Method for controlling a compressor recirculation valve

Abstract

Method for controlling a compressor recirculation valve (218) for a fuel cell system (102) during a downward operating transient, during which the output power of the fuel cell system (102) decreases, the method comprising: a stacked cathode flow setpoint is received for a gas flow supplied from an outlet of a compressor (202) to a cathode of a fuel cell (204); characterized by the fact that A recirculation valve flow setpoint for the supplied gas flow is calculated using compressor operating data and the stack cathode flow setpoint; a forward coupling term corresponding to a new recirculation valve position is calculated by using the recirculation valve flow setpoint; a feedback term corresponding to a new recirculation valve position is generated via a PID control strategy based on an error between the stacked cathode flow setpoint and the feedback term; and a recirculation valve control command is generated based on the forward coupling term and the feedback term, which causes the recirculation valve (218) to open during a downward operating transient.

Description

BACKGROUND OF THE INVENTION

[0001] The present patent application relates generally to controlling a compressor recirculation valve in a vehicle fuel cell system, and more specifically to a method for controlling a compressor recirculation valve according to the preamble of claim 1 to help satisfy a fuel cell stack cathode flow setpoint, particularly during transient operating conditions. Such a generic method is essentially known from US 2004 / 0 161 647 A1.

[0002] Furthermore, it is known from US 5 991 670 A to control a throttle valve located downstream of the cathode outlet of a fuel cell stack on the basis of a feedforward and a feedback term.

[0003] Automotive technology is rapidly evolving to find alternatives to the use of gasoline as the primary energy source in vehicle propulsion systems. One area of ​​recent interest has focused on the use of gaseous reactants as fuel. Gases such as hydrogen, due to their abundant occurrence in nature, offer a promising alternative to gasoline. In some automotive systems, propulsion can be achieved by using hydrogen gas as part of a chemical reaction within a fuel cell to generate electrical energy, which can be used to power an electric motor. The generated electricity can then be used to propel the vehicle through the engine, either alone or in conjunction with a petroleum-based internal combustion engine.Such fuel systems also generally produce less pollution than petroleum-based ones.

[0004] In a typical fuel cell, hydrogen or another reactant gas is fed to the fuel cell's anode, where the hydrogen is broken down into electrons and positively charged ions. An electrolyte layer separates the anode from the cathode, allowing the ions to pass to the cathode while preventing electrons from passing to the cathode. Instead, electrons are bypassed around the electrolyte layer, through a load, and back to the cathode to allow electrical power to be harnessed. At the cathode, the ions, electrons, and any supplied oxygen or air are typically combined to produce water and heat. Individual fuel cells can be arranged in series or parallel as a fuel cell stack to produce higher voltage or current output. Furthermore, even higher outputs can be achieved by combining more than one stack.

[0005] In a vehicle employing fuel cell technology, these reactant gases can be transported and used within a pressurized gas system. For example, stored hydrogen can be supplied to a fuel cell anode and chemically reacted to generate an electric current. Similarly, air can be taken from an air intake and supplied to the cathode of a fuel cell. Such vehicle fuel cell systems require the use of compressors to perform this pressurization function and may also include auxiliary devices such as valves, control units, or the like to regulate the flow of a reactant gas between the compressor and the fuel cell.

[0006] An inherent characteristic of such compressors (at least as they relate to cathode-side operation) is that the cathode pressure control and flow control are coupled. This coupling tends to destabilize system operation, particularly during periods of transient system operation. These problems are particularly acute during transient operating conditions, where both the flow setpoint and the pressure setpoint can exhibit near-instantaneous changes. Even more specific are the difficulties in ensuring proper recirculation flow, particularly acute during down-transient operation, since the inertial effects of a compressor being requested to decelerate prevent the compressor speed from decreasing as rapidly as necessary; this, in turn, tends to cause recirculation flow imbalance.Because water production in a fuel cell stack varies with current, a downward transient event (where the stack current is suddenly reduced) causes a corresponding rapid decrease in water production. If the stack cathode airflow is not reduced along with this current reduction, the stack will quickly dry out, resulting in damage. Likewise, the imbalance can also result from insufficient flow being delivered to the cathode. In this case, the presence of a recirculation valve can allow excessive recirculation to occur, resulting in stack flooding. The presence of only a recirculation valve, by itself, does not ensure rapid, precise control to prevent both stack overflow dry out and stack underflow flooding. SUMMARY OF THE INVENTION

[0007] The invention is based on the object of stabilizing the system in order to minimize the drying and flooding conditions mentioned above.

[0008] This object is achieved by a method having the features of claim 1.

[0009] The method includes receiving as input a stacked cathode flow setpoint for a gas flow supplied from a compressor outlet to a fuel cell cathode. The method further includes calculating a recirculation valve flow command (also known as a predicted or desired value) using operating data from the compressor and the stacked cathode flow setpoint. The method further includes generating a recirculation valve control command that causes the recirculation valve to change positions. With this method, a compressor flow setpoint can be used to calculate a recirculation flow setpoint, e.g., by taking a difference between the compressor flow setpoint and the stacked cathode flow setpoint. This provides a predicted recirculation valve flow value.

[0010] The method takes place during an operating transient of the fuel cell system. The operating transient is a down-current transient, wherein the power generated by the fuel cell changes from high to low, as does the demand for oxygen or a similar cathode reactant supplied to the fuel cell stack. Thus, in the case of a down-current transient, the method can be used to predict the future recirculation valve position necessary to satisfy the new stack cathode flow setpoint, as well as to send a command to the recirculation valve to satisfy such a new setpoint. As such, the calculations performed in accordance with the present disclosure are preferably used in a predictive function to make a decision regarding a future valve position. Feedforward terms such asthe stacked cathode flow setpoint and the compressor discharge pressure setpoint may be used, while other operating parameters (particularly feedback parameters such as compressor speed, which typically changes relatively slowly) may also be used. Significantly, feedforward control of the compressor recirculation valve helps to stabilize and control this inherently coupled system. In one form, the control described herein may be used in conjunction with feedforward compressor speed control as directed in the accompanying application entitled PREDICTIVE CATHODE COMPRESSOR SPEED CONTROL IN A FUEL CELL POWER SYSTEM (US 2014 / 0 120 445 A1), which was filed on the same day as the present application and is incorporated by reference in its entirety. In the present context, the relative speed at which a component (e.g.The transition from the compressor speed (the compressor speed) to the cooling tower (the compressor speed) does not occur slowly; rather, such a transition occurs over a period of time that, if left uncorrected, could lead to adverse humidity imbalances within the stack. For example, a downward transient for the physical compressor speed from a maximum to a minimum might be approximately 4 seconds; this is a relatively long time for the stack's cathode membranes to be exposed to the drying effects of the increased airflow. Prompt deployment of the recirculation valve, as discussed herein, ameliorates the influence of the excess airflow during this transition period.The compressor operating data may be derived from a compressor map; this data may be converted into a conventional machine-readable form, for example, via a lookup table or the like using known compressor characteristics provided by the compressor manufacturer. In a particular form, a compressor inlet gas flow setpoint may be of the type of operating data that can be determined from the compressor map. Similarly, the recirculation valve flow setpoint may be calculated using a difference between the compressor inlet gas flow setpoint and the stacked cathode flow setpoint. The stacked cathode flow setpoint may equally well be generated elsewhere, e.g., in another algorithm that may be stored in a computer or similar control device.In another optional form, the recirculation valve flow setpoint is used to calculate the new recirculation valve position, which is treated as a feedforward term for the valve position setpoint. Furthermore, a proportional-integral-derivative (PID) controller is used to generate the compensation term for the new valve position, which is treated as the feedback term of the new valve position because it is based on the stack cathode flow feedback value and attempts to eliminate the error (or difference) between the stack cathode flow setpoint and the stack cathode flow feedback value. As such, the new valve position is the sum of the feedforward (i.e., open-loop) term and the feedback (i.e., closed-loop) term, and the controller will command the valve to change position to the new setpoint using a recirculation valve control command.Those skilled in the art will appreciate that although the new valve position may utilize feedback information (e.g., compressor speed, pressure, or temperature), it cannot utilize flow feedback information, as this constitutes the target control term.

[0011] In another embodiment, a controller for a vehicle fuel cell system is disclosed. The controller includes at least one processor and a non-transitory memory, the latter storing machine instructions executed by the former to cause the controller to determine a cathode flow setpoint for a gas flow supplied from an outlet of a compressor to a cathode of one or more fuel cells used to form a stack of fuel cells in the fuel cell system. A compressor map (which may be stored in RAM or ROM of the memory), including data corresponding to the operation of the compressor, may be accessed by the processor to assist it in performing one or more calculations related to recirculation valve settings.The instructions also cause the processor to calculate a flow value for the recirculation valve so that the gas flow from the compressor outlet to the recirculation valve takes into account the stack cathode flow setpoint and the compressor map data.

[0012] Optionally, operating data corresponding to the compressor map is contained in the non-transitory memory. The compressor map data can be accessed by the controller to determine a gas flow setpoint for the compressor inlet. The instructions can also be used to cause the processor to calculate a difference between the compressor inlet gas flow setpoint and the cathode flow setpoint as a way to determine the value to which the recirculation valve should be set. As mentioned above, the algorithm used to control the valve position can use feedback terms in addition to the setpoint and similar feedforward information. Preferably, a control signal generated by the controller can be sent to an actuator to transition the valve to a wider open or closed configuration according to the needs of the vehicle fuel cell system.

[0013] In another embodiment, a vehicle fuel cell system is disclosed. The system includes a fuel cell having a cathode and a compressor having an inlet and an outlet. The outlet is operatively connected to the cathode of the fuel cell and provides a gas to the cathode of the fuel cell. The system also includes a recirculation valve operatively connected to the inlet and outlet of the compressor. The system further includes a controller having at least one processor and a non-transitory memory in signal communication with the processor. The memory stores instructions that, when executed by the processor, cause the processor to determine a cathode flow setpoint for gas flow to the cathode. The instructions further cause the processor to use operating data related to the compressor, which may be stored in the form of a compressor map.The instructions also cause the processor to calculate a recirculation valve flow value for gas flow to the recirculation valve using the compressor map and the cathode flow setpoint; this, in turn, can be used to generate a recirculation valve control command that causes the valve to change positions.

[0014] Optionally, the compressor map is used to determine a gas flow setpoint to the compressor inlet. The instructions may further cause the compressor to calculate a difference between the compressor inlet gas flow setpoint and the cathode flow setpoint as a way to determine the recirculation valve value, which in an even more specific form may form a feedforward value for the controller. As stated above, the position of the recirculation valve may be determined by the feedforward value and may also be based on one or more of a feedback-based integral value and a derivative value. Data exchange devices (e.g.,signal-carrying lines, connectors, or the like) may form part of the fuel cell system; such a device may be used to relay the recirculation valve control command between at least the control unit and the recirculation valve. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The following detailed description of specific embodiments is best understood when read in conjunction with the following drawings, in which like structures are designated by like reference numerals and in which: Fig. 1 shows a vehicle with a fuel cell system; Fig. 2 a schematic illustration of the air supply system in the Fig. 1 shown vehicle fuel cell system; Fig. 3 is an illustration of a compressor map; Fig. 4 a schematic illustration of a control unit for the vehicle fuel cell system of Fig. 1 is. Fig. 5A shows a flow control for a downward transient according to the prior art; Fig. 5B shows flow control for a downward transient according to the present disclosure; Fig. 6A shows a flow control for an up-transient according to the prior art; and Fig. 6B shows flow control for an up-transient according to the present disclosure.

[0016] The embodiments presented in the drawings are purely illustrative. Furthermore, certain aspects of the drawings and embodiments will become more apparent and better understood in light of the following detailed description. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] As stated above, the use of compressors within vehicle fuel cell systems presents additional challenges for the automotive industry, with the flow rate, as well as the related fuel cell cathode stoichiometry, being an important parameter for such a system. Cathode stoichiometry is defined as the ratio of the molar inlet flow of a reactant to the consumption rate of the reactant and is therefore directly proportional to the gas flow to the cathode. When the fuel cell output transitions from high power to low power (e.g., during an idle state), there is a lower reactant consumption, which in turn significantly increases the cathode stoichiometry. However, this increase in stoichiometry can dry out the stack and consequently affect stack performance or even cause stack damage.In any event, providing rapid opening and closing of the recirculation valve according to aspects of the invention will increase the likelihood that the system will quickly reach the flow setpoint and avoid a stack over-stoichiometry condition; these more precise commands are particularly advantageous during down-transient conditions, where latency associated with decreasing compressor speed may otherwise hinder maintaining proper stack stoichiometry.

[0018] Now referring to Fig. 1, a vehicle 100 is shown in accordance with the embodiments shown and described herein. The vehicle 100 (e.g., a car, a bus, a truck, or a motorcycle) is powered by one or more fuel cells forming the fuel cell system 102, which converts a stored gaseous fuel into electrical power for the engine 104. In a preferred embodiment, the fuel cell system 102 is a hydrogen-based fuel cell system that converts a catalytic reaction between hydrogen gas and oxygen into electrical power. The fuel cell system 102 may include one or more onboard hydrogen storage tanks or similar storage tanks (none shown), the latter being such that the reformation of a hydrogen gas precursor can take place within the vehicle.In such a case, the fuel cell system 102 may also include any number of valves, compressors, piping, temperature controllers, electrical storage devices (e.g., batteries, ultracapacitors, and the like, none of which are shown), and controllers to help ensure proper operation of the fuel cell system 102. The engine 104 may be fully electric or a hybrid engine (e.g., an engine that uses both electricity and petroleum-based combustion to propel the vehicle 100). In either configuration, the engine 104 utilizes power from the fuel cell system 102 for at least a portion of the propulsion needs of the vehicle 100.

[0019] Any number of different types of fuel cells may be used in the fuel cell system 102 (e.g., metal hydride fuel cells, alkaline fuel cells, electrogalvanic fuel cells, or other known types of fuel cells). Multiple fuel cells may also be connected in series and / or parallel as a fuel cell stack within the fuel cell system 102 to produce a higher voltage or current output by the fuel cell system 102. The electrical power produced may then be supplied directly to the engine 104 or stored in an electrical storage device for later use by the vehicle 100.

[0020] Now referring to Fig. Figure 2 shows a schematic illustration of the fuel cell system 102 in accordance with the embodiments shown and described herein. The compressor 202 takes in air at the inlet 202A, compresses it, and provides the supplied gas via the outlet 202B to the cathode inlet 204A of the fuel cell stack 204, where the gas is used to produce electrical power for the vehicle 100. The exhaust gas or effluent liquids are then removed from the cathode outlet 204B via the drain 222. In addition to the compressor 202, the fuel cell system 102 also includes a number of valves that help regulate the flow of gas throughout the system. For example, the fuel cell system 102 may include a bypass valve 214 that allows some or all of the gaseous fuel to bypass the fuel cell stack 204.The fuel cell system 102 may also include a cathode shutoff valve 216 that regulates the flow of exhaust gas from the cathode outlet 204B. The cathode shutoff valve 216 may be used, for example, to control the pressure and flow of a gas through the cathode. A recirculation valve 218 allows selective fluid communication between the compressor outlet 202B and inlet 202A. A particularly valuable use of the recirculation valve 218 is during operating transients in the fuel cell stack 204, where it can increase or decrease the amount of air present at the inlet 206 depending on the transient requirements of the stack 204.

[0021] The fuel cell system 102 further includes a number of sensors that measure the properties of the gas within the system. For example, the fuel cell system 102 may include a flow meter 220 that measures the mass air flow from the air intake system. Other non-limiting examples of sensors (not shown) in the fuel cell system 102 include sensors that measure the mass air flow, pressure, temperature, density, or similar properties of the gas(es) used in the fuel cell system 102. Further examples include sensors that measure the positions of the valves 214, 216, 218 and the speed of the compressor 202.

[0022] A downward transient may occur in the fuel system 102, meaning that the output of the fuel cell stack 204 shifts to a lower output setting based on a request from the driver of the vehicle 100. In one example, a lower output setting may correspond to idle-state operation of the vehicle 100. However, to maintain the stability of the stack 204, the cathode stoichiometry must be increased in such a way that other operating parameters of the stack 204 are not adversely affected. In accordance with the present disclosure, the recirculation valve 218 at the outlet 202B of the compressor 202 may be opened. It should be noted that control via one device in the system 102 may be integrated with control via another device. For example, a setpoint may be used to control multiple valves of the system 102.In a more specific form, an algorithm used as part of the present disclosure may help determine a flow setpoint for compressor 202 to be used to calculate the recirculation flow setpoint, which would correspond to the difference between the flow setpoints of compressor 202 and stack 204. The logic of the present method for controlling recirculation valve 218 is particularly advantageous when used in conjunction with power-down transients in fuel cell system 102. In one form, the compressor flow setpoint information may be derived from a compressor map 500.

[0023] Now referring to the Fig. 3 and Fig. 4, further details are shown in connection with the compressor map 500 and the various features of the control unit 400. With particular reference to Fig. 3, data contained in the compressor map 500 is that corresponding to the operation of the compressor 202; this information may include pressure ratios for given flow conditions (e.g., including factors related to mass flow and may also account for ambient temperature and pressure variations) at various speed curves 505. The compressor maps 500 provide a useful tool for predicting the behavior of the compressor 202 by allowing the prediction of flow setpoints based on the operating state of the compressor 202. A compressor map 500 typically relates the pressure ratio 502 to the flow 504 of gas through the compressor 202.A surge line 506 separates those operating conditions of the compressor 202 that result in stalls, flow reversals, or other undesirable surge conditions from those that do not, with points to the left of the surge line 506 being problematic, while operating points to the right of the surge line 506 correspond to generally normal operation of the compressor 202. The distance between an operating point 508, 510, or 514 (explained in more detail below) of the compressor 202 and the surge line 506 is also referred to as the surge margin and provides a measure of how close the compressor 202 is to a surge condition.In one form, the data corresponding to the various operating conditions across the compressor map 500 may be stored in a lookup table or similar data structure, which in turn may be embedded or otherwise contained in any suitable machine-accessible medium, such as a pre-programmed chip or storage device, flash memory, a hard disk drive, a CD, a DVD, a floppy disk, or a similar non-transitory structure. Thus, the machine-accessible medium may be in the form of memory 404 that forms part of the controller 400. The data in the compressor map 500 may be preloaded into the controller 400 using information provided, for example, by the manufacturer of the compressor 202.In another form, the data may be constructed using a diagnostic routine that varies the operation of compressor 202 and detects various conditions; either form is acceptable. In addition to the feedforward setpoint parameters mentioned above, compressor 202 may also be controlled by one or more feedback loops to regulate its speed, differential pressure, gas flow rate, or similar parameters in an attempt to maintain the one or more setpoints.

[0024] As mentioned above, points 508, 510, and 514 illustrate various operating conditions (or points) that compressor 202 may experience. For example, at point 508, compressor 202 is currently operating at a feedback speed. When the fuel cell system output goes to low power, a downward transient condition may exist. Then, the speed of compressor 202 may be controlled to point 510 to conserve resources, as both the x-axis flow rate 504 and the y-axis pressure ratio 502 experience a corresponding decrease. Similarly, the speed of compressor 202 may be controlled along one of the constant speed lines 505 to point 514, as the rotational speed of compressor 202 slowly decreases during a downward transient condition.Under these circumstances, the gas flow setpoint of the compressor inlet 202A at point 514 is determined by the pressure ratio setpoint 502 and the feedback speed at point 508. The flow command of the recirculation valve 218 can be calculated from the difference between the gas flow setpoint of the compressor inlet 202A and the cathode flow setpoint shown at point 510. The recirculation valve 218 can then be opened to redirect any excess airflow, rather than allowing that air to pass through the cathode of the stack 204, where, as mentioned above, it could result in an unbalanced stoichiometry for the stack 204. As will be appreciated by those skilled in the art, a preferred maximum operating point "A" for compressor 202 is located at the top of the uppermost one of the speed lines 505, which intersects point 508.The inertia within the compressor 202 causes it to maintain its last speed for a short time during the down transient; this results in the compressor flow tending toward the operating point "C" at point 514 in the map 500.

[0025] With reference to the specific example given in Fig. 3, a transient condition from point 508 (high power) to 510 (low power) in map 500 means that the stacked cathode flow setpoint changes accordingly; likewise, the pressure setpoint for compressor outlet 202B is changed, as shown moving from point 508 to point 510. As the flow setpoint changes, the position of recirculation valve 218 must be predicted and then promptly adjusted to meet the new flow requirements at point 510. Since the running characteristics of the compressor 202 are already known by following certain properties (and, as mentioned above, generally change slowly, especially over the duration of the down transient), the compressor map 500 can be used to predict the position of the recirculation valve 218 with the simplified assumption that the speed of the compressor 202 remains unchanged over the course of a single time step.Additional operating knowledge may also be used. For example, knowing that the pressure at compressor outlet 202B will decrease when recirculation valve 218 is opened allows recirculation valve 218 to open to the predicted position, allowing the compressor to run at point 514. The air flow that must be redirected through recirculation valve 218 to satisfy the new stack cathode flow setpoint 514 is equal to the difference between it and point 510.

[0026] From this, the feedforward term of the new position of the recirculation valve 218 is then determined using a general valve model using the flow through the valve (i.e., the difference between the x-axis points at points 514 and 510), the flow temperature, the pressure at the inlet and outlet of the valve 218, as well as other necessary inputs. Together with the feedback term determined by the PID controller, the new position of the recirculation valve 218 is finally determined, after which the recirculation valve 218 is commanded to move to the new position, and the compressor 202 will run at point 514 while its pressure at the outlet 202B will decrease to the pressure at setpoint 514. In such an exemplary movement, the flow of the compressor 202 is displayed at point 514, while the flow of the fuel cell stack 204 is at point 510.Accordingly, the flow diverted (by the recirculation valve 218) is the difference between points 514 and 510, shown as Δ. ISP. Thus, reducing the rotational speed of compressor 202 causes its operating state to change from point 514 to point 510. During this process, both the flow of stack 204 and the pressure continue to satisfy the new setpoint 510. The feedforward term of the new valve position ensures that valve 218 opens quickly during a down-bias transient and closes quickly during an up-bias transient, while the feedback term of the new valve position from PID controller 400 can eliminate any offset error or mitigate any overshoot during a transient condition. The method of the present disclosure will cause the transient to decay more quickly, thereby avoiding dryout of stack 204.An important positive side effect of the recirculation valve 218 is that it helps prevent the compressor 202 from running into the aforementioned surge condition treatment, which is addressed in the accompanying application entitled REACTIVE COMPRESSOR SURGE MITIGATION STRATEGY FOR A FUEL CELL POWER SYSTEM (US 2014 / 0 120 447 A1), filed on the same day as the present patent application. When the system 102 is in a steady state, the recirculation valve 218 will move to a position based on surge control to prevent the compressor 202 from moving above the surge line 506. The valve 218 will recirculate a minimum flow rate while the compressor 202 remains at the lowest required speed to achieve the desired flow.

[0027] With particular reference to Fig. 4, the controller 400 for regulating the operation of the compressor recirculation valve 218 includes a processor 402, which may be at least one processor (e.g., a microprocessor, an application-specific integrated circuit (ASIC), a reprogrammable gate array, or the like) communicatively coupled to a memory 404 and interfaces 406 and 408 (which may correspond to input and output steps, respectively). The processor 402 may determine (among other things) a cathode flow setpoint for a gas flow delivered from the compressor 202 to the fuel cell stack 204, wherein the cathode flow setpoint corresponds to a desired operating state of the stack 204, and the desired operating state may be calculated in various modes. The desired operating state may be passed to a memory 404 for storing the state.In one form, the desired operating state may be stored in a memory 404 as a lookup table, a data structure, or a similar configuration. The memory 404 may be any type of memory capable of storing machine-executable instructions that perform one or more of the functions disclosed herein when executed by the processor 402. The memory 404 may be, for example, a RAM, a ROM, a flash memory, a hard disk, an EEPROM, a CD-ROM, a DVD, other forms of non-transitory storage devices, or any combination of such storage devices. The controller 400 may also include one or more of a digital signal conditioning capability, an application-specific integrated circuit, programmable logic devices, as well as dedicated gate or transistor hardware components.In this manner, controller 400 can generate and store non-transitory signals that can be used to control the positioning of recirculation valve 218 via one or more of the feedforward values ​​mentioned above, as well as feedback (i.e., integral / derivative) terms. System operation can be influenced by opening or closing valve 218 in response to the valve control command from controller 400.

[0028] In a preferred form, at least a portion of the controller 400 is configured to function as a feedforward device in that the operation of one or more of the Fig. 2 (including the recirculation valve 218) is based on mathematical models or a similar characterization of the fuel cell system 102, wherein program elements corresponding to such models can be loaded into the memory 404. In this way, by combining the mathematical model of the operation of parts of or all of the fuel cell system 102 into the algorithm in the controller 400, the controller 400 can perform corresponding control activities based on known system parameters (such as those sensed, measured, or otherwise incorporated into the algorithm).As stated above, optimization of the operation of the fuel cell system 102 can be enhanced by combining feedforward elements with some feedback-based elements, since feedforward control is preferred for the speed of transients, while feedback is needed to reduce small steady-state errors.

[0029] The controller 400 communicates with various devices in the fuel cell system 102 and, in some cases, provides either direct or indirect control over other devices such as the recirculation valve 218, the compressor 202, and other components. The controller 400 also receives, either directly or indirectly, sensor measurements from the fuel cell system 102 to monitor the operation of the system. For example, the controller 400 may receive temperature, pressure, flow, density, valve position, speed, or other measurements from sensors associated with the compressor 202, the recirculation valve 218, or the like. In addition, the controller 400 may also receive sensor readings from other devices (e.g., a flow meter 220, sensors associated with another valve 214, 216, or the like).It should be appreciated that any number of different combinations of sensors and sensor configurations may be used without departing from the principles or teachings of the present disclosure.

[0030] Connections 410, 412, and 414 may be any combination of wired or wireless connections that provide connections between controller 400 and compressor 202, recirculation valve 218, cathode shutoff valve 216, bypass valve 214, and other devices. In some embodiments, connections 410, 412, and 414 are part of a common data line that transmits measurement data to controller 400 and control commands to the devices of fuel cell system 102 connected to controller 400. In still other embodiments, connections 410, 412, and 414 may include one or more intermediary circuits (e.g., other microcontrollers, signal filters, etc.) and provide an indirect connection between controller 400, compressor 202, recirculation valve 218, and other devices 434.

[0031] The interface 408 is configured to receive measurement data and to transmit control commands to the recirculation valve 218, the compressor 202, and other devices. The interface 408 may, for example, include one or more wireless transceivers if any of the connections 410, 412, and 414 is a wireless connection. The interface 408 may also include one or more wired connections if any of the connections 410, 412, and 414 is a wired connection. The interface 408 may also include circuitry configured to digitally sample or filter received measurement data. The interface 408 may, for example, B. Sample temperature data received from a temperature sensor of compressor 202 via connection 412 at discrete times (e.g., k, k+1, k+2, etc.) to generate discrete temperature values ​​(e.g., T(k), T(k+1), T(k+2), etc.).

[0032] In some embodiments, controller 400 may also communicate with interface devices 440 (e.g., a display, a speaker, a printer, or any other electronic device that provides data to or receives data from a user) via connection 436. Controller 400 may also communicate with other computing systems 442 (e.g., another controller, a portable electronic device, a server, or the like) via connection 438. Connections 436 and 438 may be wired or wireless connections. For example, other computing systems 442 may include a server located remotely from vehicle 100, and connection 438 may be a wireless connection. For example, server 400 may relay status condition information to the server via a cellular phone, Wi-Fi, radio, or satellite connection, or the like.Interface 406 may also include one or more transceivers configured to transmit and receive location information for vehicle 100. Interface 406 may include, for example, a GPS receiver or a mobile receiver that uses triangulation to determine the location of vehicle 100. In other embodiments, interfaces 406 and 408 may be a single interface for controller 400.

[0033] Controller 400 uses the setpoints for pressure ratio, surge margin, or flow rate to control recirculation valve 218, as well as the other valves. For example, controller 400 may determine a required valve position for recirculation valve 218 that moves the operating state of compressor 202 away from the surge line of compressor map 50. Controller 400 then generates a recirculation valve control command that causes recirculation valve 218 to open or close accordingly.

[0034] The parameters 432 that can change the functions of the controller 400 can include one or more thresholds for the speed of the compressor 202, temperature thresholds, or a surge condition time limit. These parameters 432 can also include values ​​that enable or disable functions of the controller 400. Such parameters 432 can also include, for example, those that determine whether a surge reducer is active. Likewise, the parameters 432 can further include values ​​that determine how information about the fuel cell system 102 is transmitted to the interface devices 440 or to other computing systems 442 (e.g., how often, what format, and the like).

[0035] In some cases, some or all of the parameters 432 may be preloaded into memory 404 (e.g., by the manufacturer of the controller 400, the manufacturer of the vehicle 100, or the like). In other cases, some or all of the parameters 432 may be provided to the controller 400 via the interface devices 440 or other computing systems 442. Furthermore, some or all of the parameters 432 may be updated or deleted via the interface devices 440 or other computing systems 442.

[0036] Although controller 400 is shown as a singular device, it should be understood that this is purely exemplary and not intended to be limiting. For example, processor 402 may include any number of processors of any number of computing devices that execute the instructions stored in memory 404. Likewise, memory 404 may include any number of storage devices and is not limited to storage devices residing in the same housing as processor 402. In some cases, processor 402 or memory 404 may even be located external to vehicle 100.

[0037] The flow setpoint for the recirculation valve 218 is calculated using the compressor map 500 and the cathode flow setpoint. As stated above, this flow setpoint corresponds to a predicted flow, and all such calculations are predictive. By being a predictive parameter, this flow value can reduce the time and uncertainty associated with operating transients. A gas flow setpoint for the compressor inlet 202A is determined by the compressor map 500 at a speed and pressure ratio across the compressor 500. The compressor pressure ratio setpoint PR Sp can be calculated as: PRSp=pCmprOutSppCmprInSp where pCmprOutSp the pressure setpoint at the outlet 202B of the compressor 202 is and pCmprInSp is the pressure setpoint at the inlet 202A of the compressor 202. The corrected gas flow setpoint of the compressor inlet 202A is determined by the compressor map 500 as follows: m˙CmprDchrgSpCrtd=MapCmpr(nnCmprCrtd, PRSp) where m˙CmprDchrgSpCrtd is the corrected compressor inlet gas flow setpoint determined from the speed and pressure ratio across compressor 202. The final gas flow setpoint for compressor inlet 202A is converted from the corrected value: m˙CmprDchrgSp=m˙CmprDchrgSpCrtd*pCmprinpref*TrefTCmprinSp where TCmprinSp is the temperature setpoint at the compressor inlet 202A from the recirculation valve 218. The flow value for the recirculation valve 218 is determined using the difference between the final gas flow setpoint for the compressor inlet 202A (m˙CmprDchrgSp) and the stacked cathode flow setpoint (m˙CmprAssmDchrgReq) calculated. m˙RcrcVivsp=m˙CmprDchrgSp−m˙CmprAssmDchrgReq

[0038] The flow setpoint (m˙RcrcVivsp) for the recirculation valve 218 is flattened towards zero when the final gas flow setpoint (m˙CmprDchrgSp) for the compressor inlet 202A is less than the stacked cathode flow setpoint Req (m˙CmprAssmDchrgReq).

[0039] The requested valve kVlv value for the feedforward term of the position of the recirculation valve 218 is calculated based on the valve model for subcritical gas flow as: kVlvCmprRcrcVlvReq,Subcrtcl=a*m˙RcrcVlvspMCaSplyAir*MCaSplyAir*TCmprOut(pCmprOutSp)2−(pCmprinSp)2 and for a critical gas flow calculated as follows: kVlvCmprRcrcVlvReq,Crtcl=b*m˙RcrcVlvspMCaSplyAir*MCaSplyAir*TCmprOut(pCmprOutSp)2 where M CaSplyAir the molecular weight of the gas passing through the recirculation valve 218, m˙RcrcVlvsp is the previously calculated flow diverted by the recirculation valve 218, T CmprOut the temperature at the outlet 202B of the compressor 202 is pCmprOutSp the pressure setpoint at the outlet 202B of the compressor 202 is pCmprInSp is the pressure setpoint at the inlet 202A of the compressor 202, and a and b are the conversion coefficients for the subcritical and critical gas flow units, respectively. In some embodiments, a and b are equal to 4.633 and 5.375, respectively. The maximum of the subcritical and critical gas flow values ​​can then be used as the requested position of the recirculation valve 218: kVlvCmprRcrcVlvReq=Max(kVlvCmprRcrcVlvReq,Subcrtcl, kVlvCmprRcrcVlvReq,Crtcl)

[0040] The final kVIv value of valve 218 is calculated from the feedforward kVlv value calculated above and the feedback kVlv from PID controller 400, and this is converted to the new valve position using a lookup table or other method based on the valve characteristics. The recirculation valve command is then generated and sent over interfaces 406, causing the recirculation valve 218 to change positions. The command is generated by one or both of a feedback-based (i.e., an integral / derivative) value and a feedforward-based (i.e., a setpoint) value, which is particularly well-suited for rapid response to transient flow.

[0041] Using the above valve equations, the predicted position of the recirculation valve 218 is calculated as a feedforward term to the PID controller 400 based on a valve model with input from a predetermined flow, temperature, pressure, and other inputs. In a specific form, the invention uses the information stored in the compressor map 500 to calculate the feedforward term of the position of the valve 218, which, upon command from the controller 400, can cause the valve 218 to move rapidly during down- or up-shift transients compared to a purely feedback-based control strategy.Those skilled in the art will appreciate that the above equations are exemplary, and that other equations may be used to produce the predicted position of the valve 218; such other equations, if used to help predict a future position for the recirculation valve 218 to satisfy the flow setpoint of the fuel cell stack 204, are within the scope of the present disclosure.

[0042] Next, referring to the Fig. 5A, Fig. 5B, Fig. 6A and Fig. 6B are results for a state-of-the-art recirculation valve flow control ( Fig. 5A and Fig. 6A) and the present disclosure ( Fig. 5B and Fig. 6B). Specifically referring to Fig. 5A, differences in the stack flow setpoint 610 and the stack flow feedback 620 appear as errors 630, while the dryout range 640 between the stack flow setpoint 610 and the stack flow feedback 620 is large for a considerable period of time (as shown, approximately 4 seconds); this can lead to dryout of the various cathodes of the stack 204. Referring specifically to Fig. 5B, differences in the stack flow setpoint 710 and the stack flow feedback 720 are immediately resolved; this in turn means that the dryout region 740 is minimized, corresponding to reduced cathode dryout and improved stack performance and durability.

[0043] With particular reference to Fig. 6A, the cathode stack flow setpoint 610 temporarily increases, followed by a rapid decrease. This occurs in conjunction with anode hydrogen bleed events, in which anode water and nitrogen are temporarily "burped" from the system. Similar to Fig. 5A shows Fig. 6A dry-out areas 640 for the cathode, wherein the stack flow setpoint 610 and feedback 620 are separated, while in Fig. 6B, the disclosed invention rectifies the situation by causing the setpoint 710 and the stack flow feedback 720 to correlate much better, thereby reducing the dryout region 740. List of reference symbols for Fig. 4 402 processor 404 Memory 422 System Monitoring 424 Compressor control unit 426 valve control unit 428 valve models 430 pressure drop models 432 parameters 434 Other devices 440 interface devices 442 Other computing systems

Claims

[1] A method for controlling a compressor recirculation valve (218) for a fuel cell system (102) during a down-current operating transient during which the output power of the fuel cell system (102) decreases, the method comprising: receiving a stacked cathode flow setpoint for a gas flow supplied from an outlet of a compressor (202) to a cathode of a fuel cell (204); characterized by , that a recirculation valve flow setpoint for the supplied gas flow is calculated using compressor operating data and the stacked cathode flow setpoint; a feedforward term corresponding to a new recirculation valve position is calculated by using the recirculation valve flow command value; a feedback term corresponding to a new recirculation valve position is generated via a PID control strategy based on an error between the stacked cathode flow setpoint and the feedback term; and a recirculation valve control command is generated based on the feedforward term and the feedback term, which causes the recirculation valve (218) to open during a down-operation transient. [2] The method of claim 1, wherein a compressor inlet gas flow setpoint is determined from a compressor map such that the recirculation valve flow value is calculated using a difference between the compressor inlet gas flow setpoint and the cathode flow setpoint. [3] The method of claim 2, wherein the recirculation valve flow value is calculated using a difference between the compressor inlet gas flow setpoint and the cathode flow setpoint. [4] Control unit (400) for a vehicle fuel cell system (102), comprising: at least one processor (402); and a non-transitory memory (404) in communication with the at least one processor (402), the memory (404) storing machine instructions that, when executed by the at least one processor (402), cause the at least one processor (402) to perform the method of claim 1. [5] The controller of claim 4, wherein the instructions further cause the at least one processor (402) to calculate a difference between the compressor inlet gas flow setpoint and the stacked cathode flow setpoint to determine the recirculation valve flow command value, such that the changed position of the recirculation valve (218) is determined by a feedforward value and at least one of an integral value and a derivative value in the controller. [6] Vehicle fuel cell system (102) comprising: a fuel cell stack (204) having a plurality of fuel cells, each of which comprises a cathode; a compressor (202) having an inlet and an outlet, the outlet fluidly connected to the stack (204) to provide a reactant gas to the cathodes; a recirculation valve (218) operatively connected to the inlet and outlet of the compressor (202); and a control device (400) having at least one processor (402) and a non-transitory memory (404) in signal communication with the at least one processor (402), the memory storing instructions which, when executed by the at least one processor (402), cause the at least one processor (402) to perform the method according to claim 1. [7] The vehicle fuel cell system of claim 6, wherein the instructions causing the at least one processor (402) to calculate a recirculation valve flow command further comprise instructions causing the at least one processor (402) to: calculates a feedforward term corresponding to a new recirculation valve position by using the recirculation valve flow command; and a feedback term corresponding to the new recirculation valve position is generated via a PID control strategy based on an error between the stacked cathode flow setpoint and the feedback term. [8] Vehicle fuel cell system according to claim 6 or 7, further comprising a data transmission device (408, 410) configured to transmit the recirculation valve control command between the control unit (400) and the recirculation valve (218).

Images