An underwater multifunctional automated welding repair and additive manufacturing system

Abstract

The application provides an underwater multifunctional automatic welding repair and additive manufacturing system, which comprises an ultrahigh-frequency power supply, a wire feeding device, an underwater welding robot and an industrial computer; the ultrahigh-frequency power supply comprises a power circuit, a control circuit and a wire feeding linkage system; the power circuit comprises a SiC full-bridge inverter current conversion module; the SiC full-bridge inverter current conversion module is controlled through a SiC high-frequency drive circuit; a segmented incremental PID algorithm is adopted to control the square wave signal output by the SiC high-frequency drive circuit according to the sampling result obtained by an output current sampling feedback circuit, so that current closed-loop correction is realized; in the welding process, real-time arc voltage signals are collected and transmitted to a wire feeding minimum control system; when the real-time arc voltage signals deviate from the target value, a segmented incremental PI algorithm is adopted to adjust the wire feeding drive signal, so that arc voltage closed-loop correction is realized. The system realizes "arc voltage-current double closed-loop control", deeply controls the welding arc, and thus improves the welding effect.

Description

Technical Field

[0001] This invention relates to the field of underwater welding technology, and more specifically, to an underwater multifunctional automated welding repair and additive manufacturing system. Background Technology

[0002] During the service life of marine engineering equipment, surface corrosion and cracking are common problems, requiring efficient and reliable repair technologies to ensure the integrity and reliability of the equipment structure. Underwater arc additive repair and manufacturing technology, due to its advantages of high forming quality, high forming efficiency, and flexible operation, has become a key core technology in fields such as underwater operation and maintenance of offshore wind power, oil and gas pipeline laying, and offshore ship maintenance, possessing broad application prospects and significant research value.

[0003] However, manual underwater welding operations are highly dangerous. Divers are affected by factors such as water pressure, depth, turbulence, and darkness, resulting in low welding efficiency, high costs, poor weld quality, and a high risk of accidents. Furthermore, existing specialized underwater welding robots are limited in variety, complex in design, technically demanding, and expensive, making them unsuitable as alternatives to traditional manual welding in complex curved surfaces and other challenging environments. Meanwhile, mainstream welding power supplies on the market have low control precision and lack specialized equipment or process experience for underwater welding, failing to meet the high-efficiency and high-quality welding requirements under complex and extreme underwater conditions. Current automated underwater welding solutions largely focus on achieving automated welding processes, lacking control over the depth of the welding arc. Consequently, welding results are poor under complex and extreme underwater conditions, and effective adjustments are difficult to make in a short time.

[0004] To address these issues, a fully automated underwater multifunctional welding repair and additive manufacturing system and its welding operation method suitable for complex and extreme working conditions are proposed to meet the underwater in-situ repair and manufacturing needs of key components of offshore wind power, key structures of offshore oil and gas platforms, and ship hulls in practical engineering projects. Summary of the Invention

[0005] To overcome the shortcomings and deficiencies of existing technologies, the present invention aims to provide an underwater multifunctional automated welding repair and additive manufacturing system. This system employs real-time tracking of the welding arc and converts it into arc voltage parameters. It uses a segmented incremental PI algorithm to adjust the wire feed speed in real time to maintain arc voltage stability, achieving closed-loop arc voltage control. Simultaneously, it can acquire welding current parameters in real time and achieve closed-loop current control through a segmented incremental PID algorithm, realizing "dual closed-loop control of arc voltage and current," thereby achieving in-depth control of the welding arc and improving welding performance.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: an underwater multifunctional automated welding repair and additive manufacturing system, comprising an ultra-high frequency power supply, a wire feeding device, an underwater welding robot, and an industrial control computer;

[0007] The ultra-high frequency power supply includes a power circuit, a control circuit, and a wire feeding linkage system;

[0008] The power circuit includes a power frequency rectifier and filter module, a voltage regulator module, a SiC full-bridge inverter module, a high-frequency transformer module, and a full-bridge rectifier and smoothing module connected in sequence. The SiC full-bridge inverter module uses SiC MOSFETs to form a full-bridge inverter topology. The control circuit outputs four high-frequency drive signals through four sets of SiC high-frequency drive circuits to drive the four bridge arms of the full-bridge inverter topology one-to-one.

[0009] The output terminal of the power circuit is connected to an output current sampling feedback circuit to sample the current and voltage output by the power circuit; the control circuit uses a segmented incremental PID algorithm to control the PWM square wave signal output by the SiC high-frequency drive circuit based on the sampling result obtained from the output current sampling feedback circuit, thereby realizing current closed-loop correction.

[0010] The control circuit is connected to the wire feeding linkage system via a wire feeding minimum control system. The control circuit sends wire feeding parameters to the wire feeding minimum control system. During welding, the wire feeding minimum control system outputs a motor direction signal and a wire feeding drive signal to the wire feeding motor of the wire feeding device through the wire feeding linkage system. It also uses an arc voltage closed-loop module to collect, convert, and filter real-time arc voltage signals and transmit them to the wire feeding minimum control system. The wire feeding minimum control system determines whether the real-time arc voltage signal deviates from the target value. When the real-time arc voltage signal deviates from the target value, the wire feeding minimum control system uses a segmented incremental PI algorithm to adjust the wire feeding drive signal, thereby achieving arc voltage closed-loop correction.

[0011] This invention employs an arc voltage closed-loop module to track the welding arc in real time and convert it into arc voltage parameters for acquisition. The wire feeding minimum control system and the wire feeding linkage system use a segmented incremental PI algorithm to adjust the wire feeding speed in real time to maintain arc voltage stability, thereby achieving arc voltage closed-loop control. At the same time, the ultra-high frequency power supply can acquire welding current parameters in real time and achieve current closed-loop control through a segmented incremental PID algorithm. The entire system can achieve "arc voltage-current dual closed-loop control".

[0012] Preferably, the piecewise incremental PID algorithm is as follows:

[0013]

[0014] in, The control variable for the output PWM square wave signal; The power circuit output current error values ​​at times k-2, k-1, and k are respectively.

[0015] The coefficients of P, I, and D are respectively. Based on the power circuit output current error value Perform adaptive adjustments:

[0016]

[0017]

[0018]

[0019] in, , , These represent the values ​​of the P parameter; These represent the values ​​of parameter D; Represents the value of parameter I; These represent different error ranges.

[0020] This invention employs a piecewise incremental PID algorithm. For the problem of large current oscillations caused by extreme and complex operating conditions, when the deviation between the true value and the target value is large, the proportional action is significantly increased to make the true value quickly approach the target value, while the integral action is reduced to suppress overshoot and oscillation. When the deviation between the true value and the target value is small, the proportional action is significantly reduced and the derivative action is increased to suppress the output current stress, while the proportion of the integral action is increased to improve the steady-state accuracy of the output current.

[0021] This invention enables the actual output current to quickly approach the target value and reduces the rate of change of current as it approaches the target current. It suppresses overshoot and oscillation while ensuring a sufficiently short response time, and eliminates steady-state errors, achieving rapid and high-precision control of the output current. The dynamic response time for the output current to rise from 0A to 400A is less than 200μs, and the output accuracy is within 1% across the entire range.

[0022] Preferably, the SiC full-bridge inverter module includes: eight SiC MOSFETs; every two SiC MOSFETs are connected in parallel, so that the eight SiC MOSFETs are divided into four groups of MOSFET switching groups; the four groups of MOSFET switching groups together form a full-bridge inverter topology, and each of the four groups of MOSFET switching groups is connected in parallel with an RC snubber circuit; the output terminal of the full-bridge inverter topology is connected to the primary side of the high-frequency transformer module.

[0023] The full-bridge rectifier smoothing module includes: eight Schottky diode modules, a filter capacitor bank, and a filter inductor L102; every two Schottky diode modules are connected in parallel in the same direction, so that the eight Schottky diode modules form four groups of diode bridge arms; the four groups of diode bridge arms together form a full-bridge rectifier topology, and each is connected in parallel with an RC snubber circuit II;

[0024] The secondary side of the high-frequency transformer module is connected to the input terminal of the full-bridge rectifier topology; the output terminal of the full-bridge rectifier topology is connected to the filter capacitor bank through the filter inductor L102; the filter capacitor bank is connected to the external arc load.

[0025] The control circuit outputs four high-frequency drive signals through four SiC high-frequency drive circuits to drive four MOSFET switch groups one-to-one respectively.

[0026] This invention employs a parallel structure for the power devices in each arm of the SiC full-bridge inverter module and the full-bridge rectifier smoothing module. This greatly increases the circuit's current carrying capacity while ensuring that the device size remains almost unchanged, thereby increasing the welding output current of this invention and enabling high-power welding.

[0027] Preferably, the method for dividing the eight SiC MOSFETs into four MOSFET switch groups is as follows: N SiC MOSFETs, N≥8, are subjected to dynamic and static characteristic tests using an LCR bridge and a dual-pulse test circuit to obtain the dynamic and static characteristic test results. These results include turn-off / on resistance, gate-source threshold voltage, switching time, switching loss, current / voltage change rate, and output characteristic curves. The similarity of the dynamic and static characteristic test results for each SiC MOSFET is evaluated, and the two SiC MOSFETs with the most similar dynamic and static characteristic test results are grouped into the same MOSFET switch group. The four MOSFET switch groups with the smallest difference in dynamic and static characteristic test results are used as the four MOSFET switch groups for the SiC full-bridge inverter module. Due to manufacturing errors, the SiC MOSFET grouping method of this invention can minimize the negative impact of manufacturing errors.

[0028] The four high-frequency drive signals output by the four SiC high-frequency drive circuits are PWM square wave signals; the PWM square wave signals output by the MOSFET switch groups at the upper part of the two series-connected bridge arms of the full-bridge inverter topology are complementary, and the PWM square wave signals output by the MOSFET switch groups at the lower part of the two series-connected bridge arms of the full-bridge inverter topology are complementary; the PWM square wave signals output by the diagonally opposite MOSFET switch groups of the full-bridge inverter topology are the same.

[0029] Preferably, each pair of SiC MOSFETs connected in parallel means that, within the same MOSFET switching group, the drains of the two SiC MOSFETs are connected to each other, and the Kelvin sources of the two SiC MOSFETs are connected to each other.

[0030] Each SiC high-frequency drive circuit includes an isolated high-frequency drive module (model APD202), diode D2, resistors R1, R3, R4, R5, R6, R7, R8, capacitors C3 and C4; the interfaces Reset, Fault, VI+, VI-, VDD, and GND of the isolated high-frequency drive module are connected to the control circuit respectively.

[0031] The two SiC MOSFETs in the same MOSFET switching group are SiC MOSFET 1 and SiC MOSFET 2. The interface Desat of the isolated high-frequency drive module is connected to the drains of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface Clamp of the isolated high-frequency drive module is connected to the gates of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface Vo-off of the isolated high-frequency drive module is connected to the gate of SiC MOSFET 1 via a series diode D2 and resistor R3, and also to the gate of SiC MOSFET 2 via a series diode D1 and resistor R6. The interface Vo-on of the isolated high-frequency drive module is connected to the gate of SiC MOSFET 1 via resistor R4, and also to the gate of SiC MOSFET 2 via resistor R7. The interface VSS of the isolated high-frequency drive module is connected to the Kelvin source of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface VSS of the isolated high-frequency drive module is also connected to the SiC MOSFET 2 via a parallel resistor R5 and capacitor C3. The gate of MOSFET 1 is connected; the interface VSS of the isolated high-frequency drive module is also connected to the gate of SiC MOSFET 2 through a parallel resistor R8 and capacitor C4.

[0032] Preferably, the piecewise incremental PI algorithm is as follows:

[0033]

[0034] in, The output signal controls the variable; These are the arc voltage signal error values ​​at times k-1 and k, respectively;

[0035] These are the coefficients of P and I, respectively; Based on the arc voltage signal error value Perform adaptive adjustments:

[0036]

[0037]

[0038] in, , , These represent the values ​​of the P parameter; Represents the value of parameter I; These represent different error ranges.

[0039] Preferably, the arc voltage closed-loop module includes a sampling resistor circuit, a voltage sensor U310, a voltage follower circuit, an absolute value circuit, and a clamping circuit connected in sequence; wherein, the sampling resistor circuit is connected to the output terminal of the power circuit and is used to convert the voltage signal into a sampling current; the voltage sensor U310 amplifies the sampling current of the sampling resistor circuit and converts it into a voltage signal; the voltage follower circuit is used to suppress common-mode interference of the voltage signal; the absolute value circuit converts the negative voltage component of the voltage signal into a positive voltage component with the same amplitude; the clamping circuit clamps the voltage signal output by the absolute value circuit and outputs an arc voltage signal.

[0040] Preferably, it further includes an arc vision sensing system; the arc vision sensing system includes a CMOS industrial camera, an underwater welding torch detection lens, a magnetically controlled arc device, and an integrated microprocessor; the underwater welding torch detection lens is connected to the CMOS industrial camera and is used to capture high-exposure underwater welding arc images; the CMOS industrial camera is connected to the integrated microprocessor; the magnetically controlled arc device includes a gun sleeve, an air inlet, a fixing and positioning mounting hole, a magnetic core, an excitation coil, magnetic poles, and a drive commutation module; the gun sleeve is installed on the welding torch head and serves to fix and position it, and the gun sleeve is provided with an air inlet for water and air to be introduced; the fixing and positioning mounting hole is used to fix the welding torch. The relative positions of the welding gun and the magnetically controlled arc device; the magnetic core and excitation coil are used to excite the electromagnetic field, and the magnetic field energy acts on the welding arc through the magnetic poles; the CMOS industrial camera transmits the underwater welding arc image to the integrated microprocessor in real time; the welding arc deviation is determined based on the underwater welding arc image; the integrated microprocessor generates a control signal based on the welding arc deviation and transmits it to the drive converter module; the drive converter module receives the control signal and amplifies it into a controllable current input to the excitation coil, which excites the electromagnetic field through the magnetic core, and adjusts the magnitude and direction of the magnetic field by adjusting the magnitude and direction of the controllable current, thereby realizing the welding arc attitude correction.

[0041] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0042] 1. This invention employs an arc voltage closed-loop module to track the welding arc in real time and convert it into arc voltage parameters for acquisition. The wire feeding minimum control system and the wire feeding linkage system use a segmented incremental PI algorithm to adjust the wire feeding speed in real time to maintain arc voltage stability, thereby achieving arc voltage closed-loop control. At the same time, the ultra-high frequency power supply can acquire welding current parameters in real time and achieve current closed-loop control through a segmented incremental PID algorithm. The entire system can achieve "arc voltage-current dual closed-loop control".

[0043] 2. This invention employs an arc vision sensing system, which can automatically locate and continuously track the welding area, capture and predict the welding arc posture, and correct the welding arc posture through an external electromagnetic field, thereby achieving magnetic closed-loop control of the welding arc. This is beneficial for improving the stability of the welding process and enhancing the weld formation quality.

[0044] 3. Based on the traditional full-bridge inverter topology and full-bridge rectifier topology, this invention adopts a parallel structure for the power devices of each bridge arm of the SiC full-bridge inverter converter module and the full-bridge rectifier smoothing module, which greatly increases the circuit's overcurrent capacity while ensuring that the device size remains almost unchanged.

[0045] 4. The welding system proposed in this invention can achieve fast, efficient, and high-quality fully automatic welding operations in extreme and complex underwater working conditions, effectively solving problems such as difficult underwater welding, high welding costs, poor weld quality, low efficiency and high risk of manual welding in marine engineering; it can realize multiple functions such as DC, pulse and double pulse welding, with a wide process window, and at the same time, it has high output accuracy and high output power, and has strong adaptability to different materials and working conditions. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention;

[0047] Figure 2 This is a schematic diagram of the ultra-high frequency power supply for the underwater multifunctional automated welding repair and additive manufacturing system of the present invention.

[0048] Figure 3 This is a circuit diagram of the SiC high-frequency drive circuit of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention.

[0049] Figure 4 This is a circuit diagram of the output current sampling feedback circuit of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention;

[0050] Figure 5 This is a circuit diagram of the wire feeding linkage system of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention;

[0051] Figure 6This is a circuit diagram of the arc voltage closed-loop module of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention.

[0052] Figure 7 This is a schematic diagram of the arc vision sensing system of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention;

[0053] Figure 8 This is a welding operation flowchart of the underwater multifunctional automated welding repair and additive manufacturing system of the present invention. Detailed Implementation

[0054] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0055] Example

[0056] like Figure 1 As shown in the figure, this embodiment of an underwater multifunctional automated welding repair and additive manufacturing system includes an ultra-high frequency power supply, a wire feeding device, an underwater welding robot, and an industrial control computer.

[0057] Underwater welding robots can employ existing technologies, such as the Chinese invention patent application "A magnetic wheel adsorption type underwater MIG welding robot" (publication number: CN 116118887A).

[0058] like Figure 2 As shown, the ultra-high frequency power supply includes a power circuit for outputting and modulating welding current, a control circuit based on an STM32 microprocessor, an independently operable human-machine interface system, a wire feeding linkage system for controlling the wire feeding device, a gas supply device, and a fault detection system.

[0059] The control circuit includes peripherals such as PWM, ADC, RS485, and CAN. Its working principle is as follows: The main welding process programs, including DC welding, pulse welding, and dual-pulse welding, are written using the FreeRTOS real-time operating system. These programs are then used to control the peripherals in parallel through interrupts and modular programming, achieving automated control during the welding process. Specifically, the PWM module outputs a high-frequency drive signal to drive the SiCMOSFET, enabling high-frequency inversion; the ADC module acquires, converts, and filters the current and voltage signals transmitted by the current sampling feedback circuit and the arc voltage closed-loop module, using segmented incremental PID and PI algorithms for output monitoring and closed-loop control; the RS485 module communicates with the host industrial computer and human-machine interface system via the unified MODBUS-RTU data frame protocol; and the CAN module sends wire feeding parameters to the wire feeding linkage system in real time.

[0060] The power circuit includes a power frequency rectifier and filter module, a voltage regulator module, a SiC full-bridge inverter module, a high-frequency transformer module L101, and a full-bridge rectifier and smoothing module connected in sequence. The SiC full-bridge inverter module uses SiC MOSFETs to form a full-bridge inverter topology. The control circuit outputs four high-frequency drive signals through four sets of SiC high-frequency drive circuits to drive the four bridge arms of the full-bridge inverter topology one-to-one.

[0061] The SiC full-bridge inverter module includes eight SiC MOSFETs Q101 to Q108. Two SiC MOSFETs are connected in parallel, dividing the eight SiC MOSFETs into four MOSFET switch groups. These four MOSFET switch groups together form a full-bridge inverter topology, and each of the four MOSFET switch groups is connected in parallel with an RC snubber circuit. The output of the full-bridge inverter topology is connected to the primary side of the high-frequency transformer module. The control circuit outputs four high-frequency drive signals through four SiC high-frequency drive circuits to drive each of the four MOSFET switch groups one-to-one.

[0062] The full-bridge rectifier smoothing module includes: eight Schottky diode modules D101-D108, a filter capacitor bank, and a filter inductor L102; every two Schottky diode modules are connected in parallel in the same direction, forming four groups of diode bridge arms; the four groups of diode bridge arms together form the full-bridge rectifier topology, and each is connected in parallel with an RC snubber circuit. The secondary side of the high-frequency transformer module is connected to the input terminal of the full-bridge rectifier topology; the output terminal of the full-bridge rectifier topology is connected to the filter capacitor bank through the filter inductor L102; the filter capacitor bank is connected to an external arc load.

[0063] The power circuit works as follows: the 540V DC power after three-phase rectification and regulation passes through the SiC full-bridge inverter module, and the SiC high-frequency drive circuit generates two pairs of complementary PWM drive signals ranging from -3V to 15V, which are applied to the SiC MOSFETs. Q101 to Q108 enable Q101, Q102, Q107, and Q108 to be turned on or off simultaneously, while Q103, Q104, Q105, and Q106 are turned off or on simultaneously, forming a full-bridge inverter topology. This converts 540V DC power into high-frequency AC power with a maximum frequency of 200kHz, which is then coupled to the full-bridge rectifier smoothing module via the high-frequency transformer module L101. During coupling, the voltage decreases, the current increases, and the power remains constant. D101 to D108 form a full-bridge rectifier topology, converting the high-frequency AC power back into DC power. This DC power is then regulated by the filter inductor L102 and the voltage by the filter capacitors C122, C123, and C124 before being output to the workpiece to be welded. The RC snubber circuit protects the power switching transistors and rectifier diodes, ensuring a smooth commutation process.

[0064] The method for dividing eight SiC MOSFETs into four MOSFET switch groups is as follows: N SiC MOSFETs (N≥8) are tested for dynamic and static characteristics using an LCR bridge and a dual-pulse test circuit. The dynamic and static characteristic test results include turn-off / on resistance, gate-source threshold voltage, switching time, switching loss, current / voltage change rate, and output characteristic curves. The similarity of the dynamic and static characteristic test results of each SiC MOSFET is evaluated, and the two SiC MOSFETs with the most similar dynamic and static characteristic test results are grouped into the same MOSFET switch group. The four MOSFET switch groups with the smallest difference in dynamic and static characteristic test results are used as the four MOSFET switch groups for the SiC full-bridge inverter module.

[0065] Parallel connection of two SiC MOSFETs means that, within the same MOSFET switching group, the drains of the two SiC MOSFETs are connected to each other, and the Kelvin sources of the two SiC MOSFETs are connected to each other.

[0066] like Figure 3 As shown, each SiC high-frequency drive circuit includes an isolated high-frequency drive module of model APD202, diode D2, resistors R1, R3, R4, R5, R6, R7, R8, capacitors C3 and C4; the interfaces Reset, Fault, VI+, VI-, VDD, and GND of the isolated high-frequency drive module are respectively connected to the control circuit.

[0067] The two SiC MOSFETs in the same MOSFET switching group are SiC MOSFET 1 and SiC MOSFET 2. The interface Desat of the isolated high-frequency drive module is connected to the drains of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface Clamp of the isolated high-frequency drive module is connected to the gates of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface Vo-off of the isolated high-frequency drive module is connected to the gate of SiC MOSFET 1 via a series diode D2 and resistor R3, and also to the gate of SiC MOSFET 2 via a series diode D1 and resistor R6. The interface Vo-on of the isolated high-frequency drive module is connected to the gate of SiC MOSFET 1 via resistor R4, and also to the gate of SiC MOSFET 2 via resistor R7. The interface VSS of the isolated high-frequency drive module is connected to the Kelvin source of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface VSS of the isolated high-frequency drive module is also connected to the SiC MOSFET 2 via a parallel resistor R5 and capacitor C3. The gate of MOSFET 1 is connected; the interface VSS of the isolated high-frequency drive module is also connected to the gate of SiC MOSFET 2 through a parallel resistor R8 and capacitor C4.

[0068] This invention employs an isolated high-frequency drive method to connect the control circuit and the SiC high-frequency drive circuit, as well as the SiC high-frequency drive circuit and the SiC full-bridge inverter module.

[0069] Resistors R4 and R7 are both turn-on resistors with the same resistance value to enable synchronous turn-on of two SiC MOSFETs in the same MOSFET switch group; resistors R3 and R6 are both turn-off resistors with the same resistance value to enable synchronous turn-off of two SiC MOSFETs in the same MOSFET switch group; the resistance values ​​of the turn-on and turn-off resistors are set according to the requirement that the turn-on time and turn-off time of the SiC MOSFETs be consistent.

[0070] The four high-frequency drive signals output by the four SiC high-frequency drive circuits are PWM square wave signals; the PWM square wave signals output by the MOSFET switch groups at the upper part of the two series-connected bridge arms of the full-bridge inverter topology are complementary, and the PWM square wave signals output by the MOSFET switch groups at the lower part of the two series-connected bridge arms of the full-bridge inverter topology are complementary; the PWM square wave signals output by the diagonally opposite MOSFET switch groups of the full-bridge inverter topology are the same.

[0071] The output terminal of the power circuit is connected to an output current sampling feedback circuit to sample the current and voltage output by the power circuit. Based on the sampling results obtained from the output current sampling feedback circuit, the control circuit uses a segmented incremental PID algorithm to control the PWM square wave signal output by the SiC high-frequency drive circuit, thereby realizing current closed-loop correction.

[0072] The piecewise incremental PID algorithm is as follows:

[0073]

[0074] in, The control variable for the output PWM square wave signal; The power circuit output current error values ​​at times k-2, k-1, and k are respectively.

[0075] The coefficients of P, I, and D are respectively. Based on the power circuit output current error value Perform adaptive adjustments:

[0076]

[0077]

[0078]

[0079] in, , , These represent the values ​​of the P parameter; These represent the values ​​of parameter D; Represents the value of parameter I; These represent different error ranges.

[0080] Power circuit output current error value This can be obtained using a Hall sensor signal obtained through an output current sampling feedback circuit, such as... Figure 4 As shown.

[0081] During the welding process, the minimum control system for wire feeding outputs a motor direction signal and a wire feeding drive signal to the wire feeding motor of the wire feeding device through the wire feeding linkage system. It also uses an arc voltage closed-loop module to collect, convert, and filter the real-time arc voltage signal and transmit it to the minimum control system for wire feeding. The minimum control system for wire feeding determines whether the real-time arc voltage signal deviates from the target value. When the real-time arc voltage signal deviates from the target value, the minimum control system for wire feeding uses a segmented incremental PI algorithm to adjust the wire feeding drive signal, thereby achieving arc voltage closed-loop correction.

[0082] The control circuit is connected to the minimum wire feeding control system via CAN communication; the minimum wire feeding control system is connected to the wire feeding linkage system via signal connection, serving as the control module of the wire feeding linkage system; the control circuit sends wire feeding parameters to the minimum wire feeding control system. For example... Figure 5 As shown, the wire feeding linkage system includes chips ULN301, U304, K301, U307, and U308. The minimum control system for wire feeding outputs a 0-3.3V pulse signal to the wire feeding linkage system. After being inverted by chip ULN301 and isolated and amplified by chips U307 / U308, the wire feeding linkage system generates a 0-15V pulse signal LOW. The +15V high-level signal HIGH and the pulse signal LOW are differentially divided to form the wire feeding drive signal for the wire feeding motor. The wire feeding motor adopts a common anode connection, and the wire feeding drive signal has two speed regulation modes: PWM pulse width speed regulation when the wire feeding motor is a brushless motor, and PFM frequency conversion speed regulation when the wire feeding motor is a stepper motor. The control circuit transmits the wire feeding parameters through a CAN network digital communication circuit built by chip U313. The signal is sent to the wire feeding minimum control system. During welding, the wire feeding minimum control system outputs motor direction signals and wire feeding drive signals to the wire feeding motor of the wire feeding device through a wire feeding direction circuit built by chips ULN301, U304, and K301. Since the welding arc voltage and welding arc length are positively correlated, while the welding arc length and wire feeding speed are negatively correlated, the welding arc voltage can be adjusted in real time by adjusting the motor speed. The arc voltage closed-loop module is used to collect, convert, and filter the real-time arc voltage signal and transmit it to the wire feeding minimum control system. When the real-time arc voltage signal deviates from the target value, the wire feeding minimum control system uses a segmented incremental PI algorithm to adjust the wire feeding drive signal in a timely manner, thereby quickly adjusting the wire feeding motor speed to ensure the stability of the welding arc and realize arc voltage closed-loop correction.

[0083] like Figure 6 As shown, the arc voltage closed-loop module includes a sampling resistor circuit, a voltage sensor U310, a voltage follower circuit, an absolute value circuit, and a clamping circuit connected in sequence. The sampling resistor circuit, composed of R321, R322, R323, and R324, is connected to the output of the power circuit and converts the voltage signal into a sampling current. The voltage sensor U310 amplifies the sampling current and converts it into a voltage signal through resistor R327. U311, C342, and C338 together form a voltage follower circuit to suppress common-mode interference in the voltage signal. U312, R320, R325, R326, R330, D306, D308, C337, and C342 together form an absolute value circuit that converts the negative voltage component of the voltage signal into a positive voltage component of the same amplitude. The clamping circuit D307 clamps the voltage signal output from the absolute value circuit, outputting an arc voltage signal.

[0084] The piecewise incremental PI algorithm is as follows:

[0085]

[0086] in, The output signal controls the variable; These are the arc voltage signal error values ​​at times k-1 and k, respectively;

[0087] These are the coefficients of P and I, respectively; Based on the arc voltage signal error value Perform adaptive adjustments:

[0088]

[0089]

[0090] in, , , These represent the values ​​of the P parameter; Represents the value of parameter I; These represent different error ranges.

[0091] The system of the present invention preferably further includes an arc vision sensing system; such as Figure 7 As shown, the arc vision sensing system includes a CMOS industrial camera 3, an underwater welding torch inspection lens 2, a magnetically controlled arc device, and an integrated microprocessor 4. The underwater welding torch inspection lens is connected to the CMOS industrial camera 3 and is used to capture high-exposure underwater welding arc images; the CMOS industrial camera 3 is connected to the integrated microprocessor 4; the magnetically controlled arc device includes a gun sleeve 6, an air inlet 7, a fixing and positioning mounting hole 8, a magnetic core 9, an excitation coil 10, magnetic poles 11, and a drive commutation module 5; the gun sleeve 6 is mounted on the welding torch head and serves to fix and position it; the gun sleeve 6 has an air inlet 7 for water and air intake; the fixing and positioning mounting hole 8 is used to fix the relative position of the welding torch and the magnetically controlled arc device; the magnetic core 9 and the excitation coil 10 are used to generate an electromagnetic field, and the magnetic field energy acts on the welding arc 1 through the magnetic poles 11; The CMOS industrial camera 3 transmits underwater welding arc images to the integrated microprocessor 4 in real time; it determines whether the welding arc 1 has deviated based on the underwater welding arc images; the integrated microprocessor 4 generates a control signal based on the deviation of the welding arc 1 and transmits it to the drive converter module 5; the drive converter module 5 receives the control signal and amplifies it into a controllable current input to the excitation coil 10, which, in conjunction with the magnetic core 9, excites an electromagnetic field, thereby adjusting the magnitude and direction of the magnetic field by adjusting the magnitude and direction of the controllable current, thus achieving welding arc posture correction.

[0092] The welding operation method of the underwater multifunctional automated welding repair and additive manufacturing system in this embodiment is as follows: Figure 8 As shown, it includes the following steps:

[0093] S1. Connect the UHF power supply and underwater welding robot to the industrial control computer and establish communication. Specifically: the industrial control computer acts as the MODBUS master, responsible for overall system scheduling; the UHF power supply acts as slave 1, responsible for receiving welding parameters sent by the master and feeding back real-time welding parameters to the master; the underwater welding robot acts as MODBUS slave 2, responsible for receiving motion commands from the master and feeding back the working status of modules such as motors, cameras, and sensors to the master; before the task begins, the underwater welding robot starts a self-test and feeds back the self-test status to the industrial control computer, while the UHF power supply continuously monitors for faults; in this embodiment, the underwater stepper motor and wire feeding and clamping wheel mechanism are directly mounted on the underwater welding robot.

[0094] S2. Deploy the underwater welding robot. The industrial control computer sends motion trajectory instructions to the underwater welding robot. The underwater welding robot moves according to the trajectory. The underwater panoramic camera captures and feeds back the surrounding environment to the industrial control computer in real time. The ultrasonic probe determines whether there are obstacles around the underwater welding robot. The operator continuously optimizes the motion trajectory until it reaches the vicinity of the area to be welded.

[0095] S3. The industrial control computer precisely controls the cross slide structure of the underwater welding robot, so that the local dry welding torch just covers the area to be welded. The operator plans the welding path according to the environmental video. In this embodiment, the underwater welding robot slide structure, together with the chassis itself, can realize four degrees of freedom of movement in a local area, which can meet the space requirements of most underwater structural components.

[0096] S4. The ultra-high frequency power supply is connected to an external three-phase AC input power supply; the industrial control computer sends the welding parameters and a welding start signal to the ultra-high frequency power supply; upon receiving the welding start signal, the ultra-high frequency power supply executes the welding program according to the set procedure, specifically:

[0097] S4-1. Open the drain gas solenoid valve. High-pressure drain gas fills the local dry welding torch, and the water inside the cavity is discharged from the gap between the drain cover and the workpiece to be welded.

[0098] S4-2. Open the protective gas solenoid valve. High-pressure protective gas is sprayed out from the welding nozzle through the welding torch to prevent the area to be welded from being oxidized or interfered with by the drainage gas.

[0099] S4-3, the ultra-high frequency power switch tube is working, and the no-load output is turned on; the wire feeding linkage system controls the underwater stepper motor to slowly feed the wire until the welding wire and the workpiece contact and ignite an arc.

[0100] S4-4. After successful arc ignition, the ultra-high frequency power supply sends an arc ignition success signal and real-time welding parameters to the industrial control computer. The wire feeding linkage system controls the stepper motor to feed the wire quickly. After receiving the arc ignition signal, the industrial control computer sends welding instructions to the underwater welding robot. The underwater welding robot moves at a constant welding speed according to the planned welding path. The ultra-high frequency power supply continuously outputs welding current according to the set process. The industrial control computer and the human-machine interaction system display the welding parameters and the working status of the underwater welding robot in real time.

[0101] S4-5. After the underwater welding robot reaches the end point, the industrial control computer sends a welding completion command to the ultra-high frequency power supply. The ultra-high frequency power supply stops outputting, stops wire feeding, and delays the shutdown of protective gas and drainage gas.

[0102] S5. During the welding process, the arc vision sensing system is activated, and the CMOS camera captures and feeds back the welding arc to the microprocessor in real time. The microprocessor determines whether the welding arc needs to be corrected based on the arc posture. When correction is required, the microprocessor controls the magnetic arc device to generate an electromagnetic field through the drive circulating current module to correct the welding arc posture. The magnitude and direction of the electromagnetic field are determined by the magnitude and direction of the coil current.

[0103] S6. The ultra-high frequency power supply cuts off the three-phase AC input; the cross slide moves the welding torch to a safe position, and the operator moves the underwater welding robot to the recovery area, thus ending the entire welding process.

[0104] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. An underwater multifunctional automated welding repair and additive manufacturing system, characterized in that: This includes ultra-high frequency power supplies, wire feeding devices, underwater welding robots, and industrial control computers; The ultra-high frequency power supply includes a power circuit, a control circuit, and a wire feeding linkage system; The power circuit includes a power frequency rectifier and filter module, a voltage regulator module, a SiC full-bridge inverter module, a high-frequency transformer module, and a full-bridge rectifier and smoothing module connected in sequence. The SiC full-bridge inverter module uses SiC MOSFETs to form a full-bridge inverter topology. The control circuit outputs four high-frequency drive signals through four sets of SiC high-frequency drive circuits to drive the four bridge arms of the full-bridge inverter topology one-to-one. The output terminal of the power circuit is connected to an output current sampling feedback circuit to sample the current and voltage output by the power circuit; the control circuit uses a segmented incremental PID algorithm to control the PWM square wave signal output by the SiC high-frequency drive circuit based on the sampling result obtained from the output current sampling feedback circuit, thereby realizing current closed-loop correction. The control circuit is signal-connected to the wire feeding linkage system through the minimum wire feeding control system; the control circuit sends the wire feeding parameters to the minimum wire feeding control system. During the welding process, the wire feeding minimum control system outputs a motor direction signal and a wire feeding drive signal to the wire feeding motor of the wire feeding device through the wire feeding linkage system; and uses an arc voltage closed-loop module to collect, convert, and filter the real-time arc voltage signal and transmit it to the wire feeding minimum control system; the wire feeding minimum control system determines whether the real-time arc voltage signal deviates from the target value: when the real-time arc voltage signal deviates from the target value, the wire feeding minimum control system uses a segmented incremental PI algorithm to adjust the wire feeding drive signal, thereby realizing arc voltage closed-loop correction; The SiC full-bridge inverter module includes: eight SiC MOSFETs; every two SiC MOSFETs are connected in parallel, so that the eight SiC MOSFETs are divided into four groups of MOSFET switching groups; the four groups of MOSFET switching groups together form a full-bridge inverter topology, and each of the four groups of MOSFET switching groups is connected in parallel with an RC snubber circuit; the output terminal of the full-bridge inverter topology is connected to the primary side of the high-frequency transformer module. The full-bridge rectifier smoothing module includes: eight Schottky diode modules, a filter capacitor bank, and a filter inductor L102; every two Schottky diode modules are connected in parallel in the same direction, so that the eight Schottky diode modules form four groups of diode bridge arms; the four groups of diode bridge arms together form a full-bridge rectifier topology, and each is connected in parallel with an RC snubber circuit II; The secondary side of the high-frequency transformer module is connected to the input terminal of the full-bridge rectifier topology; the output terminal of the full-bridge rectifier topology is connected to the filter capacitor bank through the filter inductor L102; the filter capacitor bank is connected to the external arc load. The control circuit outputs four high-frequency drive signals through four SiC high-frequency drive circuits to drive four MOSFET switch groups one-to-one respectively. The method for dividing the eight SiC MOSFETs into four MOSFET switch groups is as follows: N SiC MOSFETs (N≥8) are subjected to dynamic and static characteristic tests using an LCR bridge and a dual-pulse test circuit to obtain the dynamic and static characteristic test results. These results include turn-off / on resistance, gate-source threshold voltage, switching time, switching loss, current / voltage change rate, and output characteristic curves. The similarity of the dynamic and static characteristic test results for each SiC MOSFET is evaluated, and the two SiC MOSFETs with the most similar dynamic and static characteristic test results are grouped into the same MOSFET switch group. The four MOSFET switch groups with the smallest difference in dynamic and static characteristic test results are selected as the four MOSFET switch groups for the SiC full-bridge inverter module. The four high-frequency drive signals output by the four SiC high-frequency drive circuits are PWM square wave signals; the PWM square wave signals output by the MOSFET switch groups at the upper part of the two series-connected bridge arms of the full-bridge inverter topology are complementary, the PWM square wave signals output by the MOSFET switch groups at the lower part of the two series-connected bridge arms of the full-bridge inverter topology are complementary, and the PWM square wave signals output by the diagonally opposite MOSFET switch groups of the full-bridge inverter topology are the same. Parallel connection of two SiC MOSFETs means that, within the same MOSFET switching group, the drains of the two SiC MOSFETs are connected to each other, and the Kelvin sources of the two SiC MOSFETs are connected to each other. Each SiC high-frequency drive circuit includes an isolated high-frequency drive module (model APD202), diode D2, resistors R1, R3, R4, R5, R6, R7, R8, capacitors C3 and C4; the interfaces Reset, Fault, VI+, VI-, VDD, and GND of the isolated high-frequency drive module are connected to the control circuit respectively. The two SiC MOSFETs in the same MOSFET switching group are SiC MOSFET 1 and SiC MOSFET 2. The interface Desat of the isolated high-frequency drive module is connected to the drains of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface Clamp of the isolated high-frequency drive module is connected to the gates of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface Vo-off of the isolated high-frequency drive module is connected to the gate of SiC MOSFET 1 via a series diode D2 and resistor R3, and also to the gate of SiC MOSFET 2 via a series diode D1 and resistor R6. The interface Vo-on of the isolated high-frequency drive module is connected to the gate of SiC MOSFET 1 via resistor R4, and also to the gate of SiC MOSFET 2 via resistor R7. The interface VSS of the isolated high-frequency drive module is connected to the Kelvin source of SiC MOSFET 1 and SiC MOSFET 2, respectively. The interface VSS of the isolated high-frequency drive module is also connected to the SiC MOSFET 2 via a parallel resistor R5 and capacitor C3. The gate of MOSFET 1 is connected; the interface VSS of the isolated high-frequency drive module is also connected to the gate of SiC MOSFET 2 through a parallel resistor R8 and capacitor C4.

2. The underwater multifunctional automated welding repair and additive manufacturing system according to claim 1, characterized in that: The piecewise incremental PID algorithm is as follows: ； in, The control variable for the output PWM square wave signal; The power circuit output current error values ​​at times k-2, k-1, and k are respectively. The coefficients of P, I, and D are respectively. Based on the power circuit output current error value Perform adaptive adjustments: ； ； ； in, , , These represent the values ​​of the P parameter; These represent the values ​​of parameter D; Represents the value of parameter I; These represent different error ranges.

3. The underwater multifunctional automated welding repair and additive manufacturing system according to claim 1, characterized in that: The piecewise incremental PI algorithm is as follows: ； in, The output signal controls the variable; These are the arc voltage signal error values ​​at times k-1 and k, respectively; These are the coefficients of P and I, respectively; Based on the arc voltage signal error value Perform adaptive adjustments: ； ； in, , , These represent the values ​​of the P parameter; Represents the value of parameter I; These represent different error ranges.

4. The underwater multifunctional automated welding repair and additive manufacturing system according to claim 1, characterized in that: The arc voltage closed-loop module includes a sampling resistor circuit, a voltage sensor U310, a voltage follower circuit, an absolute value circuit, and a clamping circuit connected in sequence. The sampling resistor circuit is connected to the output of the power circuit and converts the voltage signal into a sampling current. The voltage sensor U310 amplifies the sampling current from the sampling resistor circuit and converts it into a voltage signal. The voltage follower circuit suppresses common-mode interference in the voltage signal. The absolute value circuit converts the negative voltage component of the voltage signal into a positive voltage component with the same amplitude. The clamping circuit clamps the voltage signal output by the absolute value circuit and outputs an arc voltage signal.

5. The underwater multifunctional automated welding repair and additive manufacturing system according to claim 1, characterized in that: It also includes an arc vision sensing system; the arc vision sensing system includes a CMOS industrial camera, an underwater welding torch detection lens, a magnetically controlled arc device, and an integrated microprocessor; the underwater welding torch detection lens is connected to the CMOS industrial camera and is used to capture high-exposure underwater welding arc images; the CMOS industrial camera is connected to the integrated microprocessor; the magnetically controlled arc device includes a gun sleeve, an air inlet, fixing and positioning mounting holes, a magnetic core, an excitation coil, magnetic poles, and a drive commutation module; the gun sleeve is installed on the welding torch head and serves to fix and position it, and the gun sleeve has an air inlet for water and air to be introduced; the fixing and positioning mounting holes are used to fix the relative position of the welding torch and the magnetically controlled arc device; the magnetic core and excitation coil are used to excite an electromagnetic field, and the magnetic field energy acts on the welding arc through the magnetic poles; the CMOS industrial camera transmits the underwater welding arc image to the integrated microprocessor in real time; the underwater welding arc image is used to determine if the welding arc has deviated; the integrated microprocessor generates a control signal based on the welding arc deviation and transmits it to the drive commutation module; The drive converter module receives the control signal and amplifies it into a controllable current input to the excitation coil. The electromagnetic field is excited by the magnetic core. The magnitude and direction of the magnetic field can be adjusted by adjusting the magnitude and direction of the controllable current, thereby realizing the welding arc posture correction.

Images