Control Strategies

In order to enable safe and convenient production of electricity and heat from a PEM fuel cell the control subsystem associated with it has to accomplish roughly four main tasks listed below.

  1. Reactant supply management
  2. Stack water management
  3. System temperature management
  4. Safety monitoring and control

In the systems constructed, the anode reactant supply is controlled electrically only by a solenoid shutoff valve. Pressure regulation is done my a manually adjustable regulator, which tries to keep the pressure on anode subsystem constant. This regulator is not electrically connected to the control system. Cathode supply is controlled by a by-pass type blower, which is controlled in a load-following manner based on current measurement form the fuel cell stack electric current output terminal. The control routine calculates the amount of air required to produce the current needed plus a certain excess ratio, which is practically needed to avoid the so-called oxygen starvation effect. Stack water management involves in one way or the other, monitoring the water balance in the system and initiating measures for excess water removal from the anode and cathode channels in the stack. In the pp8 (& pp16) systems the indicator for degrading cell performance is the individual voltage measurement from each of them. When one or more cell voltage drops below certain limit or deviate far enough from the average of all cell voltages, the system triggers a procedure to recover the performance level. This consists of briefly opening a valve at the normally closed ("dead-end") anode exchaust followed by briefly feeding high amount of excess air through the cathode channels. Other motivation for purging besides removing liquid water from the system is to counter the accumulation of inert gases, like nitrogen and impurities of the supplied hydrogen into the normally closed anode system. Temperature management routine monitors the PEMFC stack inlet and outlet temperatures, initiates cooling measures (controlling fans, valves etc.) and in case of overheating, trigger a controlled shutdown of the system. One of the most obvious safety risks when working with a gases like hydrogen is it's leakage into immediate surrounding and formation of a potentially explosive atmosphere. For this reason a hydrogen leak detecting device has been included in the system and a control routine with safe system shutdown limit has been implemented. Also, the coolant flowrate monitoring is included in order to prevent damage to the stack (and through that, a possible hydrogen leak) in case of air pockets blocking the flow.

System control strategy In it's current implementation, the hybrid system control strategy is solely based on the momentary load demand from the load. This combined with passively coupled system topology leads to severe limitations in energy flow control between the individual energy storages. Power flow control in it's current form consists of only three high current relays which can be used to

  1. Disconnect the fuel cell from the system in case of failure.
  2. Disconnect the ultracapacitor modules from the main bus in case of damaging overvoltage (or undervoltage).
  3. Connect the brake resistors integrated into the pp16 power source to main bus in case of high regenerative energy peak from the load.

The last two of these relays are connected to a separate PLC that monitors the main bus voltage, in order to assure fast reaction rates to dangerous conditions. The fuel cell relay is operated by the NutDAC control system. The fuel cell system control routines were developed utilizing the state machine approach (two figures below) and were manually transferred into C in an automatic control loop which is executed continuously while the system is running.

FSA presentations of the control software main routine The main routine consists of four states, which are stand-by, start-up, running & shutdown. While in stand-by, the control system is powered and it is waiting for a command from the user to start procuding electricity. When commanded, the software moves to start-up phase, during with basic system checks are performed and preparations made before allowing current draw from the fuel cell stack. The "running" state is the normal operating mode where the automatic control loop is executed. In shutdown state, the load is disconnected from the fuel cell and required preparations are performed before returning to stand-by mode. In addition there is a fifth state, called "emergency shutdown", which is mainly frequented when an unexpected error is detected within the system.

FSA presentations of the control software run mode As an example, one of the lower level states, called "running" is shown in the figure above. Subtasks in this routine include monitoring of cell voltages, cathode blower control, temperature monitoring and checking a hydrogen sensor reading to detect possible leaks in the system. In case the transition conditions to one of the normal states in the loop fails, the execution is halted and the main application routine is signalled of the problem encountered.

System control strategy

In it's current implementation, the hybrid system control strategy is solely based on the momentary load demand from the load. This combined with passively coupled system topology leads to severe limitations in energy flow control between the individual energy storages.

Power flow control in it's current form consists of only three high current relays which can be used to

  1. Disconnect the fuel cell from the system in case of failure.
  2. Disconnect the ultracapacitor modules from the main bus in case of damaging overvoltage (or undervoltage).
  3. Connect the brake resistors integrated into the pp16 power source to main bus in case of high regenerative energy peak from the load.

The last two of these relays are connected to a separate PLC that monitors the main bus voltage, in order to assure fast reaction rates to dangerous conditions. The fuel cell relay is operated by the NutDAC control system.

The fuel cell system control routines were developed utilizing the state machine approach (two figures below) and were manually transferred into C in an automatic control loop which is executed continuously while the system is running.


FSM representation of the main control routine

The main routine consists of four states, which are stand-by, start-up, running & shutdown. While in stand-by, the control system is powered and it is waiting for a command from the user to start procuding electricity. When commanded, the software moves to start-up phase, during with basic system checks are performed and preparations made before allowing current draw from the fuel cell stack. The "running" state is the normal operating mode where the automatic control loop is executed. In shutdown state, the load is disconnected from the fuel cell and required preparations are performed before returning to stand-by mode. In addition there is a fifth state, called "emergency shutdown", which is mainly frequented when an unexpected error is detected within the system.


FSM presentations of the control software run mode

As an example, one of the lower level states, called "running" is shown in the figure above. Subtasks in this routine include monitoring of cell voltages, cathode blower control, temperature monitoring and checking a hydrogen sensor reading to detect possible leaks in the system. In case the transition conditions to one of the normal states in the loop fails, the execution is halted and the main application routine is signalled of the problem encountered.