CN117431346A - Cordyceps fermentation temperature control method and system - Google Patents
Cordyceps fermentation temperature control method and system Download PDFInfo
- Publication number
- CN117431346A CN117431346A CN202311755119.5A CN202311755119A CN117431346A CN 117431346 A CN117431346 A CN 117431346A CN 202311755119 A CN202311755119 A CN 202311755119A CN 117431346 A CN117431346 A CN 117431346A
- Authority
- CN
- China
- Prior art keywords
- fermentation
- cordyceps sinensis
- fermentation time
- growth
- strain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000855 fermentation Methods 0.000 title claims abstract description 207
- 230000004151 fermentation Effects 0.000 title claims abstract description 207
- 238000000034 method Methods 0.000 title claims abstract description 58
- 241000190633 Cordyceps Species 0.000 title claims description 33
- 241001248610 Ophiocordyceps sinensis Species 0.000 claims abstract description 119
- 230000002159 abnormal effect Effects 0.000 claims abstract description 35
- 238000013531 bayesian neural network Methods 0.000 claims abstract description 13
- 238000007477 logistic regression Methods 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 19
- 238000013528 artificial neural network Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 230000002068 genetic effect Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000004519 manufacturing process Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000013139 quantization Methods 0.000 claims description 6
- 230000004083 survival effect Effects 0.000 claims description 5
- 238000013500 data storage Methods 0.000 claims description 4
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000001105 regulatory effect Effects 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 33
- 239000002609 medium Substances 0.000 description 14
- 210000002569 neuron Anatomy 0.000 description 9
- 238000003860 storage Methods 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 7
- 241000257303 Hymenoptera Species 0.000 description 5
- 238000010606 normalization Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000005284 excitation Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000002503 metabolic effect Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000004071 biological effect Effects 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 244000005700 microbiome Species 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013077 scoring method Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 239000004480 active ingredient Substances 0.000 description 1
- 230000000975 bioactive effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000012252 genetic analysis Methods 0.000 description 1
- 239000001963 growth medium Substances 0.000 description 1
- 239000005556 hormone Substances 0.000 description 1
- 229940088597 hormone Drugs 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000011005 laboratory method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 230000002906 microbiologic effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000007918 pathogenicity Effects 0.000 description 1
- 239000003016 pheromone Substances 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000001850 reproductive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001988 toxicity Effects 0.000 description 1
- 231100000419 toxicity Toxicity 0.000 description 1
- 231100000027 toxicology Toxicity 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention discloses a cordyceps sinensis fermentation temperature control method and system, in particular to the field of cordyceps sinensis fermentation, which are used for solving the problem that after cordyceps sinensis is fermented at an abnormal temperature for a period of time, even if the abnormal temperature is regulated, the quality of the cordyceps sinensis after the fermentation is finished is possibly reduced, and comprises the steps of determining whether the fermentation temperature is abnormal or not; if the fermentation temperature is abnormal, obtaining strain quality information of the cordyceps sinensis and growth condition information of the cordyceps sinensis, inputting the strain quality information and the growth condition information of the cordyceps sinensis into a fuzzy Bayesian neural network model, and determining a growth difference value after fermentation is completed; combining the growth difference value after the cordyceps sinensis is completely fermented and the fermentation time proportion information of the cordyceps sinensis, establishing a logistic regression model to determine whether the subsequent fermentation time needs to be adjusted; if the subsequent fermentation time is required to be adjusted, determining the adjustment quantity of the subsequent fermentation time through a fuzzy controller; thereby optimizing the growth environment and fermentation condition of the cordyceps sinensis, improving the fermentation efficiency and quality stability of the cordyceps sinensis, and ensuring the fermentation quality of the cordyceps sinensis.
Description
Technical Field
The invention relates to the technical field of cordyceps fermentation, in particular to a cordyceps fermentation temperature control method and system.
Background
In the process of cordyceps fermentation, if cordyceps is fermented for a period of time at abnormal temperature, the growth rate of cordyceps is affected, and the normal growth of hypha is blocked. In order to ensure the quality of the cordyceps sinensis after fermentation is completed, fermentation conditions and environmental parameters are required to be reasonably adjusted according to monitoring results to ensure the stable and optimized growth environment of the cordyceps sinensis, but the fermentation metabolic process of the cordyceps sinensis can be influenced due to the fact that the cordyceps sinensis is fermented at abnormal temperature for a period of time, the growth rate of the cordyceps sinensis and the accumulation rate of active ingredients are reduced, so that the quality of the cordyceps sinensis is likely to be reduced after the fermentation is completed even if the temperature of the cordyceps sinensis is adjusted, and therefore, when the cordyceps sinensis is fermented at the abnormal temperature for a period of time, the whole fermentation state of the cordyceps sinensis is required to be comprehensively judged, and further, the subsequent fermentation scheme is planned again.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide a method and a system for controlling fermentation temperature of Cordyceps, based on monitoring and analyzing growth parameters in a fermentation process of Cordyceps, evaluating and predicting differences between the growth parameters and standard parameters of Cordyceps by a machine learning algorithm to determine whether the Cordyceps needs to adjust the subsequent fermentation time, determining a specific value of the fermentation time to be adjusted by a fuzzy controller after determining the necessity of adjusting the subsequent fermentation time, and optimizing and determining control parameters of the fuzzy controller according to an ant algorithm, so as to solve the problems set forth in the background art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a Cordyceps fermentation temperature control method comprises the following steps:
step S1, measuring and recording temperature change in a fermentation bin, and determining whether fermentation temperature is abnormal or not;
s2, if the fermentation temperature is abnormal, obtaining strain quality information and growth condition information of the cordyceps sinensis, inputting fermentation time information of the cordyceps sinensis, growth condition information of the cordyceps sinensis and residual fermentation time of the cordyceps sinensis into a fuzzy Bayesian neural network model, and determining a growth difference value after complete fermentation;
step S3, establishing a logistic regression model to determine whether the subsequent fermentation time is required to be adjusted by combining the growth difference value after the cordyceps sinensis is completely fermented and the fermentation time proportion information of the cordyceps sinensis;
and S4, if the subsequent fermentation time is required to be adjusted, determining the adjustment quantity of the subsequent fermentation time through the fuzzy controller.
In a preferred embodiment, in step S2, the strain quality information of the Cordyceps is a strain quality value, which is obtained by comprehensively calculating the purity and purity of the strain, the activity and survival rate of the strain, the stability and genetic characteristics of the strain, and the safety and harmlessness of the strain, and the specific calculation expression is:；
wherein, C is a strain quality value, P, A, S, T respectively represents the evaluation values of purity, activity, stability and safety of the strain,、/>、/>、/>then the weight coefficient of the corresponding index;
the growth condition information of Cordyceps is growth condition value obtained by comparing the length of mycelium with normal length, comparing the mycelium density and mycelium diameter with corresponding normal value by the same method, and adding the ratio of the three to obtain the growth condition value.
In a preferred embodiment, in step S2, the method for constructing the fuzzy bayesian neural network model is as follows:
step S2.1, collecting sample data, including strain quality values, growth condition values and growth difference values after final fermentation in the Cordyceps sinensis history culture process;
s2.2, constructing a fuzzy Bayesian neural network model, and inputting sample data for training;
and S2.3, predicting by applying the trained model.
In a preferred embodiment, in step S2.3, the strain quality value and the growth condition value of the Cordyceps sinensis found at the abnormal fermentation temperature are input into the model, and calculated and output is performed through the forward neural network to obtain a predicted production difference value, when the production difference value is greater than or equal to 0, the subsequent fermentation time is not required to be adjusted, and when the growth difference value is less than 0, step S3 is performed.
In a preferred embodiment, in the step 3, the fermentation time proportion information of the cordyceps sinensis is a fermentation time ratio, and the ratio of the residual fermentation time of the cordyceps sinensis to the original complete time is obtained by calculating the residual fermentation time of the cordyceps sinensis compared with the complete fermentation time;
in step S3, the logistic regression model is:the method comprises the steps of carrying out a first treatment on the surface of the Wherein A is a linear combination of logistic regression, < >>For the intercept->And->Respectively the weight coefficient of the ratio of the growth difference value to the fermentation time;
the linear combination A is converted into adjustment probability through a sigmoid function and is used for judging whether the fermentation time needs to be adjusted or not.
In a preferred embodiment, in step S3, the adjustment probability is compared with a standard adjustment probability threshold, and if the adjustment probability is equal to or greater than the standard adjustment probability threshold, step S4 is performed to adjust the subsequent fermentation time of the Cordyceps, otherwise no adjustment is performed.
In a preferred embodiment, in step S4, the optimal fermentation time adjustment amount is determined by the two-dimensional fuzzy controller, and has two inputs and one output, wherein the input variables are an abnormal fermentation temperature deviation amount e and an abnormal fermentation temperature deviation change rate ec, the abnormal fermentation temperature deviation amount e is the difference between the abnormal fermentation temperature and the set temperature value, and the abnormal fermentation temperature deviation change rate is the temperature deviation change amount in unit time; the output variable U is the value of the fermentation time adjustment quantity.
In a preferred embodiment, in step S4, the selection of the control parameters of the fuzzy controller is determined by an ant algorithm, i.e. the quantization factor of the fuzzy controllerAnd->Scale factor->The optimal fermentation time adjustment amount is determined by the ant algorithm.
A Cordyceps fermentation temperature control system for the above Cordyceps fermentation temperature control method comprises:
the data acquisition module is used for acquiring strain quality information, growth condition information and fermentation time proportion information of the cordyceps sinensis in the cordyceps sinensis fermentation process;
the data analysis module is used for analyzing whether the cordyceps sinensis with abnormal fermentation temperature needs to be subjected to subsequent fermentation time adjustment or not, determining the adjustment quantity of the fermentation time through the fuzzy controller and optimizing the cordyceps sinensis fermentation process;
and the data storage module is used for storing data in the fermentation process of the historical cordyceps sinensis.
The invention has the technical effects and advantages that:
according to the invention, based on monitoring and analysis of growth parameters in the cordyceps fermentation process, the difference between the cordyceps growth parameters and standard parameters is evaluated and predicted through a machine learning algorithm to judge whether the cordyceps needs to be regulated for subsequent fermentation time, after the necessity of regulating the subsequent fermentation time is determined, a specific value of the fermentation time which needs to be regulated is further determined through a fuzzy controller, and the control parameters of the fuzzy controller are optimally determined according to an ant algorithm, so that the growth environment and fermentation conditions of the cordyceps are optimized, the fermentation efficiency and quality stability of the cordyceps are improved, and the quality of cordyceps fermentation is ensured.
Drawings
FIG. 1 is a flow chart of a cordyceps sinensis fermentation temperature control method.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention relates to a cordyceps sinensis fermentation temperature control method and system, which are based on monitoring and analyzing growth parameters in a cordyceps sinensis fermentation process, evaluate and predict differences between the growth parameters and standard parameters of cordyceps sinensis through a machine learning algorithm to judge whether the cordyceps sinensis needs to adjust subsequent fermentation time, further determine a specific value of the fermentation time which needs to be adjusted through a fuzzy controller after determining the necessity of adjusting the subsequent fermentation time, and optimally determine control parameters of the fuzzy controller according to an ant algorithm, thereby optimizing the growth environment and fermentation conditions of the cordyceps sinensis, improving the fermentation efficiency and quality stability of the cordyceps sinensis, and ensuring the quality of cordyceps sinensis fermentation.
Example 1
As shown in fig. 1, the cordyceps sinensis fermentation temperature control method comprises the following steps:
and S1, measuring and recording temperature change in the fermentation bin, and determining whether the fermentation temperature is abnormal.
Step S2, when the fermentation temperature is abnormal, strain quality information and growth condition information of the cordyceps sinensis are obtained, fermentation time information of the cordyceps sinensis, growth condition information of the cordyceps sinensis and residual fermentation time of the cordyceps sinensis are input into a fuzzy Bayesian neural network model, and a growth difference value after complete fermentation is determined.
And S3, establishing a logistic regression model to determine whether the subsequent fermentation time is required to be adjusted by combining the growth difference value after the cordyceps sinensis is completely fermented and the fermentation time proportion information of the cordyceps sinensis.
And S4, if the subsequent fermentation time is required to be adjusted, determining the adjusted fermentation time through the fuzzy controller.
Specifically, in step S1, the fermentation chamber is a closed or semi-closed space or facility for performing a cordyceps fermentation process, and is used for controlling and adjusting environmental factors such as temperature, humidity, ventilation and the like in the fermentation process, so as to promote growth, propagation and metabolic activity of microorganisms. In actual production, due to factors such as equipment cost, complexity of real-time monitoring and the like, the invention is established on the premise of regularly monitoring the fermentation temperature of cordyceps sinensis under the conditions of long-time fermentation process and relatively stable fermentation environment, so that the phenomenon of abnormal fermentation temperature at a certain monitoring moment can be caused.
In the step S2, the quality of the strain of the cordyceps directly determines the growth condition of hypha and the quality of metabolic products in the fermentation process of the cordyceps. The high-quality strain can effectively promote the growth and reproduction of cordyceps mycelia and improve the content of bioactive components of cordyceps. The strain quality information of the cordyceps sinensis is obtained by comprehensively calculating the purity and purity of the strain, the activity and survival rate of the strain, the stability and genetic characteristics of the strain and the safety and harmlessness of the strain. The calculation method adopts weighted average value calculation, and the specific calculation expression is:；
wherein C isStrain quality values, which are strain quality information, P, A, S, T respectively represent evaluation values of purity, activity, stability and safety of the strain,、/>、/>、/>the weight coefficient of the corresponding index represents the importance degree of each index in the comprehensive evaluation.
The purity and purity (P) of the strain are detected by microbiological experiments and culture media, whether other microorganisms pollute the strain is detected, and the purity and purity of the strain are evaluated;
the activity and survival rate (A) of the strain is the index of evaluating the biological activity and survival rate of the strain, including the growth rate, reproductive capacity, metabolic activity and the like of the strain through biological activity detection and culture experiments.
Stability and genetic characteristics (S) of the strain are obtained by evaluating genetic stability and genetic purity of the strain through genetic analysis and genetic characteristic detection, and the genetic stability and genetic characteristics of the strain in the fermentation process are known.
Safety and harmlessness (T) of the strain are assessed by toxicology assessment and safety detection, and the safety and harmlessness of the strain to the environment and human body are assessed, including the toxicity, pathogenicity, allergenicity and the like of the strain.
Acquisition of these indices typically requires specialized testing and assessment in a laboratory or production site in combination with related laboratory techniques and testing methods to obtain accurate data and assessment results. For each index acquisition, a corresponding detection method and standard operation procedure can be adopted to ensure the accuracy and reliability of the data. This is prior art and is not described in detail herein.
At the same time, the method comprises the steps of,、/>、/>、/>the method is set according to actual conditions, for example, an expert weighting method is adopted, that is, experts in related fields are invited to determine the weight of each index through professional opinion investigation and comprehensive evaluation, so that the weight coefficient can accurately reflect the importance of each index in strain quality evaluation. In addition, a plurality of methods such as an analytic hierarchy process, a fuzzy comprehensive evaluation method and the like can be considered to determine the weight coefficient so as to ensure the objectivity and scientificity of the weight coefficient. And will not be described in detail herein.
The growth condition of the cordyceps sinensis is characterized in that the growth condition of hypha in the fermentation process of the cordyceps sinensis is monitored, indexes such as the form, density and growth speed of the hypha are included, and the growth state and the fermentation effect of the cordyceps sinensis are primarily judged through observation and analysis of the growth condition of the hypha.
Specifically, the growth condition information of the cordyceps sinensis is quantitatively calculated through parameters such as the length, the density and the diameter of the mycelium, specifically, the length of the mycelium is compared with the normal length, the density of the mycelium and the diameter of the mycelium are compared with the corresponding normal values by the same method, and then the ratio of the mycelium to the corresponding normal values is added to obtain the growth condition value, namely the growth condition information, so that the growth condition of the cordyceps sinensis can be objectively reflected from multiple angles, and the fermentation effect and the quality condition of the cordyceps sinensis can be accurately evaluated.
It should be noted that the normal length, the normal density and the normal diameter are all normal values of the fermentation time period of the strain, and can be obtained by recording the average cordyceps length, density and diameter data of each fermentation time point in the history fermentation process.
In practical application, the method can set proper comparison standard and weight, and calculate and analyze the comparison value according to specific experimental data to evaluate the whole growth condition of the cordyceps sinensis. The determination of the specific weight coefficient is similar to the weight coefficient determination method in the calculation formula of the strain quality value, and is not described herein.
In step S2, the quality value, the growth condition value and the residual fermentation time of the strain are used as inputs of a fuzzy Bayesian neural network model, and the growth difference value of the cordyceps sinensis after the cordyceps sinensis is fermented is predicted.
The growth difference value is the difference value between the growth condition value after the cordyceps fermentation is completed and the growth condition value under the standard quality of cordyceps sinensis. It should be noted that, the growth difference value may be positive or negative, when the growth difference value is greater than or equal to 0, it indicates that the Cordyceps fermentation meets the fermentation standard, and the quality is qualified, at this time, no adjustment on the subsequent fermentation time is needed, and when the growth difference value is less than 0, it indicates that the fermentation is unqualified, and it is necessary to further determine whether to perform the subsequent fermentation time adjustment.
Further, the construction method of the fuzzy Bayesian neural network model comprises the following steps:
step S2.1, collecting sample data, including strain quality values, growth condition values and growth difference values after final fermentation in the Cordyceps sinensis history culture process;
for example, the collected sample data are data samples of one year, and are all historical data, namely 365 groups of data, but not limited to 365 groups of data;
the sample data is divided into training sample data and test sample data, and a sample pair consisting of a sample input and an expected output; in this example, 300 sets of data were used as training sample data, and 65 sets of data were used as test sample data.
In order to avoid overlarge neural network errors and prevent local neurons from reaching an oversaturated state, carrying out normalization processing on sample data so that the sample data are between 0 and 1, and obtaining an original output value by adopting inverse normalization processing on network output vectors; the normalization formula of the sample data is:the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For normalized sample data, ++>For the ratio of the original sample data to the minimum value of the original sample data, +.>The ratio of the maximum value of the original sample data to the minimum value of the original sample data is set;
obtaining a training sample set through normalization processing, namelyWhereinRepresenting the quality value of the strain,/->The value of the growth condition is indicated,indicating the remaining fermentation time,/->The growth difference value is represented, n is the data acquired for 365 sets, i.e., n=365.
S2.2, constructing a fuzzy Bayesian neural network model, and inputting sample data for training;
the neural network model for predicting the difference between the growth quality and the standard quality of the cordyceps sinensis after fermentation is completed consists of an input layer, an hidden layer and an output layer; the input layer consists of 15 neuron nodes in total and an implicit layer is determined by an empirical formula, wherein the 15 neuron nodes are composed of a strain quality value of the cordyceps sinensis and a growth condition value 1 group of the cordyceps sinensis; the output layer is the growth difference value of the cordyceps sinensis as a prediction result; the hidden layer is determined by an empirical formula:the method comprises the steps of carrying out a first treatment on the surface of the Wherein G is the number of neurons of an hidden layer, h is the number of neurons input by an input layer, m is the number of neurons output by an output layer, and a is a constant in the value range of 1-10;
establishing the fuzzy Bayesian neural network prediction model, and determining an excitation function, a training function, a learning function and a neural network performance index of the model; the excitation function selecting the sigmod function, i.eThe training function selects the traditional function, the learning function selects the Bayesian function, and the neural network performance index is:；
where n is the number of samples or 365 sets of sample data collected,representing the input vector +.>The weight component is represented by a number of components,the desired output target value, i= {1,2, 3..n }.
The pre-training process comprises the following steps: setting a training target and training step number through a pre-training function tranlm, training error precision, and selecting the optimal hidden layer neuron number according to the result;
creating a forward neural network:
net = newff (PR, [ S1, S2..sn 1], { TF1, tf2..tfn 1}, BTF, BLF, PF), wherein the vector elements range from 1 to N1; net is the creation of a new neural network; PR is a matrix formed by the maximum value and the minimum value of network input elements; [ S1, S2..SN 1] represents the number of neurons of the hidden layer and the output layer of the network; { TF1, tf2..tfn1 } represents the hidden layer and output layer excitation functions, as sigmod functions; the BTF is a training function of the network and is a tranlm function; BLF is a weight learning function of the network and is a Bayesian function; PF is a performance function, defaulting to a "mse" function;
creating a set of neural network weights: the set of weights affecting the computational accuracy and generalization ability of the neural network is represented by ω, i.e., ω= {，/>，/>，...，/>Wherein }>(i=1, 2, 3..n.) represents a weight component, n is the data of 365 sets collected, i.e., n=365;
creating a weight judgment set: and (3) fuzzifying the weight of the neural network by adopting an improved expert scoring method, scoring the neural network without communication by the expert, sorting the scoring results from large to small, negotiating by the expert from head to tail, scoring again, reordering, and the like until scoring is finished. The evaluation set is denoted by V, i.e., v= {,/>,/>,...,/>}, wherein->(i=1, 2, 3..n) represents the importance of the weight component, n being the number of 365 sets of data acquired, i.e., n=365;
expert scoring: blurring the weight of the neural network by adopting an expert scoring method;
defuzzification: and (3) performing deblurring by adopting a weighted average method to obtain the prior probability of the weight of the neural network, wherein the formula is as follows:
；
wherein P is%) A priori probability representing weights of the neural network, +.>Indicates the number of judges and the->Indicating that the judge makes possible judging results, n is acquired 365 groups of data, namely n=365;
determining a likelihood function: assuming a desired output target value,/>,/>,...,/>Is generated under Gaussian white noise, and likelihood functions are as follows: />；
Wherein,(gamma) is a normalization factor, and gamma is a super parameter; />(/>) Representing an error function;
the posterior probability of the weight is determined as (prior probability formula likelihood function)/sample distribution constant, and the specific expression is as follows:
；
where i= (1, 2,3,) n, j= (1, 2,3,) n,as an error function +.>P (D) represents a sample distribution constant;
randomly selecting a training sample set D to learn and train a fuzzy Bayesian neural network prediction model, determining each weight of an input layer, an implicit layer and an output layer by using fuzzy knowledge and Bayesian functions, and judging whether the actual output and the expected output of the output layer meet the performance index requirement of the neural network or not by using training sample data; if the requirement is not met, the number of neurons of the hidden layer is properly changed, the weights of the input layer, the hidden layer and the output layer are determined again by fuzzy knowledge and Bayesian functions, and whether the actual output and the expected output of the output layer meet the requirement of the neural network performance index is judged again through training sample data; if the requirement is met, finishing training, otherwise continuing training until the performance index requirement of the neural network is met; thus, a trained model is obtained.
S2.3, predicting by applying the trained model;
the strain quality value and the growth condition value of the cordyceps sinensis at the abnormal fermentation temperature are input into a model, and are calculated and output through a forward neural network to obtain a predicted production difference value, wherein the data at the moment is not historical data but data which is actually required to be predicted. When the production difference value is more than or equal to 0, the cordyceps sinensis fermentation meets the fermentation standard, the quality is qualified, the subsequent fermentation time is not required to be adjusted at the moment, and when the growth difference value is less than 0, the cordyceps sinensis fermentation is unqualified at the moment, and whether the subsequent fermentation time is adjusted is required to be further determined.
In the step 3, the fermentation time proportion information of the cordyceps sinensis is calculated by comparing the residual fermentation time of the cordyceps sinensis with the complete fermentation time, so that the proportion of the residual fermentation time of the cordyceps sinensis to the original complete time is obtained, and the smaller the proportion is, the later the fermentation stage of the cordyceps sinensis is, the later the fermentation time is adjusted, the influence of temperature abnormality can be corrected, but the influence of the adjustment of the fermentation time on the growth and the product quality of the cordyceps sinensis is relatively smaller because the cordyceps sinensis is already in the later stage of fermentation, namely the smaller the proportion of the residual fermentation time of the cordyceps sinensis to the original complete time is, the lower the necessity of adjusting the subsequent fermentation time is. The ratio of the residual fermentation time of the cordyceps sinensis to the original complete time is the fermentation time ratio, namely the fermentation time ratio information is the fermentation time ratio.
In step S3, the logistic regression model is:the method comprises the steps of carrying out a first treatment on the surface of the Wherein A is a linear combination of logistic regression, < >>For growth difference, F is fermentation time ratio, +.>For the intercept->And->Is a coefficient of an argument.
Then, a logic function (sigmoid function) is applied to convert the linear combination into a probability value, which is recorded as an adjustment probability, and is used for judging whether the fermentation time needs to be adjusted.
Comparing the adjustment probability with a standard adjustment probability threshold, if the adjustment probability is greater than or equal to the standard adjustment probability threshold, the method shows that the adjustment of the subsequent fermentation time has a larger influence on the growth quality of the cordyceps sinensis, and is favorable for obtaining the cordyceps sinensis with good quality, and at the moment, the subsequent fermentation time of the cordyceps sinensis is adjusted, or else, the cordyceps sinensis is not adjusted.
It should be noted that, the sigmoid function is given above, and the standard adjustment probability threshold is set according to the actual situation, for example, 0.6, etc., and the invention uses methods of maximum likelihood estimation or gradient descent, etc. to estimate the parameters of the model according to the collected actual dataThereby obtaining a specific model. Thereby judging the necessity of regulating the fermentation time of the cordyceps sinensis according to the logistic regression model.
In step S4, the present invention adopts a two-dimensional fuzzy controller having two inputs and one output, wherein the input variables are an abnormal fermentation temperature deviation e and an abnormal fermentation temperature deviation change rate ec, the abnormal fermentation temperature deviation e is the difference between the abnormal fermentation temperature and a set temperature value, and the abnormal fermentation temperature deviation change rate is the temperature deviation change amount (ec=de/dt) in unit time; the output variable U is the value of the fermentation time adjustment quantity.
It should be noted that, the fuzzy control rule of the present invention is summarized by expert knowledge and practical experience, and will not be described herein.
Further, the selection of the control parameters of the fuzzy controller is determined by ant algorithm, namely the quantization factor of the fuzzy controllerAnd->Scale factor->The optimal fermentation time adjustment amount is determined by the ant algorithm.
Specifically, the invention evaluates the indexWherein N is the number of sampling points.
The quantization factorAnd->Scale factor->As variables to be optimized are real numbers, the effective digits of the variables can be set according to the value ranges of the 3 parameters in the controller. For example, it may be assumed that they all have n significant digits. In order to facilitate the use of the ant colony algorithm, the values of the 3 parameters are determined to be expressed on the 0XY plane in an abstract mode, and the method comprises the following steps: 3n equally spaced, equally long line segments perpendicular to the X-axis are drawn, 1 to n line segments can be represented as quantization factors +.>N+1 to 2n line segments can be expressed as quantization factors +.>The 2n+1 to 3n line segments can be expressed as the scale factor +.>Is a digital value of (a). The positions of these line segments on the X-axis are denoted by the numerals 0 to 3n, respectively. Thus, each line segment has 3n nodes, and each node represents that the possible 3n+1 values of the digits represented by the line segment are 0 to 3n.
The ant algorithm comprises the following specific steps:
step S4.1, setting the number of ants to m, setting the time counter t=0, the number of loops nc=0, and setting the maximum number of loops NCmax and the concentration of the information hormone on each node at the initial timeWherein i=1 to 3n, j=0 to 3n, and +.>Setting all ants to a start point 0 and setting an array Routek (k= 1~m) of the storage paths for the ants;
step S4.2, where the set variable s=1, S represents the round;
step S4.3, by the formulaAnd calculating the probability of transferring the ants to each node on the line segment, selecting a node for each ant k (k= 1~m) on the line segment by adopting a roulette selection method according to the probability values, moving the ant k to the node, and storing the ordinate value of the node into an array.
Step S4.4, set s=s+1, go to step S4.3 if S <3n, otherwise go to step S4.5.
Step S4.5, according to the path of ant k (k=l-m), passing through the formulaCalculating the rule parameter of fuzzy controller corresponding to the path +.>And->And +.>Calculating an objective function Q corresponding to ant k, recording an optimal path in the cycle, and storing the corresponding fuzzy controller parameters>Is a kind of medium.
Step S4.6, order，/>The concentration of the pheromone substance on each node is updated and all elements in the array Routek (k= 1~m) of the ant path will be recorded clear 0.
Step S4.7 if NC<NCmax, and the entire ant colony system has not converged to the same path, then the method will againAll ants are placed at the starting point 0 and go to step S4.4 if NC<NCmax but the entire ant colony system has converged to take the same path, or NC=NCmax, then the loop ends to output the optimal path, and simultaneously calculates the corresponding optimal fuzzy controller parameters。
The fermentation time adjustment amount may be positive or negative.
The fuzzy controller with the optimized parameters by adopting the ant colony algorithm has the advantages of high response speed, no overshoot and short transition time, and is superior to a fuzzy controller with manually selected fuzzy factors.
Embodiment 2 is a system embodiment of embodiment 1, configured to implement a cordyceps sinensis fermentation temperature control method described in embodiment 1, and specifically includes:
the data acquisition module is used for acquiring strain quality information, growth condition information and fermentation time proportion information of the cordyceps sinensis in the cordyceps sinensis fermentation process.
The data analysis module is used for analyzing whether the cordyceps sinensis with abnormal fermentation temperature needs to be subjected to subsequent fermentation time adjustment or not, determining the adjustment quantity of the fermentation time through the fuzzy controller and optimizing the cordyceps sinensis fermentation process.
And the data storage module is used for storing data in the fermentation process of the historical cordyceps sinensis.
The above formulas are all formulas for removing dimensions and taking numerical calculation, and specific dimensions can be removed by adopting various means such as standardization, and the like, which are not described in detail herein, wherein the formulas are formulas for acquiring a large amount of data and performing software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present application are all or partially produced. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired or wireless means (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, ATA hard disk, tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state ATA hard disk.
It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the several embodiments provided in the present application, it should be understood that the disclosed systems, apparatuses, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a mobile ATA hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (9)
1. The cordyceps fermentation temperature control method is characterized by comprising the following steps of:
step S1, measuring and recording temperature change in a fermentation bin, and determining whether fermentation temperature is abnormal or not;
s2, if the fermentation temperature is abnormal, obtaining strain quality information and growth condition information of the cordyceps sinensis, inputting fermentation time information of the cordyceps sinensis, growth condition information of the cordyceps sinensis and residual fermentation time of the cordyceps sinensis into a fuzzy Bayesian neural network model, and determining a growth difference value after complete fermentation;
step S3, establishing a logistic regression model to determine whether the subsequent fermentation time is required to be adjusted by combining the growth difference value after the cordyceps sinensis is completely fermented and the fermentation time proportion information of the cordyceps sinensis;
and S4, if the subsequent fermentation time is required to be adjusted, determining the adjustment quantity of the subsequent fermentation time through the fuzzy controller.
2. The method for controlling the fermentation temperature of cordyceps sinensis according to claim 1, wherein the method comprises the following steps of:
in the step S2, strain quality information of the cordyceps sinensis is a strain quality value, which is obtained by comprehensively calculating the purity and purity of the strain, the activity and survival rate of the strain, the stability and genetic characteristics of the strain and the safety and harmlessness of the strain, wherein the specific calculation expression is as follows:；
wherein, C is a strain quality value, P, A, S, T respectively represents the evaluation values of purity, activity, stability and safety of the strain,、/>、/>、/>then the weight coefficient of the corresponding index;
the growth condition information of Cordyceps is growth condition value obtained by comparing the length of mycelium with normal length, comparing the mycelium density and mycelium diameter with corresponding normal value by the same method, and adding the ratio of the three to obtain the growth condition value.
3. The method for controlling the fermentation temperature of cordyceps sinensis according to claim 2, wherein the method comprises the following steps of:
in step S2, the construction method of the fuzzy bayesian neural network model is as follows:
step S2.1, collecting sample data, including strain quality values, growth condition values and growth difference values after final fermentation in the Cordyceps sinensis history culture process;
s2.2, constructing a fuzzy Bayesian neural network model, and inputting sample data for training;
and S2.3, predicting by applying the trained model.
4. A cordyceps sinensis fermentation temperature control method according to claim 3, wherein:
in step S2.3, the strain quality value and the growth condition value of the cordyceps sinensis at the abnormal fermentation temperature are input into a model, the predicted production difference value is obtained through calculation output of a forward neural network, when the production difference value is greater than or equal to 0, the subsequent fermentation time is not required to be adjusted, and when the growth difference value is less than 0, step S3 is performed.
5. The method for controlling the fermentation temperature of cordyceps sinensis according to claim 1, wherein the method comprises the following steps of:
in the step 3, the fermentation time proportion information of the cordyceps sinensis is a fermentation time proportion, and the proportion of the residual fermentation time of the cordyceps sinensis to the original complete time is obtained by comparing the residual fermentation time of the cordyceps sinensis with the complete fermentation time;
in step S3, the logistic regression model is:the method comprises the steps of carrying out a first treatment on the surface of the Wherein A is a linear combination of logistic regression,for the intercept->And->Respectively the weight coefficient of the ratio of the growth difference value to the fermentation time;
the linear combination A is converted into adjustment probability through a sigmoid function and is used for judging whether the fermentation time needs to be adjusted or not.
6. The method for controlling the fermentation temperature of cordyceps sinensis according to claim 5, wherein the method comprises the following steps of:
in step S3, the adjustment probability is compared with a standard adjustment probability threshold, if the adjustment probability is greater than or equal to the standard adjustment probability threshold, the subsequent fermentation time of the cordyceps sinensis is adjusted in step S4, otherwise, the cordyceps sinensis is not adjusted.
7. The method for controlling the fermentation temperature of cordyceps sinensis according to claim 1, wherein the method comprises the following steps of:
in step S4, determining an optimal fermentation time adjustment amount by a two-dimensional fuzzy controller, wherein the two-dimensional fuzzy controller has two inputs and one output, the input variables are an abnormal fermentation temperature deviation amount e and an abnormal fermentation temperature deviation change rate ec, the abnormal fermentation temperature deviation amount e is the difference between the abnormal fermentation temperature and a set temperature value, and the abnormal fermentation temperature deviation change rate is the temperature deviation change amount in unit time; the output variable U is the value of the fermentation time adjustment quantity.
8. The method for controlling the fermentation temperature of cordyceps sinensis according to claim 7, wherein the method comprises the following steps of: in step S4, the selection of the control parameters of the fuzzy controller is determined by ant algorithm, i.e. the quantization factor of the fuzzy controllerAnd->Scale factor->The optimal fermentation time adjustment amount is determined by the ant algorithm.
9. A cordyceps sinensis fermentation temperature control system for realizing a cordyceps sinensis fermentation temperature control method as claimed in any one of claims 1 to 8, comprising:
the data acquisition module is used for acquiring strain quality information, growth condition information and fermentation time proportion information of the cordyceps sinensis in the cordyceps sinensis fermentation process;
the data analysis module is used for analyzing whether the cordyceps sinensis with abnormal fermentation temperature needs to be subjected to subsequent fermentation time adjustment or not, determining the adjustment quantity of the fermentation time through the fuzzy controller and optimizing the cordyceps sinensis fermentation process;
and the data storage module is used for storing data in the fermentation process of the historical cordyceps sinensis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311755119.5A CN117431346A (en) | 2023-12-20 | 2023-12-20 | Cordyceps fermentation temperature control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311755119.5A CN117431346A (en) | 2023-12-20 | 2023-12-20 | Cordyceps fermentation temperature control method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117431346A true CN117431346A (en) | 2024-01-23 |
Family
ID=89555665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311755119.5A Pending CN117431346A (en) | 2023-12-20 | 2023-12-20 | Cordyceps fermentation temperature control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117431346A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201785393U (en) * | 2010-07-27 | 2011-04-06 | 河北农业大学 | Intelligent temperature control device for cordyceps militaris liquid fermentation process |
| CN103814752A (en) * | 2012-11-19 | 2014-05-28 | 无锡美新物联网科技有限公司 | Intelligent cultivation system of Cordyceps sinensis |
-
2023
- 2023-12-20 CN CN202311755119.5A patent/CN117431346A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201785393U (en) * | 2010-07-27 | 2011-04-06 | 河北农业大学 | Intelligent temperature control device for cordyceps militaris liquid fermentation process |
| CN103814752A (en) * | 2012-11-19 | 2014-05-28 | 无锡美新物联网科技有限公司 | Intelligent cultivation system of Cordyceps sinensis |
Non-Patent Citations (2)
| Title |
|---|
| 姜汶君等: "蚁 群 算 法 在 北 虫 草 发 酵 温 度 模 糊 控 制 系 统 中 的 应 用", 农机化研究, no. 7, 31 December 2012 (2012-12-31) * |
| 蔡梦萍等: "基于蚁群算法优化的黄酒发酵温度模糊控制系统", 食品与发酵工业, vol. 39, no. 10, 31 December 2013 (2013-12-31) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116757534B (en) | Intelligent refrigerator reliability analysis method based on neural training network | |
| CN113723007B (en) | Equipment residual life prediction method based on DRSN and sparrow search optimization | |
| CN108710974B (en) | Water ammonia nitrogen prediction method and device based on deep belief network | |
| CN110889085A (en) | Intelligent wastewater monitoring method and system based on complex network multiple online regression | |
| CN111144552B (en) | Multi-index grain quality prediction method and device | |
| Wang et al. | Diabetes Risk Analysis Based on Machine Learning LASSO Regression Model | |
| CN108399434B (en) | Analysis and prediction method of high-dimensional time series data based on feature extraction | |
| CN115510963A (en) | Incremental equipment fault diagnosis method | |
| CN112288137A (en) | LSTM short-term load prediction method and device considering electricity price and Attention mechanism | |
| CN114399032A (en) | Method and system for predicting metering error of electric energy meter | |
| CN111931983A (en) | Precipitation prediction method and system | |
| CN116029604B (en) | Cage-raised meat duck breeding environment regulation and control method based on health comfort level | |
| CN113393057A (en) | Wheat yield integrated prediction method based on deep fusion machine learning model | |
| CN114692507A (en) | Counting data soft measurement modeling method based on stacking Poisson self-encoder network | |
| CN107782857A (en) | Flexible measurement method based on the edible fungus fermented process key parameter for improving CS BPNN | |
| CN115841278B (en) | Method, system, equipment and medium for evaluating running error state of electric energy metering device | |
| CN113868957B (en) | Residual life prediction and uncertainty quantitative calibration method under Bayes deep learning | |
| CN117591843A (en) | Prediction method based on cross-validation machine learning | |
| CN116595465A (en) | High-dimensional sparse data outlier detection method and system based on self-encoder and data enhancement | |
| CN115060625A (en) | Method and device for obtaining slurry density, electronic equipment and storage medium | |
| CN112382382B (en) | Cost-sensitive integrated learning classification method and system | |
| CN114022678A (en) | Corn disease image identification method based on optimized convolutional neural network | |
| CN117233638A (en) | Lithium battery SOH detection method and device, electronic equipment and storage medium | |
| CN117431346A (en) | Cordyceps fermentation temperature control method and system | |
| CN116128361A (en) | Multistage efficiency calculation method, device, equipment and medium for intelligent reconnaissance system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20240123 |