A PWM control method for an inverter welding machine

Abstract

This invention belongs to the field of welding power supply technology, specifically disclosing a PWM control method for an inverter welding machine. This invention employs a PWM control method with adjustable pulse width on one bridge arm and a fixed maximum pulse width on the other. It generates a volt-second product only during conduction and clamps to 0V during freewheeling, without amplifying asymmetry. This allows the freewheeling capacitor sufficient freewheeling time and a low-impedance path to compensate for the volt-second difference, canceling out all volt-second imbalance during conduction throughout the entire cycle. Even with dynamic changes in PWM duty cycle, load jumps, or device asymmetry, volt-second balance can be quickly restored. During the freewheeling short-circuit period, the transformer primary side is approximately short-circuited, resulting in very low loop impedance. The new PWM control method allows for faster current changes during the freewheeling phase, quickly neutralizing the unidirectional residual magnetic flux of the iron core, releasing the transformer's residual magnetism, and efficiently completing flux reset, thus avoiding transformer bias. This invention fully utilizes the magnetic flux of the core, reducing the size of the transformer core, thereby miniaturizing and lightening the welding machine.

Description

Technical Field

[0001] This invention belongs to the field of welding power supply technology, and specifically relates to a PWM control method for an inverter welding machine. Background Technology

[0002] Currently, inverter welding machines using a full-bridge topology typically have four PWM controls, controlling four switching transistors Q0, Q1, Q2, and Q3 to form two bridge arms (bridge arm A and bridge arm B). Their circuit structure is as follows: Figure 1 As shown. Bridge arm A is formed by Q0 and Q1. Figure 1 The left half of the bridge shown is composed of bridge arm B, which is formed by Q2 and Q3. Figure 1 The right half of the bridge is shown in the image.

[0003] In bridge arm A, Q0 is the upper transistor, connected to the positive terminal P+ of the three-phase rectifier bridge, and Q1 is the lower transistor, connected to the negative terminal N- of the three-phase rectifier bridge. In bridge arm B, Q3 is the upper transistor, connected to the positive terminal P+ of the three-phase rectifier bridge, and Q2 is the lower transistor, connected to the negative terminal N- of the three-phase rectifier bridge. The inverter transformer and freewheeling capacitor C are connected between point A (between Q0 and Q1) and point B (between Q3 and Q2).

[0004] Traditional PWM control often adopts the "diagonal transistors alternately conduct" method, where the pulse widths of bridge arm A and bridge arm B are simultaneously adjustable. That is, a complete working cycle T is divided into four stages, and the control method of "diagonal transistors alternately conduct" with simultaneously adjustable pulse widths is used.

[0005] For the two switches on each bridge arm, the maximum on-time Ton is fixed at half a cycle, i.e., T / 2 (including dead time), meaning the maximum duty cycle D_single of a single switch is 50%, as shown in the timing diagram below. Figure 2 and Figure 3 As shown.

[0006] When using traditional PWM control, if the transformer core volume decreases and the flux utilization rate is high, there is not enough flux margin to withstand the flux offset caused by residual magnetic energy or a small DC component (any asymmetry in the circuit (such as a small difference in the conduction time of the switching transistor or asymmetry in pulse drive) may produce a small DC component), the transformer will become magnetized.

[0007] Furthermore, with traditional PWM control, the minute volt-second errors caused by device asymmetry cannot be quickly compensated by the DC blocking capacitor. The errors accumulate cycle by cycle, leading to unidirectional flux creep and transformer saturation. Simultaneously, frequent arcing in welding machines and sudden load increases due to short circuits cause instantaneous PWM jumps. During these jumps, the freewheeling phase is rigidly clamped by the bus voltage, and the freewheeling time is locked by the PWM, only occurring within the dead time. This limits the capacitor charging and discharging time window, causing minute differences to accumulate and resulting in magnetic bias.

[0008] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art. Summary of the Invention

[0009] The purpose of this invention is to propose a PWM control method for an inverter welding machine. This method adopts a novel PWM control mode in which the pulse width of one bridge arm is adjustable and the pulse width of the other bridge arm is fixed at the maximum, which effectively avoids the phenomenon of transformer bias.

[0010] To achieve the above objectives, the present invention adopts the following technical solution: A PWM control method for an inverter welding machine employs a PWM control approach where the pulse width of two switches on one side of the bridge arm is adjustable, while the pulse width of two switches on the other side of the bridge arm is fixed. The details are as follows: The bridge arm with a fixed maximum pulse width is defined as bridge arm A, and the other bridge arm is defined as bridge arm B. The two switches Q0 and Q1 in bridge arm A are set to keep their maximum pulse width constant. The conduction time of each switch in bridge arm A is fixed at half a cycle, that is, the duty cycle of a single switch is 50%. Meanwhile, the pulse widths of the two switches Q2 and Q3 in bridge arm B are adjustable, and the conduction time of each switch in bridge arm B in one cycle is less than that of each switch in bridge arm A in one cycle.

[0011] A single PWM control cycle T is divided into four working stages, which appear as a whole.

[0012] When output begins, Q0 of bridge arm A and Q2 of bridge arm B are turned on first, serving as the first valid drive waveform within the cycle; when output stops, Q1 of bridge arm A and Q3 of bridge arm B are turned off, serving as the last set of turn-off waveforms at the end of the cycle.

[0013] The four operating stages within a single PWM control cycle T are as follows: Phase 1: Q0 of bridge arm A and Q2 of bridge arm B are turned on first, while Q1 of bridge arm A and Q3 of bridge arm B are turned off. Phase 2: Q0 of bridge arm A is turned on, Q2 of bridge arm B is turned off, and at the same time, Q1 of bridge arm A and Q3 of bridge arm B remain off, entering the first freewheeling phase, which continues until Q1 and Q3 are turned on. Phase 3: Q1 of bridge arm A and Q3 of bridge arm B are turned on, while Q0 of bridge arm A and Q2 of bridge arm B are turned off. Phase 4: Q1 of bridge arm A is turned on, Q3 of bridge arm B is turned off, and at the same time, Q0 of bridge arm A and Q2 of bridge arm B remain off, entering the second freewheeling phase, thus completing a full cycle.

[0014] Furthermore, based on the aforementioned inverter welding machine PWM control method, this invention also proposes an inverter welding machine that uses the inverter welding machine PWM control method described above for PWM control.

[0015] The present invention has the following advantages: As described above, this invention discloses a PWM control method for an inverter welding machine. This method employs a novel PWM control approach where the pulse width of one bridge arm is adjustable, while the pulse width of the other bridge arm is fixed at its maximum. This PWM control method generates a volt-second product only during conduction. The positive volt-second product of stage one and the negative volt-second product of stage three completely cancel each other out within the complete cycle, thereby eliminating magnetization. During freewheeling, the voltage is clamped to 0V, preventing asymmetry amplification. This allows the freewheeling capacitor sufficient freewheeling time and a low-impedance path to compensate for the volt-second difference, canceling out all volt-second imbalances during conduction throughout the entire cycle. Even with dynamic changes in the PWM duty cycle, load jumps, or device asymmetry, the volt-second balance can be quickly restored, thus suppressing magnetization. Furthermore, during freewheeling, the primary side of the transformer is approximately short-circuited, resulting in very low circuit impedance. The residual magnetic energy of the transformer core charges the freewheeling capacitor. The current change in the novel PWM control method is faster, enabling more rapid neutralization of the unidirectional residual magnetic flux in the core, releasing the transformer's residual magnetism, efficiently completing flux reset, and continuously suppressing magnetization accumulation, thereby avoiding transformer magnetization. The novel PWM control method proposed in this invention can fully utilize the magnetic flux of the core, which helps to reduce the size of the transformer core, thereby making the welding machine smaller and lighter. In addition, by reducing the core size, the length of the primary and secondary windings is reduced, copper losses are decreased, the temperature rise of the welding machine is improved, and the overall power conversion efficiency is improved, thus enhancing the energy efficiency rating of the welding machine. Attached Figure Description

[0016] Figure 1 The circuit diagram for an inverter welding machine using a full-bridge topology; Figure 2 This is a schematic diagram of the drive waveform when the welding machine outputs high current. Figure 3 This is a schematic diagram of the drive waveform when the welding machine is outputting a small current. Figure 4 This is a schematic diagram of the flow loop for the four operating stages under traditional PWM control; where... Figure 4 In the diagram, (a), (b), (c), and (d) represent the first, second, third, and fourth stages of traditional control, respectively. Figure 5 This is a schematic diagram of the flow loop for the four working stages under the novel PWM control method proposed in this invention; wherein Figure 5 In the diagram, (a), (b), (c), and (d) represent the first, second, third, and fourth stages of the novel PWM control, respectively. Figure 6 This is a timing diagram of the novel PWM control method in an embodiment of the present invention; Figure 7This is a schematic diagram of voltage markings in stage one of the novel PWM control method proposed in this invention; Figure 8 This is a schematic diagram showing the voltage markings in stage two of the novel PWM control method proposed in this invention. Figure 9 This is a comparison graph of the current change at a certain moment between the freewheeling current of the present invention method and the traditional PWM control method. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1 This embodiment describes a PWM control method for an inverter welding machine. This method adopts a novel PWM control approach where the pulse width of one bridge arm is adjustable and the pulse width of the other bridge arm is fixed at the maximum, which effectively avoids the transformer bias phenomenon.

[0018] Before formally introducing this invention, the principle and process of the traditional PWM control method will be explained for comparison.

[0019] like Figure 4 As shown, the four operating stages of traditional PWM control are as follows: I. Traditional control stage one: Diagonal transistors Q0 and Q2 are turned on (positive half-cycle).

[0020] The PWM drive turns on Q0 and Q2 simultaneously, while turning off Q1 and Q3.

[0021] Current path: Starting from the positive terminal P+ connected to the three-phase rectifier bridge, the current flows to Q0, then through the freewheeling capacitor C, then through the primary side Np of the transformer (bridge arm A → bridge arm B), then through Q2, and finally reaches the negative terminal N- connected to the three-phase rectifier bridge.

[0022] At this time, the DC voltage Uin output by the three-phase rectifier bridge is applied to the primary winding of the transformer.

[0023] transformer primary voltage = +Uin (because) (Point B ≈ +Uin, Point B ≈ 0), the transformer is forward magnetized. A positive voltage is induced on the secondary side of the transformer, and the corresponding rectifier diode DR2 conducts, supplying energy to the load.

[0024] II. Traditional control phase two: All pipes are turned off (first dead time), first freewheeling phase.

[0025] When Q0 and Q2 are turned off, the circuit enters a brief dead zone. The transformer primary winding inductance L consists of parasitic leakage inductance Lk and main inductance Lm, generating an induced electromotive force e = -L* di / dt (positive at the top and negative at the bottom), the polarity of which attempts to maintain the original direction of current Ip.

[0026] The sustained current Ip will force the voltage at point B to be greater than that of the P+ bus, and the voltage at point A to be less than that of the N- bus.

[0027] At this point, the current will naturally switch to diodes D1 and D3, which are connected in antiparallel to Q1 and Q3.

[0028] Current path (Ip): The current first flows through capacitor D3 at transformer B, then through the P+ bus to the input capacitor Cin, then through the N- bus to diode D1, then to point A, then into the freewheeling capacitor C, and finally back to the transformer. end.

[0029] At this time, diodes D1 and D3 are turned on, connecting the two ends of the primary side of the transformer ( The voltages of B and P are clamped at N- (0V) and P+ (+Uin), respectively. Therefore = -Uin (because) Point B ≈ 0V, Point B ≈ +Uin).

[0030] This negative voltage -Uin is applied across the leakage inductance, causing the primary current Ip to decrease.

[0031] III. Traditional control stage three: Diagonal transistors Q1 and Q3 are turned on (negative half-cycle).

[0032] The PWM drive turns on Q1 and Q3 simultaneously, while turning off Q0 and Q2.

[0033] Current path: The current flows from the positive terminal P+ connected to the three-phase rectifier bridge to Q3, then through the primary winding Np of the transformer, to the freewheeling capacitor C, then to Q1, and finally back to the negative terminal N- connected to the three-phase rectifier bridge. At this time, the DC voltage Uin output by the three-phase rectifier bridge is applied in reverse to the primary winding of the transformer, and the transformer is reverse magnetized.

[0034] A negative voltage is induced on the secondary side of the transformer, and another set of rectifier diodes DR3 conducts, continuing to supply energy to the load.

[0035] IV. Traditional control phase four: All pipes are turned off (second dead time), second freewheeling phase.

[0036] Q1 and Q3 are shut down, entering the second dead zone.

[0037] Ip path: From transformer A' end, it first passes through freewheeling capacitor C, then reaches point A, then through D0, then reaches P+ bus, then through input capacitor Cin to reach N- bus, then through diode D2, and finally returns to transformer B end.

[0038] At this time, both ends of the primary side of the transformer ( The voltages of A and B are clamped at P+ (+Uin) and N- (0V), respectively.

[0039] therefore, =+Uin (because) (Point B ≈ +Uin, Point B ≈ 0V). This voltage +Uin is applied to the leakage inductance, causing the primary current Ip to decrease. The conventional control stage four repeats the freewheeling process of stage two in reverse, thus completing a full cycle.

[0040] like Figure 6 As shown in the figure, the present invention proposes a PWM control method for an inverter welding machine, which adopts a PWM control mode in which the pulse width of the two switching transistors of one side bridge arm is adjustable, and the pulse width of the two switching transistors of the other side bridge arm is fixed.

[0041] The bridge arm with a fixed maximum pulse width is defined as bridge arm A, and the other bridge arm is defined as bridge arm B.

[0042] The two switches Q0 and Q1 of bridge arm A are set to keep their maximum pulse width constant. The on-time (including dead time) of each switch on bridge arm A is fixed at half a cycle, i.e., T / 2, that is, the duty cycle of a single switch is 50%.

[0043] Meanwhile, the pulse widths of the two switches Q2 and Q3 in bridge arm B are adjustable, and the conduction time of each switch in bridge arm B in one cycle is less than that of each switch in bridge arm A in one cycle.

[0044] This invention employs a PWM control method with "diagonal transistors conducting alternately" where the pulse width is adjustable at different times.

[0045] The output current of the whole machine is controlled by the PWM modulation of each switching transistor on bridge arm B to achieve closed-loop regulation. The conduction pulse width of the two power devices on bridge arm B is corrected in real time according to the given voltage command, so as to accurately control the output current of the welding machine.

[0046] In this embodiment, the pulse widths of the two switches Q2 and Q3 in the B-arm are adjustable, and the pulse widths of both switches are adjusted using the traditional PWM pulse width adjustment method. The adjustment process is as follows: The voltage value output by PWM control is calculated using the output voltage setpoint and the output voltage feedback value. .

[0047] in To control the proportional coefficient, it is adjusted according to the actual welding conditions, representing the dynamic response speed. For voltage setting value, This is the output voltage feedback value. This is the voltage value output by the PWM control.

[0048] To control the integral coefficient, it is adjusted according to the actual welding conditions, representing the ability to eliminate accumulated residual errors and ensuring that the current remains constant during long-term welding without slowly drifting.

[0049] Then according to By controlling the on-time of switching transistors Q3 and Q2 (Unit: s).

[0050] .

[0051] in This refers to the number of turns on the primary side of the welding machine inverter transformer. This refers to the number of turns on the secondary side of the welding machine inverter transformer. This refers to the voltage on the primary side of the welding machine inverter transformer, in volts (V). This is the voltage on the secondary side of the welding machine inverter transformer, in volts (V).

[0052] The voltage drop across the secondary side of the welding machine inverter transformer (voltage drop of switching transistors, voltage drop of conductors, etc.), in volts (V). The inverter frequency of the welding machine is expressed in Hz. This is the DC voltage output by the three-phase rectifier bridge of the welding machine, in volts (V).

[0053] The above describes the traditional PWM pulse width adjustment method. This is only for ease of understanding and is not an innovation of this invention.

[0054] A single PWM control cycle T is divided into four working stages, which appear as a whole.

[0055] When output begins, Q0 of bridge arm A and Q2 of bridge arm B are turned on first, serving as the first valid drive waveform within the cycle; when output stops, Q1 of bridge arm A and Q3 of bridge arm B are turned off, serving as the last set of turn-off waveforms at the end of the cycle, as shown below. Figure 6 As shown.

[0056] The four operating stages within a single PWM control cycle T are as follows: Phase 1: Q0 of bridge arm A and Q2 of bridge arm B are turned on first, while Q1 of bridge arm A and Q3 of bridge arm B are turned off.

[0057] Phase 2: Q0 of bridge arm A is turned on, Q2 of bridge arm B is turned off, and at the same time, Q1 of bridge arm A and Q3 of bridge arm B remain off, entering the first freewheeling phase, which continues until Q1 and Q3 are turned on.

[0058] Phase 3: Q1 of bridge arm A and Q3 of bridge arm B are turned on, while Q0 of bridge arm A and Q2 of bridge arm B are turned off.

[0059] Phase 4: Q1 of bridge arm A is turned on, Q3 of bridge arm B is turned off, and at the same time, Q0 of bridge arm A and Q2 of bridge arm B remain off, entering the second freewheeling phase, thus completing a full cycle.

[0060] like Figure 5 As shown, the four operating stages of the new PWM control are as follows: Phase 1: PWM drive turns on Q0 and Q2 simultaneously, while turning off Q1 and Q3.

[0061] The path of current Ip in this stage is as follows: The current flows from the P+ bus to Q0, then through the freewheeling capacitor C, then through the primary side of the transformer, and finally through Q2 to reach N-; where the P+ bus is connected to the positive terminal of the three-phase rectifier bridge, and the N- bus is connected to the negative terminal of the three-phase rectifier bridge.

[0062] At this time, the DC voltage Uin output by the three-phase rectifier bridge is applied to the primary winding of the transformer.

[0063] transformer primary voltage = +Uin (because) (Point B ≈ +Uin, Point B ≈ 0), the transformer is forward magnetized. A positive voltage is induced on the secondary side of the transformer, and the corresponding rectifier diode DR2 conducts, supplying energy to the load.

[0064] Phase 2: The PWM drive turns Q0 on (it is not turned off at this time), turns off Q2, and turns off Q1 and Q3, entering the first freewheeling phase; this time is controllable and continues until Q1 and Q3 are turned on.

[0065] The transformer primary winding inductance L is composed of parasitic leakage inductance Lk and main inductance Lm, which will generate an induced electromotive force e = -L* di / dt (positive at the top and negative at the bottom), whose polarity attempts to maintain the original current Ip direction unchanged.

[0066] When Q0 of bridge arm A is turned on, the voltage at point A will be equal to the voltage of the P+ bus. The primary current of the transformer cannot change abruptly. The voltage at point B is greater than the voltage of the P+ bus. At this time, the current will naturally switch to the diode D3, which is connected in antiparallel with Q3.

[0067] Point A is located between Q0 and Q1; point B is located between Q2 and Q3 and is connected to one end of the transformer. The end of the transformer's primary winding connected to point B is defined as end B, and the other end of the transformer's primary winding is... end.

[0068] The path of the second current Ip in this stage is as follows: from terminal B of the transformer, it first goes to diode D3, then to the P+ bus, then through Q0 to point A, then into the freewheeling capacitor C, and finally back to the transformer. end.

[0069] At this time, Q0 and diode D3 are turned on, clamping the voltage across the primary side of the transformer at P+, i.e., +Uin; therefore, the voltage across the primary side of the transformer is... 0 (because Point B ≈ +Uin, Point B ≈ +Uin), at this time the power supply and load are decoupled and there is no energy transfer.

[0070] The residual magnetic energy of the transformer core charges the freewheeling capacitor, forming an LC resonant circuit. The oscillating current continuously circulates within the power bridge arm and the transformer winding loop and gradually decays.

[0071] The residual magnetic energy of the transformer freewheels through the anti-parallel diode D3 of the switching transistors Q0 and Q3, creating conditions for soft switching.

[0072] Phase 3: PWM drive turns on Q1 and Q3 simultaneously, while turning off Q0 and Q2.

[0073] The paths of the three currents Ip in this stage are as follows: The current flows from the P+ busbar through Q3, then through the primary side of the transformer, then through the freewheeling capacitor C, and finally through Q1 to reach N-; where the P+ busbar is connected to the positive terminal of the three-phase rectifier bridge, and the N- busbar is connected to the negative terminal of the three-phase rectifier bridge. At this time, the DC voltage Uin output by the three-phase rectifier bridge is applied to the primary winding of the transformer; transformer primary voltage = -Uin (because) Point B ≈ 0, Point B ≈ +Uin), the transformer is reverse magnetized; a negative voltage is induced on the secondary side of the transformer, and another set of rectifier diodes DR3 conducts, continuing to supply energy to the load.

[0074] Phase 4: The PWM drive turns Q1 on (it is not turned off at this time), turns off Q3, and turns off Q0 and Q2, entering the second freewheeling phase; this time is controllable and continues until Q0 and Q2 turn on.

[0075] Since Q1 in bridge arm A is turned on, the voltage at point A will be equal to the N-bus voltage. The transformer primary current cannot change abruptly, so the voltage at point B is less than the N-bus voltage. At this time, the current will naturally switch to diode D2, which is connected in anti-parallel with Q2.

[0076] Point A is located between Q0 and Q1; point B is located between Q2 and Q3 and is connected to one end of the transformer. The end of the transformer's primary winding connected to point B is defined as end B, and the other end of the transformer's primary winding is... end.

[0077] The paths of the four currents Ip in this stage are as follows: from the transformer... The current first enters the freewheeling capacitor C, then reaches point A, then passes through Q1 to reach the N-bus, then passes through diode D2, and finally returns to the B terminal of the transformer.

[0078] At this time, Q1 and diode D2 are turned on, clamping the voltage across the primary side of the transformer at N-, i.e., -Uin; therefore, the voltage across the primary side of the transformer... 0 (because Point A ≈ -Uin, Point B ≈ -Uin), at this time the power supply and load are decoupled and there is no energy transfer.

[0079] The residual magnetic energy of the transformer core charges the freewheeling capacitor, forming an LC resonant circuit. The oscillating current continuously circulates within the power bridge arm and the transformer winding loop and gradually decays.

[0080] The novel PWM control method proposed in this invention can fully utilize the magnetic flux of the magnetic core, reduce the volume of the transformer core, and thus make the welding machine smaller and lighter. Its theoretical basis is as follows: Define the flux utilization rate k of the transformer core. This represents the number of turns on the primary side of the welding machine inverter transformer. According to the following formula, with the core material remaining unchanged, as k increases, Ae decreases, thereby reducing the core volume: .

[0081] in Saturation magnetic flux density, in tons (T). Residual magnetic flux density, in tons (T). The effective cross-sectional area of ​​the magnetic core is expressed in mm. 2 ; The inverter frequency of the welding machine is expressed in Hz.

[0082] The DC voltage output by the three-phase rectifier bridge when the welding machine is at its rated input voltage, in volts (V).

[0083] In addition, by reducing the core volume, the lengths of the primary and secondary windings are reduced, copper losses are decreased, the temperature rise of the welding machine is improved, and the overall power conversion efficiency is increased, thereby improving the energy efficiency rating of the welding machine. The theoretical basis for this is as follows: Under the same effective cross-sectional area Ae and working magnetic flux, reducing the core volume is equivalent to shortening the circumference of a single-turn winding, reducing the total winding length L, decreasing the DC resistance R of the winding, and consequently reducing copper losses. ; in The resistivity of copper is expressed in Ω·mm. 2 / m; L is the total length of the copper wire, in meters; S is the cross-sectional area of ​​the copper wire, in mm². 2 .

[0084] Furthermore, this invention generates a volt-second product only when the circuit is on, and clamps it to 0V during freewheeling, without amplifying the asymmetry. This allows the freewheeling capacitor to have sufficient freewheeling time and a low-impedance path to compensate for the volt-second difference, thus canceling out all volt-second imbalances during the entire cycle.

[0085] In traditional PWM control: Phase 1 conduction phase: × ; Phase Two Continuation Phase: × ; Phase 3 conduction phase: × ; Phase 4 Continuation Phase: × Ideally, the total volt-second product of these four stages is zero. .

[0086] .

[0087] In reality, welding machines are subjected to highly dynamic loads (short circuits, arcing, spatter, and rapid current changes), and coupled with the discreteness of components, welding machines are inevitably unbalanced. During rapid changes in PWM, stage one... ≠ Stage Three Phase Two ≠ Stage Four Imbalance accumulates cycle by cycle. During freewheeling, the entire process is affected by ±U. in Clamping, the primary voltage of the transformer is ±U in Lock-up occurs when the freewheeling capacitor C attempts to generate a compensation voltage to offset Δ, but the freewheeling time is locked by the PWM, limiting the capacitor's charging and discharging time window, causing tiny differences to accumulate. Ultimately, this leads to a cycle-by-cycle accumulation of the volt-second difference, causing the DC flux of the magnetic core to drift and resulting in magnetization.

[0088] Compared with the traditional PWM control method, the present invention uses a novel PWM control method: Phase 1 conduction phase: × (A positive voltage is applied to the primary side of the transformer); Phase 2 follow-through phase: 0V × (The winding is short-circuited by the bridge arm, and the voltage is approximately 0V). Phase 3 conduction phase: × (A reverse voltage is applied to the primary side of the transformer); Phase 4 follow-through phase: 0V × (The winding is short-circuited by the bridge arm, and the voltage is approximately 0V).

[0089] During the freewheeling phase, the voltage is clamped to 0, and the volt-second difference ∆ is naturally "cut off." This is equivalent to: when the PWM changes rapidly, only the voltage of phase one is present. ≠ Stage Three Only then can an imbalance occur during the conduction phase; simultaneously, in phase one... and Phase Three The asymmetrical volt-second difference will discharge the freewheeling capacitor C until a stable compensation voltage is established. .

[0090] Initially, the volt-second product is balanced: .

[0091] .

[0092] If stage one > Phase Three The forward volt-second is greater than the reverse volt-second, the net volt-second of the period is positive, there is a DC component in the winding, and the capacitor is continuously charged during the positive half-cycle.

[0093] If stage one <Phase Three> The forward volt-second is less than the reverse volt-second, the net volt-second of the period is negative, and there is a DC component in the winding. The negative half-cycle continuously charges the capacitor. Therefore, the freewheeling capacitor C automatically establishes a compensation voltage U. C The entire cycle cancels out the volt-second difference of the conduction segment, maintaining the initial volt-second product balance.

[0094] This method generates a volt-second product only during conduction. The positive volt-second product in stage one and the negative volt-second product in stage three completely cancel each other out within the entire cycle, thus eliminating magnetization. During freewheeling, the voltage is clamped to 0V, preventing asymmetry amplification. This allows the freewheeling capacitor sufficient freewheeling time and a low-impedance path to compensate for the volt-second difference, canceling out all volt-second imbalances during conduction throughout the entire cycle. Even with dynamic changes in the PWM duty cycle, load transitions, or device asymmetry, the volt-second balance can be quickly restored, thereby suppressing magnetization.

[0095] In addition, during freewheeling, the primary side of the transformer is approximately short-circuited, and the circuit impedance is very low. The residual magnetic energy of the transformer core charges the freewheeling capacitor. The current change of the new PWM control method is faster, which can quickly neutralize the unidirectional residual magnetic flux of the core, release the residual magnetism of the transformer, efficiently complete the magnetic flux reset, and continuously suppress the accumulation of bias magnetism, thereby avoiding the phenomenon of transformer bias magnetism.

[0096] Specifically, according to Kirchhoff's voltage law (the algebraic sum of the voltages of all components along any closed loop = 0), the freewheeling phase of traditional PWM control satisfies: ,like Figure 7 As shown.

[0097] in The DC voltage output by the three-phase rectifier bridge when the welding machine is at its rated input voltage, in volts (V). This is the voltage across the freewheeling capacitor, in volts (V). This is the voltage of the primary winding of the transformer, in volts (V).

[0098] Based on the previous analysis, because Substituting into the formula above, we get: .

[0099] Similarly, the freewheeling phase of the new PWM control satisfies: ,like Figure 8 As shown.

[0100] Based on the previous analysis, because Substituting into the formula above, we get: .

[0101] As can be seen from the above, traditional PWM control... New PWM control .

[0102] After the conduction is completed, This is the DC voltage output by the three-phase rectifier bridge when the welding machine is at its rated input voltage.

[0103] At the start of freewheeling, the initial current I0 on the primary side of the transformer is the same. As I0 approaches 0... The time t required by traditional PWM control is less than that of the new PWM control method, and can be simplified to Figure 9 .

[0104] Within the same Δt time interval, the novel PWM control... (Current change) is greater than that of traditional PWM control. Therefore, the new PWM control method can more quickly neutralize the unidirectional residual magnetic flux of the iron core, release the residual magnetism of the transformer, efficiently complete the magnetic flux reset, and continuously suppress the accumulation of bias magnetism, thereby avoiding the transformer bias magnetism phenomenon. When the output stops, the PWM pulse width of the switching transistors Q2 and Q3 in bridge arm B is gradually reduced until it is reduced to 0, so that the freewheeling capacitor voltage is fully discharged. The specific operation is as follows: After acquiring the stop output signal, record the last pulse width Ton before the Q3 stop output signal. Preset N discharge cycles. In each discharge cycle, Q2 and Q3 decrease by each pulse width. The magnitude decreases until it decreases to 0.

[0105] This method can avoid the generation of bias magnetism due to the freewheeling capacitor C not being fully discharged when the output stops and then immediately starts again.

[0106] Example 2 This embodiment 2 describes an inverter welding machine, which adopts the same inventive concept as embodiment 1 above. This inverter welding machine uses the PWM control method of the inverter welding machine in embodiment 1 for PWM control.

[0107] Of course, the above description is only a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. It should be noted that any equivalent substitutions or obvious modifications made by those skilled in the art under the guidance of this specification fall within the scope of this specification and should be protected by the present invention.

Claims

1. A PWM control method for an inverter welding machine, characterized in that, The system employs a PWM control method where the pulse width of two switches on one side of the bridge arm is adjustable, while the pulse width of two switches on the other side of the bridge arm is fixed. The details are as follows: The bridge arm with a fixed maximum pulse width is defined as bridge arm A, and the other bridge arm is defined as bridge arm B. The two switches Q0 and Q1 in bridge arm A are set to maintain a fixed maximum pulse width. The conduction time of each switch in bridge arm A is fixed at half a cycle, that is, the duty cycle of a single switch is 50%. Meanwhile, the pulse widths of the two switches Q2 and Q3 in bridge arm B are adjustable, and the conduction time of each switch in bridge arm B in one cycle is less than that of each switch in bridge arm A in one cycle.

2. The inverter welding machine PWM control method according to claim 1, characterized in that, The pulse width adjustment of the two switching transistors Q2 and Q3 in the B-arm is achieved by using the traditional PWM pulse width adjustment method.

3. The inverter welding machine PWM control method according to claim 1, characterized in that, A single PWM control cycle T is divided into four working stages, and the four working stages appear as a whole. When output begins, Q0 of bridge arm A and Q2 of bridge arm B are turned on first, serving as the first valid drive waveform within the cycle; when output stops, Q1 of bridge arm A and Q3 of bridge arm B are turned off, serving as the last set of turn-off waveforms at the end of the cycle.

4. The inverter welding machine PWM control method according to claim 3, characterized in that, The four operating stages within a single PWM control cycle T are as follows: Phase 1: Q0 of bridge arm A and Q2 of bridge arm B are turned on first, while Q1 of bridge arm A and Q3 of bridge arm B are turned off. Phase 2: Q0 of bridge arm A is turned on, Q2 of bridge arm B is turned off, and at the same time, Q1 of bridge arm A and Q3 of bridge arm B remain off, entering the first freewheeling phase, which continues until Q1 and Q3 are turned on. Phase 3: Q1 of bridge arm A and Q3 of bridge arm B are turned on, while Q0 of bridge arm A and Q2 of bridge arm B are turned off. Phase 4: Q1 of bridge arm A is turned on, Q3 of bridge arm B is turned off, and at the same time, Q0 of bridge arm A and Q2 of bridge arm B remain off, entering the second freewheeling phase, thus completing a full cycle.

5. The inverter welding machine PWM control method according to claim 4, characterized in that, In the first stage, the PWM drive turns on Q0 and Q2 simultaneously, while turning off Q1 and Q3. The path of current Ip in this stage is as follows: The current flows from the P+ bus to Q0, then through the freewheeling capacitor C, then through the primary side of the transformer, and finally through Q2 to reach N-; where the P+ bus is connected to the positive terminal of the three-phase rectifier bridge, and the N- bus is connected to the negative terminal of the three-phase rectifier bridge. At this time, the DC voltage Uin output by the three-phase rectifier bridge is applied to the primary winding of the transformer; transformer primary voltage = +Uin, the transformer is positively magnetized; A positive voltage is induced on the secondary side of the transformer, which turns on the corresponding rectifier diode and delivers energy to the load.

6. The PWM control method for an inverter welding machine according to claim 5, characterized in that, In the second stage, the PWM drive turns Q0 on, turns Q2 off, and keeps Q1 and Q3 off. When Q0 of bridge arm A is turned on, the voltage at point A will be equal to the voltage of the P+ bus. The transformer primary current cannot change abruptly. The voltage at point B is greater than the voltage of the P+ bus. At this time, the current will naturally switch to the diode D3 that is connected in antiparallel with Q3. Point A is located between Q0 and Q1; point B is located between Q2 and Q3 and is connected to one end of the transformer. The end of the transformer's primary winding connected to point B is defined as end B, and the other end of the transformer's primary winding is... end; The path of the second current Ip in this stage is as follows: from terminal B of the transformer, it first goes to diode D3, then to the P+ bus, then through Q0 to point A, then into the freewheeling capacitor C, and finally back to the transformer. end; At this time, Q0 and diode D3 are turned on, clamping the voltage across the primary side of the transformer at P+, i.e., +Uin; therefore, the voltage across the primary side of the transformer is... 0 At this point, the power supply and load are decoupled, and no energy is transferred. The residual magnetic energy of the transformer core charges the freewheeling capacitor C, forming an LC resonant circuit. The oscillating current continuously circulates within the power bridge arm and the transformer winding loop and gradually decays.

7. The inverter welding machine PWM control method according to claim 4, characterized in that, In the third stage, the PWM drive turns on Q1 and Q3 simultaneously, while turning off Q0 and Q2. The paths of the three currents Ip in this stage are as follows: The current flows from the P+ busbar through Q3, then through the primary side of the transformer, then through the freewheeling capacitor C, and finally through Q1 to reach N-; where the P+ busbar is connected to the positive terminal of the three-phase rectifier bridge, and the N- busbar is connected to the negative terminal of the three-phase rectifier bridge. At this time, the DC voltage Uin output by the three-phase rectifier bridge is applied to the primary winding of the transformer; transformer primary voltage = -Uin, the transformer is reverse magnetized; A negative voltage is induced on the secondary side of the transformer, and another set of rectifier diodes conducts, continuing to supply energy to the load.

8. The inverter welding machine PWM control method according to claim 7, characterized in that, In the fourth stage, the PWM drive turns Q1 on and Q3 off, while Q0 and Q2 remain off. When Q1 in bridge arm A is turned on, the voltage at point A will be equal to the N-bus voltage. Since the transformer primary current cannot change abruptly, the voltage at point B is less than the N-bus voltage. At this time, the current will naturally switch to diode D2, which is connected in anti-parallel with Q2. Point A is located between Q0 and Q1; point B is located between Q2 and Q3 and is connected to one end of the transformer. The end of the transformer's primary winding connected to point B is defined as end B, and the other end of the transformer's primary winding is... end; The paths of the four currents Ip in this stage are as follows: from the transformer... The current first enters the freewheeling capacitor C, then reaches point A, then passes through Q1 to reach the N-bus, then passes through diode D2, and finally returns to the B terminal of the transformer; At this time, Q1 and diode D2 are turned on, clamping the voltage across the primary side of the transformer at N-, i.e., -Uin; therefore, the voltage across the primary side of the transformer is... 0 At this point, the power supply and load are decoupled, and no energy is transferred. The residual magnetic energy of the transformer core charges the freewheeling capacitor, forming an LC resonant circuit. The oscillating current continuously circulates within the power bridge arm and the transformer winding loop and gradually decays.

9. The inverter welding machine PWM control method according to claim 3, characterized in that, When the output stops, the PWM pulse width of the switching transistors Q2 and Q3 in bridge arm B is gradually reduced until it reaches 0, so that the voltage of the freewheeling capacitor C is fully discharged. The specific operation is as follows: After acquiring the stop output signal, record the last pulse width Ton before the Q3 stop output signal. Preset N discharge cycles. In each discharge cycle, Q2 and Q3 decrease by each pulse width. The magnitude decreases until it decreases to 0.

10. An inverter welding machine, characterized in that, The inverter welding machine PWM control method described in any one of claims 1 to 9 is used for PWM control.

Images