BACKGROUND OF THIS INVENTION
Multiburner Furnaces provide heat and energy. With recent improvements in furnace design the ratio of combustants to oxidants yield lower levels of flue carbon monoxide (CO) in a less fuel rich burn, producing less ash and greater efficiency; less flue nitrogen monoxide (NO) in a less fuel lean burn, producing less pollution; and, flue temperature balancing CO and NO. Day to day use may undo these improvements at a cost. It is desirable to have a device available, which automatically controls and prolongs these improvements.
BRIEF SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method and apparatus to control CO, NO, or temperature in the flue of a multiburner furnace to produce and deliver appropriate oxidants to the combustants at the burners to increase efficiency and decrease pollution. It is a further object of this invention to provide a device which will prolong all improvements.
In carrying out the above objects and other stated objects and features of the present invention a method and apparatus is provided as an Automatic Furnace for maintaining a desired CO, NO, or temperature range at the flue (referred to as flue parameters) of a multiburner furnace. and includes delivering a first oxidant (oxygen or air) dose to the combustant/oxidants at the burners of a multiburner furnace of any design producing a sequential flue parameter dose selected from one of a plurality of sequential flue parameter doses between a first flue parameter dose and a second flue parameter dose. The method includes delivering a second oxidant dosage to the burner while repeatedly sequencing through the plurality of sequential flue parameter doses beginning with the first flue parameter dose and proceeding to an adjacent flue parameter dose in the sequence after a predetermined time interval has elapsed. The second oxidant dosage is delivered until the flue parameter level attains the desirable range, at which point corresponding oxidant doses and flue parameter doses are selected from the plurality of oxidant doses and the plurality of sequential flue parameter doses. The method also includes delivering the selected oxidant dose and flue parameter dose so as to maintain the desired flue parameter range.
In the preferred embodiment the method and apparatus employs CO as the sole flue parameter. The other flue parameters may be employed as well.
The advantages of the Automatic Furnace are minimal needs for furnace shut downs, less pollution, more efficiency, and a reduction in the cost of running and maintaining a multiburner furnace.
The above objects, features, and other advantages will be readily appreciated by one of ordinary skill in the art from the following detailed description of the best mode for carrying out the invention, when taken in connection with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1/6 demonstrates a perspective view of the first embodiment of the present invention.
FIG. 2/6 is a graphical demonstration of the flow charts of the Automatic Furnace.
FIGS. 3/3-5/6 are flow charts dealing with the oxidant dosage and CO dosage strategy of the present invention for use in the Automatic Furnace.
FIG. 6/6 is a flow chart for relating parameters in the Automatic Furnace.
DETAILED DESCRIPTION OF THE INVENTION
Referring now to FIG. 1/6, a first embodiment of the present invention is shown. This embodiment indicated by reference number 1 in FIG. 1/6 is the best mode in implementing this invention and is particularly suited for use as an Automatic Furnace. FIG. 1/6 includes 2, multiburners; 3, combustants (solid, liquid, or gas): 4, oxidants (oxygen or air); 5, a furnace flue; 6, a multiburner furnace; 7, a flue parameter sensor in Vol % or degree Centigrade; 8, a band pass filter; 9, the ECU; 10, multiple variably opening solenoid valves; and,11, the oxidant entrance.
In response to flue parameter sensor 7 data, oxidant flow rates at the inlets 11 are controlled by an ECU 9 controlled variably opening solenoid valve 10 with Coulomb controlling circuits, as was taught in 877 and U.S. Pat. No. 5,008,773. They enhance or restrict combustion at the burners 2.
Referring now to FIG. 2/6, the method of device function is demonstrated graphically. The flue parameters are placed on the ordinate and time or oxidant dose are placed on the abscissa of a Cartesian plane. Maximum oxidant dosage occurs at tr on the abscissa, the significance of which will be presented later. Measured and calculated logarithmic functions are used in the preferred embodiment as flue parameter dosages, but any measured and estimated transcendental function with an inverse may be used.
Referring again to FIG. 1/6, as will be seen, conditions on CO—the preferred flue parameter—control oxidant flow rate 11 and thus CO dosage.
Referring now to FIG. 2/6, the illustrated method of oxidant dosage and CO level selection starts with the administration of an extreme oxidant gas flow rate—herein referred to as the selector dose of the oxidant gas flow rate which produces the maximum or minimum CO dosage—as in curve A or B. Curve A is represented by y=log to the base a of x. Curve A activates at x=0.
Line CG is the desired CO level—herein referred to as the selection parameter, which is a range in the actual device. At the intersection of line CG and curve A or B (call it X), line D points to point E on the abscissa as the selected oxidant dose. This is determined by graphical means and, as will be seen, the flow charts. The virtual CO dosage in Vol % is curve F, which activates at point E, the selected oxidant flow rate, and is boosted by curves A, B, H—an overshoot of curve A—and curve I—a deactivation of curve H—to produce line G, which is the selected CO level, is also a CO dosage, and is represented by y=log to the base b of tr, where tr is the t value of the flattening out of the logarithm y=log to the base b of t (curve F) at tr seconds. Line G is completely determined by the intersection (X) described above and in the flow charts, as will be seen, thus the determination of curve F and line G by the above methods is unnecessary. Curve F and line G start in the x coordinate system at x=t and in the t coordinate system at t=0, when curve A deactivates. Curve F and line G deactivate when curve A activates. Curve J is the virtual curve of curves A and H. K marks the Circulation time. It marks the time from the initial oxidant gas flow rate to the first recording of any change in the CO dosage or level. Its accuracy is essential for proper flow chart function with respect to time. Its calculation and that of tr will be demonstrated. The oxidant dose is circulation time dependent. The CO dose is not, since it is a function of time.
Before describing the flow charts it is useful to explain the terminology employed. The most recent base state keeps CO in its desirable range. The oxidant flow rate and CO level are measured in all states. The washout state washes out overshoots. It also determines the selected CO dose and oxidant flow rate, as will be seen. CO doses are functions of oxidant flow rates.
Referring now to FIGS. 3/6-5/6, flow charts are shown, which illustrate the system and method for the proper selection of oxidant flow rates and CO doses.
Referring to FIG. 3/6, Step 400 determines various system parameters, which may be predetermined and stored in memory, calculated by an ECU (such as ECU 9 in FIG. 1/6) or entered by a system operator. The system parameters include the following:
- MIN R=minimum dose of oxidant flow rate given for each range.
- MAX R=maximum dose of oxidant flow rate given for each range.
- CO=level in Vol %
- TCO1=desired CO level.
- dL=low CO level threshold.
- dH=high CO level threshold.
- Tss=series state delay time.
- Tcirc=circulation delay time.
- Twash=washout delay time.
- tr=desired response time or reaction time
- The value of dH and dL are determined by the current operating state.
As shown in FIG. 3/6 the ECU now passes control to Step 402, which measures the oxidant flow rate and CO level. At Step 404 a maximum oxidant dose of the last range is administered. This is represented graphically by curve A of FIG. 2/6 and is called the selector dose. It represents the maximum oxidant dose. The possible CO level is set for the lowest level of the lowest range.
With continuing reference to FIG. 3/6 at Step 406 the oxidant dose is maintained while pausing Tcirc seconds, then x is set to 0 seconds. Step 406 is called an adjustment state. It coordinates the flow charts with respect to time. Initial circulation times may be estimated or measured.
Referring once again to FIG. 3/6 the ECU passes control to Step 408, which continues to deliver maximum oxidant dosage to the burners. Step 408 is referred to as a series state—Tss—and is necessary to coordinate the progression through various possible CO levels within a time period determined by tr. The calculation of Tss depends on the current operating state. Some representative calculations are illustrated in FIG. 6/6 for a single ranged implementation as discussed in greater detail below.
Still referring to FIG. 3/6 a test is performed at Steps 409 and 410. It asks—is CO greater than dH?—and, is CO less than dL?, respectively. They split control into three pathways. Negative answers to both conditions direct control to Step 426, where 1. The definitive current CO level is set to the possible level, while the preliminary oxidant dose is set one circulation time into the future. 2. A pause for the circulation time takes place. Then, 3. t is set to 0. This represents preliminary oxidant dose and definitive CO dose selection.
Now referring to FIG. 4/6 processing continues with the ECU directing control to Step 428, which pauses to washout high valued functions from the selected dose. The state is completed when all involved functions equal a straight line—the selected CO level or dose. For convenience in the representation of the method in the flow charts the ECU was represented to set t=0 in Step 426. This actually occurs at the start of the washout state. The ECU directs in the washout state the determination of the selected value of point E of FIG. 1/6—the definitive selected oxidant dose—then activates this dose. The CO dose remains the selected dose as line G in FIG. 1/6. Both of the above dosages continue until activation of MIN R or MAX R. FIG. 430 measures CO values for the Conditions below. Steps 409 and 410 represent a second test and ask the same questions as the above mentioned first test—Is CO greater than dH or less than dL, respectively? If either answer yes, control is directed to Steps 431 and 434, respectively, where a predetermined fraction of tr is either subtracted or added, respectively to tr. This pathway determines tr only if the circulation time is correct. The circulation time is calculated by keeping the last three base state values in memory. When control is directed to or beyond a noncontiguous base state from which control was originally assumed a predetermined amount of time is added to the circulation time. This will correct abnormally short circulation times. For abnormally long circulation times—if control passes consecutively to two ascending or descending base states, a predetermined amount of time is subtracted from the circulation time.
Referring now to FIG. 5/6, if both conditions in the second test answer no, the ECU places control in Step 436, the base state. Steps 438 and 440 represent the third test and ask the same questions (is CO>dH or <dL?) as those of the previous tests with different consequences. They determine the stability of the base state (both conditions answer no if it is stable). If it is unstable, the ECU directs control to either Step 463, if Step 438 answers yes, or 446, which 1. Minimizes or maximizes the current dose, respectively 2. Pauses for the circulation time, then 3. Sets x=0. These doses continue until dose selection. It should be noted that Steps 431, 434, the yes part of 418, and the no part of Steps 433 and 440 all yield control to Step 436, the base state. The ECU then directs control from Step 463 to Step 411, and from Step 446 to Step 412.
Referring again to FIG. 3/6, the ECU directs control from Step 464 (evaluated later), and if Step 414 in FIG. 4/6 (the first condition of fourth test to be elucidated soon) answers no, to Step 408 to maintain the current CO dose for Tss. Control is then directed to Step 409, which, if along with Step 410—the first test—the answer is yes to both conditions, control is passed to Steps 411 and 412, respectively, which decrement and increment the possible dose, respectively, then both pass control to Condition 414.
Referring now to FIG. 4/6, Steps 414 and 418 represent the fourth and final test with different conditions than the other tests. These conditions ask if the present possible dose is the last dose available, and if the present range is the last one available, respectively. If Step 414 answers no, control is directed by the ECU to Step 408 in FIG. 3/6, which maintains a current dose for Tss. If the condition answers yes, control is directed to Step 418, which determines if the present range is the last range available. If it answers no, control is directed to Step 464, in which control enters a new range, sets the current oxidant and CO dose to MAX R or MIN R of the new range, pauses for the circulation time, then sets x=0. Control is then directed to Step 408, which maintains a current oxidant and CO dose for Tss. If Step 418 answers yes, the ECU directs control to Step 436, the base state.
Referring now to FIG. 6/6 a flow chart is shown illustrating representative calculations of Tss according to the present invention. One of these calculations or an analogous calculation is performed for each series state of FIGS. 3/6-5/6, such as illustrated at Steps 408, 411, and 412.
Returning to FIG. 6/6 at Step 480 a test is performed to determine if the system has reached a base state. If not, the series state delay is estimated as: Tss=tr/IR. If the result is true, the process continues with Step 484, where a test is performed to determine whether CO<dL. If true, then Step 486 determines whether the most recent base state is a minimum for the current range. If it is true, the series state delay is calculated by Step 488 as Tss=tr/(IR−1). Step 498 then returns control to the series state which initiated the calculation.
With continuing reference to FIG. 6/6, if the test at Step 486 is false, then the series state delay is calculated by Step 490 as Tss=tr(MAX R−MIN R)/(IR−1)(MAX R−BASE STATE) before control is released to the series state via Step 498. If the test performed at Step 484 is false, then Step 492 performs a test to determine if the most recent base state is the maximum for the current range. If the result of Step 492 is true, then Step 496 calculates the series state delay as Tss=tr/(IR−1). Control is then returned to the appropriate series state via Step 498. If the result of the test at Step 492 is false, then the series state delay is calculated by Step 494 as Tss=tr(MAX R−MIN R)/(IR−1)(BASE STATE−MIN R). Step 498 then returns control to the appropriate series state. FIG. 6/6 applies to a single range. One of ordinary skill in the art should appreciate that the calculations may be modified to accommodate a number of possible ranges.