A tool warehouse-in and warehouse-out management and damage defect detection system
Patent Information
- Application Number
- CN202510520809.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-16
- Publication Date
- 2026-10-23
AI Technical Summary
这种方式不仅效率低下，容易出现人为记录错误，而且难以实时准确地掌握刀具库存数量和位置，在生产急需刀具时，常常需要花费大量时间寻找，导致生产停滞；在刀具损伤缺陷检测方面，以往多采用人工目检或简单的量具测量
本发明采用二维码和RFID识别技术，出入库识别模块可快速读取刀具信息，结合实时数据更新的出入库管理软件，实现了刀具出入库的高效自动化管理。工作人员无需手动记录繁琐信息，查询刀具时通过多条件查询功能即可迅速获取所需信息，大大缩短了出入库操作时间和查找刀具的时间，提高了生产效率；通过损伤缺陷检测模块，利用高分辨率工业相机，以及基于深度学习算法的损伤缺陷分析软件，能够高精度检测刀具的磨损、崩刃、裂纹等损伤缺陷，并根据刀具累计的加工量和所加工工件的材料等信息，与系统设定的加工量标准值和所加工工件的材料所需刀具精度进行比对判别，如加工量达到相应标准值，系统可降低该刀具加工标准或报废，由精加工降级为粗加工，机械臂和可旋转吸盘将该刀具转入粗加工库存货架，报废的刀具转入报废货架，有效避免因刀具问题导致的产品质量缺陷，保障了产品加工精度和质量稳定性；精准的库存管理避免了不必要的刀具采购成本。此外，系统的自动化和智能化减少了人工操作，降低了人工成本；本发明与企业现有的企业资源规划(ERP)系统，生产执行(MES)系统等进行集成，实现数据共享和业务流程协同。
Smart Images
Figure CN122966145A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of tool management technology, specifically to a tool entry and exit management and damage / defect detection system. Background Technology
[0002] In the machining industry, cutting tools are critical processing tools, and their management and condition monitoring have a significant impact on production efficiency, product quality, and production costs. Traditional cutting tool inventory management relies primarily on manual records, where staff manually register tool numbers, entry / exit times, and the work group using the tools. This method is not only inefficient and prone to human error, but also struggles to accurately track tool inventory quantities and locations in real time. When tools are urgently needed, significant time is often spent searching, leading to production stoppages. Regarding tool damage and defect detection, the past primarily relied on visual inspection or simple measuring tools. Visual inspection is limited by human visual acuity and subjective judgment, making it difficult to detect minor wear, cracks, and other damage, easily resulting in missed detections. Simple measuring tools can only detect some dimensional parameters of the tool, failing to comprehensively detect damage and defects, and cannot meet the stringent requirements of high-precision machining for tool condition. With the continuous improvement of machining precision and automation, traditional tool management and inspection methods can no longer meet the needs of modern production, necessitating an efficient and accurate tool inventory management and damage / defect detection system. Summary of the Invention
[0003] The purpose of this invention is to provide a tool entry and exit management and damage and defect detection system, which realizes intelligent and automated management of tool entry and exit, as well as high-precision detection of tool damage and defects, thereby improving production efficiency, ensuring product quality, and reducing production costs.
[0004] This invention is achieved by providing the following technical solution:
[0005] A tool entry / exit management and damage / defect detection system, characterized by comprising a hardware system and a software system: The hardware system includes an entry / exit identification module, a damage / defect detection module, and an inventory management shelf. The entry / exit identification module employs QR code identification and RFID radio frequency identification technologies, affixing a unique QR code label or RFID chip to each tool. QR code scanners or RFID readers are installed on both sides of the entry / exit channel to quickly read data such as tool number, model, specifications, entry time, and cumulative processing volume, and transmit this data to the tool entry / exit management system of the Enterprise Resource Planning (ERP) system. The damage / defect detection module includes a robotic arm, a rotatable suction cup, and a high-resolution industrial camera. When a tool enters the warehouse, the robotic arm and rotatable suction cup place the tool within the camera's field of view. The camera takes high-definition photos of the tool from multiple angles, acquiring surface image information, which is then transmitted to damage / defect analysis software. The software automatically identifies whether the tool is clean; unclean tools trigger an alarm and are returned to the worker for cleaning before being put back into the warehouse. The software system includes interfaces for Enterprise Resource Planning (ERP) systems. The RP (Resource Processing) system includes inbound / outbound management software and damage / defect analysis software. The inbound / outbound management software features real-time data updates and supports multi-condition queries based on tool number, inbound / outbound time range, tool type, tool machining quantity, and workpiece material. It also includes an inventory warning function, automatically sending a warning when the tool inventory falls below a set safety threshold. The damage / defect analysis software uses deep learning algorithms to analyze tool images captured by industrial cameras, identifying wear, chipping, cracks, and other damage defects. It uses a data analysis model to determine the degree of tool damage and remaining lifespan. Based on the tool's cumulative machining quantity and the workpiece material, it compares this information with the system's set machining quantity standard and the required tool precision for the workpiece material. If the machining quantity reaches the corresponding standard value, the system can reduce the tool's machining standard or scrap it, downgrading it from finishing to roughing. A robotic arm and a rotatable suction cup transfer the tool to a roughing inventory shelf, while scrapped tools are transferred to a scrap shelf. The inventory management shelf is divided into finishing tool area, roughing tool area, and scrap tool area, with multiple tool storage positions in each area.
[0006] Preferably, the inbound / outbound identification module of the present invention uses a high-speed QR code scanner with a scanning rate of over 50 scans per second, capable of recognizing QR codes as small as 1mm × 1mm, meeting the rapid reading requirements of QR code labels for knives of different specifications. The RFID reader operates at a frequency of 13.56MHz, with an effective identification distance adjustable within the range of 0.1-1 meter, supporting the simultaneous reading of information from multiple RFID chips, ensuring that data such as the knife's serial number, model, specifications, and entry time are accurately and in real-time transmitted to the Enterprise Resource Planning (ERP) management system when the knife passes through the inbound / outbound channel.
[0007] Preferably, the industrial camera has a resolution of 5 megapixels and a lens aperture of f / 1.8.
[0008] Preferably, the damage and defect analysis software of the present invention uses a deep learning algorithm based on a convolutional neural network (CNN) architecture, employing ResNet50 as the base model, and optimized for the characteristics of tool damage and defect detection. During model training, a large amount of tool image data containing different types of damage and defects is used, and data augmentation techniques such as rotation, scaling, and cropping are applied to improve the model's generalization ability. The data analysis model combines support vector machines (SVM) and decision tree algorithms, comparing the cumulative machining amount of the tool with the system's standard machining amount. If the machining amount reaches the standard value, the system can reduce the machining standard of the tool or scrap it.
[0009] Preferably, when a tool is taken out of the warehouse, the system determines the tool to be taken out based on the tool inventory, preset production priority rules, and the machining accuracy required by the tool for the workpiece material being processed, and generates a take-out instruction, which is then identified and the take-out operation is completed by the robotic arm and the rotatable suction cup.
[0010] Compared with the prior art, the present invention has the following advantages: This invention employs QR code and RFID identification technologies. The inbound / outbound identification module can quickly read tool information. Combined with real-time data-updated inbound / outbound management software, it achieves highly efficient and automated management of tool inbound / outbound operations. Workers no longer need to manually record tedious information; when searching for tools, they can quickly obtain the required information through multi-condition queries, significantly reducing inbound / outbound operation time and tool retrieval time, thus improving production efficiency. Through the damage and defect detection module, utilizing a high-resolution industrial camera and damage and defect analysis software based on deep learning algorithms, it can accurately detect tool wear, chipping, cracks, and other damage and defects. Based on the tool's cumulative machining volume and the material of the workpiece being processed, it compares this information with the system's set machining volume standard value and the tool precision required for the workpiece material. If the machining volume reaches the corresponding standard value, the system can reduce the tool's machining standard or scrap it, downgrading it from finishing to roughing. A robotic arm and a rotating suction cup transfer the tool to the roughing inventory shelf, while scrapped tools are transferred to the scrap shelf. This effectively avoids product quality defects caused by tool problems, ensuring product machining accuracy and quality stability. Precise inventory management avoids unnecessary tool procurement costs. Furthermore, the system's automation and intelligence reduce manual operations and lower labor costs; this invention integrates with existing enterprise resource planning (ERP) systems, manufacturing execution (MES) systems, etc., to achieve data sharing and business process collaboration. Attached Figure Description
[0011] Figure 1This is a hardware architecture diagram of the present invention, showing the connection relationship and layout of hardware devices such as the inbound identification module, robotic arm, rotatable suction cup, and inventory management shelf. Detailed Implementation
[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0013] Hardware installation and debugging of this invention: According to design requirements, QR code scanners and RFID readers are installed on both sides of the inbound / outbound channel, ensuring the scanning range covers the entire channel. Multiple tests are conducted by adjusting the installation angle and position to ensure accurate reading of tool information. When installing the inventory management shelves, ensure the shelves are level and stable. At the inspection station, install robotic arms, rotatable suction cups, high-resolution industrial cameras, and other equipment. Based on the inspection station diagram, precisely adjust the equipment position and angle. Calibration and debugging are performed using standard tools and simulated cutting devices to ensure clear and reliable image information acquired by the equipment.
[0014] The software deployment and configuration of this invention involves deploying the inventory management software and damage / defect analysis software of the Enterprise Resource Planning (ERP) management system onto a server. The server is configured with at least an 8-core CPU, 16GB of RAM, and a 500GB hard drive. System initialization settings are performed, including setting user permissions and assigning corresponding operating permissions based on the responsibilities of different personnel. For example, administrators have the highest permissions and can perform system settings, data management, and other operations; ordinary staff only have permissions for tool inventory management and querying. Inventory warning thresholds are set, and safety thresholds and warning levels are reasonably set based on the company's production needs and tool procurement cycles. Damage / defect detection parameters are set, such as the shooting parameters of industrial cameras and the thresholds of deep learning models.
[0015] This invention is based on the following working principle: Tool warehousing: Workers place the tools to be stored in the inbound / outbound channel. The inbound / outbound identification module reads the tool information at a high-speed scanning rate. Using a robotic arm and a rotating cleaning tray, the tool is placed within the photographing range of an industrial camera. The industrial camera takes high-definition photos of the tool from multiple angles, acquiring surface image information. This image information is transmitted to the damage and defect analysis software, which automatically identifies whether the tool is clean. Unclean tools trigger an alarm and are returned to the worker for cleaning before being stored. The damage and defect analysis software analyzes the tool images captured by the industrial camera based on deep learning algorithms, identifying damage defects such as wear, chipping, and cracks. It uses a data analysis model to determine the degree of damage and remaining life of the tool. Based on the tool's cumulative machining volume and the material of the workpiece being processed, it compares the results with the system's set machining volume standard value and the tool precision required for the workpiece material. If the machining volume reaches the corresponding standard value, the system can reduce the processing standard of the tool or scrap it, downgrading it from finishing to roughing. The robotic arm and rotating suction cup transfer the tool to the roughing inventory shelf, while scrapped tools are transferred to the scrap shelf. Tool issuance: Staff input issuance requests into the management system. The system determines the tools to be issued based on the tool inventory, preset production priority rules, and the material requirements of the workpiece to be processed in the production task, and generates an issuance instruction. The robotic arm and a rotatable suction cup identify and complete the issuance operation. At the same time, the management system updates the inventory data and records information such as the issuance time and the work group using the tool. Damage and Defect Detection: Periodically or after a certain number of uses, the tool is placed in the inspection station. The industrial camera of the damage and defect detection module takes high-definition photos of the tool from multiple angles according to preset shooting angles and parameters, and transmits the tool image information to the damage and defect analysis software. The damage and defect analysis software first preprocesses the images captured by the industrial camera, including image enhancement and noise reduction, and then inputs them into a CNN-based deep learning model for damage and defect identification. The data analysis model further determines the degree of tool damage and remaining life, and compares it with the system's set machining volume standard value and the tool precision required by the material of the workpiece being machined. If the machining volume reaches the corresponding standard value, the system can reduce the machining standard of the tool or scrap it, downgrading it from finishing to roughing. The robotic arm and a rotatable suction cup transfer the tool to the roughing inventory shelf, and the scrapped tool is transferred to the scrap shelf.
[0016] The above provides a detailed description of a tool entry and exit management and damage defect detection system provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, based on the ideas of the present invention, several modifications and improvements can be made to the specific implementation methods and application scope, and these all fall within the protection scope of the present invention.
Claims
1. A tool entry / exit management and damage / defect detection system, characterized in that, The system includes hardware and software systems: the hardware system includes an inbound / outbound identification module, a damage / defect detection module, and an inventory management shelf. The inbound / outbound identification module uses QR code identification technology and RFID radio frequency identification technology. A unique QR code label or RFID chip is affixed to each tool. QR code scanners or RFID readers are installed on both sides of the inbound / outbound channel to quickly read data such as tool number, model, specifications, inbound time, and cumulative processing volume, and transmit them to the tool inbound / outbound management system of the Enterprise Resource Planning (ERP) system. The damage and defect detection module includes a robotic arm, a rotatable suction cup, and a high-resolution industrial camera. When tools are put into storage, the robotic arm and rotatable suction cup place the tool within the camera's field of view. The camera takes high-definition photos of the tool from multiple angles, acquiring surface image information. This image information is then transmitted to the damage and defect analysis software, which automatically identifies whether the tool is clean. Unclean tools trigger an alarm and are returned to the worker for cleaning before being put back into storage. The software system includes inventory management software that interfaces with an Enterprise Resource Planning (ERP) system and damage and defect analysis software. The inventory management software features real-time data updates and supports multi-condition queries based on tool number, inventory time range, tool type, machining quantity, and workpiece material. It also includes an inventory warning function. When the quantity falls below a set safety threshold, an automatic warning message is sent. The damage and defect analysis software analyzes and processes tool images captured by industrial cameras based on deep learning algorithms to identify tool wear, chipping, cracks, and other damage defects. It uses a data analysis model to determine the degree of tool damage and remaining lifespan. Based on the tool's cumulative machining volume and the material of the workpiece being machined, it compares the result with the system's set machining volume standard value and the tool precision required for the workpiece material. If the machining volume reaches the corresponding standard value, the system can reduce the tool's machining standard or scrap it, downgrading it from finishing to roughing. The robotic arm and rotatable suction cup transfer the tool to the roughing inventory shelf, while scrapped tools are transferred to the scrap shelf. The inventory management shelf has a finishing tool area, a roughing tool area, and a scrapped tool area, with multiple tool storage positions in each area.
2. The tool entry / exit management and damage / defect detection system according to claim 1, characterized in that, The QR code scanner has a scanning rate of no less than 50 times per second and can recognize QR codes with a minimum size of 1mm×1mm. The RFID reader operates at a frequency of 13.56MHz and has an effective recognition distance that is adjustable within the range of 0.1-1 meters. It supports the simultaneous reading of information from multiple RFID chips.
3. The tool entry / exit management and damage / defect detection system according to claim 1, characterized in that, The industrial camera has a resolution of 5 megapixels and a lens aperture of f / 1.
8.
4. The tool entry / exit management and damage / defect detection system according to claim 1, characterized in that, The deep learning algorithm of the damage and defect analysis software is based on the convolutional neural network (CNN) architecture, uses ResNet50 as the basic model, and combines support vector machine (SVM) and decision tree algorithms for data analysis.
5. The tool entry / exit management and damage / defect detection system according to claim 1, characterized in that, When a tool is taken out of the warehouse, the system determines the tool to be taken out based on the tool inventory, preset production priority rules, and the machining accuracy required for the workpiece material. It then generates a take-out instruction, which is recognized by the robotic arm and a rotatable suction cup to complete the take-out operation.
6. The tool entry / exit management and damage / defect detection system according to claim 1, characterized in that, During damage and defect detection, the damage and defect analysis software first preprocesses the images captured by the industrial camera, including image enhancement and noise reduction, and then inputs them into a deep learning model for damage and defect identification.