International Journal of Advanced Manufacturing Technology最新文献

The direction in which wire and powder feedstock are fed influences deposit quality of as-built parts produced using laser-direct energy deposition (L-DED). While lateral wire feed has been explored in existing L-DED investigations, limitations like process instability persist, especially in achieving the required connection between the wire and the melt pool. Co-axial feedstock deposition offers a potential solution, enabling higher manufacturing flexibility and efficiency by co-axially feeding wire or powder. However, the full potential of L-DED using co-axial feeding for metal components remains underexplored due to equipment limitations. This study systematically evaluates the printability of stainless steel (SS) 316L and compares the microstructures and microhardness properties between co-axial powder-fed and wire-fed L-DED specimens. Utilizing the MELTIO M450 L-DED system in an argon environment, single-layer three-track specimens are produced with different combinations of process parameters. Comprehensive characterization, employing optical and scanning electron microscopy alongside microhardness testing, reveals powder-fed specimens exhibit greater melt pool depth and cooling rates, while wire-fed counterparts display fewer oxide inclusions and smoother surfaces. Microstructural differences include higher δ-ferrite content in wire-fed specimens. Microhardness values between powder-fed and wire-fed specimens are comparable. These findings hold implications for sequential powder and wire deposition, enabling the production of diverse mechanical structures with distinct characteristics. Overall, this paper provides an insight into feedstock selection for efficient metallic part production via co-axial feedstock deposition and recommends a range of process parameters suitable for fabricating SS316L parts using co-axial deposition in L-DED.

Abstract

The thermal error suppression rate depends on the cooling effect of the water cooling system, and the cooling water flow rate is a direct factor affecting the cooling effect. To better reduce the thermal error, a numerical model of cooling water is established to solve for the optimal cooling water flow rate. Firstly, a numerical model of thermal deformation of the pendulum angle milling head is established based on thermoelasticity theory to determine the main heat sources leading to thermal deformation. Then, a numerical analysis model of the cooling water flow rate is established to investigate the cooling water flow rate that has the best effect on the suppression of thermal errors. Finally, five flow rates are used for cooling experiments to verify the accuracy of the numerical model. The results show that the temperature of each measurement point increases with the flow rate from a significant decrease to the basic constant trend of gradual saturation. The reduction rate of thermal error at v=54 cm/s is as high as 73.4%, providing a theoretical basis for enterprises to optimize water cooling system parameters.

The transition towards renewable electricity provides opportunities for manufacturing companies to save electricity costs through participating in demand response programs. End-to-end implementation of demand response systems focusing on manufacturing power consumers is still challenging due to multiple stakeholders and subsystems that generate a heterogeneous and large amount of data. This work develops an approach utilizing artificial intelligence for a demand response system that optimizes industrial consumers’ and prosumers’ production-related electricity costs according to time-variable electricity tariffs. It also proposes a semantic middleware architecture that utilizes an ontology as the semantic integration model for handling heterogeneous data models between the system’s modules. This paper reports on developing and evaluating multiple machine learning models for power generation forecasting and load prediction, and also mixed-integer linear programming as well as reinforcement learning for production optimization considering dynamic electricity pricing represented as Green Electricity Index (GEI). The experiments show that the hybrid auto-regressive long-short-term-memory model performs best for solar and convolutional neural networks for wind power generation forecasting. Random forest, k-nearest neighbors, ridge, and gradient-boosting regression models perform best in load prediction in the considered use cases. Furthermore, this research found that the reinforcement-learning-based approach can provide generic and scalable solutions for complex and dynamic production environments. Additionally, this paper presents the validation of the developed system in the German industrial environment, involving a utility company and two small to medium-sized manufacturing companies. It shows that the developed system benefits the manufacturing company that implements fine-grained process scheduling most due to its flexible rescheduling capacities.

Abstract

Closed-die forging preforms are usually made by open die forging operations, which are subject to significant variabilities. A sensitivity study covering a wide range of influencing parameters has highlighted the predominant influence of the initial billet geometry. The forging die strokes were also highly influential, while their fidelity is sufficient to use them as control parameters in order to compensate the geometrical dispersions of the billet. Consequently, their optimization was performed by taking a nominal preform geometry as the target. Polynomial surrogate models have been constructed to enable real-time optimization. A specific preform was used as a demonstrator in this study, while the approach was generic. The surrogate models were built using data from finite element simulations, which were first validated with an experimental campaign. On the one hand, this approach introduced agility by allowing changes in the billet geometry, and on the other hand, it allowed individual customization of the specific route to each billet.

The forming grinding process is a crucial method for helical flute grinding of the end mill and screw tap, and the grinding wheel profile is the key to ensure the precision of the helical flute, which is constrained by the grinding wheel posture. However, the existing methods have some deficiencies such as the invalidity or uneven wear problems. In this paper, an optimized calculation method of the grinding wheel profile for the forming grinding of the helical flute is proposed, including the suitable grinding posture correspondingly. Firstly, the parametric models of the helical cutting edge and the helical flute surface are established. Secondly, the contact condition between the grinding wheel and the flute is deduced to calculate the grinding wheel profile by the envelop theory. Thirdly, an optimization method for grinding wheel profile is proposed, which could predict of the solution interval, avoid the profile intersection, and improve wear resistance. Finally, the method was verified by a series of simulations and experiments, and the results show that the method could meet the grinding precision requirements and expand the application range of forming grinding technology in helical flute.

The inclusion of additive manufacturing (AM) as an automated repair method leads to a sustainable remanufacturing process, which is known as additive repair. Despite its potential in improving the efficiency of repair and restoration, additive repair remains in its infancy and requires a thorough investigation on part design and process parameters. The major concern raised in additive repair is the capability to create perfect bonding between two metals, which will affect the mechanical properties of the complete repaired part. Hence, performing evaluation from the beginning is crucial to validate the feasibility of the process through appropriate structural analysis and to obtain deformation and stress results. Brake caliper housing is selected as a remanufacturable component for case exemplary purposes. Prior to analysis, the potential damages and failures of the brake caliper component were initially evaluated through literature surveys and direct interviews with industry experts where two types of damages were identified, namely, cracks and broken or fractured parts. Then, the validation focuses on comparative analysis of two different conditions of the brake caliper housing: original, and repaired caliper model using finite element analysis in ANSYS. Results indicate that the strength of the repaired caliper model shows equal and higher strength compared with the original model. This result confirms that the repair process through AM can retain or improve the quality of the remanufactured brake caliper housing. Therefore, this paper provides a systematic framework for the evaluation of mechanical properties in multi-metal additive repair with the integration of failure analysis techniques.

Ball-end milling cutters are commonly used in the finishing processes of curved-side milling for titanium alloys; however, several issues arise during machining, such as poor cutting conditions at the bottom of the end teeth, low cutting speeds, and limited chip space. Given the above issues, the research on the design and manufacture of conical arc side-edge milling cutter for titanium alloy processing was carried out in this paper; the mathematical model of the vital structure of conical arc side-edge milling cutter was established; the grinding trajectory equations of tool front and flank were deduced; the tool-workpiece kinematics of ultrasonic vibration applied to conical arc side edge was studied; and the comparative experimental study of the conical arc side-edge milling cutter cutting titanium alloy with and without ultrasonic vibration was carried out. The experiment results indicate that in comparison to conventional milling techniques, ultrasonic vibration cutting significantly decreases cutting force, plastic deformation of the chip, and wear rate of the flank face. The tool wear band is both longer and more uniform, bonding phenomena in titanium alloys are distinctly reduced, and tool performance is improved.

Accurate tool wear prediction is of great significance to improve production efficiency, ensure product quality and reduce machining cost. This paper proposes a hybrid physics data-driven model-based fusion framework for tool wear prediction to improve low prediction accuracy of physical model and poor interpretation of data-driven model. In this framework, physical information and local features of sensor measurement signals are used as inputs to build a hybrid physics data-driven (HPDD) model. And data mining and physics principles are effectively integrated by using unlabeled samples for data expansion. Piecewise prediction is introduced to reduce difficulty in parameter estimation. Then, in order to manage prediction uncertainty of physical information and HPDD method, two prediction results are gradually combined based on Bayesian fusion mechanism to eliminate prediction error. Finally, the effectiveness of the proposed method is verified by experiment. Compared with existing methods, this method significantly improves prediction. The mean values of root mean square error (RMSE) and mean relative error (MARE) for tool wear prediction results are respectively 2.28 and 1.85.

Owing to low machining efficiency, poor machining accuracy, and surface quality in traditional electrical discharge machining (EDM) of TC4 titanium alloy holes, an EDM trepanning method was developed using a deionized water medium and a flushing liquid in tube electrodes (denoted by EDM-TFD). Several experiments and research mechanisms were conducted on EDM-TFD. Compared with traditional EDM, ED milling, and EDM trepanning, EDM-TFD featured increased hole machining efficiency by more than five times, improved machining taper by more than 57%, and improved surface quality by more than 40%. The optimal processing technology was determined through process experiments using copper tube electrodes, a flushing liquid with a speed of 2 m/s, pulse width (T_on) of 150 μs, pulse interval (T_off) of 150 μs, and peak current (I_p) of 15 A. Under the optimized process, the hole exhibited a feed rate of 1.1 mm/min, machining efficiency of 38 mm³/min, machining taper of 40 μm, and surface roughness of 5.2 μm.

Shaft sleeve assembly is a common task in industrial manufacturing. The fitting approach for shaft sleeve assembly is usually interference fit, which requires significant contact forces. Conventional assembly methods, though focused on safety, often struggle to achieve high efficiency. Reinforcement learning can effectively select appropriate assembly actions through interaction with the environment, making it well-suited for shaft sleeve assembly tasks. Firstly, a comprehensive workflow for shaft sleeve assembly is formulated, including system initialization, insertion, push, and completion. Our research focuses mainly on the insertion process. Secondly, the core control algorithm adopts a deep reinforcement learning method based on the Actor-Critic architecture. The reward function includes safety reward, step length reward, and step reward. Safety reward ensures assembly security, while step length and step reward enhance assembly efficiency from different perspectives. Finally, real-world experiments on shaft sleeve assembly are conducted, including ablation experiments, parameter tuning experiments on reward function, and comparative experiments with conventional methods. The results of the ablation experiments and parameter tuning experiments indicate that after combining safety reward, step length reward, and step reward, the assembly effect achieves the best, verifying the effectiveness of the proposed reward function. Comparative experimental results demonstrate that our approach not only enhances safety compared to conventional methods but also significantly improves assembly efficiency, indicating the feasibility of this method.