Key Takeaways & Executive Findings
- •• A novel deep learning retrieval method integrates process design features (modulus, wall thickness, symmetry, length-to-height ratio) with CLIP-based shape features, significantly improving process reuse. • Incorporating process design features boosts retrieval accuracy by 6.09% to 97.82% and increases process similarity by 30.25% on a database of 230 production casting models. • The retrieved process, validated via EasyCast simulation, yields minimal shrinkage in the target model, demonstrating superior process reuse capability. • The method requires no large training dataset, making it highly applicable to casting process design and related manufacturing.
Abstract
Accurate retrieval of casting 3D models is crucial for process reuse. Current methods primarily focus on shape similarity, neglecting process design features, which compromises reusability. In this study, a novel deep learning retrieval method for process reuse was proposed, which integrates process design features into the retrieval of casting 3D models. This method leverages the comparative language-image pretraining (CLIP) model to extract shape features from the three views and sectional views of the casting model and combines them with process design features such as modulus, main wall thickness, symmetry, and length-to-height ratio to enhance process reusability. A database of 230 production casting models was established for model validation. Results indicate that incorporating process design features improves model accuracy by 6.09%, reaching 97.82%, and increases process similarity by 30.25%. The reusability of the process was further verified using the casting simulation software EasyCast. The results show that the process retrieved after integrating process design features produces the least shrinkage in the target model, demonstrating this method’s superior ability for process reuse. This approach does not require a large dataset for training and optimization, making it highly applicable to casting process design and related manufacturing processes.
1. Introduction
Casting is crucial in industries such as aviation, aerospace, transportation, and shipbuilding, supporting the innovation of numerous products and high-end equipment. Typical castings, such as satellite brackets, missile shells, engine casings, and aerospace engine blades, feature complex structures like irregular surfaces, internal channels, and structural holes, making process design highly challenging and time-consuming.
In industrial manufacturing, over 75% of product designs are based on case studies [1]. Reusing and refining mature processes from existing products can significantly shorten production cycles. However, current process reuse relies on manual retrieval of casting process cards, which is inefficient and dependent on experience [2]. With the advancement of Industry 4.0 and digital intelligent manufacturing [3, 4], 3D model-based process design is widely adopted in casting enterprises [5]. Thus, accurate retrieval of 3D models is essential for effective process reuse.
Loading authentic research manuscript (Pages 1–5)...
Xiao-long Pei, Hua Hou, Li-wen Chen, Zhi-qiang Duan, Yu-hong Zhao (2025). Deep learning retrieval of 3D casting models combined with professional knowledge for process reuse. China Foundry. https://doi.org/10.1007/s41230-025-4145-7
Research & Educational Purpose Only:The translations, structured abstracts, analytical annotations, and data reports provided by SinoTechIntel are intended exclusively for academic research, internal corporate R&D, and educational benchmarking. They do not constitute formal engineering, chemical safety, legal, or professional advice.
Copyright & Intellectual Property Notice: Original copyright of the underlying source articles and experimental data remains with the respective authors, institutions, and original publishing journals. SinoTechIntel claims intellectual property only over its proprietary translations, analytical syntheses, and AEO structured enhancements in accordance with international fair use and academic citation principles.
Frequently Asked Questions
What is the main contribution of this paper?
The paper proposes a novel deep learning retrieval method that integrates process design features with shape features extracted by CLIP, significantly improving the accuracy and reusability of casting 3D model retrieval for process reuse.
How does the method improve retrieval accuracy?
By incorporating process design features such as modulus, main wall thickness, symmetry, and length-to-height ratio, the method achieves a 6.09% improvement in accuracy (reaching 97.82%) and a 30.25% increase in process similarity compared to shape-only methods.
What dataset was used for validation?
A database of 230 production casting models was established to validate the proposed method.
How was process reusability verified?
The reusability was verified using the casting simulation software EasyCast, showing that the process retrieved with integrated process design features produces the least shrinkage in the target model.
Does the method require a large training dataset?
No, the method does not require a large dataset for training and optimization, making it highly applicable to casting process design and related manufacturing processes.
Related Technical Papers & Translations
Pull-out capacity and energy absorption of cable bolts under impact loading
This study investigates the performance of high-strength cable bolts under impact loading conditions representative of rock bursts in underground environments. Although widely used, the dynamic behaviour of these cable bolts has received limited experimental attention, and their effectiveness in seismically active zones remains a subject of ongoing debate. To address this gap, a reverse pull-out test machine integrated with a drop hammer rig was employed. Tests were conducted on 70-t SUMO bulbed and non-bulbed cable bolts with encapsulation lengths of 300 and 450 mm, subjected to an impact energy of 14.52 kJ. Results indicate that non-bulbed cables, despite showing lower initial peak loads (average 218 vs. 328 kN for bulbed cables at 300 mm encapsulation), demonstrated superior energy absorption (average 11.26 vs. 8.75 kJ) and displacement capacity (average 48.40 vs. 36.25 mm). Increasing the encapsulation length for bulbed cables led to a reduction in initial peak load but improved displacement and energy absorption. The dominant failure mechanism was debonding at the cable-grout interface, characterised by frictional sliding and cable rotation. These findings provide new insights into the energy dissipation mechanisms of cables and support the development of more resilient ground support systems for dynamically active conditions.
Effect of eutectic content on microstructure and mechanical properties of Al-Zn-Mg-Cu alloys
The 7xxx series aluminum alloys have emerged as a particularly promising class of lightweight structural materials. However, the inherent strength of these materials is primarily influenced by the content and type of alloying elements added during the manufacturing process, as well as casting defects. The present study investigated the effects of eutectics formed by solute atoms (Zn, Mg, and Cu), with equal mass ratios (Zn/Mg=2, Mg/Cu=3) but varying overall contents, on the liquid film thickness, crack propagation depth, and the mechanical properties of the Al-Zn-Mg-Cu alloy after heat treatment. The results from gravity casting indicate that the intergranular liquid film thickness increases with the increase of eutectic content. A thick intergranular liquid film in the casting can accommodate greater strain during grain contraction, thereby preventing liquid film rupture and subsequent hot tearing. Concurrently, during the solution treatment at 475 °C, the residual eutectic fraction in the Al-7Zn-3.5Mg-1.18Cu alloy diminishes from 9.1% at 10 h to 0.35% at 40 h. At 165 °C, the Al-6Zn-3.0Mg-1.0Cu alloy exhibits the optimal mechanical properties, with a peak aging tensile strength of 510 MPa and an elongation of 6.4%. The incorporation of lower concentrations of solute atoms (Zn, Mg, and Cu) serves to reduce the barrier to dislocation precipitation, thereby enhancing alloy plasticity. However, when the proportion of alloying elements exceeds the solubility limit of the α-Al matrix at specific heat treatment temperatures, coarse residual phases remain intergranular, thereby significantly impairing the mechanical properties of the alloy. This study provides a reference for the optimal addition level of the main strengthening elements in Al‑Zn‑Mg‑Cu alloys.
Achieving optimal strength-conductivity balance in cast Al-2.3Fe-Mg-Si alloys via Mg/Si ratio regulation
The Al-2.3Fe eutectic alloy is regarded as a promising substitute for Cu conductors in automotive motors owing to its excellent castability and low resistivity. However, its application is restricted by the mutually exclusive relationship between electrical conductivity and mechanical strength. The microstructure and mechanical properties of Al-2.3Fe alloy were modified through Mg/Si alloying combined with T6 heat treatment in this work, leading to the development of a high-performance cast Al-2.3Fe-Mg-Si alloy. In the Al-2.3Fe-0.40Mg-0.72Si (Mg/Si=0.56) alloy subjected to T6 treatment, an electrical conductivity of (52.5±0.6)% IACS is achieved, while the ultimate tensile strength is significantly enhanced to 309.5±5.6 MPa. The addition of Mg and Si brings about marked changes in the solidification process of the Al-2.3Fe alloy, resulting in considerable variations in both the morphology of the second phase and its phase constitution. The aging behavior of the alloy is governed by second phase and solid solubility. Through optimization of the Mg/Si ratio, the aging response can be effectively enhanced. At the ratio of Mg/Si=0.56, a balance is achieved between solid solubility and precipitation, while simultaneously minimizing the detrimental impact on electrical conductivity and reaching the best mechanical properties and electrical conductivity in peak-aged Al-2.3Fe-xMg-ySi alloy. This work providing valuable insights for developing advanced conductor materials.