CN115938493A - Online optimization method of high-temperature solid oxide electrolytic hydrogen production system - Google Patents
Online optimization method of high-temperature solid oxide electrolytic hydrogen production system Download PDFInfo
- Publication number
- CN115938493A CN115938493A CN202211468436.4A CN202211468436A CN115938493A CN 115938493 A CN115938493 A CN 115938493A CN 202211468436 A CN202211468436 A CN 202211468436A CN 115938493 A CN115938493 A CN 115938493A
- Authority
- CN
- China
- Prior art keywords
- hydrogen production
- voltage
- model
- solid oxide
- temperature solid
- 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
- 239000001257 hydrogen Substances 0.000 title claims abstract description 83
- 229910052739 hydrogen Inorganic materials 0.000 title claims abstract description 83
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 title claims abstract description 81
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 67
- 238000005457 optimization Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000007787 solid Substances 0.000 title claims abstract description 27
- 230000007246 mechanism Effects 0.000 claims abstract description 17
- 238000013499 data model Methods 0.000 claims abstract description 9
- 238000012843 least square support vector machine Methods 0.000 claims abstract description 6
- 238000005096 rolling process Methods 0.000 claims abstract description 6
- 230000000737 periodic effect Effects 0.000 claims abstract description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 34
- 230000006870 function Effects 0.000 claims description 22
- 238000005868 electrolysis reaction Methods 0.000 claims description 19
- 239000002245 particle Substances 0.000 claims description 11
- 239000007789 gas Substances 0.000 claims description 8
- 230000010287 polarization Effects 0.000 claims description 6
- 230000002441 reversible effect Effects 0.000 claims description 6
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 5
- 239000001301 oxygen Substances 0.000 claims description 5
- 229910052760 oxygen Inorganic materials 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 239000003792 electrolyte Substances 0.000 claims description 3
- 230000003213 activating effect Effects 0.000 claims description 2
- 150000002431 hydrogen Chemical class 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000012706 support-vector machine Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000005299 abrasion Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- INJRKJPEYSAMPD-UHFFFAOYSA-N aluminum;silicic acid;hydrate Chemical compound O.[Al].[Al].O[Si](O)(O)O INJRKJPEYSAMPD-UHFFFAOYSA-N 0.000 description 1
- 239000012223 aqueous fraction Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000012943 hotmelt Substances 0.000 description 1
- 230000013016 learning Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009326 social learning Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/30—Hydrogen technology
- Y02E60/36—Hydrogen production from non-carbon containing sources, e.g. by water electrolysis
Landscapes
- Electrolytic Production Of Non-Metals, Compounds, Apparatuses Therefor (AREA)
Abstract
The invention relates to an online optimization method of a high-temperature solid oxide electrolytic hydrogen production system, which comprises the following steps: s1, initializing a model correction period parameter and an online optimization period parameter; s2, establishing a mechanism model and a data model of field data errors by adopting a Least Square Support Vector Machine (LSSVM) as error correction of electrolytic cell voltage prediction, setting an online optimization period, and performing periodic rolling correction on electrolytic cell voltage prediction; s3, establishing a power model of each auxiliary engine component of the high-temperature solid oxide electrolytic hydrogen production system, calculating the hydrogen production efficiency of the electrolytic hydrogen production system as a target function, and setting upper and lower limits of each operation parameter; s4, obtaining the optimal operation parameters of the hydrogen production efficiency in the current optimization period, and sending the optimal operation parameters to a high-temperature hydrogen production device control system for execution.
Description
Technical Field
The invention relates to the field of control optimization of high-temperature electrolytic hydrogen production, in particular to an online optimization method of a high-temperature electrolytic hydrogen production system.
Background
The high-temperature Solid Oxide (SOEC) electrolytic hydrogen production system runs in a high-temperature environment, and the periphery of the galvanic pile is required to be provided with a pump, a steam generator, an electric heater, a subsequent gas compression system and other power consumption auxiliary machines, so that the hydrogen production efficiency of the system is reduced. Therefore, in the actual operation process, on-line optimization of the system line efficiency is needed under the condition that the hydrogen production rate meets the requirement, the power consumption of the auxiliary engine is reduced, and the hydrogen production efficiency of the system is improved.
Disclosure of Invention
The invention aims to provide an online optimization method of a high-temperature solid oxide electrolysis hydrogen production system, which is used for compensating voltage errors caused by equipment abrasion or galvanic pile attenuation, reducing the energy consumption of auxiliary machines of the system and improving the hydrogen production efficiency.
The invention provides an online optimization method of a high-temperature solid oxide electrolytic hydrogen production system, which is characterized by comprising the following steps of:
a1: initializing a model correction period parameter and an online optimization period parameter, and reading a historical operating parameter and historical voltage data;
a2: using the sum of the voltage value predicted by the mechanism model and the voltage value predicted by the LSSVM model as error correction of electrolytic bath voltage prediction, and performing periodic rolling correction on the sum according to the set online optimization period; the LSSVM model is a data model which is established by adopting a least square support vector machine to predict a voltage error between a voltage value predicted by a mechanism model and historical voltage data;
a3: establishing power models of the galvanic pile and each auxiliary engine component, and calculating hydrogen production efficiency;
a4: setting upper and lower limits of each operation parameter by taking the hydrogen production efficiency as a target function; and (4) obtaining the optimal operation parameters of the hydrogen production efficiency in the current optimization period, and sending the optimal operation parameters to a control system of the high-temperature hydrogen production device for execution.
The online optimization method of the high-temperature solid oxide electrolytic hydrogen production system has the advantages that:
(1) Establishing a data model of an error between a predicted voltage and an actual voltage of a mechanism model by adopting an LSSVM (least squares support vector machine), and realizing periodical rolling correction of the predicted model, thereby compensating a voltage error caused by equipment abrasion or galvanic pile attenuation;
(2) A power model of the high-temperature hydrogen production system is established, and the optimal operating conditions are sought by optimizing the hydrogen production heat value and the power consumption ratio of the whole system as an optimization target, so that the energy consumption of auxiliary machines of the system is effectively reduced, and the hydrogen production efficiency is improved.
Drawings
FIG. 1 is a flow chart of an online optimization method for a high temperature hydrogen production system.
FIG. 2 is a flow chart of a particle swarm optimization method.
FIG. 3 is a schematic diagram of an online optimization system of a high temperature hydrogen production system.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments.
As shown in FIG. 1, the invention provides an online optimization method of a high-temperature solid oxide electrolysis hydrogen production system, which comprises the following steps:
s1: setting model on-line correction period parameter t co And on-line optimization of the period parameter t opt Reading relevant variable historical data needing to participate in modeling from a control system of the high-temperature hydrogen production device, wherein the relevant variable historical data comprise voltage V, current I, temperature T and cathode flow F c Anode flow rate F a Cathode water vapor ratio χ H2O 。
S2: establishing a voltage prediction model of the solid oxide electrolytic cell:
V cell,co (t)=V cell +ΔV(t)
wherein, V cell The total electrolysis voltage is a voltage value predicted by a mechanism model. And the delta V (t) is the error between the voltage value predicted by the mechanism model and the historical voltage data, and a data model of the voltage error is established by adopting a Least Square Support Vector Machine (LSSVM).
Specifically, an electrochemical mechanism model of the high-temperature solid oxide electrolytic cell is constructed according to an electrochemical principle, and the total electrolytic voltage V is cell The sum of the reversible voltage, the activation polarization voltage, the ohmic polarization voltage and the concentration polarization voltage can be used for obtaining the following results:
V cell (t)=E rev (t)+η act (t)+η conc (t)+η ohm (t)
wherein E is rev Is a reversible voltage, η act Activating the overpotential, eta, for the cathode and anode conc Is the over-potential of the cathode and anode concentration eta ohm Ohmic overpotential for the electrolyte.
Reversible voltage E rev :
Wherein E is 0 Represents a standard voltage, R represents a gas equilibrium constant, F represents a Faraday constant,and &>The partial pressures of hydrogen, oxygen and water vapor are shown.
Activation overpotential η act :
Wherein J represents an electrolytic current density, J 0,i Indicates the exchange current density, a represents the anode and c represents the cathode.
Concentration overpotential η conc The method comprises the following steps:
wherein,and &>Is the concentration of hydrogen and steam, respectively, on the three-phase surface>Is the concentration of oxygen at the surface of the three phases.
Ohmic overpotential η ohm :
Wherein, d e The electrolyte layer thickness is indicated.
For the error Δ V (t) between the voltage predicted by the mechanism model and the historical voltage data, the process of LSSVM modeling includes:
the identification model for coefficient regression is set as follows:
y=ω T ψ(x(t i ))+b
wherein y represents the predicted voltage error value output by the identification model, ω is a weight coefficient, ω is T Bias of weight coefficients,. Psi. (. Cndot.) is a kernel function, b represents a bias term, x (t) i ) Inputting parameters for time dimension characteristics, specifically comprising:
wherein, Δ V (T), I (T), T (T), F c (t)、F a (t) is voltage error, current, temperature, cathode inlet flow, anode inlet flow, respectively; p, q, r, m, n and s represent sample data from the current time t to the previous time p, q, r, m, n and s;
according to the model complexity and the mean square error, setting a risk function of an optimized structure as follows:
st.y(t i )=ω T ψ(x(t i ))+b+e(t i ),i=1,2,…N
where γ is the regularization parameter used to determine the trade-off between model complexity and accuracy, e (t) i ) Representing the regression error between the actual and predicted values of the output.
To solve the above optimization problem, a corresponding lagrangian function is constructed:
wherein alpha is i Is a Lagrange multiplier, by pairing ω, b, e (t) i )、α i The derivation is equal to zero, and the condition for the optimal solution of the problem can be obtained:
eliminate omega and e (t) for the above formula i ),α i And can be found by:
wherein y = [ y = 1 ,…,y N ],α=[α 1 ,…,α N ],E=[1,…,1]Ω is a symmetric matrix of N × N kernel functions:
Ω=ψ(x(t)) T ψ(x(t i ))=K(x(t),x(t i ))
wherein K (x (t), x (t) i ) Is a kernel function, using the radial basis function:
where δ is the coefficient of the basis function.
The prediction model for the voltage error calculated based on the selected kernel function is as follows:
s3: calculating the hydrogen production efficiency of the high-temperature solid oxide electrolytic hydrogen production system:
wherein, HHV H2 (t) represents the high calorific value of the system outlet hydrogen; p el (t) represents the power consumption of the electrolysis of the electrolytic cell; p is pump (t) represents power consumption of the water pump; p vapor (t) represents the power consumption of the water vapor generator; p is heater (t) power of the electric heaters of the positive and negative poles; p compressor (t) represents the pressure of air and hydrogen at the rear endThe power consumption of the compressor. The power consumption calculation steps of each part comprise the following steps:
calculating the electrolysis power consumption of the electrolytic cell:
P el (t)=V cell,co (t)·I(t)
and calculating the power consumption of the cathode side water pump:
P P (t)=F c (P P,out -P P,in )/(η p,e ×η p,m )
wherein, P P,in And P P,out Is the inlet pressure and outlet pressure, eta, of the water pump P,e Is the effective power of the pump, eta P,m Is the mechanical power of the pump.
Calculating the power consumption of the electric heaters on the two sides of the cathode and the anode:
wherein,indicating the hot melt of the streams on both sides of the cathode and anode,. DELTA.T i Represents the temperature difference between the inlet and outlet of the flow of the positive and negative electric heaters heater,i The efficiency of the electric heater with anode and cathode electrodes is shown, wherein a represents the anode and c represents the cathode.
Calculating the power consumption of the hydrogen and air compressor:
wherein, P HC,in,i And P HC,out,i Respectively compressor inlet and outlet pressure, K representing the adiabatic index of the gas, V HC,in,i Refers to compressor inlet volumetric flow.
S4: setting the upper limit and the lower limit of each operation parameter by taking the hydrogen production efficiency as a target function:
min-γ H2 (t)
st.Var i,min ≤Var i (t)≤Var i,max ,i={I,T,F c ,F a ,χ H2O }
wherein i represents each optimized operating parameter, χ H2O Representing the water fraction in the cathode stream.
Illustratively, a Particle Swarm Optimization (PSO) algorithm is adopted to obtain the optimal operating parameters of the hydrogen production efficiency in the current Optimization period, and the optimal operating parameters are sent to the high-temperature hydrogen production device control system for execution. As shown in fig. 2, the detailed process of the particle swarm optimization algorithm is as follows:
(1) Initializing a population P, wherein the size of the population is NP; initializing the velocity of each particleAnd random position>
(2) Evaluation of the fitness value Val for each particle fit,i
(3) If the current adaptive value Val of a certain particle fit,i Better than previously recorded solution of Pbest for the particle i Preferably, pbest is updated;
(4) If the current adaptive value Val of a certain particle fit,i Updating the gbest if the global optimal solution gbest is better than the previously recorded global optimal solution gbest;
(5) And if the gbest meets the requirement, ending. If gbest does not meet the requirements, the particles update their speed and new position according to the following formula:
wherein i represents a particle number and d represents a dimension number; ω represents an inertia factor, i.e. the inertia of the current flight speed; c. C 1 ,c 2 As empirical weight, c 1 Also known as individual learning factors, c 2 Also known as social learning factors; r is 1 ,r 2 Represents a discount factor; alpha represents a constraint factor, controlling the weight of the speed.
In the following, a high-temperature solid oxide hydrogen production system is taken as an example, and a detailed calculation and operation flow is given for the invention. Example relates to a system structure as shown in fig. 3, the high-temperature solid oxide electrolysis hydrogen production system comprises a high-temperature hydrogen production device and an online optimization system. The exemplary high-temperature hydrogen production plant comprises a control system, a water pump 1, a steam generator 2, a first electric heater 3, a second heater 4, an air compressor 6, a hydrogen compressor 7, an electrolytic bath 5, a first heat exchanger 8, a second heat exchanger 9 and a steam-water separator 10, wherein the water pump, the steam generator, the first electric heater 3, the second heater 4, the air compressor 6 and the hydrogen compressor 7 are in signal connection with the control system. At the cathode, pure water enters a steam generator 2 through a water pump 1 to generate steam, the steam is mixed with part of backflow hydrogen at the outlet of a rear-end hydrogen compressor 7 and then enters a first heat exchanger 8, the mixed gas enters a first electric heater 3 to be further heated to an electrolysis temperature after exchanging heat with the gas at the cathode outlet of an electrolytic cell, the mixed gas enters an electrolytic cell 5 to be electrolyzed, and the generated hydrogen is dehydrated through a steam-water separator 10 and then is compressed to a storage system through a compressor. At the anode, after being pressurized by a compressor, the air enters a second heat exchanger 4 to exchange heat with the gas at the anode outlet of the electrolytic cell, enters a second electric heater 4 to be further heated to the electrolysis temperature, enters an electrolytic cell 5 to be electrolyzed, and the generated oxygen-enriched air is evacuated after heat is recovered through heat exchange. The control system is used for controlling the work of the devices, collecting required historical data and real-time data from the devices and sending the historical data and the real-time data to the online optimization execution system; the online optimization system executes the method of the invention to obtain the optimal operating parameters of the hydrogen production efficiency in the current optimization period, and sends the optimal operating parameters to the control system of the high-temperature hydrogen production device to drive the devices controlled by the control system to execute correspondingly.
The high temperature hydrogen plant shown in FIG. 3 is a pilot plant with an electrolytic power of about 20kW, and during normal operation, the amount of hydrogen production is determined (about 5-6 Nm) in order to meet the production demand 3 H), so that the electrolytic current is determined within a certain range, and detailed calculation procedures and operation procedures are given below. The embodiment is implemented on the premise of the technical scheme of the invention.
As shown in FIG. 1, the online optimization process of the high-temperature hydrogen production system mainly comprises the following steps:
the method comprises the following steps: initializing the model correction period parameter and the online optimization period parameter, and reading the historical operation parameter and the voltage data.
In combination with specific examples, since the time for equipment wear and battery decay is relatively long, the model correction period is set to 360h, i.e., once in half a month; the online optimization period is set to 24h, namely once a day, and in addition, when large working condition adjustment (such as change of electrolysis power) occurs, the optimization can be immediately carried out, and the working condition is determined again.
Reading historical data of each parameter of a period at the moment k, including the voltage V (t) k-1 ) Current I (t) k-1 ) Temperature T (T) k-1 ) Cathode flow rate F c (t k-1 ) Anode flow rate F a (t k-1 ) Cathode water vapor ratio χ H2O (t k-1 )。
Step two: judging whether the online correction requirement is met: if the correction period is reached, establishing a data model of the error between the mechanism model predicted voltage and the field voltage measured data by adopting a Least Square Support Vector Machine (LSSVM) as the error correction of the electrolytic cell voltage prediction, setting an online optimization period, and performing periodic rolling correction on the electrolytic cell voltage prediction.
Combining with specific calculation example, substituting the historical data read from the control system into the electrolysis bath mechanism model to obtain the predicted voltage V Theory of the invention (t k-1 ) To obtain a voltage prediction error Δ V (t) k-1 )=V(t k-1 )-V Theory of the invention (t k-1 )。
For the present embodiment, Δ V (t) is established in combination with the established LSSVM regression method k ) And variable x [ Delta V (t) ] k-1 ) I(t k-1 ) T(t k-1 ) F c (t k-1 ) F a (t k-1 ) χ H2O (t k-1 )]The data model f (x (t)) of (d), so the final predicted voltage:
V cell,co (t k )=V theory of the invention (t k )+f(x(t k ))
Step three: judging whether the online optimization requirement is met, if the online optimization period is reached, performing the following steps:
step 3-1: reading the current hydrogen production requirement, namely the electrolysis current value. This is because the amount of hydrogen produced is generally constant during actual operation, and is determined by the electrolysis current. Therefore, the electrolytic current is a constant value in the optimization process and does not participate in the optimization.
In combination with the specific calculation example, the current set value of the electrolysis current is read as I from the controller k 。
Step 3-2: and setting an objective optimization function and a constraint function.
With specific examples, the objective function set in this example is as follows:
min-γ H2 (t)
st.Var i,min ≤Var i (t)≤Var i,max ,i={T,F c ,F a ,χ H2O }
the main power consumption components of the embodiment comprise a water pump, a steam generator, a cathode and anode electric heater, a solid oxide electrolytic cell, air and a hydrogen compressor. Therefore, the hydrogen production efficiency calculation formula is as follows:
the constraint function mainly includes the upper and lower limits of each optimization variable, and the optimization variables selected in this example include: temperature T of electrolytic cell and cathode inlet flow F c Anode inlet flow F a Cathode water vapor ratio χ H2O 。
Step 3-3: and (4) obtaining the optimal operation parameters of the hydrogen production efficiency in the current optimization period, and sending the optimal operation parameters to a high-temperature hydrogen production device control system for execution.
In combination with a specific example, the PSO algorithm is used for optimization in this example, and specific optimization steps are shown in fig. 2. Feeding each optimized variable back to the control system for execution, and specifically operating as follows:
the temperature T of the electrolytic cell is used as the set temperature of the positive and negative electric heaters, and the temperature of the inlet of the galvanic pile is controlled to reach an optimized value;
cathode inlet flow F c And cathode water vapor ratio χ H2O Determining the set flow rate (F) of the water pump c *χ H2O );
Anode inlet flow F a The set flow rate after the compressor at the steam air inlet is determined.
Step four: and returning to the step one, and calculating the next week.
In conclusion, the invention provides an online optimization method of a high-temperature solid oxide electrolytic hydrogen production system, which can make up for voltage prediction errors caused by stack attenuation by performing rolling correction on a model based on historical data and improve the model precision; and an online optimization model with the hydrogen production efficiency as a target is established, so that the operation process of the system is better guided.
While the present invention has been described in detail with reference to the preferred embodiments, it should be understood that the above description should not be taken as limiting the invention. Various modifications and alterations to this invention will become apparent to those skilled in the art upon reading the foregoing description. Accordingly, the scope of the invention should be determined from the following claims.
Claims (10)
1. An online optimization method for a high-temperature solid oxide electrolytic hydrogen production system is characterized by comprising the following steps:
a1: initializing a model correction period parameter and an online optimization period parameter, and reading a historical operating parameter and historical voltage data;
a2: using the sum of the voltage value predicted by the mechanism model and the voltage value predicted by the LSSVM model as error correction of electrolytic bath voltage prediction, and performing periodic rolling correction on the sum according to the set online optimization period; the LSSVM model is a data model which is established by adopting a least square support vector machine to predict a voltage error between a voltage value predicted by a mechanism model and historical voltage data;
a3: establishing power models of the galvanic pile and each auxiliary engine component, and calculating hydrogen production efficiency;
a4: setting upper and lower limits of each operation parameter by taking the hydrogen production efficiency as a target function; and (4) obtaining the optimal operation parameters of the hydrogen production efficiency in the current optimization period, and sending the optimal operation parameters to a control system of the high-temperature hydrogen production device for execution.
2. The on-line optimization method for the high-temperature solid oxide electrolysis hydrogen production system according to claim 1, wherein in the step A1, a model correction period t is input co Inputting the on-line optimization period t opt (ii) a The read historical operating parameters and historical voltage data include: voltage V, current I, temperature T, cathode flow F c Anode flow rate F a Cathode water vapor ratio χ H2O 。
3. The online optimization method for the high-temperature solid oxide electrolysis hydrogen production system according to claim 2, wherein in the step A2, the sum of the voltage value predicted by the mechanism model and the voltage value predicted by the LSSVM model is as follows:
V cell,co (t)=V cell +ΔV(t)
wherein, V cell The total electrolytic voltage is obtained through mechanism model prediction; and delta V (t) is a voltage error value predicted by the LSSVM model.
4. The on-line optimization method for the high-temperature solid oxide electrolytic hydrogen production system according to claim 3, characterized in that an electrochemical mechanism model of the high-temperature solid oxide electrolytic cell is constructed according to an electrochemical principle, and the total electrolytic voltage predicted by the mechanism model is obtained by the sum of the reversible voltage, the activation polarization voltage, the ohmic polarization voltage and the concentration polarization voltage:
V cell (t)=E rev (t)+η act (t)+η conc (t)+η ohm (t)
wherein, E rev Is a reversible voltage, eta act Activating the overpotential, eta, for the cathode and anode conc Is the cathode and anode concentration overpotential eta ohm For ohmic contact of electrolytesAn electrical potential.
5. The on-line optimization method for high-temperature solid oxide electrolytic hydrogen production system according to claim 4, wherein reversible voltage E is rev :
Wherein E is 0 Denotes a standard voltage, R denotes a gas equilibrium constant, F denotes a Faraday constant, and &>The partial pressure of hydrogen, oxygen and water vapor is expressed;
activation overpotential η act :
Wherein J represents an electrolytic current density, J 0,i Denotes the exchange current density, a denotes the anode, c denotes the cathode;
concentration overpotential η conc ,:
Wherein,and &>In the three-phase surface, respectively, hydrogen and steam>The concentration of oxygen on the three-phase surface;
ohmic overpotential η ohm :
Wherein, d e The electrolyte layer thickness is indicated.
6. The online optimization method for the high-temperature solid oxide electrolytic hydrogen production system according to claim 4, wherein a data model of a voltage error between a voltage value predicted by a mechanism model and historical voltage data is established by using a least squares support vector machine, and the method further comprises the following steps:
the identification model for coefficient regression is set as follows:
y=ω T ψ(x(t i ))+b
y represents the predicted voltage error value output by the identification model, omega is a weight coefficient, omega T For the bias of the weight coefficients, ψ (-) is the kernel function, b represents the bias term; x (t) i ) Parameters are input for the time dimension features:
wherein, Δ V (T), I (T), T (T), F c (t)、F a (t) is the voltage error, current, temperature, cathode inlet flow, anode inlet flow, respectively; p, q, r, m, n, s represent samples from the current time t to the previous p, q, t, m, n, sData;
according to the complexity of the model and the mean square error, setting a risk function of an optimized structure as follows:
st.y(t i )=ω T ψ(x(t i ))+b+e(t i ),i=1,2,…N
where γ is the regularization parameter used to determine the trade-off between model complexity and accuracy, e (t) i ) Representing the regression error between the actual and predicted values of the output.
7. The on-line optimization method for the high-temperature solid oxide electrolysis hydrogen production system according to claim 6, wherein corresponding Lagrangian functions are constructed:
wherein alpha is i Is a Lagrange multiplier, by pairing ω, b, e (t) i )、α i And (3) obtaining the optimal solution condition of the optimization problem when the derivative is equal to zero:
eliminate omega and e (t) for the above formula i ) α and b are found by the following formula:
wherein y = [ y = 1 ,…,y N ],α=[α 1 ,…,α N ],E=[1,…,1],
Ω is a symmetric matrix of NxN kernel functions:
Ω=ψ(x(t)) T ψ(x(t i ))=K(x(t),x(t i ))
K(x(t),x(t i ) Is a selected kernel function, using the radial basis function:
wherein δ is a coefficient of the basis function; the data model for predicting voltage error is:
8. the on-line optimization method for the high-temperature solid oxide electrolysis hydrogen production system according to claim 6, wherein the hydrogen production efficiency in the step A3 is as follows:
wherein HHV H2 (t) represents the high calorific value of the hydrogen at the outlet of the system; p is el (t) represents the power consumption of electrolysis in the electrolytic cell; p pump (t) represents power consumption of the water pump; p is vapor (t) represents the power consumption of the water vapor generator; p heater (t) represents a positive and negative electric heaterThe power of (d); p compressor (t) represents the power consumption of the air compressor and the back-end hydrogen compressor.
9. The on-line optimization method for the high-temperature solid oxide electrolysis hydrogen production system according to claim 8, wherein in the step A4, an optimization model of the hydrogen production efficiency is established:
min-γ H2 (t)
st.Var i,min ≤Var i (t)≤Var i,max ,i={I,T,F c ,F a ,χ H2O }
wherein, the current I, the temperature T and the cathode flow F c Anode flow rate F a Cathode water vapor ratio χ H2O Is the operating parameter to be optimized.
10. The on-line optimization method for the high-temperature solid oxide electrolytic hydrogen production system according to claim 9, wherein in the step A4, the particle swarm optimization algorithm is adopted for the optimization model of the hydrogen production efficiency to obtain the optimal operating parameters of the hydrogen production efficiency in the current optimization cycle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211468436.4A CN115938493A (en) | 2022-11-22 | 2022-11-22 | Online optimization method of high-temperature solid oxide electrolytic hydrogen production system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211468436.4A CN115938493A (en) | 2022-11-22 | 2022-11-22 | Online optimization method of high-temperature solid oxide electrolytic hydrogen production system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115938493A true CN115938493A (en) | 2023-04-07 |
Family
ID=86655096
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211468436.4A Pending CN115938493A (en) | 2022-11-22 | 2022-11-22 | Online optimization method of high-temperature solid oxide electrolytic hydrogen production system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115938493A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116595868A (en) * | 2023-05-05 | 2023-08-15 | 中国长江三峡集团有限公司 | Method and device for optimizing direct current energy consumption, electronic equipment and storage medium |
CN118223074A (en) * | 2024-05-24 | 2024-06-21 | 山东国创燃料电池技术创新中心有限公司 | Method and device for controlling temperature of electrolytic tank of electrolytic water hydrogen production system |
-
2022
- 2022-11-22 CN CN202211468436.4A patent/CN115938493A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116595868A (en) * | 2023-05-05 | 2023-08-15 | 中国长江三峡集团有限公司 | Method and device for optimizing direct current energy consumption, electronic equipment and storage medium |
CN118223074A (en) * | 2024-05-24 | 2024-06-21 | 山东国创燃料电池技术创新中心有限公司 | Method and device for controlling temperature of electrolytic tank of electrolytic water hydrogen production system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115938493A (en) | Online optimization method of high-temperature solid oxide electrolytic hydrogen production system | |
CN110970642B (en) | Air system control method of fuel cell | |
CN109524693A (en) | Fuel battery air feed system model predictive control method | |
CN107681181B (en) | Performance diagnosis method of fuel cell | |
Chen et al. | Artificial intelligence-aided model predictive control for a grid-tied wind-hydrogen-fuel cell system | |
CN113403645B (en) | Method and device for determining working state of electrolytic cell and controller | |
CN110008435B (en) | Method for calculating optimal power point of fuel cell | |
Cai et al. | Construction and analysis of photovoltaic directly coupled conditions in PEM electrolyzer | |
CN111948562A (en) | Full life cycle monitoring and evaluating system for fuel cell | |
CN112713288B (en) | Control system and control method for fuel cell bubbling humidifier | |
CN109827866B (en) | Soft measurement and closed-loop control method for cathode side water content of proton exchange membrane fuel cell | |
CN114551944A (en) | Method and system for rapidly controlling water content in proton exchange membrane fuel cell | |
CN116384823A (en) | Reliability evaluation method and system for hydrogen electric coupling system | |
EP3927866A1 (en) | System and method for controlling a multi-state electrochemical cell | |
CN111628196A (en) | Air system control method of fuel cell | |
CN213692125U (en) | Fuel cell monitoring system | |
CN113745607A (en) | Fuel cell control method and system | |
Han et al. | Modeling of interfacial resistance effects on the performance and efficiency for electrolyzer energy storage | |
CN115859590A (en) | Method for establishing PEM (proton exchange membrane) electrolytic tank simulation model based on power input characteristics | |
CN114488821B (en) | Method and system for predicting and controlling interval economic model of fuel cell oxygen passing ratio | |
CN114781188A (en) | Dynamic modeling method for solid oxide electrolytic hydrogen production system for fluctuating electric energy consumption | |
JP2007012549A (en) | Fuel cell system | |
CN114976128A (en) | PEMFC fault prediction and health management system and method based on five-dimensional digital twin technology | |
CN107576913B (en) | Hydrogen load operation analysis method in active power distribution network | |
CN113991155A (en) | Method and device for operating a fuel cell system by means of a machine learning method |
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 |