Process parameter abnormity alarm method, analysis method and system

Abstract

The embodiment of the invention provides a process parameter abnormity alarm method, analysis method and system, and belongs to the technical field of process control. Comprising the steps that material information and production parameter data in the equipment operation process are collected in real time; sending the collected data to a process parameter matching rule base, and generating a dynamic alarm range corresponding to equipment target parameters through matching operation of the rule base; and according to the acquired data of the target parameter and the dynamic alarm range, carrying out comparison and judgment to determine whether an abnormal alarm is triggered or not. According to the method, a configurable process parameter matching rule base is constructed, a multi-dimensional condition matching mechanism is introduced, monitoring standards can be dynamically adjusted according to various variables in actual production, and intelligent monitoring of key process parameters and more accurate abnormity early warning are achieved.

Description

Technical Field

[0001] This invention relates to the field of process control technology, and more specifically to a method, analysis method, and system for alarming abnormal process parameters. Background Technology

[0002] In the rolling and processing of non-ferrous metals such as aluminum and copper, numerous process parameters are affected by various factors, including environment, equipment, and manual operation. Changes in these parameters during production can impact the quality of the final product. Generally, non-ferrous metal processing enterprises develop corresponding process flows and parameter control standards based on parameters such as the raw materials used, the target product alloy, and the target specifications to ensure the stability and consistency of product quality. Precise control and optimization of parameters at each stage can effectively reduce fluctuations in the production process and improve the yield rate of finished products.

[0003] In actual production scenarios, various process parameters inevitably fluctuate due to changes in temperature and humidity, improper operator handling, and other factors. These fluctuations can ultimately lead to poor product quality and damage the company's economic benefits. Therefore, it is necessary to monitor changes in process parameters during production and, based on the company's specified process parameter standards, issue timely alerts for abnormal situations, thereby reminding process personnel to make timely adjustments and reduce economic losses.

[0004] However, existing technologies often use a one-size-fits-all static standard, which cannot adapt to the actual impact of different environmental and material variables on process parameters, resulting in the inability to identify abnormal situations in a timely manner or the generation of a large number of false alarms. Summary of the Invention

[0005] The purpose of this invention is to provide a method, analysis method, and system for alarming abnormal process parameters, aiming to overcome the deficiencies of the prior art. It aims to build a configurable process parameter matching rule base, introduce a multi-dimensional condition matching mechanism, and dynamically adjust the monitoring standards according to various variables in actual production, thereby realizing intelligent monitoring of key process parameters and more accurate abnormal early warning.

[0006] To achieve the above objectives, on the one hand, the present invention provides a method for alarming abnormal process parameters. The alarm method includes: real-time collection of material information and production parameter data during equipment operation; sending the collected data to a process parameter matching rule library; generating a dynamic alarm range for the corresponding target parameter of the equipment through matching calculations in the rule library; and comparing and judging whether an abnormal alarm is triggered based on the collected data of the target parameter and the dynamic alarm range.

[0007] Optionally, the process parameter matching rule base is constructed based on the following method: configuring matching conditions for the equipment according to multiple production parameters associated with the equipment; configuring standard threshold values ​​for each matching condition, including parameter definitions for configuration condition thresholds, process parameters to be alarmed, and corresponding process parameter alarm baseline values; and dynamically generating dynamic alarm threshold ranges for the target parameters according to the real-time activation status of the matching conditions.

[0008] Optionally, the production parameters include at least one of the following: equipment type, seasonal parameters; the material information includes at least one of the following: alloy type, incoming material specifications, finished product specifications; and the target parameters include at least one of the following: rolling force, temperature, liquid level, and current.

[0009] Optionally, the step of comparing and determining whether to trigger an abnormal alarm based on the collected data of the target parameter and the dynamic alarm range includes: when the target parameter is a single parameter, executing a single-parameter independent alarm for the target parameter; or when the target parameter is multiple parameters, executing a multi-parameter linkage combined alarm based on the parameter combination relationship of the multiple target parameters.

[0010] Optionally, the single-parameter independent alarm includes: if the real-time collected data of the target parameter exceeds the range of the dynamic alarm threshold, an abnormal alarm is directly triggered; or if the real-time collected data of the target parameter exceeds the range of the dynamic alarm threshold, a parameter over-limit violation is recorded, and an abnormal alarm is triggered when the number of violations reaches a first set threshold or the proportion of violations within a set time period reaches a first preset proportion threshold.

[0011] Optionally, the multi-parameter linkage combined alarm includes: for each target parameter in the parameter combination, comparing its real-time collected data with its corresponding dynamic alarm threshold range to filter out abnormal target parameters that exceed its dynamic alarm range; and triggering an abnormal alarm when the number of abnormal target parameters reaches a second set threshold or the proportion in the parameter combination reaches a second preset proportion threshold.

[0012] Optionally, the target parameters in the parameter combination satisfy a validity verification, wherein the validity verification is performed by employing a time-series correlation verification method.

[0013] Optionally, the dynamic alarm range includes a set value and at least one set of upper and lower limits, wherein when the dynamic alarm range includes multiple sets of upper and lower limits, the abnormal alarm includes multiple alarm levels.

[0014] On the other hand, embodiments of the present invention provide a process parameter anomaly alarm system, the alarm system comprising: a production parameter acquisition device for real-time acquisition of material information and production parameter data during the operation of production equipment; a rule matching processing device for sending the acquired data to a process parameter matching rule library, and generating a dynamic alarm range for the corresponding equipment target parameter through matching operations in the rule library; and an anomaly determination alarm device for comparing and determining whether an anomaly alarm is triggered based on the acquired data of the target parameter and the dynamic alarm range.

[0015] Another aspect of the present invention provides a process parameter analysis method, the analysis method comprising: outputting abnormal alarm records through the process parameter abnormal alarm method; and performing multi-dimensional statistical analysis on the abnormal alarm records to identify characteristic abnormal patterns.

[0016] Through the above technical solutions, the present invention provides a method, analysis method and system for alarming abnormal process parameters, aiming to build a configurable process parameter matching rule base, introducing a multi-dimensional condition matching mechanism, which can dynamically adjust the monitoring standards according to various variables in actual production, and realize intelligent monitoring of key process parameters and more accurate abnormal early warning.

[0017] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the process parameter abnormality alarm method provided in this embodiment of the invention.

[0019] Figure 2 This is a schematic diagram illustrating the enabling of matching conditions provided in an embodiment of the present invention.

[0020] Figures 3a-3b This is a schematic diagram of the process parameters and their upper and lower limits provided in the embodiments of the present invention.

[0021] Figure 4 This is a schematic diagram of the process parameter abnormality alarm system provided in an embodiment of the present invention.

[0022] Figures 5a-5d This is a schematic diagram of alarm record statistical analysis provided in an embodiment of the present invention. Detailed Implementation

[0023] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0024] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0025] Before introducing the various embodiments of the present invention, a brief description of the prior art is provided. Currently, a commonly used process monitoring method is based on static parameter ranges. Its core principle is to preset fixed upper and lower limits for each process parameter. The system collects production data in real time and compares it with these preset thresholds. Once a parameter is detected to exceed the allowable range, an alarm is triggered. This system typically consists of three key components: first, engineers set fixed parameter standard values ​​and their allowable deviation ranges based on experience or historical data; second, real-time process data is continuously collected through a field sensor network; and finally, an over-limit alarm is implemented through simple logical judgment.

[0026] However, the applicant found that although this solution seemed simple to implement and quick to respond, its inherent defects were very obvious: due to the use of a one-size-fits-all static standard, it could not adapt to the actual impact of variables such as different alloy materials, seasonal environment, and incoming material specifications on process parameters; at the same time, it lacked the ability to analyze the coordinated changes of multiple parameters, making it difficult to identify complex anomalies; in addition, each time a new product specification was added or production conditions were changed, the thresholds needed to be manually readjusted, resulting in high maintenance costs.

[0027] To address this issue, this invention proposes an intelligent process parameter alarm scheme based on multi-dimensional dynamic matching, applicable to the non-ferrous metal rolling processing industry. This scheme achieves intelligent monitoring and anomaly warning for key process parameters such as rolling force and temperature by constructing a configurable process parameter matching rule library. Compared to traditional fixed threshold alarm methods, this scheme innovatively introduces a multi-dimensional condition matching mechanism, which can dynamically adjust monitoring standards based on variables such as alloy type, seasonal environment, and incoming material specifications in actual production.

[0028] First, this invention discloses a process parameter anomaly alarm method 100, the core of which lies in achieving precise monitoring and intelligent alarm of process parameters through a dynamic matching mechanism driven by multi-dimensional production conditions. For example... Figure 1 As shown, the anomaly alarm method of the present invention may include steps S110-S130: Step S110: Collect material information and production parameter data in real time during equipment operation.

[0029] Process parameter alarms require a crucial data foundation: the production plan. Before production begins for each material, a plan must be developed. This plan includes not only basic attributes of the finished product such as material specifications and alloy composition, but also links to process parameters in the "Process Knowledge Base." A "process parameter identifier" establishes the connection between production materials and process parameter data, and production personnel must operate according to these parameters. During production plan execution, the system monitors production record data from the MES in real time. Upon receiving a production start event, it retrieves the corresponding "process parameter identifier" from the production plan based on the material number. Using this identifier, it locates the corresponding process parameters and standards for the current material in the "Process Knowledge Base," determining the upper and lower limits for each parameter alarm. Based on this, the system acquires various process parameters during equipment production in real time, determines whether an alarm is violated based on the alarm standards specific to the current material, generates an alarm message, and pushes it to the front-end page, reminding the user to correct production operations or preparing an early warning plan for the next production execution. Details will be provided in subsequent steps.

[0030] The production parameters may include at least one of the following: equipment type, seasonal parameters, etc., specifically including equipment operation number, seasonal identifier (spring / summer / autumn / winter, or accurate to the specific month). Material information includes at least one of the following: alloy type, incoming material specifications, finished product specifications, etc., specifically including alloy grade (e.g., A3003, C1100), incoming material width and thickness range, and finished product target width and thickness range. To ensure consistency in technical descriptions, in this document, "incoming material specifications" and "finished product specifications" refer to the dimensions of the raw material before entering the rolling process and the target dimensions of the final product, respectively, with ranges quantified in millimeters (mm), such as incoming material thickness: 1.0–1.5mm, finished product width: 800–1200mm.

[0031] Taking an aluminum foil rolling production process parameter alarm system as an example, the implementation of this invention is completed through the following steps: First, a sensor network needs to be deployed on various equipment in the production site. These sensors are used to collect production parameter data and key process parameters such as rolling force, temperature, liquid level, and current as target parameters for subsequent matching. Then, the production parameter and process parameter data collected by the sensor network can be transmitted to the data platform server through an industrial internet gateway and stored in the database to ensure the real-time performance and reliability of data collection.

[0032] Furthermore, the process parameter alarm system can also synchronize material information and production result data from the MES (Manufacturing Execution System), including information such as the equipment where the production process occurred, alloy, incoming material specifications, and product specifications, for subsequent process parameter alarm rule matching. Specifically, it can monitor newly added production task records synchronously with the MES. Whenever a new record is generated, the system will automatically search the process knowledge base, using conditions such as alloy, season, incoming material width / thickness, and finished product width / thickness to match conditions in the knowledge base, thereby finding the corresponding standard values ​​and upper and lower limits of process parameters.

[0033] Step S120: The collected data is sent to the process parameter matching rule base. Through the matching calculation of the rule base, the dynamic alarm range of the corresponding equipment target parameter is generated.

[0034] The process parameter matching rule base in this step is the core engine for realizing intelligent process monitoring. Its construction process can be completed through a three-step closed loop of "condition definition - standard binding - dynamic activation", thereby transforming static process experience into reusable and adaptive digital knowledge assets. Among them, the target parameters, as mentioned above, can include at least one of the following: rolling force, temperature, liquid level, current, etc. More parameters can be introduced according to production needs, which will not be elaborated here.

[0035] The process parameter matching rule base in step S120 can be constructed based on two steps: matching condition configuration and process parameter standard configuration. Specifically, it can include the following steps S121-S123: Step S121: Configure matching conditions for the device based on multiple production parameters associated with the device.

[0036] First, you can configure process parameter matching conditions. You can create structured matching condition groups in the rule library on a per-equipment basis, adding matching conditions such as alloy (e.g., A3003, C1100), season (spring / summer / autumn / winter), incoming material thickness range, incoming material width range, finished product thickness range, and finished product width range to the equipment.

[0037] Additionally, users can choose whether to enable certain conditions based on actual circumstances, creating a set of customized matching criteria. For example... Figure 2 As shown, the rule base supports enabling or disabling each dimension independently. For example, when producing thin aluminum foil, the "finished product thickness" matching item can be turned off, leaving only the triple constraints of "alloy + season + incoming material thickness" to form a precise production scenario profile. All configurations are stored in the database as structured data, ensuring that the rules are traceable and reusable.

[0038] Step S122: Configure the standard threshold for each matching condition, including the configuration condition threshold, the parameter definition of the process parameter to be alarmed and the corresponding process parameter alarm reference value.

[0039] Next, configure the condition values, alarm process parameters, and process parameter alarm settings for the matching conditions. The matching condition values ​​are used to determine the standards that the process parameters must meet during the production process for each production task record, based on basic information such as product alloy and specifications. Alarm process parameters are displayed as names, and the system backend maps these names to fields in the database. Specifically, based on each set of matching conditions, the corresponding process parameter alarm standards are configured.

[0040] Reference Figures 3a-3b As shown, the rule base allows binding multiple target parameters (such as temperature, rolling force, current, and liquid level) to each condition group, and defines three core values ​​for each parameter: target value (ideal operating point), lower limit, and upper limit, which together constitute the dynamic alarm range. For example, for the combination of "winter + A3003 + incoming material thickness 1.2mm", the target temperature value can be set to 128℃, and the upper and lower limits are 120℃~135℃. The alarm parameter name (such as "heating furnace temperature") is displayed in a user-friendly manner on the front-end interface, and the back-end automatically maps it to the sensor data field (such as sensor_temp_01) to achieve semantic consistency between humans and machines. All standard values ​​support version management to ensure that process changes are traceable and rollbackable.

[0041] Step S123: Based on the real-time activation status of the matching conditions, dynamically generate the dynamic alarm threshold range of the target parameters.

[0042] As mentioned earlier, the rule base supports flexible configuration of the activation status of each matching item. The dynamic alarm range includes a set value and at least one set of upper and lower limits to identify the normal range of each process parameter value. When the dynamic alarm range includes multiple sets of upper and lower limits, abnormal alarms can include multiple alarm levels. That is, alarms adopt a hierarchical mechanism, classifying alarms into three levels—yellow, orange, and red—from minor to severe based on production process knowledge (manually configured). When the system pushes alarm messages, severe alarms will be automatically specified and highlighted with a background color.

[0043] Ultimately, for each processing equipment, multi-dimensional matching items can be set, including conditions such as alloy grade, seasonal parameters, incoming material width / thickness range, and finished product width / thickness range. The system also supports flexible configuration of the activation status of each matching item. These matching conditions are stored in a structured form in the system database. For each set of matching conditions, the matching value for each matching item, as well as the target values ​​and upper and lower limits of the process parameters requiring alarms, are configured, forming a reusable process knowledge base.

[0044] As can be seen, all the above matching conditions and parameter standards together constitute a reusable process knowledge base, becoming the "intelligent decision-making brain." This knowledge base not only supports real-time alarms, but also serves as the underlying data asset for process optimization, personnel training, and digital twin modeling, achieving a fundamental shift from "experience-driven" to "data-driven" approaches.

[0045] Step S130: Based on the collected data of the target parameters and the dynamic alarm range, determine whether to issue an abnormal alarm.

[0046] This step involves monitoring the fluctuations of target parameters (process parameters), recording the collected data for each production task, and monitoring whether it conforms to the matched process parameter standards to identify and issue alarms for abnormal process parameters. Alarms can be selected in two forms: single-value alarms and indicator group alarms. In single-value mode, an alarm is triggered when a single process parameter exceeds the normal range; in indicator group mode, an alarm is generated when several process parameters configured as an indicator group are simultaneously abnormal.

[0047] Specifically, this step may include the following steps S131 or S132: Step S131: When the target parameter is a single parameter, execute a single-parameter independent alarm for that target parameter.

[0048] Step S132: When there are multiple target parameters, execute a multi-parameter linkage combination alarm based on the parameter combination relationship of the multiple target parameters.

[0049] The single-parameter independent alarm includes: 1) If the real-time collected data of the target parameter exceeds the dynamic alarm threshold range, an abnormal alarm is directly triggered; or 2) If the real-time collected data of the target parameter exceeds the range of the dynamic alarm threshold, a parameter over-limit violation is recorded. When the number of violations reaches the first set threshold or the proportion of violations within the set time period reaches the first preset proportion threshold, an abnormal alarm is triggered.

[0050] In other words, when monitoring a single process parameter (such as temperature or rolling force), this invention supports two alarm strategies: one is an immediate over-limit alarm, which triggers an alarm immediately once the real-time data exceeds the set upper and lower limit thresholds; the other is a cumulative violation alarm, which requires recording each over-limit event. An alarm is only triggered when the number of violations reaches a preset first threshold (such as 5 consecutive times) or the proportion of violations within a specified time period (such as 10 minutes) exceeds a first preset proportion (such as 30%), effectively filtering out instantaneous fluctuation interference.

[0051] The multi-parameter linkage combined alarm includes: 1) For each target parameter in the parameter combination, compare its real-time collected data with its corresponding dynamic alarm threshold range, and filter out abnormal target parameters that exceed their dynamic alarm range; and 2) If the number of abnormal target parameters reaches the second set threshold or the proportion in the parameter combination reaches the second preset proportion threshold, an abnormal alarm is triggered.

[0052] In other words, when multiple parameter combinations are involved (such as temperature + pressure + current), this invention activates a multi-parameter linkage alarm mechanism: First, it independently determines whether each parameter in the combination exceeds its dynamic threshold, filtering out abnormal parameters; then, if the number of abnormal parameters reaches a second set threshold (such as 2 out of 3) or the proportion of abnormal parameters in the total number of combinations exceeds a second preset threshold (such as 60%), it is determined to be a systemic process risk, triggering a high-level alarm. This mechanism significantly improves the ability to identify complex anomalies.

[0053] Furthermore, to ensure the scientific validity of combined alarms, for process parameter groups requiring combined alarms, the target parameters within the parameter combination must undergo validity verification to prove whether the multiple process parameters selected by the user warrant combined alarms. For example, parameters whose individual alarms can reflect a certain abnormal situation in production, without needing to be combined with additional parameters as evidence of the abnormality, are considered to have no need for combined alarms.

[0054] One approach to validity verification is to employ a time-series correlation verification method. Specifically, this method uses historical data of the selected process parameters as the verification dataset. That is, it calculates the Pearson correlation coefficient between the parameters using historical data. If the verification reveals that two process parameters are unrelated—for example, if their fluctuations show no significant correlation (correlation coefficient below 0.3)—the user is advised that the selected combination of process parameters may not require a combined alarm. For instance, a message might be displayed stating, "This combination has no synergistic anomaly characteristics; it is recommended to cancel the combined alarm," thus preventing ineffective configuration from causing alarm fatigue. This verification mechanism upgrades the combination rules from "empirical assumptions" to "data-driven" approaches, ensuring the reliability and interpretability of the alarm logic.

[0055] In summary, this invention constructs an intelligent process knowledge base and a dynamic alarm mechanism, which is divided into matching condition maintenance and process parameter alarm condition maintenance. This enables dynamic maintenance of process parameter thresholds, reduces the false alarm rate caused by changes in influencing factors such as seasons and specifications, and significantly improves the convenience of process parameter alarm configuration. It fundamentally reconstructs the process monitoring mode in the field of non-ferrous metal rolling processing.

[0056] Specifically, the beneficial effects of this invention may include: 1. Traditional solutions use fixed thresholds, which cannot distinguish between normal process drift caused by material properties or environmental fluctuations and genuine anomalies. Existing technologies often configure a static threshold for each process parameter, failing to respond to variations in conditions such as different alloys, seasons, and incoming material specifications, thus generating false alarms. This invention, however, uses a condition-driven threshold mapping mechanism, enabling the system to automatically select the optimal alarm interval under different production conditions, significantly reducing the false alarm rate caused by non-fault factors such as seasonal temperature differences and batch composition fluctuations, and improving alarm accuracy.

[0057] 2. This invention overcomes the limitations of independent monitoring of single parameters, supporting collaborative monitoring based on multi-variable joint alarm rules. In some practical situations, anomalies in a single process parameter cannot reflect the actual abnormal conditions in the production site. It is often necessary to monitor multiple process parameters simultaneously. An anomaly is considered to have occurred only when a set of monitored process parameters simultaneously meets the abnormal conditions. However, existing technologies often cannot support combined alarms. To address this, this invention can support the system triggering a composite abnormal event when a set of predefined process parameters (such as temperature, pressure, and flow rate) simultaneously meet preset logical combination conditions (such as AND / OR relationships). This accurately identifies complex coupled process faults and improves the detection coverage of latent anomalies.

[0058] In another embodiment, the present invention also provides a process parameter abnormality alarm system 200, such as... Figure 4 As shown, the alarm system 200 may include: a production parameter acquisition device 210, used to collect material information and production parameter data during the operation of production equipment in real time; a rule matching processing device 220, used to send the collected data to the process parameter matching rule library, and generate a dynamic alarm range for the corresponding equipment target parameter through the matching operation of the rule library; and an anomaly determination alarm device 230, used to compare and determine whether an anomaly alarm is triggered based on the collected data of the target parameter and the dynamic alarm range.

[0059] In another embodiment, the present invention also provides a process parameter analysis method, which includes: outputting abnormal alarm records through the process parameter abnormal alarm method described above; and performing multi-dimensional statistical analysis on the abnormal alarm records to identify characteristic abnormal patterns.

[0060] This method aims to systematically identify high-frequency anomaly patterns from massive alarm data, providing data-driven decision-making support for process parameter optimization and production stability improvement. The method uses structured alarm records output by an existing intelligent alarm system as input, and achieves multi-dimensional, in-depth pattern mining through an integrated big data analytics platform.

[0061] For example, an analysis platform based on big data technology can be built to support in-depth mining of historical alarm data by multiple dimensions such as date, process, indicator, and equipment, identify high-frequency anomaly patterns, and provide data support for process optimization. Furthermore, this invention can also comprehensively analyze alarm conditions for process parameters, statistically calculating all alarm records that meet the criteria such as time period, process, and equipment.

[0062] like Figures 5a-5d As shown, four statistical results are presented respectively: 1. For example Figure 5a As shown, the daily alarm quantity trend chart for each process is displayed, which summarizes the alarm records generated each day and for each process by date. The alarm quantity trend for each process is shown by a line. 2. For example Figure 5b As shown, the alarm distribution of each process is presented in the form of a pie chart, which shows the percentage of alarms generated by each process in the statistical data. 3. For example Figure 5c As shown in the figure, the alarm indicator distribution is presented in the form of a bar chart, which compares and shows the number of alarms generated by each indicator. 4. For example Figure 5d As shown, the distribution of alarm devices is displayed in a bar chart, which compares and shows the number of alarm records generated by each device.

[0063] The beneficial effects of the process parameter anomaly alarm system and process parameter analysis method provided in this application can be referred to the above description of the process parameter anomaly alarm method, and will not be repeated here.

[0064] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0065] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for alarming abnormal process parameters, characterized in that, The alarm method includes: Real-time acquisition of material information and production parameter data during equipment operation; The collected data is sent to the process parameter matching rule base. Through the matching calculation of the rule base, the dynamic alarm range of the target parameters of the corresponding equipment is generated. Based on the collected data of the target parameters and the dynamic alarm range, a comparison is made to determine whether an abnormal alarm is triggered.

2. The process parameter abnormality alarm method according to claim 1, characterized in that, The process parameter matching rule base is constructed based on the following method: Based on the multiple production parameters associated with the device, complete the configuration of matching conditions for the device; Configure the standard threshold for each matching condition, including the configuration condition threshold, the process parameter to be alarmed, and the corresponding process parameter alarm baseline value. Based on the real-time activation status of the matching conditions, the dynamic alarm threshold range of the target parameters is dynamically generated.

3. The method for alarming abnormal process parameters according to claim 1, characterized in that, The production parameters include at least one of the following: equipment type, seasonal parameters, The material information includes at least one of the following: alloy type, incoming material specifications, and finished product specifications. The target parameters include at least one of the following: rolling force, temperature, liquid level, and current.

4. The process parameter abnormality alarm method according to claim 3, characterized in that, The step of comparing and determining whether an abnormal alarm is triggered based on the collected data of the target parameters and the dynamic alarm range includes: When the target parameter is a single parameter, execute a single-parameter independent alarm for that target parameter; or When there are multiple target parameters, a multi-parameter linkage alarm is executed based on the parameter combination relationship of the multiple target parameters.

5. The process parameter abnormality alarm method according to claim 4, characterized in that, The single-parameter independent alarm includes: If the real-time collected data of the target parameter exceeds the dynamic alarm threshold range, an abnormal alarm is directly triggered; or If the real-time collected data of the target parameter exceeds the range of the dynamic alarm threshold, a parameter over-limit violation is recorded. When the number of violations reaches the first set threshold or the proportion of violations within a set time period reaches the first preset proportion threshold, an abnormal alarm is triggered.

6. The process parameter abnormality alarm method according to claim 4, characterized in that, The multi-parameter linkage combined alarm includes: For each target parameter in the parameter combination, its real-time collected data is compared with its corresponding dynamic alarm threshold range to filter out abnormal target parameters that exceed its dynamic alarm range; and An anomaly alarm is triggered when the number of abnormal target parameters reaches a second set threshold or the proportion of the parameter combination reaches a second preset proportion threshold.

7. The process parameter abnormality alarm method according to claim 4 or 6, characterized in that, The target parameters in the parameter combination satisfy the validity verification. The validity verification is performed by employing a time-series correlation verification method.

8. The method for alarming abnormal process parameters according to claim 1, characterized in that, The dynamic alarm range includes a set value and at least one set of upper and lower limits. Wherein, when the dynamic alarm range includes multiple sets of upper and lower limits, the abnormal alarm includes multiple alarm levels.

9. A method for analyzing process parameters, characterized in that, The analytical method includes: The abnormal alarm method for process parameters as described in any one of claims 1-8 outputs an abnormal alarm record; and Multi-dimensional statistical analysis is performed on the abnormal alarm records to identify characteristic abnormal patterns.

10. A process parameter abnormality alarm system, characterized in that, The alarm system includes: Production parameter acquisition device, used to collect material information and production parameter data in real time during the operation of production equipment; The rule matching processing device sends the collected data to the process parameter matching rule base, and generates a dynamic alarm range for the corresponding equipment target parameters through matching calculations in the rule base; and An anomaly detection and alarm device is used to compare and determine whether an anomaly alarm is triggered based on the collected data of the target parameters and the dynamic alarm range.

Images