Quality health degree evaluation and cross-machine group early warning method based on multi-process roll-by-roll detection

Abstract

The application discloses a quality health degree evaluation and cross-machine-group early warning method based on multi-process roll-by-roll detection and relates to the technical field of metallurgical process quality management and control.The application realizes roll-by-roll actual measurement on each roll of material in each key process, collects quality data with the finished product roll number as the main key, automatically evaluates the quality health degree, and automatically issues a quality early warning before the roll enters the next process, so that the downstream process can know the risk in advance and adjust the process in advance.The application realizes the change from "blind box type production" to "transparent and pre-known production", effectively reduces batch waste, improves the yield, reduces quality loss, and is suitable for various metallurgical roll continuous production lines.

Description

Technical Field

[0001] This invention relates to the fields of metallurgical process quality control, process collaboration and intelligent early warning technology, specifically to a method for evaluating quality health through multi-process, roll-by-roll measured data and providing early warnings for the next process. Background Technology

[0002] Traditional metallurgical production is mostly carried out in a "blind production" mode, which has the following prominent problems: The previous process had problems such as uneven thickness, poor plate shape, and local deviations in the raw materials / semi-finished products, which the next process was unaware of beforehand; The production process relies on experience and luck, and is similar to the blind box model, with anomalies only being discovered after a batch of abnormalities occur. Quality incidents can only be traced and the responsibility determined after the fact; they cannot be avoided in advance. There is a lack of overall quality and health assessment of each roll at each stage of the process; The lack of information sharing between processes prevents downstream industries from adjusting their processes and taking preventative measures in advance.

[0003] Existing technologies cannot achieve roll-by-roll quality evaluation, transparent transfer between processes, or proactive early warning. Summary of the Invention

[0004] Purpose of the invention This paper presents a quality health evaluation and cross-unit early warning method based on multi-process roll-by-roll inspection, realizing "every roll has a physical examination, transparency between processes, and early warning of abnormalities", completely eliminating the need for blind box production. Technical solution

[0005] In key processes such as hot rolling, pickling, rolling mill, galvanizing / tin plating, leveling, and tension straightening, data is collected for each roll of material, including thickness, width, plate shape, crown, local high / low points, zinc layer, coating amount, and surface defects. Using the finished roll number as the primary key, the measured data of each process are summarized and stored in a unified quality database; The system automatically calculates the quality health of the roll based on preset standards (upper and lower limits of each parameter, defect level threshold), and divides it into three levels: qualified (no out-of-tolerance items), slightly abnormal (single item slightly out of tolerance, not affecting subsequent processing), and seriously abnormal (multiple out-of-tolerance items or single item seriously out of tolerance, which may lead to scrap). Before the next production process begins, the system automatically identifies all upstream quality results based on the finished roll number; For rolls with minor / serious abnormalities, an automatic warning is issued to the next process, clearly indicating the risk points (such as "0.1mm thinner in the middle", "0.2mm thicker at the edge", "surface scratches"). The next process operator adjusts the process parameters (such as adjusting the mill pressure and tension) based on the warning to avoid producing defective products; The entire process of quality data, health status, early warning records, and process adjustment records are automatically stored, forming a complete traceability chain.

[0006] Core Innovation Points Each roll is inspected individually, and the health assessment is conducted to achieve a full-process "physical examination" of every roll of material, thus eliminating blind mass production. Transparent transmission of quality information between processes and early warning of upstream anomalies shifts the focus from "post-event accountability" to "pre-event prevention." The early warning information is accurate to specific risk points, guiding downstream processes to make targeted adjustments to their processes and reduce scrap rates; The closed-loop traceability of data throughout the entire process provides complete data support for quality analysis and process optimization. Beneficial effects

[0007] Batch scrap rate reduced by more than 60%, yield significantly improved, and quality loss reduced; Inter-process communication costs are reduced by 80%, eliminating the need for manual transmission of quality information and preventing information omissions; Quality problems can be quickly pinpointed to specific processes and parameters, facilitating process optimization. The production process is transparent, and managers can monitor the quality status of each roll of material in real time. It is compatible with all metallurgical coil continuous production lines, has strong versatility, and brings significant economic benefits after implementation. Detailed Implementation

[0008] Testing equipment (thickness gauge, plate shape gauge, zinc coating gauge, etc.) is configured in each key process. After each roll of material is produced, the measured data is automatically collected, bound to the finished roll number, and uploaded to the quality database. The system presets qualified thresholds for each parameter (such as allowable thickness deviation ±0.05mm), slight abnormality threshold (±0.05-0.1mm), severe abnormality threshold (above ±0.1mm), and surface defect level standards; The system aggregates measured data from all processes of a specific material roll, compares each item against thresholds, and automatically determines the quality and health status. Pass: All parameters are within the acceptable threshold, and there are no surface defects; Minor anomaly: One parameter is within the minor anomaly threshold, with no serious defects; Severe anomaly: ≥1 parameter is within the severe anomaly threshold, or ≥2 parameters are within the minor anomaly threshold, or there is a severe surface defect; When the coil of material enters the next process (such as from the rolling mill to the galvanizing process), the HMI of the galvanizing process will automatically pop up a warning window: "Coil number CJ20240501002, health status: slight abnormality, risk point: edge thickness 0.08mm, suggested adjustment: reduce galvanizing temperature by 5℃"; Operators adjust process parameters based on warnings and record the adjustment results in the system after production is completed; All data (measured values, health status, early warning information, and adjustment records) are linked to the finished product roll number, supporting query and traceability by roll number, process, and health status level. Quality management personnel can regularly analyze abnormal trends and optimize process standards.

Claims

1. A method for quality health evaluation and cross-unit early warning based on multi-process roll-by-roll inspection, characterized in that, include: (1) Collect quality data for each coil in multiple processes such as hot rolling, pickling, rolling mill, galvanizing / tin plating, leveling, and tension straightening; (2) Using the finished roll number as the primary key, summarize the inspection results of each process and automatically calculate the quality and health status; (3) Before the roll enters the next process, the system automatically identifies and issues a quality warning to the downstream process; (4) The next process should adjust the process in advance based on the warning content to avoid quality risks; (5) The entire process of detection records, health status, and early warning information are automatically saved and traceable.

2. The method according to claim 1, characterized in that, The collected data includes thickness, width, plate shape, convexity, zinc layer, coating amount, and surface defects.

3. The method according to claim 1, characterized in that, The warning includes the location of the anomaly, the degree of the anomaly, and risk alerts, which are used to guide downstream processes to adjust their processes in advance.