period and input it into the coefficient of variation model. The process is as follows:

[0054] A1. Calculate the average value of the dissolved oxygen amounts in all fermentation areas within the regional dissolution sequence, and output the regional dissolved oxygen average value ;

[0055] A2. Calculate the standard deviation of the dissolved oxygen amounts in all fermentation areas within the regional dissolution sequence, and output the regional dissolved oxygen standard deviation value ;

[0056] A3. Input the regional dissolved oxygen average value and the regional dissolved oxygen standard deviation value into the coefficient of variation model (Equation 1), and output the period coefficient;

[0057] Equation 1: ;

[0058] It can be understood that the purpose of the coefficient of variation model is that the average values of the dissolved oxygen amounts in different fermentation areas may be different. Simply comparing the standard deviations cannot accurately judge the dispersion degree of the dissolved oxygen amounts in each area. However, the coefficient of variation standardizes the standard deviation relative to the average value, which can better compare the fluctuations of the dissolved oxygen amounts in different areas and reflect to a certain extent the yeast growth situation between different fermentation areas during the monitoring period;

[0059] Input the period coefficients corresponding to adjacent monitoring periods within the aerobic fermentation cycle into the Euclidean distance calculation model (Equation 2), and output the uniformity determination value , and the process is as follows:

[0060] Equation 2: ;

[0061] Among them, represents one less than the total number of monitoring periods within the aerobic fermentation cycle, , respectively represent the period coefficients corresponding to adjacent monitoring periods;

[0062] It can be understood that the purpose of the Euclidean distance calculation model is as follows: the time period coefficient is obtained through the coefficient of variation model, which reflects the relative dispersion degree of the dissolved oxygen content in different fermentation regions within each monitoring time period, indirectly reflecting the uniformity of yeast growth within that time period. Inputting the time period coefficients of adjacent monitoring time periods into the Euclidean distance calculation model reflects the difference in the uniformity of yeast growth between two adjacent time periods;

[0063] Specifically, the meaning represented by the uniformity determination value is as follows: it is calculated by quantifying the time period coefficient, reflecting the stability degree of the change in the time period coefficient by analyzing different monitoring time periods in the time dimension, and embodying the stability of whether the yeast growth is uniform. Specifically, the smaller the value, the more stable the change in the uniformity of yeast growth between adjacent time periods, that is, the yeast growth is relatively stable and uniform; the larger the value, the more unstable the change in the uniformity of yeast growth between adjacent time periods, indicating that there may be some factors affecting the growth uniformity of yeast in different regions;

[0064] Compare the uniformity determination value with the uniformity determination threshold, and the process is as follows:

[0065] If the uniformity determination value is less than or equal to the uniformity determination threshold, it indicates that the yeast growth is relatively stable and uniform, and a growth uniformity signal is generated;

[0066] If the uniformity determination value is greater than the uniformity determination threshold, it indicates that the change in the uniformity of yeast growth between adjacent time periods is relatively unstable, and a growth non-uniformity signal is generated;

[0067] It should be noted that the uniformity determination threshold is set by those skilled in the art;

[0068] Summary of the solution of this embodiment: During the aerobic fermentation cycle, use an optical fiber dissolved oxygen sensor to monitor the dissolved oxygen content in different regions, divide the fermentation cycle into equally spaced monitoring time periods, integrate the dissolved oxygen content in different regions of each time period to obtain a regional dissolution sequence, calculate the time period coefficient through the coefficient of variation model, quantify the dispersion degree of the dissolved oxygen content in each time period to reflect the yeast growth situation, and then use the Euclidean distance calculation model to obtain the uniformity determination value, reflecting the difference in the uniformity of yeast growth between adjacent time periods, and solving the problem of how to evaluate whether the yeast growth in different fermentation regions in the fermentation tank is uniform.

[0069] Embodiment 2

[0070] Please refer to Figure 1As shown, the present invention is a method for monitoring faults in the microbial fermentation process of food production. According to the yeast growth-limited regions screened in the first embodiment, the reasons for the limitation are investigated. For example, if the reason for the limitation is uneven oxygen supply between different fermentation regions, it will cause changes in the yeast metabolic pathway. In addition to producing alcohol, some by-products may also accumulate, such as acetaldehyde, acetic acid, glycerol, etc. The accumulation of these by-products may be toxic to yeast cells and inhibit the growth and metabolism of yeast. Therefore, taking the oxygen supply as an example, it is analyzed whether the oxygen supply affects the yeast growth limitation in the yeast growth-limited region. Therefore, the following steps need to be performed:

[0071] Step 2: If it is uneven, then according to the dissolved oxygen amounts in different fermentation regions at different monitoring time periods, the growth-limited regions are screened out, and the oxygen supply amounts between the growth-limited regions and the regions with normal growth are compared and analyzed to determine whether the oxygen supply causes the yeast growth limitation in the growth-limited regions. If the oxygen supply causes it, an oxygen supply influence signal is generated;

[0072] In a preferred embodiment, any one fermentation region is selected randomly;

[0073] The dissolved oxygen amounts in the fermentation region at different monitoring time periods are extracted, and an averaging calculation is performed to output the average dissolved oxygen amount for the time period;

[0074] If the average dissolved oxygen amount for the time period is greater than or equal to the preset dissolved oxygen amount, then the yeast in the fermentation region grows normally at different monitoring time periods, that is, it is a region with normal growth;

[0075] If the average dissolved oxygen amount for the time period is less than or equal to the preset dissolved oxygen amount, then the yeast in the fermentation region is growth-limited at different monitoring time periods, that is, it is a growth-limited region;

[0076] Obtain the oxygen supply amounts in the growth-limited regions at different monitoring time periods, and sort them according to the time series, and correspondingly integrate them into a limited oxygen supply sequence;

[0077] Obtain the oxygen supply amounts in the regions with normal growth at different monitoring time periods, and sort them according to the time series, and correspondingly integrate them into a normal oxygen supply sequence;

[0078] Randomly select one growth-limited region and one region with normal growth;

[0079] Input the oxygen supply amounts corresponding to the same monitoring time periods in the limited oxygen supply sequence and the normal oxygen supply sequence into the Pearson distance model, and output the regional deviation value ;

[0080] It should be noted that the meaning of the regional deviation value is as follows: it reflects the difference in oxygen passing amount between the growth-restricted area and the normal growth area during the same monitoring period, thereby indirectly reflecting that the yeast growth restriction in the growth-restricted area is caused by the oxygen passing amount, further determining the cause of the growth-restricted area, and providing data support for the subsequent adjustment and optimization of the growth-restricted area;

[0081] Specifically, the total number of oxygen passing amount elements in the restricted oxygen passing sequence is the same as that in the normal oxygen passing sequence;

[0082] Input into the Pearson distance model, and the steps are as follows:

[0083] B1. Obtain the covariance of the restricted oxygen passing sequence and the normal oxygen passing sequence through the covariance calculation formula (Formula III) ;

[0084] Formula III: ;

[0085] Among them, m represents the total number of the restricted oxygen passing sequence or the normal oxygen passing sequence, represents the oxygen passing amount corresponding to the th monitoring period in the restricted oxygen passing sequence, represents the average dissolved oxygen value of the corresponding period in the growth-restricted area, represents the oxygen passing amount corresponding to the th monitoring period in the normal oxygen passing sequence, represents the average dissolved oxygen value of the corresponding period in the normal growth area;

[0086] B2. Obtain the standard deviations of the restricted oxygen passing sequence and the normal oxygen passing sequence respectively through the standard deviation calculation formula 、 ;

[0087] B3. Input the standard deviation of the restricted oxygen passing sequence, the standard deviation of the normal oxygen passing sequence, and the covariance into the Pearson distance model formula (Formula IV), specifically as follows:

[0088] Formula IV: ;

[0089] Specifically, the purpose of inputting into the Pearson distance model is to quantify the difference in oxygen ventilation volume between the growth-restricted area and the normal growth area within the same monitoring period. It comprehensively considers the covariance of the oxygen ventilation sequences of the two areas and their respective standard deviations, reflecting the deviation of the oxygen ventilation volume in different areas and different periods as a whole, and indirectly reflecting the difference in oxygen ventilation volume between the growth-restricted area and the normal growth area. Furthermore, it can determine whether the yeast growth restriction in the growth-restricted area is caused by the oxygen ventilation volume, providing strong data support for the subsequent adjustment and optimization of the growth-restricted area, and helping to take targeted measures, such as adjusting the oxygen ventilation volume, to improve the fermentation process, enhance the uniformity of yeast growth, and fermentation efficiency;

[0090] Compare the regional deviation value with the regional deviation threshold. The process is as follows:

[0091] If the regional deviation value is greater than the regional deviation threshold, it indicates that the difference in oxygen ventilation volume between the growth-restricted area and the normal growth area is large, and an oxygen ventilation influence signal is generated;

[0092] If the regional deviation value is less than or equal to the regional deviation threshold, it indicates that the difference in oxygen ventilation volume between the growth-restricted area and the normal growth area is small, and an oxygen ventilation non-influence signal is generated;

[0093] Summary of the solution in this embodiment: Screen the growth-restricted and normal areas according to the dissolved oxygen volume in different monitoring periods of the fermentation area, obtain and integrate the oxygen ventilation sequences of the two, input the oxygen ventilation volume in the same monitoring period into the Pearson distance model, calculate the covariance and standard deviation to obtain the regional deviation value, so as to obtain the difference in oxygen ventilation volume between the growth-restricted area and the normal growth area in the same monitoring period through the regional deviation value, thereby indirectly reflecting that the yeast growth restriction in the growth-restricted area is caused by the oxygen ventilation volume, further determining the cause of the growth-restricted area, and providing data support for the subsequent adjustment and optimization of the growth-restricted area.

[0094] Embodiment 3

[0095] Please refer to Figure 1 As shown, the present invention is a method for monitoring faults in the microbial fermentation process of food production. According to the analysis in Embodiment 2, the yeast growth restriction in the growth-restricted area is caused by the oxygen ventilation volume. Therefore, it is also necessary to adjust the oxygen ventilation volume in the growth-restricted area according to the spatial position of different growth-restricted areas, so as to avoid the restricted growth of yeast in the growth-restricted area, resulting in insufficient yeast quantity or inhibited activity, which leads to a slowdown in the conversion rate of fermentation substrates into products and a decrease in the amount of alcohol produced in the later anaerobic stage. Therefore, the following steps need to be further performed:

[0096] Step 3: Based on the oxygen ventilation influence signal, obtain the oxygen ventilation flow rate adjustment amount and adjust the oxygen ventilation flow rate in the growth-restricted area;

[0097] In a preferred embodiment, a two-dimensional spatial model of the region is constructed based on different fermentation regions in the fermenter;

[0098] It can be understood that the two-dimensional spatial model of the region includes a region with normal growth and a region with restricted growth;

[0099] In the two-dimensional spatial model of the region, the coordinates of the central points in all regions with restricted growth are obtained, and the coordinates of the central points in any two regions with restricted growth are combined and connected to obtain multiple sets of distance analysis groups;

[0100] Arbitrarily select a set of distance analysis groups;

[0101] Input the distance analysis group into the coordinate distance calculation model (Equation Five) to obtain the distance between restricted regions ;

[0102] Equation Five: ;

[0103] Wherein, 、 are the coordinates of the central points in the regions with restricted growth within the distance analysis group;

[0104] Compare the distance between restricted regions corresponding to the distance analysis group with a preset restricted distance. The specific process is as follows:

[0105] If the distance between restricted regions is greater than the preset restricted distance, it indicates that the distance between the regions with restricted growth within the distance analysis group is relatively large, and the distance analysis group is marked as a separated distance group;

[0106] If the distance between restricted regions is less than or equal to the preset restricted distance, it indicates that the distance between the regions with restricted growth within the distance analysis group is relatively small, and the distance analysis group is marked as an adjacent distance group;

[0107] It should be noted that the preset restricted distance is twice the distance between the central points of adjacent fermentation regions, where adjacent fermentation regions include, but are not limited to, adjacent regions with normal growth, adjacent regions with restricted growth, and regions with normal growth and regions with restricted growth;

[0108] Exemplarily, for the adjacent distance group, obtain the oxygen supply adjustment amount and perform the following process:

[0109] In the adjacent distance group, obtain the oxygen supply amounts of the regions with restricted growth at different monitoring time periods respectively, and input them into the Euclidean distance calculation model (Equation Six) to output the adjacent oxygen supply deviation value ;

[0110] Equation Six: ;

[0111] Among them, is expressed as the adjacent oxygen passing deviation value corresponding to the th adjacent distance group, is expressed as the total number of the separated distance groups, is expressed as the oxygen passing amount corresponding to one of the restricted areas in the th adjacent distance group at the th monitoring period, is expressed as the oxygen passing amount corresponding to another restricted area in the th adjacent distance group at the th monitoring period, is expressed as the total number of the adjacent distance groups;

[0112] It should be noted that the purpose of using the Euclidean distance calculation model is as follows: to quantify the difference degree of the oxygen passing amounts of adjacent growth restricted areas at different time periods, accurately reflect the discrete situation of the oxygen passing amounts of these areas numerically, assist in determining the oxygen passing adjustment strategy for adjacent growth restricted areas, improve the adjustment efficiency of yeast growth restriction in the growth restricted areas, and solve the problem of yeast growth restriction caused by uneven oxygen passing amounts;

[0113] If the adjacent oxygen passing deviation value is greater than the preset adjacent oxygen passing deviation value, then a single adjustment signal;

[0114] When generating a single adjustment signal, respectively, the minimum oxygen passing amount and the maximum oxygen passing amount of the growth restricted areas in the adjacent distance group are averaged, and then the difference is taken with the preset oxygen passing amount, and the absolute value is taken to obtain the oxygen passing adjustment amount;

[0115] If the adjacent oxygen passing deviation value is less than or equal to the preset adjacent oxygen passing deviation value, then a centralized adjustment signal;

[0116] When generating a centralized adjustment signal, the oxygen passing amounts of all the growth restricted areas in the adjacent distance group are averaged, and then the difference is taken with the preset oxygen passing amount, and the absolute value is taken to obtain the oxygen passing adjustment amount;

[0117] Specifically, the minimum oxygen passing amount of the growth restricted areas in the adjacent distance group is the minimum value extracted from the restricted oxygen passing sequence;

[0118] Similarly, the maximum oxygen passing amount of the growth restricted areas in the adjacent distance group is the maximum value extracted from the restricted oxygen passing sequence;

[0119] It is further explained that the restricted area adjustment sub - value only corresponds to one restricted area. If the restricted area is repeated, the averaging calculation is not performed to obtain the oxygen passing adjustment amount;

[0120] Exemplarily, for the separated distance group, to obtain the oxygen passing adjustment amount, the following process is executed:

[0121] Within the separated distance group, obtain the oxygen ventilation amounts of the growth-restricted regions during different monitoring time periods respectively, and perform averaging calculation to obtain the oxygen ventilation adjustment amount.

[0122] Summary of this embodiment: Based on the oxygen ventilation influence signal, construct a regional two-dimensional space model with different fermentation regions in the fermenter, obtain the central point coordinates of the growth-restricted regions and combine them into a distance analysis group. Through the coordinate distance calculation model, obtain the distance between the restricted regions, and accordingly divide them into adjacent distance groups and separated distance groups. For the adjacent distance group, use the Euclidean distance calculation model to quantify the difference in oxygen ventilation amounts at different time periods to obtain the adjacent oxygen ventilation deviation value, and then calculate the average values of the minimum and maximum oxygen ventilation amounts within the group to obtain the restricted region adjustment sub-value, and further calculate the oxygen ventilation adjustment amount. For the separated distance group, directly perform averaging calculation on the oxygen ventilation amounts of each growth-restricted region in the group at different time periods to obtain the oxygen ventilation adjustment amount. Use these oxygen ventilation adjustment amounts to accurately adjust the oxygen ventilation flow rate of the growth-restricted regions, solve the problem of yeast growth restriction caused by uneven oxygen ventilation amounts during the aerobic fermentation stage, improve the adjustment efficiency of the yeast growth restriction problem, improve the poor fermentation condition caused by uneven oxygen ventilation amounts, and optimize the yeast growth environment.

[0123] Embodiment 4

[0124] Please refer to Figure 1 As shown, the present invention is a fault monitoring system for the microbial fermentation process in food production, including the following modules:

[0125] Uniform growth evaluation module: During the aerobic fermentation cycle, monitor the dissolved oxygen amount in the fermentation region during the monitored time period after division, and perform comprehensive analysis through the coefficient of variation model and the Euclidean distance model to evaluate the yeast growth uniformity.

[0126] In a preferred embodiment, divide the aerobic fermentation cycle into several monitored time periods with equal time intervals.

[0127] Obtain the dissolved oxygen amounts of different fermentation regions during the monitored time period in real time through an optical fiber dissolved oxygen sensor, and sort and integrate the dissolved oxygen amounts of different fermentation regions according to the time series to obtain a regional dissolution sequence.

[0128] Arbitrarily select the regional dissolution sequence corresponding to the monitored time period, input it into the coefficient of variation model, and then input it into the Euclidean distance calculation model, and output to obtain a uniform determination value.

[0129] If the uniform determination value is less than or equal to the uniform determination threshold, it indicates that the yeast growth is relatively stable and uniform, and a growth uniformity signal is generated.

[0130] If the uniform determination value is greater than the uniform determination threshold, it indicates that the change in yeast growth uniformity between adjacent time periods is relatively unstable, and a growth non-uniformity signal is generated.

[0131] Limited growth analysis module: If the yeast growth is uneven, screen out the growth-limited area and the normal growth area according to the dissolved oxygen content in the fermentation area, obtain the oxygen ventilation amounts of the growth-limited area and the normal growth area respectively, input them into the Pearson distance model, and judge whether the oxygen ventilation amount affects the yeast growth;

[0132] Extract the dissolved oxygen content in the fermentation area at different monitoring time periods, perform averaging calculation, and output the average dissolved oxygen content in the time period;

[0133] If the average dissolved oxygen content in the time period is greater than or equal to the preset dissolved oxygen content, the yeast in the fermentation area grows normally during different monitoring time periods, that is, the normal growth area;

[0134] If the average dissolved oxygen content in the time period is less than or equal to the preset dissolved oxygen content, the yeast in the fermentation area is growth-limited during different monitoring time periods, that is, the growth-limited area;

[0135] Obtain the oxygen ventilation amounts in the growth-limited area at different monitoring time periods, sort them according to the time series, and correspondingly integrate them into a limited oxygen ventilation sequence;

[0136] Obtain the oxygen ventilation amounts in the normal growth area at different monitoring time periods, sort them according to the time series, and correspondingly integrate them into a normal oxygen ventilation sequence;

[0137] Arbitrarily select a growth-limited area and a normally growing area;

[0138] Input the oxygen ventilation amounts corresponding to the same monitoring time period in the limited oxygen ventilation sequence and the normal oxygen ventilation sequence into the Pearson distance model, and output the regional deviation value;

[0139] If the regional deviation value is greater than the regional deviation threshold, it indicates that the difference in oxygen ventilation amounts between the growth-limited area and the normal growth area is relatively large, and a oxygen ventilation influence signal is generated;

[0140] If the regional deviation value is less than or equal to the regional deviation threshold, it indicates that the difference in oxygen ventilation amounts between the growth-limited area and the normal growth area is relatively small, and a oxygen ventilation non-influence signal is generated;

[0141] Oxygen ventilation limited adjustment module: If it affects the yeast growth, construct a regional two-dimensional space model, obtain the adjacent distance group and the separated distance group according to the center point coordinate positions of different growth-limited areas, and respectively obtain the oxygen ventilation adjustment amounts for the adjacent distance group and the separated distance group;

[0142] Based on different fermentation areas in the fermentation tank, construct a regional two-dimensional space model;

[0143] In the regional two-dimensional space model, obtain the center point coordinates in all growth-limited areas, arbitrarily combine and connect the center point coordinates in two growth-limited areas to obtain multiple groups of distance analysis groups;

[0144] Arbitrarily select a set of distance analysis groups;

[0145] Input the distance analysis group into the coordinate distance calculation model (Equation 5) to obtain the restricted area spacing;

[0146] If the restricted area spacing is greater than the preset restricted spacing, it indicates that the spacing between the growth restricted areas in the distance analysis group is large, then mark the distance analysis group as the separated distance group;

[0147] If the restricted area spacing is less than or equal to the preset restricted spacing, it indicates that the spacing between the growth restricted areas in the distance analysis group is small, then mark the distance analysis group as the adjacent distance group;

[0148] For the adjacent distance group, respectively obtain the oxygen ventilation amounts of the growth restricted areas at different monitoring time periods, and input them all into the Euclidean distance calculation model, and output to obtain the adjacent oxygen ventilation deviation value;

[0149] If the adjacent oxygen ventilation deviation value is greater than the preset adjacent oxygen ventilation deviation value, then generate a single adjustment signal;

[0150] When generating a single adjustment signal, respectively select the minimum oxygen ventilation amount and the maximum oxygen ventilation amount of the growth restricted areas in all adjacent distance groups, and perform averaging calculations respectively to obtain the oxygen ventilation adjustment amount;

[0151] If the adjacent oxygen ventilation deviation value is less than or equal to the preset adjacent oxygen ventilation deviation value, then generate a centralized adjustment signal;

[0152] Respectively select the minimum oxygen ventilation amount and the maximum oxygen ventilation amount of the growth restricted areas in all adjacent distance groups, and perform averaging calculations respectively to obtain the restricted area adjustment sub-value;

[0153] Perform averaging calculations on all restricted area adjustment sub-values to obtain the oxygen ventilation adjustment amount;

[0154] The minimum oxygen ventilation amount of the growth restricted areas in the adjacent distance group is to extract the minimum value in the restricted oxygen ventilation sequence;

[0155] Similarly, the maximum oxygen ventilation amount of the growth restricted areas in the adjacent distance group is to extract the maximum value in the restricted oxygen ventilation sequence;

[0156] For the separated distance group, respectively obtain the oxygen ventilation amounts of the growth restricted areas at different monitoring time periods, and perform averaging calculations to obtain the oxygen ventilation adjustment amount.

[0157] Concept of the technical solution of the present invention: The dissolved oxygen content in different regions is collected by an optical fiber dissolved oxygen sensor, and the uniformity of yeast growth is evaluated through the calculation of the coefficient of variation and the Euclidean distance model to determine whether the growth is stable and uniform. When the yeast growth is uneven, the growth-limited and normal regions are screened according to the dissolved oxygen content, and the influence of the oxygen supply amount on yeast growth is analyzed through the Pearson distance model. If the oxygen supply amount causes yeast growth limitation, a two-dimensional space model of the region is constructed, grouped according to the central point coordinates of the growth-limited region, and the oxygen supply adjustment amount is calculated respectively, solving the problems of how to accurately evaluate the uniformity of yeast growth, determine the reasons for growth limitation, and targetedly adjust the oxygen supply, etc., so as to be able to timely detect fermentation abnormalities, accurately locate the root cause of the problem, effectively adjust the oxygen supply to optimize the yeast growth environment, and improve the fermentation efficiency and product quality.

[0158] The above has described in detail an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope covered by the present invention.

Claims

1. A method for monitoring faults in the microbial fermentation process of food production, characterized in that: It includes the following steps: Divide the aerobic fermentation cycle into several monitoring periods with equal time intervals, obtain the dissolved oxygen content in different fermentation regions during the monitoring periods, sort and integrate the dissolved oxygen content in different fermentation regions according to the time series to obtain the regional dissolution sequence. Arbitrarily select the regional dissolution sequence corresponding to the monitoring period, input it into the coefficient of variation model, and output the period coefficient. Input the period coefficients corresponding to adjacent monitoring periods into the Euclidean distance calculation model, and output the uniformity determination value for evaluating the yeast growth uniformity; If the yeast growth is uneven, the growth-limited area and the normal growth area are selected according to the dissolved oxygen content in the fermentation area. The oxygen passing amounts in the growth-limited area and the normal growth area at different monitoring time periods are obtained respectively, and sorted in the order of the monitoring time periods to obtain a restricted oxygen passing sequence and a normal oxygen passing sequence. The covariance between the restricted oxygen passing sequence and the normal oxygen passing sequence is obtained through the covariance calculation formula ; , where m represents the total number of the restricted oxygen passing sequence or the normal oxygen passing sequence, represents the oxygen passing amount corresponding to the th monitoring time period in the restricted oxygen passing sequence, represents the average dissolved oxygen content in the corresponding time period of the growth-limited area, represents the oxygen passing amount corresponding to the th monitoring time period in the normal oxygen passing sequence, represents the average dissolved oxygen content in the corresponding time period of the normal growth area; The standard deviation of the restricted oxygen supply sequence , the standard deviation of the normal oxygen supply sequence and the covariance are input into the Pearson distance model formula, and the output is the regional deviation value , which is used to determine whether the oxygen supply affects yeast growth; If it affects yeast growth, construct a regional two-dimensional space model, and based on the central point coordinate positions of different growth-limited regions, determine whether the growth-limited regions can be adjusted centrally. If so, obtain the centralized oxygen supply adjustment amount. If not, obtain the single oxygen supply adjustment amount.

2. The method for monitoring faults in the microbial fermentation process of food production according to claim 1, wherein: If the uniformity determination value is greater than the uniformity determination threshold, generate a growth non-uniformity signal.

3. The method for monitoring the faults in the microbial fermentation process of food production according to claim 1, wherein: The screening process of the growth-limited regions and the normal growth regions is as follows: Calculate the average value of the dissolved oxygen content in the fermentation region during different monitoring periods, and output the period dissolved oxygen average value; If the period dissolved oxygen average value is greater than or equal to the preset dissolved oxygen content, it is recorded as a normal growth region; If the period dissolved oxygen average value is less than or equal to the preset dissolved oxygen content, it is recorded as a growth-limited region.

4. The method for monitoring faults in the microbial fermentation process of food production according to claim 1, characterized in that: Judge whether the oxygen supply amount affects yeast growth. The judgment process is as follows: If the regional deviation value is greater than the regional deviation threshold, generate an oxygen supply influence signal.

5. A method for monitoring faults in the microbial fermentation process of food production according to claim 1, characterized in that: Based on the central point coordinate positions of different growth-limited regions, obtain the adjacent distance group and the separated distance group. The execution process is as follows: In the regional two-dimensional space model, arbitrarily combine the central point coordinates in two growth-limited regions to obtain multiple distance analysis groups, and input them into the coordinate distance calculation model respectively to obtain the distance between the limited regions; If the distance between the limited regions is greater than the preset limited distance, it is the separated distance group; If the distance between the limited regions is less than or equal to the preset limited distance, it is the adjacent distance group.

6. A method for monitoring faults in the microbial fermentation process of food production according to claim 5, characterized in that: Perform the following operations on the adjacent distance group: Respectively obtain the oxygen supply amounts of the growth-limited regions in the adjacent distance group during the same monitoring period, and input them into the Euclidean distance calculation model, and output the adjacent oxygen supply deviation value; If the adjacent oxygen supply deviation value is greater than the preset adjacent oxygen supply deviation value, generate a single adjustment signal; If the adjacent oxygen supply deviation value is less than or equal to the preset adjacent oxygen supply deviation value, generate a centralized adjustment signal.

7. A method for monitoring faults in the microbial fermentation process of food production according to claim 6, characterized in that: For the adjacent distance group, obtain the oxygen supply adjustment amount. The execution process is as follows: If a single adjustment signal is generated, respectively calculate the average value of the minimum oxygen supply amount and the maximum oxygen supply amount of the growth-limited regions in the adjacent distance group, subtract the preset oxygen supply amount from it, and take the absolute value to obtain the centralized oxygen supply adjustment amount; If a centralized adjustment signal is generated, calculate the average value of the oxygen supply amounts of all the growth-limited regions in the adjacent distance group, subtract the preset oxygen supply amount from it, and take the absolute value to obtain the centralized oxygen supply adjustment amount.

8. A method for monitoring faults in the microbial fermentation process of food production according to claim 5, characterized in that: Perform the following operations on the separated distance group: In the separated distance group, respectively obtain the oxygen supply amounts of the growth-limited regions during different monitoring periods, calculate the average value, subtract the preset oxygen supply amount from it, and take the absolute value to obtain the single oxygen supply adjustment amount.

9. A fault monitoring system for the microbial fermentation process of food production, characterized in that: It includes the following modules: Uniform growth assessment module: Divide the aerobic fermentation cycle into several monitoring periods with equal time intervals, obtain the dissolved oxygen content in different fermentation regions during the monitoring periods, sort and integrate the dissolved oxygen content in different fermentation regions according to the time series to obtain the regional dissolution sequence. Arbitrarily select the regional dissolution sequence corresponding to the monitoring period, input it into the coefficient of variation model, output the period coefficient, input the period coefficients corresponding to adjacent monitoring periods into the Euclidean distance calculation model, output the uniform determination value, and then evaluate the uniformity of yeast growth; Restricted growth analysis module: If the yeast growth is uneven, the growth-restricted area and the normal growth area are screened according to the dissolved oxygen content in the fermentation area. The oxygen ventilation amounts in the growth-restricted area and the normal growth area at different monitoring time periods are obtained respectively, and sorted in the order of the monitoring time periods to obtain the restricted oxygen ventilation sequence and the normal oxygen ventilation sequence. The covariance between the restricted oxygen ventilation sequence and the normal oxygen ventilation sequence is obtained through the covariance calculation formula ; , where m represents the total number of the restricted oxygen ventilation sequence or the normal oxygen ventilation sequence, represents the oxygen ventilation amount corresponding to the th monitoring time period in the restricted oxygen ventilation sequence, represents the mean dissolved oxygen content in the corresponding time period of the growth-restricted area, represents the oxygen ventilation amount corresponding to the th monitoring time period in the normal oxygen ventilation sequence, represents the mean dissolved oxygen content in the corresponding time period of the normal growth area; The standard deviation of the restricted oxygen supply sequence , the standard deviation of the normal oxygen supply sequence and the covariance are input into the Pearson distance model formula, and the regional deviation value is output, so as to judge whether the oxygen supply amount affects yeast growth; Oxygen supply limited adjustment module: If it affects yeast growth, construct a regional two-dimensional space model, and judge whether the growth-limited region can be adjusted centrally based on the central point coordinate positions of different growth-limited regions. If it can, obtain the centralized oxygen supply adjustment amount; if not, obtain the single oxygen supply adjustment amount.

Images