Search

CN-121981621-A - Complex network-based steel production modeling method

CN121981621ACN 121981621 ACN121981621 ACN 121981621ACN-121981621-A

Abstract

The invention discloses a steel production modeling method based on a complex network, and relates to the technical field of industrial production process modeling. The method comprises the steps of deconstructing the whole flow of a steel production system, abstracting three types of units of production processing, energy/medium conversion and storage and material input/output/buffering into network nodes, extracting the physical connection of equipment and the transmission relation of production flow, defining directed edges, constructing a directed complex network model and visualizing, calculating degree distribution, global average path length, improved average path length and aggregation coefficient, and verifying the characteristics of no scale and small world. The invention realizes overall process global topology characterization and system characteristic quantification of steel production through a double-layer index system of a basic theoretical layer-engineering application layer, provides model support for overall process optimization, has both academic rigor and engineering practicability, and can be popularized to process industrial systems such as cement, thermal power and the like.

Inventors

  • LONG YAN
  • ZHAO YINGNA
  • HUA TIANYI
  • LI YUHANG
  • LU QIUYAN

Assignees

  • 华中科技大学

Dates

Publication Date
20260505
Application Date
20260119

Claims (10)

  1. 1. The steel production modeling method based on the complex network is characterized by comprising the following steps of: (1) The whole flow deconstructing is carried out on the target steel production system, and a production processing unit, an energy/medium conversion and storage unit and a material input/output/buffer unit in the system are abstracted into nodes in a complex network model; (2) Analyzing and extracting physical connection relations and/or production flow transmission relations among the nodes in the step (1), and abstracting the relations into directed edges connected with the corresponding nodes; (3) System complex network construction and visualization, namely integrating abstract nodes in the step (1) and directed edges defined in the step (2), and constructing a directed complex network model for representing the whole process of steel production Wherein For a set of nodes, The network model is visualized as a directed edge set; (4) Network topology feature calculation and verification by calculating topology feature parameters of the directed complex network model to verify its scaleless and small world characteristics and calculating an improved average path length for guiding production efficiency assessment 。
  2. 2. The method of claim 1, wherein in step (1), the production processing unit comprises at least one of a sintering machine, a coke oven, a blast furnace, a converter, a continuous casting machine, and a rolling mill, the energy/medium conversion and storage unit comprises at least one of a gas tank, a generator set, and a water treatment facility, and the material input/output/buffer unit comprises at least one of a raw material warehouse, a semi-finished product warehouse, and a final product warehouse.
  3. 3. The method of claim 1, wherein in step (2) the physical connection is determined based on physical connection pipes or channels between the devices, and wherein the production flow transfer is determined based on the actual flow of material or energy between the units.
  4. 4. The method of claim 1, wherein in step (4), the verifying the scale-free property comprises computing a degree distribution of the network and fitting the degree distribution data to a power law function When the fitness distribution accords with the power law distribution characteristics, the network is judged to have the scale-free characteristic.
  5. 5. The method of claim 1, wherein in step (4), the verifying small world properties comprises: calculating an average aggregation coefficient C and a global average path length L of the network; Calculating average aggregation coefficient of random network under same average degree Average path length ; Calculating small world index ; When (when) When >2, the determination network has small world characteristics.
  6. 6. The method of claim 5, wherein the directed complex network model is treated as a non-directed network for the time being when calculating the average aggregation factor C and the global average path length L.
  7. 7. The method of claim 1, wherein in step (4), the improved average path length The calculation formula of (2) is as follows: Where N is the total number of network nodes, The shortest path length from node i to the total product library node.
  8. 8. The method of claim 1, wherein the step (4) calls NetworkX libraries for complex network model construction and topology feature computation, and the step (3) uses Gephi software for network topology visualization.
  9. 9. The method according to any one of claims 1 to 8, wherein a two-layer index system from basic theoretical verification to engineering application is constructed in step (4), wherein the global average path length L and the average concentration coefficient C constitute a basic theoretical layer index for verifying the small world characteristics of the network, and wherein the improved average path length And forming engineering application layer indexes for directly representing the average process quantity of converting materials into products.
  10. 10. The method of claim 1, wherein the method is suitable for modeling and analysis of cement production or thermal power production flow industrial systems.

Description

Complex network-based steel production modeling method Technical Field The invention relates to the technical field of modeling of industrial production processes, in particular to a steel production modeling method based on a complex network, which is particularly suitable for systematic modeling, system characteristic analysis and global optimization guidance of the whole process of steel production, and can be popularized to modeling and analysis scenes of other complex process industrial systems such as cement, thermal power and the like. Background The steel industry belongs to a typical flow type industry, and the production flow has the remarkable characteristics of long chain, complicated working procedures and tight equipment coupling, and the material circulation, energy transfer and information interaction in the system show high nonlinearity and dynamic complexity. In the prior art, modeling means aiming at steel production are mainly focused on three types of mathematical planning, system simulation or mechanism models of a single process. Although the method achieves a certain application effect in the optimization level of local working procedures, obvious technical limitations generally exist, on one hand, the method is used for analyzing a plurality of independent sub-problems of the whole production system, and is difficult to completely describe and quantify global topological association structures among working procedures and units, and on the other hand, the whole characteristics of the production system emerging due to a network topological structure cannot be effectively revealed, and the collaborative optimization decision of the whole process is difficult to support. Complex network theory is a powerful tool for resolving complex system structures, which is good at researching the overall behavior of the system from the perspective of association. However, the application of the theory in the steel industry is focused on macroscopic steel logistics or supply chain network research targeting enterprises and regions, and the research of carrying out refined network modeling and topology characteristic analysis on all production units and material flows from raw materials to finished products is blank when the theory is applied to the interior of a specific steel production flow. Therefore, the prior art lacks an effective means for fully characterizing the steel production flow from the global topological structure level and performing scientific quantitative analysis on the steel production flow. Disclosure of Invention Object of the invention Aiming at the technical defects of 'local optimization guide, global relativity deficiency, unquantifiable system overall characteristics' and the like of the existing steel production modeling method, the invention aims to provide a steel production modeling method based on a complex network, which is characterized in that a discrete steel production system is abstracted into a unified and quantifiable complex network model, and scientific verification is carried out on the topological structure characteristics of the system, a model foundation is laid for whole-flow analysis, key node identification, vulnerability assessment and global optimization of the steel production system, and a brand new technical path is provided for realizing production flow optimization and energy efficiency improvement of steel enterprises. (II) technical scheme In order to achieve the aim of the invention, the technical scheme adopted by the invention is a steel production modeling method based on a complex network, which specifically comprises the following steps: (1) Production system deconstructing and node abstraction And (3) performing full-flow deconstructing on the target steel production system, identifying all units with production, conversion and storage core functions in the system, and abstracting each key production device and main input and output materials into nodes in a complex network model. The node covers at least three types of units: 1. The production processing unit comprises core production equipment such as a sintering machine, a coke oven, a blast furnace, a converter, a continuous casting machine, a rolling mill and the like; 2. The energy/medium conversion and storage unit comprises auxiliary guarantee units such as a gas cabinet, a generator set, a water treatment facility and the like; 3. The material input/output/buffer unit comprises a material turnover unit for raw material coal, iron ore, a semi-finished product warehouse, a final product warehouse and the like. (2) Relationship identification and definition of edges And the association relation among the nodes in the first step is analyzed and extracted, wherein the association relation comprises two types, namely a physical connection relation among the equipment and a transfer relation of leading production flow among the equipment and materials, and the association relation