Patent Application
20240326149
The present disclosure relates to pulse width modulation applied to power inverters in a welding system.
Inverter-based welding and cutting power supplies typically control the power in a welding operation by dynamically adjusting the pulse width modulation (PWM) applied to a power inverter based on feedback indicating welding current and voltage levels during a welding/cutting operation. The power inverter comprises a set of high-speed semiconductor switching devices that are switched on and off at a switching frequency of the PWM to create a current that is supplied to a welding/cutting apparatus through a transformer. A conventional technique for controlling the PWM computes fixed on-times of cycles of the PWM on a per cycle basis, i.e., only once per cycle, and in advance of the cycles. For example, the conventional technique may perform the following sequence of operations: (1) pre-compute a first fixed on-time prior to a first cycle, and then generate the first cycle with the first fixed on-time; (2) pre-compute a second fixed on-time prior to a second cycle, and then generate the second cycle with the second fixed on-time; and so on, for subsequent cycles. Thus, during each cycle, the pre-computed fixed on-time cannot be adjusted dynamically to accommodate a change in demand on the welding current during each cycle. For example, the conventional technique cannot increase or decrease the on-time dynamically during each cycle to accommodate a corresponding demand to increase or decrease the welding current during each cycle. This results in a sluggish response time to a demand to increase or decrease the welding current, which negatively impacts the welding process.
In an embodiment, a method of controlling pulse width modulation (PWM) applied to at least one power inverter to generate a weld current for cutting or welding comprises: receiving current values indicative of the weld current during a cycle of the PWM; receiving a target current value indicative of a target current for the weld current; during the cycle, turning ON a pulse to energize the at least one power inverter; and after turning ON the pulse, repeating multiple times during the cycle: determining whether to adjust an on-time of the pulse based on the current values and the target current value in order to drive the weld current toward the target current; and adjusting the on-time responsive to determining.
With reference to
Electrode feeder 106 includes an electrode feeder 116 to feed an electrode from a coiled electrode 120 through cable assembly 108 and through contact tip 111 of welding torch 110, which is in electrical contact with the electrode. Under control of PSC 104, power supply 102 generates weld power that drives the welding process/operation. In welding operations that involve a pulsed or periodic waveform, the weld power typically includes a series of weld current pulses. Power supply 102 provides the weld power from an output terminal 130a of the power supply to the electrode, through electrode feeder 116, cable assembly 108, and welding torch 110, while the cable assembly also delivers a shielding gas from gas container 112 to the welding torch. Return path/cable 115 provides an electrical return path from workpiece 114 to an input terminal 130b of power supply 102
During a welding operation, an electrode tip 118 of the electrode is brought into contact or near contact with workpiece 114, and the weld power (i.e., current and voltage) supplied by power supply 102 to the welding torch creates an arc between workpiece 114 and electrode tip 118 (also referred to as an “electrode stick”) extending through the contact tip. To control the welding process, PSC 104 controls power supply 102 to generate the weld power (e.g., current) at a desired level for the welding process, based on feedback in the form of measurements of the current and voltage (e.g., arc voltage) supplied by the power supply to the welding process. The measurements may be produced by current and voltage sense points in power supply 102 and/or at sense points that are remote from the power supply, such as in cable assembly 108 or welding torch 110.
Inverter 204 comprises a set of high-speed semiconductor switching devices (i.e., power switches) that are pulse width modulated (i.e., switched on and off at a switching frequency) responsive to pulse width modulation (PWM) waveforms 210 (also referred to as “PWM signals”), generated by PSC 104 and applied to control terminals of the switching devices, to convert the DC bus voltage to an AC (power) signal or waveform including a voltage and a primary current IL that flows into transformer 206. Such operation is referred to as “PWM operation” of inverter 204. Example switching frequencies may be in a range from 1 kHz-100 kHz, although other switching frequencies above and below this range may be used. Inverter 204 supplies the AC signal to transformer 206. Transformer 206 converts the voltage and current of the AC signal from inverter 204 to a transformed AC signal having desired levels of a voltage and a secondary current Is for the welding operation, and supplies the transformed AC signal to rectifier 208. Rectifier 208 rectifies the transformed AC signal to produce the weld power based on the secondary current Is and supplies the same to the welding process.
As shown in
PSC 104 receives the current and voltage measurements. To control primary current IL (and correspondingly secondary current Is) generated by power supply 102, PSC 104 controls or dynamically adjusts PWM waveforms 210 applied to inverter 204 based on a desired or target primary current (or secondary current), and based on the current and/or voltage measurements from the welding process. For example, PSC 104 may increase an on-time of cycles of PWM waveforms 210 applied to inverter 204 to increase primary/secondary current IL/Is, and vice versa. In this way, PSC 104 implements a sample-based feedback control loop to control PWM waveforms 210 based on the current and voltage measurements.
Inverter 204 includes a four-quadrant inverter, such as an H-bridge inverter, for example. In other examples, other types of inverters may be employed. Inverter 204 is sometimes referred to as a single “power block” of power supply 102. Inverter 204 includes input voltage rails P1 and P2 to receive the DC bus voltage generated by AC/DC converter 202, and also includes a set of high-frequency semiconductor (current) switches Q1-Q4 (more generally referred to as “power switches” or simply “switches”) coupled to the input voltage rails and arranged in an H configuration. A given switch Qi may be referred to simply as “Qi.” Q1 and Q2 are connected in series with each other between input voltage rails P1 and P2 to form a first/left leg of the inverter, and (ii) Q3 and Q4 are connected in series with each other between the input voltage rails to form a second/right leg of the inverter. Q1 and Q2 are connected to each other at a left terminal LT, and Q3 and Q4 are connected to each other at a right terminal RT. Terminals LT, RT are respectively connected to opposing sides of input coil IC of transformer 206, while opposing sides of output coil OC of the transformer are coupled to rectifier 208. Q1-Q4 are controlled (i.e., turned ON or closed and turned OFF or opened) responsive to PWM waveforms 210 generated by PSC 104 and applied to respective control terminals of the switches. PWM waveforms 210 are typically configured as periodic square wave pulses, although other waveforms are possible.
In the example of
In an example, each switch Qi may be implemented as a switching transistor, such as an insulated gate bipolar transistor (IGBT) having an emitter-collector current path that is turned ON or OFF responsive to a voltage applied to a gate (i.e., the control terminal) of the IGBT, for example. Alternatively, each switch Qi may be implemented as a field effect transistor (FET) (e.g., a metal oxide semiconductor (MOS) FET (MOSFET)) having a source-drain current path that is turned ON or OFF responsive to a voltage applied to a gate of the FET, for example. Each switch Qi may also include an internal diode that is connected in parallel with the controlled current path of the switch. Other types of power switches may be used, as would be appreciated by one of ordinary skill in the relevant arts.
As mentioned above, a conventional technique for controlling a PWM waveform computes and implements fixed on-times for cycles of the PWM once per cycle of the PWM waveform, which results in sluggish and inflexible responses to demands to change welding current during a welding or cutting operation. Embodiments presented herein to generate/control the PWM waveform overcome the aforementioned disadvantages and offer additional advantages described below. More specifically, the embodiments generate/control the PWM waveform to have an on-time that can be adjusted many times during each cycle in correspondence with/to accommodate demands on the primary/secondary current that change during each cycle. In other words, the embodiments provide repetitive, intra-cycle on-time adjustments. To implement such control, the embodiments repeat, multiple times during each cycle (i.e., per cycle), “determining-and-adjusting” operations that include (i) determining whether to adjust an on-time of a pulse generated during the cycle, and (ii) adjusting the on-time of the pulse in a direction indicated by each determining. The determining-and-adjusting occurs at a rate (e.g., at a mega sample per second rate) that exceeds the switching frequency (e.g., a kHz rate) of the PWM waveform, i.e., at a time interval that is less than a period of each cycle. As a result, the on-time may be adjusted (e.g., successively increased) many times during each cycle to accommodate changing current demands during the cycle, i.e., intra-cycle. The embodiments are described in further detail below.
PSC 104 includes a regulator 402, a PWM cycle generator 406, and a driver 408 (i.e., a signal driver) coupled to one another and to power supply 102. PWM cycle generator 406 includes a comparator 410 (which may be an analog comparator), a clock 412, and a latch or flip-flop 414. At a high level, regulator 402 and PWM cycle generator 406 interact to generate PWM waveform PWM_1 and apply the PWM waveform to switches Q1 and Q4 of inverter 204 through driver 408. In particular, regulator 402 and PWM cycle generator 406 control an on-time or pulse width of each cycle of PWM waveform PWM_1, as described below. PSC 104 generates PWM waveform PWM_2 in substantially the same way as the PSC generates PWM waveform PWM_1, except that PWM waveform PWM_2 is phase-shifted with respect to PWM waveform PWM_1. Therefore, the ensuing description for generating PWM waveform PWM_1 shall suffice for PWM waveform PWM_2.
Regulator 402 includes an analog-to-digital converter (ADC) 418 and an ADC 420 to receive current measurement I and voltage measurement V representative of secondary current IL and the output voltage produced by power supply 102, as shown. In another arrangement, ADC 418 and ADC 420 may be external to PSC 104. ADC 418 samples current measurement I at a sample rate SR, to produce current values 422 at the sample rate, and provides the current values for use by PSC 104. Sample rate SR may be in the MHz range, for example. Similarly, ADC 420 samples voltage measurement V at the sample rate SR, to produce voltage values 424 at the sample rate, and provides the voltage values for use by PSC 104. Current values 422 and voltage values 424 represent digitized current and voltage measurements, which provide real-time feedback from the welding process to PSC 104. The current values 422 and the voltage values 424 may also be referred to as “current samples” and “voltage samples,” respectively.
Regulator 402 also receives a target current value TC (also referred to as a current “set point”) that represents a target current for primary current IL. The target current value TC may be configured on welding system 100 through an external source (not shown), or may be varied responsive to demands on primary IL, for example.
PSC 104 performs operations described next at sample rate SR, which is greater than a switching frequency of PWM waveform PWM_1. The operations may be performed at a rate other than sample rate SR, provided that the rate is greater than the switching frequency. At sample rate SR, regulator 402 processes current values 422, voltage values 424, and the target current value TC using a proportional-integral-derivative (PID) algorithm, for example, to produce values of a signal U_setvalue (also referred to as a “trip value”) at the sample rate, and provides the values of U_setvalue to a first input of comparator 410 (also referred to as a “trip comparator 410”). The values of U_setvalue may be voltage values, for example. When regulator 402 employs the PID algorithm, the regulator may operate as a proportional integral (PI) controller or regulator, for example.
Each value of U_setvalue (i.e., “U_setvalue”) is indicative of whether to adjust primary current IL, so as to drive the same toward the target current represented by target current value TC. When current values 422<TC, regulator 402 produces successive values of U_setvalue that increase one to the next. On the other hand, when current values 422>TC, regulator 402 produces successive values of U_setvalue that decrease one to the next. When current values 422 ˜ TC, regulator 402 produces successive values of U_setvalue that are constant.
Because sample rate SR is greater than the switching frequency of PWM waveform PWM_1, regulator 402 produces multiple values of U_setvalue per cycle of the PWM waveform. Stated otherwise, during each cycle of PWM waveform PWM_1, regulator 402 determines/computes N values of U_setvalue that are separated one from the next by a time interval ti (1/SR) that is less than a period P of each cycle, where N=P/ti.
A second input of comparator 410 receives representative signal U a IL supplied by the intermediate voltage sense point on input coil IC of transformer 206. Comparator 410 acts as an analog comparator to continuously compare successive values of U_setvalue to successive values of representative signal U α IL to produce compare result INHIBIT+/−. That is, INHIBIT+/− represents a result of the compare operation performed by comparator 410. When U>U_setvalue, INHIBIT+/− is high (e.g., a logic 1), which indicates to decrease primary current IL. Alternatively, when U≤U_setvalue, INHIBIT+/− is low (e.g., a logic 0), which indicates to increase primary current IL.
Flip-flop 414 includes a reset input R, a set input S, and a latch output Q. Comparator 410 applies INHIBIT+/− to reset input R of flip-flop 414. Clock 412 generates a train of short trigger pulses (referred to collectively and individually as “TP”) and applies the trigger pulses to set input S of flip-flop 414. Trigger pulses TP have a frequency FS that represents the switching frequency (where FS<SR) and a period (1/FS) of each cycle of PWM waveform PWM_1. That is, trigger pulses TP have a pulse repetition interval (PRI) (measured from a rising edge of one trigger pule to a rising edge of a next trigger pulse) equal to the period.
A rising edge of each trigger pulse TP latches output Q high. On the other hand, a rising edge of INHIBIT+/− applied to reset input R resets output Q, i.e., latches output Q low. Thus, output Q of flip-flop 414 produces successive cycles of PWM waveform PWM_1 with period 1/FS responsive to successive trigger pulses TP. Additionally, an on-time or pulse width of each cycle varies in response to the timing of trigger pulses TP in relation to activity of INHIBIT+/−. The start of a trigger pulse TP triggers a cycle of PWM waveform PWM_1, and starts an on-time (i.e., turns ON a pulse) of the cycle. Subsequently, while INHIBIT+/− is low during the cycle, the pulse of the cycle remains ON. In contrast, a rising edge of INHIBIT+/− turns OFF the pulse before the cycle ends to establish a final or total on-time or pulse width of the cycle. The process repeats with the start of a next trigger pulse TP, and so on.
In the example described above, current values 422, target current value TC, representative signal U, and U_setvalue are representative of or based on primary current IL. In another example, current values 422, target current value TC, representative signal U, and U_setvalue may be representative of or based on secondary current Is, rather than primary current IL.
Referring to
In a scenario in which target current value TC suddenly increases (i.e., “jumps-up” in value) during a cycle (to reflect a demand to suddenly increase primary current IL or secondary current IS), U_setvalue responds accordingly, and will also suddenly increase during the cycle to meet the demand. The sudden increase in U_setvalue relative to representative signal U during the cycle will cause the on-time to increase during the cycle. This will cause the primary and second currents (and thus representative signal U) to increase over subsequent cycles and eventually “catch-up” to and exceed U_setvalue, at which point INHIBIT+/− will trigger to shorten on-times during the subsequent cycles. Additionally, U_setvalue may decrease over the subsequent cycles to limit the primary and secondary currents.
Operations described above in connection with
The operations described above control the on-time during the cycle without actually computing a value for the on-time. Rather, during each cycle, the operations repeatedly determine/evaluate whether to adjust the on-time multiple times and then adjust the on-time during the cycle based on each determination.
An example of PSC 814 is shown in an expanded view EV of
Thus, 908 includes repeating, multiple times (i) determining whether to adjust the on-time during the cycle, and (ii) adjusting the on-time responsive to each determining.
Once the pulse is turned OFF, the pulse may remain OFF until a start of a next cycle. In other words, adjusting may include adjusting the on-time by increasing the on-time (i.e., keeping the pulse ON responsive to each determining to increase the on-time), until adjusting the on-time includes turning OFF the pulse responsive to determining to decrease the on-time, after which the pulse remains turned OFF.
Determining whether to adjust the on-time may include:
Adjusting the on-time responsive to determining may include:
Method 900 further incudes, for each of successive cycles of the PWM, repeating turning on the pulse, and repeating the multiple times (i) determining, and (ii) adjusting to drive the weld current to the target current.
In a single power inverter embodiment, the one or more power inverters include only one power inverter. In a multiple power inverter embodiment, the one or more power inverters includes multiple power inverters configured to generate multiple currents responsive to the PWM. The multiple power inverter embodiment includes summing the multiple currents to produce the weld current as a combined current, and providing the PWM to each of the multiple power inverters based on the combined current.
With reference to
In some aspects, the techniques described herein relate to a method of controlling pulse width modulation (PWM) applied to at least one power inverter to generate a weld current for cutting or welding, including: receiving current values indicative of the weld current during a cycle of the PWM; receiving a target current value indicative of a target current for the weld current; during the cycle, turning ON a pulse to energize the at least one power inverter; and after turning ON the pulse, repeating multiple times during the cycle: determining whether to adjust an on-time of the pulse based on the current values and the target current value in order to drive the weld current toward the target current; and adjusting the on-time responsive to determining.
In some aspects, the techniques described herein relate to a method, wherein: repeating the multiple times includes repeating the multiple times at a time interval that is less than a period of the cycle.
In some aspects, the techniques described herein relate to a method, wherein: determining includes determining whether to increase or decrease the on-time; and adjusting includes: when it is determined to increase the on-time, increasing the on-time by not turning OFF the pulse; and when it is determined to decrease the on-time, decreasing the on-time by turning OFF the pulse.
In some aspects, the techniques described herein relate to a method, further including: for each of successive cycles of the PWM, repeating (i) turning ON the pulse, and (ii) repeating the multiple times in order to drive the weld current to the target current.
In some aspects, the techniques described herein relate to a method, wherein determining includes: processing the target current value and the current values using a proportional-integral-derivative (PID) algorithm.
In some aspects, the techniques described herein relate to a method, wherein: turning ON the pulse includes turning ON the pulse at a start of the cycle.
In some aspects, the techniques described herein relate to a method, wherein: adjusting include adjusting the on-time until adjusting the on-time includes turning OFF the pulse responsive to determining to decrease the on-time.
In some aspects, the techniques described herein relate to a method, wherein determining includes: computing a set value indicative of whether to increase or decrease the on-time based on the current values and the target current value; and comparing the set value to a representative signal indicative of the weld current, wherein adjusting includes adjusting based on results of comparing.
In some aspects, the techniques described herein relate to a method, wherein adjusting includes: when the representative signal does not exceed the set value, increasing the on-time by not turning OFF the pulse; and when the representative signal exceeds the set value, decreasing the on-time by turning OFF the pulse.
In some aspects, the techniques described herein relate to a method, wherein the at least one power inverter includes multiple power inverters configured to generate multiple currents responsive to the PWM, and the method further includes: summing the multiple currents to produce the weld current; and providing the PWM to each of the multiple power inverters.
In some aspects, the techniques described herein relate to an apparatus including: at least one power inverter of a welding system; and a controller configured to perform controlling pulse width modulation (PWM) applied to the at least one power inverter to generate a weld current for cutting or welding by: receiving current values indicative of the weld current during a cycle of the PWM; receiving a target current value indicative of a target current for the weld current; during the cycle, turning ON a pulse to energize the at least one power inverter; and after turning ON the pulse, repeating multiple times during the cycle: determining whether to adjust an on-time of the pulse based on the current values and the target current value in order to drive the weld current toward the target current; and adjusting the on-time responsive to determining.
In some aspects, the techniques described herein relate to an apparatus, wherein the controller is further configured to perform: repeating the multiple times includes repeating the multiple times at a time interval that is less than a period of the cycle.
In some aspects, the techniques described herein relate to an apparatus, wherein: the controller is configured to perform determining by determining whether to increase or decrease the on-time; and the controller is configured to perform adjusting by: when it is determined to increase the on-time, increasing the on-time by not turning OFF the pulse; and when it is determined to decrease the on-time, decreasing the on-time by turning OFF the pulse.
In some aspects, the techniques described herein relate to an apparatus, wherein the controller is further configured to perform: for each of successive cycles of the PWM, repeating (i) turning ON the pulse, and (ii) repeating the multiple times in order to drive the weld current to the target current.
In some aspects, the techniques described herein relate to an apparatus, wherein: the controller is configured to perform processing the target current value and the current values using a proportional-integral-derivative (PID) algorithm.
In some aspects, the techniques described herein relate to an apparatus, wherein: the controller is configured to perform turning ON the pulse by turning ON the pulse at a start of the cycle.
In some aspects, the techniques described herein relate to an apparatus, wherein: the controller is configured to perform adjusting by adjusting the on-time until adjusting the on-time includes turning OFF the pulse responsive to determining to decrease the on-time.
In some aspects, the techniques described herein relate to an apparatus, wherein the controller is configured to perform determining by: computing a set value indicative of whether to increase or decrease the on-time based on the current values and the target current value; and comparing the set value to a representative signal indicative of the weld current, wherein the controller is configured to perform adjusting by adjusting based on results of comparing.
In some aspects, the techniques described herein relate to an apparatus, wherein the controller is configured to perform adjusting by: when the representative signal does not exceed the set value, increasing the on-time by not turning OFF the pulse; and when the representative signal exceeds the set value, decreasing the on-time by turning OFF the pulse.
In some aspects, the techniques described herein relate to an apparatus, wherein the at least one power inverter includes multiple power inverters configured to generate multiple currents responsive to the PWM, and the apparatus further includes: a summer to sum the multiple currents to produce the weld current; and the controller is configured to perform providing the PWM to each of the multiple power inverters.
The descriptions of the various embodiments have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
