This research study describes a machine learning (ML)-driven model for producing smart structural materials via additive manufacturing (AM) by extrusion. A 3D concrete printing system was used to make cementitious composites that were reinforced with carbon nanotubes (CNTs) and graphene nanoplatelets (GNPs). Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN) models were used to undergo supervised learning on an experimental dataset consisting of 320 specimens to predict compressive strength, electrical conductivity, and print quality as dependent on process parameters and material composition. The highest R2 of compressive strength prediction of SVM was 0.946, whereas RF had the highest R2 of 0.987, which was used to predict electrical conductivity. Optimization of parameters guided by ML had a 61.8% enhancement of compressive strength and 30.5 times increase in electrical conductivity in comparison to non-optimized baselines. Nanomaterial networks were also found to be conductive, allowing individual networks to detect their strain levels through changes in current at a strain of 0.1%, which facilitates real-time structural health sensing. The artificial system showed a 31% decrease in CO2 emissions and a 58.8% decrease in material wastage compared with the usual way of building, proving to be a valid route towards intelligent and sustainable infrastructure.
This review methodically expounds on the genesis, distribution characteristics, and control methodologies of residual stress (RS) in additive/subtractive hybrid manufacturing (A/SHM). RS, originating from non-uniform temperature fields during manufacturing, rapid solidification of the molten pool, and complex thermal cycling, are key factors causing component deformation, performance degradation, and even cracking. It is evident that significant limitations are imposed on the industrial implementation of A/SHM technology in the domain of high-end equipment manufacturing. This review methodically unveils the influence patterns of process conditions, such as scanning strategies and laser parameters, on RS distribution. It elucidates the intrinsic relationship between microstructural evolution and RS and summarizes effective approaches to regulating RS through process optimization, post-heat treatment, and material modification. This paper proactively proposes a development direction for precise RS regulation through intelligent monitoring and control. This approach provides a theoretical foundation and technical support to enhance the reliability of A/SHM components and advance their industrial applications.
A metal forming technique called equal channel angular pressing is used to produce alloys and metals with ultrafine grain and nanocrystalline structure. Using this method, grain refining to the nano or submicron-scale is possible in materials with high strain super plasticity without affecting the size of the workpiece. One of the greatest techniques for creating bulk materials with ultra-fine grains is equal channel angular pressing. During this procedure, metal is continuously pushed through a channel die that has been particularly made with intersecting channels at different angles. The material is pass through a die in this procedure that has two channels that meet at a particular angle. Finer grains are formed as a result of the material’s deformation when it passes through the die. The creation of ultra-fine grains is influenced by a number of die design characteristics. The effects of processing route, corner angle, channel angle, and number of passes in die design on grain refinement. After comparing the results of several parameters, it was found that (90°) is the ideal channel angle for producing the maximum shear strain, and this strain reduces as the channel angle increases. The die was designed and produced in the lab with ideal design specifications, including a corner angle of (20°) and a channel angle of (90°). The mechanical characteristics of AA5083 were examined both before and after the Equal Channel Angular Pressing method. This study examines and analyses the mechanical behaviour of AA5083 that is treated through the use of an ECAP die that has ideal design specifications. Pressing was done between 0 and 2 times when using the (BC) path. According to the results, the grain size decreased from 480 nm to 170 nm, and the tensile strength increased from 225.8 MPa to 358.4 MPa after two ECAP runs.
