CN117598391B - Determination method of fermentation process conditions and feed fermentation process monitoring method - Google Patents
Determination method of fermentation process conditions and feed fermentation process monitoring method Download PDFInfo
- Publication number
- CN117598391B CN117598391B CN202410074415.7A CN202410074415A CN117598391B CN 117598391 B CN117598391 B CN 117598391B CN 202410074415 A CN202410074415 A CN 202410074415A CN 117598391 B CN117598391 B CN 117598391B
- Authority
- CN
- China
- Prior art keywords
- fermentation
- matrix
- fermentation process
- steady
- submatrix
- 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.)
- Active
Links
- 238000000855 fermentation Methods 0.000 title claims abstract description 222
- 230000004151 fermentation Effects 0.000 title claims abstract description 222
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000012544 monitoring process Methods 0.000 title claims abstract description 23
- 239000011159 matrix material Substances 0.000 claims abstract description 109
- 239000002994 raw material Substances 0.000 claims abstract description 61
- 238000005070 sampling Methods 0.000 claims abstract description 48
- 230000007613 environmental effect Effects 0.000 claims abstract description 38
- 230000004044 response Effects 0.000 claims abstract description 5
- 238000006243 chemical reaction Methods 0.000 claims description 18
- 239000007788 liquid Substances 0.000 claims description 17
- 239000007787 solid Substances 0.000 claims description 8
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 6
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims description 6
- KDYFGRWQOYBRFD-UHFFFAOYSA-N succinic acid Chemical compound OC(=O)CCC(O)=O KDYFGRWQOYBRFD-UHFFFAOYSA-N 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 5
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 5
- 229910052760 oxygen Inorganic materials 0.000 claims description 5
- 239000001301 oxygen Substances 0.000 claims description 5
- 244000046052 Phaseolus vulgaris Species 0.000 claims description 4
- 235000010627 Phaseolus vulgaris Nutrition 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 102000004169 proteins and genes Human genes 0.000 claims description 4
- 108090000623 proteins and genes Proteins 0.000 claims description 4
- 244000063299 Bacillus subtilis Species 0.000 claims description 3
- 235000014469 Bacillus subtilis Nutrition 0.000 claims description 3
- 241000606125 Bacteroides Species 0.000 claims description 3
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 claims description 3
- 150000001413 amino acids Chemical class 0.000 claims description 3
- 239000001569 carbon dioxide Substances 0.000 claims description 3
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 239000008103 glucose Substances 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 238000002156 mixing Methods 0.000 claims description 3
- 239000010902 straw Substances 0.000 claims description 3
- 239000000758 substrate Substances 0.000 claims description 3
- 239000001384 succinic acid Substances 0.000 claims description 3
- 235000015099 wheat brans Nutrition 0.000 claims description 3
- GAMYVSCDDLXAQW-AOIWZFSPSA-N Thermopsosid Natural products O(C)c1c(O)ccc(C=2Oc3c(c(O)cc(O[C@H]4[C@H](O)[C@@H](O)[C@H](O)[C@H](CO)O4)c3)C(=O)C=2)c1 GAMYVSCDDLXAQW-AOIWZFSPSA-N 0.000 claims description 2
- 229930003944 flavone Natural products 0.000 claims description 2
- 150000002212 flavone derivatives Chemical class 0.000 claims description 2
- 235000011949 flavones Nutrition 0.000 claims description 2
- VHBFFQKBGNRLFZ-UHFFFAOYSA-N vitamin p Natural products O1C2=CC=CC=C2C(=O)C=C1C1=CC=CC=C1 VHBFFQKBGNRLFZ-UHFFFAOYSA-N 0.000 claims description 2
- 241000894006 Bacteria Species 0.000 claims 2
- JVTAAEKCZFNVCJ-UHFFFAOYSA-N lactic acid Chemical compound CC(O)C(O)=O JVTAAEKCZFNVCJ-UHFFFAOYSA-N 0.000 claims 2
- 239000004310 lactic acid Substances 0.000 claims 1
- 235000014655 lactic acid Nutrition 0.000 claims 1
- 238000007619 statistical method Methods 0.000 abstract description 8
- 238000001311 chemical methods and process Methods 0.000 abstract description 2
- 230000006698 induction Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 5
- 230000000813 microbial effect Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 238000013499 data model Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000010438 heat treatment Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 241000193830 Bacillus <bacterium> Species 0.000 description 2
- 239000002028 Biomass Substances 0.000 description 2
- 241000186660 Lactobacillus Species 0.000 description 2
- 235000019764 Soybean Meal Nutrition 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 210000001035 gastrointestinal tract Anatomy 0.000 description 2
- 229940039696 lactobacillus Drugs 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 210000004767 rumen Anatomy 0.000 description 2
- 239000002893 slag Substances 0.000 description 2
- 239000004455 soybean meal Substances 0.000 description 2
- 238000013179 statistical model Methods 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000001580 bacterial effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005842 biochemical reaction Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 239000011344 liquid material Substances 0.000 description 1
- 230000037353 metabolic pathway Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013386 optimize process Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000000376 reactant Substances 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
- C12Q3/00—Condition responsive control processes
-
- A—HUMAN NECESSITIES
- A23—FOODS OR FOODSTUFFS; TREATMENT THEREOF, NOT COVERED BY OTHER CLASSES
- A23K—FODDER
- A23K10/00—Animal feeding-stuffs
- A23K10/10—Animal feeding-stuffs obtained by microbiological or biochemical processes
- A23K10/12—Animal feeding-stuffs obtained by microbiological or biochemical processes by fermentation of natural products, e.g. of vegetable material, animal waste material or biomass
-
- A—HUMAN NECESSITIES
- A23—FOODS OR FOODSTUFFS; TREATMENT THEREOF, NOT COVERED BY OTHER CLASSES
- A23K—FODDER
- A23K10/00—Animal feeding-stuffs
- A23K10/30—Animal feeding-stuffs from material of plant origin, e.g. roots, seeds or hay; from material of fungal origin, e.g. mushrooms
- A23K10/37—Animal feeding-stuffs from material of plant origin, e.g. roots, seeds or hay; from material of fungal origin, e.g. mushrooms from waste material
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/10—Analysis or design of chemical reactions, syntheses or processes
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/80—Data visualisation
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biotechnology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Polymers & Plastics (AREA)
- Zoology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Organic Chemistry (AREA)
- Molecular Biology (AREA)
- Crystallography & Structural Chemistry (AREA)
- Theoretical Computer Science (AREA)
- Wood Science & Technology (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Microbiology (AREA)
- Biochemistry (AREA)
- Physiology (AREA)
- Animal Husbandry (AREA)
- Food Science & Technology (AREA)
- Immunology (AREA)
- Mycology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Biomedical Technology (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Sustainable Development (AREA)
- General Engineering & Computer Science (AREA)
- Botany (AREA)
- Analytical Chemistry (AREA)
- Genetics & Genomics (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Apparatus Associated With Microorganisms And Enzymes (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention discloses a method for determining fermentation process conditions and a method for monitoring a feed fermentation process, and belongs to the technical field of statistical analysis of chemical processes. The method for determining fermentation process conditions can record the environmental characteristic matrix, the raw material characteristic matrix and the product characteristic matrix of a plurality of fermentation batches, determine the steady-state intervals of the fermentation process according to the change of matrix data, and then generate the control intervals of the fermentation process conditions of each steady-state interval. The characteristic data of the same steady-state interval have strong relevance, and the invention can improve the monitoring precision of the fermentation process by distinguishing the fermentation process conditions of each steady-state interval, and can avoid the influence of the data fluctuation of the unsteady-state interval on the induction of statistics. And meanwhile, the time-lag period number is determined according to the sampling period and the response delay parameter, so that the loss of effective data in a steady-state interval due to the sampling delay of product characteristics is avoided.
Description
Technical Field
The invention relates to the technical field of statistical analysis of chemical processes, in particular to a method for determining fermentation process conditions and a method for monitoring a feed fermentation process.
Background
The fermentation process can be carried out in an advantageous direction by accurately monitoring and controlling each factor affecting the fermentation level according to the fermentation mechanism. The feed fermentation products have various types and complex mechanisms, and the difficulty of analyzing each fermentation reaction in detail is great. In the prior art, an appropriate mathematical model is constructed through each parameter characteristic of the fermentation process, model parameters are iterated continuously, and finally, the parameter conditions of the fermentation process are optimized. For example, chinese patent publication No. CN105629958B, CN103853152B discloses a process monitoring method for an intermittent fermentation process, first, normal operation data is used as a training sample driven by data, an optimal process offline model is obtained by a matrix analysis method, and then, whether the current production process is normal is determined by combining with a parameter matrix of the current process.
The above prior art documents fit the entire fermentation process to the same data model. Because the reactants in the fermentation process are numerous, the states in different time intervals are unstable, and the error of fitting the fermentation process into the same data model is large. Therefore, it is necessary to divide the fermentation process into a plurality of steady state intervals according to the parameters of the core and then determine the corresponding mathematical model. Such as quantitative relation model and application, fermentation monitoring method, device, system and equipment disclosed in Chinese patent application with the application number of CN 202310295650.2. Before determining quantitative relation, the method carries out linear fitting on all historical sample parameters under different fermentation working conditions, different production strains and different fermentation time states. However, this document does not give a method for determining the fermentation status. It is therefore desirable to provide a method for indirectly determining the fermentation stability interval, and thus the optimized process conditions, in combination with fermentation characteristics. Furthermore, the abnormal value of the current fermentation process is identified according to the historical fermentation process conditions, so that continuous monitoring of the fermentation process is realized.
Disclosure of Invention
Aiming at the problems, the invention provides a method for determining fermentation process conditions, which predicts the interval of fermentation steady state and establishes a condition model of the fermentation process. Furthermore, the invention also provides a method for monitoring the fermentation process of the feed, which monitors the fermentation process according to the relation between the current fermentation parameters and the fermentation process conditions.
The aim of the invention can be achieved by the following technical means:
a method for determining fermentation process conditions, comprising the steps of:
step 1: the anaerobic tank executes fermentation batch k, the circulating pipeline conveys liquid raw materials from the bottom of the anaerobic tank to the top of the anaerobic tank, and the rotary table extracts fermentation products from the anaerobic tank;
step 2: an environment collector is arranged in the anaerobic tank, a component collector is arranged in the circulating pipeline, an image collector is arranged in the rotary table, the environment collector periodically collects I environmental characteristics of the anaerobic tank, the component collector periodically collects J raw material characteristics of the anaerobic tank, and the image collector periodically collects H product characteristics of the anaerobic tank;
step 3: after the fermentation batch K is completed, determining a reaction delay parameter T according to the fermentation time length, and storing the environmental characteristics, the raw material characteristics and the product characteristics of each sampling time, if K is more than or equal to K, entering a step 4, otherwise, returning to the step 1, wherein K is the total batch of the fermentation samples, and if K is more than or equal to K;
step 4: constructing an environmental feature matrix X (KXNxI) based on the plurality of environmental features, constructing a raw material feature matrix Y (KXNxJ) based on the plurality of raw material features, and constructing a product feature matrix Z (KXNxH) based on the plurality of product features, N being the total number of sampling periods;
step 5: expanding a product characteristic matrix into Z (N multiplied by KH) along the sampling moment, setting a reaction boundary matrix B (M multiplied by H), wherein M is the number of reaction boundaries, and dividing the product characteristic matrix into M fermentation intervals according to the reaction boundary matrix;
step 6: expanding a raw material characteristic matrix into Y (N multiplied by KJ) along the sampling moment, dividing the raw material characteristic matrix into M first sub-matrixes according to a fermentation interval, and determining a steady-state interval of any first sub-matrix according to the parameter stability of the first sub-matrix;
step 7: expanding the environment characteristic matrix into X (NxKI) along the fermentation batch, and extracting a second submatrix from the environment characteristic matrix according to a steady-state interval, wherein the second submatrix corresponding to the steady-state interval m is X m (N m ×KI),N m For the number of samples in the steady-state interval M, m=1, 2,;
step 8: determining a time delay period number N' according to the sampling period T and the response delay parameter T to obtain an extended submatrix X m ( (N m +N'). Times. KI), developing an expanded submatrix to X along the fermentation batch m (K×(N m +N')I);
Step 9: generating a reference value of the fermentation process condition and a control interval of the fermentation process condition of the steady-state interval m according to the extended submatrix.
In the present invention, the environmental features include: one or more of pH value, temperature, carbon dioxide content and liquid level.
In the invention, the raw material characteristics comprise one or more of flavone concentration, ethanol concentration, glucose concentration and dissolved oxygen.
In the present invention, the product features include one or more of particle size features, color features, and pattern features.
In the present invention, in step 6, the matrix is selected from the first submatrix Y m (N m X KJ) to decompose the time-series submatrix Y of the raw material feature j mj (N m X K), J is less than or equal to J, and calculating a time sequence submatrix Y mj (N m X K) covariance sigma j Calculating the matrix of the first sub-matrix in the sampling period n and the sampling period n-1Similarity D n According to covariance sigma j Similarity to matrix D n And constructing a steady-state interval.
In the present invention, in step 9, the extended submatrix X for the steady-state interval m m ( (N m +N') x KI), calculating a principal component matrix of a steady-state interval m in each fermentation batch, predicting a reference value of a fermentation process condition according to the principal component matrix, and calculating a control interval of the fermentation process condition according to a preset confidence coefficient alpha.
A method of monitoring a fermentation process of a feed in accordance with a method of determining conditions of the fermentation process, comprising the steps of:
step 100: generating a control interval of fermentation process conditions of each steady-state interval according to the determination method of the fermentation process conditions;
step 200: mixing solid raw materials and fermentation strains, putting the mixed materials into an anaerobic tank, injecting liquid raw materials, conveying the liquid raw materials from the bottom of the anaerobic tank to the top of the anaerobic tank through a circulating pipeline, and extracting fermentation products from the anaerobic tank through a rotary table;
step 300: periodically collecting environmental characteristics of the anaerobic tank and raw material characteristics of the circulating pipeline, and constructing an environmental characteristic matrix and a raw material characteristic matrix;
step 400: dividing a raw material characteristic matrix into a first submatrix according to a fermentation interval, determining a steady-state interval according to the first submatrix, and extracting a second submatrix from an environmental characteristic matrix according to the steady-state interval;
step 500: if the second submatrix is within the control interval of the fermentation process condition, returning to the step 200, otherwise, generating a fermentation abnormality notification.
In the invention, the solid raw material is one or more of straw, bran, bean pulp and wheat bran.
In the invention, the fermentation strain is one or more of succinic acid-producing filamentous bacillus, rumen bacteroides, bacillus natto and lactobacillus.
In the invention, the fermentation product is one or more of protein, amino acid and fermentation matrix.
The method for determining the fermentation process conditions and the method for monitoring the fermentation process of the feed have the beneficial effects that: and determining a steady-state interval according to the consumption conditions of the product and the liquid raw material in the normal fermentation state, and generating a control interval of fermentation process conditions in each steady-state interval, so that the induction of data fluctuation influence statistics in the unsteady-state interval can be avoided. The characteristic data of the same steady-state interval have strong correlation, and the monitoring precision of the fermentation process can be improved by distinguishing the fermentation process conditions of the steady-state intervals. Further, the invention determines the time-lag period number according to the sampling period and the response delay parameter, and the second submatrix is expanded through the time-lag period to avoid losing effective data in a steady-state interval due to sampling delay of product characteristics.
Drawings
FIG. 1 is a schematic diagram of the equipment composition of an anaerobic fermentation process;
FIG. 2 is a flow chart of a method of determining fermentation process conditions of the present invention;
FIG. 3 is a block diagram of an anaerobic tank of the present invention;
FIG. 4 is a line graph of a preferred temperature profile of the present invention;
FIG. 5 is a line graph of preferred colony features of the present invention;
FIG. 6 is a schematic diagram of a product feature matrix development process according to the present invention;
FIG. 7 is a schematic diagram of a fermentation zone of the present invention;
FIG. 8 is a schematic diagram of the determination of steady-state intervals from fermentation intervals according to the present invention;
FIG. 9 is a schematic diagram of a process for extracting a second submatrix according to the present invention;
FIG. 10 is a schematic diagram of an expansion sub-matrix expansion process according to the present invention;
FIG. 11 is a flow chart of a preferred extraction product feature of the present invention;
FIG. 12 is a flow chart of a method of monitoring a feed fermentation process of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
Microbial fermentation can convert biomass feedstock into a desired fermented feed via a specific metabolic pathway. The microbial fermentation is applied to feed production, and the main biomass solid raw materials are bran coat, bean pulp and the like, so that the absorption of effective components such as proteins by animal intestinal tracts can be promoted, and intestinal tract diseases in the feeding process can be reduced. Microbial fermentation is typically performed in an anaerobic environment. As shown in fig. 1, the common anaerobic fermentation process equipment comprises an anaerobic tank 11, a storage bin 12, a substrate tank 13, a liquid storage tank 14, a circulating tank 15, a slag discharge tank 16 and a heating tank 17. The solid raw materials are stored in a stock bin 12, the fermentation broth is stored in a substrate tank 13, and the liquid raw materials are stored in a liquid storage tank 14. During fermentation, the recycle tank 15 draws liquid feed from the bottom of the anaerobic tank and returns to the top of the anaerobic tank. The slag discharging bath 16 is used for discharging the fermentation residues. The heating tank 17 supplies a heating medium to the interlayer of the anaerobic tank for adjusting the fermentation temperature of the anaerobic tank. The anaerobic tank 11 has strict limitation on oxygen content and bacterial environment, and the outside cannot pay attention to the fermentation state all the time, so that only a fermentation product sample can be periodically extracted. In general, microbial fermentation is regarded as an intermittent process consisting of a plurality of fermentation states, the fermentation states are counted according to the environmental characteristics of the anaerobic tank 11 in the prior art, the optimal interval of the fermentation states is determined, and finally whether the current environmental characteristics are abnormal or not is judged. The biochemical reaction mechanism in the fermentation process is complex, and the statistical analysis method has lower accuracy.
Example 1
According to the invention, different fermentation intervals and steady-state intervals in the fermentation intervals are distinguished according to the product condition and the raw material consumption condition of the normal fermentation process, and the control interval of the fermentation process condition of each steady-state interval is generated, so that the influence of parameter fluctuation of the unsteady-state interval on the monitoring result can be reduced. As shown in fig. 2 to 10, the method for determining the fermentation process conditions according to the present invention comprises the following steps.
Step 1: the anaerobic tank performs a fermentation batch k, and the circulation pipe 21 conveys the liquid raw material from the bottom of the anaerobic tank to the top of the anaerobic tank, from which the rotary table 22 extracts the fermentation product. k is initialized to 1. The circulation pipes 21 are connected to both sides of the circulation tank. The anaerobic tank is provided with a lifting machine 23, the lifting machine 23 collects fermentation products from the tank bottom and then places the fermentation products into a rotary table 22, and a partition plate 24 of the rotary table 22 spreads the fermentation products. After the sampling is completed, the rotary table 22 is reversely rotated again, and the fermentation product is returned to the anaerobic tank. At the sampling moment, the contact surface between the anaerobic tank and the air is small, and at the non-sampling moment, the anaerobic tank is in a closed state.
Step 2: an environmental collector is arranged in the anaerobic tank, a component collector is arranged in the circulating pipeline, an image collector is arranged in the rotary table, the environmental collector 25 periodically collects I environmental characteristics of the anaerobic tank, the component collector 26 periodically collects J raw material characteristics of the anaerobic tank, and the image collector 27 periodically collects H product characteristics of the anaerobic tank. The environmental collectors 25 are disposed at various locations of the anaerobic tank, such as pH detectors, temperature sensors, level detectors, and the like. The corresponding environmental features include: one or more of pH value, temperature, carbon dioxide content and liquid level. The temperature profile of a particular fermentation process is shown in FIG. 4, which shows a steady state over a period of time. The ingredient collector 26 is, for example, a spectrometer, and near infrared light is irradiated on a transparent region of the circulation pipe, and the characteristics of the material are determined from the spectral interval of the liquid material. The raw material characteristics comprise one or more of colony count, ethanol concentration, glucose concentration and dissolved oxygen. The shannon index of colony count for a particular fermentation process is shown in figure 5. The colony count showed a stable state for a certain period of time. The image collector 27 obtains the product features including one or more of granularity, color, and pattern features. And (5) extracting the characteristics of the fermented product from the image of the fermented product of the rotary table. For example, in the fermentation process of soybean meal, the product turns from yellow to black, gradually thickens from dry granular and gradually dries. The fermentation status can be predicted from the image of the fermentation product. The detailed product feature extraction method is described with reference to example two.
Step 3: after the fermentation batch K is completed, determining a reaction delay parameter T according to the fermentation time, storing the environmental characteristics, the raw material characteristics and the product characteristics at each sampling time, if K is more than or equal to K, entering a step 4, otherwise, returning to the step 1, wherein K is the total batch of the fermentation samples, and if K is more than or equal to K+1. Because the fermentation state and the sampling result have delay, the reaction delay parameter T is set to expand the length of the statistical steady-state interval, so that the key characteristic parameters are prevented from being lost due to the sampling delay. In this example, the reaction delay parameter t=fermentation duration×0.01. For microbial fermentation processes of feeds such as soybean meal, the fermentation time period is usually 44 to 52 hours, and K can be preset to 100.
Step 4: an environmental feature matrix X (KXNxI) is constructed based on the plurality of environmental features, a raw material feature matrix Y (KXNxJ) is constructed based on the plurality of raw material features, and a product feature matrix Z (KXNxH) is constructed based on the plurality of product features, N being the total number of sampling periods. The environment feature matrix, the raw material feature matrix and the product feature matrix are all three-dimensional matrices. For example, the environment feature matrix is formed by 10 groups of sampling points of 100 batches of 300 sampling moments, and the corresponding data volume is 100×300×10=3×10 5 . The invention confirms the parameter range of the normal steady-state interval through the statistical analysis of the data, and provides guidance for the follow-up industrial production monitoring. N is the total sampling frequency of one fermentation process, the sampling interval is 10min, and N can be preset to 300.
Step 5: and expanding the product characteristic matrix into Z (N multiplied by KH) along the sampling moment, setting a reaction boundary matrix B (M multiplied by H), wherein M is the number of reaction boundaries, and dividing the product characteristic matrix into M fermentation intervals according to the reaction boundary matrix. As shown in fig. 6, the product feature matrix Z (kxnxh) is expanded into Z (nxkh), and the expanded product feature matrix is a two-dimensional matrix. The number of rows of the matrix is N, and the number of columns is KH. This way of unfolding facilitates comparison with the reaction boundary matrix. After expansion, Z (NXKH) is defined by K Z k (n×h) composition, k=1, 2,..k.
The reaction boundary matrix refers to the critical point of the fermentation product for distinguishing between different fermentation states. The mth data group of B (MXH) is [ B ] m1 , b m2 ,..., b mh ,...,b mH H is less than or equal to H. Traversing Z along sampling time n k (N H) each column, find out that [ b ] is satisfied m1 , b m2 ,..., b mh ,...,b mH The starting column of ], divide Z by this column k (N×H), thereby completing the division of Z (N×KH). As in fig. 7, fermentation intervals M, m=1, 2, & M were obtained after segmentation. Specifically, if all product features of any data set areIf the data set is greater than or equal to the lower limit of the mth fermentation state and less than the lower limit of the (if any) m+1th fermentation state, the data set is considered to belong to the fermentation zone m. For M data sets of B (MxH), the product feature matrix is divided into M parts along the corresponding segments, corresponding to M fermentation intervals.
Step 6: and expanding the raw material characteristic matrix into Y (N multiplied by KJ) along the sampling time, dividing the raw material characteristic matrix into M first sub-matrixes according to the fermentation interval, and determining the steady-state interval of any first sub-matrix according to the parameter stability of the first sub-matrix. And (3) expanding the raw material feature matrix by adopting the method in the step (5), wherein the expanded raw material feature matrix corresponds to the sampling time of the product feature matrix one by one. For any one fermentation batch, the raw material characteristic matrix can be divided into M parts according to the position of the fermentation zone.
At the edge time of the fermentation interval, the fermentation process is in an unstable state, and in order to improve the accuracy of statistical analysis, unstable data on both sides needs to be eliminated. The present embodiment employs covariance to estimate parameter stability. From the first submatrix Y m (N m X KJ) to decompose the time-sequential submatrix Y of sampling points j mj (N m X K), J is less than or equal to J. Calculating a time sequence submatrix Y mj (N m X K) covariance σ of component characteristics at each sampling time j . The sum of covariance of J sample points is. The larger the covariance, the worse the data stability. The larger the matrix similarity, the better the data stability. Calculating matrix similarity D of sampling time n and sampling period n-1 n Matrix similarity D n =diss(Y n-1 , Y n )。Y n-1 And Y is equal to n The time sequence submatrices are respectively composed of raw material characteristics of sampling time n and sampling period n-1. Matrix similarity D n Preferably 0.5, 0.4 < D n Preferably less than 0.6. According to covariance sigma j Similarity to matrix D n And constructing a steady-state interval. Stability lower limit value for sampling instant n>. The first sub-matrix is traversed, starting with n=1, up to CSI n And not less than the stability lower limit. And then decrementing from n=n, traversing each row of the first sub-matrix. If CSI of the nth row n Deleting the data group of the nth row until the CSI is smaller than the stability lower limit value n The stability lower limit is, for example, 3 to 7. And searching M steady-state intervals corresponding to the M fermentation intervals.
Step 7: expanding the environment characteristic matrix into X (NxKI) along the fermentation batch, and extracting a second submatrix from the environment characteristic matrix according to a steady-state interval, wherein the second submatrix corresponding to the steady-state interval m is X m (N m ×KI),N m For the number of samples of steady-state interval M, m=1, 2. Referring to fig. 8, the steady-state interval m corresponds to the fermentation interval m one by one. For any fermentation batch, after the first submatrix of the raw material characteristic matrix is extracted, a steady-state interval of the first submatrix is identified, then according to the initial sampling time and the ending sampling time of the steady-state interval, the corresponding sampling time of the environmental characteristic matrix is searched, and finally the second submatrix of each fermentation batch is extracted. As shown in FIG. 9, a second submatrix X can be obtained by summing up K fermentation batches m (N m ×KI)。
Step 8: determining a time delay period number N' according to the sampling period T and the response delay parameter T to obtain an extended submatrix X m ((N m +N'). Times. KI), developing an expanded submatrix to X along the fermentation batch m (K×(N m +n') I). N' =mod (T/T), mod () is a modulo operation. The number of rows of the second sub-matrix is extended forward according to the hysteresis number N'. Expanded submatrix X m ( (N m +N') ×KI) is N m +N'. The product features inevitably exhibit sampling delays relative to the constituent features and environmental features. Therefore, the second sub-matrix is expanded forward, that is, the environmental features of N' sampling periods in front of the second sub-matrix are integrated into the second sub-matrix, and an expanded sub-matrix is generated, so that the loss of effective data of the environmental features is avoided.
Step 9: generating a reference value of the fermentation process condition and a control interval of the fermentation process condition of the steady-state interval m according to the extended submatrix. Extended submatrix X m ( (N m +N'). Times. KI) is a sample of the final statistical analysis of the steady state interval m. As shown in FIG. 10, to facilitate statistical analysis, a submatrix X is extended m ( (N m +N'). Times. KI) are developed in the direction of the fermentation batch. The present invention preferably uses entropy analysis (KECA) to analyze the feature data of the extended submatrices. Firstly, carrying out standardization processing on an expansion submatrix of a steady-state interval m, then calculating a main component matrix of the steady-state interval m in each fermentation batch, predicting a reference value of a fermentation process condition according to the main component matrix, and calculating a control interval of the fermentation process condition according to a preset confidence coefficient alpha. And if the monitored environment characteristic matrix does not meet the control interval of the fermentation process condition of the steady-state interval m, the fermentation is considered to be abnormal.
After the steady-state intervals are determined, the invention can predict the reference value of the fermentation process condition of each steady-state interval through statistical analysis, and determine the allowable control interval of the normal fermentation process on the basis of the reference value. The present invention is not limited to a model for calculating a reference value and a control interval of fermentation process conditions from historical data of the fermentation process conditions. In one particular embodiment, a KECA-based statistical model may be employed. The reference value of the fermentation process condition is the model statistic, and the control interval of the fermentation process condition is the control limit of the model statistic. In another embodiment, a statistical model based on angular structure (Cauchy-Schwarz) may be used to calculate the statistic C of the sampling instant n in the steady-state interval m n . Regeneration of reference values for fermentation process conditions in steady-state interval m。N m +N' is the number of sampling instants of the steady-state interval m. The confidence coefficient alpha is preset to be 5 percent, and the control interval of the fermentation process condition is [ C-alpha C, C+alpha C ]. I.e. if the statistics calculated by the current environmental matrix are greater than 105% c or less than 95% c, the fermentation is considered abnormal. Besides the analysis method disclosed by the embodiment, the invention can also adopt methods such as a support vector data model (SVDD), a Long-short-time memory model (Long-Short Term Memory) and the like to solve the data analysis of a steady-state interval.
Example two
The present example further discloses the method of extracting product features in step 7. The product features of this embodiment include pattern features, texture features, and gray scale features, i.e., h=3. And presetting lower limit values of pattern features, texture features and gray scale features of M fermentation states according to the product patterns, textures and gray scales of the fermentation states to obtain a reaction boundary matrix B (MxH). The data set of the m-th row is [ b ] m1 ,b m2 ,b m3 ,]. And if the data set of the current product characteristic matrix simultaneously meets the lower limit values of the pattern characteristic, the texture characteristic and the gray level characteristic of the mth data set, the fermentation is considered to enter a fermentation state m. As shown in fig. 11, the method of extracting the product features includes the following steps.
And (5) preprocessing an image. The image collector shoots an original image of the fermentation product. The original image is an RGB space image, which is converted into HNV space for better extraction of image data features. And screening abnormal pixel points of the original image based on the 3 sigma principle, and carrying out normalization processing on the brightness of each pixel point to obtain a corrected image.
Extracting pattern features. The self-adaptive threshold value segments the corrected image, and fills in the holes of the segmented image to generate a connected region. Extracting boundary pixels of a connected region of the corrected image to obtain pattern features b m1 。
And extracting texture features. Performing two-dimensional discrete wavelet decomposition on the corrected image based on haar wavelet, and decomposing the corrected image into sub-bands GD in the horizontal direction and the vertical direction 2 And DG 2 The pixel mean value for each subband is calculated. The pixel mean value is texture feature b m2 。
And extracting gray scale characteristics. And generating a gray level histogram of the corrected image, obtaining a statistical moment of the gray level histogram, and calculating an average gray level value and a gray level standard deviation. The gray standard deviation is the gray characteristic b m3 。
Furthermore, the invention can also divide pattern features into a plurality of types according to the forms of different fermentation products. The color sensitive card can also be arranged according to the volatile gas in the fermentation product. Determining the product characteristics of the fermentation product according to the color development reflection of different color sensitive cards.
Example III
According to the feed fermentation process monitoring method based on the fermentation process condition determining method, the abnormality of the current fermentation process is predicted according to the fermentation process condition generated by statistics. According to different fermentation processes, the invention can generate different fermentation process conditions. Firstly, a standardized fermentation operation flow is determined according to fermentation process requirements, and related process can refer to technical literature of related fermented feeds, which is not described herein. As shown in fig. 12, the feed fermentation process monitoring method of the present embodiment includes the following steps.
Step 100: and generating a control interval of the fermentation process conditions of each steady-state interval according to the determination method of the fermentation process conditions. In the sample collection process, relevant data without abnormality in the fermentation process is reserved, a corresponding steady-state interval is determined according to the method of the first embodiment, statistics of environmental characteristics in the steady-state interval are regenerated, and finally a control interval of fermentation process conditions is generated.
Step 200: mixing solid raw materials and fermentation strains, putting into an anaerobic tank, injecting liquid raw materials, conveying the liquid raw materials from the bottom of the anaerobic tank to the top of the anaerobic tank through a circulating pipeline, and extracting fermentation products from the anaerobic tank through a rotary table. The solid raw materials in the fermentation process are one or more of straw, bran, bean pulp and wheat bran. The fermentation strain is one or more of succinic acid-producing filamentous bacillus, rumen bacteroides, bacillus natto and lactobacillus. The fermentation product is one or more of protein, amino acid and fermentation matrix.
Step 300: and periodically collecting the environmental characteristics of the anaerobic tank and the raw material characteristics of the circulating pipeline, and constructing an environmental characteristic matrix and a raw material characteristic matrix. The method for collecting the environmental characteristics and the raw material characteristics can be described in the first embodiment. To avoid oxygen from entering the anaerobic tank, this embodiment omits the use of a lift to collect the fermentation product.
Step 400: dividing the raw material characteristic matrix into a first submatrix according to the fermentation interval, determining a steady-state interval according to the first submatrix, and extracting a second submatrix from the environmental characteristic matrix according to the steady-state interval. The present embodiment partitions the first submatrix directly using the average fermentation interval determined from the sample data in step 100. In another embodiment, to obtain more accurate statistics, the product characteristics of the fermentation product may be re-collected, predicting the fermentation interval.
Step 500: if the second submatrix is within the control interval of the fermentation process condition, returning to the step 200, otherwise, generating a fermentation abnormality notification. I.e. the statistic calculated from the environmental characteristics of the second submatrix is within the fermentation process condition control interval, the fermentation is considered normal and the monitoring is continued by returning to step 200. Otherwise, entering an error-reporting early-warning program and sending error-reporting notices to the on-site alarm, the control center and the mobile terminal.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (5)
1. A method for determining fermentation process conditions, comprising the steps of:
step 1: the anaerobic tank executes fermentation batch k, the circulating pipeline conveys liquid raw materials from the bottom of the anaerobic tank to the top of the anaerobic tank, and the rotary table extracts fermentation products from the anaerobic tank;
step 2: setting up the environment collector in the anaerobic tank, setting up the composition collector at circulation pipeline, setting up the image collector at the revolving platform, the I environmental characteristic in anaerobic tank is periodically gathered to the environment collector, and J raw materials characteristic in anaerobic tank is periodically gathered to the composition collector, and the H product characteristic in anaerobic tank is periodically gathered to the image collector, wherein, the environmental characteristic includes: one or more of pH value, temperature, carbon dioxide content and liquid level height, wherein the raw material characteristics comprise one or more of flavone concentration, ethanol concentration, glucose concentration and dissolved oxygen amount, and the product characteristics comprise one or more of granularity characteristics, color characteristics and pattern characteristics;
step 3: after the fermentation batch K is completed, determining a reaction delay parameter T according to the fermentation time length, and storing the environmental characteristics, the raw material characteristics and the product characteristics of each sampling time, if K is more than or equal to K, entering a step 4, otherwise, returning to the step 1, wherein K is the total batch of the fermentation samples, and if K is more than or equal to K;
step 4: constructing an environmental feature matrix X (KXNxI) based on the plurality of environmental features, constructing a raw material feature matrix Y (KXNxJ) based on the plurality of raw material features, and constructing a product feature matrix Z (KXNxH) based on the plurality of product features, N being the total number of sampling periods;
step 5: expanding a product characteristic matrix into Z (N multiplied by KH) along the sampling moment, setting a reaction boundary matrix B (M multiplied by H), wherein M is the number of reaction boundaries, and dividing the product characteristic matrix into M fermentation intervals according to the reaction boundary matrix;
step 6: expanding a raw material characteristic matrix into Y (N multiplied by KJ) along a sampling time, dividing the raw material characteristic matrix into M first sub-matrices according to a fermentation interval, and determining a steady-state interval of any first sub-matrix according to parameter stability of the first sub-matrix, wherein the raw material characteristic matrix is selected from the first sub-matrix Y m (N m X KJ) to decompose the time-series submatrix Y of the raw material feature j mj (N m X K), J is less than or equal to J, and calculating a time sequence submatrix Y mj (N m X K) covariance sigma j Calculating the matrix similarity D of the first sub-matrix in the sampling period n and the sampling period n-1 n According to covariance sigma j Similarity to matrix D n Constructing a steady-state interval;
step 7: expanding the environment characteristic matrix into X (NxKI) along the fermentation batch, and extracting a second submatrix from the environment characteristic matrix according to a steady-state interval, wherein the second submatrix corresponding to the steady-state interval m is X m (N m ×KI),N m For the number of samples in the steady-state interval M, m=1, 2,;
step 8: determining a time delay period number N' according to the sampling period T and the response delay parameter T to obtain an extended submatrix X m ( (N m +N'). Times. KI), developing an expanded submatrix to X along the fermentation batch m (K×(N m +N')I);
Step 9: generating a reference value of fermentation process conditions and a control interval of fermentation process conditions of the steady-state interval m according to the expansion submatrix, wherein the steady-state intervalM-between extended submatrix X m ( (N m +N') x KI), calculating a principal component matrix of a steady-state interval m in each fermentation batch, predicting a reference value of a fermentation process condition according to the principal component matrix, and calculating a control interval of the fermentation process condition according to a preset confidence coefficient alpha.
2. A method for monitoring the fermentation process of a feed according to the method for determining the conditions of a fermentation process of claim 1, comprising the steps of:
step 100: generating a control interval of fermentation process conditions of each steady-state interval according to the determination method of the fermentation process conditions;
step 200: mixing solid raw materials and fermentation strains, putting the mixed materials into an anaerobic tank, injecting liquid raw materials, conveying the liquid raw materials from the bottom of the anaerobic tank to the top of the anaerobic tank through a circulating pipeline, and extracting fermentation products from the anaerobic tank through a rotary table;
step 300: periodically collecting environmental characteristics of the anaerobic tank and raw material characteristics of the circulating pipeline, and constructing an environmental characteristic matrix and a raw material characteristic matrix;
step 400: dividing a raw material characteristic matrix into a first submatrix according to a fermentation interval, determining a steady-state interval according to the first submatrix, and extracting a second submatrix from an environmental characteristic matrix according to the steady-state interval;
step 500: if the second submatrix is within the control interval of the fermentation process condition, returning to the step 200, otherwise, generating a fermentation abnormality notification.
3. The method for monitoring the fermentation process of feed according to claim 2, wherein the solid raw material is one or more of straw, bran, bean pulp and wheat bran.
4. The method for monitoring the fermentation process of the feed according to claim 2, wherein the fermentation strain is one or more of filamentous succinic acid-producing bacteria, bacteroides ruminalis, bacillus natto and lactic acid bacteria.
5. The method for monitoring the fermentation process of feed according to claim 2, wherein the fermentation product is one or more of protein, amino acid and fermentation substrate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410074415.7A CN117598391B (en) | 2024-01-18 | 2024-01-18 | Determination method of fermentation process conditions and feed fermentation process monitoring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410074415.7A CN117598391B (en) | 2024-01-18 | 2024-01-18 | Determination method of fermentation process conditions and feed fermentation process monitoring method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117598391A CN117598391A (en) | 2024-02-27 |
CN117598391B true CN117598391B (en) | 2024-04-16 |
Family
ID=89944621
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410074415.7A Active CN117598391B (en) | 2024-01-18 | 2024-01-18 | Determination method of fermentation process conditions and feed fermentation process monitoring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117598391B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118151620A (en) * | 2024-05-10 | 2024-06-07 | 合肥师范学院 | Intelligent storage monitoring system suitable for fruit wine fermentation production |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103970092A (en) * | 2014-04-13 | 2014-08-06 | 北京工业大学 | Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm |
CN107895224A (en) * | 2017-10-30 | 2018-04-10 | 北京工业大学 | A kind of MKECA fermentation process fault monitoring methods based on extension nuclear entropy load matrix |
WO2019089730A1 (en) * | 2017-10-31 | 2019-05-09 | Locus Ip Company, Llc | Matrix fermentation systems and methods for producing microbe-based products |
WO2019129891A1 (en) * | 2017-12-29 | 2019-07-04 | F. Hoffmann-La Roche Ag | Predicting the metabolic condition of a cell culture |
CN110794782A (en) * | 2019-11-08 | 2020-02-14 | 中国矿业大学 | Batch industrial process online quality prediction method based on JY-MKPLS |
CN113031553A (en) * | 2021-03-09 | 2021-06-25 | 北京工业大学 | Intermittent process quality prediction method based on automatic clustering combined with partial least squares |
CN113311796A (en) * | 2021-06-04 | 2021-08-27 | 北京工业大学 | Fermentation process stage division method based on joint typical variable matrix |
CN115165884A (en) * | 2022-08-03 | 2022-10-11 | 湖南粒丰生物科技有限公司 | Biological fermentation feed quality detection method and device, electronic equipment and storage medium |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9541471B2 (en) * | 2012-04-06 | 2017-01-10 | Mks Instruments, Inc. | Multivariate prediction of a batch manufacturing process |
-
2024
- 2024-01-18 CN CN202410074415.7A patent/CN117598391B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103970092A (en) * | 2014-04-13 | 2014-08-06 | 北京工业大学 | Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm |
CN107895224A (en) * | 2017-10-30 | 2018-04-10 | 北京工业大学 | A kind of MKECA fermentation process fault monitoring methods based on extension nuclear entropy load matrix |
WO2019089730A1 (en) * | 2017-10-31 | 2019-05-09 | Locus Ip Company, Llc | Matrix fermentation systems and methods for producing microbe-based products |
WO2019129891A1 (en) * | 2017-12-29 | 2019-07-04 | F. Hoffmann-La Roche Ag | Predicting the metabolic condition of a cell culture |
CN110794782A (en) * | 2019-11-08 | 2020-02-14 | 中国矿业大学 | Batch industrial process online quality prediction method based on JY-MKPLS |
CN113031553A (en) * | 2021-03-09 | 2021-06-25 | 北京工业大学 | Intermittent process quality prediction method based on automatic clustering combined with partial least squares |
CN113311796A (en) * | 2021-06-04 | 2021-08-27 | 北京工业大学 | Fermentation process stage division method based on joint typical variable matrix |
CN115165884A (en) * | 2022-08-03 | 2022-10-11 | 湖南粒丰生物科技有限公司 | Biological fermentation feed quality detection method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
Title |
---|
"Analysis of the Actuality and Obstacles of Organizational Learning Based on Knowledge Fermenting Theory";Jing Sui;《IEEE》;20091228;第1-5页 * |
基于RMKMFDA的间歇过程监测与故障诊断;肖应旺;杨军;张承忠;姚美银;杜瑛;;计算机与应用化学;20140128(第01期);第79-85页 * |
Also Published As
Publication number | Publication date |
---|---|
CN117598391A (en) | 2024-02-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117598391B (en) | Determination method of fermentation process conditions and feed fermentation process monitoring method | |
Lopes et al. | Chemometrics in bioprocess engineering: process analytical technology (PAT) applications | |
CN108664009B (en) | Stage division and fault detection method based on correlation analysis | |
Mjolsness et al. | From coexpression to coregulation: an approach to inferring transcriptional regulation among gene classes from large-scale expression data | |
CN111949012A (en) | Intermittent process fault detection method based on double-weight multi-neighborhood preserving embedding algorithm | |
CN111506041A (en) | Neighborhood preserving embedding intermittent process fault detection method based on diffusion distance improvement | |
CN115240772A (en) | Method for analyzing active pathway in unicellular multiomics based on graph neural network | |
CN109144039A (en) | A kind of batch process fault detection method keeping extreme learning machine based on timing extension and neighborhood | |
CN111008726A (en) | Class image conversion method in power load prediction | |
Sajedi et al. | An image analysis-aided method for redundancy reduction in differentiation of identical Actinobacterial strains | |
Besenhard et al. | A multivariate process monitoring strategy and control concept for a small-scale fermenter in a PAT environment | |
Potočnik et al. | Empirical modeling of antibiotic fermentation process using neural networks and genetic algorithms | |
Karim et al. | Data‐based modeling and analysis of bioprocesses: some real experiences | |
CN112348253B (en) | Solid state fermentation utility prediction method based on full-connection neural network | |
Cleary et al. | Composite measurements and molecular compressed sensing for highly efficient transcriptomics | |
He et al. | Improving KNN method based on reduced relational grade for microarray missing values imputation | |
Chan et al. | An efficient greedy k-means algorithm for global gene trajectory clustering | |
CN116665805A (en) | Digital twinning-based white spirit fermentation prediction and feedback intervention system | |
CN117690506B (en) | Prediction method of process boundary and evaluation method of feed production process | |
CN113341926B (en) | Multi-stage intermittent process fault detection method based on sparse weighted neighborhood preserving embedding | |
Mohammadi et al. | Estimating missing value in microarray data using fuzzy clustering and gene ontology | |
Fernández-Fernández et al. | Realistic model of a solid substrate fermentation packed-bed pilot bioreactor | |
Song et al. | A linear discrete dynamic system model for temporal gene interaction and regulatory network influence in response to bioethanol conversion inhibitor HMF for ethanologenic yeast | |
Bicciato et al. | Fermentation diagnosis by multivariate statistical analysis | |
Rafii et al. | Data pre-processing and reducing for microarray data exploration and analysis |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |