US3597518A

US3597518A - Electric arc furnace control

Info

Publication number: US3597518A
Application number: US15035A
Authority: US
Inventors: Roland W Roberts
Original assignee: Robicon Corp
Current assignee: Robicon Corp; Sanwa Business Credit Corp
Priority date: 1970-02-27
Filing date: 1970-02-27
Publication date: 1971-08-03
Anticipated expiration: 1988-08-03

Abstract

A control system for electric arc furnaces incorporating proportional plus integral control characteristics. The system, which is designed to maintain constant arc impedance, develops an output voltage proportional to input impedance error and then adds a second voltage which increases with time at a rate proportional to the impedance error magnitude. When the combined voltages for a furnace electrode exceed a preset level, the drive motor for that electrode is actuated to move the electrode upwardly or downwardly to maintain the desired arc impedance. In this way, the system will respond quickly to large impedance errors, is insensitive to short term impedance fluctuations, and at the same time has a very high sensitivity to continuing small errors.

Description

United States Patent Primary Examiner-Harold Broome Assistant Examiner-R. N. Envali, Jr. Attorney-Brown. Murray, Flick & Peckham ABSTRACT: A control system for electric arc furnace: incor poratin; proportional plus integral control ChlflClCliliiCL The system, which is designed to maintain constant are impedance. develops an output voltage proportional to input impedanca error and then adda a second voltage which increnscs with time at a rate proportional to the impedance error mag nitude. When the combined voltagea for a furnace electrode exceed a preset level. the drive motor for that electrode is actuated to move the electrode upwardly or downwardly to maintain the desired are impedance. In this way. the ryrtem will respond quickly to large impedance errors, is insensitive to short term impedance fluctuations, and at the same time has a veryhigh sensitivity to continuing small erron.

INTI! TIM POL AND M ROBERTS Altorneys liilLli'lCTlltllC ARC 1F URNACE CONTROL BACKGROUND OF THE INVENTION In the usual control of an electric arc furnace, arc voltage and current are measured on each electrode in servosystems and used to develop error signals proportional to any deviation from desired arc impedance. If the arc voltage signal is V and the current signal is KXI, then by subtracting the two in opposed polarity relationship to develop an error signal, a zero error signal means that Vl'(1=0, or V/l=i the arc impedance. Since W1 is the arc impedance, the system will move the electrodes as necessary to maintain constant arc impedance at a set value.

Constant arc impedance, in turn, means that the power factor is maintained at a set value. At constant line voltage, the power input to the furnace is also constant. By adjusting the impedance control, the furnace operator can also set the power factor and KVA input.

in the past, control systems of this type usually employed a balanced beam relay for the purpose of comparing the signals proportional to V and K1, the arrangement being such that when the two did not balance, the beam of the relay would move in one direction or the other to cause the drive motor for the electrode to move it upwardly or downwardly until the system was again balanced. Instead of using a balanced beam relay for controlling the electrode positioning motors, proportional control was sometimes-provided by means of proportional controllers such as Amplidynes (Trademark) or Rototrols (Trademark).

In the prior art control systems of this type, fluctuations in arc impedance which occur rapidly under startup conditions, for example, will cause the system to move the electrodes through short increments at a rapid rate, causing extensive contactor wear as well as wear on mechanical components of the system. Slowing the response of the system to eliminate this condition will prevent rapid movement of the electrodes in response to large error signals. Of course, the balanced beam relay often used in such systems carries with it the attendant problems of any relay system, namely the possibility of dirty or sticking contacts.

SUMMARY OF THE INVENTlON in accordance with the present invention, a solid-state control system for an electric arc furnace is provided which overcomes the aforementioned and other difficulties of prior art systems. Specifically, this is accomplished by applying an error signa, proportional to a deviation from desired arc impedance, to a proportional plus integrating controller. One such controller is provided for each electrode which is controlled independently of the others. The output of each controller is the sum of a step or proportional function and a ramp function. When the output of the controller exceeds a predetermined maximum value, the drive motor for its associated electrode is actuated to move it upwardly or downwardly until the impedance error is again zero. When a large impedance error occurs, the proportional characteristic of the controller causes its output to rise immediately to the aforesaid predetermined maximum value, and initiate corrective action on the electrode. On the other hand, continuing small errors of a given polarity will not cause a series of discrete corrective actions, but rather will cause the output of the controller to rise according to the time integral of these errors until its output exceeds the aforesaid predetermined magnitude, at which point corrective action will be taken. Short term random impedance fluctuations of varying polarity will have no effect on the system if they are not of sufficient magnitude to cause the controller to reach the predetermined trip point immediately since they are integrated to zero over a period of time.

The above and other objects and features of the invention will become apparent from the following detailed description taken in connection with the accompanying drawings which form a part ofthis specification and in which:

FIG. 1 is a schematic block diagram of an electric arc furnace control system employing the principles of the invention;

FIG. 2 is a detailed schematic circuit diagram of the integrating error regulator of the invention as applied to a single electrode, it being understood that the systems for the other two electrodes are identical; and

HO. 3 is a graph illustrating the proportional plus integrating output characteristics of the regulator of FIG. 2.

With reference now to the drawings, and particularly to FIG. i, the system shown includes three input terminals A, B and C adapted for connection of the three phases of a threephase power supply, not shown. The terminals A, B and C are connected through a three-phase transformer 10 to the three electrodes 12, 14 and 16 of an electric arc furnace 18.

The furnace 18 includes a melting chamber or bath 20 which contains a molten pool of metal 22. Normally, electrodes ll2-16 are spaced from the surface of the molten bath 22 so as to establish arcs between the two, the heat of the arcs serving to melt the molten metal and maintain it in a molten condition. Ordinarily, in starting a melt, scrap is placed within the bottom of the melting chamber 20 and the electrodes 12-- 16 lowered until an arc is struck. During the initial melting period, the electrodes 12-16 will move upwardly and downwardly rapidly until a sufficient pool of molten metal is established to maintain an arc of the desired impedance characteristics.

Connected between the top of each electrode and ground is a

resistor

24, 26 or 28 such that the voltages appearing across the

resistors

24, 26 and 28 are proportional to the arc voltages. These voltages are applied through

potentiometers

30, 32 and 34 and calibrating

potentiometers

36, 38 and 40 across

resistors

42, 44 and 46 respectively. The voltages across

resistors

42, 44 and 46, which are proportional to the arc voltages established by the three electrodes 12, i4 and 116, are applied through

transformers

48, 50 and 52 to the input of integrating error regulator 54, hereinafter described in detail.

Arc currents are sensed by

current transformers

56, 58 and 60 and applied across three

resistors

62, 64 and 66 which develop voltages proportional to arc current. These voltages are applied through three potentiometers 68, 70 and 72 and

transformers

74, 76 and 78 to the integrating error regulator 54.

The integrating error regulator is actually divided into three separate circuits A-ll, B-ll, and C1, each circuit being adapted to control an associated drive motor for one of the electrodes 12, 14 or 16. All of the circuits A-ll, B-1 and C-1 are identical in operation and only one will be described herein in detail with reference to FIG. 2. EAch portion A-l, B-ll or C! of the integrating error amplifier 54 is connected to an associated

motor control circuit

80, 82 and 84 which, in turn, controls an associated

drive motor

86, 88 or 90 for the three electrodes 12, t4 and 16, respectively. As will be seen, the

motor control circuits

80, 82 and 84 may be of the relay type wherein the motors are caused to move upwardly or downwardly at a fixed speed when a relay is energized, or may be of the proportional type wherein the motors are driven upwardly or downwardly at a rate proportional to the magnitude of the error signal.

With reference now to FIG. 2, one of the three sections of the integrating error regulator 54 is shown, it being understood that the other two portions are identical and control their associated drive motors independently of the others. It will be assumed that the circuitry of FIG. 2 controls the drive motor 86 for electrode 12. The signals developed across the

transformers

48 and 74 which are proportional to are voltage and arc current, respectively, are applied to full-

wave bridge rectifiers

92 and 94. Consequently, a voltage proportional to are voltage will appear across resistor 96; while a voltage will appear across resistor 98 proportional to are current.

Smoothing capacitors

100 and 102 are connected in shunt with the

resistors

96 and 98, as shown, The midpoint between

resistors

96 and 98 is connected to ground; while the opposite ends are connected through

equal resistors

104 and 106 to a summation point 108. In this manner, when the voltages across

resistors

96 and 98 are equal, the voltage at point 108 will be zero and will not deviate from zero until either the arc voltage or are current changes, indicating a change in are impedance.

The summation point 108 is connected to one input of an operational amplifier 110, the other input of which is connected to ground through resistor 112. The amplifier 110 is provided with a feedback path consisting of capacitor 116 in series with resistor 113. In this manner, the output of the amplifier 110 comprises a step or a proportional function followed by an integrating or ramp function. This will be explained more in detail hereafter.

The output of operational amplifier 110 is applied to an absolute value stage 120 comprising an operational amplifier 122 having one input connected to the output of operational amplifier 110 and its other input connected to ground through resistor 124. Amplifier 122 is provided with two feedback paths, one of which includes diode 126, and the other of which includes diode 128 in series with resistor 130. The absolute value stage 120 ensures that the voltage on lead 132 will always be of one polarity, namely negative, regardless of the polarity at the output of operational amplifier 110. ln this respect, the value of resistor 134 is substantially twice that of resistor 136.

If the signal at the output of amplifier 110 is positive, the signal at the output of amplifier 122 is negative; and current flows through diode 123 and resistor 130. The voltage on lead 132 is now negative by virtue of the fact that resistor 136 is one-half the size of resistor 134. On the other hand, if the output of amplifier 110 is negative, the output of amplifier 122 is positive and current is bypassed through feedback diode 126.

Under these circumstances, no voltage is developed across resistor 136; however the negative voltage on lead 132 is equal to what it would be if the signal at the output of amplifier 110 were positive. This is so by virtue of the fact that resistor 134 is double in size of resistor 136.

The negative signal on lead 132 is applied to the input of a third operational amplifier 138 which is operated without feedback at very high gain. Also applied to the input of amplifier 138 is a positive bias signal applied through resistor 144 and normally closed contacts RY 1-1 of relay RY1. The positive signal applied to the input of amplifier 138 through contacts RY1-1 will be greater than the negative signal on lead 132 until it is desired to actuate the drive motor for an electrode. Before the drive motor is actuated, the net positive input signal to amplifier 138 produces a negative output signal which is bypassed through diode 146 and resistor M8 to ground. However, when the negative signal on lead 132 exceeds the positive bias applied through contacts RY1-1, the output of operational amplifier 138 becomes positive, and relay RY1 is energized. When relay RY1 is energized, normally closed contacts RY1-1 open, thereby removing the positive bias until the signal on lead 132 falls to zero, the output of the amplifier 138 becomes zero and the relay RY1 becomes deenergized.

When relay RY1 is energized, as when the negative signal on lead 132 exceeds that from the positive bias source, it closes contacts RY1-2 to connect lead 132 to a source of positive potential through resistors 150 and 152, these resistors having a much higher resistance value than resistor 144. At the same time, energization of relay RY1 closes contacts RY1-3 to provide a shunt path around integrating capacitor 116 through resistor 154. Finally, when relay RY1. is energized, it closes contacts RY1- 1 to connect the output of operational amplifier 110 to a relay control circuit 156 which causes either up relay RY2 or down relay RY3 to become energized, depending upon the polarity of the signal at the output of operational amplifier 110.

Operation of the circuit can best be understood by reference to FIG. 3 showing the output of operational amplifier 110. Assuming the existence of a continuing step error signal, the output of the operational amplifier is a continuing proportional step function indicated by the vertical portion 158 of the curve, plus an output proportional to the time integral of the error function as indicated by the line 160. Whenever the output of operational amplifier 110, as transferred in absolute value by operational amplifier 122, exceeds the bias level at point indicated by the line 162 in FIG. 3, the relay RY1 will become energized, thereby causing relay control 156 to energize up relay RY2 or down relay RY3 of motor control circuit 80, depending upon the polarity ofthe signal at the output of the amplifier 110. As explained above, however, the signal on lead 132 which is compared with the positive bias on point 145 is always negative. If the deviation in impedance from a desired impedance is relatively slight, the output of amplifier 110 may appear as the full line curve of FIG. 3 wherein the signal level output increases along line 158 abruptly and then increases along the integrating or ramp portion of the curve until the voltage level at 162 is reached, whereupon relay RY1 is tripped.

However, if the deviation in impedance from a desired impedance is large, as it might be under startup conditions, the proportional function of the amplifier 110 will cause the signal output level to rise abruptly as indicated by the dotted outline 163 in FIG. 3, again followed by integration for a continuing error. In this case, the relay RY1 will be triggered at time 1 which, of course, is a much shorter time elapse from the starting time 1, than the time elapse between t, and for the case of a plurality of slight variations in impedance. Thus, large deviations in arc impedance from the desired value will cause the relay RY1 to trip rapidly; whereas a series of small deviations of one polarity will be integrated into the slowly rising ramp function which reaches the bias level 162 at a much later time. Momentary small deviations in arc impedance of varying polarity will have no effect since their integrated value is zero. A continuing average error signal, however, will ultimately be integrated to a voltage equal to the trip level 162 and corrective action will be initiated.

Once the relay RY1 trips, it will remain energized until the output of amplifier 110 is zero, indicating that the arc im pedance is at the desired value. This, of course, is accomplished by actuating either relay RY2 or relay RY3 to cause the electrode to move upwardly or downwardly and bring the arc impedance to a point where the voltage at summing point 108 is again zero.

When relay RY1 trips, it closes contacts RY1-3 in shunt with integrating capacitor 116. Amplifier 110 now becomes a proportional amplifier whereby any change in the input to the amplifier will be reflected immediately at its output, and the corrective action in positioning the electrodes will take place rapidly.

Instead of using a relay system such'as that incorporating up and down relays RY2 and RY3, a proportional control can be used such as that indicated by the reference numeral 164 in FIG. 2. Regardless of the magnitude of the error, the motor will be actuated immediately, at full voltage instead of building up current in its windings for a period of time before it starts as is characteristic in proportional motor controls. This is so by virtue of the fact that capacitor 116 is discharged through resistor 154 by closure of contacts RY1-3 at the beginning of the corrective action. As capacitor 116 discharges through resistor 154, the output of amplifier 110 and consequent output from proportional control 164 decrease from full output to a value proportional to the instantaneous input error signal. Proportional action continues until the input error is reduced at zero at which time the system is reset and begins a new cycle of integrating action.

Although the invention has been shown in connection with a certain specific embodiment, it will be readily apparent to those skilled in the art that various changes in form and arrangement of parts can be made to suit requirements without departing from the spirit and scope of the invention. In this respect, it will be apparent that the system of the invention can be used equally well for consumable electrode furnaces, in which case only one electrode and one section of the integrating error regulator will be employed. It is equally apparent The resultant voltage is proportional to arc voltage and is subv stantially independent of circuit inductance.

I claim as my invention: 1. In combination with an electric arc furnace of the type in which at least one electrode is positioned adjacent a molten pool of metal and a potential gradient established between the two to form an are; a system for controlling the position of said electrode with respect-to said molten pool of metal so as to maintain a desired arc impedance, said system comprising means for generating an error signal which varies as a function of a deviation in arc impedance from said desired impedance, a proportional plus integrating controller, means for applying said error signal to the input of said controller, means for comparing the output of said controller with a reference potential, means for moving said electrode toward said molten pool when the magnitude of said output exceeds the magnitude of said reference potential and is of one polarity, and means for moving said electrode away from said molten pool when the magnitude of said output exceeds the magnitude of said reference potential and is of the other polarity.

2. The systemof claim 1 wherein the output of said controller is a step function followed by a ramp function.

3. The system of claim 1 wherein the means for generating said error signal comprises means for developing a first signal which varies as a function of arc voltage, means for developing a second signal which varies as a functional arc current, and means for comparing said first and second signals to develop said error signal.

4. The system of claim 3 wherein said first signal is proportional to are voltage V and said second signal is proportional to K! where I represents arc current and K represents the desired arc impedance, said first and second signals being of opposite polarity, and said comparing means subtracts said first and second signals to derive said error signal. A

5. The system of claim 4 wherein said first and second signals are produced by rectifying alternating current voltages derived from across said are and from current transformers coupled to a conductor supplying current to said electrodes.

6. The system of claim 1 wherein the means for comparing the output of said controller with a reference potential includes anabsolute value stage for converting signals at the output of said controller of two polarities to a signal of one polarity which is compared with said reference potential.

7. The system of claim 1 including a relay device actuable when the magnitude of the output of said controller exceeds said reference potential.

8. The system of claim 7 wherein said means for moving said electrode comprises a motor and a motor control circuit therefor, and including means operable when said relay device is actuated for connecting the output of said controller to said motor control circuit and for converting said controller to a proportional controller.

9. The system of claim 8 wherein said relay device remains actuated after initial actuation until said error signal is zero.

UNITED STATES PATENT AND TRADEMARK OFFICE Certificate Patent No. 3,597,518 Patented August 3, 1971 Roland W. Roberts Application having been made by Roland W. Roberts, the inventor named in the patent above identified, for the issuance of a certificate under the provisions of Title 35, Section 256, of the United States Code, adding the name of Richard S. DIppolito as a joint inventor, and a showing and proof of the facts satisfying the requirements of the said section having been submitted, it is this 8th day of March 1983, certified that the name of the said Richard S. DIppolito is hereby added to the said patent as a joint inventor with the said Roland W. Roberts.

Fred W. Sherling, Associate Solicitor

Claims

1. In combination with an electric arc furnace of the type in which at least one electrode is positioned adjacent a molten pool of metal and a potential gradient established between the two to form an arc; a system for controlling the position of said electrode with respect to said molten pool of metal so as to maintain a desired arc impedance, said system comprising means for generating an error signal which varies as a function of a deviation in arc impedance from said desired impedance, a proportional plus integrating controller, means for applying said error signal to the input of said controller, means for comparing the output of said controller with a reference potential, means for moving said electrode toward said molten pool when the magnitude of said output exceeds the magnitude of said reference potential and is of one polarity, and means for moving said electrode away from said molten pool when the magnitude of said output exceeds the magnitude of said reference potential and is of the other polarity.

2. The system of claim 1 wherein the output of said controller is a step function followed by a ramp function.

3. The system of claim 1 wherein the means for generating said error signal comprises means for developing a firsT signal which varies as a function of arc voltage, means for developing a second signal which varies as a function of arc current, and means for comparing said first and second signals to develop said error signal.

4. The system of claim 3 wherein said first signal is proportional to arc voltage V and said second signal is proportional to KI where I represents arc current and K represents the desired arc impedance, said first and second signals being of opposite polarity, and said comparing means subtracts said first and second signals to derive said error signal.

5. The system of claim 4 wherein said first and second signals are produced by rectifying alternating current voltages derived from across said arc and from current transformers coupled to a conductor supplying current to said electrodes.

6. The system of claim 1 wherein the means for comparing the output of said controller with a reference potential includes an absolute value stage for converting signals at the output of said controller of two polarities to a signal of one polarity which is compared with said reference potential.