Intelligent manufacturing originates from the study of artificial intelligence. It is generally believed that intelligence is the sum of knowledge and intelligence. The former is the foundation of intelligence, while the latter refers to the ability to acquire and apply knowledge to solve problems.

Intelligent manufacturing should include intelligent manufacturing technology and intelligent manufacturing systems. Intelligent manufacturing systems can not only continuously enrich the knowledge base in practice, but also have self-learning functions, as well as the ability to collect and understand environmental information and its own information, and to analyze, judge and plan its own behavior.

Manufacturing principle

Starting from the essential characteristics of intelligent manufacturing systems, and based on the fundamental principles of distributed integration, this paper applies the theories and methods of multi-agent systems in distributed artificial intelligence to achieve flexible intelligent integration of manufacturing units and network-based manufacturing systems within a distributed manufacturing network environment. Based on the isomorphic characteristics of distributed systems, this paper, based on a localized implementation of intelligent manufacturing systems, also reflects the implementation model of intelligent manufacturing systems within the global manufacturing network environment based on the Internet.

Distributed Networking

The essential characteristics of an intelligent manufacturing system are the autonomy of individual manufacturing units and the self-organizing capabilities of the system as a whole. Its basic structure is a distributed multi-agent intelligent system. Based on this concept and considering the global manufacturing network environment based on the Internet, a basic framework for a distributed networked IMS suitable for small and medium-sized enterprises can be proposed. On the one hand, agents empower each manufacturing unit with autonomy, enabling them to be independent and fully functional. On the other hand, collaboration and cooperation between agents imbue the system with self-organizing capabilities.

Based on the above framework, combined with the CNC machining system, the developed distributed network prototype system can be composed of four nodes: system manager, task planning, design and producer.

The system manager node includes two database servers, the database server and the system agent, which are responsible for managing the entire global database. It can be used by nodes with permission in the prototype system to query, read, store and retrieve data, and provide a public place for each node to exchange and share data. The system agent is responsible for the interaction between the system and the outside world on the network. The system's homepage is published on the Internet through the Web server. Online users can obtain relevant information about the system by visiting the homepage and decide whether the system should meet these needs based on their own needs. The system agent is also responsible for monitoring the interactive activities between each node on the prototype system, such as recording and displaying in real time the situation of sending and receiving messages between nodes, the execution status of tasks, etc.

The task planning node consists of a task manager and its agent (task manager agent). Its main function is to plan the tasks obtained from the Internet, decompose them into several subtasks, and then allocate these tasks to various nodes through bidding.

A design node consists of a CAD tool and its agent (design agent), providing a user-friendly human-computer interface that allows designers to effectively interact with the computer and complete design tasks. The CAD tool helps designers design products according to user requirements; the design agent is responsible for network registration, deregistration, database management, interaction with other nodes, deciding whether to accept design tasks, and submitting tasks to the task sender.