CN-122021043-A - Powder preparation optimizing system based on predictive control algorithm

CN122021043ACN 122021043 ACN122021043 ACN 122021043ACN-122021043-A

Abstract

The invention relates to the field of powder preparation prediction, and discloses a powder preparation optimizing system based on a prediction control algorithm, which is used for providing a systematic and intelligent solution for improving the powder preparation working efficiency. The method comprises the steps of generating a dynamic prediction model of the milling efficiency through historical milling data analysis, establishing an organization dynamics energy balance prediction framework, generating an organization state prediction result, forming a milling work execution prediction scheme meeting the requirements of multiple resource constraints and system specifications, generating a refined prediction report based on the scheme, developing a geometric optimization-based milling decision method according to the report, and outputting a milling work efficiency improvement optimization proposal scheme. According to the invention, the problem of insufficient prediction precision in traditional administrative optimization is solved by establishing a dynamic association model of the pulverizing process and the organization operation, a refined workflow improvement scheme is provided by a geometric optimization method, and the overall operation efficiency of the pulverizing system is remarkably improved.

Inventors

GONG GUIHUA
LU JIAN
ZHU LEIYUAN
WANG HUA
Kan jian
MAO ZHEFENG
XU XIAOHUI
LIU LING

Assignees

上海上电漕泾发电有限公司

Dates

Publication Date: 20260512
Application Date: 20260209

Claims (10)

1. The powder preparation optimizing system based on the predictive control algorithm is characterized by comprising the following components: The data module is used for generating a powder preparation efficiency dynamic prediction model through analysis of historical powder preparation data; the prediction module is used for establishing a tissue dynamics energy balance prediction framework based on the powder preparation efficiency dynamic prediction model and generating a tissue state prediction result; The generation module is used for designing a pulverizing constraint algebraic prediction system by utilizing the organization state prediction result to form a pulverizing work execution prediction scheme meeting the requirements of multiple resource constraints and system specifications; The report module is used for generating a refined prediction report based on the powder process execution prediction scheme; And the decision module is used for developing a powder preparation decision method based on geometric optimization according to the refined prediction report and outputting a powder preparation work efficiency improvement optimization proposal scheme.
2. The powder process optimization system based on predictive control algorithm as set forth in claim 1, comprising: generating an organization operation state data set by collecting organization structure data and workflow data; analyzing the cooperative relationship and the energy flow path among departments based on the tissue running state data set to generate a tissue energy distribution characteristic analysis result; constructing an information flow network model by utilizing the tissue energy distribution characteristic analysis result to form a tissue dynamics model; based on the tissue dynamics model, an energy balance prediction framework is established, and an energy balance prediction result is generated; And combining the energy balance prediction result with a powder preparation efficiency dynamic prediction model to generate a comprehensive tissue state prediction result.
3. The predictive control algorithm-based pulverizing optimization system of claim 2, wherein said analyzing the collaboration relationship and energy flow paths between departments to produce tissue energy profile analysis results comprises: generating an organization cooperation network characteristic data set by collecting cooperation frequency and communication intensity data among departments; Generating an organization energy flow path analysis result based on the interaction relation among the organization cooperative network characteristic data set analysis departments; constructing an organization energy distribution characteristic map by utilizing the analysis result of the organization energy flow path and combining with the resource allocation data of each department to form a visual analysis report; Verifying the tissue energy distribution characteristic map through historical pulverizing efficiency data to generate a tissue energy distribution characteristic analysis result; performing association analysis on the tissue energy distribution characteristic analysis result and the milling efficiency dynamic prediction model to form an association mapping relation between tissue energy distribution and milling efficiency; And identifying key energy nodes influencing the milling efficiency based on the association mapping relation, and generating an organization energy optimization key region identification report.
4. The powder process optimization system based on predictive control algorithm as set forth in claim 1, comprising: analyzing the relation between powder preparation resource supply and demand based on the organization state prediction result to generate a resource constraint analysis result; Carrying out system compliance analysis on the tissue state prediction result to generate a system specification analysis result; Constructing a powder preparation constraint algebraic prediction network by using the resource constraint analysis result and the system specification analysis result to form a constraint network model; Based on the constraint network model, performing simulation prediction on the powder process execution process to generate a preliminary powder process execution prediction scheme; and optimizing and adjusting the preliminary prediction scheme through a conflict detection mechanism to generate a final powder preparation work execution prediction scheme.
5. The powder optimization system based on predictive control algorithm as set forth in claim 4, wherein said optimizing said preliminary prediction scheme by a collision detection mechanism to produce a final powder job execution prediction scheme includes: Based on the preliminary pulverizing work execution prediction scheme, potential conflict points between resource allocation and a system specification are identified, and a conflict point analysis report is generated; Constructing a conflict resolution strategy library by utilizing the conflict point analysis report and combining with a tissue running state prediction result to form a strategy set; sorting the conflict resolution schemes by a priority sorting algorithm to generate a conflict resolution scheme sequence based on importance and urgency classification; based on the conflict resolution scheme sequence, adjusting and optimizing a preliminary pulverizing work execution prediction scheme to generate an optimized pulverizing work execution prediction scheme; and testing the optimized powder process execution prediction scheme to form a powder process execution prediction scheme.
6. The powder process optimization system based on predictive control algorithm as set forth in claim 1, comprising: Based on the powder process execution prediction scheme, extracting time characteristic parameters of the powder process, and generating a time characteristic analysis result of the powder process; separating the speed dynamic characteristics of the powder making work by using the time characteristic analysis result to generate a time scale separation result; based on the time scale separation result, a multi-time scale prediction coordination mechanism is established, and a preliminary multi-scale prediction result is generated; carrying out fusion treatment on the preliminary multi-scale prediction results to form powder preparation efficiency trend prediction; and correcting the milling efficiency trend prediction based on actual milling work data to generate a refined prediction report.
7. The predictive control algorithm based pulverizing optimization system of claim 6, wherein the short term dynamics are separated by high pass filtering and the long term dynamics are extracted by low pass filtering: Wherein, the For the original sequence of the powder preparation efficiency, As a low-frequency component of the light, As the high-frequency component of the wave, Is a slowly varying component; Wherein, the As the integrated predicted value after the fusion, As a short-term predictor value of the prediction, As the predicted value of the mid-term, Is a long-term predictive value.
8. The powder process optimization system based on predictive control algorithm as set forth in claim 1, comprising: generating a capacity distribution map based on the milling efficiency trend analysis data in the refined prediction report; carrying out geometric path planning of powder preparation resource allocation by utilizing the capability distribution map to generate a resource optimization allocation scheme; based on the resource optimization configuration scheme, combining historical workflow data to form a topology optimization scheme with improved flow; verifying the flow improvement topology optimization scheme to generate a work flow improvement scheme; And integrating the resource optimal allocation scheme and the workflow improvement scheme to form a complete powder preparation work efficiency improvement optimal proposal scheme.
9. The powder process optimization system based on a predictive control algorithm of claim 1, further comprising a comparison module: Based on the difference analysis of the actual milling work execution data and the prediction result, generating a milling efficiency prediction accuracy assessment report; carrying out parameter correction on the powder production efficiency dynamic prediction model by using the powder production efficiency prediction precision evaluation report to generate an optimized powder production efficiency prediction model; re-executing the tissue dynamics energy balance prediction through the optimized pulverizing efficiency prediction model to generate an updated tissue state prediction result; dynamically adjusting a pulverizing constraint algebraic prediction system based on the updated tissue state prediction result to form a self-adaptive constraint prediction scheme; Generating a dynamically optimized powder preparation work efficiency improvement proposal scheme by utilizing the self-adaptive constraint prediction scheme; and comparing and analyzing the dynamic optimized proposal with the original optimized proposal to form a continuous improvement mechanism of the powder preparation work efficiency optimization proposal.
10. The powder manufacturing optimization system based on a predictive control algorithm as set forth in claim 9, wherein the powder manufacturing efficiency predictive accuracy assessment report uses a percentage of mean absolute error As a core evaluation index: Where n is the number of samples, For the actual measurement at time t, Is the model predicted value at time t.

Description

Powder preparation optimizing system based on predictive control algorithm Technical Field The invention relates to the field of powder preparation prediction, in particular to a powder preparation optimizing system based on a prediction control algorithm. Background In industrial production, the pulverizing system is a key link in the fields of thermal power generation, metallurgy, grain processing and the like, and the operation efficiency directly influences the overall energy efficiency and the economical efficiency. Meanwhile, the efficient operation of the pulverizing process is highly dependent on administrative cooperation and resource allocation across departments. However, the existing optimization methods still have obvious disadvantages. The prior art lacks accurate dynamic prediction capabilities. The traditional powder preparation efficiency evaluation is mostly based on a static model and historical data, and a dynamic association model between a powder preparation process and an organization running state cannot be established, so that a prediction result has larger deviation from an actual working condition, and the method is difficult to adapt to a production environment which changes in real time; the existing method has insufficient processing capacity for multiple constraints. The powder preparation optimization involves multiple constraints such as resource supply, system specification, department cooperation and the like, but the existing system often adopts a simplified processing mode, lacks systematic constraint algebraic modeling and analysis mechanisms, and causes the defect of the generated optimization scheme in terms of feasibility; There is a lack of sophisticated decision support. For the identified efficiency bottleneck, the existing method is difficult to provide a refined solution based on the geometric optimization theory, and particularly has insufficient capability in terms of topological optimization and resource path planning of the workflow, so that optimization suggestions often stay at the theoretical level and have limited operability. Therefore, we propose a powder process optimization system based on predictive control algorithm to solve the above problems. Disclosure of Invention The invention provides a powder preparation optimizing system based on a predictive control algorithm, which is used for providing a systematic and intelligent solution for improving the powder preparation working efficiency. The invention provides a pulverizing optimization system based on a predictive control algorithm, which comprises a data module, a prediction module, a generation module, a report module and a decision module, wherein the data module is used for generating a pulverizing efficiency dynamic prediction model through historical pulverizing data analysis, the prediction module is used for establishing an organization dynamic energy balance prediction framework based on the pulverizing efficiency dynamic prediction model to generate an organization state prediction result, the generation module is used for designing a pulverizing constraint algebraic prediction system by utilizing the organization state prediction result to form a pulverizing work execution prediction scheme meeting multiple resource constraint and system specification requirements, the report module is used for generating a refined prediction report based on the pulverizing work execution prediction scheme, and the decision module is used for developing a pulverizing decision method based on geometric optimization according to the refined prediction report to output a pulverizing work efficiency improvement optimization proposal scheme. Optionally, in a first implementation manner of the first aspect of the present invention, an organization operation state dataset is generated by collecting organization structure data and workflow data, a collaboration relationship and an energy flow path between departments are analyzed based on the organization operation state dataset to generate an organization energy distribution characteristic analysis result, an information flow network model is constructed by using the organization energy distribution characteristic analysis result to form an organization dynamics model, an energy balance prediction framework is established based on the organization dynamics model to generate an energy balance prediction result, and the energy balance prediction result is combined with a powdering efficiency dynamic prediction model to generate a comprehensive organization state prediction result. Optionally, in a second implementation manner of the first aspect of the present invention, the analyzing the collaboration relationship and the energy flow path between each department generates an organization energy distribution feature analysis result, which includes generating an organization collaboration network feature dataset by collecting collaboration frequency and communi