CN110643485A - Automatic control system and method for temperature field in grain fermentation process with intelligent prediction compensator - Google Patents
Automatic control system and method for temperature field in grain fermentation process with intelligent prediction compensator Download PDFInfo
- Publication number
- CN110643485A CN110643485A CN201910905192.3A CN201910905192A CN110643485A CN 110643485 A CN110643485 A CN 110643485A CN 201910905192 A CN201910905192 A CN 201910905192A CN 110643485 A CN110643485 A CN 110643485A
- Authority
- CN
- China
- Prior art keywords
- temperature
- control
- intelligent
- compensator
- intelligent prediction
- 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
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12M—APPARATUS FOR ENZYMOLOGY OR MICROBIOLOGY; APPARATUS FOR CULTURING MICROORGANISMS FOR PRODUCING BIOMASS, FOR GROWING CELLS OR FOR OBTAINING FERMENTATION OR METABOLIC PRODUCTS, i.e. BIOREACTORS OR FERMENTERS
- C12M41/00—Means for regulation, monitoring, measurement or control, e.g. flow regulation
- C12M41/12—Means for regulation, monitoring, measurement or control, e.g. flow regulation of temperature
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12M—APPARATUS FOR ENZYMOLOGY OR MICROBIOLOGY; APPARATUS FOR CULTURING MICROORGANISMS FOR PRODUCING BIOMASS, FOR GROWING CELLS OR FOR OBTAINING FERMENTATION OR METABOLIC PRODUCTS, i.e. BIOREACTORS OR FERMENTERS
- C12M41/00—Means for regulation, monitoring, measurement or control, e.g. flow regulation
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12M—APPARATUS FOR ENZYMOLOGY OR MICROBIOLOGY; APPARATUS FOR CULTURING MICROORGANISMS FOR PRODUCING BIOMASS, FOR GROWING CELLS OR FOR OBTAINING FERMENTATION OR METABOLIC PRODUCTS, i.e. BIOREACTORS OR FERMENTERS
- C12M41/00—Means for regulation, monitoring, measurement or control, e.g. flow regulation
- C12M41/48—Automatic or computerized control
-
- 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
Abstract
The invention discloses an automatic control system and method for a temperature field in a grain fermentation process, which comprises an intelligent temperature compensator based on an RNN model, wherein in a fermentation tank temperature control system, actually measured temperature data of a fermentation tank body is input into the intelligent prediction compensator, the intelligent prediction compensator outputs a control signal to a rear-end heating or cold control valve, the intelligent prediction compensator adopts a deep learning algorithm to realize a temperature control strategy of the fermentation tank which is continuously optimized on the basis of manual control experience, and after certain control data are accumulated, the intelligent temperature compensator can stably control the temperature of the tank body, eliminates instability phenomena such as hysteresis, overshoot and the like in the traditional control technology, and has high control stability.
Description
Technical Field
The invention relates to automatic control of a temperature field in a grain fermentation process with an intelligent prediction compensator, which is a control technology of a biological fermentation tank.
Background
Biological fermentation is a key process for producing seasonings, and the metabolic activity of microbial flora decomposes, converts and synthesizes different substances to generate metabolic products required by people. The fermentation process is a very complex biochemical reaction process, the key details of which are not known yet, and the influence of the change of the fermentation temperature on the fermentation process is very important according to the existing knowledge and experience.
The temperature change in the fermentation process is accompanied by the heat absorption and heat release reaction of the fermentation biochemical reaction and the comprehensive action of the temperature, humidity, wind field, light field and air pressure around the fermentation tank. The temperature reaction of the tank body is the result of the comprehensive action, lags behind the fermentation biochemical reaction, and the temperature condition of the central part of the tank body cannot be accurately reflected, so that the obtained temperature data is not only lagged but also macroscopically reflected after the multi-factor synthesis compared with the real-time temperature of the fermentation biochemical reaction based on the limitation of the existing measuring means. This fact forces the temperature control of the fermenter to be still relatively dependent on manual experience.
The traditional automatic control, especially the control according to the established program is not easy to adapt to the change of the fermentation process, the implemented temperature control always lags behind the fermentation biochemical reaction, and the practical application result shows that the control using the traditional established program strategy can only reduce the physical force of manual operation, is difficult to improve the fermentation quality, cannot achieve the satisfactory control effect, often has the systematic response lag or overshoot, cannot rapidly reach the stable state, and even has the phenomenon of repeated oscillation.
Disclosure of Invention
The invention aims to provide an automatic control system and method for a temperature field in a grain fermentation process, which are used for improving the traditional method, realizing real-time self-adaptive online control of the whole fermentation process in a fermentation tank, and can pre-judge the future temperature control of the tank body according to the change of environmental working conditions and the prior experience so as to control the temperature and meet the requirement of the real-time required temperature of fermentation biochemical reaction.
In order to achieve the above object of the present invention, the method of the present invention is as follows.
The automatic control method of the temperature field in the grain fermentation process with the intelligent prediction compensator comprises the temperature intelligent compensator based on an RNN model, in a fermentation tank temperature control system, the actually measured temperature data of a fermentation tank body is input into the intelligent prediction compensator, the intelligent prediction compensator outputs a control signal to a rear end heating or cold control valve, and a deep learning model of the intelligent prediction compensator is implemented according to the following steps:
a. establishing a deep learning time series RNN network, wherein the final output of the mathematical model of the structural network at the time t is as the following formula,
ht＝Uxt+Wst-1
st＝f(ht)
ot＝g(Vst)
w, U, V, wherein the weights are respectively the weight coefficients to be optimized by the network;
b. optimizing the network weight coefficient based on back propagation by using a gradient descent method through the error sum;
c. learning and optimizing the network by using the measured fermentation measurement temperature s of the measured temperature field and the behavior input x of the temperature regulation control valve;
the prediction of the control behavior of the intelligent prediction compensator is implemented as follows:
d. establishing a reinforcement learning model; the optimized control of the temperature regulating valve of the fermentation tank is realized by utilizing a reinforcement learning model; several key elements of the reinforcement learning model are respectively Agent, Environment, Reword r, Action a and State s.
e. Establishing a Q-table, wherein the row and the column of the Q-table respectively represent the values of state(s) and action (a), and the value Q (s, a) of the Q-table is used for measuring how well the current state(s) takes action (a); the value of Q (s, a) is judged and determined by the prediction result of the intelligent predictor based on RNN network learning in the previous step;
f. in the training process, the Bellman equation is adopted to update the Q-table as a discount coefficient; continuously updating the Q-table in a circulating mode until the change of the value of the Q-table is smaller than a given critical value;
g. and selecting a control strategy for optimally adjusting the temperature field valve according to the optimal Q-table value, so as to realize accurate regulation and control of the temperature of the fermentation tank.
The system of the present invention is as follows.
An automatic control system with an intelligent prediction compensator for a temperature field in a grain fermentation process is composed of the intelligent prediction compensator, a valve actuating mechanism, a flow regulating valve and a fermentation tank, and is also provided with an angular displacement sensor for detecting the angular displacement of the flow regulating valve and feeding back the detection result to the actuating mechanism, and a temperature sensor for detecting the temperature of the fermentation tank and feeding back the detection result to the intelligent prediction compensator; the intelligent prediction compensator is an intelligent prediction compensator with an RNN deep learning model configured according to the method of the invention.
The intelligent temperature compensator with the RNN deep learning function is adopted, the experience data of manually controlling the temperature can be introduced, the temperature of the fermentation tank is monitored, the temperature control strategy of the fermentation tank is continuously optimized on the basis of the manual control experience, and after certain control data are accumulated, the intelligent temperature compensator can stably control the temperature of the tank body, the instability phenomena of hysteresis, overshoot and the like in the traditional control technology are eliminated, and the control stability is high.
Drawings
FIG. 1 is a schematic diagram of a temperature measurement control arrangement of a fermenter according to an embodiment of the present invention;
FIG. 2 is a control system diagram of an embodiment of the present invention;
FIG. 3 is a RNN structural diagram according to an embodiment of the present invention;
fig. 4 is an expanded view of a hidden layer according to an embodiment of the present invention.
Detailed Description
As shown in fig. 1, the temperature measurement control layout of the fermentation tank of the embodiment is constructed, three temperature measurement points are arranged on the fermentation tank, three cooling modules are correspondingly arranged, each cooling module comprises a cooling jacket, a flow valve and a valve actuating mechanism installed on the flow valve, and the cooling module adopts water with the temperature of below 25 ℃.
The control system of the present embodiment is set up as shown in fig. 2.
As shown in fig. 3, a deep learning time series RNN network is established. Wherein the structure of the hierarchical development diagram of the hidden layer is shown in fig. 4, the final output of the structure network mathematical model at the time t is shown in the following formula,
ht＝Uxt+Wst-1
st＝f(ht)
ot＝g(Vst)
w, U, V, wherein the weights are respectively the weight coefficients to be optimized by the network; optimizing the network weight coefficient based on back propagation by using a gradient descent method through the error sum; the learning and optimization of the network are realized by using the measured fermentation temperature s of the temperature field and the behavior input x of the temperature adjusting control valve.
The prediction of the control behavior of the intelligent prediction compensator is implemented as follows:
establishing a reinforcement learning model; the optimized control of the temperature regulating valve of the fermentation tank is realized by utilizing a reinforcement learning model; several key elements of the reinforcement learning model are respectively Agent, Environment, Reword r, Action a and State s.
Establishing a Q-table, wherein the row and the column of the Q-table respectively represent the values of state(s) and action (a), and the value Q (s, a) of the Q-table is used for measuring how well the current state(s) takes action (a); the value of Q (s, a) is determined and decided by the prediction result of the RNN learning-based intelligent predictor in the previous step.
In the training process, the Bellman equation is adopted to update the Q-table, which is a discount coefficient. And continuously and circularly updating the Q-table until the change of the value of the Q-table is smaller than a given critical value.
g. And selecting a control strategy for optimally adjusting the temperature field valve according to the optimal Q-table value, so as to realize accurate regulation and control of the temperature of the fermentation tank.
Through the system and the method of the embodiment, the normalization execution is performed in the production process, the consistency of the actually measured temperature of the fermentation system and the process target temperature required by the fermentation can be realized, the instability phenomena of hysteresis, overshoot and the like in the traditional control technology are eliminated, and the control stability is high.
Claims (2)
1. The automatic control method of the temperature field in the grain fermentation process with the intelligent prediction compensator comprises the temperature intelligent compensator based on an RNN model, in a fermentation tank temperature control system, the actually measured temperature data of a fermentation tank body is input into the intelligent prediction compensator, and the intelligent prediction compensator outputs a control signal to a rear end heating or cold control valve, and is characterized in that a deep learning model of the intelligent prediction compensator is implemented according to the following steps:
a. establishing a deep learning time series RNN network, wherein the final output of the mathematical model of the structural network at the time t is as the following formula,
ht＝Uxt+Wst-1
st＝f(ht)
ot＝g(Vst)
w, U, V, wherein the weights are respectively the weight coefficients to be optimized by the network;
b. optimizing the network weight coefficient based on back propagation by using a gradient descent method through the error sum;
c. learning and optimizing the network by using the measured fermentation measurement temperature s of the measured temperature field and the behavior input x of the temperature regulation control valve;
the prediction of the control behavior of the intelligent prediction compensator is implemented as follows:
d. establishing a reinforcement learning model; the optimized control of the temperature regulating valve of the fermentation tank is realized by utilizing a reinforcement learning model; several key elements of the reinforcement learning model are respectively Agent, Environment, Reword r, Action a and State s.
e. Establishing a Q-table, wherein the row and the column of the Q-table respectively represent the values of state(s) and action (a), and the value Q (s, a) of the Q-table is used for measuring how well the current state(s) takes action (a); the value of Q (s, a) is judged and determined by the prediction result of the intelligent predictor based on RNN network learning in the previous step;
f. in the training process, the Bellman equation is adopted to update the Q-table as a discount coefficient; continuously updating the Q-table in a circulating mode until the change of the value of the Q-table is smaller than a given critical value;
g. and selecting a control strategy for optimally adjusting the temperature field valve according to the optimal Q-table value, so as to realize accurate regulation and control of the temperature of the fermentation tank.
2. An automatic control system with an intelligent prediction compensator for a temperature field in a grain fermentation process comprises the following steps: the intelligent prediction system consists of an intelligent prediction compensator, a valve actuating mechanism, a flow regulating valve and a fermentation tank, and is also provided with an angular displacement sensor for detecting the angular displacement of the flow regulating valve and feeding back a detection result to the actuating mechanism, and a temperature sensor for detecting the temperature of the fermentation tank and feeding back the detection result to the intelligent prediction compensator; the intelligent prediction compensator is an intelligent prediction compensator with an RNN deep learning model configured according to the method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910905192.3A CN110643485A (en) | 2019-09-24 | 2019-09-24 | Automatic control system and method for temperature field in grain fermentation process with intelligent prediction compensator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910905192.3A CN110643485A (en) | 2019-09-24 | 2019-09-24 | Automatic control system and method for temperature field in grain fermentation process with intelligent prediction compensator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110643485A true CN110643485A (en) | 2020-01-03 |
Family
ID=68992410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910905192.3A Pending CN110643485A (en) | 2019-09-24 | 2019-09-24 | Automatic control system and method for temperature field in grain fermentation process with intelligent prediction compensator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110643485A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113325897A (en) * | 2021-06-10 | 2021-08-31 | 西北农林科技大学 | Automatic cooling control system and method for wine fermentation |
| CN116144489A (en) * | 2023-04-19 | 2023-05-23 | 山东土木启生物科技有限公司 | Automatic control system for microbial fermentation |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101370926A (en) * | 2006-01-28 | 2009-02-18 | Abb研究有限公司 | Method for on-line future performance estimation of fermentation apparatus |
| US20100257866A1 (en) * | 2007-04-12 | 2010-10-14 | Daniel Schneegass | Method for computer-supported control and/or regulation of a technical system |
| CN107367929A (en) * | 2017-07-19 | 2017-11-21 | 北京上格云技术有限公司 | Update method, storage medium and the terminal device of Q value matrixs |
| CN109753872A (en) * | 2018-11-22 | 2019-05-14 | 四川大学 | Intensified learning units match Recognition with Recurrent Neural Network system and its training and prediction technique |
| KR20190096311A (en) * | 2019-06-04 | 2019-08-19 | 엘지전자 주식회사 | A device for generating a temperature prediction model and a method for providing a simulation environment |
| CN110187727A (en) * | 2019-06-17 | 2019-08-30 | 武汉理工大学 | A kind of Glass Furnace Temperature control method based on deep learning and intensified learning |
-
2019
- 2019-09-24 CN CN201910905192.3A patent/CN110643485A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101370926A (en) * | 2006-01-28 | 2009-02-18 | Abb研究有限公司 | Method for on-line future performance estimation of fermentation apparatus |
| US20100257866A1 (en) * | 2007-04-12 | 2010-10-14 | Daniel Schneegass | Method for computer-supported control and/or regulation of a technical system |
| CN107367929A (en) * | 2017-07-19 | 2017-11-21 | 北京上格云技术有限公司 | Update method, storage medium and the terminal device of Q value matrixs |
| CN109753872A (en) * | 2018-11-22 | 2019-05-14 | 四川大学 | Intensified learning units match Recognition with Recurrent Neural Network system and its training and prediction technique |
| KR20190096311A (en) * | 2019-06-04 | 2019-08-19 | 엘지전자 주식회사 | A device for generating a temperature prediction model and a method for providing a simulation environment |
| CN110187727A (en) * | 2019-06-17 | 2019-08-30 | 武汉理工大学 | A kind of Glass Furnace Temperature control method based on deep learning and intensified learning |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113325897A (en) * | 2021-06-10 | 2021-08-31 | 西北农林科技大学 | Automatic cooling control system and method for wine fermentation |
| CN116144489A (en) * | 2023-04-19 | 2023-05-23 | 山东土木启生物科技有限公司 | Automatic control system for microbial fermentation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109581870B (en) | Dynamic matrix control method for temperature in energetic material reaction kettle | |
| CN102096373B (en) | Microwave drying PID (proportion integration differentiation) control method based on increment improved BP (back propagation) neural network | |
| Trelea et al. | Predictive modelling of brewing fermentation: from knowledge-based to black-box models | |
| Arauzo-Bravo et al. | Automatization of a penicillin production process with soft sensors and an adaptive controller based on neuro fuzzy systems | |
| CN104597842A (en) | BP neutral network heavy machine tool thermal error modeling method optimized through genetic algorithm | |
| CN110643485A (en) | Automatic control system and method for temperature field in grain fermentation process with intelligent prediction compensator | |
| CN112947088B (en) | Modeling and control method of temperature and humidity system based on closed space | |
| CN103123460A (en) | Temperature control system and temperature control method | |
| CN110609476B (en) | Multivariate nonlinear dynamic system model prediction control method based on Gaussian process model | |
| CN110097929A (en) | A kind of blast furnace molten iron silicon content on-line prediction method | |
| Zhou et al. | Application of an adaptive PID controller enhanced by a differential evolution algorithm for precise control of dissolved oxygen in recirculating aquaculture systems | |
| CN111123871B (en) | Prediction function control method for genetic algorithm optimization of chemical process | |
| Andrášik et al. | On-line tuning of a neural PID controller based on plant hybrid modeling | |
| CN104950670A (en) | Integrated multi-model method for controlling CSTRs (continuous stirred tank reactors) | |
| CN101639902A (en) | Modeling method of support vector machine (SVM)-based software measurement instrument in biological fermentation process | |
| CN101126917A (en) | Groove type reactor non-linear predication control method based on multi-kernel support vector machine | |
| CN105676645A (en) | Double-loop water tank liquid level prediction control method based on function type weight RBF-ARX model | |
| CN103760931A (en) | Oil-gas-water horizontal type three-phase separator pressure control method optimized through dynamic matrix control | |
| CN113126488B (en) | Intelligent substance mixing system | |
| Mao et al. | Coating thickness modeling and prediction for hot-dip galvanized steel strip based on GA-BP neural network | |
| CN107168066B (en) | Greenhouse environment self-adaptive control method | |
| CN116880201A (en) | Water network channel state control system based on fuzzy neural network | |
| Julien et al. | Bioreactor monitoring, modeling, and simulation | |
| Kiran et al. | Control of continuous fed-batch fermentation process using neural network based model predictive controller | |
| Arpaia et al. | Model predictive control strategy based on differential discrete particle swarm optimization |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200103 |