Automatic cutting system for valve seal welding seam and control method

Abstract

According to the automatic cutting system and control method for the valve sealing weld joints, the system integrates digital twin positioning, self-adaptive clamping, dual-mode cutting, multi-dimensional monitoring and block chain tracing technologies, and high cutting precision is achieved; through a three-level safety response mechanism, the safety accident rate is greatly reduced, the scrap recovery rate is high, and dust explosion and high-temperature burning risks are effectively prevented; the system adapts to a full-size valve, the universality is improved, and the requirement for customizing tools is reduced; the block chain platform realizes tampering-free storage of cut data in a full life cycle, and generates a traceable report in combination with machine vision quality evaluation, so that the maintenance traceability requirement of nuclear power equipment is met; a self-adaptive learning algorithm dynamically optimizes a cutting path and process parameters based on historical data, and a'data driving-intelligent optimization 'closed-loop ecology is formed. According to the method, multi-dimensional breakthrough of precision, efficiency, safety and universality is achieved through technical innovation, and remarkable economic benefits and social values are achieved.

Description

Technical Field

[0001] This invention relates to the field of valve sealing weld cutting, and in particular to an automatic valve sealing weld cutting system and control method. Background Technology

[0002] There are four major technical bottlenecks in cutting traditional valve sealing welds:

[0003] Precision bottleneck: Manual hand-held cutting equipment results in a precision fluctuation of ±1.5mm, which is far higher than the ≤0.5mm weld reinforcement standard required for nuclear power equipment, and is prone to problems such as uneven weld reinforcement and deviation of cutting bevel angle.

[0004] Thermal management challenges: Controlling the heat input during cutting is difficult; local temperatures exceeding 200°C can easily lead to intergranular corrosion in the heat-affected zone of austenitic stainless steel, reducing the material's mechanical properties.

[0005] Limited compatibility: Existing clamping devices are only compatible with valves with a diameter of 200-400mm. Small-diameter valves (<100mm) and large-diameter valves (>600mm) require customized tooling, resulting in poor versatility.

[0006] Safety risks: Significant safety hazards such as flying cutting debris, high-temperature burns, and toxic gas leaks result in a high accident rate. Summary of the Invention

[0007] To address the technical problems mentioned in the background section, the present invention provides an automatic cutting system and control method for valve sealing welds, the system comprising:

[0008] Digital twin intelligent positioning module: It uses a dual-modal sensor of laser contour scanner and industrial camera to construct a three-dimensional point cloud model of valve, and achieves sub-millimeter positioning of weld seam through edge detection algorithm, and eliminates assembly error with feature point matching algorithm;

[0009] Adaptive hydraulic-mechanical hybrid clamping module: It adopts a three-jaw elastic expansion sleeve structure, realizes stepless adaptation of valve shaft diameter through hydraulic cylinder drive, and deploys a pressure sensor array in conjunction with PID control algorithm to realize adaptive adjustment of clamping force;

[0010] Dual-mode cutting execution module: includes coarse cutting mode and fine cutting mode. Coarse cutting adopts pulsed plasma arc cutting process, and fine cutting adopts ultrasonic vibration cutting, which is combined with a laser rangefinder to realize closed-loop control of cutting depth.

[0011] Multi-dimensional intelligent monitoring module: integrates acoustic emission sensor, vibration sensor, and temperature sensor to build a three-level safety response mechanism: Level I vibration warning, Level II temperature warning, and Level III emergency shutdown, and works with negative pressure adsorption + cyclone separation debris collection device to achieve a high debris recovery rate;

[0012] Digital twin quality traceability module: Integrates a cutting process data acquisition module with a blockchain platform to record key data such as cutting parameters, vibration waveforms, and temperature curves, achieving tamper-proof storage and quality traceability of cutting data throughout its entire lifecycle.

[0013] The digital twin intelligent positioning module also includes a virtual debugging platform, which supports cutting path pre-simulation and process parameter optimization.

[0014] The adaptive hydraulic-mechanical hybrid clamping module also includes a discontinuous locking mechanism, which achieves dual axial / radial positioning through a wedge-shaped force-enhancing structure.

[0015] The dual-modal cutting execution module also includes a spiral cutting trajectory optimization algorithm, which generates the optimal path based on the three-dimensional model of the weld to reduce heat accumulation.

[0016] The multi-dimensional intelligent monitoring module also includes negative pressure adsorption and cyclone separation technology for the debris collection device.

[0017] The digital twin quality traceability module also includes a machine vision recognition module for identifying cut width, surface roughness, and weld reinforcement.

[0018] The system also includes an adaptive learning algorithm module, which optimizes the cutting path and process parameters based on historical operating data to improve cutting efficiency and quality consistency.

[0019] The method includes the following steps:

[0020] S1: Positioning and clamping stage: A three-dimensional model of the valve is constructed by laser scanning to locate the starting point of the weld. The hydraulic system applies clamping force and adjusts it dynamically.

[0021] S2: Cutting execution stage: Roughing uses pulsed plasma arc cutting, and finishing switches to ultrasonic vibration cutting, combined with laser ranging to achieve cutting depth control;

[0022] S3: Monitoring and Protection Phase: Real-time monitoring of vibration and temperature data, triggering a three-level safety response mechanism to achieve debris recycling;

[0023] S4: Quality Inspection and Traceability Stage: Machine vision is used to identify cutting parameters, and the blockchain platform stores the data to generate traceability reports.

[0024] The positioning and clamping stage also includes a wedge-shaped force-enhancing structure to achieve dual axial / radial positioning; the cutting execution stage also includes a spiral cutting trajectory optimization algorithm to reduce heat accumulation; the monitoring and protection stage also includes negative pressure adsorption and cyclone separation technology of the debris collection device; the quality inspection and traceability stage also includes an adaptive learning algorithm to optimize the cutting path and process parameters.

[0025] The beneficial effects of this invention are as follows:

[0026] By integrating digital twin positioning, adaptive clamping, dual-modal cutting, multi-dimensional monitoring, and blockchain traceability technologies, cutting accuracy is improved, cutting time is shortened, and efficiency is increased compared to traditional manual operation. A three-level safety response mechanism reduces the accident rate and increases the debris recovery rate, effectively preventing the risks of dust explosions and high-temperature burns. It is compatible with all-size valves, improving versatility and reducing the need for customized tooling. The blockchain platform enables tamper-proof storage of cutting data throughout its entire lifecycle. Combined with machine vision quality assessment, it generates traceable reports, meeting the stringent traceability requirements of nuclear power equipment maintenance and improving operational credibility. The adaptive learning algorithm dynamically optimizes the cutting path and process parameters based on historical data, continuously iterating to improve cutting efficiency and quality stability, forming a closed-loop ecosystem of "data-driven - intelligent optimization". Detailed Implementation

[0027] The preferred embodiments of the present invention will now be described in detail so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0028] The system includes:

[0029] Digital twin intelligent positioning module: It uses a dual-modal sensor of laser contour scanner and industrial camera to construct a three-dimensional point cloud model of valve, and achieves sub-millimeter positioning of weld seam through edge detection algorithm, and eliminates assembly error with feature point matching algorithm;

[0030] Adaptive hydraulic-mechanical hybrid clamping module: It adopts a three-jaw elastic expansion sleeve structure, realizes stepless adaptation of valve shaft diameter through hydraulic cylinder drive, and deploys a pressure sensor array in conjunction with PID control algorithm to realize adaptive adjustment of clamping force;

[0031] Dual-mode cutting execution module: includes coarse cutting mode and fine cutting mode. Coarse cutting adopts pulsed plasma arc cutting process, and fine cutting adopts ultrasonic vibration cutting, which is combined with a laser rangefinder to realize closed-loop control of cutting depth.

[0032] Multi-dimensional intelligent monitoring module: integrates acoustic emission sensor, vibration sensor, and temperature sensor to build a three-level safety response mechanism: Level I vibration warning, Level II temperature warning, and Level III emergency shutdown, and works with negative pressure adsorption + cyclone separation debris collection device to achieve a high debris recovery rate;

[0033] Digital twin quality traceability module: Integrates a cutting process data acquisition module with a blockchain platform to record key data such as cutting parameters, vibration waveforms, and temperature curves, achieving tamper-proof storage and quality traceability of cutting data throughout its entire lifecycle.

[0034] The digital twin intelligent positioning module also includes a virtual debugging platform, which supports cutting path pre-simulation and process parameter optimization.

[0035] The adaptive hydraulic-mechanical hybrid clamping module also includes a discontinuous locking mechanism that achieves dual axial / radial positioning through a wedge-shaped force-enhancing structure.

[0036] The dual-modal cutting execution module also includes a spiral cutting trajectory optimization algorithm, which generates the optimal path based on the three-dimensional model of the weld to reduce heat accumulation.

[0037] The multi-dimensional intelligent monitoring module also includes negative pressure adsorption and cyclone separation technology for the debris collection device.

[0038] The digital twin quality traceability module also includes a machine vision recognition module, used to identify cut width, surface roughness, and weld reinforcement.

[0039] The system also includes an adaptive learning algorithm module, which optimizes the cutting path and process parameters based on historical operating data to improve cutting efficiency and quality consistency.

[0040] The method includes the following steps:

[0041] S1: Positioning and clamping stage: A three-dimensional model of the valve is constructed by laser scanning to locate the starting point of the weld. The hydraulic system applies clamping force and adjusts it dynamically.

[0042] S2: Cutting execution stage: Roughing uses pulsed plasma arc cutting, and finishing switches to ultrasonic vibration cutting, combined with laser ranging to achieve cutting depth control;

[0043] S3: Monitoring and Protection Phase: Real-time monitoring of vibration and temperature data, triggering a three-level safety response mechanism to achieve debris recycling;

[0044] S4: Quality Inspection and Traceability Stage: Machine vision is used to identify cutting parameters, and the blockchain platform stores the data to generate traceability reports.

[0045] The positioning and clamping stage also includes a wedge-shaped force-enhancing structure to achieve dual axial / radial positioning; the cutting execution stage also includes a spiral cutting trajectory optimization algorithm to reduce heat accumulation; the monitoring and protection stage also includes negative pressure adsorption and cyclone separation technology of the debris collection device; and the quality inspection and traceability stage also includes an adaptive learning algorithm to optimize the cutting path and process parameters.

[0046] Taking the overhaul of the main feedwater regulating valve (DN200, made of 304L stainless steel) of a nuclear power plant as an example, the entire process is implemented as follows:

[0047] 1. Intelligent positioning and adaptive clamping stage

[0048] The digital twin positioning system uses an LJ-V7000 laser profile scanner to construct a three-dimensional point cloud model of the valve, and uses the Canny edge detection algorithm to locate the starting point of the weld (error ≤ 0.2mm), combined with a feature point matching algorithm to eliminate assembly errors.

[0049] The adaptive clamping system employs a three-jaw elastic expansion sleeve structure, with hydraulic cylinder drive enabling stepless adaptation of shaft diameters from Ф50mm to Ф600mm. A pressure sensor array (1N resolution) provides real-time feedback on the clamping force, which is dynamically adjusted to the range of 280-320N by a PID controller. A wedge block force amplification structure achieves dual axial / radial positioning.

[0050] The virtual debugging platform can simulate the cutting path and optimize process parameters (such as cutting speed and feed rate) to reduce on-site debugging time.

[0051] 2. Dual-modal cutting execution phase

[0052] Roughing mode: Employs 60A pulsed plasma arc cutting at a speed of 25mm / min. A temperature sensor monitors the tool temperature in real time (controlled between 120-150℃). The cutting path uses a spiral trajectory, generating the optimal path based on the weld seam 3D model to reduce heat accumulation, keeping the heat-affected zone within 45μm.

[0053] Fine-tuning mode: Switch to ultrasonic vibration cutting (amplitude 25μm, frequency 20kHz), combined with a laser rangefinder (accuracy ±0.05mm) to achieve closed-loop control of cutting depth. During the cutting process, the air-cooling system continuously cools the material to prevent annealing.

[0054] The adaptive learning algorithm module optimizes the cutting path based on historical running data, improving cutting efficiency and quality consistency.

[0055] 3. Multi-dimensional intelligent monitoring and security protection stage

[0056] Acoustic emission sensor (detection sensitivity ≤10⁻) 6 The feed rate is monitored in real time (m / s²) to detect microcracks, and a vibration sensor monitors the cutting vibration (automatically reducing the feed rate by 20% when the feed rate is ≥4g).

[0057] A temperature sensor monitors the tool temperature, activating the air-cooling system when it exceeds 180°C and triggering an emergency shutdown when it exceeds 220°C. The chip collection device achieves a chip recovery rate of over 95% through negative pressure adsorption (vacuum degree -80kPa) combined with cyclone separation technology, preventing the risk of dust explosion.

[0058] A three-level safety response mechanism: Level I vibration warning (≥3g), Level II temperature warning (≥180℃), and Level III emergency shutdown (≥5g or ≥220℃) to ensure the cutting process is safe and controllable.

[0059] 4. Digital Twin Quality Traceability and Intelligent Optimization Stage

[0060] The machine vision system (5μm resolution) identifies the kerf width (2.9mm), surface roughness (Ra2.8μm), and weld reinforcement height (0.4mm), meeting the requirement of ≤0.5mm weld reinforcement height for nuclear power equipment.

[0061] The cutting data is stored through the Hyperledger Fabric blockchain platform, generating an immutable quality traceability report that includes key data such as cutting parameters, vibration waveforms, and temperature curves.

[0062] The adaptive learning algorithm module optimizes the cutting path and process parameters based on historical operating data, improving cutting efficiency and quality consistency, and saving 1,200 hours of maintenance time per unit per year.

[0063] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. An automatic valve seal weld cut system, characterized by: The system comprises: Digital twin intelligent positioning module: a laser profile scanner and an industrial camera dual-mode sensor are used to construct a valve three-dimensional point cloud model, and an edge detection algorithm is used to realize sub-millimeter level positioning of the weld position, and a feature point matching algorithm is used to eliminate assembly errors; Self-adaptive hydraulic-mechanical hybrid clamping module: a three-jaw elastic expansion sleeve structure is used, and stepless adaptation of the valve shaft diameter is realized through the driving of a hydraulic cylinder, and a pressure sensor array is deployed to realize self-adaptive adjustment of the clamping force through a PID control algorithm; Dual-mode cutting execution module: including a rough cutting mode and a finishing mode, the rough cutting adopts a pulse plasma arc cutting process, and the finishing adopts ultrasonic vibration cutting, and a laser range finder is used to realize closed-loop control of the cutting depth; Multi-dimensional intelligent monitoring module: integrating acoustic emission sensors, vibration sensors, and temperature sensors, a three-level safety response mechanism is constructed: level I vibration warning, level II temperature warning, and level III emergency shutdown, and a negative pressure adsorption + cyclone separation debris collection device is used to realize high recovery rate of debris; Digital twin quality traceability module: integrating a cutting process data acquisition module and a blockchain platform, key data such as cutting parameters, vibration waveform, and temperature curve are recorded to realize tamper-proof storage and quality traceability of the cutting data throughout the life cycle.

2. The automatic valve seal weld cut system of claim 1, wherein: The digital twin intelligent positioning module further comprises a virtual debugging platform, which supports cutting path pre-performance and process parameter optimization.

3. The automatic valve seal weld cut system of claim 1, wherein: The self-adaptive hydraulic-mechanical hybrid clamping module further comprises a non-continuous locking mechanism, which realizes axial / radial double positioning through a wedge block reinforcement structure.

4. The automatic valve seal weld cut system of claim 1, wherein: The dual-mode cutting execution module further comprises a spiral cutting trajectory optimization algorithm, which generates an optimal path based on the weld three-dimensional model to reduce heat accumulation.

5. The automatic valve seal weld cut system of claim 1, wherein: The multi-dimensional intelligent monitoring module further comprises negative pressure adsorption and cyclone separation technology of the debris collection device.

6. The automatic valve seal weld cut system of claim 1, wherein: The digital twin quality traceability module further comprises a machine vision recognition module for identifying slit width, surface roughness, and weld reinforcement.

7. The automatic valve seal weld cut system of claim 1, wherein: The system further comprises an adaptive learning algorithm module, which optimizes the cutting path and process parameters based on historical operation data to improve cutting efficiency and quality consistency.

8. A method of automatic cutting control of valve seal welds, characterized by: The method comprises the following steps: S1: positioning and clamping stage: a three-dimensional model of the valve is constructed by laser scanning, the weld starting point is located, and the hydraulic system applies clamping force and dynamically adjusts; S2: cutting execution stage: rough cutting adopts pulse plasma arc cutting, and finishing switches to ultrasonic vibration cutting, and a laser range finder is used to realize cutting depth control; S3: monitoring and protection stage: real-time monitoring of vibration and temperature data triggers a three-level safety response mechanism to realize debris recovery; S4: quality detection and traceability stage: slit parameters are identified by machine vision, and a blockchain platform stores data to generate a traceability report.

9. The method of claim 8, wherein: The positioning and clamping stage further comprises a wedge block reinforcement structure to realize axial / radial double positioning; the cutting execution stage further comprises a spiral cutting trajectory optimization algorithm to reduce heat accumulation; the monitoring and protection stage further comprises negative pressure adsorption and cyclone separation technology of the debris collection device; and the quality detection and traceability stage further comprises an adaptive learning algorithm to optimize the cutting path and process parameters.