Plastic Parts Research Articles

Three-Dimensional Printing of Metallic Parts by Means of Fused Filament Fabrication (FFF)

Obtaining metallic parts via Additive Manufacturing can yield several advantages over using other traditional manufacturing methods such as machining. Material extrusion (MEX) can handle complex shapes with porous structures and, at the present time, much low-end and desktop equipment is available. In the present work, different industrial and medical applications of metallic Fused Filament Fabrication (FFF) parts are presented. First, an overview of the process, equipment, and of the metal-filled filaments currently available is provided. Then, the properties of parts and different applications are shown. For example, metal-filled filaments with a low metal content that can be used to obtain plastic parts with metallic appearance (with either steel, copper, or bronze), and filaments with a high metallic content allow obtaining metallic parts with high mechanical strength after a sintering operation. The present contribution aims to be an up-to-date panorama for current industrial and medical results and lessons learnt from the application of FFF to obtain metallic parts.

Use of pneumatic waste in thermal insulation systems in arid áreas

Recent decades have seen a major global concern to protect the environment from non-biodegradable waste as little as possible. On the other hand, the problem of energy saving has emerged. Builders in arid areas and particularly in the Algerian Sahara are always faced with a very difficult choice: “build with traditional and local materials promoting a certain thermal comfort or build with modern materials offering more convenience, aesthetics and space but with less or no thermal comfort and very high consumption of air conditioning energy. The appearance of new materials and the innovation of waste in construction have not only improved the mechanical properties of materials but also energy saving through insulation, waterproofing and even the appearance of the construction. In the present work, an investigation was undertaken in order to respond to the problem of thermal comfort of construction while getting rid of tires which, through their long life, occupy large environmental spaces and through their incineration contribute to environmental pollution. Numerous tests have been carried out on reduced or real models representing either the floors or the partition walls which represent the two elements which are the main players in the heat transfer of the building. In comparison with the usual materials used in thermal insulation, the innovation of pneumatic waste has shown technical performances and thermal insulation properties superior to those existing on the market, which has the effect of increasing the energy efficiency of the buildings thus generating great economic interest. It is necessary to point out that only the plastic part of the tires was used while the metal part constituting the structural hoops of the tires has a wide use as a reinforcement system.

Small Punch Test to Estimate the Threshold Stress in Aggressive Environments by Incremental Step Loading

The present work is a relevant advance in the validation of the incremental step loading technique (ASTM F1624 standard) when applied to Small Punch tests (SPT) for the threshold load determination of medium- and high-strength steels in aggressive environments, as a novel alternative to conventional time-consuming tests under constant load. It completes previous works by the authors on this topic, extending a methodology to estimate the threshold stress from SPT tests in aggressive environments, covering the whole range of hardness marked by ASTM F1624 as the main goal. This is achieved by calibrating a model of the material’s hardness by the use of a coefficient in function of it. For this purpose, four medium- and high-strength steels of 33, 35, 50 and 60 HRC (Hardness Rockwell C) are exposed to three different cathodic polarization hydrogen embrittlement environments of 1, 5 and 10 mA/cm2 in 1N H2SO4 acid electrolyte connected to a platinum anode. Threshold stresses in these circumstances are obtained by uniaxial specimens following ASTM F1624 and compared to their homologous threshold loads obtained by Small Punch tests according to the authors’ original methodology proposal. Finally, the aforementioned model, consisting of a correlation based on composing an elastic and a plastic part, is calibrated for a hardness ranging 33–60 HRC, this being the main original contribution of this work; the elastic part is dependent just on the elastic-to-plastic transition SPT load, while the plastic part is ruled by a material hardness-dependent coefficient. This technique supposes an advance in engineering tools, due to its applicability in situations of material shortage, such as in-service components, welded joints, local areas, complex geometries, small thicknesses, etc., often present in aerospace, automotive or oil–gas, among others.

Mechanical Strength of Additive Manufactured and Standard Polymeric Components Joined Through Structural Adhesives.

The main aim of this work is to evaluate the mechanical properties of additive manufactured polymeric parts joined with standard plastic parts through structural adhesives. The primary advantage of this technique is its ability to significantly increase the size of the final assembly by using additive manufacturing (AM) for complex joints and inexpensive, reliable extruded plastic parts for load-bearing components. This hybrid assembly combines the flexibility and shape adaptability of AM with the structural strength and cost-effectiveness of extruded polymer parts, resulting in a final design that performs comparably to the base material. The materials used in the paper are rigid acrylic adhesive and toughened acrylic, both applicable with almost no surface preparation and fast curing. The 3D-printed parts are produced in ABS, while the standard parts are in PVC. First, the work is devoted to estimating the performance of the adhesives using pin-collar joints and a combined numerical and experimental methodology. The second section presents and discusses the results of two more realistic applications of adhesive bonding to hybrid complex joints. For the pin-collar joints, the results show failure mostly in the adhesive, with an average shear stress of 11.5 MPa and 5.22 MPa and a stiffness of 4449 N/mm and 3649 N/mm for the rigid and toughened adhesives, respectively. The results of the adhesive bonding of structural joints show that the adhesive is always capable of providing the load-carrying capacity required to achieve the strength of traditionally manufactured polymeric parts. The paper shows that adhesives are a feasible way to expand the potential of 3D-printed equipment to obtain larger hybrid parts partially realized with traditional technology, especially with inexpensive off-the-shelf bars and sections.

Facilitating the Production of 3D-Printed Spare Parts in the Design of Plastic Parts: A Design Requirement Review

Using additive manufacturing for spare part production can ensure that spare parts are available for a long time. However, spare parts are currently not designed for additive manufacturing. This study aimed to find how the production of 3D-printed spare parts can be facilitated in the design of plastic parts. We used a literature review and illustrative case to find how the design requirements for standard injection moulded plastic parts relate to the manufacturing capabilities of additive manufacturing for spare parts. The design requirements were defined by assigning corresponding structural and material properties. These requirements were then used to construct and evaluate the capabilities of additive manufacturing compared to injection moulding. It was found that additive manufacturing is especially suitable for requirements like Accuracy, Heat resistance, and Chemical resistance. However, to fully enable 3D-printed spare parts, certain design challenges still need to be tackled. Designers should pay careful attention to the synergies and trade-offs between design requirements and the challenges that might arise from the combination of certain requirements. Also, designers should ensure products are easily reparable before considering 3D-printed spare parts. If we target these challenges in the design phase, we can facilitate 3D-printed spare parts that enable product repairability.

Time reduction in changing of plastic injection mold part bottom cover injection machine size 380 ton

This study focuses on enhancing efficiency by reducing the time required to adjust plastic injection molds on a 380-ton machine. The setup time for mold change was 61 minutes, exceeding the company's target of 30 minutes. The research employed a quasi-experimental approach, capturing pre-improvement data, measuring the time for each mold change, and documenting sub-work steps using a Flow Process Chart and a Man-Machine Chart. The main activities identified were mold change and Jig robot adjustment using the Quick changeover technique, involving internal and external activities. The study revealed that some mold-changing steps could be prepared in advance, and applying SMED (Single Minute Exchange of Die) and ECRS (Eliminate, Combine, Rearrange, Simplify) principles led to the identification of seven external adjustment steps. Assigning another employee to assist in these steps significantly reduced the average mold adjustment time to 28 minutes, representing a 54.09% improvement.

Demystifying Velopharyngeal Dysfunction for Plastic Surgery Trainees Part 3: Objective Assessment and Surgical Decision-Making.

Velopharyngeal dysfunction (VPD) is the inability to achieve proper closure of the velopharyngeal (VP) port, affecting speech and swallowing. After an auditory-perceptual speech evaluation by a speech-language pathologist, objective assessment of the VP port is required to determine the need for surgical intervention. This 3-part series provides a comprehensive discussion on (1) the anatomy and physiology of the velopharyngeal mechanism; (2) fundamental speech terminology and principles of perceptual speech assessment for VPD; and (3) techniques for objective evaluation of the VP port and surgical decision-making process. In part 3, the authors focus on the modalities for objective VP port assessment, including both direct and indirect methods. Direct imaging techniques such as videofluoroscopy, nasoendoscopy, and MRI are detailed for their strengths and limitations in visualizing VP port function and preoperative planning. Indirect assessments, including nasometry and aerodynamic measurements, are also briefly discussed. The decision-making process for surgical intervention is explored, emphasizing factors such as the severity and etiology of VPD, VP closure patterns, palatal length, orientation of the levator veli palatini, and other patient-specific considerations. The authors review the surgical options for repair including palatoplasty procedures (Furlow palatoplasty, straight-line intravelar veloplasty, and palatal lengthening buccal myomucosal flaps) and pharyngoplasty procedures (posterior pharyngeal flaps and sphincter pharyngoplasty), highlighting their indications, techniques, and potential complications. This series serves as an accessible resource, providing the foundational knowledge required for surgical trainees new to this topic.

Demystifying Velopharyngeal Dysfunction for Plastic Surgery Trainees Part 2: Speech Fundamentals and Perceptual Speech Assessment.

Velopharyngeal dysfunction (VPD) is the inability to achieve proper closure of the velopharyngeal (VP) port, affecting speech and swallowing. The gold standard for diagnosis is auditory-perceptual speech evaluation by a specialized speech-language pathologist. This 3-part series provides a comprehensive discussion on (1) the anatomy and physiology of the velopharyngeal mechanism, (2) fundamental speech terminology and principles of perceptual speech assessment for VPD, and (3) techniques for objective evaluation of the VP port and surgical decision-making process. In part 2, the authros begin with the concepts and terminology required to understand the perceptual speech assessment of VPD, including resonance, articulation, and the differentiation between consonants and vowels. We review the types of speech samples used for evaluation such as single-word articulation tests, syllable repetition, sentence repetition, and spontaneous connected speech. Finally, we discuss the auditory-perceptual speech assessment for VPD, including the assessment of resonance, nasal air emission, articulation, and voice quality. The use of rating scales like the Pittsburgh Weighted Speech Scale (PWSS) and the Cleft Audit Protocol for Speech-Augmented-Americleft Modification (CAPS-A-AM) is highlighted. In addition, the significance of intraoral examinations, visual-tactile-auditory evaluations, and the assessment of voice quality are covered. Most textbook chapters discussing this topic assume a foundational knowledge of speech-language pathology, which a surgical trainee may lack. This study aimed to bridge the gap between surgical training and speech pathology, providing a comprehensive resource to enhance the understanding and management of VPD.

Thermal and Catalytic Recycling of Plastics from Waste Electrical and Electronic Equipment-Challenges and Perspectives.

The amount of end-of-life electrical and electronic devices has been widely increased, globally. This emphasizes how recycling waste electric and electronic equipment (WEEE) is essential in order to reduce the amount of WEEE that is disposed of directly in the environment. Plastics account for a big percentage in WEEE, almost 20%. As a result, the application of recycling methods on plastics gathered from WEEE is of great importance since, in this way, landfill disposal can be reduced. Nevertheless, despite the advantages, there are a lot of difficulties, such as the variety of different plastics present in the plastic mix and the existence of various additives in the plastic parts, for instance, brominated flame retardants that need special attention during their treatments, which restricts their wide application. Considering all these, this review aims to provide readers with all the current techniques and perspectives that are available for both the thermal and the catalytic recycling of plastics retrieved from WEEE. Apart from the up-to-date information on the recycling methods, in this review, emphasis is also given on the advantages each method offers and also on the difficulties and the limitations that may prevent them from being applied on a large scale. Current challenges are critically examined, including the use of mechanical or thermo-chemical recycling, the treatment of individual polymers or polymer blends and the separation of harmful additives before recycling or not. Finally, emerging technologies are briefly discussed.

A multistrategy differential evolution algorithm combined with Latin hypercube sampling applied to a brain–computer interface to improve the effect of node displacement

Injection molding is a common plastic processing technique that allows melted plastic to be injected into a mold through pressure to form differently shaped plastic parts. In injection molding, in-mold electronics (IME) can include various circuit components, such as sensors, amplifiers, and filters. These components can be injected into the mold to form a whole within the melted plastic and can therefore be very easily integrated into the molded part. The brain–computer interface (BCI) is a direct connection pathway between a human or animal brain and an external device. Through BCIs, individuals can use their own brain signals to control these components, enabling more natural and intuitive interactions. In addition, brain–computer interfaces can also be used to assist in medical treatments, such as controlling prosthetic limbs or helping paralyzed patients regain mobility. Brain–computer interfaces can be realized in two ways: invasively and noninvasively, and in this paper, we adopt a noninvasive approach. First, a helmet model is designed according to head shape, and second, a printed circuit film is made to receive EEG signals and an IME injection mold for the helmet plastic parts. In the electronic film, conductive ink is printed to connect each component. However, improper parameterization during the injection molding process can lead to node displacements and residual stress changes in the molded part, which can damage the circuits in the electronic film and affect its performance. Therefore, in this paper, the use of the BCI molding process to ensure that the node displacement reaches the optimal value is studied. Second, the multistrategy differential evolutionary algorithm is used to optimize the injection molding parameters in the process of brain–computer interface formation. The relationship between the injection molding parameters and the actual target value is investigated through Latin hypercubic sampling, and the optimized parameters are compared with the target parameters to obtain the optimal parameter combination. Under the optimal parameters, the node displacement can be optimized from 0.585 to 0.027 mm, and the optimization rate can reach 95.38%. Ultimately, by detecting whether the voltage difference between the output inputs is within the permissible range, the reliability of the brain–computer interface after node displacement optimization can be evaluated.

Enhancing of Surface Quality of FDM Moulded Materials through Hybrid Techniques.

With the rapid advancement of 3D-printing technology, additive manufacturing using FDM extrusion has emerged as a prominent method in manufacturing. However, it encounters certain limitations, notably in surface quality and dimensional accuracy. Addressing issues related to stability and surface roughness necessitates the integration of 3D-printing technology with traditional machining, a strategy known as the hybrid technique. This paper presents a study of the surface geometric parameters and microstructure of plastic parts produced by FDM. Sleeve-shaped samples were 3D-printed from polyethylene terephthalate glycol material using variable layer heights of 0.1 mm and 0.2 mm and then subjected to the turning process with PVD-coated DCMT11T304 turning inserts using variable cutting parameters. The cutting depth was constant at 0.82 mm. Surface roughness values were correlated with the cutting tool feed rate and the printing layer height applied. The selected specimen's microstructure was studied with a Zeiss EVO MA 15 scanning electron microscope. The roundness was measured with a Keyence VR-6200 3D optical profilometer. The research results confirmed that the additional application of turning, combined with a reduction in the feed rate (0.0506 mm/rev) and the height of the printed layer (0.1 mm), reduced the surface roughness of the sleeve (Ra = 1.94 μm) and increased its geometric accuracy.

Accurate 3D reconstruction of highly reflective mechanical part surfaces using a structured light method with combined phase shift and gray code

The structured light measurement method is applied in quality control across various sectors including manufacturing, medicine, fashion, and technology. This approach facilitates the generation of rapid and consistent evaluation reports by analyzing errors, geometric dimensions, and deviations in measured parts, aligning them with design specifications. Such capabilities significantly enhance the speed and quality of production processes, particularly in the mechanical industry. This paper proposes an effective alternative solution for the inaccurate measurement of shiny mechanical surfaces. It explores ambient light to achieve optimal calibration and enhance measurement accuracy. The appropriate exposure time for surface measurement was determined by analyzing the image intensity histogram. This step helps avoid saturation in the detection device and ensures the generation of high-quality point clouds by merging them at suitable exposure times. To validate the effectiveness of the proposed system, two comparisons were conducted: one between the 3D profile of the part measured by commercial software versus the proposed method, and another between the 3D profile of the plastic part as measured by the proposed method versus its designed profile. The results demonstrate the efficiency of the proposed method in accurately testing full-field reflective surfaces, without requiring additional equipment or hardware.

Electrochemical deposition of coloured Ni/Cu2O bilayer films on large area plastic substrates for decorative applications

ABSTRACT In this study, coloured Ni/Cu2O bilayer films were electrochemically deposited on large area automobile plastic parts (7 × 13 cm) for decorative applications. Such coatings may become an alternative to ubiquitously used Cr coatings for plastics. Purple, dark cyan and gold colours with good homogeneity and high brightness could be produced by tuning the deposition time of the Cu2O layer. Colouration of the bilayer film can be ascribed to interference of light through the metal/dielectric layer, where the thickness of the dielectric layer controls the wavelength of the reflected light. Cu2O film has a compact structure and low roughness, which is essential to obtain uniform colours. Moreover, the Ni/Cu2O bilayer exhibits good adhesion behaviour, environmental stability and enhanced corrosion resistance. This process is industrially viable and enables the large scale production of such decorative coatings.

Predicting 3D printed plastic part properties: A deep learning approach with thermographic and vibration data fusion

Additive Manufacturing (AM) holds transformative potential for the manufacturing industry, yet its widespread adoption is hindered by inconsistent product properties. This study addresses this challenge by pioneering a novel predictive method to assess the impact of in-process sensing on part property prediction in Fused Deposition Modeling (FDM), serving as a representative case study. Focused on five critical mechanical properties, including surface roughness, tensile strength, elongation at break, micro-hardness, and warpage, we systematically explore various sensor combinations by integrating Inertial Measurement Unit (IMU) sensors and a thermal camera with machine setting parameters. Utilizing deep learning models, specifically hybrid Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) architectures, we encode time-series sensor data into signal images and evaluate performance across different sensor configurations. Our best-performing model achieves an impressive 99% correlation in predicting tensile strength, albeit with less robust performance in predicting warpage and micro-hardness, suggesting additional influencing factors. Furthermore, the employed signal imaging technique outperforms a handcrafted feature selection alternative. These findings carry significant implications for advancing the broader adoption of AM in critical applications, addressing challenges associated with inconsistent product properties and unlocking its full potential.

In‐mold rheology and automated process control for injection molding of recycled polypropylene

AbstractManufacturing plastic parts with secondary feedstocks has risen to the forefront of importance in recent years. However, the variation in molecular weight and rheology of secondary feedstock can lead to inconsistent part quality. This work evaluates the effectiveness of a novel closed‐loop adaptive process control system that adjusts nozzle pressure in response to in‐mold pressure data. Five different recycled polypropylene blends, with a broad distribution of flow properties, were evaluated to determine the effectiveness of the control system at reducing processing variation. The experimental results show that the process control strategy reduced the variation within the mold, as seen by in‐mold pressure curves and calculated in‐mold viscosity values. Additionally, the parameters that control the automated process adjustments were investigated, showing the importance of optimization. The analysis of the correlation between in‐mold rheology and mechanical properties showed a slight variation in the mechanical properties and parts weight with a coefficient of variation of under 5%. Overall, the results demonstrate the ability of pressure‐controlled molding and automated viscosity adjustment to reduce the variability when molding a secondary feedstock.Highlights Pressure‐controlled injection molding of recycled polypropylene. Automated closed‐loop adaptive process control methodology. Methodology resulted in a reduction in pressure variation during molding. Changes in mechanical properties and in‐mold viscosity were investigated. Results show the potential of pressure‐controlled molding at reducing variation.

Investigation of Recycled Expanded Polyamide Beads through Artificial Ageing and Mechanical Recycling as a Proof of Concept for Circular Economy.

Car manufacturers are currently challenged with increasing the sustainability of their products and production to comply with sustainability requirements and legislation. One way to enhance product sustainability is by reducing the carbon footprint of fossil-based plastic parts. Particle foams are a promising solution to achieve the goal of using lightweight parts with minimal material input. Ongoing developments involve the use of expanded particle foam beads made from engineering plastics such as polyamide (EPA). To achieve this, a simulated life cycle was carried out on virgin EPA, including mechanical recycling. The virgin material was processed into specimens using a steam-free method. One series was artificially aged to replicate automotive life cycle stresses, while the other series was not. The mechanical recycling and re-foaming of the minipellets were then carried out, resulting in an EPA particle foam with 100% recycled content. Finally, the thermal and chemical material properties were comparatively analysed. The study shows that the recycled EPA beads underwent polymer degradation during the simulated life cycle, as evidenced by their material properties. For instance, the recycled beads showed a more heterogeneous molecular weight distribution (an increase in PDI from two to three), contained carbonyl groups, and exhibited an increase in the degree of crystallization from approximately 24% to 36%.

Optimization Design of Injection Mold Conformal Cooling Channel for Improving Cooling Rate

Optimizing the design of the conformal cooling channel can increase the cooling rate of injection mold. The aim of this study was the problem of low cooling efficiency of injection mold for deep-cavity plastic parts under the conventional cooling channel. Based on the analysis of the heat transfer principle of the injection mold, a mathematical description of the cooling time of the conformal cooling channel was made. By designing orthogonal experiments and using simulation methods in the Autodesk Moldflow 2019 software, with the minimum cooling time of the mold as the optimization goal, experimental optimization was carried out for the three design variables of the channel, thereby obtaining the optimal combination of design variables for the conformal cooling channel. The conformal cooling channel layout was innovatively designed, and through computer simulation experiments, it was concluded that the conformal cooling channel adopted a series flat-head layout, which has the shortest cooling time and the fastest cooling rate. Metal additive manufacturing technology was used to complete the manufacturing of the mold insert with the conformal cooling channel. After the trial production of the conformal cooling injection mold, the molding cycle was obviously shortened, and the injection molding production efficiency was significantly improved.

Design of hybrid deep learning using TSA with ANN for cost evaluation in the plastic injection industry

In the plastic injection industry, plastic injection molding is one of the most extensively used mass production technologies and has been continuously increasing in recent years. Cost evaluation is essential in corporate operations to increase the market share and lead in plastic part pricing. The complexity of the plastic parts and manufacturing data resulted in a long data waiting time and inaccurate cost evaluation. Therefore, the aim of this research is to apply a cost evaluation approach that combines hybrid deep learning of a tunicate swarm algorithm (TSA) with an artificial neural network (ANN) for the cost evaluation of complicated surface products in the plastic injection industry to achieve a faster convergence rate for optimal solutions and higher accuracy. The methodology entails the ANN, which applies feature-based extraction of 3D-model complicated surface products to develop a cost evaluation model. The TSA is used to construct the initial weight into the learning model of the ANN, which can generate faster-to-convergent optimal solutions and higher accuracy. The result shows that the new hybrid deep learning TSA combined with the ANN provides more accurate cost evaluation than the ANN. The prediction accuracy of cost evaluation is approximately 96.66% for part cost and 93.75% for mold cost. The contribution of this research is the development of a new hybrid deep learning model combining the TSA with the ANN that includes the calculation of the number of hidden layers specifically for complicated surface products, which are unavailable in the literature. The cost evaluation approach can be practically applied and is accurate for complicated surface products in the plastic injection industry.

Influence of Polymer Flow on Polypropylene Morphology, Micro-Mechanical, and Tribological Properties of Injected Part

This research investigates the micro-mechanical and tribological properties of injection-molded parts made from polypropylene. The tribological properties of polymers are a very interesting area of research. Understanding tribological processes is very crucial. Considering that the mechanical and tribological properties of injected parts are not uniform at various points of the part, this research was conducted to explain the non-homogeneity of properties along the flow path. Non-homogeneity can be influenced by numerous factors, including distance from the gate, mold and melt temperature, injection pressure, crystalline structure, cooling rate, the surface of the mold, and others. The key factor from the micro-mechanical and tribological properties point of view is the polymer morphology (degree of crystallinity and size of the skin and core layers). The morphology is influenced by polymer flow and the injection molding process conditions. Gained results indicate that the indentation method was sufficiently sensitive to capture the changes in polypropylene morphology, which is a key parameter for the resulting micro-mechanical and tribological properties of the part. It was proven that the mechanical and tribological properties are not equal in varying regions of the part. Due to cooling and process parameters, the difference in the indentation modulus in individual measurement points was up to 55%, and the tribological properties, in particular the friction coefficient, showed a difference of up to 20%. The aforementioned results indicate the impact this finding signifies for injection molding technology in technical practice. Tribological properties are a key property of the part surface and, together with micro-mechanical properties, characterize the resistance of the surface to mechanical failure of the plastic part when used in engineering applications. A suitable choice of gate location, finishing method of the cavity surface, and process parameters can ensure the improvement of mechanical and tribological properties in stressed regions of the part. This will increase the stiffness and wear resistance of the surface.

Real-time detection of plastic part surface defects using deep learning- based object detection model

In this study, it was aimed to detect defects in plastic parts produced in a company operating in the automotive sub-industry using the YOLOv8 object detection model. The defect types seen in plastic parts were evaluated with the help of Pareto analysis, and scratches, stains and shine were selected as the most common defect types, and data on the three defect types were collected. YOLOv8 models were trained using faulty part images. As a result of the training, the highest mean average precision value of 0.990 was obtained in the YOLOv8s model, and the shortest training time was obtained in the YOLOv8n model. In the YOLOv8s model, which gave the highest mAP value, hyperparameter adjustment was made according to the batch size and learning rate values. The testing phase was carried out with the hyperparameter values that gave the best results and the mAP value was obtained as 0.902.