A device and method for predicting the remaining life of a hybrid fuel cell based on a nonlinear observer.
By estimating fuel cell degradation factors in real time using a nonlinear observer-based method, the problem of accurate prediction of fuel cell remaining life is solved, enabling safe and stable operation and economical application of fuel cells, which is applicable to marine hybrid power systems.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HARBIN ENG UNIV
- Filing Date
- 2026-05-29
- Publication Date
- 2026-07-03
AI Technical Summary
Existing technologies cannot predict the remaining lifespan of fuel cells in real time and accurately, making it difficult for energy management systems to allocate power reasonably. This may lead to overuse of batteries and navigation safety risks, and methods that rely on massive amounts of historical data have weak generalization ability.
A nonlinear observer-based approach is adopted to estimate the fuel cell degradation factor and calculate the remaining lifetime by measuring parameters such as fuel cell output current and bus voltage. This data is then used as a control constraint to adjust the fuel cell output current and power. Combined with the master-slave control strategy of the energy storage system, real-time prediction and protection are achieved.
It enables real-time and accurate prediction of the remaining life of fuel cells, reduces the risk of overuse of batteries, ensures the safety and stability of ship operation, and has good economic efficiency and universality, without relying on historical data for training.
Smart Images
Figure CN122330751A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a marine hybrid power control device and method, specifically a power fuel cell control device and method. Background Technology
[0002] With increasingly stringent environmental protection requirements in the global shipping industry and the rapid development of clean energy technologies, the electrification and clean energy conversion of ship propulsion systems have become the mainstream trend in the industry. Among these, fuel cells, with their significant advantages such as high energy conversion efficiency, low operating noise, and zero emissions, are finding increasingly widespread application in ship propulsion, especially in hybrid-powered vessels. Despite their promising prospects, the cost and lifespan of fuel cells remain key obstacles to their widespread adoption. In actual operation, factors such as load fluctuations inevitably cause physicochemical degradation in key components of fuel cells, such as catalysts and proton exchange membranes, leading to a gradual decline in performance.
[0003] In marine hybrid power systems, fuel cells serve as the primary or auxiliary power source, and accurately determining their health status and remaining lifespan is crucial for energy management and navigation safety. If the remaining lifespan of the fuel cells cannot be obtained in real time and accurately, the energy management system will struggle to formulate reasonable power allocation strategies, potentially leading to overuse and accelerated aging of the batteries, or even malfunctions during navigation, severely impacting the ship's operational safety and reliability.
[0004] However, in practical applications, the remaining lifespan of fuel cells cannot be directly measured. Existing technologies typically require indirect estimation using externally measurable parameters such as voltage and current. Current fuel cell lifespan prediction methods are mainly based on data-driven approaches. While widely used, data-driven methods rely heavily on massive amounts of historical experimental data for offline training, placing extremely high demands on the quality and scale of the dataset, and resulting in weak model generalization ability. Summary of the Invention
[0005] The purpose of this invention is to provide a hybrid fuel cell remaining lifetime prediction device and method based on a nonlinear observer, which is highly versatile and has the ability to predict remaining lifetime in real time, without relying on training based on historical test time.
[0006] The objective of this invention is achieved as follows: This invention discloses a hybrid fuel cell remaining life prediction device based on a nonlinear observer, characterized by comprising a fuel cell, an energy storage system, a bidirectional DC / DC converter, a boost DC / DC converter, a DC bus, a load, a first current sensor, a second current sensor, a first voltage sensor, a second voltage sensor, a third voltage sensor, an energy management controller, and a fuel cell life observer. The fuel cell is connected to the DC bus via the boost DC / DC converter, the energy storage device is connected to the DC bus via the bidirectional DC / DC converter, and the load is connected to the DC bus. The first current sensor measures the fuel cell output current, the second current sensor measures the energy storage system output current, the first voltage sensor measures the fuel cell voltage, the second voltage sensor measures the energy storage system voltage, and the third voltage sensor measures the DC bus voltage. The energy management controller sends the target output power of the fuel cell and the energy storage system. The fuel cell life observer estimates the remaining life of the fuel cell and calculates the maximum output current and maximum output power of the fuel cell, which are then sent as control constraints to the boost DC / DC converter and the energy management controller, respectively.
[0007] This invention discloses a method for predicting the remaining life of a hybrid fuel cell based on a nonlinear observer, characterized by the following steps: (1) Establish a marine hybrid power system as described in claim 1; (2) Collect the fuel cell output current and bus voltage; (3) Design a nonlinear observer to estimate the degradation factor of the fuel cell; (4) Calculate the remaining life based on the relationship between the fuel cell degradation factor and the remaining life; (5) Calculate the maximum output current of the fuel cell based on the fuel cell degradation factor and pass it to the fuel cell controller as a control constraint.
[0008] The remaining lifetime prediction method for a hybrid fuel cell based on a nonlinear observer according to the present invention may further include: 1. In step (1), the fuel cell and its DC / DC converter adopt a power closed-loop control method: Fuel cell DC / DC power closed loop: in, This is the reference output power for the fuel cell; This is the reference output current for the fuel cell; This refers to the actual output power of the fuel cell; This refers to the terminal voltage of the fuel cell. This is the proportionality coefficient; The integral coefficient; For the capacitor of the fuel cell DC converter; Duty cycle of the fuel cell DC-DC converter; This is a degradation factor for fuel cells; Energy storage system DC / DC voltage-current dual closed loop: in, This is the reference value for the DC bus voltage; This serves as the reference output current for the energy storage system. This represents the actual output current of the energy storage system. This is the voltage loop proportionality coefficient; The voltage loop integral coefficient; This is the proportionality coefficient of the current loop; The integral coefficient of the current loop; This refers to the duty cycle of the DC-DC converter in the energy storage system.
[0009] 2. Step (3) uses a nonlinear model to characterize the fuel cell voltage: in, This refers to the number of fuel cell wafers connected in series. This is the reversible voltage of the fuel cell; This refers to the internal resistance of the fuel cell. This refers to the current density of the fuel cell. To activate the exchange current associated with the loss; The limiting current is related to concentration loss; , It is a constant; For fuel cell temperature; This is a degradation factor for fuel cells; The nonlinear observer for the degradation factor of fuel cells is designed as follows: in, The state equation for the fuel cell current; , , , As an intermediate variable; This is the adjustment coefficient; These are the observed values for the degradation factor.
[0010] 3. Step (4): Based on the conversion relationship between fuel cell degradation factor and fuel cell remaining life, the remaining life is predicted as follows: in, For the remaining lifespan of the fuel cell; This represents the longest lifespan of a fuel cell. This represents the maximum degradation factor for fuel cell lifespan. This represents the internal resistance of the fuel cell.
[0011] 4. Step (5) Calculate the maximum output current of the fuel cell after performance degradation: This maximum current is used as a physical constraint upper limit and passed to the fuel cell and its converter.
[0012] The advantages of this invention are: 1. The proposed method for predicting the remaining life of fuel cells in hybrid power systems based on a nonlinear observer employs an aging factor to assess the degree of fuel cell degradation. This invention can detect the degree of fuel cell degradation in real time, providing life information for fuel cell repair or replacement, and promoting the application of fuel cells in marine power systems.
[0013] 2. The master-slave control method employed in this invention uses the energy storage system as the main power source to ensure the stability of the DC bus, while the fuel cell employs follower control, which ensures the smooth operation of the fuel cell and thus reduces fuel cell lifespan degradation. Furthermore, the energy storage system, as the main power source, can provide electrical energy for the fuel cell's startup, thereby solving the problem of the fuel cell's inability to start automatically.
[0014] 3. This invention can predict fuel cell degradation in real time and calculate the maximum operating current and maximum output power of the fuel cell. It provides physical constraints for the control system, ensuring the safety of the system during ship operation.
[0015] 4. This invention does not require additional sensors, only essential current and voltage sensors, which are typically necessary for the system. Therefore, this invention is highly economical.
[0016] 5. This invention has the capability to run in real time. The lifespan predictor has low computational requirements and low controller computing power requirements, enabling it to run in mass-produced controllers. Furthermore, the design of the fuel cell lifespan predictor in this invention does not rely on training with historical data, thus possessing strong versatility. Attached Figure Description
[0017] Figure 1 This is a flowchart of the remaining lifetime prediction method of the present invention; Figure 2This is a schematic diagram of the remaining life prediction device of the present invention; Figure 3 Figure showing the observation results of fuel cell degradation factors; Figure 4 This is a graph showing the predicted remaining lifespan of the fuel cell. Detailed Implementation
[0018] The invention will now be described in more detail with reference to the accompanying drawings: Combination Figure 1-4 The fuel cell remaining life prediction device of the present invention includes a fuel cell 1, an energy storage system 2, a bidirectional DC / DC converter 3, a boost DC / DC converter 4, a DC bus 5, a load 6, a first current sensor 7, a second current sensor 8, a first voltage sensor 9, a second voltage sensor 10, a third voltage sensor 11, an energy management controller 12, and a fuel cell life observer 13. The fuel cell 1 is connected to the DC bus 5 via the boost DC / DC converter 4, and the energy storage system 2 is connected to the DC bus 5 via the bidirectional DC / DC converter 3. The load 6 represents the load of various electrical equipment in the ship. The first current sensor 7 measures the output current of the fuel cell, the second current sensor 8 measures the output current of the energy storage system, the first voltage sensor 9 measures the voltage of the fuel cell, the second voltage sensor 10 measures the voltage of the energy storage system, and the third voltage sensor measures the DC bus voltage. The energy management controller 12 is responsible for sending the target output power of the fuel cell 1 and the energy storage system 2. The fuel cell life observer 13 is responsible for estimating the remaining life of the fuel cell and calculating the maximum output current and maximum output power of the fuel cell, which are then sent as control constraints to the boost DC / DC converter 4 and the energy management controller 12, respectively.
[0019] The fuel cell remaining life prediction method of the present invention includes the following steps: S1: A marine hybrid power system consisting of a fuel cell and a battery. The fuel cell and its converter adopt power closed-loop control, while the battery and its converter adopt voltage-current dual closed-loop control. S2: Collects fuel cell output current and bus voltage; S3: Design a nonlinear observer to estimate the degradation factor of the fuel cell; S4: Calculate the remaining life based on the relationship between the fuel cell degradation factor and the remaining life; S5: Calculate the maximum output current of the fuel cell based on the fuel cell degradation factor and pass it to the fuel cell controller as a control constraint.
[0020] In step S1, the DC grid formed by the fuel cell and energy storage system in the hybrid power system adopts a fully active architecture. The fuel cell and energy storage system are connected to the DC bus via DC / DC converters, and the load is connected to the DC bus. The hybrid power system adopts master-slave control, with the energy storage system stabilizing the bus voltage and the fuel cell controlling the grid. The fuel cell and its DC / DC converter employ a power closed-loop control method, including: Fuel cell DC / DC power closed loop: (1) (2) (3) in, This is the reference output power for the fuel cell; This is the reference output current for the fuel cell; This refers to the actual output power of the fuel cell; This refers to the terminal voltage of the fuel cell. This is the proportionality coefficient; The integral coefficient; For the capacitor of the fuel cell DC converter; Duty cycle of the fuel cell DC-DC converter; This is a degradation factor for fuel cells.
[0021] Energy storage system DC / DC voltage-current dual closed loop: (4) (5) in, This is the reference value for the DC bus voltage; This serves as the reference output current for the energy storage system. This represents the actual output current of the energy storage system. This is the voltage loop proportionality coefficient; The voltage loop integral coefficient; This is the proportionality coefficient of the current loop; The integral coefficient of the current loop; This refers to the duty cycle of the DC-DC converter in the energy storage system.
[0022] In step S2, the fuel cell current and bus voltage are obtained through current sensors and voltage sensors.
[0023] In step S3, based on the fuel cell polarization curve, a nonlinear model is fitted to characterize the fuel cell voltage: (6) in, This refers to the number of fuel cell wafers connected in series. This is the reversible voltage of the fuel cell; This refers to the internal resistance of the fuel cell. To activate the exchange current associated with the loss; The limiting current is related to concentration loss; , It is a constant; This refers to the temperature of the fuel cell.
[0024] Design a fuel cell lifetime predictor. Based on the nonlinear voltage equation of the fuel cell, design a nonlinear observer for the fuel cell degradation factor as follows: (7) (8) (9) (10) (11) in, , , , As an intermediate variable; This is the adjustment coefficient; These are the observed values for the degradation factor.
[0025] In step S4, the conversion relationship between the fuel cell degradation factor and the remaining life of the fuel cell is as follows: (12) in, For the remaining lifespan of the fuel cell; This represents the longest lifespan of a fuel cell. This represents the maximum degradation factor for fuel cell lifespan. This represents the internal resistance of the fuel cell.
[0026] In step S5, after performance degradation, the maximum output current of the fuel cell becomes: (13) This maximum current is used as a physical constraint upper limit and passed to the fuel cell and its converter to protect the safety of the fuel cell during ship operation.
Claims
1. A nonlinear observer-based hybrid fuel cell remaining life prediction device, characterized by: The system includes a fuel cell, an energy storage system, a bidirectional DC / DC converter, a boost DC / DC converter, a DC bus, a load, a first current sensor, a second current sensor, a first voltage sensor, a second voltage sensor, a third voltage sensor, an energy management controller, and a fuel cell lifetime observer. The fuel cell is connected to the DC bus via the boost DC / DC converter, and the energy storage device is connected to the DC bus via the bidirectional DC / DC converter. The DC bus is connected to the load. The first current sensor measures the fuel cell output current, the second current sensor measures the energy storage system output current, the first voltage sensor measures the fuel cell voltage, the second voltage sensor measures the energy storage system voltage, and the third voltage sensor measures the DC bus voltage. The energy management controller sends the target output power of the fuel cell and the energy storage system. The fuel cell lifetime observer estimates the remaining life of the fuel cell and calculates the maximum output current and maximum output power of the fuel cell, which are then sent as control constraints to the boost DC / DC converter and the energy management controller, respectively.
2. A method for predicting the remaining life of a hybrid fuel cell based on a nonlinear observer, characterized by: Includes the following steps: (1) Establish a marine hybrid power system as described in claim 1; (2) Collect the fuel cell output current and bus voltage; (3) Design a nonlinear observer to estimate the degradation factor of the fuel cell; (4) Calculate the remaining life based on the relationship between the fuel cell degradation factor and the remaining life; (5) Calculate the maximum output current of the fuel cell based on the fuel cell degradation factor and pass it to the fuel cell controller as a control constraint.
3. The method of claim 2, wherein the method is characterized by: In step (1), the fuel cell and its DC / DC converter employ a power closed-loop control method: Fuel cell DC / DC power closed loop: in, This is the reference output power for the fuel cell; This is the reference output current for the fuel cell; This refers to the actual output power of the fuel cell; This refers to the terminal voltage of the fuel cell. This is the proportionality coefficient; The integral coefficient; For the capacitor of the fuel cell DC converter; Duty cycle of the fuel cell DC-DC converter; This is a degradation factor for fuel cells; Energy storage system DC / DC voltage-current dual closed loop: in, This is the reference value for the DC bus voltage; This serves as the reference output current for the energy storage system. This represents the actual output current of the energy storage system. This is the voltage loop proportionality coefficient; The voltage loop integral coefficient; This is the proportionality coefficient of the current loop; The integral coefficient of the current loop; This refers to the duty cycle of the DC-DC converter in the energy storage system.
4. The method for predicting the remaining life of a hybrid fuel cell based on a nonlinear observer according to claim 2, characterized in that: Step (3) uses a nonlinear model to characterize the fuel cell voltage: in, This refers to the number of fuel cell wafers connected in series. This is the reversible voltage of the fuel cell; This refers to the internal resistance of the fuel cell. This refers to the current density of the fuel cell. To activate the exchange current associated with the loss; The limiting current is related to concentration loss; , It is a constant; For fuel cell temperature; This is a degradation factor for fuel cells; The nonlinear observer for the degradation factor of fuel cells is designed as follows: in, The state equation for the fuel cell current; , , , As an intermediate variable; This is the adjustment coefficient; These are the observed values for the degradation factor.
5. The method for predicting the remaining life of a hybrid fuel cell based on a nonlinear observer according to claim 2, characterized in that: Step (4) Based on the conversion relationship between fuel cell degradation factor and fuel cell remaining life, the remaining life is predicted as follows: in, For the remaining lifespan of the fuel cell; This represents the longest lifespan of a fuel cell. This represents the maximum degradation factor for fuel cell lifespan. This represents the internal resistance of the fuel cell.
6. The method for predicting the remaining life of a hybrid fuel cell based on a nonlinear observer according to claim 2, characterized in that: Step (5) Calculate the maximum output current of the fuel cell after performance degradation: This maximum current is used as a physical constraint upper limit and passed to the fuel cell and its converter.