A method and system for predicting electric power consumption of a hydrogen production plant

Abstract

The application discloses a hydrogen production plant electric power prediction method and system, and belongs to the technical field of hydrogen production plant electric power prediction. The technical problem to be solved is that the existing technology estimates auxiliary power of a hydrogen production plant by using an experience curve or a static load ratio, and cannot depict a dynamic coupling relationship between the auxiliary power and renewable power generation power, resulting in large energy consumption estimation deviation and inaccurate hydrogen production cluster operation optimization. The technical solution points are that a hydrogen production plant auxiliary system mechanism model is constructed and total theoretical power is synthesized; a fusion prediction model combining the mechanism model and an ARMA time sequence is established, parameters are fitted through historical data; a hydrogen production unit scheduling plan is input, and accurate plant electric power prediction results are output through the fusion prediction model. The system sets various function modules corresponding to the above method, and efficient prediction is realized.

Description

Technical Field

[0001] This invention belongs to the field of power consumption prediction technology for hydrogen production plants, and specifically relates to a method and system for predicting power consumption in hydrogen production plants. Background Technology

[0002] Hydrogen production from renewable energy power generation has gradually moved from theoretical research to large-scale engineering practice, with green electricity hydrogen production projects both domestically and internationally entering a phase of rapid construction. Hydrogen production plants that utilize a high proportion, or even 100%, of renewable energy sources such as wind and solar power have become the core direction for promoting the green and low-carbon development of the hydrogen energy industry, and are also one of the key paths to achieving clean energy substitution.

[0003] However, wind power, photovoltaic power, and other new energy power generation have inherent intermittent and fluctuating characteristics. The rapid and random changes in their power output directly affect the operating status of the electrolyzer (such as current and voltage stability), and further transmit this to key auxiliary equipment (plant power) systems in the hydrogen production system, such as chillers, circulating water pumps, alkali pumps, and compressors. These auxiliary equipment are the core support for ensuring the continuous and stable operation of the hydrogen production process. Their load characteristics are constrained by multiple process stages, including hydrogen production reaction rate, temperature control, and pressure regulation, exhibiting complex characteristics of multi-timescale response and nonlinear changes.

[0004] When renewable energy power supply fluctuates drastically or hydrogen production system operating conditions change rapidly, traditional auxiliary power estimation methods using fixed proportional coefficients (e.g., 5%–10% of the hydrogen production load) or empirical curves are insufficient to characterize the dynamic coupling relationship between auxiliary power and renewable power generation, failing to reflect the true dynamic load level. This problem directly leads to significant power deviations between the hydrogen production plant's operation scheduling plan and actual operating conditions, affecting not only the safety and stability of the hydrogen production process but also reducing energy utilization efficiency and the accuracy of optimized operation of the hydrogen production cluster. Therefore, conducting research on accurate modeling and prediction of auxiliary power in hydrogen production plants under fluctuating renewable energy power supply conditions, and improving the accuracy of dynamic prediction of auxiliary power, has significant engineering practical implications for ensuring the safe and stable operation of hydrogen production plants, optimizing scheduling strategies, and improving energy utilization efficiency.

[0005] Relevant non-patent literature retrieved:

[0006] (1) Journal title: Journal of Electrical Engineering Technology, document title: Optimization control strategy of off-grid wind power hydrogen production alkaline electrolyzer array considering electrothermal characteristics, volume number: 36, publication date: 2021. This document discloses the use of a constant proportional coefficient to model the plant power load of the hydrogen production plant, setting the plant power load to 5%~10% of the hydrogen production load, without considering the dynamic impact of renewable energy fluctuations on auxiliary power. (2) Journal title: China Electric Power, document title: Day-ahead output plan of hydrogen production system considering electrolyzer start-up and shutdown characteristics, volume number 55, publication date: 2022. This document discloses the same technical solution of estimating the power load of hydrogen production plant using a constant proportional coefficient, but does not involve dynamic modeling of auxiliary equipment system. (3) Journal title: IEEE Transactions on Power Delivery, document title: Hydrogene Electrolyzer Load Modelling for Steady-State Power System Studies, Volume 38, Publication date: 2023. This document discloses the steady-state calculation model of the compressor and water pump in the power supply of a hydrogen production plant, which is only applicable to stable operating conditions. (4) Journal title: Applied Energy, article title: Numerical modeling and analysis of the effect of pressure on the performance of an alkaline water electrolysis system, volume number 287, publication date: 2021. This article discloses a technical solution for detailed simulation of hydrogen production auxiliary equipment process based on chemical software, without considering the dynamic interference of renewable energy fluctuations. (5) Journal title: Applied Energy, article title: Active pressure and flow rate control of alkaline water electrolyzer based on wind power prediction and 100% energy utilization in off-grid wind-hydrogen coupling system, volume number 328, publication date: 2022. This article discloses a technical solution for modeling and optimizing the control system of hydrogen production auxiliary equipment, but does not involve the dynamic prediction of auxiliary equipment power.

[0007] The prior art represented by the aforementioned documents has at least the following unresolved technical problems or defects: 1. It is impossible to characterize the dynamic coupling relationship between auxiliary power and renewable power generation. The relevant evidence is that existing technologies are all based on steady-state or specific operating conditions modeling. However, renewable energy sources such as wind power and photovoltaics have significant intermittency and volatility. Rapid changes in their power generation will be directly transmitted to the auxiliary system, causing the auxiliary load to exhibit multi-timescale response and nonlinear change characteristics. Fixed proportional coefficients or steady-state models cannot adapt to this dynamic change. 2. The auxiliary equipment power estimation has a large deviation. The relevant evidence is that the existing technology uses empirical curves or static load ratios to estimate the auxiliary equipment load, which ignores the impact of changes in parameters such as temperature, pressure and electrolyte concentration in the hydrogen production process on the auxiliary equipment power, resulting in a significant power deviation between the hydrogen production plant operation scheduling plan and the actual operation. 3. The optimization accuracy of hydrogen production cluster operation is insufficient. The relevant evidence is that due to inaccurate prediction of auxiliary machine power, it is impossible to provide accurate data support for the dynamic optimization and scheduling of the hydrogen production system, which affects energy utilization efficiency and system operation stability.

[0008] In view of this, the present invention is hereby proposed. Summary of the Invention

[0009] To address the aforementioned technical problems in the existing technology, this invention provides a method and system for predicting the power consumption of hydrogen production plants. The existing technology uses empirical curves or static load ratios to estimate the auxiliary power of hydrogen production plants, which cannot depict the dynamic coupling relationship between auxiliary power and renewable power generation, resulting in large deviations in energy consumption estimation and inaccurate optimization of hydrogen production cluster operation.

[0010] To achieve the above objectives, the technical solution of the present invention is as follows: Firstly, a method for predicting the power consumption of a hydrogen production plant includes: S1. Construct a mechanistic model of the auxiliary equipment system within the hydrogen production plant; S2. Based on the mechanism model of the auxiliary system, the total theoretical power consumption of the hydrogen synthesis plant; S3. Construct a fusion prediction model that combines the aforementioned mechanism model with ARMA time series data; S4. Input the scheduling plan of the hydrogen production unit, and output the predicted power consumption of the hydrogen production plant through the fusion prediction model.

[0011] Furthermore, in step S1, the mechanism model of the auxiliary system includes a basic process correlation model, which includes a hydrogen production model, a hydrogen production water consumption model, and an electrolyzer thermal model. The hydrogen production model calculates the hydrogen molar flow rate and mass flow rate based on the DC current and Faraday efficiency of the electrolyzer stack. The hydrogen production water consumption model calculates the actual water consumption based on the electrode reaction stoichiometric relationship and the water consumption amplification factor. The electrolytic cell thermal model calculates the heat power generated by the electrolytic cell based on the input electrical power, reaction enthalpy, and natural heat dissipation power.

[0012] Furthermore, the auxiliary system includes a cooling system power model, which includes a heat exchange model, a circulating water tank model, a chiller model, and a chilled water pump model. The heat exchange model calculates the cooling load based on the mass flow rate of cold water, the specific heat at constant pressure, and the temperature difference between the supply and return water. The circulating water tank model includes a liquid level dynamic model based on the inlet and outlet water volume flow rate and a temperature dynamic model based on energy balance. The chiller model calculates chiller power based on load ratio, ambient temperature, and coefficient of performance. The chilled water pump model calculates the chilled water pump's electrical power based on the total head requirement, flow rate, and unit efficiency.

[0013] Furthermore, the auxiliary system also includes an alkali circulation system model, which includes an alkali circulation flow rate model and an alkali circulation pump power model; The alkaline solution circulation flow model is based on proportional control of the electrolysis current and sets minimum and maximum flow constraints. The power model for the alkali circulation pump is based on the total head requirement, alkali circulation flow rate, and unit efficiency to calculate the electric power of the alkali circulation pump.

[0014] Furthermore, the auxiliary system includes a water replenishment system model, which includes a target water replenishment flow rate model and a water replenishment pump power model; The target water replenishment flow rate model is calculated and corrected based on the actual water consumption during hydrogen production and the error in the alkali solution level. The power model for the water supply pump is based on the total head requirement, the target water supply flow rate, and the unit efficiency to calculate the electric power of the water supply pump.

[0015] Furthermore, the auxiliary system includes a hydrogen compressor model, which calculates the compressor power based on the number of compressor stages, isentropic efficiency, hydrogen specific heat ratio, inlet and outlet pressures, and hydrogen mass flow rate.

[0016] Furthermore, the specific process of constructing the fusion prediction model includes: A discrete-time mathematical model for hydrogen production stations is established, and the difference between the actual power consumption of the hydrogen production station and the total theoretical power is defined as the error sequence. Using the scheduling power of the hydrogen production unit as input and the error sequence as output, an ARMA time series equation is established. The model parameters of the ARMA time series equation are fitted using historical operating data.

[0017] Furthermore, the model parameters include AR order, MA order, input lag order, autoregressive coefficient, input coefficient, and moving average coefficient, and the fitting process is based on the principle of minimizing residuals; The specific process of outputting the predicted power consumption of the hydrogen production plant includes: predicting the error residual at future times based on the fusion prediction model, and superimposing the total theoretical power with the error residual to obtain the predicted power consumption value of the hydrogen production plant.

[0018] Furthermore, the process for determining the total theoretical power consumption of the hydrogen synthesis plant is as follows: summing the power of the cooling system, the power of the alkali circulation pump, the power of the water replenishment pump, the power of the hydrogen compressor, and the power of other auxiliary equipment to obtain the total theoretical power.

[0019] Secondly, a power consumption prediction system for a hydrogen production plant includes: The mechanism model building module is used to build mechanism models of auxiliary systems within hydrogen production plants. The total theoretical power synthesis module is used to synthesize the total theoretical power consumption of the hydrogen production plant based on the mechanism model of the auxiliary system. The fusion prediction model construction module is used to construct a fusion prediction model that combines the mechanistic model with the ARMA time series. The power prediction output module is used to input the scheduling plan of the hydrogen production unit and output the power prediction result of the hydrogen production plant through the fusion prediction model.

[0020] The present invention has at least the following beneficial effects: (1) Improve the accuracy of power consumption prediction for hydrogen production plants: By constructing a mechanism model of auxiliary equipment system, the coupling relationship between auxiliary equipment such as chillers and circulating water pumps and hydrogen production process parameters is accurately characterized. Combined with ARMA time series model to correct the mechanism model error, the problem that traditional static proportional estimation or steady-state modeling cannot adapt to the fluctuation of renewable energy is solved, and the energy consumption estimation deviation is significantly reduced. (2) Supporting the optimized scheduling of hydrogen production plants: Taking the scheduling plan of hydrogen production units as input, the output is the high-precision prediction result of plant power consumption, which provides accurate data support for the optimization of hydrogen production cluster operation and power resource allocation, reduces the power deviation between the scheduling plan and the actual operation, and improves the stability and economy of system operation; (3) Adapting to fluctuating power supply scenarios: Specifically designed for the intermittent and fluctuating characteristics of renewable energy sources such as wind power and photovoltaics, the model can dynamically respond to the fluctuations in auxiliary machine load caused by changes in power generation, and meet the actual operation requirements of hydrogen production plants with a high proportion of renewable energy power supply. (4) Convenient and efficient application: The fusion architecture of the mechanism model and the ARMA model is simple. Parameter fitting can be completed based on historical operating data without the need for complex hardware modifications. It is easy to promote and apply in existing hydrogen production plants, taking into account both prediction speed and accuracy. (5) Ensure system safety and stability: Accurate auxiliary power prediction can predict load change trends in advance, avoid auxiliary overload or abnormal operation due to power fluctuations, reduce the risk of hydrogen production process interruption, and ensure the long-term safe and reliable operation of the hydrogen production system. Attached Figure Description

[0021] Figure 1A flowchart of a method for predicting the power consumption of a hydrogen production plant provided in an embodiment of the present invention; Figure 2 This is an architecture diagram of a hydrogen production plant power prediction system provided in an embodiment of the present invention. Detailed Implementation

[0022] The technical solution of the present invention will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention. All other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0023] It should be noted that, unless otherwise specifically stated, the relative arrangement and numerical expressions of the components and steps described in these embodiments should not be construed as limiting the scope of the invention.

[0024] The following description of exemplary embodiments is merely illustrative and is not intended to limit the invention or its application or use in any way. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail herein, but where applicable, such techniques, methods, and apparatus should be considered part of this specification.

[0025] Example 1 See Figure 1 , Figure 1 This is a flowchart of a method for predicting the power consumption of a hydrogen production plant proposed in this invention. Specific steps may include: S1. Constructing a mechanistic model of the auxiliary equipment system within the hydrogen production plant; specifically including: S11, Basic Process Association Model, specifically including: S111, Hydrogen Production Model: Based on the DC current and Faraday efficiency of the electrolyzer stack, the molar flow rate and mass flow rate of hydrogen are calculated. The specific model is as follows:

[0026] in, Hydrogen molar flow rate (unit: mol / s); For Faraday efficiency; It is the Faraday constant (approximately 96485 C / mol); The current in the electrolytic cell stack is DC current (unit: A). The hydrogen mass flow rate model is as follows:

[0027] in, Hydrogen mass flow rate (unit: kg / s); The molar mass of hydrogen (approximately 2 × 10⁻⁶) -(³kg / mol).

[0028] S112, Hydrogen Production Water Consumption Model: Actual water consumption is calculated based on the electrode reaction stoichiometric relationship and water consumption amplification factor. According to the electrode reaction stoichiometric coefficients, theoretically, producing 1 mole of hydrogen by electrolysis consumes approximately 1 mole of pure water. Therefore, the ideal metered water consumption for hydrogen production by electrolysis can be modeled as follows:

[0029] in, The theoretical water consumption (unit: kg / s) is calculated solely based on electrode reaction measurements. The molar mass of water (approximately 18 × 10⁻⁶) - (³kg / mol); However, in actual system operation, there are also: evaporation loss, entrainment, sewage discharge / venting, etc., and the actual water consumption can be modeled as:

[0030] in, This is the water consumption amplification factor, and it has... ≥1 is used to describe additional water loss caused by comprehensive consideration of evaporation, entrainment, sewage discharge, etc.

[0031] S113. Electrolyte Thermal Model: The thermal power generated by the electrolyzer is calculated based on the input electrical power, reaction enthalpy, and natural heat dissipation power. A portion of the input electrical power is converted into the chemical energy (reaction enthalpy) of hydrogen, and the remainder is converted into heat within the electrolyzer. To maintain the reaction temperature, the electrolyzer needs to be cooled. Its thermal dynamics can be modeled as follows:

[0032]

[0033] in, , These are the electrolytic cell voltage (unit: V) and the electrolytic cell power (unit: W), respectively. The current in the electrolytic cell stack is DC current (unit: A). The heat power generated by the electrolytic cell (unit: W); For the electrolytic cell in Molar enthalpy of reaction for hydrolysis / hydrogen generation (unit: J / mol); The natural heat dissipation power of the electrolytic cell (unit: W).

[0034] S12, Cooling system power model, specifically including: S121. Heat transfer model: The cooling load is calculated based on the mass flow rate of the cold water, the specific heat at constant pressure, and the temperature difference between the supply and return water. The heat generated by the electrolytic cell is transferred to the cold water through the heat exchanger, causing a change in the temperature difference between the supply and return water. The cooling load model for the cold water under steady-state or quasi-steady-state conditions is as follows:

[0035] in, The cooling load transferred from the electrolyzer to the chilled water (unit: W); The mass flow rate of cold water is expressed in kg / s. For the isobaric specific heat of cold water, it is usually taken as ; The cold water return temperature (unit: K); Cold water supply temperature (unit: K); Among them, cold water mass flow rate The calculation formula is:

[0036] in, This represents the change in cold water temperature (unit: K). S122, Circulating Water Tank Model: This includes a dynamic model of liquid level based on inlet and outlet water volume flow rates and a dynamic model of temperature based on energy balance. Cold water output from the heat exchanger enters the circulating water tank. Assuming the circulating water tank is a vertical cylindrical shape or an equivalent constant cross-sectional area tank, the water inside the tank is completely mixed and exchanges heat with the ambient temperature. The water level change in the circulating water tank can then be expressed as:

[0037] in, The instantaneous liquid level height of the circulating water tank (unit: ); Cross-sectional area of ​​circulating water tank (unit: ); , These are the volumetric flow rates entering the circulating water tank (unit: m³ / s). The temperature dynamics of the circulating water tank can be modeled using the energy balance formula as follows:

[0038] in, Density of pure water (unit: ); The volume of water in the circulating water tank (unit: ); The mixing temperature of the water inside the circulating water tank (unit: ); The overall heat transfer coefficient between the circulating water tank and the environment (unit: ). ); External ambient temperature of circulating water tank (unit: ).

[0039] S123, Chiller Model: Chiller power is calculated based on load ratio, ambient temperature, and coefficient of performance (COP). The chiller cools water to a set temperature through compression refrigeration or an electric refrigeration heat pump cycle. Its energy efficiency is determined by the COP, which is affected by ambient temperature and the partial load ratio. The formula for calculating the chiller load ratio is:

[0040]

[0041] in, This refers to the load ratio of the chiller. The optimal load distribution ratio for the chiller can be found in the manual; Rated cooling capacity of the chiller (unit: W); This represents the performance coefficient of the chiller under the current operating conditions. The performance coefficient is for reference operating conditions; The linear derating factor of COP is the effect of ambient temperature. For reference ambient temperature (unit: ); This is the second derating factor for COP when the partial load deviates from the optimal point.

[0042] chiller power The expression is:

[0043] in, The cooling load transferred from the electrolyzer to the chilled water (unit: W); This represents the chiller load ratio.

[0044] S124. Chilled Water Pump Model: The electric power of the chilled water pump is calculated based on the total head requirement, flow rate, and unit efficiency. The chilled water pump ensures that the heat exchanger has sufficient flow to remove the heat generated by the electrolytic cell stack. Its model can be represented as follows:

[0045]

[0046] in, Power of the cold water circulation pump (unit: W); The total head requirement for the chilled water circuit, including pipe losses and equipment pressure drop (unit: ); Zero flow head (unit: ); For the efficiency of the chilled water pump-motor unit; Density of pure water (unit: ); Cooling water volumetric flow rate (unit: ); This is the acceleration due to gravity, approximately 9.81 m / s². , It is a constant coefficient.

[0047] S13, Alkali circulation system model, specifically including: S131, Alkali Circulation Flow Model: Based on proportional control of the electrolysis current, with minimum and maximum flow constraints. In industrial alkaline electrolyzers, alkali circulation control needs to remove reaction heat, expel bubbles from the electrode surface, and maintain a uniform electrolyte concentration. To simplify control in industry, an alkali flow rate control that increases proportionally with load / current / hydrogen production is typically used, while providing minimum and maximum flow constraints. In this case, the alkali circulation flow control can be modeled as follows:

[0048] in, Alkali solution circulation volumetric flow rate (unit: ); , , These are the minimum, maximum, and rated alkali flow rates allowed by the system (unit: ); Rated electrolysis current (unit: A).

[0049] S132, Alkali Circulation Pump Power Model: The electrical power of the alkali circulation pump is calculated based on the total head requirement, alkali circulation flow rate, and unit efficiency; the power of the alkali circulation pump can be modeled as follows:

[0050]

[0051] in, Power of the alkaline solution circulation pump (unit: W); Total head requirement for the chilled water circuit (unit: ); The volumetric flow rate of the alkali solution (unit: ); Zero flow head (unit: ); The density of the alkaline solution (unit: ); This is the acceleration due to gravity, approximately 9.81 m / s². For the efficiency of the chilled water pump-motor unit; , It is a constant coefficient.

[0052] S14, Water replenishment system model, specifically including: S141, Target water replenishment flow rate model, based on actual water consumption in hydrogen production and correction calculations for alkali solution level error; the water replenishment system maintains the alkali solution level within the allowable range. Considering proportional control, the water flow rate of the water replenishment pump can be expressed as the electrolysis water consumption and the correction reference amount based on the level error, specifically:

[0053] in, Target water replenishment flow rate (unit: ); For the proportional control parameters of alkali solution level (unit: ); , These are the reference and actual liquid level values.

[0054] S142, Makeup Water Pump Power Model: The power of the makeup water pump is calculated based on the total head requirement, target makeup water flow rate, and unit efficiency. The power of the makeup water pump can be modeled as follows:

[0055]

[0056] in, Power of the water supply pump (unit: W); Total head requirement for the makeup water pump circuit (unit: ); The efficiency of the motor unit for the chilled water pump; , It is a constant coefficient.

[0057] S15. Hydrogen compressor model: The hydrogen compressor model calculates compressor power based on the number of compressor stages, isentropic efficiency, hydrogen specific heat ratio, inlet and outlet pressures, and hydrogen mass flow rate. The hydrogen compressor uses a multi-stage compression method to compress hydrogen into the hydrogen storage tank. The compressor power model can be expressed as:

[0058] in, Compressor power (unit: W); The number of compressor stages; The isentropic efficiency of the compressor; The total efficiency of the compressor's mechanical components and motor; The specific heat ratio of hydrogen; The gas constant of hydrogen is approximately 4124. ; Compressor inlet temperature (unit: K); The compressibility factor of hydrogen gas at the compressor inlet state; , These are the compressor inlet and outlet pressures (unit: Pa); This is the hydrogen mass flow rate (unit: kg / s).

[0059] S2. Based on the mechanism model of the auxiliary equipment system, the total theoretical power consumption of the hydrogen production plant is calculated. Combining the power consumption of each auxiliary equipment system, the total theoretical power consumption of the hydrogen production plant is:

[0060] S3. Construct a fusion prediction model combining the aforementioned mechanism model and ARMA time series; the electricity consumption of hydrogen production plants is affected by the hydrogen production process, exhibiting multi-time cross-coupling characteristics, and the process parameters are difficult to estimate accurately, leading to deviations between the calculated results of the mechanism model and the actual auxiliary power, especially under the influence of fluctuations in renewable power generation, making accurate modeling quite difficult. Therefore, a prediction method based on combining the plant auxiliary power mechanism model with the ARMA process is proposed, with specific steps including: S31. Establish a discrete-time mathematical model for the power consumption of hydrogen production plants:

[0061]

[0062] in, The number of hydrogen production units. Number of compressors; Power of the chiller (unit: W); Power of the chilled water pump (unit: W); Power of the water supply pump (unit: W); Compressor power (unit: W); S32. The difference between the actual power consumption of a hydrogen production plant and the theoretical power consumption is:

[0063] in, These represent the actual power consumption of the hydrogen production plant. This represents the theoretical power consumption of the plant.

[0064] S33, Order , Establish time series as follows:

[0065] in, The dispatch power of the hydrogen production unit; ; Let be the order of AR; Let MA be the order of MA; Input the lag order; For the first Autoregressive coefficient of order; For the first First-order input coefficients; For the first The moving average coefficient.

[0066] S34. Utilizing historical data Based on the following formula Fitted prediction model parameters:

[0067]

[0068]

[0069] in, These are model parameters; The model parameters were estimated. The number of samples; This is the residual obtained from the current calculation; The target sequence being modeled; For exogenous input sequences; , These represent the order of autoregression and the order of the moving average, respectively. This represents the lag order of the exogenous input. After estimating this parameter, a predictive model for electricity consumption in hydrogen production plants based on the ARMA process can be obtained.

[0070] S4. Input the scheduling plan for the hydrogen production unit, and output the predicted power consumption of the hydrogen production plant through the fusion prediction model; at time k, use historical data and the already estimated ARMA model to predict the residual for the next step. and :

[0071]

[0072] In this formula, the meanings of each variable are the same as above, and This is to compensate for the mean of the prediction error.

[0073] At this point, if the scheduling curves of the hydrogen production units in the hydrogen production plant are known, the prediction error of the plant's power consumption can be obtained. At this point, the complete power prediction model for the internal power consumption of the hydrogen production plant can be obtained as follows:

[0074] in, Predicted power consumption for the hydrogen production plant (unit: W).

[0075] Example 2 See Figure 2 , Figure 2 An architecture diagram of a power prediction system for a hydrogen production plant provided by the present invention includes: M1, Mechanism Model Construction Module, is used to construct the mechanism model of the auxiliary system in the hydrogen production plant; M2, Total Theoretical Power Synthesis Module, is used to synthesize the total theoretical power consumption of the hydrogen production plant based on the mechanism model of the auxiliary system; M3, the fusion prediction model construction module, is used to construct a fusion prediction model that combines the mechanistic model with the ARMA time series; M4, the power prediction output module, is used to input the scheduling plan of the hydrogen production unit and output the predicted power consumption of the hydrogen production plant through the fusion prediction model.

[0076] The above specific embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for predicting the power consumption of a hydrogen production plant, characterized in that, include: S1. Construct a mechanistic model of the auxiliary equipment system within the hydrogen production plant; S2. Based on the mechanism model of the auxiliary system, the total theoretical power consumption of the hydrogen synthesis plant; S3. Construct a fusion prediction model that combines the aforementioned mechanism model with ARMA time series data; S4. Input the scheduling plan of the hydrogen production unit, and output the predicted power consumption of the hydrogen production plant through the fusion prediction model.

2. The method for predicting the power consumption of a hydrogen production plant according to claim 1, characterized in that, In step S1, the mechanism model of the auxiliary system includes a basic process correlation model, which includes a hydrogen production model, a hydrogen production water consumption model, and an electrolyzer thermal model. The hydrogen production model calculates the hydrogen molar flow rate and mass flow rate based on the DC current and Faraday efficiency of the electrolyzer stack. The hydrogen production water consumption model calculates the actual water consumption based on the electrode reaction stoichiometric relationship and the water consumption amplification factor. The electrolytic cell thermal model calculates the heat power generated by the electrolytic cell based on the input electrical power, reaction enthalpy, and natural heat dissipation power.

3. The method for predicting the power consumption of a hydrogen production plant according to claim 1 or 2, characterized in that, The auxiliary system includes a cooling system power model, which includes a heat exchange model, a circulating water tank model, a chiller model, and a chilled water pump model. The heat exchange model calculates the cooling load based on the mass flow rate of cold water, the specific heat at constant pressure, and the temperature difference between the supply and return water. The circulating water tank model includes a liquid level dynamic model based on the inlet and outlet water volume flow rate and a temperature dynamic model based on energy balance. The chiller model calculates chiller power based on load ratio, ambient temperature, and coefficient of performance. The chilled water pump model calculates the chilled water pump's electrical power based on the total head requirement, flow rate, and unit efficiency.

4. The method for predicting the power consumption of a hydrogen production plant according to claim 1 or 2, characterized in that, The auxiliary system also includes an alkaline solution circulation system model, which includes an alkaline solution circulation flow rate model and an alkaline solution circulation pump power model. The alkaline solution circulation flow model is based on proportional control of the electrolysis current and sets minimum and maximum flow constraints. The power model for the alkali circulation pump is based on the total head requirement, alkali circulation flow rate, and unit efficiency to calculate the electric power of the alkali circulation pump.

5. The method for predicting the power consumption of a hydrogen production plant according to claim 1 or 2, characterized in that, The auxiliary system includes a water supply system model, which includes a target water supply flow rate model and a water supply pump power model. The target water replenishment flow rate model is calculated and corrected based on the actual water consumption during hydrogen production and the error in the alkali solution level. The power model for the water supply pump is based on the total head requirement, the target water supply flow rate, and the unit efficiency to calculate the electric power of the water supply pump.

6. The method for predicting the power consumption of a hydrogen production plant according to claim 1 or 2, characterized in that, The auxiliary system includes a hydrogen compressor model, which calculates the compressor power based on the number of compressor stages, isentropic efficiency, hydrogen specific heat ratio, inlet and outlet pressures, and hydrogen mass flow rate.

7. The method for predicting the power consumption of a hydrogen production plant according to claim 1, characterized in that, In step S3, the specific process of constructing the fusion prediction model includes: A discrete-time mathematical model for hydrogen production stations is established, and the difference between the actual power consumption of the hydrogen production station and the total theoretical power is defined as the error sequence. Using the scheduling power of the hydrogen production unit as input and the error sequence as output, an ARMA time series equation is established. The model parameters of the ARMA time series equation are fitted using historical operating data.

8. The method for predicting the power consumption of a hydrogen production plant according to claim 7, characterized in that, The model parameters include AR order, MA order, input lag order, autoregressive coefficient, input coefficient, and moving average coefficient. The fitting process is based on the principle of minimizing residuals. The specific process of outputting the predicted power consumption of the hydrogen production plant includes: predicting the error residual at future times based on the fusion prediction model, and superimposing the total theoretical power with the error residual to obtain the predicted power consumption value of the hydrogen production plant.

9. The method for predicting the power consumption of a hydrogen production plant according to claim 1, characterized in that, The process of obtaining the total theoretical power consumption of the synthetic hydrogen production plant is as follows: summing the power of the cooling system, the power of the alkali circulation pump, the power of the water replenishment pump, the power of the hydrogen compressor, and the power of other auxiliary equipment to obtain the total theoretical power.

10. A power consumption prediction system for a hydrogen production plant, characterized in that, include: The mechanism model building module is used to build mechanism models of auxiliary systems within hydrogen production plants. The total theoretical power synthesis module is used to synthesize the total theoretical power consumption of the hydrogen production plant based on the mechanism model of the auxiliary system. The fusion prediction model construction module is used to construct a fusion prediction model that combines the mechanistic model with the ARMA time series. The power prediction output module is used to input the scheduling plan of the hydrogen production unit and output the power prediction result of the hydrogen production plant through the fusion prediction model.

Images