Multi-variety small-batch adaptive process planning method based on process knowledge graph

Abstract

The invention relates to the technical field of digital manufacturing and intelligent process planning. The invention discloses a multi-variety small-batch adaptive process planning method based on a process knowledge graph. The method comprises the following steps: constructing a dynamic knowledge graph of processing process-equipment capability-quality characteristics; in combination with order parameters and equipment states, performing real-time reasoning based on the dynamic knowledge graph to generate an optimal process path; and continuously optimizing the remodeling strategy by utilizing reinforcement learning. According to the method, a dynamic knowledge graph of processing technology-equipment capability-quality characteristics is constructed, an optimal technology path is generated through real-time reasoning in combination with order parameters and equipment states, and a remodeling strategy is continuously optimized through reinforcement learning. The bottleneck that a traditional CAPP system depends on artificial experience, is slow in response and cannot adapt to multi-variety small-batch mixed line production is solved, the process planning time is shortened from the hour level to the minute level, and the remodeling efficiency is improved by 40% or above.

Description

Technical Field

[0001] This invention relates to the field of digital manufacturing and intelligent process planning technology, specifically to a multi-variety, small-batch adaptive process planning method based on process knowledge graphs. Background Technology

[0002] With the trend of digital and intelligent transformation in the manufacturing industry, the multi-variety, small-batch production model has become the mainstream in the market and is widely used in aerospace, precision machinery, customized equipment and other fields. This model can quickly respond to customers' personalized needs and reduce inventory costs, but it also poses a serious challenge to traditional process planning systems (CAPP).

[0003] Traditional process planning methods rely heavily on engineers' experience, requiring the development of process paths for each product model. This not only leads to low planning efficiency but also makes it prone to inconsistencies in process solutions and poor quality stability due to differences in experience. Furthermore, traditional CAPP systems lack the ability to dynamically perceive real-time production data, failing to adapt to dynamic scenarios such as changes in equipment operating status (e.g., malfunctions, load fluctuations) and adjustments to order parameters (e.g., changes in delivery dates, upgraded quality requirements). This results in a disconnect between process paths and actual production needs, making it difficult to meet the flexible requirements of multi-variety mixed-line production.

[0004] Furthermore, frequent product changeovers are a prominent issue in multi-variety, small-batch production. Traditional changeover strategies lack scientific optimization mechanisms, often resulting in long changeover times, high costs, and significant fluctuations in product quality after changeovers, further hindering production efficiency and cost control. Existing process planning technologies fail to effectively integrate multi-dimensional knowledge such as processing technology, equipment capabilities, and quality characteristics, and also lack deep integration with artificial intelligence technologies, leading to slow response speeds and weak adaptability. This has become a core bottleneck restricting the efficient operation of multi-variety, small-batch production.

[0005] To address this issue, we propose a multi-variety, small-batch adaptive process planning method based on process knowledge graphs. Summary of the Invention

[0006] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a multi-variety, small-batch adaptive process planning method based on process knowledge graphs, which solves the problems mentioned in the background technology.

[0007] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a multi-variety, small-batch adaptive process planning method based on process knowledge graphs, comprising the following steps: Construct a dynamic knowledge graph of processing technology, equipment capabilities, and quality characteristics; Combining order parameters and equipment status, the optimal process path is generated in real time based on the dynamic knowledge graph; Continuously optimize the transformation strategy using reinforcement learning.

[0008] Preferably, the step of constructing a dynamic knowledge graph of processing technology, equipment capability, and quality characteristics specifically includes: Collect data and information related to the processing technology, including the operation steps and process parameter ranges of various processing technologies; Collect data and information related to equipment capabilities, including equipment models, equipment performance parameters, and the range of processes that the equipment can process; Collect data and information related to quality characteristics, including various quality indicators and quality standard requirements of the product; The collected data is structured and a dynamic knowledge graph is constructed based on the relationships between processing technology, equipment capabilities, and quality characteristics.

[0009] Preferably, in the step of generating the optimal process path in real time based on the dynamic knowledge graph by combining order parameters and equipment status, the order parameters include product model, order quantity, delivery date, and quality requirements; the equipment status includes equipment operating status, remaining processing capacity, and equipment fault information.

[0010] Preferably, the step of generating the optimal process path based on the dynamic knowledge graph in real-time reasoning specifically includes: Order parameters and device status are input data into the dynamic knowledge graph; By leveraging the relationships and reasoning rules between nodes in a dynamic knowledge graph, multi-dimensional analysis and matching can be performed. Taking into account factors such as the feasibility of the processing technology, the efficiency of equipment use, product quality requirements, and delivery time, the optimal process path that meets the order requirements is generated.

[0011] Preferably, the step of continuously optimizing the transformation strategy using reinforcement learning specifically includes: Define the state, action, and reward functions during the transformation process; After each transformation operation, the reward value is calculated based on the transformation result and the reward function; The policy parameters in the reinforcement learning model are updated based on the reward value. By iteratively optimizing the changeover policy, the changeover time and cost can be reduced.

[0012] Preferably, the defined state during the changeover process includes the current equipment status, product information to be processed, and completed processing steps; actions include selecting changeover equipment, adjusting equipment parameters, and changing cutting tools or fixtures; the reward function is set based on factors such as changeover time, changeover cost, and product quality stability after changeover.

[0013] Preferably, the method further includes a step of updating the dynamic knowledge graph in real time, dynamically adjusting and optimizing the node information and relationships in the dynamic knowledge graph based on newly collected processing technology data, equipment status data and quality feedback data during actual production.

[0014] Preferably, the real-time reasoning of the dynamic knowledge graph includes process compatibility rules, equipment load balancing rules, and quality compliance priority rules. The reasoning process adopts a combination of forward reasoning and backward reasoning to improve the generation efficiency of the optimal process path.

[0015] (III) Beneficial Effects Compared with existing technologies, this invention provides a multi-variety, small-batch adaptive process planning method based on process knowledge graphs, which has the following beneficial effects: The invention constructs a dynamic knowledge graph of processing technology, equipment capabilities, and quality characteristics. It combines order parameters and equipment status to generate the optimal process path in real time, and continuously optimizes changeover strategies through reinforcement learning. This solves the bottlenecks of traditional CAPP systems, which rely on manual experience, have slow response times, and cannot adapt to mixed production lines with multiple product types and small batches. It reduces process planning time from hours to minutes, improving changeover efficiency by over 40%. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. 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.

[0017] A multi-variety, small-batch adaptive process planning method based on process knowledge graph includes the following steps: Constructing a dynamic knowledge graph of processing technology, equipment capabilities, and quality characteristics, specifically including: Collect data and information related to the processing technology, including the operation steps and process parameter ranges of various processing technologies; Collect data and information related to equipment capabilities, including equipment models, equipment performance parameters, and the range of processes that the equipment can process; Collect data and information related to quality characteristics, including various quality indicators and quality standard requirements of the product; The collected data is structured and a dynamic knowledge graph is constructed based on the relationships between processing technology, equipment capabilities, and quality characteristics. Combining order parameters and equipment status, the optimal process path is generated in real time based on the dynamic knowledge graph. The order parameters include product model, order quantity, delivery date, and quality requirements; the equipment status includes equipment operating status, remaining processing capacity, and equipment fault information. The specific steps for generating the optimal process path based on the dynamic knowledge graph in real-time reasoning are as follows: Order parameters and device status are input data into the dynamic knowledge graph; By leveraging the relationships and reasoning rules between nodes in a dynamic knowledge graph, multi-dimensional analysis and matching can be performed. Taking into account factors such as the feasibility of processing technology, equipment utilization efficiency, product quality requirements, and delivery time, the optimal process path that meets the order requirements is generated. The dynamic knowledge graph real-time reasoning includes process compatibility rules, equipment load balancing rules, and quality compliance priority rules. The reasoning process adopts a combination of forward and backward reasoning to improve the efficiency of generating the optimal process path. Continuously optimize the transformation strategy using reinforcement learning, specifically including: Define the state, actions, and reward function during the changeover process. The state during the changeover process includes the current equipment status, information about the product to be processed, and the processing steps that have been completed. The actions include selecting changeover equipment, adjusting equipment parameters, and changing tools or fixtures. The reward function is set based on factors such as changeover time, changeover cost, and the stability of product quality after changeover. After each transformation operation, the reward value is calculated based on the transformation result and the reward function; The policy parameters in the reinforcement learning model are updated based on the reward value. The switching policy is continuously optimized through iteration to reduce switching time and cost. The steps for real-time updating of the dynamic knowledge graph involve dynamically adjusting and optimizing the node information and relationships in the dynamic knowledge graph based on newly collected processing technology data, equipment status data, and quality feedback data during actual production.

[0018] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-variety, small-batch adaptive process planning method based on process knowledge graph, characterized in that, Includes the following steps: Construct a dynamic knowledge graph of processing technology, equipment capabilities, and quality characteristics; Combining order parameters and equipment status, the optimal process path is generated in real time based on the dynamic knowledge graph; Continuously optimize the transformation strategy using reinforcement learning.

2. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 1, characterized in that, The steps for constructing a dynamic knowledge graph of processing technology, equipment capabilities, and quality characteristics specifically include: Collect data and information related to the processing technology, including the operation steps and process parameter ranges of various processing technologies; Collect data and information related to equipment capabilities, including equipment models, equipment performance parameters, and the range of processes that the equipment can process; Collect data and information related to quality characteristics, including various quality indicators and quality standard requirements of the product; The collected data is structured and a dynamic knowledge graph is constructed based on the relationships between processing technology, equipment capabilities, and quality characteristics.

3. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 1, characterized in that, In the step of generating the optimal process path in real time based on the dynamic knowledge graph by combining order parameters and equipment status, the order parameters include product model, order quantity, delivery date, and quality requirements; the equipment status includes equipment operating status, remaining processing capacity, and equipment fault information.

4. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 3, characterized in that, The specific steps for generating the optimal process path based on the dynamic knowledge graph in real-time reasoning are as follows: Order parameters and device status are input data into the dynamic knowledge graph; By leveraging the relationships and reasoning rules between nodes in a dynamic knowledge graph, multi-dimensional analysis and matching can be performed. Taking into account factors such as the feasibility of the processing technology, the efficiency of equipment use, product quality requirements, and delivery time, the optimal process path that meets the order requirements is generated.

5. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 1, characterized in that, The steps for continuously optimizing the transformation strategy using reinforcement learning specifically include: Define the state, action, and reward functions during the transformation process; After each transformation operation, the reward value is calculated based on the transformation result and the reward function; The policy parameters in the reinforcement learning model are updated based on the reward value. By iteratively optimizing the changeover policy, the changeover time and cost can be reduced.

6. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 5, characterized in that, The defined state during the changeover process includes the current equipment status, product information to be processed, and completed processing steps; actions include selecting changeover equipment, adjusting equipment parameters, and changing cutting tools or fixtures; the reward function is set based on factors such as changeover time, changeover cost, and product quality stability after changeover.

7. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 1, characterized in that, The method also includes a step of updating the dynamic knowledge graph in real time, dynamically adjusting and optimizing the node information and relationships in the dynamic knowledge graph based on newly collected processing technology data, equipment status data and quality feedback data during actual production.

8. The multi-variety, small-batch adaptive process planning method based on process knowledge graph according to claim 1, characterized in that, The dynamic knowledge graph real-time reasoning includes process compatibility rules, equipment load balancing rules, and quality compliance priority rules. The reasoning process adopts a combination of forward and backward reasoning to improve the efficiency of generating the optimal process path.