Additive Build Planning for Predictable Metal Weld Bead Properties
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Additive manufacturing of metal materials faces challenges in predicting the properties of built objects due to complex processes and high degrees of freedom in manufacturing conditions, making it difficult to efficiently predict and achieve desired properties.
Innovation Solution
A building plan assistance method and device that generate mathematical models to relate input information (material, welding conditions, and welding tracks) to output properties, creating a database for searching and presenting optimal conditions for target property values, with multiple subitems and property values, and optionally using temperature history as intermediate output information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If additive manufacturing is used to manufacture built objects with desired shape, then manufacturing flexibility and design freedom are improved, but predicting material properties becomes difficult and requires enormous arithmetic processes
Solution Approach 1:
The patent segments the complex additive manufacturing process into discrete welding conditions (welding method, heat input, travel speed, etc.) and evaluates each segment's impact on material properties separately. This allows systematic analysis of how individual process parameters affect outcomes without requiring enormous arithmetic processes for the entire complex process.
Solution Approach 2:
The patent performs preliminary experimentation to establish the relationship between welding conditions and material properties before actual manufacturing. By pre-determining these relationships through controlled experiments, the system avoids complex real-time calculations during production while maintaining accurate property prediction capability.
2Manufacturing precision
If empirical knowledge and trial and error are used to adjust manufacturing conditions, then desired shape and properties can be obtained, but the process requires enormous time and effort
Solution Approach 1:
The patent implements a feedback mechanism where material properties are measured after manufacturing, and this information is fed back to adjust welding conditions for subsequent production. This systematic feedback loop replaces random trial and error with directed optimization, significantly reducing adjustment time while maintaining high property accuracy.
Solution Approach 2:
The patent systematically varies welding parameters (heat input, travel speed, welding method) to establish their relationships with material properties. By pre-determining these parameter relationships, the system can quickly adjust manufacturing conditions without time-consuming trial and error during actual production.
3Extent of automation
If machine learning is used to determine weldment specifications, then property determination becomes automated, but the process requires extensive data preparation and processing
Solution Approach 1:
The patent performs preliminary data collection and processing to establish relationships between welding conditions and material properties before automation is implemented. This pre-processing creates a knowledge base that enables subsequent automated property determination without requiring complex real-time data processing.
Solution Approach 2:
The patent introduces an intermediate database or knowledge base that stores the relationships between welding conditions and material properties. This intermediary structure simplifies the automation process by providing pre-processed information that machine learning algorithms can efficiently query, reducing the complexity of direct data processing.
4Adaptability or versatility
If the degree of freedom in manufacturing conditions is increased, then various combinations of built object properties can be achieved, but property prediction requires enormous arithmetic processes
Solution Approach 1:
The patent segments the numerous manufacturing conditions into distinct categories (welding method, heat input, travel speed, etc.) and evaluates their individual and combined effects systematically. This segmentation allows the system to handle high degrees of freedom without requiring enormous arithmetic processes by analyzing parameters in an organized manner rather than as a monolithic complex system.
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
Enables efficient prediction of built object properties with reduced effort, assisting in creating more appropriate building plans and improving the accuracy of property predictions.
Implementation Method 1
the weld beads formed by melting and solidifying a filler metal fed from a welding head
Implementation Method 2
by use of a heat source such as a laser or an arc, and depositing the weld metal (weld beads)
Data Source
AI summary
A building plan assistance method by which a mathematical model is used to associate input information, including the respective items of the material of a build object, a weld condition of weld beads, and a weld track, with output information including a characteristic value of the build object when additively manufactured under the condition of the input information, and a database is created using the mathematical model. The database is searched for a build object material, a weld condition, and a weld track corresponding to a target characteristic value of the build object to be manufactured, and the obtained build object material, weld condition, and weld track are presented. In the creating of the database, each of input sub-items of the input information items is associated with an individual characteristic value of the output information by means of the mathematical model.