CN114134031A - Anaerobic fermentation temperature control system and method based on RBF neural network prediction - Google Patents

Anaerobic fermentation temperature control system and method based on RBF neural network prediction Download PDF

Info

Publication number
CN114134031A
CN114134031A CN202111421280.XA CN202111421280A CN114134031A CN 114134031 A CN114134031 A CN 114134031A CN 202111421280 A CN202111421280 A CN 202111421280A CN 114134031 A CN114134031 A CN 114134031A
Authority
CN
China
Prior art keywords
temperature
neural network
anaerobic fermentation
heat exchanger
waste heat
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
Application number
CN202111421280.XA
Other languages
Chinese (zh)
Inventor
朱建军
王强
张雁茹
张巍
王振江
祁晓乐
赵鹏翔
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NATIONAL BIO ENERGY GROUP CO LTD
Original Assignee
NATIONAL BIO ENERGY GROUP CO LTD
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by NATIONAL BIO ENERGY GROUP CO LTD filed Critical NATIONAL BIO ENERGY GROUP CO LTD
Priority to CN202111421280.XA priority Critical patent/CN114134031A/en
Publication of CN114134031A publication Critical patent/CN114134031A/en
Pending legal-status Critical Current

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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Wood Science & Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Organic Chemistry (AREA)
  • Artificial Intelligence (AREA)
  • Zoology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Microbiology (AREA)
  • Sustainable Development (AREA)
  • Biotechnology (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Genetics & Genomics (AREA)
  • Thermal Sciences (AREA)
  • Processing Of Solid Wastes (AREA)
  • Purification Treatments By Anaerobic Or Anaerobic And Aerobic Bacteria Or Animals (AREA)

Abstract

The invention provides an anaerobic fermentation temperature control system and method based on RBF neural network prediction, the system mainly comprises a flue gas waste heat exchanger, a slag discharge waste heat exchanger, an exhaust steam waste heat exchanger, a heating circulation heat exchanger, a temperature monitor, a constant temperature water storage tank and an RBF neural network controller, and the control method comprises the following steps: and constructing a neural network model, acquiring a target temperature, an actual temperature and a temperature control parameter sample of the anaerobic fermentation system, performing neural network learning, identifying a system control rule on line, setting a controller by using a neural network learning result, and adjusting a temperature control parameter to enable the actual temperature of the anaerobic fermentation to be consistent with the target temperature. The temperature control system utilizes the waste heat of the power plant to maintain the anaerobic fermentation temperature, has obvious energy-saving benefit, overcomes the uncertainty and time-varying interference of the system by utilizing the extremely strong nonlinear fitting capability of the neural network, ensures the temperature control precision and accuracy, and has better robustness.

Description

Anaerobic fermentation temperature control system and method based on RBF neural network prediction
Technical Field
The invention belongs to the field of biomass energy utilization, and particularly relates to an anaerobic fermentation temperature control system and method based on RBF neural network prediction.
Background
The anaerobic fermentation is a process of obtaining methane-rich methane by decomposing organic matters through microorganisms under a proper condition, is one of large-scale and favorable biomass energy utilization modes, and is particularly suitable for high-efficiency utilization of low-calorific-value biomass such as livestock excrement, wet straws and the like. In the anaerobic fermentation process, the temperature is one of the key factors influencing the biogas production, and according to the temperature difference of the biogas digester, the anaerobic fermentation is generally divided into normal-temperature fermentation (10-30 ℃), medium-temperature fermentation (30-40 ℃) and high-temperature fermentation (50-60 ℃). In addition, temperature fluctuation has a great influence on the anaerobic fermentation efficiency, generally speaking, the temperature fluctuation of the anaerobic fermentation per day is controlled within +/-2 ℃, when the temperature fluctuation reaches +/-3 ℃, the anaerobic fermentation rate is inhibited, and when the temperature variation reaches +/-5 ℃, the gas yield is obviously reduced. In northern cold areas, the temperature is low, and the temperature difference between winter, summer and day and night is large, in order to keep stable and efficient gas production, appropriate heating and heat preservation measures need to be taken, and the anaerobic fermentation temperature is strictly controlled so as not to be interfered by factors such as external environment temperature and the like. Therefore, the reasonable selection and design of the anaerobic fermentation heating and heat preservation mode and the control method are the problems to be solved urgently in developing large and medium-sized biogas projects in northern cold areas, and are the key points for the application and popularization of the biogas projects in the cold areas.
At present, anaerobic fermentation in cold regions is mainly carried out by burning methane partially to maintain the optimal anaerobic fermentation temperature, and the anaerobic fermentation has the defects of high energy consumption, reduction of methane yield and the like. In addition, the conventional anaerobic fermentation temperature control system is mainly a traditional PID control algorithm, and the anaerobic fermentation process has the characteristics of large fluctuation, strong time variation, many external interference factors and the like, so that the PID algorithm is easy to have the defects of large overshoot, long stabilization time and the like, and the requirement of anaerobic fermentation on the temperature control precision is difficult to meet. Therefore, it is necessary to develop a system and a method for controlling anaerobic fermentation temperature in cold regions with obvious energy-saving benefit.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an anaerobic fermentation temperature control system and method based on RBF neural network prediction.
The system recovers the low-grade waste heat generated by biomass direct-fired power generation, uses the low-grade waste heat for heating and temperature maintenance of an anaerobic fermentation system, dynamically predicts the system by using the nonlinear fitting capacity of the RBF neural network so as to control the anaerobic fermentation temperature, and overcomes the uncertainty, time-varying property and volatility of the direct-fired power generation and anaerobic fermentation systems, so that the anaerobic fermentation temperature is quickly, accurately and effectively controlled.
In order to achieve the above objects and achieve the above technical effects, the present invention is implemented by the following technical solutions: an anaerobic fermentation temperature control system based on RBF neural network prediction comprises a flue gas waste heat exchanger, a slag discharge waste heat exchanger, an exhaust steam waste heat exchanger, a heating circulation heat exchanger, a temperature monitor, a constant temperature water storage tank and a neural network controller; the flue gas waste heat exchanger is connected with the constant-temperature water storage tank, and flue gas waste heat of the direct-fired power generation system is recovered to heat circulating water and send the circulating water to the constant-temperature water storage tank; the deslagging waste heat exchanger is connected with the constant-temperature water storage tank, and heating circulating water by recycling deslagging waste heat of the direct-combustion power generation system is sent to the constant-temperature water storage tank; the exhaust steam waste heat exchanger is connected with the constant-temperature water storage tank, and the exhaust steam waste heat recovered from the direct-combustion power generation system heats circulating water and sends the circulating water to the constant-temperature water storage tank; the heating circulation heat exchanger is connected with the constant-temperature water storage tank and the anaerobic fermentation system, hot water is input from the constant-temperature water storage tank by the heat exchanger to be supplied to the anaerobic fermentation system, and then cold water is input back to the constant-temperature water storage tank; the temperature monitor detects the temperature of the constant-temperature water storage tank, the circulating water and the anaerobic fermentation tank constantly and outputs the result to the neural network controller, and the controller controls the output of the heating circulating heat exchanger according to the difference value between the set target temperature and the actually measured temperature, so that the actual temperature is consistent with the target temperature.
Preferably, in the anaerobic fermentation temperature control system based on RBF neural network prediction, the flue gas waste heat exchanger adopts a low-temperature economizer.
Preferably, in the anaerobic fermentation temperature control system based on RBF neural network prediction, the slagging waste heat exchanger adopts a water-cooled slag cooler which is one of a jacketed roller slag cooler or a film-wall roller slag cooler.
Preferably, in the anaerobic fermentation temperature control system based on RBF neural network prediction, the exhaust steam waste heat exchanger adopts one of an absorption heat pump waste heat recovery technology or a low vacuum waste heat recovery technology to recover waste heat.
The embodiment of the invention also provides a control method of the anaerobic fermentation temperature control system, which comprises the following steps:
step 1, constructing an RBF neural network;
step 2, sampling to obtain a target temperature, an actual temperature and a temperature control parameter real-time sample of the anaerobic fermentation system;
step 3, learning a neural network by using the samples, and identifying a system control rule on line;
step 4, calculating the output u of the neural network controller by using the neural network learning result, namely the output of the heating cycle heat exchanger;
and 5, adjusting the output of the temperature control parameters, and repeating the steps 2-5 until the actual temperature of the anaerobic fermentation system is consistent with the target temperature.
Preferably, in the control method of the anaerobic fermentation temperature control system, the RBF neural network in step 1 includes three layers of structures, namely an input layer, a hidden layer and an output layer, and the mathematical expression of the network is as follows:
Figure BDA0003377483260000031
in the formula: xmRepresenting the input vector, yjFor the output parameter, R is the hidden layer Gaussian kernel function output, ciIs the central point of the Gaussian kernel function, sigma is the spreading constant,w ijthe connection weight of the hidden layer and the output layer.
Preferably, in the control method of the anaerobic fermentation temperature control system, the target anaerobic fermentation temperature in step 2 is an optimal active temperature of an anaerobic fermentation strain, the actual temperature is a measured temperature of an anaerobic fermentation tank, and the temperature control parameter is a circulating water flow of a heating circulating heat exchanger.
Preferably, in the control method of the anaerobic fermentation temperature control system, the online identification system control rule in step 3 includes the following steps:
1) calculating neural network identification output according to the sampled data samples;
Figure BDA0003377483260000041
in the formula, k is sampling time, Nf and Ng are output of RBF neural network, and Tm(k) For neural network identification output, To(k) For the actual output of the system, n and m are the output and input delay orders respectively, and n is more than or equal to m.
2) Calculating the error E between the output of the neural network and the actual output, and adjusting the parameters of the neural network according to a gradient descent method;
Figure BDA0003377483260000042
3) repeating the steps 1) and 2) until the error is minimum or iteration exceeds 10 times.
Preferably, in the control method of the anaerobic fermentation temperature control system, the output u of the neural network controller in the step 4 is calculated as follows:
Figure BDA0003377483260000043
wherein Nf and Ng are respectively identified RBF neural network, TrIs the target output.
The invention has the beneficial effects that:
1. the exhaust tail gas, ash and exhaust steam of the direct-fired power generation system have rich low-grade waste heat, and the waste heat recovery system is arranged for supplying heat to the anaerobic fermentation system through gradient utilization of energy, so that methane or other fuels can be prevented from being combusted to provide heat to maintain the anaerobic fermentation temperature, and the direct-fired power generation system has remarkable energy-saving benefit.
2. The RBF neural network can approach any nonlinear function, can process the regularity which is difficult to analyze in the system, has good generalization capability and high learning convergence speed, is particularly suitable for data analysis and processing with time-varying characteristics, has the characteristics of multiple interference factors, large parameter fluctuation, strong time variation and the like of a direct-fired power generation system and an anaerobic fermentation system, and can be used for carrying out prediction control and dynamic adjustment on the system by establishing the RBF neural network, thereby greatly improving the response speed and accuracy of the control system.
Drawings
FIG. 1 is a schematic diagram of an anaerobic fermentation temperature control system according to the present invention;
FIG. 2 is a structural diagram of a temperature control method for anaerobic fermentation according to the present invention;
fig. 3 is a block diagram of an RBF neural network according to the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific embodiments. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In this embodiment, a biomass direct-fired power plant and an anaerobic fermentation system are taken as examples, and the system and the method of the invention are used for recovering the waste heat of the power plant and for heating and temperature control of anaerobic fermentation.
Example 1
An anaerobic fermentation temperature control system based on RBF neural network prediction is shown in figure 1 and comprises a flue gas waste heat exchanger, a slag discharge waste heat exchanger, an exhaust steam waste heat exchanger, a heating circulation heat exchanger, a temperature monitor, a constant temperature water storage tank and a neural network controller; the flue gas waste heat exchanger is connected with the constant-temperature water storage tank, and flue gas waste heat of the direct-fired power generation system is recovered to heat circulating water and send the circulating water to the constant-temperature water storage tank; the deslagging waste heat exchanger is connected with the constant-temperature water storage tank, and heating circulating water by recycling deslagging waste heat of the direct-combustion power generation system is sent to the constant-temperature water storage tank; the exhaust steam waste heat exchanger is connected with the constant-temperature water storage tank, and the exhaust steam waste heat recovered from the direct-combustion power generation system heats circulating water and sends the circulating water to the constant-temperature water storage tank; the heating circulation heat exchanger is connected with the constant-temperature water storage tank and the anaerobic fermentation system, hot water is input from the constant-temperature water storage tank by the heat exchanger to be supplied to the anaerobic fermentation system, and then cold water is input back to the constant-temperature water storage tank; the temperature monitor detects the temperature of the constant-temperature water storage tank, the circulating water and the anaerobic fermentation tank and outputs the result to the neural network controller, and the controller controls the output of the heating circulating heat exchanger according to the difference value between the set target temperature and the actually measured temperature, so that the actual temperature is consistent with the target temperature.
The flue gas waste heat exchanger of the preferred embodiment adopts a low-temperature economizer, the deslagging waste heat exchanger adopts a membrane wall roller slag cooler, and the exhaust steam waste heat exchanger adopts a low-vacuum waste heat recovery technology.
As shown in fig. 2, this embodiment further provides a control method of the anaerobic fermentation temperature control system, including the following steps:
step 1, constructing an RBF neural network. As shown in fig. 3, the RBF neural network includes three layers of structures, i.e. an input layer, a hidden layer and an output layer, and the mathematical expression of the network is as follows:
Figure BDA0003377483260000061
in the formula: xmRepresenting the input vector, yjIn order to output the parameters, the parameters are,r is the hidden layer Gaussian kernel function output, ciIs the central point of the Gaussian kernel function, sigma is the spreading constant,w ijthe connection weight of the hidden layer and the output layer.
And 2, sampling to obtain target temperature, actual temperature and temperature control factors of the anaerobic fermentation system. Setting the optimal active temperature of the anaerobic fermentation strain as a target temperature according to the anaerobic fermentation strain, measuring the actual temperature of the anaerobic fermentation tank in real time by a thermocouple arranged in the anaerobic fermentation tank, and taking the circulating water flow of the heating circulating heat exchanger as a temperature control variable.
Step 3, calculating the input and output of each layer of neuron of the neural network, and performing online identification learning on the control rule of the anaerobic fermentation system, wherein the online identification learning method comprises the following steps:
1) calculating the output of the neural network by using the following formula 2 according to the data sample obtained by sampling;
Figure BDA0003377483260000062
in the formula, k is sampling time, Nf and Ng are output of RBF neural network, and Tm(k) For neural network identification output, To(k) Is the actual output of the system.
2) Calculating error E between the output of the neural network and the actual output by using the following formula 3, and adjusting the parameters of the neural network, including data center c of radial basis function, by using a gradient descent methodiAnd the expansion constant sigma, the weight of the output nodew ij；
Figure BDA0003377483260000071
3) Repeating the steps 1) and 2) until the error is minimum or iteration exceeds 10 times.
Step 4, calculating the output u of the neural network controller according to the following formula 4 by using the learning result of the neural network, namely the circulating water flow of the heating circulating heat exchanger;
Figure BDA0003377483260000072
wherein Nf and Ng are respectively identified RBF neural network, TrIs the target output.
And 5, adjusting the flow of circulating water, and repeating the steps 2-4 until the actual temperature of anaerobic fermentation is equal to the target temperature.
Through the operation of the steps, the method controls the fluctuation of the anaerobic fermentation temperature not to exceed +/-0.5 ℃, and the adjustment time not to exceed 8 min.
Example 2
An anaerobic fermentation temperature control system based on RBF neural network prediction is shown in figure 1 and comprises a flue gas waste heat exchanger, a slag discharge waste heat exchanger, an exhaust steam waste heat exchanger, a heating circulation heat exchanger, a temperature monitor, a constant temperature water storage tank and a neural network controller; the flue gas waste heat exchanger is connected with the constant-temperature water storage tank, and flue gas waste heat of the direct-fired power generation system is recovered to heat circulating water and send the circulating water to the constant-temperature water storage tank; the deslagging waste heat exchanger is connected with the constant-temperature water storage tank, and heating circulating water by recycling deslagging waste heat of the direct-combustion power generation system is sent to the constant-temperature water storage tank; the exhaust steam waste heat exchanger is connected with the constant-temperature water storage tank, and the exhaust steam waste heat recovered from the direct-combustion power generation system heats circulating water and sends the circulating water to the constant-temperature water storage tank; the heating circulation heat exchanger is connected with the constant-temperature water storage tank and the anaerobic fermentation system, hot water is input from the constant-temperature water storage tank by the heat exchanger to be supplied to the anaerobic fermentation system, and then cold water is input back to the constant-temperature water storage tank; the temperature monitor detects the temperature of the constant-temperature water storage tank, the circulating water and the anaerobic fermentation tank and outputs the result to the neural network controller, and the controller controls the output of the heating circulating heat exchanger according to the difference value between the set target temperature and the actually measured temperature, so that the actual temperature is consistent with the target temperature.
The flue gas waste heat exchanger of the preferred embodiment adopts a low-temperature economizer, the deslagging waste heat exchanger adopts a membrane wall roller slag cooler, and the exhaust steam waste heat exchanger adopts a low-vacuum waste heat recovery technology.
As shown in fig. 2, this embodiment further provides a control method of the anaerobic fermentation temperature control system, including the following steps:
step 1, constructing an RBF neural network. As shown in fig. 3, the RBF neural network includes three layers of structures, i.e. an input layer, a hidden layer and an output layer, and the mathematical expression of the network is as follows:
Figure BDA0003377483260000081
in the formula: xmRepresenting the input vector, yjFor the output parameter, R is the hidden layer Gaussian kernel function output, ciIs the central point of the Gaussian kernel function, sigma is the spreading constant,w ijthe connection weight of the hidden layer and the output layer.
And 2, sampling to obtain target temperature, actual temperature and temperature control factors of the anaerobic fermentation system. Setting the optimal active temperature of the anaerobic fermentation strain as a target temperature according to the anaerobic fermentation strain, measuring the actual temperature of the anaerobic fermentation tank in real time by a thermocouple arranged in the anaerobic fermentation tank, and taking the circulating water flow of the heating circulating heat exchanger as a temperature control variable.
Step 3, calculating the input and output of each layer of neuron of the neural network, and performing online identification learning on the control rule of the anaerobic fermentation system, wherein the online identification learning method comprises the following steps:
1) calculating the output of the neural network by using the following formula 2 according to the data sample obtained by sampling;
Figure BDA0003377483260000082
in the formula, k is sampling time, Nf and Ng are output of RBF neural network, and Tm(k) For neural network identification output, To(k) Is the actual output of the system.
2) Calculating error E between the output of the neural network and the actual output by using the following formula 3, and adjusting the parameters of the neural network, including data center c of radial basis function, by using a gradient descent methodiAnd the expansion constant sigma, the weight of the output nodew ij；
Figure BDA0003377483260000091
3) Repeating the steps 1) and 2) until the error is minimum or iteration exceeds 10 times.
Step 4, calculating the output u of the neural network controller according to the following formula 4 by using the learning result of the neural network, namely the circulating water flow of the heating circulating heat exchanger;
Figure BDA0003377483260000092
wherein Nf and Ng are respectively identified RBF neural network, TrIs the target output.
And 5, adjusting the flow of circulating water, and repeating the steps 2-4 until the actual temperature of anaerobic fermentation is equal to the target temperature.
Through the operation of the steps, the method controls the fluctuation of the anaerobic fermentation temperature not to exceed +/-0.4 ℃, and the adjustment time not to exceed 10 min.

Claims (9)

1. An anaerobic fermentation temperature control system based on RBF neural network prediction is characterized by comprising a flue gas waste heat exchanger, a slag discharge waste heat exchanger, an exhaust steam waste heat exchanger, a heating circulation heat exchanger, a temperature monitor, a constant temperature water storage tank and an RBF neural network controller; the flue gas waste heat exchanger is connected with the constant-temperature water storage tank, and flue gas waste heat of the direct-fired power generation system is recovered to heat circulating water and send the circulating water to the constant-temperature water storage tank; the deslagging waste heat exchanger is connected with the constant-temperature water storage tank, and heating circulating water by recycling deslagging waste heat of the direct-combustion power generation system is sent to the constant-temperature water storage tank; the exhaust steam waste heat exchanger is connected with the constant-temperature water storage tank, and the exhaust steam waste heat recovered from the direct-combustion power generation system heats circulating water and sends the circulating water to the constant-temperature water storage tank; the heating circulation heat exchanger is connected with the constant-temperature water storage tank and the anaerobic fermentation system, hot water is input from the constant-temperature water storage tank by the heat exchanger to be supplied to the anaerobic fermentation system, and then cold water is input back to the constant-temperature water storage tank; the temperature monitor detects the temperature of the constant-temperature water storage tank, the circulating water and the anaerobic fermentation tank and outputs the temperature measurement result to the RBF neural network controller, and the controller controls the output of the heating circulation heat exchanger according to the difference value between the set target temperature and the actually measured temperature, so that the actual temperature is consistent with the target temperature.
2. The anaerobic fermentation temperature control system based on RBF neural network prediction as claimed in claim 1, wherein said flue gas waste heat exchanger employs a low temperature economizer.
3. The system for controlling the temperature of anaerobic fermentation based on RBF neural network prediction as claimed in claim 1, wherein said residual heat exchanger is a water-cooled slag cooler selected from a jacketed roller slag cooler and a film-wall roller slag cooler.
4. The anaerobic fermentation temperature control system based on RBF neural network prediction as claimed in claim 1, wherein said exhaust steam waste heat exchanger adopts one of absorption heat pump waste heat recovery technology or low vacuum waste heat recovery technology for waste heat recovery.
5. A method for controlling the temperature of anaerobic fermentation using the system of claim 1, comprising the steps of:
step 1, constructing an RBF neural network;
step 2, sampling to obtain a target temperature, an actual temperature and a temperature control parameter real-time sample of the anaerobic fermentation system;
step 3, learning a neural network by using the samples, and identifying a system control rule on line;
step 4, calculating the output u of the neural network controller by using the neural network learning result, namely the output of the heating cycle heat exchanger;
and 5, adjusting the output of the temperature control parameters, and repeating the steps 2-5 until the actual temperature of the anaerobic fermentation system is consistent with the target temperature.
6. The method for controlling the temperature of anaerobic fermentation according to claim 5, wherein the RBF neural network in step 1 comprises three layers of an input layer, a hidden layer and an output layer, and the mathematical expression of the network is as follows:
Figure FDA0003377483250000021
in the formula: xmRepresenting the input vector, yjFor the output parameter, R is the hidden layer Gaussian kernel function output, ciIs the central point of the Gaussian kernel function, sigma is the spreading constant,w ijthe connection weight of the hidden layer and the output layer.
7. The method as claimed in claim 5, wherein the anaerobic fermentation target temperature in step 2 is an optimum activity temperature of anaerobic fermentation bacteria, the actual temperature is a measured temperature of the anaerobic fermentation tank, and the temperature control parameter is a circulation water flow rate of the heating circulation heat exchanger.
8. The method for controlling the temperature of anaerobic fermentation according to claim 5, wherein the online identification system control rule in step 3 comprises the following steps:
1) calculating neural network identification output according to the sampled data samples;
Figure FDA0003377483250000022
in the formula, k is sampling time, Nf and Ng are output of RBF neural network, and Tm(k) For neural network identification output, To(k) N and m are output and input delay orders respectively for actual output of the system, and n is more than or equal to m;
2) calculating the error E between the output of the neural network and the actual output, and adjusting the parameters of the neural network according to a gradient descent method;
Figure FDA0003377483250000031
3) repeating the steps 1) and 2) until the error is minimum or iteration exceeds 10 times.
9. The method for controlling the temperature of anaerobic fermentation according to claim 5, wherein the output u of the neural network controller in the step 4 is calculated by:
Figure FDA0003377483250000032
wherein Nf and Ng are respectively identified RBF neural network, TrIs the target output.
CN202111421280.XA 2021-11-26 2021-11-26 Anaerobic fermentation temperature control system and method based on RBF neural network prediction Pending CN114134031A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111421280.XA CN114134031A (en) 2021-11-26 2021-11-26 Anaerobic fermentation temperature control system and method based on RBF neural network prediction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111421280.XA CN114134031A (en) 2021-11-26 2021-11-26 Anaerobic fermentation temperature control system and method based on RBF neural network prediction

Publications (1)

Publication Number Publication Date
CN114134031A true CN114134031A (en) 2022-03-04

Family

ID=80387951

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111421280.XA Pending CN114134031A (en) 2021-11-26 2021-11-26 Anaerobic fermentation temperature control system and method based on RBF neural network prediction

Country Status (1)

Country Link
CN (1) CN114134031A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118274650A (en) * 2024-06-03 2024-07-02 四川德润钢铁集团航达钢铁有限责任公司 Steel production waste heat recovery optimization method based on heat exchanger

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118274650A (en) * 2024-06-03 2024-07-02 四川德润钢铁集团航达钢铁有限责任公司 Steel production waste heat recovery optimization method based on heat exchanger

Similar Documents

Publication Publication Date Title
CN114134017A (en) Anaerobic fermentation temperature control system and method based on BP neural network prediction
Kana et al. Modeling and optimization of biogas production on saw dust and other co-substrates using artificial neural network and genetic algorithm
Willquist et al. Design of a novel biohythane process with high H2 and CH4 production rates
CN106842955B (en) CO after burning with exhaust gas volumn Disturbance Rejection2Trapping system forecast Control Algorithm
Chen et al. Energy, exergy, and economic analysis of a centralized solar and biogas hybrid heating system for rural areas
CN114153140A (en) Self-adaptive PID (proportion integration differentiation) anaerobic fermentation temperature control system and method based on Elman neural network
Han et al. Novel long short-term memory neural network considering virtual data generation for production prediction and energy structure optimization of ethylene production processes
CN114134031A (en) Anaerobic fermentation temperature control system and method based on RBF neural network prediction
CN114492041A (en) Optimization method and system for biogas cogeneration system of farm
Walid et al. Modeling and optimization of anaerobic digestion: a review
CN113341716B (en) Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system
Wang et al. NOx emission prediction of coal-fired power units under uncertain classification of operating conditions
Shokrkar et al. Model‐based evaluation of continuous bioethanol production plant
Huang et al. Fermentative hydrogen production using a real-time fuzzy controller
CN114958395B (en) Waste heat utilization method and system for biomass carbonization furnace
Nassef et al. Optimal operating parameter determination and modeling to enhance methane production from macroalgae
Zhang et al. Furnace temperature prediction using optimized kernel extreme learning machine
da Costa et al. Modeling, control and optimization of ethanol fermentation process
CN112734451B (en) Green house multi-energy system based on non-cooperative game and optimization method
Lin et al. Process design for very-high-gravity ethanol fermentation
CN114912363A (en) Soft measurement method for yarrowia lipolytica fermentation based on BP-LSTM model
Filali et al. Optimization of the interval approach for Chlorella vulgaris biomass estimation
Qing et al. Prediction model of the NOx emissions based on long short-time memory neural network
Bartee et al. Industrial application of nonlinear model predictive control technology for fuel ethanol fermentation process
Xu et al. Short-term power load forecasting based on deep learning

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