Helical springs have been widely used in various engineering applications for centuries. For many years, there is no significant development in the design methods of helical springs. Recently, a renewed interest is raised from the industry in exploring new designs for the helical springs with novel configurations due to the requirements of customised properties, such as specific spring stiffness and natural frequency for better performance of valve train systems. In this paper, an innovative method which combines the techniques of Finite Element Analysis (FEA), constrained Latin Hypercube sampling (cLHS) and Genetic Programming (GP) is developed to design and analyse helical springs with arbitrary shapes. cLHS method is applied to generate 300 sets of spring samples within a constrained design domain, and FE analysis is conducted on these spring samples. Two meta-models are developed from the 300 sets of FE results by using GP. They successfully describe the relationships between the design parameters and the overall mechanical performances including compression force and fundamental natural frequency of helical springs. The results show that the developed models have robust abilities on designing helical springs with arbitrary shapes, which significantly expands the design domain of the engineering design methods and potential for precise optimization of helical springs.

Helical springs with nonlinear geometric parameters nowadays have shown great advantages over classical linear springs, especially due to their superior performance in diminishing dynamic responses in high-speed situations. However, existing studies are mostly available for springs with linear properties, and the sole FE spring models using solid elements occupy significant computational resources. This study presents an FE spring model based on Timoshenko beam theory, which allows for high-speed dynamic simulations of nonlinear springs using a beehive valve spring sample. The dynamic results are also compared with the results of the FE model using solid elements and the results of the engine head test and indicate that the proposed FE model can accurately predict dynamic spring forces and the phenomenon of coil clash when simulating the beehive spring at engine speeds of both 5600 and 8000 RPM. The results also indicate that rapid coil impact brings significant spike forces. It should also be noted that the FE spring model using beam elements displays sufficient accuracy in predicting the dynamic responses of nonlinear springs while occupying much fewer computational resources than the FE model using solid elements.

Mechanical design is one of the essential disciplines in engineering applications, while inspirations of design ideas highly depend on the ability and prior knowledge of engineers or designers. With the rapid development of machine learning (ML) techniques, artificial intelligence (AI)‐based design methods are promising tools for the design of advanced engineering systems. So far, there have been some studies of 2D patterns and structural designs based on ML techniques. However, a particular challenge remains in allowing complex 3D mechanical designs using ML techniques. Herein, a novel and experience‐free method to equip ML models with 3D design capabilities by combining a convolutional neuron network (CNN) with a deep convolutional generative adversarial network (DCGAN) is developed. The model directly receives 2D image‐based training data that define the complex 3D structures of a specific machine part. After the training process, an infinite number of new 3D designs can be generated by the proposed model, with their geometric and mechanical properties being accurately predicted at the same time. Moreover, the generated new designs can be fed back to expand the original input datasets for further ML model training and updating.

The growing use of composites in automotive and aerospace fields highlights the need for effective joining of dissimilar materials. Adhesive bonding offers significant advantages over traditional methods. Therefore, comprehensively exploring the relationship between multiple design variables and joint strength, and subsequently achieving accurate prediction of joint strength based on this understanding, is essential for enhancing the effectiveness and efficiency of adhesive joint structural optimization. However, the joint—the critical yet weakest part—has strength governed by complex structural variables that are not fully understood, limiting optimization potential. Based on the effectiveness of finite element simulation in tensile fracture mechanics, this study developed a deep neural network (DNN). Combining the DNN model with a genetic algorithm (GA), both single-objective and multi-objective optimization were conducted. The single-objective optimization focused solely on maximizing joint strength, while the multi-objective GA further quantified the Pareto optimal trade-offs between joint strength and bond area, identifying compromise solutions. The effectiveness of the optimized parameters was validated, demonstrating higher efficiency and accuracy compared to traditional optimization methods such as response surface methodology (RSM). This integrated approach provides a robust framework for predicting joint strength and achieving effective optimization of bonded structures.