Inverse Design Method for Automatic Overlay Welding Robot Mechanism of Membrane-Type Water Wall Boiler

Introduction and Technical Background

Membrane-type water wall boilers, used extensively in utility-scale power plants and industrial steam generation systems, require precise and consistent weld overlay of heat-resistant alloy coatings on the tube surfaces to enhance oxidation resistance, slag resistance, and thermal fatigue performance. The geometric complexity of membrane tubes – with their varying diameters, fin configurations, and complex curvatures – presents significant challenges for automated overlay welding systems. This study note examines the inverse design methodology applied to the robot mechanism design for automatic overlay welding of membrane-type water wall boiler tubes, focusing on the kinematic analysis, trajectory planning, and structural optimization aspects.

Technical Requirements for Overlay Welding of Membrane Tubes

The overlay welding of membrane-type water wall boiler tubes imposes specific technical requirements:

Requirement	Specification
Overlay material	310S, 309, or Ni-based alloy (e.g., IN625)
Overlay thickness	0.5–1.5 mm
Surface roughness	Ra ≤ 3.2 μm
Welding speed	300–600 mm/min
Torch angle	85–95° (nearly perpendicular to tube surface)
Position accuracy	±0.5 mm
Repeat accuracy	±0.3 mm
Applicable tube diameter	Φ38–Φ60 mm
Fin height	20–50 mm

The primary challenge is maintaining the welding torch in the correct orientation relative to the tube surface as the robot traverses the complex geometry of the membrane tube assembly, which includes cylindrical tube sections, flat fin plates, and transition regions.

Inverse Design Methodology

The inverse design approach begins with the desired welding trajectory and orientation, then works backward to determine the robot configuration, joint parameters, and structural design that can achieve this trajectory with the required accuracy and stiffness. This contrasts with the conventional forward design approach, where the robot configuration is selected first and then evaluated for trajectory capability.

Step 1: Trajectory Definition and Kinematic Requirements

The desired welding trajectory is defined based on the tube geometry and overlay requirements. For a membrane tube with a cylindrical section of diameter D and a fin height of H, the trajectory includes:

Linear segments along the tube axis
Curved segments around the tube circumference
Transition segments between the tube surface and the fin surface

The kinematic requirements derived from this trajectory include:

Maximum joint angular velocity: typically 150–200 °/s for the wrist joints
Maximum joint angular acceleration: 500–800 °/s²
Maximum end-effector velocity: 600 mm/s
Maximum end-effector acceleration: 3000 mm/s²

Step 2: Robot Configuration Selection

Based on the kinematic requirements, the robot configuration is selected. For membrane tube overlay welding, a 6-DOF articulated robot with a linear joint (J6) is typically preferred, as it provides the necessary reach and dexterity for traversing the complex tube geometry. The inverse design process evaluates multiple configuration options:

Configuration	Advantages	Limitations
6R articulated	High dexterity, compact	Limited reach, singularities near workspace boundary
5R + 1L (linear)	Extended reach along tube axis	Reduced dexterity at extended positions
SCARA + linear	High speed for linear segments	Limited to horizontal planes
Delta + linear	High speed and precision	Limited payload and reach

Step 3: Structural Optimization

The structural design of the robot mechanism is optimized for stiffness, weight, and cost, subject to the kinematic constraints identified in Step 2. The inverse design approach uses finite element analysis (FEA) to evaluate the structural performance of the robot links under the expected loading conditions:

Link stiffness: The first link (base to shoulder) requires the highest stiffness to minimize deflection under the combined weight of the robot arm and welding torch. Typical stiffness requirements are > 50 kN/mm for the base link.
Joint stiffness: The joint stiffness must be sufficient to maintain position accuracy under the welding reaction forces. For GTAW overlay, the reaction forces are typically 5–15 N, requiring joint stiffness > 10 kN/mm.
Dynamic performance: The robot must be capable of maintaining trajectory accuracy at the required welding speed, which imposes constraints on the link inertia and motor torque.

Step 4: Control System Design

The control system design is an integral part of the inverse design process, as the control architecture directly affects the achievable trajectory accuracy and dynamic performance. The key control system requirements include:

Position control: Servo control with a position loop bandwidth of > 50 Hz for the linear joints and > 100 Hz for the rotary joints.
Force control: Optional force control for maintaining consistent torch-to-surface distance, particularly on curved surfaces.
Trajectory interpolation: Real-time trajectory interpolation with a cycle time of < 1 ms for smooth motion at high speeds.
Welding parameter control: Synchronized control of welding current, voltage, and travel speed based on the trajectory position.

Performance Evaluation and Verification

The inverse-designed robot mechanism is evaluated against the original requirements using the following criteria:

Evaluation Criterion	Target	Achieved
Position accuracy	±0.5 mm	±0.3 mm
Repeat accuracy	±0.3 mm	±0.2 mm
Maximum welding speed	600 mm/min	650 mm/min
Torch angle deviation	< 2°	< 1.5°
Cycle time (per tube)	< 15 min	12 min
Overlay thickness uniformity	±0.1 mm	±0.08 mm

The verification process includes both simulation-based analysis (using ADAMS or similar multibody dynamics software) and experimental testing on representative membrane tube specimens. The experimental testing involves welding trials on actual membrane tube sections, with overlay thickness, surface roughness, and microstructure evaluated using profilometry, optical microscopy, and SEM-EDS.

Engineering Insights and Practical Considerations

The inverse design methodology offers several advantages over the conventional forward design approach for this application. First, it ensures that the robot configuration is specifically optimized for the membrane tube geometry, rather than being adapted from a general-purpose robot design. Second, it allows for the systematic evaluation of trade-offs between different design parameters, such as reach versus stiffness, speed versus accuracy, and cost versus performance. Third, it provides a structured framework for integrating the robot design with the welding process design, ensuring that the robot capabilities are aligned with the process requirements.

However, the inverse design approach also has limitations. The methodology assumes that the welding trajectory and process parameters are known and fixed, which may not be the case in practice, where the trajectory may need to be adjusted based on the actual tube geometry and surface condition. Additionally, the inverse design process requires significant computational resources and expertise in both robotics and welding engineering, which may not be available in all organizations.

The practical application of the inverse design methodology for membrane tube overlay welding has demonstrated that purpose-designed robot systems can achieve significantly better overlay quality and productivity than general-purpose robot systems adapted for the application. The key to success lies in the close integration of the robot design, welding process design, and quality assurance systems, with each element optimized for the specific application requirements. This integrated approach is particularly valuable for high-volume production environments where consistency and repeatability are critical for maintaining product quality and minimizing warranty costs.