Search

CN-121979135-A - Intelligent scheduling method for plastic production

CN121979135ACN 121979135 ACN121979135 ACN 121979135ACN-121979135-A

Abstract

The application provides an intelligent scheduling method for plastic production, which relates to the technical field of plastic production. And constructing an inference decision module which comprises pattern matching, an inference machine and a learning loop. And reading the production scheduling task, performing task reasoning and global planning through reasoning decision, and determining an initial scheduling strategy. The strategy is constrained and decomposed by a dispatch buffer zone to form a pre-dispatch strategy. The strategy is executed by a cascade controller, and comprises an inner control loop and an outer control loop, so that the control of the production scheduling task is realized. The application can solve the problems that the production scheduling strategy is difficult to adapt to complex and changeable production environments due to the lack of dynamic adjustment capability in the prior art, further influences the production efficiency and the product quality, realizes that various changes in the production process can be dealt with in real time, and improves the flexibility and the robustness of the production scheduling.

Inventors

  • WANG YANBO

Assignees

  • 群冠(南通)精密科技有限公司

Dates

Publication Date
20260505
Application Date
20251231

Claims (8)

  1. 1. An intelligent scheduling method for plastic production, which is characterized by comprising the following steps: Basic configuration information of the interactive plastic production workshop, wherein the basic configuration information comprises equipment configuration and system configuration; Combining the basic configuration information to construct an inference decision module, wherein the inference decision module comprises a pattern matcher, an inference machine and a learning loop, and the learning loop takes a deduction network as a basic pattern; Reading a production scheduling task, wherein the production scheduling task comprises at least one rigid task or flexible task for plastic production; Combining the reasoning decision module, performing task reasoning and global overall planning on the production scheduling task, and determining an initial scheduling strategy; performing strategy point scheduling constraint on the initial scheduling strategy through a scheduling buffer zone, and performing strategy decomposition based on direct control and indirect control to determine a pre-scheduling strategy; the prescheduling strategy responds to a cascade controller to perform scheduling control based on the production scheduling task, wherein the cascade controller comprises a control inner ring and a control outer ring.
  2. 2. An intelligent scheduling method for plastic production according to claim 1, wherein the learning circuit is based on a deductive network, comprising: The learning loop comprises an external interaction branch, and task records and newly added task classes are retrieved based on a preset period; Performing inductive deduction based on deduction network for the task record and the newly added task class, and determining a push chain; and based on the push chain, performing self-learning management of the reasoning decision module.
  3. 3. An intelligent scheduling method for plastic production according to claim 1, wherein said determining an initial scheduling strategy comprises: identifying the production scheduling task, and if the task magnitude is greater than 1, determining a task overall planning mode by combining the mode matcher based on task characteristics and task correlation; Based on the task magnitude, matching an inference engine and performing transient integration to construct an inference architecture, wherein the inference architecture has a one-to-one correspondence with the production scheduling task; and determining the initial scheduling strategy based on the reasoning architecture and the task orchestration mode.
  4. 4. A method of intelligent scheduling of plastic production according to claim 3, wherein deciding to determine the initial scheduling policy comprises: Determining an inference mechanism based on the task characteristics, wherein the inference mechanism comprises a forward inference mechanism and a reverse inference mechanism; Based on the reasoning architecture, carrying out reasoning decision on the production scheduling task by combining a mapped reasoning mechanism, and determining a single task strategy; and carrying out coordination and overall planning on the single task strategy based on the task overall planning mode, and determining the initial scheduling strategy.
  5. 5. The intelligent scheduling method for plastic production of claim 1, wherein performing policy point scheduling constraints on the initial scheduling policy comprises: identifying the initial scheduling strategy, analyzing scheduling stability aiming at each strategy point, and determining a stability coefficient; Traversing the stability coefficient, and determining a strategy point dead zone, wherein the strategy point dead zone comprises a vector value and acceleration; and carrying out mapping identification on the initial scheduling strategy based on the strategy point dead zone, and carrying out dead zone zero setting treatment.
  6. 6. An intelligent scheduling method for plastic production according to claim 1, wherein determining the pre-scheduling strategy based on strategy decomposition of direct control and indirect control comprises: Traversing the initial scheduling strategy, and determining a scheduling trend strategy, wherein the scheduling trend strategy is used for indirect auxiliary control; traversing the initial scheduling strategy, and determining a scheduling vector strategy, wherein the scheduling vector strategy is used for direct control; And determining the pre-scheduling strategy based on the scheduling trend strategy and the scheduling vector strategy.
  7. 7. The intelligent scheduling method of plastic production according to claim 6, wherein the control inner ring is configured based on the scheduling trend strategy, the control outer ring is configured based on the scheduling vector strategy, and the control inner ring and the control outer ring are in communication connection with the intelligent central control system.
  8. 8. The intelligent scheduling method for plastic production according to claim 5, wherein after performing scheduling control based on the production scheduling task, comprising: synchronously carrying out task scheduling monitoring, measuring strategy response deviation, and determining deviation data; Positioning a preset feedback point based on the deviation data by taking the strategy point dead zone as constraint; performing production scheduling feedback adjustment and overshoot oscillation judgment based on the preset feedback points, and determining a feedback scheduling strategy, wherein the overshoot oscillation comprises a single point and a global point; and performing task feedback scheduling management based on the feedback scheduling strategy.

Description

Intelligent scheduling method for plastic production Technical Field The application relates to the technical field of plastic production, in particular to an intelligent scheduling method for plastic production. Background Rapid changes in market demand require that the production system be able to respond quickly. The traditional dispatching system has low response speed, and is difficult to meet market demands with short exchange period and multiple changes. The intelligent scheduling method can shorten the production preparation time and the response time through rapid data processing and decision support. In the plastic production process, the product quality is affected by various factors, and the traditional scheduling method has difficulty in comprehensively considering the factors. The intelligent scheduling method can continuously learn and optimize the scheduling strategy through the learning loop, reduce the occurrence of quality problems and improve the product quality. In the traditional production scheduling, due to the lack of global overall planning and optimization, the problems of unbalanced equipment utilization rate, production line bottleneck and the like often occur. The intelligent scheduling method can better optimize resource allocation and improve equipment utilization rate through global overall planning and strategy point scheduling constraint. Currently, existing plastic production scheduling systems are mostly based on static rules or simple algorithms, and lack flexibility and self-adaptive capacity. In summary, the lack of dynamic adjustment capability in the prior art makes it difficult for the production scheduling strategy to adapt to complex and variable production environments, thereby further affecting production efficiency and product quality. Disclosure of Invention The application aims to provide an intelligent scheduling method for plastic production, which is used for solving the problem that the production scheduling strategy is difficult to adapt to complex and changeable production environments due to the lack of dynamic adjustment capability in the prior art, and further influencing the production efficiency and the product quality. In view of the above problems, the present application provides an intelligent scheduling method for plastic production. The application provides an intelligent scheduling method of plastic production, which comprises the steps of interacting basic configuration information of a plastic production workshop, constructing an inference decision module by combining the basic configuration information, wherein the inference decision module comprises a pattern matcher, an inference machine and a learning loop, the learning loop takes a deduction network as a basic pattern, reading a production scheduling task, the production scheduling task comprises at least one rigid task or flexible task which is produced by plastic, performing task inference and global overall planning on the production scheduling task by combining the inference decision module, determining an initial scheduling strategy, performing strategy point scheduling constraint on the initial scheduling strategy through a scheduling buffer zone, performing strategy decomposition based on direct control and indirect control, and determining a pre-scheduling strategy, wherein the pre-scheduling strategy is responsive to a cascade controller, and performing scheduling control based on the production scheduling task, wherein the cascade controller comprises a control inner ring and a control outer ring. One or more technical schemes provided by the application have at least the following technical effects or advantages: The method comprises the steps of establishing a reasoning decision module through interaction of basic configuration information of a plastic production workshop, establishing a reasoning decision module through combination of the basic configuration information, wherein the reasoning decision module comprises a pattern matcher, a reasoning machine and a learning loop, the learning loop takes a deduction network as a basic mode, reading a production scheduling task, the production scheduling task comprises at least one rigid task or flexible task which is produced by plastic, carrying out task reasoning and global overall planning on the production scheduling task through combination of the reasoning decision module, determining an initial scheduling strategy, carrying out strategy point scheduling constraint on the initial scheduling strategy through a scheduling buffer zone, carrying out strategy decomposition based on direct control and indirect control, and determining a pre-scheduling strategy, and the pre-scheduling strategy responds to a cascade controller to carry out scheduling control based on the production scheduling task, wherein the cascade controller comprises a control inner ring and a control outer ring, so that the problem that the production sc