Turbine Blade Straightening With 3D Feedback and Force Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional manufacturing processes for turbine or compressor blades often fail to achieve the required twist or sag, necessitating a straightening process that is not automated, leading to inconsistent results.
Innovation Solution
A method involving a self-learning algorithm that uses three-dimensional data acquisition and force application to deform blades into compliance with nominal dimensions, with the algorithm training on multiple parts to optimize the straightening process, incorporating gripping zones and various force types like bending and torque.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional manufacturing processes (machining, forging, or extrusion) are used to produce blades, then production efficiency is maintained, but the required twist or deflection cannot be obtained, resulting in poor manufacturing precision
Solution Approach 1:
The patent applies preliminary action by intentionally introducing controlled defects (twist or deflection) during manufacturing, then correcting them through a subsequent straightening process. This approach allows efficient mass production while achieving precise final dimensions, as the straightening process compensates for the intentionally introduced imperfections.
Solution Approach 2:
The patent utilizes parameter changes by applying controlled mechanical forces (bending moments, torques) to modify the blade's geometric parameters (twist angle, deflection) during the straightening process. The self-learning algorithm dynamically adjusts these force parameters to achieve the target geometry while accounting for material non-linearities.
2Manufacturing precision
If a straightening process is applied to correct blade shape, then manufacturing precision is improved, but automation is lacking leading to inconsistent results and high device complexity
Solution Approach 1:
The patent implements feedback through a self-learning algorithm that uses measured data from actual blades to update and refine the straightening model. The system measures blade geometry, compares it to target specifications, determines required corrections, applies forces, and uses the results to improve future corrections, creating a closed-loop automated system that reduces inconsistency.
Solution Approach 2:
The patent applies self-service by enabling the system to automatically learn and improve its straightening capabilities without external intervention. The self-learning algorithm processes measurement data and updates the straightening model autonomously, allowing the system to serve itself by continuously improving its own performance through accumulated experience.
3Device complexity
If manual straightening processes are used, then device complexity is reduced, but measurement precision and result consistency deteriorate
Solution Approach 1:
The patent replaces manual mechanical operations with automated measurement and control systems. Optical or laser measurement systems substitute for manual measurement, and computer-controlled force application substitutes for manual straightening operations, thereby improving measurement precision while maintaining relatively simple physical equipment.
4Manufacturing precision
If iterative force application is used to achieve compliance, then manufacturing precision is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating the straightening forces using a self-learning algorithm based on measured blade geometry and a library of previous cases. This allows the system to determine optimal force parameters before actual straightening, reducing the need for extensive iterative adjustments and thereby decreasing processing time while maintaining precision.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method automates the blade straightening process, improving consistency and accuracy by iteratively adjusting forces based on data comparisons, leading to blades that meet precise aerodynamic profiles.
Implementation Method 1
determination by a self-learning algorithm trained on a plurality of parts of the same type of a force to be applied to the part to deform said part with sufficient probability
Implementation Method 2
application of the force determined in the previous step to the part so as to obtain the part in a deformed conformation
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to a method for shaping a turbine blade-type equipment part (12), comprising the following steps: - supplying a part (12) comprising a blade (14) in an initial conformation; - supplying a nominal definition representing the part in a nominal conformation; - comparing the initial conformation to the nominal definition in order to determine a conforming or non-conforming characteristic; - for a non-conforming data, determining a force to be applied to the part to deform said part; - applying the force so as to obtain the part in a deformed conformation; - comparing the deformed conformation to the nominal definition in order to determine a conforming or non-conforming characteristic; - training a self-learning algorithm (82).