Creating a control arrangement with constraints
As we noted previously it is important to have a good measurement of the air flow through the system. We need to be able to accommodate the design criteria and respect the system limits. We will need to compensate for the inlet temperature but will not actively compensate for the barometric pressure. The following diagram will remind us of the process arrangement. We will also assume that the process is grinding coal for a combustion process.
We generate a control block arrangement for measuring the air flow with temperature compensation.
Now we need to add it to a controller with set point demand and limiters. Note that there are two dampers and we need to increase or decrease both the hot and cold air dampers at the same time to provide the corresponding increase or decrease in air flow.
Let us define the relationship between coal flow and the air flow (set point). We need to respect the minimum air flow to prevent fall-out in the coal pipes. We also need to move the air flow from minimum air flow to design air and coal flows through the mill. There are essentially three ways this can be accommodated in the set point curve.
The first way is to continuously maintain the air flow at the design air flow through the system at all fuel loads. This has the problem that it relies on the mill storage equilibrium to reflect any changes from the input to the output. Dynamically this is not a useful approach when we need to regulate the flow of fuel to the furnace. This approach will be discarded for our pulverized coal application. (If, however, we were delivering pulverized glass or flour to a baghouse/receiver for further processing, this may be a useful approach.)
The next way is to continuously move the air flow from minimum through design air flow as the feeder moves from minimum coal flow to design coal flow over the whole active range.
The final way is to hold the air flow at the minimum position until the fuel demands more air to keep the air/solids ratio balanced.
The chart below reflects each of these approaches.
For our purposes we will select the curve that begins to move as the coal feeder moves from minimum coal flow (orange line). This will allow us to have a little margin above minimum air flow and the design air flow/coal flow ratio to accommodate any dynamics that we need to enable. This curve will populate the function generator [f(x)] from the feeder flow to the air flow setpoint. The operator will have the ability to bias the air flow setpoint (within limits) to accommodate minor changes in the system process. The resulting damper demand will go to the final elements.
We have a demand for the flow and now we need to control the temperature entering the pulverizer to achieve the proper temperature air/solid mixture exiting the pulverizer.
If we look at the process diagram we again note that there are only two dampers on the primary air system. Any change in one damper to adjust outlet mixture temperature would require the other damper to move in the opposite direction to maintain the desired air flow. So if the hot air damper position is increased the cold air damper would be decreased by a comparable amount.
We also have an opportunity to enhance and maintain the outlet temperature during pulverizer loading increases and decreases. Since these pulverizers generally have a large thermal mass in addition to the recirculating storage mass, the discharge temperature will respond slowly to changes in the temperature at the inlet. For any pulverizer loading we can characterize the inlet temperature required to maintain the outlet temperature. Also as the temperature demand at the inlet is changed the actual inlet temperature will respond relatively quickly.
By applying some simple thermodynamic energy balances we can model the inlet temperatures required to dry the material. This will allow us to anticipate the needed inlet temperature to maintain the proper coal-air discharge temperature. We can also go one step further and model the damper demand position needed to obtain these temperatures. Any errors in the modeling will be corrected by the outer loop temperature control of the discharge temperature of the air-coal mixture.
Combining both the flow control and temperature control provides a complete control arrangement for the pulverizer as noted below.
Since there have been some questions regarding psychometrics, I will discuss this in the next submittal. This has occurred both in cooling towers and baking/drying pasta. The next submittal will cover some psychometrics basics.
Basic humidity made easy