Electrolytic cell power loss analysis method and system based on coupling bonding diagram

Abstract

The invention relates to the technical field of electrolytic cell monitoring, in particular to an electrolytic cell power loss analysis method and system based on a coupling bonding graph. The method comprises the following steps: acquiring a potential variable signal and a flow variable signal in an operation process of an electrolytic cell, and converting into a modulation potential source and a modulation flow source; constructing a coupling bonding graph reference model comprising an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model and a fluid sub-model; configuring an energy storage element and a resistive element in the model by adopting a differential causal relationship; inputting the modulation source to a corresponding node to calculate and obtain an energy residual error; the instantaneous power loss is obtained by multiplying the residual error by the measurement signal, and the fault type is positioned according to the positive and negative characteristics of the power loss and the occurrence position. According to the invention, dependence on initial conditions is eliminated, online real-time monitoring and accurate fault positioning are realized, and a physical basis is provided for health management and life prediction of the electrolytic cell.

Description

Technical Field

[0001] This invention relates to the field of electrolytic cell monitoring technology, specifically to a method and system for analyzing power loss in electrolytic cells based on coupled bond graphs. Background Technology

[0002] In the context of the current global energy transition, using renewable energy to electrolyze water to produce "green hydrogen" is a key path to achieve deep decarbonization. As the core equipment of water electrolysis technology, the electrolyzer faces issues of safe production and efficiency management when coupled with renewable energy. However, the coupling between multiple physical fields of "electricity-heat-fluidity-mechanics-chemistry" within the electrolyzer makes simulating its characteristics extremely difficult.

[0003] For performance evaluation and health monitoring of electrolyzers, existing technologies mainly employ offline testing, simulation modeling, and data-driven approaches. Offline testing requires a controlled experimental environment and cannot achieve online monitoring. Simulation modeling methods analyze by establishing electrochemical or thermal models, but heavily rely on initial state data and are difficult to adapt to the fluctuating operating conditions of renewable energy. Data-driven methods utilize algorithms such as neural networks for prediction, but lack physical interpretability and cannot directly quantify power loss or pinpoint its sources. Summary of the Invention

[0004] This invention provides a method and system for analyzing power loss in electrolyzers based on coupled bond graphs, enabling online real-time monitoring of electrolyzer power loss and accurate location of fault sources.

[0005] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for analyzing power loss in electrolyzers based on coupled bond graphs, comprising: S100: Collects potential variable signals and current variable signals during the operation of the electrolytic cell, and converts the potential variable signals into a modulation potential source and the current variable signals into a modulation current source; S200: Construct a coupling bond graph reference model for an electrolyzer, wherein the sub-models construct a multi-physics coupled energy flow network through power bond connections. The sub-models include an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model, and a fluid sub-model. S300: The energy storage element and resistive element in the coupled bond graph reference model are configured using differential causality; S400: The modulation potential source is input to the 1-node of the coupled bond graph reference model to calculate the flow variable residual, or the modulation current source is input to the 0-node to calculate the potential variable residual; S500: Multiply the residual of the current variable with the potential variable signal, or multiply the residual of the potential variable with the current variable signal, to obtain the instantaneous power loss. Based on the sign and location of the instantaneous power loss in different sub-models, locate the fault type of the electrolytic cell.

[0006] As a preferred embodiment of the present invention, the potential variable signals include electrolytic cell voltage, ambient temperature, electrolytic cell inlet water temperature, and electrolytic cell outlet water temperature; the flow variable signals include operating current, water flow rate, hydrogen production rate, and oxygen production rate.

[0007] As a preferred embodiment of the present invention, the electrothermal model includes: 1- Junction connects reversible voltage elements, ohmic resistance elements, activation overvoltage elements, concentration overpotential elements, and double-layer capacitor elements; The double-layer capacitor element is connected to the I-junction; The electrolytic cell voltage at the 1-junction node is equal to the sum of the reversible voltage, ohmic overpotential, activation overvoltage, and concentration overpotential. The ohmic resistor, activation overvoltage element, and concentration overpotential element are resistive elements, and the irreversible heat flow generated is output to the thermal model through the power bond; The reversible voltage element is connected to the electrochemical sub-model via a conversion element.

[0008] As a preferred embodiment of the present invention, the electro-chemical sub-model includes: 1-Gibberish free energy of the junctional reaction, chemical affinity of the product hydrogen, chemical affinity of the product oxygen, and chemical affinity of the reactant water; The reaction flow rate of the 1-junction node is used as a flow variable to control the rate of gas and liquid reaction; The electrochemical sub-model is connected to the fluid sub-model through a conversion element that uses stoichiometric coefficients and molar mass to convert the reaction flow rate into a mass flow rate.

[0009] As a preferred embodiment of the present invention, the thermal model includes: 0-junction connects thermal capacitance elements, thermal resistance elements, and thermal-fluid coupling resistor elements; The 0-junction node connects the inlet fluid enthalpy, the outlet fluid enthalpy, the reaction heat flux, and the irreversible heat flux from the electrothermal model; The thermal resistance element is connected between the 0-junction and the ambient temperature source; The thermal-fluid coupling resistor element connects the O-junction to the anode-side O-junction and cathode-side O-junction of the fluid sub-model.

[0010] As a preferred embodiment of the present invention, the fluid sub-model includes: Anode-side 0-junction and cathode-side 0-junction; The anode side 0-junction connects the anode inlet mass flow, the anode outlet mass flow, the oxygen generation flow, and the diffusion flow; The cathode side 0-junction connects the cathode inlet mass flow, the cathode outlet mass flow, the hydrogen generation flow, and the diffusion flow; The anode-side 0-junction and the cathode-side 0-junction are connected by a diffusion resistor element.

[0011] As a preferred embodiment of the present invention, the energy storage element and resistive element in the coupled bond graph reference model are configured using differential causality, including: The energy storage element is configured as a differential causal relationship, such that one of the potential variable and the flow variable is the dependent variable and the other is the effect variable, where the effect variable is the derivative of the dependent variable with respect to time. A resistive element maintains the causal direction determined by its constitutive relation; The differential causal configuration makes the computation of the coupled bond graph reference model independent of the system's initial state data.

[0012] As a preferred embodiment of the present invention, step S400 includes: The modulation potential source is input to the 1-node via a strong key, and a flow variable detector is deployed at the 1-node to obtain the flow variable residual. Alternatively, the modulated current source can be input to the 0-node via a strong key, and a potential variable detector can be deployed at the 0-node to obtain the potential variable residual.

[0013] As a preferred embodiment of the present invention, the step of locating the fault type includes: When the instantaneous power loss is negative in the electrothermal model, it is determined to be due to increased ohmic resistance, membrane electrode aging, or increased contact resistance. When the instantaneous power loss is positive in the thermal model, it is determined that the thermal resistance has increased abnormally. When the instantaneous power loss is negative in the thermal model, it is determined to be heat leakage caused by insulation layer damage. When the instantaneous power loss is negative in the fluid sub-model, it is determined to be fluid leakage caused by seal failure or internal gas leakage caused by diaphragm perforation. When the instantaneous power loss is positive in the fluid sub-model, it is determined to be a flow channel blockage. This invention also proposes an electrolyzer power loss analysis system based on coupled bond graphs, comprising: The data acquisition module is used to acquire potential variable signals and current variable signals during the operation of the electrolytic cell, and convert the potential variable signals into a modulation potential source and the current variable signals into a modulation current source. The model building module is used to build a coupling bond graph reference model of the electrolyzer, wherein the sub-models construct a multi-physics coupled energy flow network through power bond connections. The sub-models include an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model, and a fluid sub-model. The causal configuration module is used to configure the energy storage element and resistive element in the coupled bond graph reference model using differential causal relationships; The residual calculation module is used to input the modulation potential source to the 1-node of the coupled bond graph reference model to calculate the flow variable residual, or to input the modulation current source to the 0-node to calculate the potential variable residual; The loss diagnosis module is used to multiply the residual of the current variable with the potential variable signal, or multiply the residual of the potential variable with the current variable signal, to obtain the instantaneous power loss. Based on the sign and location of the instantaneous power loss in different sub-models, the fault type of the electrolytic cell is located.

[0014] The beneficial effects of this invention are: 1. This invention employs differential causal relationships to configure energy storage elements in a coupled bond graph reference model, directly calculating the rate of change of state variables, and using real-time acquired modulation sources as boundary conditions for model calculation. This method eliminates dependence on initial system state data, enabling monitoring to begin at any point during electrolyzer operation, and adapting to frequent start-up and shutdown conditions caused by renewable energy fluctuations.

[0015] 2. This invention constructs a multiphysics field coupled bond graph reference model encompassing four domains: electrical, thermal, fluid, and chemical. It utilizes the energy conservation properties of 0-junctions and 1-junctions to calculate the energy residual and convert it into instantaneous power loss. This method can decouple efficiency degradation into specific physical losses. Based on the positive or negative characteristics of the power loss and its location, it accurately pinpoints whether the fault originates from the electrical, thermal, or fluid domain, and distinguishes specific fault types such as increased ohmic resistance, abnormal thermal resistance, fluid leakage, or flow channel blockage.

[0016] 3. This invention employs bond graph theory to uniformly represent energy exchange across different physical domains as power transfer, making this method widely applicable to various types of electrolytic cell systems, such as proton exchange membranes and alkaline systems. It can be adapted to different types and specifications of electrolytic cells simply by adjusting the model parameters, exhibiting excellent versatility and scalability. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram illustrating the basic principle of bond graphs; Figure 2This is a schematic flowchart of an electrolytic cell power loss analysis method based on coupling bond graphs according to the present invention. Figure 3 This is a reference model diagram of the electrolytic cell coupling bonding diagram of the present invention; Figure 4 This is a schematic diagram of the residual analysis nodes of the present invention; Figure 5 This is a schematic diagram of the structure of an electrolytic cell power loss analysis system based on coupling bond graphs according to the present invention. Detailed Implementation

[0018] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0019] Example 1: This invention employs bond graph theory to perform multiphysics coupling modeling of an electrolyzer. A bond graph is a graphical modeling method for representing energy flow in a physical system. Its core idea is to unify the energy exchange between different physical domains as power transfer. In a bond graph, power consists of two factors, expressed as: ; in Represents potential variables (Effort), such as voltage, pressure, temperature, etc. Represents flow variables, such as current, velocity, gas-liquid flow rate, etc. This represents the instantaneous power flow through the bond. The direction of the actual power flow is determined by the signs of the potential variable e and the flow variable f.

[0020] like Figure 1 Part (a) shows a basic schematic diagram of the bond graph, which includes several key elements. Power bonds are represented by half-arrows, indicating the positive direction of power flow. The basic element includes an energy source S, where... Represents the source of power. The energy source (R) represents the energy source; the energy consumption (R) is a resistive element that consumes energy but does not store it; the energy storage (C) is a capacitive element that stores potential energy; the energy conversion (TF) is a conversion element that can convert the potential or current between different systems. Connection points include 0-junctions (equipotential junctions) and 1-junctions (equicurrent junctions). At a 0-junction, the potential variables of each bond are equal, and the sum of the current variables is zero. At a 1-junction, the current variables of each bond are equal, and the sum of the potential variables is zero.

[0021] At the end of a power bond, there is a short vertical dash, also known as a causal dash. The dash represents the direction of the causal relationship. For example... Figure 1As shown in section (b), there are two types of causal relationships. In integral causality, the potential variable e and the current variable f are the cause and effect, respectively, and the current variable f is the integral of the potential variable e with respect to time. This is applicable when the initial state data of the system is known. In differential causality, the effect is the derivative of the cause with respect to time and does not depend on the initial state data of the system. This invention uses a differential causal relationship configuration bond graph model to eliminate the dependence on initial conditions.

[0022] Based on the above bond graph theory, such as Figure 2 As shown, the present invention provides a method for analyzing the power loss of an electrolyzer based on a coupled bond graph, comprising the following steps: S100: Collects potential variable signals and current variable signals during the operation of the electrolytic cell, and converts the potential variable signals into a modulation potential source and the current variable signals into a modulation current source; Furthermore, the potential variable signals include electrolyzer voltage, ambient temperature, electrolyzer inlet water temperature, and electrolyzer outlet water temperature; the flow variable signals include operating current, water flow rate, hydrogen production rate, and oxygen production rate.

[0023] Specifically, the computer acquires the operating signals of the electrolytic cell in real time, and there are two types of signals: potential variable signals. Including electrolytic cell voltage Ambient temperature Temperature of inlet and outlet water of the electrolytic cell The temperatures of the inlet and outlet water of the electrolytic cell include the inlet water temperature. and outlet water temperature Stream variable signal Including the operating current of the electrolytic cell Water input flow rate Hydrogen / Oxygen Production Rate and .

[0024] Potential variables represent the energy potential in a system, such as voltage, temperature, and pressure. Flow variables represent the rate of energy flow in a system, such as current, velocity, and mass flow rate. These signals are acquired through corresponding sensors: voltage and current are measured by voltage and current sensors, temperature by temperature sensors, and flow rate by flow sensors.

[0025] The sensor will change over time This is converted into a modulation signal MSe, which is then used as the potential variable input at a specific node of the coupled-bond graph reference model. Specifically, the electrolyzer voltage... Converted into a modulation potential source input to the 1-junction node of the electrothermal model, ambient temperature. Converted into a modulation potential source input to the thermal model, inlet water temperature and outlet water temperature Used to calculate the inlet and outlet fluid enthalpies and input them into the thermodynamic model.

[0026] The sensor will change over time This is converted into a modulation signal MSf, which is then set as the current variable input at a specific node of the coupled-bond graph reference model. Specifically, the operating current... Converted to a modulated flow source input to an electrothermal model, the input flow rate of water Converted to the modulation flow source input to the anode-side 0-junction and cathode-side 0-junction of the fluid submodel, the hydrogen production rate and oxygen production rate These are respectively converted into modulated flow source inputs and fed to the corresponding nodes of the fluid sub-model.

[0027] S200: Construct a coupling bond graph reference model for an electrolyzer, wherein the sub-models construct a multi-physics coupled energy flow network through power bond connections. The sub-models include an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model, and a fluid sub-model. Specifically, the computer runs a pre-built reference model of the electrolyzer's coupled bond graph. For example... Figure 3 As shown, the model internally comprises four sub-models: an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model, and a fluid sub-model. These sub-models are connected by power bonds, forming a multiphysics coupled energy flow network.

[0028] Furthermore, the electrothermal model includes: 1- Junction connects reversible voltage elements, ohmic resistance elements, activation overvoltage elements, concentration overpotential elements, and double-layer capacitor elements; The double-layer capacitor element is connected to the I-junction; The electrolytic cell voltage at the 1-junction node is equal to the sum of the reversible voltage, ohmic overpotential, activation overvoltage, and concentration overpotential. The ohmic resistor, activation overvoltage element, and concentration overpotential element are resistive elements, and the irreversible heat flow generated is output to the thermal model through the power bond; The reversible voltage element is connected to the electrochemical sub-model via a conversion element.

[0029] Furthermore, the electrochemical sub-model includes: 1-Gibberish free energy of the junctional reaction, chemical affinity of the product hydrogen, chemical affinity of the product oxygen, and chemical affinity of the reactant water; The reaction flow rate of the 1-junction node is used as a flow variable to control the rate of gas and liquid reaction; The electrochemical sub-model is connected to the fluid sub-model through a conversion element that uses stoichiometric coefficients and molar mass to convert the reaction flow rate into a mass flow rate.

[0030] Furthermore, the thermal model includes: 0-junction connects thermal capacitance elements, thermal resistance elements, and thermal-fluid coupling resistor elements; The 0-junction node connects the inlet fluid enthalpy, the outlet fluid enthalpy, the reaction heat flux, and the irreversible heat flux from the electrothermal model; The thermal resistance element is connected between the 0-junction and the ambient temperature source; The thermal-fluid coupling resistor element connects the O-junction to the anode-side O-junction and cathode-side O-junction of the fluid sub-model.

[0031] Furthermore, the fluid sub-model includes: Anode-side 0-junction and cathode-side 0-junction; The anode side 0-junction connects the anode inlet mass flow, the anode outlet mass flow, the oxygen generation flow, and the diffusion flow; The cathode side 0-junction connects the cathode inlet mass flow, the cathode outlet mass flow, the hydrogen generation flow, and the diffusion flow; The anode-side 0-junction and the cathode-side 0-junction are connected by a diffusion resistor element.

[0032] Specifically, the electrothermal model is built around a 1-junction. The electrolytic cell voltage at the 1-junction node is equal to the sum of the voltage drops of each component, as shown in the following formula: ; in It is a reversible voltage used to participate in the electrochemical reaction that generates hydrogen. It is the Ohmic overpotential; These are the activation overvoltages of the anode and cathode, respectively. It is the concentration overpotential.

[0033] The electrothermal model includes reversible voltage elements, ohmic resistance elements, activation overvoltage elements, concentration overpotential elements, and double-layer capacitor elements, all connected at a 1-junction. (Double-layer capacitor element) This formula is used to simulate the charge accumulation effect at the electrode-electrolyte interface, thereby describing the dynamic response. ; in It is the dielectric constant. It's the area. It has a double-layer thickness.

[0034] The electrothermal model includes resistive elements, including ohmic resistors, activation overvoltages, and concentration overpotentials, corresponding to the ohmic overpotential, activation overvoltage, and concentration overpotential in the above formulas, respectively. The R element describes the process of electrical energy dissipating into heat energy, capable of converting electrical energy across physical domains into irreversible heat flow. As shown in the formula: ; Heat flow is directly delivered to the thermal model via power bonds. TF elements are used to connect electrochemical sub-models. At the electrochemical interface, current is converted into mass flow, and energy is stored in chemical bonds in the form of Gibbs free energy, as shown in the following formula: ; in It is the reaction flow rate. Faraday's constant is the constant for the conversion between charge and substance in an electrochemical reaction. It is Gibbs free energy.

[0035] A flow rate residual analysis node is set at the current input terminal, and the actual flow rate of the electrolytic cell material is input. Used for fault diagnosis.

[0036] In the electro-chemical sub-model, within the chemical domain of the bond graph, the 1-junction represents a specific chemical reaction event, and as an isocurrent junction, the reaction flow rate... The rate of gas generation in the chemical domain is determined by this node, which satisfies the following formula: ; in It is the Gibbs free energy of the reaction. It is the sum of the chemical affinities of the products hydrogen and oxygen. It is the chemical affinity of the reactant water.

[0037] The electrochemical sub-model includes the Gibbs free energy of the 1-junction reaction, the chemical affinity of the product hydrogen, the chemical affinity of the product oxygen, and the chemical affinity of the reactant water. The reaction flux at the 1-junction node is also considered. As a flow variable, it controls the rate of gas-liquid reaction. At the 1-junction, the reaction rate... As a flow variable, it controls the rate of all gas / liquid reactions.

[0038] TF components utilize stoichiometry and molar mass Calculate mass flow rate This connects the fluid domain. The electrochemical sub-model connects to the fluid sub-model via conversion elements, which convert reaction flow rates to mass flow rates using stoichiometric coefficients and molar mass. The coefficients in the TF elements... It is a pre-set parameter, and the formula is as follows: ; The thermal model is constructed in the thermal domain. Heat capacity. This refers to the ability of the electrolytic cell as a whole to absorb heat and raise its temperature, as shown in the following formula: ; in For inflow enthalpy, For outflow enthalpy, For Ohmic heat flow, For the reaction heat flow, It serves as a heat dissipation flow for the environment.

[0039] The thermodynamic model includes 0-junction connected thermal capacity elements, thermal resistance elements, and thermal-fluid coupling resistance elements. The 0-junction node connects the inlet fluid enthalpy, outlet fluid enthalpy, reaction heat flux, and irreversible heat flux from the electrothermal model.

[0040] thermal resistance It is the thermal resistance between the battery body and the surrounding environment, and is the factory data of the electrolytic cell. The thermal resistance element is connected between the 0-junction and the ambient temperature source.

[0041] Thermal-fluid coupling resistor Connecting the fluid sub-model inlet / outlet to the thermal sub-model reflects the process of heat transfer as the fluid flows, as shown in the following formula: ; ; in It is the enthalpy flow rate at the inlet and outlet. It is the mass flow rate. It is the specific heat capacity of the fluid. This refers to the water temperature inside the electrolytic cell. The thermal-fluid coupling resistor element connects the 0-junction to the anode-side 0-junction and the cathode-side 0-junction of the fluid sub-model.

[0042] Irreversible heat flow It comes from the electrothermal model and is the waste heat converted from electrical energy consumed by resistors.

[0043] Reaction heat flow It is modulated by the current, i.e., the reaction rate, as shown in the following formula: ; in It is a thermodynamic constant related to enthalpy change or entropy change.

[0044] Inlet fluid enthalpy include , which is the heat of the cold water that just enters the electrolytic cell, is expressed by the following formula: ; in It is the mass flow rate from the fluid model. It is the specific heat capacity of the fluid. It is the fluid inlet temperature.

[0045] outlet fluid enthalpy include , which is the heat carried away by the generated gas and unreacted water, as shown in the following formula: ; in Mass flow rate from the fluid model, It is the specific heat capacity of the fluid. It is the fluid outlet temperature.

[0046] Set a potential variable residual analysis node at the temperature input end, and input the actual electrolytic cell material. Used for fault diagnosis.

[0047] The fluid sub-model is constructed within the fluid domain of the bond graph and is responsible for managing the mass flow and pressure dynamics within the electrolyzer. At the core of the model are two 0-junctions, or equipotential junctions, representing the fluid cavities on the anode and cathode sides, respectively, and obeying the law of conservation of mass.

[0048] The fluid sub-model includes an anode-side 0-junction and a cathode-side 0-junction. At the anode-side 0-junction, the input mass flow includes the anode supply flow rate. and the oxygen flow rate generated by the electrochemical reaction The output mass flow includes the mixing rate flowing towards the outlet. and the permeate flow rate that diffuses through the diaphragm to the cathode The anode side 0-junction connects the anode inlet mass flow, anode outlet mass flow, oxygen generation flow, and diffusion flow. Anode side pressure. fluid capacity The net mass accumulation is determined by the following formula: ; in It is the anode-side fluid capacitance parameter, representing the volumetric effect; The source signal is modulated by an external pump. The cathode-side 0-junction connects the cathode inlet mass flow, cathode outlet mass flow, hydrogen generation flow, and diffusion flow; the same applies to the cathode.

[0049] diffusion resistance The 0-junction connecting the anode and cathode simulates the diffusion phenomenon of gas or liquid passing through the diaphragm under pressure difference, and the resulting diffusion flow rate... The formula is as follows: ; in It is a diffusion resistance. The anode-side 0-junction and the cathode-side 0-junction are connected through a diffusion resistance element.

[0050] At the outlet, flow resistance element and Describe the flow resistance of the anode and cathode outlet pipes respectively. Taking the anode side as an example, 1-junction represents the pressure drop relationship when the fluid flows out, and the outlet flow rate... The relationship with stress is as follows: ; in This is the external environmental pressure set by the back pressure valve, corresponding to the Se source in the model. A potential variable residual analysis node is set at the pressure node (0-node) of the fluid sub-model, and the actual measured pressure is input for subsequent residual calculations.

[0051] Each sub-model achieves cross-domain energy transfer through power bonds. The electro-thermal sub-model is connected to the electro-chemical sub-model via TF elements, realizing the conversion of electrical energy into chemical energy. The irreversible heat flow generated by the electro-thermal sub-model through R elements is transported to the thermal sub-model via power bonds. The electro-chemical sub-model is connected to the fluid sub-model via TF elements, realizing the conversion of chemical reaction flow rate to mass flow rate. The thermal sub-model is connected to the fluid sub-model via a thermo-fluid coupling resistor, realizing the transfer of heat with the fluid.

[0052] S300: The energy storage element and resistive element in the coupled bond graph reference model are configured using differential causality; Furthermore, configuring the energy storage element and resistive element in the coupled bond graph reference model using differential causality includes: The energy storage element is configured as a differential causal relationship, such that one of the potential variable and the flow variable is the dependent variable and the other is the effect variable, where the effect variable is the derivative of the dependent variable with respect to time. A resistive element maintains the causal direction determined by its constitutive relation; The differential causal configuration makes the computation of the coupled bond graph reference model independent of the system's initial state data.

[0053] Specifically, such as Figure 1 As shown in section (b), the bond graph contains two types of causal relationships: integral causality and differential causality. When the initial state data of the system is known, integral causality is typically used. In this case, the potential variable e and the current variable f are the cause and effect, respectively, and the current variable f is the integral of the potential variable e with respect to time. In differential causality, the effect is the derivative of the cause with respect to time.

[0054] This invention configures the energy storage element in the coupled-bonded graph reference model as having a differential causal relationship, such that one of the potential variable and the current variable is the dependent variable, and the other is the effect variable, where the effect variable is the derivative of the dependent variable with respect to time. The resistive element maintains the causal direction determined by its constitutive relation.

[0055] For the double-layer capacitor element in the electrothermal model After configuring the differential causal relationship, voltage is the dependent variable and current is the effect variable, with current being the derivative of voltage with respect to time. For the heat capacity element in the thermodynamic model... After configuring using differential causality, the rate of temperature change is used as the output, without needing to know the initial temperature value beforehand. This applies to the fluid capacitive elements in the fluid sub-model. and By using differential causality configuration, the pressure change rate is used as the output, without needing to know the initial pressure value in advance.

[0056] For resistive elements, including ohmic resistance elements, activation overvoltage elements, and concentration overpotential elements in the electrothermal model, and thermal resistance elements in the thermal model. Thermo-fluid coupling resistor element and the diffusion resistance element in the fluid sub-model. Outlet flow resistance element and These elements maintain the causal direction determined by their constitutive relations, that is, the causal relationship between potential variables and current variables is determined according to their physical properties.

[0057] By employing differential causality to configure energy storage elements, the calculation of the coupled bond graph reference model does not depend on the system's initial state data. Traditional integral causality requires knowledge of initial states such as the initial voltage of the capacitor, the initial temperature of the heat capacity, and the initial pressure of the fluid cavity. In contrast, differential causality directly calculates the rate of change of state variables. Model calculation can be performed using the modulation potential source and modulation current source input in real-time in step S100 as boundary conditions, without the need for pre-setting initial conditions. This allows the present invention to begin monitoring at any time during the operation of the electrolyzer, adapting to frequent start-up and shutdown conditions caused by renewable energy fluctuations.

[0058] S400: The modulation potential source is input to the 1-node of the coupled bond graph reference model to calculate the flow variable residual, or the modulation current source is input to the 0-node to calculate the potential variable residual; Further, step S400 includes: The modulation potential source is input to the 1-node via a strong key, and a flow variable detector is deployed at the 1-node to obtain the flow variable residual. Alternatively, the modulated current source can be input to the 0-node via a strong key, and a potential variable detector can be deployed at the 0-node to obtain the potential variable residual.

[0059] Specifically, in this embodiment, step S400 involves inputting the modulation potential source to the 1-node of the coupled bond graph reference model to calculate the flow variable residual, or inputting the modulation flow source to the 0-node to calculate the potential variable residual.

[0060] The computer uses the energy conservation properties of the 0-junctions and 1-junctions in the bond graph model as a verification mechanism. Since a coupled bond graph reference model based on the real-time rate of change has been constructed in step S200, step S400 couples this theoretical value with the physical system to calculate the energy flow difference between the theoretical and actual conditions of the system, i.e., the energy residual.

[0061] Depending on the type of residual analysis node, such as Figure 4 As shown, the computer can execute the following two solution methods in parallel or selectively.

[0062] For residual analysis of potential variables, such as Figure 4 As shown in section (a), the measured values ​​obtained from the electrolytic cell physical system MSe is obtained after modulation by the sensor. The modulated potential source MSe is forced into a 1-junction node via a strong bond, where the flow detector Df shows no potential change and can only measure the flow variable value. Measured values. All input 1-junctions are used as potential sources e in the computation of the coupled-bond graph reference model. The flux residuals are obtained at the 1-junction nodes using a flux detector. .

[0063] For residual analysis of flow variables, such as Figure 4 As shown in section (b), the measured values ​​obtained from the electrolytic cell physical system MSf is obtained after modulation by the sensor. The modulated current source MSf is forced into the 0-node via a strong key, where the potential detector De has no current change and can only measure the potential variable value. Measured value All input 0-junctions are used as flow sources f in the computation of the coupled-bond graph reference model. Potential variable residuals are obtained at the 0-junction nodes through a potential variable detector. .

[0064] Specifically, in the electrothermal model, the flow variable residual analysis node set at the current input terminal will be used to analyze the actual measured operating current. As the modulation potential source input, the flux residual is calculated based on the energy conservation relation of the 1-junction. In the thermodynamic model, the potential variable residual analysis node set at the temperature input terminal will use the actual measured electrolytic cell temperature. As the input to the modulation potential source, the residual potential variable is calculated based on the energy conservation relationship of the 0-junction. In the fluid sub-model, a potential variable residual analysis node is set at the pressure node, i.e., the 0-junction. The actual measured pressure is used as the input of the modulation potential source, and the potential variable residual is calculated based on the energy conservation relationship of the 0-junction. .

[0065] In this way, the coupled bond graph reference model calculates the theoretical energy flow based on the real-time input modulation source signal and differential causality. The residuals generated at the 0-junction or 1-junction after the actual system's measured values ​​are input through strong bonds reflect the difference between the theoretical model and the actual system. This difference is the source of power loss.

[0066] S500: Multiply the residual of the current variable with the potential variable signal, or multiply the residual of the potential variable with the current variable signal, to obtain the instantaneous power loss. Based on the sign and location of the instantaneous power loss in different sub-models, locate the fault type of the electrolytic cell.

[0067] Furthermore, the steps for locating the fault type include: When the instantaneous power loss is negative in the electrothermal model, it is determined to be due to increased ohmic resistance, membrane electrode aging, or increased contact resistance. When the instantaneous power loss is positive in the thermal model, it is determined that the thermal resistance has increased abnormally. When the instantaneous power loss is negative in the thermal model, it is determined to be heat leakage caused by insulation layer damage. When the instantaneous power loss is negative in the fluid sub-model, it is determined to be fluid leakage caused by seal failure or internal gas leakage caused by diaphragm perforation. When the instantaneous power loss is positive in the fluid sub-model, it is determined to be a flow channel blockage.

[0068] Specifically, the computer converts the residual value obtained in step S400 into a unified physical unit, namely watt, to obtain instantaneous power loss data. The result is then output to the display terminal. The calculation formula is: ; or ; The first formula corresponds to the case of potential variable residual analysis, which involves measuring the potential variable signal. Compared with the calculated flow variable residuals Multiplication; the second formula corresponds to the case of residual analysis of the flow variables, that is, the measured flow variable signal. The calculated potential variable residual Multiply.

[0069] The system generates a graph of power loss over time on the display interface, which can be observed... The time-varying curve quantifies the impact of power loss on the system.

[0070] The system calculates power losses based on different sub-models. The positive and negative characteristics and their locations are used to physically locate and diagnose the health status of the electrolytic cell.

[0071] When instantaneous power loss is negative in the electrothermal model, it represents additional energy consumption, indicating an increase in the ohmic resistance of the electrolyzer or an abnormal activation overpotential. In this case, the actual input electrical energy is not completely converted into chemical energy, but rather more is converted into waste heat, which is determined to be an increase in ohmic resistance, membrane electrode aging, or increased contact resistance, indicating degradation of the core components, namely the electrode / membrane.

[0072] When instantaneous power loss is positive in the thermal model, it manifests as energy accumulation / power gain. This does not violate energy conservation, but rather indicates that the model-predicted heat loss is greater than the actual measured value. This is judged as an abnormal increase in thermal resistance, meaning that the thermal resistance... Abnormal increases in temperature can be caused by factors such as a malfunctioning cooling fan, obstructed cooling water circulation, or dust accumulation on the heat sink, preventing heat from being effectively dissipated into the environment and causing an abnormal rise in battery temperature.

[0073] When the instantaneous power loss is negative in the thermal model, it is determined to be heat leakage caused by insulation layer damage, indicating the existence of unexpected heat leakage, such as insulation layer damage.

[0074] When instantaneous power loss is negative in the fluid submodel, it manifests as pressure energy loss, indicating either fluid leakage due to seal failure or internal gas leakage due to diaphragm perforation. This signifies the presence of fluid leakage (i.e., seal failure) or internal gas leakage (i.e., diaphragm perforation) in the system. Invalid.

[0075] When the instantaneous power loss is positive in the fluid sub-model, it is determined to be a flow channel blockage, indicating that there is a flow channel blockage in the system, which leads to an abnormal increase in fluid transport resistance.

[0076] Through the aforementioned location and diagnosis of power losses, the system can decouple the general efficiency decline into specific physical sources of loss, providing a physical basis for the health management and lifespan prediction of the electrolyzer. Due to power losses... Corresponding to the loss of system efficiency, successful prediction The evolution of the electrolytic cell can be used to estimate its remaining service life.

[0077] Example 2: Figure 5 As shown, this embodiment provides a power loss analysis system for electrolyzers based on coupled bond graphs, including: The data acquisition module is used to acquire potential variable signals and current variable signals during the operation of the electrolytic cell, and convert the potential variable signals into a modulation potential source and the current variable signals into a modulation current source. The model building module is used to build a coupling bond graph reference model of the electrolyzer, wherein the sub-models construct a multi-physics coupled energy flow network through power bond connections. The sub-models include an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model, and a fluid sub-model. The causal configuration module is used to configure the energy storage element and resistive element in the coupled bond graph reference model using differential causal relationships; The residual calculation module is used to input the modulation potential source to the 1-node of the coupled bond graph reference model to calculate the flow variable residual, or to input the modulation current source to the 0-node to calculate the potential variable residual; The loss diagnosis module is used to multiply the residual of the current variable with the potential variable signal, or multiply the residual of the potential variable with the current variable signal, to obtain the instantaneous power loss. Based on the sign and location of the instantaneous power loss in different sub-models, the fault type of the electrolytic cell is located.

[0078] It should be noted that the electrolytic cell power loss analysis system based on coupled bond graph provided in this embodiment of the invention is used to execute all the process steps of the electrolytic cell power loss analysis method based on coupled bond graph in the above embodiment. The working principle and beneficial effects of the two are one-to-one, so they will not be described again.

[0079] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for analyzing power loss of an electrolytic cell based on a coupled bond graph, characterized by, The method comprises the following steps: S100: collecting potential variable signals and flow variable signals in the electrolytic cell running process, and converting the potential variable signals into a modulated potential source and the flow variable signals into a modulated flow source; S200: constructing a coupling bond graph reference model of the electrolytic cell, wherein a sub-model is used to construct an energy flow network of multi-physical field coupling through a power bond connection relationship, and the sub-model comprises an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model and a fluid sub-model; S300: configuring energy storage elements and resistive elements in the coupling bond graph reference model by using a differential causal relationship; S400: inputting the modulated potential source into a 1-junction node of the coupling bond graph reference model to calculate a flow variable residual error, or inputting the modulated flow source into a 0-junction node to calculate a potential variable residual error; S500: multiplying the flow variable residual error with the potential variable signal or multiplying the potential variable residual error with the flow variable signal to obtain an instantaneous power loss, and positioning a fault type of the electrolytic cell according to the numerical value of the instantaneous power loss and the occurrence position in different sub-models.

2. The method of claim 1, wherein, The potential variable signals comprise an electrolytic cell voltage, an ambient temperature, an electrolytic cell inlet water temperature and an electrolytic cell outlet water temperature. The flow variable signals comprise a working current, a water flow rate, a hydrogen production rate and an oxygen production rate.

3. The method of claim 1, wherein, The electro-thermal sub-model comprises: a 1-junction connected reversible voltage element, ohmic resistance element, activation overvoltage element, concentration overpotential element and double-layer capacitance element; the double-layer capacitance element is connected to the 1-junction; an electrolytic cell voltage at the 1-junction node is equal to the sum of reversible voltage, ohmic overpotential, activation overvoltage and concentration overpotential; the ohmic resistance element, activation overvoltage element and concentration overpotential element are resistive elements, and the generated irreversible heat flow is output to the thermal sub-model through a power bond; the reversible voltage element is connected to the electro-chemical sub-model through a conversion element.

4. The method of claim 1, wherein, The electro-chemical sub-model comprises: a 1-junction connected reaction Gibbs free energy, product hydrogen chemical affinity, product oxygen chemical affinity and reactant water chemical affinity; a reaction flow rate at the 1-junction node is used to control the rate of gas and liquid reactions; the electro-chemical sub-model is connected to the fluid sub-model through a conversion element, and the conversion element uses stoichiometric number and molar mass to convert the reaction flow rate into a mass flow rate.

5. The method of claim 1, wherein, The thermal sub-model comprises: a 0-junction connected heat capacity element, thermal resistance element and thermal-fluid coupling resistance element; the 0-junction node is connected to an inlet fluid enthalpy, an outlet fluid enthalpy, a reaction heat flow and an irreversible heat flow from the electro-thermal sub-model; the thermal resistance element is connected between the 0-junction and an ambient temperature source; the thermal-fluid coupling resistance element connects the 0-junction with an anode side 0-junction and a cathode side 0-junction of the fluid sub-model.

6. The method of claim 1, wherein, The fluid sub-model comprises: an anode side 0-junction and a cathode side 0-junction; the anode side 0-junction is connected to an anode inlet mass flow, an anode outlet mass flow, an oxygen generation flow and a diffusion flow; the cathode side 0-junction is connected to a cathode inlet mass flow, a cathode outlet mass flow, a hydrogen generation flow and a diffusion flow; the anode side 0-junction is connected to the cathode side 0-junction through a diffusion resistance element.

7. The method of claim 1, wherein, Configuring energy storage elements and resistive elements in the coupled bond graph reference model using differential causal relationships comprises: Configuring energy storage elements as differential causal relationships, such that one of the potential variable and the flow variable is the dependent variable and the other is the independent variable, and the independent variable is the derivative of the dependent variable with respect to time; Resistive elements maintain their constitutive relationship determined causal direction; The differential causal relationship configuration makes the calculation of the coupled bond graph reference model independent of system initial state data.

8. The method of claim 1, wherein, The steps of the S400 include: Inputting the modulated potential source into a 1-junction node through a strong bond, and deploying a flow variable detector at the 1-junction node to obtain a flow variable residual error; Or inputting the modulated flow source into a 0-junction node through a strong bond, and deploying a potential variable detector at the 0-junction node to obtain a potential variable residual error.

9. The method of claim 1, wherein, The steps of locating the fault type include: When the instantaneous power loss is negative in the electro-thermal sub-model, it is determined to be an increase in ohmic resistance, membrane electrode aging, or an increase in contact resistance; When the instantaneous power loss is positive in the thermal sub-model, it is determined to be an abnormal increase in thermal resistance; When the instantaneous power loss is negative in the thermal sub-model, it is determined to be thermal leakage caused by damage to the thermal insulation layer; When the instantaneous power loss is negative in the fluid sub-model, it is determined to be fluid leakage caused by seal failure or internal gas leakage caused by diaphragm perforation; When the instantaneous power loss is positive in the fluid sub-model, it is determined to be flow channel blockage.

10. A power loss analysis system for electrolytic cells based on coupled bond graphs, characterized in that, It includes: A data acquisition module for acquiring potential variable signals and flow variable signals during the operation of an electrolytic cell, and converting the potential variable signals into a modulated potential source and the flow variable signals into a modulated flow source; A model construction module for constructing a coupled bond graph reference model of the electrolytic cell, wherein sub-models are connected by power bond connection relationships to construct an energy flow network coupled by multiple physical fields, and the sub-models include an electro-thermal sub-model, an electro-chemical sub-model, a thermal sub-model, and a fluid sub-model; A causal configuration module for configuring energy storage elements and resistive elements in the coupled bond graph reference model using differential causal relationships; A residual error calculation module for inputting the modulated potential source into a 1-junction node of the coupled bond graph reference model to calculate a flow variable residual error, or inputting the modulated flow source into a 0-junction node to calculate a potential variable residual error; A loss diagnosis module for multiplying the flow variable residual error by the potential variable signal, or multiplying the potential variable residual error by the flow variable signal to obtain an instantaneous power loss, and locating the fault type of the electrolytic cell according to the numerical value and occurrence position of the instantaneous power loss in different sub-models.

Images