引言
本文在企业数字化、智能化转型升级的探索和实践，发展智能化、柔性化、规模化的产业体系特别是智能化无人生产体系的建设中，对涉及的关键要素进行总结分析，对相关企业数字化转型建设具有一定的借鉴参考作用。
一、建设无人化生产体系原因
某企业的产品要求高精密、高质量、高可靠，生产模式是典型的多品种、小批量离散生产模式。经过多年来数字化转型建设实践，已经具备了较强的自动化加工与信息化管理能力。从哑设备改造、产线的少人化探索，再到智能产线的升级，每一次提升都是基于企业发展到一定阶段，遇到瓶颈后的主动改造、自我进化。 数字化转型网www.szhzxw.cn
在企业发展到当前阶段，生产过程仍然面临四个方面问题需要解决：①生产模式限制产能提升；②计划准时完成率不高；③质量提升不明显，追溯不易；④生产成本不易压缩。
因此需要使用新模式、新体系、新方法和新工具来变革组织，实现发展。从 2022 年起，以自动化生产线为基础，应用数字孪生、大数据分析、边缘计算和 5G 等新技术建设动态计划、智能控制、柔性调度系统，将人工执行升级为数据驱动的自动执行；围绕基础建设、生产准备、生产执行和生产追溯四方面来建设整个生产系统，形成无人化生产体系的探索和实践。
二、无人化生产体系建设实践
在黑灯工厂建设的探索过程中，基于当前面临的问题，通过充分梳理需求，整理出包括减少人力提高生产效率、降低生产成本、提高产品质量、提升计划准确完成率，以及提高产能五方面目标，通过分解实施，达到了预期效果，并构建出一整套无人生产体系。 数字化转型网www.szhzxw.cn
1. 重构业务流程
尽管此次智能化改造之前，企业已经具备了先进生产设备，但业务流程的设计仍然是依据传统生产组织模式进行的，为了实现生产过程的无人化，彻底解放人工，必须重构生产业务流程，对信息系统底层逻辑进行全面优化，减少生产准备阶段人员的工作量，优化生产、提升协同管理能力。重构业务流程示意图如图 1 所示。
为此，在“黑灯工厂”无人化生产模式下，对生产资料等要素进行了升级，生产的组织管理模式需要变革，生产的业务流程需要重构；以车间 MES 和智能柔性管控系统为核心，以网络安全防护为保障，实现计划到执行的业务流、信息流、物料流的流程统一；以生产数据为基础的数字孪生系统，反向驱动、优化生产及协同管理。 数字化转型网www.szhzxw.cn
2. 设备和生产线的改造
在规划了智能车间及无人化生产体系，并梳理了无人条件下的业务流程后，经过严谨论证、方案规划和评审，进行了包括工艺布局、车间物流和机器人干涉等仿真验证，作为产线正式建设的指导。
按机床加工特性进行分组布局建设产线，针对不同的加工能力，设计实施不同的线内上下料机器人及辅助设备；基于产线主控 PLC 逻辑实现线内设备动作执行，同时建设工业环网，结合物联网支撑机床与各生产要素互联互通。升级智能生产线如图 2 所示。
1）设备智能化改造。黑灯工厂的基本元素是智能设备。为适配无人化操作，对车间设备进行了智能化改造，通过增加电磁阀、传感器、通信口和机床 PLC 控制逻辑，使机床具备感知、执行、检测、通信和远程控制等关键能力，为智能产线建设打下坚实基础。 数字化转型网www.szhzxw.cn
2）自动上下料设备实施。减少人工转运的时间浪费，降低人员伤害，提升生产效率。
3）线边库实施。实现原料、成品线内暂存，满足产线自动生产短期物料供应。
4）线边接驳台实施。满足产线与车间物流对接，以连续供应产线物料、提升换型效率。
5）安全防护等辅助设施实施。隔离人员活动与自动化设备工作区，保障人员和设备运行安全
3. 智能仓储建设
智能仓储系统是物料的起点和终点，是物料转运的十字路口，是智能工厂的核心要素。建设的智能仓储系统的特点是具备货位存储的动态规划能力，可以根据产线拉动式需求做自适应式任务调度，并与 MES 进行集成，实现物料齐套信息的自动传输；基于数字孪生技术实现了库存周转分析和货位分析优化。智能仓储系统如图 3 所示。
4. 建设 5G 专网及三网安全防护
为满足车间数据采集、AGV 物流控制及后期视频监控、直播等应用，实现园区内智能化，建设了覆盖园区、车间的 5G 专网，对园区主要工作场所进行 100% 全覆盖。区别于三大运营商的 5G 网络，该专网将信令系统和数据系统一并下沉于企业内部，保证了数据不出园区，将数据通过虚拟加密通道传输到内网服务器，保障数据安全。 数字化转型网www.szhzxw.cn
5. 车间智能物流系统
替代人工是无人生产必要要素，智能物流系统基于5G 技术实现 AGV 对物料在立体库、装调间、产线之间的转运。系统具备自动避障、自主充电、多 AGV 智能调度和二维码自动循迹等特点，实现了生产过程物料调度转运无人化，减少了生产待料时间，提高生产效率，降低了生产运营成本。
6. 智能制造控制中心
智能制造控制中心是黑灯工厂业务执行的核心，在控制中心实现计划、装调等生产准备工作，实现生产准备的前置化，使无人生产区与生产准备区分离；并依托建设于此的企业大脑、数字孪生，为车间管理提供分析、决策依据。如图 4 所示。
建设数字孪生，应用 AI+ 大数据技术，多源异构数据处理、数据驱动的建模，基于无限产能识别瓶颈资源，基于有限产能构建瓶颈网络优化模型。支持工厂运行监控、预警及敏捷决策。
1）采集产线、机床、AGV 和机器人等物理实体运行数据，基于边缘云实现产线监测及故障预测性分析。
2）基于设备状态预测、数字孪生工时优化进行动态排程，实现生产任务动态调度优化和产线运行优化。
3）计划准时完成率从 70% 提高到 95% ，产品一次交验合格率从 95% 提升到 99.9%。
结合业务实际需要，通过采集 ERP、PDM、MES 和OA 等核心系统的离散业务数据，形成企业数字大脑。企业大脑涵盖了企业战略运营、人力、财务、销售、采购、设备、生产和质量等多个维度。为企业生产管理和精准决策提供技术支撑，为科研生产数据赋能赋智。 数字化转型网www.szhzxw.cn
7. 车间 MES
针对车间生产计划达成率、准时完成率较低、临时插单多等问题，建设了车间级 MES。基于柔性、动态算法进行计划排产并自动驱动机床加工，实现了车间智能主导的生产，产品准时完成率明显提高。
8. 柔性管控系统
建设柔性管控系统，如图 5 所示，实现生产过程的柔性、自动化控制，工人在智能制造控制中心即可实现毛坯自动上下料、机床自动加工、在线检测和成品自动返回装调间等操作。系统的设计完全契合企业产品特点及生产组织模式，并对比、吸收了国内外优秀产品设计思路，具有一定的行业先进性。
9. 建设边缘云平台
基于边缘云平台，采集设备数据，实现设备的健康管理和预测性维护；依托数字孪生的分析优化及时实现生产工时优化，动态调度生产资源，提高计划准时完成率。
10. 建设机床主轴碰撞缓冲系统
建设机床主轴碰撞缓冲系统，通过边缘云平台对主轴振动数据采集，对主轴安全运行状态进行监控，并通过边缘云平台反控及时防止设备主轴碰撞损坏，实现机床设备的健康管理，并有效避免产线无人化后，设备严重故障时，无人及时应对的问题，有效减少设备故障造成的损失。
11. 建设机加产品机内在线检测系统
建设机加产品机内在线检测系统，系统依据预先编制的 NC 程序按加工需求进行调用，快速测量和记录关于零件几何特征的形状和位置信息，经后处理测量宏程序，自动保存相关测量数据，实时采集产品加工质量参数，实现无人化产品检测，并有效提升产品质量。在线检测系统如图 6 所示。
三、结束语
从技术创新、商业模式创新、管理创新三个方面来论述，本文旨在总结建设一个可复制、可推广，具有普适性的，适用于多数黑灯工厂建设的模式。下一步将继续加快推动黑灯工厂建设，从无人化生产模式复制及推广、智能装配线建设等方面入手，持续探索建设具有自感知、自学习、自决策、自执行和自适应等特征的智能化无人生产模式未来工厂。 数字化转型网www.szhzxw.cn
翻译：
Practice and analysis of unmanned production system construction in black light factory
Introduction
In the exploration and practice of enterprise digitalization and intelligent transformation and upgrading, the development of intelligent, flexible and large-scale industrial system, especially in the construction of intelligent unmanned production system, this paper summarizes and analyzes the key elements involved, and has a certain reference role for the digital transformation construction of relevant enterprises.
The construction of unmanned production system reasons
The products of an enterprise require high precision, high quality and high reliability, and the production mode is a typical multi-variety, small-batch discrete production mode. After years of digital transformation construction practice, it has a strong ability of automatic processing and information management. From the transformation of dumb equipment, the less humanized exploration of the production line, and then to the upgrade of the intelligent production line, each upgrade is based on the active transformation and self-evolution of the enterprise after it has reached a certain stage of development. 数字化转型网www.szhzxw.cn
In the development of the enterprise to the current stage, the production process still faces four problems to be solved: (1) the production mode limits the capacity improvement; ② The on-time completion rate of the plan is not high; ③ The quality improvement is not obvious, and the traceability is not easy; ④ Production costs are not easy to compress.
Therefore, it is necessary to use new models, new systems, new methods and new tools to transform the organization and achieve development. From 2022, on the basis of automated production lines, new technologies such as digital twin, big data analysis, edge computing and 5G will be applied to build dynamic planning, intelligent control and flexible scheduling systems, upgrading manual execution to data-driven automatic execution; The whole production system is built around the four aspects of infrastructure construction, production preparation, production execution and production traceability, and the exploration and practice of unmanned production system is formed.
Unmanned production system construction practice
In the exploration process of the construction of the black light factory, based on the current problems, by fully combing the needs, sorting out five goals including reducing manpower to improve production efficiency, reducing production costs, improving product quality, improving the accurate completion rate of the plan, and improving production capacity, through decomposition and implementation, to achieve the expected effect, and build a complete set of unmanned production system.
Reconfigure the service process
Although the enterprise already has advanced production equipment before the intelligent transformation, the design of business process is still based on the traditional production organization mode. In order to realize the unmanned production process and completely liberate the labor force, it is necessary to reconstruct the production business process, comprehensively optimize the underlying logic of the information system, and reduce the workload of personnel in the production preparation stage. Optimize production and enhance collaborative management capabilities. The diagram of refactoring business processes is shown in Figure 1. 数字化转型网www.szhzxw.cn
Figure 1 Schematic diagram of refactoring business processes
Therefore, in the unmanned production mode of “black light factory”, the production means and other factors are upgraded, the production organization and management mode need to be changed, and the production business process needs to be reconstructed. With workshop MES and intelligent flexible management and control system as the core and network security protection as the guarantee, the process unification of business flow, information flow and material flow from plan to execution is realized. Digital twin system based on production data, reverse drive, optimize production and collaborative management.
Transformation of equipment and production line
After planning the intelligent workshop and unmanned production system, and sorting out the business process under unmanned conditions, through rigorous demonstration, program planning and review, simulation verification including process layout, workshop logistics and robot interference is carried out as a guide for the formal construction of the production line.
According to the machining characteristics of the machine, the production line is constructed in a group layout. According to different processing capabilities, different in-line loading and unloading robots and auxiliary equipment are designed and implemented. Based on the main control PLC logic of the production line, the operation of the equipment in the line is implemented, and the industrial ring network is built, and the interconnection between the machine tool and various production factors is supported by the Internet of Things. The upgrade of intelligent production line is shown in Figure 2.
Figure 2 Upgrade the smart production line 数字化转型网www.szhzxw.cn
Transformation of equipment and production line
1) Intelligent transformation of equipment. The basic element of a black light factory is smart equipment. In order to adapt to unmanned operation, the workshop equipment has been intelligently transformed, by adding solenoid valves, sensors, communication ports and PLC control logic of machine tools, so that the machine tools have key capabilities such as perception, execution, detection, communication and remote control, laying a solid foundation for the construction of intelligent production line.
2) Implementation of automatic loading and unloading equipment. Reduce manual transfer time waste, reduce personnel injury, improve production efficiency.
3) Implementation of line side library. Realize the temporary storage of raw materials and finished products in the line to meet the short-term material supply of automatic production line.
4) Implementation of line side connection platform. To meet the production line and workshop logistics docking, in order to continuously supply production line materials, improve the conversion efficiency.
5) Implementation of auxiliary facilities such as safety protection. Isolate personnel activities from automated equipment work areas to ensure safe operation of personnel and equipment
Intelligent storage construction
Intelligent storage system is the starting point and end point of materials, the crossroads of material transfer, and the core element of smart factory. The construction of intelligent storage system is characterized by the dynamic planning ability of storage, can do adaptive task scheduling according to the pull demand of the production line, and integrate with MES to realize the automatic transmission of material assembly information; Based on digital twin technology, inventory turnover analysis and location analysis are optimized. The intelligent storage system is shown in Figure 3.
Figure 3 Intelligent storage system 数字化转型网www.szhzxw.cn
Build 5G private network and three-network security protection
In order to meet the workshop data collection, AGV logistics control and later video surveillance, live broadcasting and other applications, to achieve intelligent in the park, a 5G private network covering the park and workshops has been built, and 100% full coverage of the main workplaces in the park has been achieved. Different from the 5G networks of the three major operators, the private network sinks the signaling system and data system inside the enterprise together, ensuring that data does not leave the park, and transmits data to the Intranet server through a virtual encrypted channel to ensure data security.
Workshop intelligent logistics system
Replacing labor is a necessary element of unmanned production, and the intelligent logistics system is based on 5G technology to realize the transfer of AGV materials between three-dimensional warehouses, assembly rooms and production lines. The system has the characteristics of automatic obstacle avoidance, autonomous charging, multi-AGV intelligent scheduling and QR code automatic tracking, etc., which realizes the unmanned material scheduling and transfer in the production process, reduces the production waiting time, improves the production efficiency and reduces the production operation cost.
Intelligent manufacturing control center
The intelligent manufacturing control center is the core of the business execution of the black light factory, which realizes the production preparation work such as planning, installation and adjustment in the control center, realizes the preprocessing of production preparation, and separates the unmanned production area from the production preparation area. And relying on the enterprise brain and digital twin built here, it provides analysis and decision-making basis for workshop management. As shown in Figure 4. 数字化转型网www.szhzxw.cn
FIG. 5 Flexible control system
Build digital twins, apply AI+ big data technology, multi-source heterogeneous data processing, data-driven modeling, identify bottleneck resources based on unlimited capacity, and build bottleneck network optimization models based on limited capacity. Support plant operation monitoring, early warning and agile decision making.
1) Collect operation data of production line, machine tool, AGV, robot and other physical entities, and realize production line monitoring and fault predictive analysis based on edge cloud.
2) Perform dynamic scheduling based on equipment state prediction and digital twin working hour optimization to realize dynamic scheduling optimization of production tasks and operation optimization of production lines.
3) The planned on-time completion rate was increased from 70% to 95%, and the pass rate of product delivery was increased from 95% to 99.9%. 数字化转型网www.szhzxw.cn
According to the actual needs of business, the enterprise digital brain is formed by collecting discrete business data of core systems such as ERP, PDM, MES and OA. The enterprise brain covers many dimensions such as enterprise strategic operation, human resources, finance, sales, procurement, equipment, production and quality. Provide technical support for enterprise production management and accurate decision-making, and empower scientific research and production data.
Workshop MES
Aiming at the problems of workshop production plan achievement rate, low on-time completion rate, temporary order insertion and so on, workshop level MES was built. Based on flexible and dynamic algorithm, production planning and scheduling are carried out and machine tool processing is automatically driven to realize intelligent workshop production, and the on-time completion rate of products is significantly improved.
Flexible control system
Build a flexible control system, as shown in Figure 5, to achieve flexible and automatic control of the production process. Workers can realize automatic loading and unloading of blanks, automatic machining of machine tools, online testing and automatic return of finished products to the assembly and adjustment room in the intelligent manufacturing control center. The design of the system fully fits the characteristics of the enterprise product and the production organization mode. And compares and absorbs the excellent product design ideas at home and abroad. And has a certain advanced nature of the industry. 数字化转型网www.szhzxw.cn
Build edge cloud platform
Collect equipment data based on edge cloud platform to realize health management and predictive maintenance of equipment. Relying on the analysis and optimization of digital twins, the production hours can be optimized in time. And the production resources can be dynamically scheduled to improve the on-time completion rate of the plan.
Construction of machine tool spindle collision buffer system
Build a spindle collision buffer system for machine tools, collect spindle vibration data through the edge cloud platform, monitor the safe operation status of the spindle, and prevent spindle collision damage in time through the edge cloud platform counter-control, realize the health management of machine tools and equipment. And effectively avoid the problem of no one responding to serious equipment failures after the production line is unmanned. Effectively reduce the loss caused by equipment failure.
Build the internal line detection system of the machine
The internal line inspection system of the machine is built, which is called according to the pre-prepared NC program according to the processing requirements. And quickly measures and records the shape and position information of the geometric features of the parts. After the post-processing measurement macro program, the relevant measurement data is automatically saved. The product processing quality parameters are collected in real time, and the unmanned product inspection is realized. And the product quality is effectively improved. The online detection system is shown in Figure 6.
Figure 6 Online detection system 数字化转型网www.szhzxw.cn
Concluding remarks
From three aspects of technological innovation, business model innovation and management innovation, this paper aims to sum up the construction of a model that can be replicated, promoted. And is universal and suitable for the construction of most black light factories. The next step will continue to accelerate the construction of the black light factory, from the unmanned production mode replication and promotion, intelligent assembly line construction and other aspects, continue to explore the construction of intelligent unmanned production mode of the future factory with self-perception, self-learning, self-decision-making, self-execution and adaptive characteristics.
本文由数字化转型网（www.szhzxw.cn）转载而成，来源于新工业网；编辑/翻译：数字化转型网宁檬树。
免责声明: 本网站(https://www.szhzxw.cn/)内容主要来自原创、合作媒体供稿和第三方投稿，凡在本网站出现的信息，均仅供参考。本网站将尽力确保所提供信息的准确性及可靠性，但不保证有关资料的准确性及可靠性，读者在使用前请进一步核实，并对任何自主决定的行为负责。本网站对有关资料所引致的错误、不确或遗漏，概不负任何法律责任。
本网站刊载的所有内容(包括但不仅限文字、图片、LOGO、音频、视频、软件、程序等) 版权归原作者所有。任何单位或个人认为本网站中的内容可能涉嫌侵犯其知识产权或存在不实内容时，请及时通知本站，予以删除。
