KR20230033470A - Apparatus and method for controlling fuel cell - Google Patents

Abstract

A fuel cell control device according to an embodiment of the present invention includes: a hydrogen supply valve for controlling the supply of hydrogen to a fuel cell stack; a hydrogen pressure sensor that detects the pressure of hydrogen supplied to the fuel cell stack; and a control unit that controls the opening amount of the hydrogen supply valve based on the hydrogen pressure detected by the hydrogen pressure sensor, and controls the opening amount of the hydrogen supply valve so that hydrogen of a predetermined flow rate can be supplied after learning the opening amount of the hydrogen supply valve matched to a predetermined hydrogen pressure.

Description

Fuel cell control device and method {APPARATUS AND METHOD FOR CONTROLLING FUEL CELL}

The present invention relates to a fuel cell control device and method.

In general, a fuel cell system includes a fuel cell stack that generates electrical energy, a fuel supply system that supplies hydrogen to the fuel cell stack, an air supply system that supplies oxygen, an oxidant required for an electrochemical reaction, to the fuel cell stack, and a fuel cell stack. It consists of a heat and water management system that controls the operating temperature of the

High-pressure compressed hydrogen of about 700 bar is stored in the hydrogen tank provided in the fuel supply system, and the stored compressed hydrogen is released to the high-pressure line according to the on/off of the high-pressure regulator installed at the inlet of the hydrogen tank, Passing through the hydrogen cut-off valve and the hydrogen supply valve, the pressure is reduced and supplied to the fuel cell stack.

The hydrogen supply valve plays a role in reducing the hydrogen fuel of 17 bar supplied from the hydrogen shut-off valve to 1 to 2 bar and supplying it to the fuel cell stack.

The hydrogen supply valve is a solenoid valve in which the plunger moves according to the magnetic field, and the valve opening amount is adjusted by controlling the duty ratio.

When the plunger height is adjusted according to the duty ratio in the hydrogen supply valve, the flow rate changes and the hydrogen electrode pressure is formed. control the rain

On the other hand, in an ideal situation, the hydrogen electrode pressure is constant according to the opening amount of the hydrogen supply valve, but the opening amount of the hydrogen supply valve may vary depending on the temperature or environment even at the same hydrogen electrode pressure.

That is, the hydrogen supply valve draws a rattle-shaped hysteresis graph due to external factors such as heat generation, and the control model of the hydrogen supply valve is not constant even in the steady-state of the fuel cell stack. It can cause problems with output and durability.

One object of the present invention is to predict a hysteresis function by applying a learning algorithm to control a hydrogen supply valve, so that a constant output of a fuel cell system can be expected through stable hydrogen supply, and durability and fuel efficiency of the fuel cell system can be improved. An object of the present invention is to provide a fuel cell control device and method.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

A fuel cell control device according to an embodiment of the present invention includes a hydrogen supply valve for controlling the supply of hydrogen to a fuel cell stack, a hydrogen pressure sensor for sensing the pressure of hydrogen supplied to the fuel cell stack, and detection by the hydrogen pressure sensor. Controls the opening amount of the hydrogen supply valve based on the hydrogen pressure obtained, and controls the opening amount of the hydrogen supply valve to supply hydrogen at a predetermined flow rate after learning the opening amount of the hydrogen supply valve matching the predetermined hydrogen pressure A control unit may be included.

In one embodiment, the hydrogen supply valve may be a solenoid valve whose opening amount is controlled by adjusting the duty ratio.

In an embodiment, the control unit may control the opening amount of the hydrogen supply valve based on a hysteresis graph in which a predetermined duty ratio is mapped to a predetermined hydrogen flow rate.

In an embodiment, the control unit may generate a learned hysteresis graph after a predetermined time period after learning the duty ratio mapped to a predetermined hydrogen flow rate using a support vector machine model.

In one embodiment, the control unit may control the opening amount of the hydrogen supply valve based on the duty ratio mapped to the learned hysteresis graph when a hydrogen flow rate within a predetermined range is required.

In one embodiment, the control unit may control the opening amount of the hydrogen supply valve based on a duty ratio mapped to a hysteresis graph prior to learning when a hydrogen flow rate out of the predetermined range is required.

A fuel cell control method according to another embodiment of the present invention includes controlling the opening amount of a hydrogen supply valve based on the hydrogen pressure detected by a hydrogen pressure sensor, and determining the opening amount of the hydrogen supply valve matched to a predetermined hydrogen pressure. The method may include learning and controlling an opening amount of the hydrogen supply valve to supply hydrogen at a predetermined flow rate based on the learning result.

In one embodiment, the step of controlling the opening amount of the hydrogen supply valve based on the hydrogen pressure sensed by the hydrogen pressure sensor controls the opening amount of the solenoid valve controlling the supply of hydrogen to the fuel cell stack by adjusting the duty ratio. and sensing the pressure of hydrogen supplied to the fuel cell stack by a hydrogen pressure sensor.

In one embodiment, the step of controlling the opening amount of the hydrogen supply valve based on the hydrogen pressure detected by the hydrogen pressure sensor is based on a hysteresis graph in which a predetermined duty ratio is mapped to a predetermined hydrogen flow rate of the hydrogen supply valve. A step of controlling the amount of opening may be included.

In one embodiment, the step of learning the opening degree of the hydrogen supply valve that matches the predetermined hydrogen pressure is learned after a predetermined time after learning the duty ratio mapped to a predetermined hydrogen flow rate using a support vector machine model. It may include generating a hysteresis graph.

In one embodiment, the step of controlling the opening amount of the hydrogen supply valve so that a predetermined flow rate of hydrogen is supplied based on the learning result is mapped to the learned hysteresis graph when a hydrogen flow rate within a predetermined range is required. Controlling an opening amount of the hydrogen supply valve based on the duty ratio may be included.

In one embodiment, the step of controlling the opening amount of the hydrogen supply valve so that a predetermined flow rate of hydrogen is supplied based on the learning result is mapped to a hysteresis graph before learning when a hydrogen flow rate outside the predetermined range is required. and controlling an opening amount of the hydrogen supply valve based on the duty ratio.

This technology applies a learning algorithm to the hydrogen supply valve control to predict the hysteresis function, so that a constant output of the fuel cell system can be expected through stable hydrogen supply, and the durability and fuel efficiency of the fuel cell system can be improved.

In addition to this, various effects identified directly or indirectly through this document may be provided.

1 is a diagram showing a fuel cell control device according to an embodiment of the present invention;
2 is a block diagram for explaining hydrogen supply valve control according to an embodiment of the present invention;
3 and 4 are diagrams for explaining the application of a learning algorithm for controlling a hydrogen supply valve according to an embodiment of the present invention;
5 is a flowchart illustrating a fuel cell control method according to an embodiment of the present invention.

Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that this is not intended to limit the present invention to the specific embodiments, and includes various modifications, equivalents, and/or alternatives of the embodiments of the present invention.

Various embodiments of this document and terms used therein are not intended to limit the technical features described in this document to specific embodiments, and should be understood to include various modifications, equivalents, or substitutes of the embodiments.

In connection with the description of the drawings, like reference numbers may be used for like or related elements. The singular form of a noun corresponding to an item may include one item or a plurality of items, unless the relevant context clearly dictates otherwise.

In this document, "A or B", "at least one of A and B", "at least one of A or B", "A, B or C", "at least one of A, B and C", and "A Each of the phrases such as "at least one of , B, or C" may include any one of the items listed together in that phrase, or all possible combinations thereof.

Terms such as "first", "second", or "first" or "secondary" may simply be used to distinguish a given component from other corresponding components, and may be used to refer to a given component in another aspect (eg, importance or order) is not limited. A (e.g., first) component is said to be "coupled" or "connected" to another (e.g., second) component, with or without the terms "functionally" or "communicatively." When mentioned, it means that the certain component may be connected to the other component directly (eg by wire), wirelessly, or through a third component.

1 is a diagram showing a fuel cell control device according to an embodiment of the present invention, FIG. 2 is a block diagram for explaining control of a hydrogen supply valve according to an embodiment of the present invention, and FIGS. 3 and 4 are It is a diagram for explaining application of a learning algorithm to control a hydrogen supply valve according to an embodiment of the present invention.

1 and 2, the fuel cell control device according to an embodiment of the present invention includes a hydrogen supply unit H2, a hydrogen shutoff valve 110, a hydrogen supply valve 120, an ejector 130, and a fuel cell stack. (140), hydrogen purge valve (150), hydrogen pressure sensor (155), water trap (160), condensate valve (170), BPCU (180), air supplier (190), humidifier (200), air shutoff valve ( It can be configured to include an Air Cut-off Valve (ACV) (210) and a control unit (220).

The hydrogen supply unit H2 supplies hydrogen as fuel to the fuel cell stack 140 and may include a hydrogen tank (not shown) and a regulator (not shown) in which hydrogen is stored, and a hydrogen supply system extending from the hydrogen tank. Hydrogen may flow to the fuel cell stack 140 along the piping.

The hydrogen tank can store high-pressure hydrogen gas of about 700 bar inside the container.

The regulator can reduce the high-pressure hydrogen stored in the hydrogen tank to about 17 bar, and supply the reduced-pressure hydrogen to the rear end of the regulator.

A hydrogen shutoff valve 110 and a hydrogen supply valve 120 may be provided at a rear end of the regulator along the pipe of the hydrogen supply system.

The hydrogen shutoff valve 110 may be a normally-closed (NC) type valve. The hydrogen shutoff valve 110 may be a valve for blocking hydrogen that may be discharged from the hydrogen tank in an emergency.

The hydrogen supply valve 120 may supply the fuel cell stack 140 by reducing the hydrogen supply of 17 bar from the hydrogen shutoff valve 110 to 1 to 2 bar. The hydrogen supply valve 120 may be configured as a solenoid type valve that can be driven by an electromagnet. The hydrogen supply valve 120 may be connected to the fuel cell stack 140 by a pipe formed at a rear end.

The ejector 130 may supply low-pressure hydrogen supplied through the hydrogen supply unit H2 to the fuel cell stack 140 .

The ejector 130 may mix high-temperature and high-humidity hydrogen discharged from the hydrogen electrode of the fuel cell stack 140 and dry hydrogen supplied from the hydrogen supply unit H2 and supply the mixture to the hydrogen electrode of the fuel cell stack 140 .

The fuel cell stack 140 is capable of generating electricity by using a chemical reaction between hydrogen and oxygen, and includes a Membrane Electrode Assembly (MEA) to which a catalyst electrode layer in which an electrochemical reaction occurs on both sides of an electrolyte membrane is attached, and a membrane It may be composed of an anode stacked on one surface of the electrode assembly and supplied with hydrogen as a fuel, and a cathode stacked on the other surface of the membrane electrode assembly and supplied with oxygen as an oxidizing agent.

The condensed water generated by the electrochemical reaction in the fuel cell stack 140 is generated inside the fuel cell stack 140 and must be smoothly discharged to the outside of the fuel cell stack 140 .

When the condensate is not well discharged from the inside of the fuel cell stack 140, that is, when it is in a flooding state, the supply of hydrogen, which is the fuel, is hindered, and the power generation performance of the fuel cell stack 140 is deteriorated. In serious cases, the fuel cell Damage to the stack 140 may occur.

In order to discharge the condensed water from the hydrogen electrode, the flow rate of the fluid (mixed gas containing moisture) inside the fuel cell stack 140 must be increased by increasing the flow rate of hydrogen inside the fuel cell stack 140 .

At this time, the most used is periodic hydrogen purging. That is, when moisture in the fuel cell stack 140 is to be removed, the hydrogen flow rate in the fuel cell stack 140 can be temporarily increased by purging through the condensate valve 170 .

Gas containing hydrogen discharged from the hydrogen electrode may be recycled back to the hydrogen electrode through the ejector 130, and the rest may be discharged to the outside through the water trap 160 and the condensate valve 170.

The hydrogen purge valve 150 may release hydrogen unnecessary for reaction in the fuel cell stack 140 into the atmosphere.

The hydrogen pressure sensor 155 may detect pressure of hydrogen supplied to the fuel cell stack 140 through the hydrogen supply valve 120 .

The water trap 160 may store condensed water.

The condensate valve 170 is a valve for discharging the condensed water stored in the water trap 160 to the outside. The condensate valve 170 may be configured as a solenoid-type valve that can be driven by an electromagnet.

The BPCU 180 may receive DC power from a high voltage battery (240V BAT) and convert it into AC power. The BPCU 180 charges the DC power output from the high voltage battery in a high voltage capacitor (not shown) and transfers it to the air supply motor 191 of the air supply unit 190 to control the driving of the air supply motor 191. there is.

For reference, the current of the high voltage battery may be controlled by a negative (-) high voltage main relay, a high voltage pre-charge relay, and a positive (+) high voltage main relay.

The negative (-) high voltage main relay is installed on the (-) power line connecting the high voltage battery and the BPCU (180), and the positive (+) high voltage main relay is (+) connecting the high voltage battery and the BPCU (180). It is installed in the power line, and the high voltage precharge relay can be installed in parallel with the positive (+) high voltage main relay.

In order to charge the high voltage capacitor with the current of the high voltage battery, when the negative (-) high voltage main relay is shorted and the high voltage precharge relay is shorted, an inrush current is generated and a certain voltage is applied to the high voltage capacitor, so charges are gathered and energy is generated. can be charged. Next, when the high voltage pre-charge relay is opened after shorting the positive (+) high voltage main relay, the high voltage capacitor can be charged.

When the voltage of the high voltage capacitor reaches a predetermined set voltage or higher, it may be determined that charging is completed. For example, when the voltage of the high voltage capacitor is 80% or more of the high voltage battery voltage, it may be determined that charging is completed.

The air supplier 190 includes an air supply motor 191 which is a brushless direct current motor (BLDC) three-phase motor using a high voltage, and rotates the air supply motor 191 to generate compressed air to the humidifier 200. can supply The air supply motor 191 receives AC power converted by the BPCU 180.

Meanwhile, when the air supplier 190 rotates at a predetermined RPM, whether or not it rotates can be checked through the air flow sensor 195.

The humidifier 200 may be provided between an air supply line connected to the air electrode of the fuel cell stack 140 and an air discharge line to humidify air supplied to the air electrode along the air supply line.

The humidifier 200 exchanges moisture between air introduced in a dry state through the air supplier 190 and flowing along the air supply line and air discharged from the air electrode in a wet state and flowing along the air discharge line, so that the air supply line can humidify the air flowing through it.

The air shutoff valve 210 may perform an opening or closing operation to allow or block the flow of air supplied from the humidifier 200 to the fuel cell stack 140 .

When the air shutoff valve 210 is in an open state during vehicle startup (IGN ON), the air supplied from the humidifier 200 is supplied to the air electrode of the fuel cell stack 140 or the air used in the fuel cell stack 140 Can be discharged to the humidifier (200). The air shutoff valve 210 can block air from entering the fuel cell stack 140 by being closed at the end of vehicle startup (IGN Off) and blocking the flow of air.

The control unit 220 may include a fuel control unit (FCU) and may process signals transmitted between components of the fuel cell control device.

The controller 220 may be connected to the fuel cell stack 140 to receive state information of the fuel cell stack 140 such as voltage, current, and temperature of the fuel cell stack 140 .

The control unit 220 may control the BPCU 180 to charge the high voltage capacitor with a voltage or control the high voltage capacitor to discharge the charged voltage. The control unit 220 may generate a speed control command for the air supply motor 191 and provide it to the BPCU 180 .

The BPCU 180 may adjust the flow rate of air generated in the air supplier 190 with a speed control command for the air supply motor 191 generated by the control unit 220 . Accordingly, the air supply unit 190 may supply the air in an adjusted amount to the humidifier 200 under the control of the BPCU 180 .

The control unit 220 is electrically connected to the hydrogen shutoff valve 110, the hydrogen supply valve 120, the hydrogen purge valve 150, the condensate valve 170, etc., and controls the opening or closing of each valve by a signal or communication. can

The controller 220 controls the opening amount of the hydrogen supply valve 120 based on the hydrogen pressure detected by the hydrogen pressure sensor 155, and learns the opening amount of the hydrogen supply valve 120 matched to the predetermined hydrogen pressure. The opening amount of the hydrogen supply valve 120 may be controlled so that a predetermined flow rate of hydrogen is supplied.

The control unit 220 may control pulse width modulation (PWM) and adjust the opening amount of the hydrogen supply valve 120 by controlling a duty ratio.

In the hydrogen supply valve 120, the opening amount is adjusted by adjusting the plunger height based on the duty, and the hydrogen flow rate is changed based on the opening amount control, thereby forming a hydrogen electrode pressure.

Eventually, the control unit 220 may set the control target value of the hydrogen supply valve 120 as the hydrogen electrode pressure and adjust the duty ratio while receiving feedback of the hydrogen electrode pressure through the hydrogen pressure sensor 155. there is.

For reference, the hydrogen flow rate and the hydrogen electrode pressure can be expressed as equivalent concepts. That is, when the hydrogen flow rate increases, the hydrogen electrode pressure may also increase, and when the hydrogen flow rate decreases, the hydrogen electrode pressure may also decrease.

Referring to FIG. 3 , the control unit 220 may control the opening amount of the hydrogen supply valve 120 based on a hysteresis graph in which a predetermined duty ratio is mapped to a predetermined hydrogen flow rate.

Meanwhile, a rattle-shaped hysteresis graph may be generated by external environmental factors such as heat generation, and the hysteresis graph may continuously change when the duty ratio is increased or decreased to control the opening amount of the hydrogen supply valve 120. there is.

That is, even when the flow rate of hydrogen supplied when the duty ratio is increased and the flow rate of hydrogen supplied when the duty ratio is decreased are the same, the opening amount of the hydrogen supply valve 120 may be different.

Therefore, the controller 220 learns the duty ratio mapped to a predetermined hydrogen flow rate using a learning model, and generates a hysteresis graph learned after a predetermined time, so that the hydrogen opening amount is constantly adjusted according to the required hydrogen flow rate. The supply valve 120 can be controlled.

Referring to FIG. 3 , learning may be performed in a section in which the flow rate compared to the duty ratio rapidly changes, for example, in a section between a duty ratio of 40% and 65%. This section may be a normal running state of the vehicle.

Since the learning model for learning is non-linear control rather than linear control, the support vector machine algorithm can be applied among the supervised models, and the polynomial version model among the non-linear control kernel techniques can be applied. learning can be performed.

In this case, the control unit 220 may perform learning when the hydrogen supply valve 120 is in a steady state, that is, when the hydrogen purge valve 150 is closed.

This is because the hydrogen electrode pressure is not constant in a complex situation such as control of the hydrogen supply valve 120 in a state in which the hydrogen purge valve 150 is open, that is, in a hydrogen purge state.

Referring to FIG. 4 , a separating function is required to distinguish an upward curve S1 and a downward curve S2 in the hysteresis graph.

In the separation function, the hydrogen electrode pressure, which is a value that can be actually measured by the control unit 220, may be placed on the Y-axis, and the duty ratio of the hydrogen supply valve 120 may be placed on the X-axis.

As a process for calculating the separation function, it is necessary to calculate a support vector called an alpha value, which is values closest to the separation function.

The support vector can be calculated with reference to Equation 1.

Here, the tuning parameter C is set to 1, and the degree q is set to 4 to use a 4th order polynomial, and the kernel function K is as shown in Equation 2.

Meanwhile, Degree q is not limited to 4, and the degree can be set differently depending on the situation.

Subsequently, a separation function may be calculated by substituting only the values classified as support vectors from the measured duty ratio of the hydrogen supply valve 120 and the hydrogen pole pressure data into Equation 3.

For example, assuming that 0.997x4-2.335x3+2.15x2-0.17x+b is derived as f(x) in Equation 3, the values classified as support vectors applied to Equation 1 are returned to Equation 3 After obtaining b by substitution, f(x) can be finally calculated.

For reference, the reason for applying the support vector machine is to linearize it on a hyperplane by increasing the dimension through a kernel function in order to distinguish non-linearly distributed two-dimensional data. That is, if the function of the existing hysteresis graph is f(x), pi(x) generated through the kernel function can be multidimensional.

In addition, it is declared that a kernel function in the form of Equation 2 will be used, and a large number of data measured based on Equation 2 is applied to Equation 1, which is the pi (x) function, to obtain a linear function with a maximum margin value Find it, and apply this function to Equation 3 again to calculate it in the f (x) dimension.

Then, since the finally calculated f(x) is a function of the hydrogen flow rate and duty ratio supplied through the hydrogen supply valve 120, the final function Y should be calculated with the hydrogen electrode pressure as a control parameter.

Therefore, the final function Y can be calculated with reference to Equation 4.

Therefore, the controller 220 determines the opening amount of the hydrogen supply valve 120 while receiving feedback on the hydrogen electrode pressure based on the control model of the hydrogen supply valve 120 through the final function Y with reference to Equation 4. You can control it.

On the other hand, in the section other than the normal driving state, for example, in the section where the duty ratio is less than 40% and the section where the change in flow rate is relatively small compared to the duty ratio, and in the section where the duty ratio exceeds 65%, there is little learning effect, so learning is unnecessary. do.

Therefore, in a section other than the normal driving state, the opening amount of the hydrogen supply valve 120 can be controlled based on the duty ratio mapped to the hysteresis graph before learning.

By generating the hysteresis graph learned in this way and controlling the hydrogen supply valve 120 so that the opening amount is constantly adjusted according to the required hydrogen flow rate, a constant output of the fuel cell system through stable hydrogen supply of the hydrogen supply system (FPS) is expected It can improve the durability and fuel efficiency of the fuel cell system, and make a predictable fuel cell model in the power system.

In addition, the controller 220 may control at least one other component (eg, a hardware or software component) of the fuel cell control device and may perform various data processing or calculations.

According to an embodiment, as at least part of data processing or operation, the control unit 220 stores commands or data received from other components (eg, sensors) in a volatile memory, and processes the commands or data stored in the volatile memory. And the resulting data can be stored in non-volatile memory.

According to an embodiment, the control unit 220 may include a main processor (eg, a central processing unit or an application processor) or a secondary processor (eg, a graphic processing unit, an image signal processor, a sensor hub processor, or a communication processor) that may operate independently or together with the main processor. processor). For example, when the controller 220 includes a main processor and an auxiliary processor, the auxiliary processor may use less power than the main processor or may be set to be specialized for a designated function. A secondary processor may be implemented separately from, or as part of, the main processor.

Although not shown in the drawings, according to embodiments, the fuel cell control device may further include a storage unit.

The storage unit may store a command for controlling the fuel cell control device, a control command code, control data, or user data. For example, the storage unit may include at least one of an application program, an operating system (OS), middleware, or a device driver.

The storage unit may include one or more of volatile memory and non-volatile memory.

Volatile memory includes dynamic random access memory (DRAM), static RAM (SRAM), synchronous DRAM (SDRAM), phase-change RAM (PRAM), magnetic RAM (MRAM), resistive RAM (RRAM), and ferroelectric RAM (FeRAM). can include

The nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, and the like.

The storage unit may further include a nonvolatile medium such as a hard disk drive (HDD), a solid state disk (SSD), an embedded multimedia card (eMMC), and a universal flash storage (UFS). there is.

Hereinafter, a fuel cell control method according to another embodiment of the present invention will be described in detail with reference to FIG. 5 .

5 is a flowchart illustrating a fuel cell control method according to an embodiment of the present invention.

Hereinafter, it is assumed that the fuel cell control apparatus of FIGS. 1 and 2 performs the process of FIG. 5 .

First, after starting (S110), with the hydrogen shutoff valve 110 closed, the operation of the hydrogen supply valve 120 can be checked by setting the hydrogen supply valve 120 to the maximum open state (S120) .

Subsequently, hydrogen is supplied by opening the hydrogen shutoff valve 110, the control unit 220 performs PWM control, receives feedback of the hydrogen electrode pressure through the hydrogen pressure sensor 155, and controls the duty ratio to supply the hydrogen supply valve ( 120) can be adjusted.

At this time, the control unit 220 may control the opening amount of the hydrogen supply valve 120 based on a hysteresis graph in which a predetermined duty ratio is mapped to a predetermined flow rate of hydrogen (S130).

In order to perform learning, the control unit 220, when the hydrogen purge valve 150, which is a steady state of the hydrogen supply valve 120, is closed (S140), is a section in which the flow rate compared to the duty ratio rapidly changes. For example, in a section between 40% and 65% of the duty ratio (S150), the flow rate value against the duty ratio may be continuously measured and stored (S160).

On the other hand, when the hydrogen purge valve 150 is in a closed state and the period in which the change in flow rate compared to the duty ratio is relatively small, for example, in the period where the duty ratio is less than 40% and in the period where the duty ratio exceeds 65%, the learning effect is Since there is little, learning is unnecessary (S170).

Subsequently, the controller 220 learns the duty ratio mapped to a predetermined hydrogen flow rate using a learning model, and generates a hysteresis graph learned after a predetermined time, so that the hydrogen opening amount is constantly adjusted according to the required hydrogen flow rate. The supply valve 120 can be controlled.

Since the learning model for learning is non-linear control rather than linear control, the support vector machine algorithm can be applied among the supervised models, and the polynomial version model among the non-linear control kernel techniques can be applied. It is possible to perform learning by doing (S180).

Referring to FIG. 4 , a separating function is required to distinguish an upward curve S1 and a downward curve S2 in the hysteresis graph.

In the separation function, the hydrogen electrode pressure, which is a value that can be actually measured by the control unit 220, may be placed on the Y-axis, and the duty ratio of the hydrogen supply valve 120 may be placed on the X-axis.

As a process for calculating the separation function, it is necessary to calculate a support vector called an alpha value, which is values closest to the separation function.

Subsequently, f(x) may be finally calculated after calculating a separation function using values classified as support vectors from the measured duty ratio of the hydrogen supply valve 120 and the hydrogen pole pressure data.

Subsequently, since the finally calculated f(x) is a function of the duty ratio for the hydrogen flow rate supplied through the hydrogen supply valve 120, the final function Y should be calculated with the hydrogen electrode pressure as a control parameter.

Based on the control model of the hydrogen supply valve 120 through the final function Y, the control unit 220 may control the opening amount of the hydrogen supply valve 120 while receiving feedback on the hydrogen electrode pressure (S190).

Various embodiments of the present document may be implemented as software (eg, a program) including one or more instructions stored in a storage medium (eg, internal memory or external memory) readable by a machine. For example, the device may call at least one command among one or more commands stored from a storage medium and execute it. This enables the device to be operated to perform at least one function according to the at least one command invoked. The one or more instructions may include code generated by a compiler or code executable by an interpreter.

The device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-temporary' only means that the storage medium is a tangible device and does not contain a signal (e.g. electromagnetic wave), and this term refers to the case where data is stored semi-permanently in the storage medium. It does not discriminate when it is temporarily stored.

According to one embodiment, the method according to various embodiments disclosed in this document may be included and provided in a computer program product. Computer program products may be traded between sellers and buyers as commodities. A computer program product is distributed in the form of a machine-readable storage medium (e.g. compact disc read only memory (CD-ROM)), or distributed via an application store or directly between two user devices, online (e.g. : can be downloaded or uploaded). In the case of online distribution, at least part of the computer program product may be temporarily stored or temporarily created in a storage medium readable by a device such as a manufacturer's server, an application store server, or a relay server's memory.

According to various embodiments, each component (eg, module or program) of the components described above may include a single object or a plurality of objects, and some of the multiple objects may be separately disposed in other components. .

According to various embodiments, one or more components or operations among the aforementioned components may be omitted, or one or more other components or operations may be added.

Alternatively or additionally, a plurality of components (eg modules or programs) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by a corresponding component of the plurality of components prior to the integration. .

According to various embodiments, operations performed by modules, programs, or other components are executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations are executed in a different order, omitted, or , or one or more other operations may be added.

As described above, according to the present invention, by applying a learning algorithm to the hydrogen supply valve control to predict the hysteresis function, a constant output of the fuel cell system can be expected through stable hydrogen supply, and the durability and fuel efficiency of the fuel cell system can be improved. can be improved

The above description is merely an example of the technical idea of the present invention, and various modifications and variations can be made to those skilled in the art without departing from the essential characteristics of the present invention.

Therefore, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention, but to explain, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be construed according to the claims below, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

Claims (12)

a hydrogen supply valve controlling the supply of hydrogen to the fuel cell stack;
a hydrogen pressure sensor for sensing the pressure of hydrogen supplied to the fuel cell stack; and
Based on the hydrogen pressure sensed by the hydrogen pressure sensor, the opening amount of the hydrogen supply valve is controlled, and after learning the opening amount of the hydrogen supply valve matched to the predetermined hydrogen pressure, the hydrogen supply valve is supplied with a predetermined flow rate. A control unit that controls the amount of opening of
A fuel cell control device comprising a. The method of claim 1,
The hydrogen supply valve,
A fuel cell control device characterized in that it is a solenoid valve whose opening amount is controlled by adjusting the duty ratio. The method of claim 1,
The control unit,
A fuel cell control device characterized in that the opening amount of the hydrogen supply valve is controlled based on a hysteresis graph in which a predetermined duty ratio is mapped to a predetermined hydrogen flow rate. The method of claim 3,
The control unit,
A fuel cell control device characterized by generating a learned hysteresis graph after a predetermined time after learning the duty ratio mapped to a predetermined hydrogen flow rate using a support vector machine model. The method of claim 4,
The control unit,
When a hydrogen flow rate within a predetermined range is required, the fuel cell control device controls an opening amount of the hydrogen supply valve based on a duty ratio mapped to the learned hysteresis graph. The method of claim 4,
The control unit,
When a hydrogen flow rate outside the predetermined range is required, the fuel cell control device controls the opening amount of the hydrogen supply valve based on a duty ratio mapped to a hysteresis graph prior to learning. controlling an opening amount of a hydrogen supply valve based on the hydrogen pressure detected by the hydrogen pressure sensor;
learning an opening amount of the hydrogen supply valve matched to a predetermined hydrogen pressure; and
Controlling the opening amount of the hydrogen supply valve so that a predetermined flow rate of hydrogen is supplied based on the learning result.
A fuel cell control method comprising a. The method of claim 7,
The step of controlling the opening amount of the hydrogen supply valve based on the hydrogen pressure detected by the hydrogen pressure sensor,
Controlling the opening amount of a solenoid valve that controls the supply of hydrogen to the fuel cell stack by adjusting the duty ratio, and sensing the pressure of hydrogen supplied to the fuel cell stack by a hydrogen pressure sensor. Fuel cell control method. The method of claim 7,
The step of controlling the opening amount of the hydrogen supply valve based on the hydrogen pressure detected by the hydrogen pressure sensor,
and controlling an opening amount of the hydrogen supply valve based on a hysteresis graph in which a predetermined duty ratio is mapped to a predetermined hydrogen flow rate. The method of claim 9,
The step of learning the opening amount of the hydrogen supply valve matched to the predetermined hydrogen pressure,
A fuel cell control method comprising the step of learning the duty ratio mapped to a predetermined hydrogen flow rate using a support vector machine model and then generating a learned hysteresis graph after a predetermined time. The method of claim 10,
Controlling the opening amount of the hydrogen supply valve so that a predetermined flow rate of hydrogen is supplied based on the learning result,
and controlling an opening amount of the hydrogen supply valve based on a duty ratio mapped to the learned hysteresis graph when a hydrogen flow rate within a predetermined range is required. The method of claim 10,
Controlling the opening amount of the hydrogen supply valve so that a predetermined flow rate of hydrogen is supplied based on the learning result,
and controlling an opening amount of the hydrogen supply valve based on a duty ratio mapped to a hysteresis graph prior to learning when a hydrogen flow rate outside the predetermined range is required.

Images