Alarm rule evaluation and optimization method, system, electronic device, and storage medium

Abstract

The application provides an alarm rule evaluation and optimization method and system, electronic equipment and a storage medium. The method comprises the following steps: obtaining original alarm rules and a plurality of sets of historical strip steel production data; if a pre-warning index parameter in any set of historical strip steel production data meets a constraint condition, it is recorded as an un-pre-warning event; if the pre-warning index parameter in any set of historical strip steel production data does not meet the constraint condition, it is recorded as a pre-warning event; the first event quantity of the un-pre-warning event and the second event quantity of the pre-warning event are counted; if a quality attribute parameter corresponding to the un-pre-warning event indicates that a target defect does not occur, it is recorded as a quality normal event; if the quality attribute parameter corresponding to the pre-warning event indicates that the target defect occurs, it is recorded as a quality abnormal event; the third event quantity of the quality normal event and the fourth event quantity of the quality abnormal event are counted to complete rule evaluation and optimization; and the method better realizes quantitative evaluation and optimization of the original alarm rules.

Description

Technical Field

[0001] This application relates to the field of process rule evaluation technology, and in particular to an alarm rule evaluation and optimization method, system, electronic device and storage medium. Background Technology

[0002] In traditional process manufacturing industries such as steel, chemicals, and building materials, the stability, safety, and consistency of production processes and product quality highly depend on the real-time monitoring of massive amounts of process parameters (the technological parameters involved in product production). Therefore, most factories deploy automated systems such as product data acquisition and monitoring systems, with alarm functionality being a crucial feature. The alarm function monitors key process parameters in real time, promptly alerting relevant personnel when these parameters deviate from normal conditions to facilitate intervention, thereby preventing production anomalies and ensuring stable product quality. The core logic of this alarm function is based on "alarm rules." For example, when a key process parameter (such as the deviation of the pinch roll gap) is detected as not meeting preset constraints (such as deviating from a preset alarm threshold), the system triggers an alarm. Therefore, the rationality of these constraints directly determines the effectiveness and reliability of the alarm function.

[0003] However, in current industrial practice, the setting of these constraints typically relies on human experience, which has significant drawbacks, including: 1. The threshold settings in the constraints are subjective, lacking unified evaluation standards, making it difficult to quantify and optimize alarm rules. 2. In pursuit of "safety and reliability," a common practice is to set the thresholds in the constraints too strictly. This leads to a large number of deviations within the normal fluctuation range that do not have alarm significance triggering alarms, causing "alarm fatigue." 3. Setting the threshold range in the constraints too broadly may adversely affect production. Summary of the Invention

[0004] This application provides an alarm rule evaluation and optimization method, system, electronic device, and storage medium to solve the technical problem that it is difficult to quantitatively evaluate and optimize alarm rules in the prior art.

[0005] This application provides a method for evaluating and optimizing alarm rules, the method comprising: The system acquires original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether the strip steel has a target defect, as well as early warning index parameters for the target defect. The original alarm rules include triggering an alarm if the early warning index parameters do not meet preset constraints. If any set of historical strip steel production data contains a warning indicator parameter that meets the constraint condition, it is recorded as a non-warning event. If any set of historical strip steel production data contains a warning indicator parameter that does not meet the constraint condition, it is recorded as a warning event. The number of first events of non-warning events and the number of second events of warning events are counted. If the quality attribute parameter corresponding to the non-warning event indicates that the target defect has not appeared, it is recorded as a normal quality event. If the quality attribute parameter corresponding to the warning event indicates that the target defect has appeared, it is recorded as a quality abnormal event. The number of third events of the normal quality events and the number of fourth events of the quality abnormal events are counted. The original alarm rules are evaluated and optimized based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event.

[0006] In one embodiment of this application, evaluating and optimizing the original alarm rule based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event includes: The ratio of the number of the third event to the number of the first event is determined as the first target parameter; The ratio between the number of the fourth event and the number of the second event is determined as the second target parameter; Based on the first target parameter and the second target parameter, the original alarm rule is evaluated and optimized.

[0007] In one embodiment of this application, the evaluation of the original alarm rule based on the first target parameter and the second target parameter includes: Based on the first target parameter and the second target parameter, the original alarm rule is scored to obtain the target score of the original alarm rule. The target score is positively correlated with the first target parameter and positively correlated with the second target parameter.

[0008] In one embodiment of this application, scoring the original alarm rule based on the first target parameter and the second target parameter to obtain a target score for the original alarm rule includes: The sum of the first target parameter and the second target parameter is determined as the target score, or... The product between the first target parameter and the preset first weight is determined as the first product value, the product between the second target parameter and the preset second weight is determined as the second product value, and the sum of the first product value and the second product value is determined as the target score.

[0009] In one embodiment of this application, optimizing the original alarm rule based on the first target parameter and the second target parameter includes: The threshold parameter in the constraint is modified at least once to obtain the modified constraint. The rule containing the modified constraint is the new alarm rule, wherein each modification corresponds to a new alarm rule. Based on the new alarm rules, the number of non-alert events, the number of alert events, the number of normal quality events, and the number of abnormal quality events are re-statistically analyzed to obtain a new target score for the new alarm rules. The original alarm rule is optimized based on the target score and at least one of the new target scores.

[0010] In one embodiment of this application, optimizing the original alarm rule based on the target score and at least one new target score includes: If the number of new target scores is one and the new target score is greater than the target score, then the alarm rule optimization is completed according to the new alarm rule corresponding to the new target score. If there are multiple new target scores, and these new target scores are different, the maximum value among the multiple new target scores is determined as the score to be compared. If the score to be compared is greater than the target score, the new alarm rule or new alarm rule set corresponding to the score to be compared is determined as the target solution. The new alarm rule set includes multiple new alarm rules with the same score. Based on the target solution, the alarm rule optimization is completed.

[0011] In one embodiment of this application, optimizing the original alarm rule based on the first target parameter and the second target parameter includes: If the first target parameter of any new alarm rule is greater than the first target parameter of the original alarm rule, and the second target parameter of the new alarm rule is greater than the second target parameter of the original alarm rule, then the current new alarm rule is determined to be the Pareto improved solution of the original alarm rule; or, if any of the new target scores is greater than the target score, then the new alarm rule corresponding to the current new target score is determined to be the Pareto improved solution of the original alarm rule. Based on multiple Pareto improved solutions, the Pareto optimal solution is obtained; Based on the Pareto optimal solution, the original alarm rules are optimized.

[0012] This application also provides an alarm rule evaluation and optimization system, the system comprising: The data acquisition module is used to acquire the original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether the strip steel has a target defect, as well as early warning index parameters for the target defect. The original alarm rules include triggering an alarm if the early warning index parameters do not meet preset constraints. The first statistical module is used to record a non-warning event if any set of historical strip steel production data meets the constraint conditions, and a warning event if any set of historical strip steel production data does not meet the constraint conditions. The module also counts the number of first non-warning events and the number of second warning events. The second statistics module is used to record a normal quality event if the quality attribute parameter corresponding to the non-warning event indicates that the target defect has not appeared, and to record an abnormal quality event if the quality attribute parameter corresponding to the warning event indicates that the target defect has appeared. The module is used to count the number of third events of the normal quality events and the number of fourth events of the abnormal quality events. The evaluation and optimization module is used to evaluate and optimize the original alarm rules based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event.

[0013] This application also provides an electronic device, including a processor, a memory, and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the alarm rule evaluation and optimization method as described in any of the above claims.

[0014] This application also provides a computer-readable storage medium having a computer program stored thereon, the computer program being used to cause a computer to execute the alarm rule evaluation and optimization method as described in any of the preceding claims.

[0015] The beneficial effects of this application are as follows: This application proposes an alarm rule evaluation and optimization method, system, electronic device, and storage medium. The method includes: acquiring original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether a target defect has occurred in the strip steel, and early warning index parameters for the target defect. The original alarm rules include: if the early warning index parameters do not meet preset constraints, an alarm is triggered; if the early warning index parameters in any set of historical strip steel production data meet the constraints, it is recorded as a non-alarm event; if the early warning index parameters in any set of historical strip steel production data do not meet the constraints, it is recorded as an early warning event; the number of first non-alarm events and the number of second early warning events are counted; if the quality attribute parameters corresponding to a non-alarm event indicate that no target defect has occurred, it is recorded as a normal quality event; if the quality attribute parameters corresponding to an early warning event indicate that a target defect has occurred, it is recorded as a quality abnormal event; the number of third normal quality events and the number of fourth quality abnormal events are counted; and the original alarm rules are evaluated and optimized based on the number of first, second, third, and fourth events. This method uses the number of first events, second events, third events, and fourth events as evaluation indicators for the original alarm rules, thereby achieving a quantitative evaluation of the original alarm rules. This method is highly reliable and helps to optimize the original alarm rules in the future. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] In the attached diagram: Figure 1 This is a flowchart illustrating the alarm rule evaluation and optimization method provided in one embodiment of this application; Figure 2 This is a flowchart illustrating the process of evaluating and optimizing the original alarm rules based on the number of events in an alarm rule evaluation and optimization method provided in one embodiment of this application. Figure 3 This is a schematic diagram of the process for optimizing the original alarm rules in the alarm rule evaluation and optimization method provided in one embodiment of this application. Figure 1 ; Figure 4 This is a schematic diagram of the process for optimizing the original alarm rules in the alarm rule evaluation and optimization method provided in one embodiment of this application. Figure 2 ; Figure 5 This is an example diagram of the Pareto front in the alarm rule evaluation and optimization method provided in one embodiment of this application; Figure 6 This is a schematic diagram of the structure of an alarm rule evaluation and optimization system provided in one embodiment of this application; Figure 7 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0018] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0019] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0020] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0021] Please see Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the alarm rule evaluation and optimization method provided in this application, as shown below. Figure 1 As shown, the method includes: S110: Obtain the original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether the strip steel has a target defect, as well as early warning index parameters for the target defect. The original alarm rules include an alarm if the early warning index parameters do not meet preset constraints.

[0022] In some examples of this embodiment, the alarm rule evaluation and optimization method described herein can be applied to the evaluation and optimization of alarm rules in the strip steel production process. The target defect can be a tower-like defect (a process defect in which the head of the strip steel exhibits unevenness and progressively higher coils during the coiling process, forming a pagoda-like layered structure, which is a type of tower-shaped defect). The warning indicator parameter for the target defect refers to the product parameter associated with the target defect. For example, when this warning indicator parameter exceeds a preset value, the product is more likely to have the target defect. This warning indicator parameter can be, for example, the deviation of the pinch roll gap in the strip steel.

[0023] In the actual production process of strip steel, early warning indicators are often monitored to prevent target defects from occurring. For example, if the deviation of the pinch roll gap of the strip steel is greater than the preset value, an alarm is issued to prompt the staff to intervene and prevent defects such as head stacking of the strip steel.

[0024] In some examples of this embodiment, the constraints can be set according to the actual situation, such as the warning indicator parameter needing to be greater than a preset threshold parameter. This threshold parameter is the target for subsequent optimization.

[0025] In some examples of this embodiment, the number of historical strip steel production data can be set according to the actual situation, such as 1000 sets.

[0026] S120: If any set of historical strip steel production data contains a warning indicator parameter that meets the constraint condition, it is recorded as a non-warning event. If any set of historical strip steel production data contains a warning indicator parameter that does not meet the constraint condition, it is recorded as a warning event. The number of first events of non-warning events and the number of second events of warning events are counted.

[0027] S130: If the quality attribute parameter corresponding to the non-warning event indicates that the target defect has not appeared, it is recorded as a normal quality event. If the quality attribute parameter corresponding to the warning event indicates that the target defect has appeared, it is recorded as a quality abnormal event. The number of third events of the normal quality events and the number of fourth events of the quality abnormal events are counted.

[0028] Specifically, in this embodiment, events that conform to the original alarm rules, i.e., the early warning indicator parameters meet the constraints, are determined as non-alarm events; events that do not conform to the original alarm rules, i.e., the early warning indicator parameters do not meet the constraints, are determined as early warning events; events that conform to the optimization goal, i.e., the quality attribute parameters indicate that no target defect has appeared, are determined as normal quality events; and events that do not conform to the optimization goal, i.e., the quality attribute parameters indicate that a target defect has appeared, are determined as abnormal quality events.

[0029] By employing steps S120 and S130, the evaluation metrics for the original alarm rules are effectively quantified. Specifically, the number of first, second, third, and fourth events are used as evaluation metrics for the original alarm rules, facilitating subsequent evaluation based on these metrics. Furthermore, obtaining the aforementioned number of first, second, third, and fourth events allows for subsequent optimization of the original alarm rules with high accuracy.

[0030] S140: Evaluate and optimize the original alarm rules based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event.

[0031] It is understandable that the alarm rule evaluation and optimization method in this embodiment provides a unified evaluation standard for the original alarm rules, realizes the quantitative evaluation and optimization of the original alarm rules, and has strong feasibility.

[0032] After identifying and statistically analyzing unannounced events, announced events, normal quality events, and abnormal quality events, the following conditional probability model can be obtained based on probability theory: (Equation 1) (Equation 2) in, This refers to the probability of a normal quality event occurring among unannounced events. This refers to the probability of a quality anomaly occurring among unannounced events. This refers to the probability of a normal quality event occurring within a warning event. This refers to the probability of a quality anomaly occurring in an unannounced event.

[0033] Taking 1000 coils of strip steel produced historically by a steel mill as an example, assuming 1000 sets of historical strip steel production data are obtained, with each coil corresponding to one set of historical strip steel production data, among these, the warning indicator parameters (assuming the pinch roll gap deviation) in 920 sets of historical strip steel production data meet the constraint condition (e.g., less than or equal to 0.8 mm), while the warning indicator parameters in 80 sets of historical strip steel production data do not meet the constraint condition. Of these 920 sets of historical strip steel production data, the quality attribute parameters in 95 sets indicate the presence of a target defect (assuming a head tower defect), while the quality attribute parameters in 825 sets indicate the absence of a target defect. Of the aforementioned 80 sets of historical strip steel production data, the quality attribute parameters in 51 sets indicate the presence of a target defect, while the quality attribute parameters in 29 sets indicate the absence of a target defect. Based on this, the following probability results can be obtained: (Equation 3) (Equation 4) (Equation 5) (Equation 6) Based on the above probability results, it can be concluded that if the alarm rules can effectively influence the optimization objective (i.e., the headless tower defect), then... and It should be as big as possible, and correspondingly and It should be as small as possible.

[0034] Therefore, this embodiment is based on and The original alarm rules are evaluated and optimized. Please refer to [the relevant documentation / reference]. Figure 2 In some embodiments, evaluating and optimizing the original alarm rules based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event includes: S1401: The ratio between the number of the third event and the number of the first event is determined as the first target parameter.

[0035] S1402: The ratio between the number of the fourth event and the number of the second event is determined as the second target parameter.

[0036] S1403: Based on the first target parameter and the second target parameter, complete the evaluation and optimization of the original alarm rule.

[0037] Understandably, the first target parameter is... The second objective parameter is The evaluation of the original alarm rules based on the first and second target parameters demonstrates high reliability. Optimizing the original alarm rules based on these parameters improves the accuracy of rule optimization, enabling precise optimization of the threshold parameters within the original alarm rules.

[0038] In some embodiments, the evaluation of the original alarm rule based on the first target parameter and the second target parameter includes: Based on the first target parameter and the second target parameter, the original alarm rule is scored to obtain the target score of the original alarm rule. The target score is positively correlated with the first target parameter and positively correlated with the second target parameter.

[0039] Understandably, scoring the original alarm rules using the above method yields high accuracy.

[0040] In some embodiments, scoring the original alarm rule based on the first target parameter and the second target parameter to obtain a target score for the original alarm rule includes: The sum of the first target parameter and the second target parameter is determined as the target score, or... The product between the first target parameter and the preset first weight is determined as the first product value, the product between the second target parameter and the preset second weight is determined as the second product value, and the sum of the first product value and the second product value is determined as the target score.

[0041] Understandably, by adopting the above method, the scoring of the original alarm rules can be completed relatively well.

[0042] Please refer to Figure 3 In some embodiments, optimizing the original alarm rule based on the first target parameter and the second target parameter includes: S310: Modify the threshold parameter in the constraint at least once to obtain the modified constraint. The rule containing the modified constraint is the new alarm rule, wherein each modification corresponds to a new alarm rule.

[0043] S320: Based on the new alarm rule, re-count the number of non-warning events, the number of warning events, the number of normal quality events, and the number of abnormal quality events to obtain a new target score for the new alarm rule.

[0044] S330: Based on the target score and at least one of the new target scores, optimize the original alarm rule.

[0045] In some examples of this embodiment, if the new target score is greater than the target score of the original alarm rule, it can be determined that the new alarm rule is superior to the original alarm rule. Therefore, the new alarm rule can be recommended to the user based on this.

[0046] Understandably, by adopting the above methods, the original alarm rules can be optimized more effectively.

[0047] In some embodiments, optimizing the original alarm rule based on the target score and at least one of the new target scores includes: S3301: When the number of new target scores is one and the new target score is greater than the target score, the alarm rule optimization is completed according to the new alarm rule corresponding to the new target score.

[0048] S3302: When there are multiple new target scores and the multiple new target scores are different, the maximum value among the multiple new target scores is determined as the score to be compared. If the score to be compared is greater than the target score, the new alarm rule or new alarm rule set corresponding to the score to be compared is determined as the target solution. The new alarm rule set includes multiple new alarm rules with the same score. Based on the target solution, the alarm rule optimization is completed.

[0049] In some examples of this embodiment, if the score to be compared is less than or equal to the target score of the original alarm rule, the corresponding new alarm rule can be determined to be an invalid rule and the invalid rule can be deleted.

[0050] In some examples of this embodiment, the target solution can be recommended to the user to complete the optimization of alarm rules.

[0051] Understandably, by adopting the above methods, the original alarm rules can be optimized quite well.

[0052] Please refer to Figure 4 In some embodiments, optimizing the original alarm rule based on the first target parameter and the second target parameter includes: S410: If the first target parameter of any new alarm rule is... The first target parameter is greater than the original alarm rule. And the second target parameter of the new alarm rule The second target parameter is greater than the original alarm rule. If the new alarm rule is determined to be the Pareto improved solution of the original alarm rule, then if any of the new target scores is greater than the target score, then the new alarm rule corresponding to the new target score is determined to be the Pareto improved solution of the original alarm rule.

[0053] Taking 1000 coils of strip steel produced historically by a steel mill as an example, assuming the constraint condition is changed from less than or equal to 0.8 mm to less than or equal to 1 mm, a new alarm rule is obtained. Based on this new alarm rule, it can be statistically determined that the warning indicator parameter (pinch roll gap deviation) in 940 sets of historical strip steel production data meets the new constraint condition (less than or equal to 1 mm), while the warning indicator parameter in 60 sets of historical strip steel production data does not meet the new constraint condition. Of these 940 sets of historical strip steel production data, the quality attribute parameters in 96 sets indicate the presence of a target defect (assuming it is a head tower defect), while the quality attribute parameters in 844 sets indicate the absence of a target defect. Of the aforementioned 60 sets of historical strip steel production data, the quality attribute parameters in 50 sets indicate the presence of a target defect, while the quality attribute parameters in 10 sets indicate the absence of a target defect. Based on this, the following probability results can be obtained: (Equation 7) (Equation 8) (Equation 9) (Equation 10) Comparing Equation 7 and Equation 3, and simultaneously comparing Equation 10 and Equation 6, we can obtain ,and Therefore, the new alarm rule that includes the new constraint (less than or equal to 1 mm) is superior to the original alarm rule and is a Pareto improvement solution of the original alarm rule.

[0054] S420: Based on multiple Pareto improved solutions, obtain the Pareto optimal solution.

[0055] In some examples of this embodiment, a Pareto optimal solution refers to a solution to which no possible Pareto improvement exists. In this case, any effort to improve one objective will lead to a decrease in at least one other objective. Among all possible solutions, the Pareto optimal solution is not necessarily unique; there may also exist a solution space, i.e., a set of Pareto optimal solutions. The set of all Pareto optimal solutions in the solution space is called the Pareto front. Figure 5 For an example diagram of the Pareto front in the alarm rule evaluation and optimization method provided in one embodiment of this application, please refer to... Figure 5 , Figure 5 R1, R2, R3, and R4 in the solution space are the different Pareto optimal solutions. That is .

[0056] S430: Based on the Pareto optimal solution, optimize the original alarm rule.

[0057] In some examples of this embodiment, Pareto optimal solutions can be recommended to the user, who only needs to select the most suitable parameters based on the actual situation.

[0058] The alarm rule evaluation and optimization method in the above embodiments provides a new paradigm for objective and transparent alarm rule decision-making. It transforms the previously ambiguous judgment of whether an alarm rule is "good" or "bad" into precise calculations based on conditional probability, shifting the evaluation process from subjective experience to objective data. Simultaneously, the introduction of Pareto improvement provides a clear and rigorous logical standard for "whether a new rule should be adopted," effectively avoiding ambiguity and controversy in the decision-making process. Furthermore, the alarm rule evaluation and optimization method in the above embodiments is not only applicable to alarm rule evaluation and optimization in strip steel production but also to other industrial production scenarios, demonstrating strong adaptability. In addition, the alarm rule evaluation and optimization method in the above embodiments provides a scientific link for efficient collaboration between equipment, processes, and production in traditional process industries. By adopting this method, production sites can flexibly formulate more scientific alarm rules according to different critical stages, helping to improve product quality and process management levels, and achieving high-quality scientific development.

[0059] Please refer to Figure 6 This embodiment also provides an alarm rule evaluation and optimization system, the system comprising: The data acquisition module 610 is used to acquire the original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether the strip steel has a target defect, as well as early warning index parameters for the target defect. The original alarm rules include that if the early warning index parameters do not meet the preset constraints, an alarm will be triggered. The first statistical module 620 is used to record a non-warning event if any set of the historical strip steel production data meets the constraint conditions, and a warning event if any set of the historical strip steel production data does not meet the constraint conditions. The module also records the number of first events of non-warning events and the number of second events of warning events. The second statistics module 630 is used to record a normal quality event if the quality attribute parameter corresponding to the non-warning event indicates that the target defect has not appeared, and to record an abnormal quality event if the quality attribute parameter corresponding to the warning event indicates that the target defect has appeared, and to count the number of third events of the normal quality events and the number of fourth events of the abnormal quality events. The evaluation and optimization module 640 is used to evaluate and optimize the original alarm rules based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event.

[0060] It should be noted that the alarm rule evaluation and optimization method and the alarm rule evaluation and optimization system provided in the above embodiments belong to the same concept. The specific methods of execution of each module have been described in detail in the method embodiments and will not be repeated here. In practical applications, the alarm rule evaluation and optimization system provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0061] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call logic instructions in the memory 730 to execute an alarm rule evaluation and optimization method. This method includes: acquiring original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters characterizing whether a target defect has occurred in the strip steel, and early warning index parameters for the target defect. The original alarm rules include: if the early warning index parameters do not meet preset constraints, an alarm is triggered; if the early warning index parameters in any set of historical strip steel production data meet the constraints, it is recorded as a non-alarm event; if the early warning index parameters in any set of historical strip steel production data do not meet the constraints, it is recorded as an early warning event; the number of first non-alarm events and the number of second early warning events are counted; if the quality attribute parameters corresponding to a non-alarm event indicate that no target defect has occurred, it is recorded as a normal quality event; if the quality attribute parameters corresponding to an early warning event indicate that a target defect has occurred, it is recorded as a quality abnormal event; the number of third normal quality events and the number of fourth quality abnormal events are counted; and the original alarm rules are evaluated and optimized based on the number of first, second, third, and fourth events.

[0062] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0063] Furthermore, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, this computer program is implemented to perform the alarm rule evaluation and optimization methods provided by the aforementioned methods. This method includes: acquiring original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters characterizing whether a target defect has occurred in the strip steel, and early warning index parameters for the target defect. The original alarm rules include triggering an alarm if the early warning index parameters do not meet preset constraints; if any set of historical strip steel production data contains early warning index parameters that meet the constraints, it is recorded as a non-alarm event. If any set of historical strip steel production data fails to meet the constraints of the early warning indicator parameters, it is recorded as an early warning event. The number of first events (no early warning events) and the number of second events (early warning events) are counted. If the quality attribute parameter corresponding to a non-early warning event indicates that the target defect has not appeared, it is recorded as a normal quality event. If the quality attribute parameter corresponding to an early warning event indicates that the target defect has appeared, it is recorded as a quality abnormal event. The number of third events (normal quality events) and the number of fourth events (abnormal quality events) are counted. Based on the number of first, second, third, and fourth events, the original alarm rules are evaluated and optimized.

[0064] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0065] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0066] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for evaluating and optimizing alarm rules, characterized in that, The method includes: The system acquires original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether the strip steel has a target defect, as well as early warning index parameters for the target defect. The original alarm rules include triggering an alarm if the early warning index parameters do not meet preset constraints. If any set of historical strip steel production data contains a warning indicator parameter that meets the constraint condition, it is recorded as a non-warning event. If any set of historical strip steel production data contains a warning indicator parameter that does not meet the constraint condition, it is recorded as a warning event. The number of first events of non-warning events and the number of second events of warning events are counted. If the quality attribute parameter corresponding to the non-warning event indicates that the target defect has not appeared, it is recorded as a normal quality event. If the quality attribute parameter corresponding to the warning event indicates that the target defect has appeared, it is recorded as a quality abnormal event. The number of third events of the normal quality events and the number of fourth events of the quality abnormal events are counted. The original alarm rules are evaluated and optimized based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event.

2. The alarm rule evaluation and optimization method according to claim 1, characterized in that, The step of evaluating and optimizing the original alarm rules based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event includes: The ratio of the number of the third event to the number of the first event is determined as the first target parameter; The ratio between the number of the fourth event and the number of the second event is determined as the second target parameter; Based on the first target parameter and the second target parameter, the original alarm rule is evaluated and optimized.

3. The alarm rule evaluation and optimization method according to claim 2, characterized in that, The evaluation of the original alarm rule based on the first target parameter and the second target parameter includes: Based on the first target parameter and the second target parameter, the original alarm rule is scored to obtain the target score of the original alarm rule. The target score is positively correlated with the first target parameter and positively correlated with the second target parameter.

4. The alarm rule evaluation and optimization method according to claim 3, characterized in that, The step of scoring the original alarm rule based on the first target parameter and the second target parameter to obtain the target score of the original alarm rule includes: The sum of the first target parameter and the second target parameter is determined as the target score, or... The product between the first target parameter and the preset first weight is determined as the first product value, the product between the second target parameter and the preset second weight is determined as the second product value, and the sum of the first product value and the second product value is determined as the target score.

5. The alarm rule evaluation and optimization method according to claim 3, characterized in that, The optimization of the original alarm rule based on the first target parameter and the second target parameter includes: The threshold parameter in the constraint is modified at least once to obtain the modified constraint. The rule containing the modified constraint is the new alarm rule, wherein each modification corresponds to a new alarm rule. Based on the new alarm rules, the number of non-alert events, the number of alert events, the number of normal quality events, and the number of abnormal quality events are re-statistically analyzed to obtain a new target score for the new alarm rules. The original alarm rule is optimized based on the target score and at least one of the new target scores.

6. The alarm rule evaluation and optimization method according to claim 5, characterized in that, The optimization of the original alarm rule based on the target score and at least one new target score includes: If the number of new target scores is one and the new target score is greater than the target score, then the alarm rule optimization is completed according to the new alarm rule corresponding to the new target score. If there are multiple new target scores, and these new target scores are different, the maximum value among the multiple new target scores is determined as the score to be compared. If the score to be compared is greater than the target score, the new alarm rule or new alarm rule set corresponding to the score to be compared is determined as the target solution. The new alarm rule set includes multiple new alarm rules with the same score. Based on the target solution, the alarm rule optimization is completed.

7. The alarm rule evaluation and optimization method according to claim 5, characterized in that, The optimization of the original alarm rule based on the first target parameter and the second target parameter includes: If the first target parameter of any new alarm rule is greater than the first target parameter of the original alarm rule, and the second target parameter of the new alarm rule is greater than the second target parameter of the original alarm rule, then the current new alarm rule is determined to be the Pareto improved solution of the original alarm rule; or, if any of the new target scores is greater than the target score, then the new alarm rule corresponding to the current new target score is determined to be the Pareto improved solution of the original alarm rule. Based on multiple Pareto improved solutions, the Pareto optimal solution is obtained; Based on the Pareto optimal solution, the original alarm rules are optimized.

8. An alarm rule evaluation and optimization system, characterized in that, The system includes: The data acquisition module is used to acquire the original alarm rules and multiple sets of historical strip steel production data. The historical strip steel production data includes quality attribute parameters used to characterize whether the strip steel has a target defect, as well as early warning index parameters for the target defect. The original alarm rules include triggering an alarm if the early warning index parameters do not meet preset constraints. The first statistical module is used to record a non-warning event if any set of historical strip steel production data meets the constraint conditions, and a warning event if any set of historical strip steel production data does not meet the constraint conditions. The module also counts the number of first non-warning events and the number of second warning events. The second statistics module is used to record a normal quality event if the quality attribute parameter corresponding to the non-warning event indicates that the target defect has not appeared, and to record an abnormal quality event if the quality attribute parameter corresponding to the warning event indicates that the target defect has appeared. The module is used to count the number of third events of the normal quality events and the number of fourth events of the abnormal quality events. The evaluation and optimization module is used to evaluate and optimize the original alarm rules based on the number of the first event, the number of the second event, the number of the third event, and the number of the fourth event.

9. An electronic device, characterized in that, It includes a processor, a memory, and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the alarm rule evaluation and optimization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that enables the computer to perform the alarm rule evaluation and optimization method as described in any one of claims 1 to 7.

Images