SCADA flexible alarm configuration method and system based on multi-dimensional mapping

Abstract

The invention discloses an SCADA flexible alarm configuration method and system based on multi-dimensional mapping, and relates to the technical field of alarm configuration, and the method comprises the steps: S1, collecting industrial alarm mapping data in real time, and carrying out the preprocessing of the collected data; s2, constructing a product name and specification mapping table, evaluating the consistency of working conditions in the current production unit, and performing optimization verification on the product name and specification mapping table; s3, an alarm parameter loading index is established, an alarm parameter template of the current production unit is retrieved, an original alarm threshold value is corrected, and an alarm parameter set is generated; and S4, constructing an alarm rule candidate set, performing conflict degree evaluation on the alarm candidate rules, eliminating conflict rules to obtain an alarm rule set, and writing the alarm rule set into an SCADA alarm rule configuration interface. The problems that alarm rule affiliation and parameter loading are often misplaced and rule version tracing is difficult due to the fact that a unit level lacks specification-based consistency constraint in existing alarm configuration are solved.

Description

Technical Field

[0001] This invention relates to the field of alarm configuration technology, specifically to a SCADA flexible alarm configuration method and system based on multi-dimensional mapping. Background Technology

[0002] With the advancement of digital transformation in industrial enterprises, SCADA systems are widely used in various continuous and discrete manufacturing scenarios for process monitoring, equipment management, and operational safety assurance. Together with information system integration services, they form a crucial support system for both the enterprise's field and management levels. Industrial fields contain a variety of key operating parameters, which vary significantly depending on product specifications and operating conditions. To ensure safe and stable production, it is typically necessary to configure corresponding alarm thresholds and rules. However, in complex and ever-changing production environments, fixed thresholds and static rules often fail to meet the alarm requirements of multiple scenarios, products, and conditions, resulting in a large workload for alarm configuration, insufficient adaptability, and high maintenance costs.

[0003] For example, the invention with publication number CN116386293A provides an alarm device and alarm system, which relates to the technical field of computer applications. The alarm device includes: a controller, and a network module and a monitoring module connected to the controller; the network module is communicatively connected to a data center; the monitoring module includes at least one data monitoring unit for collecting monitoring data of the environment in which the alarm device is located; the controller is used to generate alarm information when it determines that the monitoring data meets the pre-configured alarm conditions, and sends it to the data center through the network module.

[0004] For example, the invention disclosed in CN113191677B provides a dynamically configurable on-board battery alarm method. This method establishes an alarm configuration table for the on-board battery and compares and matches the collected real-time default parameters with the alarm configuration table to generate alarm commands of corresponding levels. This method can configure corresponding alarm rules for different parameter indicators of the on-board battery, compare and match alarm rules, and complete the collection of alarm data for different types and levels of batteries based on these rules. It automatically provides alarm information to battery users, preventing accidents caused by battery failures, fundamentally ensuring the safety and reliability of the battery, and keeping the battery in its healthiest state. This extends the battery's lifespan, improves vehicle energy management, enhances vehicle comfort, significantly reduces operating and maintenance costs, and ultimately increases the economic value of the battery.

[0005] Although the above technical solutions can generate alarm information and dynamically match alarm rules based on environmental monitoring data or equipment parameters, the existing technologies generally have the following limitations: (1) The setting of alarm thresholds depends on manual configuration or fixed parameter tables, and lacks the ability to adapt to differences in production unit operating conditions and changes in product specifications; (2) Most methods can only compare real-time parameters with preset conditions, without considering parameter fluctuation patterns, evaluation window behavior and multi-dimensional mapping relationships, and cannot support flexible alarm adjustment across products; (3) Existing methods usually generate rules for a single device or a single data source, which makes it difficult to achieve overall coordination of multiple types of industrial alarm mapping data, and easily leads to conflicts, overlaps or redundancies between alarm rules.

[0006] Therefore, in order to address the above issues, there is an urgent need for a flexible alarm configuration method and system for SCADA based on multi-dimensional mapping. Summary of the Invention

[0007] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a SCADA flexible alarm configuration method and system based on multi-dimensional mapping. This solves the problem that the lack of specification-based consistency constraints at the unit level in existing alarm configurations often leads to misalignment of alarm rule attribution and parameter loading, and causes difficulties in rule version tracing.

[0008] Technical solution To achieve the above objectives, this invention provides the following technical solution: a SCADA flexible alarm configuration method based on multi-dimensional mapping, comprising: S1, real-time acquisition of industrial alarm mapping data, and performing time synchronization, noise suppression, anomaly removal, missing data imputation, and normalization processing on the acquired data to obtain preprocessed industrial alarm mapping data; S2, constructing a product name and specification mapping table based on the preprocessed industrial alarm mapping data, establishing a window to evaluate the consistency of operating conditions within the current production unit, determining whether the mapping segmentation process is triggered, and optimizing and verifying the product name and specification mapping table; S3, establishing an alarm parameter loading index based on the product name and specification mapping table, retrieving alarm parameter templates for the current production unit, correcting the original alarm thresholds in the templates, generating an alarm parameter set, and performing continuous legality verification; S4, constructing an alarm rule candidate set based on the alarm parameter loading index, evaluating the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, removing conflicting rules to obtain an alarm rule set, writing the alarm rule set into the SCADA alarm rule configuration interface, and archiving the successfully written alarm rules.

[0009] Furthermore, the industrial alarm mapping data is collected in real time, and the collected data undergoes time synchronization, noise suppression, anomaly removal, missing data imputation, and normalization processing to obtain preprocessed industrial alarm mapping data. The specific steps are as follows: Real-time collection of industrial alarm mapping data, including material feed temperature, material discharge temperature, material temperature inside the tank, absolute pressure inside the tank, pipeline inlet pressure, pipeline outlet pressure, instantaneous material flow rate, liquid level, and pressure difference before and after the filter; Time synchronization and timing alignment are achieved using a unified clock source for the collected industrial alarm mapping data, and noise suppression and abrupt change smoothing are performed using a Kalman filter algorithm; Outlier detection is performed using a Grubbs anomaly detection algorithm to filter out false anomalies caused by electromagnetic interference and equipment start-up / shutdown transients; For missing data caused by sampling intervals and transmission delays, a linear interpolation algorithm is used for data completion and sequence continuity; The industrial alarm mapping data is normalized using a range normalization algorithm to unify the numerical scale and eliminate dimensional differences.

[0010] Furthermore, the specific steps for constructing a product name and specification mapping table based on the preprocessed industrial alarm mapping data are as follows: taking the smallest production unit as the organizational boundary, classify and map the preprocessed industrial alarm mapping data according to the production unit, and establish a unique production unit identifier code for the data stream within the same production unit; read and parse the product name identifier point and the product specification identifier point within each smallest production unit, and structurally bind them with the corresponding production unit identifier code to construct a product name and specification mapping table.

[0011] Furthermore, the specific steps for establishing a window for evaluating the consistency of operating conditions within the current production unit are as follows: Design a fixed time step as an evaluation window; calculate the mean of each industrial alarm mapping data item in the previous and current evaluation windows for each production unit, and extract the maximum and minimum values ​​of each industrial alarm mapping data item in the current evaluation window; take the absolute value of the difference between the mean of the i-th industrial alarm mapping data item in the current evaluation window and the corresponding mean in the previous evaluation window, divide it by the sum of the differences between the maximum and minimum values ​​of the i-th industrial alarm mapping data item in the current evaluation window, and add the minimum term to obtain the normalized difference term; square the normalized difference terms of all industrial alarm mapping data items, sum them, and divide by the number of industrial alarm mapping data categories to obtain the window difference mean square term; take the square root of the window difference mean square term and multiply it by the control sensitivity coefficient to obtain the sensitivity modulation term; apply a hyperbolic tangent function to the sensitivity modulation term to obtain the switching fluctuation evaluation value.

[0012] Further, the specific steps for determining whether to trigger the mapping segmentation process and optimizing the product name and specification mapping table are as follows: Real-time comparison of the switching fluctuation assessment value and the fluctuation threshold. When the switching fluctuation assessment value is less than the fluctuation threshold, it is determined that the current operating condition within the production unit is stable, maintaining the existing binding relationship between the product name code and the product specification code, and the assessment window continues to slide backward at a fixed time step. Otherwise, it is determined that the operating condition within the production unit is fluctuating, triggering the mapping segmentation process: closing the current assessment window and creating a new assessment window, while updating the corresponding record in the product name and specification mapping table. The record content includes the production unit identifier code, product name code, product specification code, assessment window number, window start and end time, and switching fluctuation assessment value, and marking the previous assessment window status as completed. For adjacent assessment windows within the same production unit, consistency verification is performed on the product name code and product specification code. When there is an overlap in time intervals, the product name code and product specification code are overwritten and retained according to the sampling timestamp order, and the overwritten record is marked as conflict resolved.

[0013] Furthermore, based on the product name and specification mapping table, an alarm parameter loading index is established to retrieve the alarm parameter template for the current production unit. The specific steps for correcting the original alarm thresholds in the template are as follows: Based on the real-time product name and specification mapping table, the product name code and product specification code information corresponding to the current production unit are read, and an alarm parameter loading index is established; using the combination of product name code, product specification code, and production unit identifier code as a unique index key, the corresponding alarm parameter template is retrieved from the alarm parameter template library, and the alarm parameter template is loaded into the memory instance as the initial parameter set for the current production unit; the original alarm thresholds of each industrial alarm mapping data in the loaded alarm parameter template are extracted, and the alarm thresholds of each item in the current production unit are corrected: the normalized difference term of the i-th industrial alarm mapping data in the current evaluation window is calculated, and the normalized difference term is corrected according to... The amplitude modulation amount is obtained by exponentially multiplying the alarm response coefficient by the sum of the switching fluctuation assessment value and the constant 1. The negative of the amplitude modulation amount is used as the exponent, and an exponential operation is applied to the natural constant e. The result of the exponential operation is then subtracted from the constant 1 to obtain the threshold scaling term. The difference between the maximum and minimum values ​​of the i-th industrial alarm mapping data in the current assessment window is multiplied by the threshold scaling term and added to the original alarm threshold of the i-th industrial alarm mapping data in the template to obtain the alarm threshold correction value of the i-th industrial alarm mapping data. The alarm threshold correction value of each industrial alarm mapping data is calculated, and legality verification and boundary constraints are performed. When threshold overlap and out-of-bounds are detected, the historical upper and lower limits of the corresponding industrial alarm mapping data and the alarm parameter template definition rules are extracted to perform a second correction on the alarm threshold correction value. The corrected alarm threshold correction value is used as the final alarm threshold.

[0014] Furthermore, the specific steps for generating an alarm parameter set and performing continuous validity verification are as follows: Extract all final alarm thresholds to form the alarm parameter set for the current production unit; during the effective period of the alarm parameter set, when any change is detected in the product name code, product specification code, or production unit identifier code, terminate the use of the current alarm parameter set, and summarize the current alarm parameter set, the corresponding alarm parameter loading index, and the evaluation window interval information to build a historical parameter archive; based on the latest product name code and product specification code, re-trigger the parameter loading process to generate a new alarm parameter set, and output it to the alarm rule configuration unit in the form of a message structure, combining the corresponding alarm parameter loading index, alarm threshold correction record, and historical parameter archive identifier.

[0015] Furthermore, based on the alarm parameter loading index, an alarm rule candidate set is constructed. Conflict assessment is performed on multiple alarm candidate rules corresponding to the same industrial alarm mapping data, and conflicting rules are eliminated to obtain the alarm rule set. The specific steps are as follows: After receiving the message structure, the alarm rule configuration unit parses the alarm rule basic template of the current production unit using the alarm parameter loading index as the unique matching key to form an alarm rule candidate set, and generates a rule identifier code and version number for each rule. In the alarm rule candidate set, multiple alarm candidate rules belonging to the same industrial alarm mapping data are sorted according to the alarm threshold size, and the upper and lower safety limits of the i-th type of industrial alarm mapping data in the alarm parameter template are extracted, and the difference between the two is calculated. Let be the alarm interval span; square the difference in alarm thresholds between two adjacent alarm candidate rules in the i-th type of industrial alarm mapping data to obtain the threshold difference squared term; multiply the square of the alarm interval span of the i-th type of industrial alarm mapping data by the number of alarm candidate rules corresponding to the i-th type of industrial alarm mapping data in the current production unit minus one, and add a minimum term to obtain the normalized denominator term; subtract the ratio of the threshold difference squared term to the normalized denominator term from the constant one to obtain the alarm rule conflict degree evaluation value of the j-th alarm candidate rule; compare the alarm rule conflict degree evaluation value and the rule conflict threshold in real time, and remove all alarm candidate rules in the alarm rule candidate set whose alarm rule conflict degree evaluation value is greater than the rule conflict threshold to obtain a complete alarm rule set.

[0016] Furthermore, the specific steps for writing alarm rule sets into the SCADA alarm rule configuration interface and archiving successfully written alarm rules are as follows: A rule publishing task queue is constructed based on the production unit identifier and rule identifier. Alarm rules are written into the SCADA alarm rule configuration interface according to the order of the rule publishing task queue. After the interface returns a successful response, the corresponding rule identifier is recorded as the real-time alarm rule version of the current production unit. After the alarm rule is published, the corresponding alarm rule identifier, alarm threshold, version timestamp, and rule publishing task queue processing result are extracted and summarized into the historical rule archive to form an alarm rule version record. The current rule version number is then sent back to the alarm rule configuration unit in the form of a message structure for traceable management of alarm rules.

[0017] The second solution of this invention provides a SCADA flexible alarm configuration system based on multi-dimensional mapping, including: a data acquisition and preprocessing module, a product name and specification identification and binding module, an alarm parameter flexible mapping module, and an alarm rule configuration and publishing module. The data acquisition and preprocessing module is used to acquire industrial alarm mapping data in real time and perform time synchronization, noise suppression, anomaly removal, missing data imputation, and normalization processing on the acquired data to obtain preprocessed industrial alarm mapping data. The product name and specification identification and binding module is used to construct a product name and specification mapping table based on the preprocessed industrial alarm mapping data and establish a window to evaluate the consistency of the current production unit's operating conditions, determining whether... Whether the mapping segmentation process is triggered, the product name and specification mapping table is optimized and verified; the alarm parameter flexible mapping module is used to build an alarm parameter loading index based on the product name and specification mapping table, retrieve the alarm parameter template of the current production unit, perform correction on the original alarm threshold in the template, generate an alarm parameter set and perform continuous legality verification; the alarm rule configuration and release module is used to build an alarm rule candidate set based on the alarm parameter loading index, evaluate the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, eliminate conflicting rules to obtain the alarm rule set, write the alarm rule set into the SCADA alarm rule configuration interface, and archive the alarm rules that are successfully written.

[0018] Beneficial effects

[0019] The present invention has the following beneficial effects: (1) A flexible SCADA alarm configuration method and system based on multi-dimensional mapping. This method constructs a multi-dimensional mapping relationship using product name code, product specification code, and production unit identification code, and achieves accurate allocation of alarm parameters within the same production unit based on dynamic segmented evaluation windows for consistent operating conditions. This mechanism fundamentally eliminates the problem of alarm rule misalignment, improves the coupling between alarms and process status, and enhances the clarity of process traceability.

[0020] (2) A flexible alarm configuration method and system for SCADA based on multi-dimensional mapping uses mathematical models such as normalized difference terms, alarm response coefficients, exponential function modulation, and operating condition consistency coupling to dynamically correct template alarm thresholds, enabling the thresholds to automatically expand and contract with changes in operating conditions, thereby overcoming the limitations of traditional fixed threshold strategies. This method can maintain appropriate sensitivity across different process states, achieving a precise balance between avoiding over-triggering and missing critical alarms.

[0021] (3) A flexible SCADA alarm configuration method and system based on multi-dimensional mapping introduces a normalized conflict degree calculation formula based on the square of the threshold difference, the alarm interval span, and the number of rules to quantitatively analyze the differences between candidate alarm rules of the same type and automatically eliminate redundant and conflicting rules. This mechanism effectively avoids problems such as rule overlap and unclear logical mutual exclusion caused by experience configuration, realizes mathematical pruning and structural optimization of the rule set, and significantly improves the reliability of rule release.

[0022] (4) A flexible alarm configuration method and system for SCADA based on multi-dimensional mapping. By constructing a rule release task queue with alarm parameter loading index as the core, and combining release results, rule version numbers and archived records, the system realizes automatic writing, version tracking and historical backtracking of alarm rules. This mechanism provides the SCADA system with auditable, rollbackable and comparable rule management capabilities, solving prominent problems such as untrackable rule changes and chaotic status in existing systems. Attached Figure Description

[0023] Figure 1 A flowchart of a SCADA flexible alarm configuration method based on multi-dimensional mapping; Figure 2 A structural diagram of a SCADA flexible alarm configuration system based on multi-dimensional mapping; Figure 3 A flowchart for dynamically generating alarm parameters based on the mapping between product name and specifications; Figure 4 This is a screening diagram for candidate alarm rules based on the alarm rule conflict degree evaluation value. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figures 1-4This invention provides a technical solution: a SCADA flexible alarm configuration method based on multi-dimensional mapping, comprising: S1, real-time acquisition of industrial alarm mapping data, and performing time synchronization, noise suppression, anomaly removal, missing data imputation and normalization processing on the acquired data to obtain preprocessed industrial alarm mapping data; S2, constructing a product name and specification mapping table based on the preprocessed industrial alarm mapping data, establishing a window to evaluate the consistency of operating conditions within the current production unit, determining whether the mapping segmentation process is triggered, and optimizing and verifying the product name and specification mapping table; S3, establishing an alarm parameter loading index based on the product name and specification mapping table, retrieving the alarm parameter template of the current production unit, correcting the original alarm threshold in the template, generating an alarm parameter set and performing continuous legality verification; S4, constructing an alarm rule candidate set based on the alarm parameter loading index, evaluating the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, removing conflicting rules to obtain an alarm rule set, writing the alarm rule set into the SCADA alarm rule configuration interface, and archiving the successfully written alarm rules.

[0026] Specifically, the process involves real-time acquisition of industrial alarm mapping data, followed by time synchronization, noise suppression, anomaly removal, missing data imputation, and normalization to obtain preprocessed industrial alarm mapping data. The specific steps are as follows: Real-time acquisition of industrial alarm mapping data, including material feed temperature, material discharge temperature, material temperature inside the tank, absolute pressure inside the tank, pipeline inlet pressure, pipeline outlet pressure, instantaneous material flow rate, liquid level, and pressure difference before and after the filter. The material feed temperature is acquired by a platinum resistance temperature sensor installed at the inlet of the feed pipeline; the material discharge temperature is acquired by a sensor installed on the discharge pipe. Thermocouple temperature sensors at the outlet section are used to collect data; the material temperature inside the tank is collected by immersion temperature probes located at measuring points on the inner wall of the tank; the absolute pressure inside the tank is collected by an absolute pressure transmitter located at the top flange of the tank; the pipeline inlet pressure is collected by a pressure transmitter installed upstream of the inlet pipeline; the pipeline outlet pressure is collected by a pressure transmitter installed downstream of the outlet pipeline; the instantaneous flow rate of the material is collected by an electromagnetic flowmeter connected in series in the main process pipeline; the liquid level is collected by a static pressure level gauge installed on the side wall of the tank; and the pressure difference before and after the filter is collected by two differential pressure transmitters located before and after the filter. For the collected industrial alarm mapping data, time synchronization and timing alignment are achieved using a unified clock source. Kalman filtering is then used for noise suppression and abrupt smoothing. During the filtering process, the state covariance is updated based on the transient fluctuation distribution of each industrial alarm mapping data point, and the gain matrix is ​​dynamically adjusted to maintain steady-state consistency in the filtered output. Outlier detection is performed using the Grubbs anomaly detection algorithm to filter out false anomalies caused by electromagnetic interference and equipment start-up / shutdown transients. Multi-frame comparison is performed in conjunction with the continuity characteristics of alarm monitoring points, allowing isolated outliers to be eliminated during the detection phase. For sampling intervals and transmission delays... For missing data, a linear interpolation algorithm is used to complete the data and make the sequence continuous. During the interpolation process, the interpolation is refined according to the time interval between effective sampling and the data gradient characteristics to ensure that the interpolation results are consistent with the actual working conditions in terms of trend. The industrial alarm mapping data is normalized by the range normalization algorithm to unify the numerical scale and eliminate the difference in dimensions. At the same time, it ensures that the distribution range of material feed temperature, material discharge temperature, material temperature in tank, absolute pressure in tank, pipeline inlet pressure, pipeline outlet pressure, instantaneous material flow rate, liquid level height and pressure difference before and after the filter are comparable after normalization.

[0027] In this implementation plan, by uniformly collecting, timing-checking, noise suppression, outlier detection, missing data completion, and normalization of material feed temperature, material discharge temperature, tank / vessel temperature, tank absolute pressure, pipeline inlet pressure, pipeline outlet pressure, instantaneous material flow rate, liquid level, and pressure difference before and after the filter, all industrial alarm mapping data have a consistent time reference and stable numerical scale before entering subsequent processes. This ensures that the monitoring data meets the input requirements for alarm processing in terms of continuity, reliability, and comparability, reduces the risk of deviation caused by acquisition errors, improves the accuracy of alarm threshold determination, parameter correction, and rule generation, and provides a higher-quality data foundation for subsequent alarm configuration throughout the entire process.

[0028] Specifically, the steps for constructing a product name and specification mapping table based on preprocessed industrial alarm mapping data are as follows: Using the smallest production unit as the organizational boundary, the preprocessed industrial alarm mapping data is classified and mapped according to the production unit. The smallest production unit is determined based on the equipment boundaries of the process flow, the logic of process segment division, and the material transport path of the continuous production line, ensuring that the same production unit maintains structural independence in process execution and material closed-loop. During the classification process, consistency checks are performed on the timestamps and sampling stability of material inlet temperature, material outlet temperature, tank / vessel temperature, tank absolute pressure, pipeline inlet pressure, pipeline outlet pressure, instantaneous material flow rate, liquid level, and pressure difference before and after the filter, to ensure data continuity and traceability within the production unit. A unique production unit identifier is established for the data stream within the same production unit. This unique identifier uses a hierarchical coding structure composed of process segment number, equipment asset code, and production unit serial number. A unique mapping relationship is generated by combining the inherent attributes of the equipment on the production site with the process topology, thus ensuring that all industrial alarm mapping data have a clear source boundary within the same production unit. Within each smallest production unit, the product name identifier and product specification identifier are read and parsed. Both the product name identifier and the product specification identifier are string-based acquisition channels, provided by the upper-level process instruction channel. During the parsing process, the real-time value, sampling time, and validity status of the identifier are verified, and then structurally bound to the corresponding production unit identifier code to construct a product name and specification mapping table.

[0029] In this implementation plan, the preprocessed industrial alarm mapping data is classified and mapped according to the smallest production unit, so that each piece of industrial alarm mapping data has a clear production unit affiliation and product characteristic association. This significantly improves the matching accuracy of alarm-related data in spatial and business dimensions, enhances the uniqueness constraint effect of subsequent parameter loading and alarm rule configuration, avoids parameter mismatch caused by unclear data affiliation, and enhances the stability and traceability of the entire alarm configuration process.

[0030] Specifically, the steps for establishing a window to assess the consistency of operating conditions within the current production unit are as follows: Design a fixed time step as an assessment window, ensuring that the industrial alarm mapping data forms continuous segments in the time dimension, so as to periodically evaluate the operating status of the production unit; calculate the mean value of each industrial alarm mapping data item in the previous and current assessment windows for each production unit, so that this mean value can characterize the overall stability of material feed temperature, material discharge temperature, tank / reservoir material temperature, tank absolute pressure, pipeline inlet pressure, pipeline outlet pressure, instantaneous material flow rate, liquid level, and pressure difference before and after the filter over different time periods. The system calculates the level and extracts the maximum and minimum values ​​of each industrial alarm mapping data item within the current evaluation window, ensuring these values ​​reflect the fluctuation limits of the data within this window. It then takes the absolute value of the difference between the mean of the i-th industrial alarm mapping data item in the current evaluation window and the corresponding mean in the previous evaluation window. This absolute value measures the actual variation of each industrial alarm mapping data item between adjacent windows. This value is then divided by the sum of the differences between the maximum and minimum values ​​of the i-th industrial alarm mapping data item in the current evaluation window, plus a minimum term, to obtain a normalized difference term, which characterizes the variation of the i-th industrial alarm mapping data item within the current evaluation window. The relative degree of change between adjacent windows; where the minima are non-zero, extremely small positive real numbers, so that the denominator can reflect the local fluctuation scale of the data and avoid calculation instability under extremely small amplitude conditions; the normalized difference terms of all industrial alarm mapping data are squared and then summed; where the squaring operation strengthens the influence of larger difference terms, the summation value can comprehensively reflect the overall fluctuation intensity of each industrial alarm mapping data; the summation value is divided by the number of industrial alarm mapping data categories to keep the contribution of the averaging process to different data categories balanced, thus obtaining the window difference mean square term; the square root of the window difference mean square term is taken, so that... Its dimensional structure is restored to be consistent with the normalized difference term and the interpretability of the results is enhanced. It is then multiplied with the control sensitivity coefficient to obtain the sensitivity modulation term. The control sensitivity coefficient is obtained by fitting the change correlation between historical industrial alarm mapping data and historical switching fluctuation assessment values ​​through the least squares regression algorithm, and its value ranges from zero to one. The hyperbolic tangent function is applied to the sensitivity modulation term so that the hyperbolic tangent function can map the sensitivity modulation term to the [0,1] interval and enhance the ability to compress large fluctuations, thereby obtaining the switching fluctuation assessment value, which is used to comprehensively reflect the operating stability of the current production unit within the continuous window.

[0031] The specific formula for calculating the switching volatility assessment value is as follows: ; In the formula, This indicates a switch to the volatility assessment value. Indicates the control sensitivity coefficient. This represents the mean of the i-th industrial alarm mapping data within the current evaluation window. This represents the mean of the i-th industrial alarm mapping data within the previous evaluation window. This represents the maximum value of the i-th industrial alarm mapping data within the current evaluation window. This represents the minimum value of the i-th industrial alarm mapping data within the current evaluation window. Indicates minterms, This indicates the number of categories in the industrial alarm mapping data.

[0032] In this implementation scheme, by constructing continuous evaluation windows for various industrial alarm mapping data at a fixed time step, a multi-level analysis chain is formed, consisting of mean, maximum, minimum, normalized difference term, window difference mean square term, sensitivity modulation term, and switching fluctuation evaluation value. This allows changes in operating conditions to be quantified on a uniform scale, makes industrial alarm mapping data from different sources comparable in time series, and ensures a smooth transition in the stability of production unit operating status between continuous windows. This structured window evaluation method significantly improves the discriminative power of operating condition identification, transforming fluctuation trends from local features to global representations. This provides a reliable basis for subsequent alarm parameter correction and alarm rule selection, improving the accuracy and robustness of operating condition consistency judgment.

[0033] Specifically, the steps for determining whether to trigger the mapping segmentation process and optimizing the product name and specification mapping table are as follows: Real-time comparison of the switching fluctuation assessment value and fluctuation threshold. During the comparison, a high-precision timestamp is used as the sole time-series benchmark to ensure the continuity of the judgment process. When the switching fluctuation assessment value is less than the fluctuation threshold, the current operating condition within the production unit is determined to be stable. The existing binding relationship between the product name code and the product specification code remains unchanged, and the assessment window continues to slide forward at a fixed time step. Simultaneously, the start and end times of the assessment window are continuously recorded to ensure the closure of the window division. Otherwise, fluctuations in the operating condition within the production unit are determined, triggering the mapping segmentation process: closing the current assessment window and creating a new assessment window. During the creation process, a new assessment window is generated synchronously. Numbering is used to ensure the orderliness of the window sequence, while updating the corresponding records in the product name and specification mapping table. The record content includes the production unit identifier code, product name code, product specification code, evaluation window number, window start and end time, and switching fluctuation evaluation value. The status of the previous evaluation window is marked as completed to make it traceable. For product name codes and product specification codes of adjacent evaluation windows within the same production unit, consistency verification is performed. In the consistency verification, the sampling timestamp is used as the unique comparison benchmark. When there is an overlap in time intervals, the product name codes and product specification codes are overwritten and retained according to the order of sampling timestamps, and the overwritten records are marked as conflict resolved to ensure the uniqueness and continuity of the product name and specification mapping table.

[0034] In this implementation scheme, a precise time-driven decision-making mechanism is introduced to ensure a continuous and stable differentiation capability between product name codes and product specification codes within the production unit. By updating the product name and specification mapping table in real time when the fluctuation evaluation value changes and strictly maintaining the integrity of the evaluation window sequence, the delineation of specification switching boundaries becomes clearer. By implementing a sampling timestamp-based overwrite retention strategy for product name codes and product specification codes in adjacent evaluation windows, potential conflict records can be accurately identified and resolved. The overall effect is that the specification identification link at the production unit level maintains uniqueness and consistency, and the structure of the product name and specification mapping table remains clear and traceable in a dynamic production environment, laying a stable foundation for the accurate association of subsequent alarm parameter loading and alarm rule configuration.

[0035] Specifically, the steps for establishing an alarm parameter loading index based on the product name and specification mapping table, retrieving the alarm parameter template for the current production unit, and correcting the original alarm thresholds in the template are as follows: Figure 3As shown, firstly, based on the real-time product name and specification mapping table, the product name code and product specification code information corresponding to the current production unit are read to establish an alarm parameter loading index. During the reading process, the data integrity of the product name code and product specification code is simultaneously verified to ensure that both coding fields are consistent with the latest evaluation window record, so as to ensure that the alarm parameter loading index can accurately point to the target product in the actual production state. Using the combination of product name code, product specification code, and production unit identifier code as a unique index key, the corresponding alarm parameter template is retrieved from the alarm parameter template library, and the alarm parameter template is loaded into the memory instance as the initial parameter set of the current production unit. During the loading process, the field names of all industrial alarm mapping data are compared for consistency to avoid field drift caused by historical template versions. Its purpose is to ensure that each piece of industrial alarm mapping data has a matching value in the subsequent threshold calculation process. The basic field configuration is as follows: The original alarm thresholds of various industrial alarm mapping data in the loaded alarm parameter template are extracted, and the alarm thresholds of the current production unit are corrected to ensure that the correction action is always based on the real business template. The normalized difference term of the i-th industrial alarm mapping data in the current evaluation window is calculated. The normalized difference term is exponentially operated on according to the alarm response coefficient, and then multiplied by the sum of the switching fluctuation evaluation value and a constant to obtain the amplitude modulation amount. During the calculation process, the temporal consistency of the normalized difference term, the alarm response coefficient, and the switching fluctuation evaluation value is maintained, ensuring that the threshold correction source is consistent within the same evaluation window. This operation transforms the production status fluctuation into a quantifiable threshold adjustment intensity. The alarm response coefficient is obtained by fitting the gradient regression algorithm based on the sensitivity correlation between the normalized difference term of historical industrial alarm mapping data and historical alarm triggering behavior, with a value range of zero to one.The amplitude modulation quantity is inversely multiplied by an exponent. An exponential operation is then applied to the natural constant e, and the result is subtracted from the constant to obtain the threshold scaling term. This ensures that the value of the threshold scaling term always falls within the range of zero to one. This calculation suppresses over-adjustment through an exponential decay mechanism, keeping the alarm threshold scaling behavior smooth and controllable. The difference between the maximum and minimum values ​​of the i-th industrial alarm mapping data within the current evaluation window is multiplied by the threshold scaling term, and then added to the original alarm threshold of the i-th industrial alarm mapping data in the template to obtain the alarm threshold correction value for the i-th industrial alarm mapping data. During the calculation, the maximum and minimum values ​​are kept consistent with the boundaries of the current evaluation window. This operation serves to... The key is to map the fluctuation range within the window to a scalable threshold adjustment range, enabling dynamic adaptation of the alarm threshold. It calculates the alarm threshold correction value for each industrial alarm mapping data item and performs legality checks and boundary constraints. When threshold overlap or out-of-bounds is detected, it extracts the historical upper and lower limits of the corresponding industrial alarm mapping data and the alarm parameter template definition rules to perform a secondary correction on the alarm threshold correction value. The corrected alarm threshold correction value is used as the final alarm threshold. After the correction is completed, the triggering conditions and historical parameter range of the secondary correction process are recorded, ensuring that the source of the alarm threshold has complete traceability. This step ensures that the alarm threshold is continuous, interpretable, and always meets business boundary requirements within the historical range.

[0036] The specific formula for calculating the alarm threshold correction value is as follows: ; In the formula, This represents the alarm threshold correction value for the i-th industrial alarm mapping data. This represents the original alarm threshold of the i-th industrial alarm mapping data in the template. This represents the maximum value of the i-th industrial alarm mapping data within the current evaluation window. This represents the minimum value of the i-th industrial alarm mapping data within the current evaluation window. This represents the mean of the i-th industrial alarm mapping data within the current evaluation window. This represents the mean of the i-th industrial alarm mapping data within the previous evaluation window. This indicates a switch to the volatility assessment value. Indicates minterms, This represents the alarm response coefficient.

[0037] In this implementation scheme, the method forms a complete and continuous parameter adaptive link in each stage, including alarm parameter loading index, alarm parameter template retrieval, alarm threshold correction, threshold scaling calculation, threshold validity verification, and secondary correction. This enables the original alarm threshold to be robustly and dynamically adjusted under the real-time operating conditions of the production unit. By jointly indexing and managing product name codes, product specification codes, and production unit identification codes, the accuracy of alarm parameter retrieval is improved. The amplitude modulation quantity, driven by normalized difference terms, alarm response coefficients, and switching fluctuation evaluation values, achieves a sensitive response to operating condition fluctuations in threshold changes. Exponential threshold scaling control keeps the threshold adjustment process smooth and controlled. Boundary constraints and secondary correction mechanisms ensure that the alarm threshold of each industrial alarm mapping data remains stable and maintains business continuity after dynamic adjustment, thereby improving the overall matching accuracy and reliability between alarm thresholds and real-time operating conditions.

[0038] Specifically, the steps for generating an alarm parameter set and performing continuous validity checks are as follows: Extract all final alarm thresholds to form the alarm parameter set for the current production unit. During the extraction process, the final alarm thresholds of each industrial alarm mapping data are reorganized according to the field order to ensure that the field arrangement within the alarm parameter set is consistent with the alarm parameter template, thus providing a stable parameter structure for subsequent threshold calls. During the effective period of the alarm parameter set, if any change is detected in the product name code, product specification code, or production unit identification code, the current alarm parameter set is terminated. Before termination, an integrity check is performed on all fields of the current alarm parameter set and its effective time range to ensure the parameter set maintains a closed structure before exiting the effective state. Simultaneously, the current alarm parameter set and the corresponding alarm parameters are... The system loads the index and summarizes the evaluation window range information to build a historical parameter archive. During the archiving process, a unique archive identifier is generated for each alarm parameter record, ensuring that the parameter traceability has a verifiable and unique source. Based on the latest product name code and product specification code, the parameter loading process is re-triggered to generate a new set of alarm parameters. During the generation process, the correspondence between the latest product name code, the latest product specification code and the current production unit identifier is simultaneously verified to ensure that the three fields of the index key are completely matched. Combined with the corresponding alarm parameter loading index, alarm threshold correction record and historical parameter archive identifier, the data is output to the alarm rule configuration unit in the form of a message structure. Before output, the field order, data type and timestamp format of the message structure are processed for consistency to ensure that the output result can be stably parsed by subsequent configuration steps.

[0039] This implementation scheme ensures verifiable stability of the alarm parameter set at each stage of its lifecycle by linking the final alarm threshold, alarm parameter loading index, product name code, product specification code, and production unit identifier code throughout the entire process. Clear constraints on the exit, archiving, and reloading conditions of the alarm parameter set ensure a clear boundary for the alarm threshold's flow path. By constructing a unique archiving identifier for the historical parameter archive and maintaining consistency in the message structure output format, the source of the alarm thresholds is traceable. These combined effects enable the alarm parameter set to maintain high reliability in response to dynamic changes in the production unit, thereby improving the accuracy and controllability of alarm threshold management.

[0040] Specifically, the alarm rule candidate set is constructed based on the alarm parameter loading index. The conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data is evaluated, and conflicting rules are eliminated to obtain the alarm rule set. The specific steps are as follows: After receiving the message structure, the alarm rule configuration unit parses the alarm rule basic template of the current production unit using the alarm parameter loading index as the unique matching key to form the alarm rule candidate set. A rule identifier code and version number are generated for each rule. During the generation process, the uniqueness of the rule identifier code is simultaneously verified to ensure the alarm rule candidate set is distinguishable. In the alarm rule candidate set, multiple alarm candidate rules belonging to the same industrial alarm mapping data are sorted according to their alarm threshold values. The upper and lower safety limits of the i-th type of industrial alarm mapping data in the alarm parameter template are extracted, and the difference between them is calculated as the alarm interval span. Simultaneously, the alarm interval span is used as a benchmark to measure the magnitude of alarm threshold changes, ensuring that the subsequent normalization results have a unified dimension. The square of the alarm threshold difference between two adjacent alarm candidate rules in the i-th type of industrial alarm mapping data is obtained as the threshold difference square. The system employs a method to determine the conflict degree of an alarm rule. The squared threshold difference term is used as a sensitive measure of the closeness of separation between adjacent rules, making rule pairs with smaller differences more easily identified as potential conflict sources. A normalized denominator term is obtained by multiplying the square of the alarm interval span of the i-th type of industrial alarm mapping data by the number of alarm candidate rules corresponding to the i-th type of industrial alarm mapping data in the current production unit minus one, and then adding a minimum term. This normalized denominator term is used to suppress evaluation fluctuations caused by changes in the number of alarm candidate rules, ensuring comparability of conflict degree results under different rule numbers. The ratio of the squared threshold difference term to the normalized denominator term is subtracted from a constant to obtain the alarm rule conflict degree evaluation value of the j-th alarm candidate rule. This evaluation value is used as a quantitative indicator to measure the density of alarm thresholds, explicitly identifying rules with excessively close adjacent thresholds. The system compares the alarm rule conflict degree evaluation value with the rule conflict threshold in real time, eliminating all alarm candidate rules in the alarm rule candidate set whose evaluation value is greater than the rule conflict threshold. This ensures that only alarm candidate rules with reasonable threshold distribution and sufficient discrimination remain in the alarm rule set, resulting in a structurally complete alarm rule set.

[0041] The specific formula for calculating the alarm rule conflict degree evaluation value is as follows: ; In the formula, This represents the alarm rule conflict evaluation value of the j-th candidate alarm rule. This represents the number of alarm rules corresponding to the i-th type of industrial alarm mapping data in the current production unit. This represents the alarm threshold of the i-th type of industrial alarm mapping data in the (j+1)-th alarm candidate rule. This represents the alarm threshold of the i-th type of industrial alarm mapping data in the j-th alarm candidate rule. This represents the alarm interval span of the i-th industrial alarm mapping data in the alarm parameter template. Indicates a minus term.

[0042] In this embodiment, Table 1 is a data table of alarm rule conflict evaluation values, listing the alarm threshold elements and corresponding alarm rule conflict evaluation results for five alarm candidate rules under the same rule configuration conditions. The alarm threshold elements used in this embodiment include: the alarm threshold of adjacent alarm candidate rules, the alarm threshold of the next alarm candidate rule, the alarm interval span in the corresponding alarm parameter template, and the alarm rule conflict evaluation value obtained based on the calculation relationship. Specifically: Alarm candidate rule 1: The alarm threshold of this rule is 0.52, the alarm threshold of the next adjacent rule is 1.13, the alarm interval span is 1.0, and the final calculated alarm rule conflict evaluation value is 0.907. Alarm candidate rule 2: The alarm threshold of this rule is 0.67, the alarm threshold of the next adjacent rule is 1.15, the alarm interval span is 1.0, and the final calculated alarm rule conflict evaluation value is 0.942. Alarm candidate rule 3: The alarm threshold for this rule is 0.75, the alarm threshold for the next adjacent rule is 1.16, the alarm interval span is 1.0, and the final calculated alarm rule conflict evaluation value is 0.958. Alarm candidate rule 4: The alarm threshold for this rule is 0.99, the alarm threshold for the next adjacent rule is 1.17, the alarm interval span is 1.0, and the final calculated alarm rule conflict evaluation value is 0.992. Alarm candidate rule 5: The alarm threshold for this rule is 1.06, the alarm threshold for the next adjacent rule in this embodiment is 1.18 for completion calculation, the alarm interval span is 1.0, and the final alarm rule conflict evaluation value is 0.990.

[0043] Table 1. Alarm Rule Conflict Degree Evaluation Value Data Table

[0044] like Figure 4As shown in the figure, the conflict evaluation values ​​of five candidate alarm rules and the corresponding rule retention decisions are displayed. The points in the line graph are distinguished by different colors: green points indicate that the conflict evaluation value of the alarm rule does not exceed the conflict threshold and is retained as an alarm rule; red points indicate that the conflict evaluation value of the alarm rule exceeds the conflict threshold and is removed from the candidate alarm rules. The conflict threshold line used for rule filtering is marked with a blue dashed line in the figure. It can be seen from the figure that candidate alarm rules 1 and 2 are retained as alarm rules. Candidate alarm rules 3, 4, and 5 are removed from the candidate alarm rules. Figure 4 The system intuitively demonstrates the alarm rule conflict assessment mechanism. By quantitatively describing the threshold differences between adjacent rules, it enables automatic screening of alarm candidate rules, providing a reliable basis for generating a more concise and consistent alarm rule set in the future.

[0045] In this implementation scheme, continuous quantification processing—including alarm threshold sorting, alarm interval span calculation, threshold difference squared term calculation, normalized denominator construction, alarm rule conflict degree evaluation value generation, and rule conflict threshold comparison—completely quantifies the distribution characteristics of candidate alarm rules under the same industrial alarm mapping data, thereby achieving accurate identification of dense alarm threshold segments. This process explicitly removes potential rule congestion points from the alarm threshold distribution, ensuring that the remaining alarm rules have sufficient threshold spacing, thus guaranteeing higher trigger discrimination clarity in the alarm rule set during execution. Through this mechanism, the generated alarm rule set maintains stable threshold discrimination capability under the same industrial alarm mapping data conditions, improving the reliability of alarm triggering behavior, making subsequent alarm judgments more explicit, avoiding false alarms caused by multiple rules triggering simultaneously due to excessively close thresholds, and enhancing the overall controllability and consistency of alarm threshold configuration.

[0046] Specifically, the steps for writing alarm rule sets into the SCADA alarm rule configuration interface and archiving successfully written alarm rules are as follows: A rule publishing task queue is constructed based on the production unit identifier and rule identifier. During the construction process, the legality of the production unit identifier and the uniqueness of the rule identifier are synchronously verified to ensure that any record in the publishing task queue has unique mapping capability. A failure retry mechanism, a timeout judgment mechanism, and a rollback mechanism are set within the task queue. If a write failure, communication timeout, sequence abnormality, or rule identifier conflict occurs during the publishing process, the write action can be retried according to the retry count rule. When the maximum number of retries is exceeded, a task rollback is triggered, returning the task queue state to its initial, re-executable position, ensuring the stability of the rule publishing process. Alarm rules are written to the SCADA alarm rule configuration interface in the order of the task queue according to the rules. During the writing process, the alarm threshold field, alarm trigger condition field, and alarm level field of each alarm rule are compared for consistency to ensure that the data structure submitted to the SCADA alarm rule configuration interface remains intact. After the SCADA alarm rule configuration interface returns a successful response, the response type returned by the interface is distinguished as a synchronous ACK response and an asynchronous acknowledgment response. The rule is only confirmed to have been written successfully after both types of responses have arrived. Subsequently, the corresponding rule identifier code is recorded as the real-time alarm rule version of the current production unit, and the response timestamp of the SCADA alarm rule configuration interface is recorded synchronously. The response timestamp is generated based on a unified time zone and in a monotonically increasing manner, so that the version sequence has a strict time order. After an alarm rule is published, the corresponding alarm rule identifier, alarm threshold, version timestamp, and rule publication task queue processing results are extracted and summarized into the historical rule archive. During the writing process, the version range of the alarm threshold and the effective range of the rule identifier are checked to ensure that the archived data has strict temporal continuity. After the alarm rule version record is formed, the current rule version number is sent back to the alarm rule configuration unit in the form of a message structure for traceable management of alarm rules. A unique verification fingerprint of the version record is added during the return process to enable rapid verification of version matching in subsequent traceability.

[0047] This implementation scheme strengthens the data integrity, submission consistency, and version effectiveness accuracy of alarm rules during the release process by constructing a rule release task queue using production unit identifiers and rule identifiers, maintaining a consistent rule release order, and combining this with the synchronization of response timestamps from the SCADA alarm rule configuration interface with real-time rule identifier version records. This ensures that alarm rule sets have stable effectiveness after being written to the SCADA alarm rule configuration interface. Furthermore, a historical rule archive is constructed using alarm thresholds, version timestamps, rule release task queue processing results, and the continuity of the effective scope. Combined with the returned rule version number and corresponding unique verification fingerprint, this creates a clear timeline for alarm rule version records, ensuring that each rule update forms a verifiable source path. This enables precise traceability of alarm rules throughout the entire lifecycle of the production unit, providing a unified, reliable, and verifiable management capability for parameter evolution records in continuous production scenarios.

[0048] like Figure 2 As shown, the second solution of this invention provides a SCADA flexible alarm configuration system based on multi-dimensional mapping, including: a data acquisition and preprocessing module, a product name and specification identification and binding module, an alarm parameter flexible mapping module, and an alarm rule configuration and publishing module. The data acquisition and preprocessing module is used to acquire industrial alarm mapping data in real time and perform time synchronization, noise suppression, anomaly removal, missing data imputation, and normalization processing on the acquired data to obtain preprocessed industrial alarm mapping data. The product name and specification identification and binding module is used to construct a product name and specification mapping table based on the preprocessed industrial alarm mapping data and establish a window to evaluate the consistency of operating conditions within the current production unit, determining... Whether to trigger the mapping segmentation process, optimize and verify the product name and specification mapping table; the alarm parameter flexible mapping module is used to build an alarm parameter loading index based on the product name and specification mapping table, retrieve the alarm parameter template of the current production unit, correct the original alarm threshold in the template, generate an alarm parameter set and perform continuous legality verification; the alarm rule configuration and release module is used to build an alarm rule candidate set based on the alarm parameter loading index, evaluate the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, eliminate conflicting rules to obtain the alarm rule set, write the alarm rule set into the SCADA alarm rule configuration interface, and archive the alarm rules that are successfully written.

[0049] This implementation plan integrates four technical capabilities—data acquisition and preprocessing, product specification identification and binding, flexible alarm parameter mapping, and alarm rule configuration and publishing—into a single information system integration service. This creates a structured, verifiable, and traceable mapping link between industrial alarm mapping data, product name codes, product specification codes, alarm parameter loading indexes, alarm threshold correction values, and alarm rule candidate sets throughout the entire process. This achieves dynamic consistency between alarm parameters and alarm rules at the production unit level, across specification changes, and under varying operating conditions. This information system integration service can automatically complete alarm threshold correction, alarm rule filtering, alarm rule publishing, and alarm rule archiving in a data-driven manner during industrial production. It transforms alarm configuration capabilities from static settings to adaptive adjustments based on evolving operating conditions, significantly improving the accuracy and adaptability of alarm configurations, the stability and reliability of rule publishing, and the transparency of rule version traceability. This provides highly integrated, highly consistent, and highly verifiable technical support for flexible alarm management in continuous production scenarios.

[0050] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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.

[0051] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for SCADA flexible alarm configuration based on multi-dimension mapping, characterized in that, The method comprises the following steps: S1, real-time collection of industrial alarm mapping data, and performing time synchronization, noise suppression, abnormality rejection, missing interpolation and normalization processing on the collected data to obtain preprocessed industrial alarm mapping data; S2, constructing a product name and specification mapping table based on the preprocessed industrial alarm mapping data, and establishing a window to evaluate the working condition consistency in the current production unit to determine whether to trigger the mapping segmentation process and optimize and verify the product name and specification mapping table; S3, establishing an alarm parameter loading index based on the product name and specification mapping table, retrieving the alarm parameter template of the current production unit, performing correction on the original alarm threshold in the template, generating an alarm parameter set and performing continuous and legal verification; S4, constructing an alarm rule candidate set according to the alarm parameter loading index, evaluating the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, eliminating the conflict rules to obtain an alarm rule set, writing the alarm rule set into a SCADA alarm rule configuration interface, and archiving the alarm rules written successfully.

2. The method of claim 1, wherein: The specific steps of real-time collection of industrial alarm mapping data and performing time synchronization, noise suppression, abnormality rejection, missing interpolation and normalization processing on the collected data to obtain preprocessed industrial alarm mapping data are as follows: Real-time collection of industrial alarm mapping data, which includes material feeding temperature, material discharging temperature, tank internal material temperature, tank absolute pressure, pipeline inlet pressure, pipeline outlet pressure, material instantaneous flow, liquid level and filter front and rear pressure difference; The collected industrial alarm mapping data is time-synchronized and time-aligned with a unified clock source, and noise suppression and mutation smoothing are performed in combination with Kalman filtering algorithm; abnormal value detection is performed through Grubbs abnormality detection algorithm to filter out pseudo abnormal points caused by electromagnetic interference and equipment start-stop transient; for missing data caused by sampling intermittence and transmission delay, linear interpolation algorithm is used for data completion and sequence continuity; the industrial alarm mapping data is normalized through range normalization algorithm to unify the numerical scale and eliminate the dimension difference.

3. The method of claim 1, wherein: The specific steps of constructing a product name and specification mapping table based on the preprocessed industrial alarm mapping data are as follows: The preprocessed industrial alarm mapping data is classified and mapped according to production units with the minimum production unit as the organizational boundary, and a unique production unit identification code is established for the data stream in the same production unit; product name identification points and product specification identification points are read and analyzed within each minimum production unit, and are structurally bound with the corresponding production unit identification code to construct a product name and specification mapping table.

4. The method of claim 1, wherein: The specific steps of establishing a window to evaluate the working condition consistency in the current production unit are as follows: The fixed time step is designed as an evaluation window, the mean value of each industrial alarm mapping data in the previous evaluation window and the current evaluation window is calculated for each production unit, and the maximum value and the minimum value of each industrial alarm mapping data in the current evaluation window are extracted; the absolute value of the difference between the mean value of the i-th industrial alarm mapping data in the current evaluation window and the corresponding mean value in the previous evaluation window is taken, and then divided by the sum of the difference between the maximum value and the minimum value of the i-th industrial alarm mapping data in the current evaluation window, and then added to the minimum item, to obtain the normalized difference item; the sum of the squares of all normalized difference items of the industrial alarm mapping data is obtained, and then divided by the number of industrial alarm mapping data categories to obtain the window difference mean square item; the square root of the window difference mean square item is taken, and then multiplied by the control sensitivity coefficient to obtain the sensitive modulation item; the hyperbolic tangent function operation is applied to the sensitive modulation item to obtain the switching fluctuation evaluation value.

5. The multi-dimensional mapping based SCADA flexible alarm configuration method of claim 1, wherein: The specific steps of determining whether to trigger the mapping segment process and optimizing and checking the product name and specification mapping table are as follows: The switching fluctuation evaluation value and the fluctuation threshold value are compared in real time, and when the switching fluctuation evaluation value is less than the fluctuation threshold value, it is determined that the working condition in the current production unit is stable, the existing product name code and product specification code binding relationship is kept unchanged, and the evaluation window continues to slide backward at a fixed time step; otherwise, it is determined that the working condition in the production unit fluctuates, and the mapping segment process is triggered: the current evaluation window is closed and a new evaluation window is created, and the corresponding record in the product name and specification mapping table is updated, including the production unit identification code, the product name code, the product specification code, the evaluation window number, the window start and end time, and the switching fluctuation evaluation value, and the last evaluation window state is marked as completed; The product name code and the product specification code of adjacent evaluation windows in the same production unit are checked for consistency, and when there is an overlapping time interval, the product name code and the product specification code are retained in the order of the sampling time stamp, and the covered record is marked as a conflict handled state.

6. The multi-dimensional mapping based SCADA flexible alarm configuration method of claim 1, wherein: The specific steps of establishing an alarm parameter loading index based on the product name and specification mapping table, retrieving the alarm parameter template of the current production unit, and correcting the original alarm threshold value in the template are as follows: Based on the real-time product name and specification mapping table, the product name code and product specification code information corresponding to the current production unit are read to establish an alarm parameter loading index; the combination of the product name code, the product specification code and the production unit identification code is used as a unique index key to retrieve the corresponding alarm parameter template from the alarm parameter template library, and the alarm parameter template is loaded into the memory instance as the initial parameter set of the current production unit; the original alarm threshold value of each industrial alarm mapping data in the loaded alarm parameter template is extracted, and the correction of each alarm threshold value of the current production unit is performed: The amplitude modulation quantity is obtained by calculating the normalized difference item of the i-th industrial alarm mapping data in the current evaluation window, performing power operation on the normalized difference item according to the alarm response coefficient, and multiplying the sum of the switching fluctuation evaluation value and the constant one; the threshold scaling item is obtained by taking the opposite of the amplitude modulation quantity as an index, performing exponential operation on the natural constant e, and subtracting the exponential operation result from the constant one; the i-th industrial alarm mapping data alarm threshold correction value is obtained by multiplying the difference between the maximum value and the minimum value of the i-th industrial alarm mapping data in the current evaluation window by the threshold scaling item, and adding the original alarm threshold of the i-th industrial alarm mapping data in the template; The alarm threshold correction value of each industrial alarm mapping data is calculated, and legality verification and boundary constraint are performed; when threshold overlap and out-of-bound are detected, the historical parameter upper and lower limits of the corresponding industrial alarm mapping data and the alarm parameter template definition rule are extracted to perform secondary correction on the alarm threshold correction value, and the corrected alarm threshold correction value is taken as the final alarm threshold.

7. The multi-dimensional mapping based SCADA flexible alarm configuration method of claim 1, wherein: The specific steps of generating the alarm parameter set and performing continuous legality verification are as follows: All final alarm thresholds are extracted to form an alarm parameter set of the current production unit; during the validity period of the alarm parameter set, when it is detected that any one of the product name code, the product specification code and the production unit identification code changes, the use state of the current alarm parameter set is terminated, and the current alarm parameter set, the corresponding alarm parameter loading index and the evaluation window interval information are summarized to build a historical parameter archive; based on the latest product name code and product specification code, the parameter loading process is retriggered to generate a new alarm parameter set, and the corresponding alarm parameter loading index, alarm threshold correction record and historical parameter archive identification are output to the alarm rule configuration unit in the form of a message structure.

8. The multi-dimensional mapping based SCADA flexible alarm configuration method of claim 1, wherein: The specific steps of constructing the alarm rule candidate set according to the alarm parameter loading index, evaluating the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, and removing the conflict rules to obtain the alarm rule set are as follows: After the alarm rule configuration unit receives the message structure, the alarm parameter loading index is taken as the unique matching key to parse the alarm rule basic template of the current production unit, form the alarm rule candidate set, and generate the rule identification code and version number for each rule; In the alarm rule candidate set, multiple alarm candidate rules belonging to the same industrial alarm mapping data are sorted according to the alarm threshold size, and the safety upper limit and the safety lower limit of the i-th industrial alarm mapping data in the alarm parameter template are extracted; the difference between the two is calculated as the alarm interval span; the square of the difference between the alarm thresholds of the adjacent two alarm candidate rules of the i-th industrial alarm mapping data is calculated as the threshold difference square item; the square of the alarm interval span of the i-th industrial alarm mapping data is multiplied by the number of alarm candidate rules corresponding to the i-th industrial alarm mapping data in the current production unit minus one, and then the minimum item is added to obtain the normalized denominator item; the alarm rule conflict degree evaluation value of the j-th alarm candidate rule is obtained by subtracting the ratio of the threshold difference square item to the normalized denominator item from the constant one. The alarm rule set is written into the SCADA alarm rule configuration interface, and the specific steps of archiving the alarm rules written successfully are as follows:

9. The multi-dimensional mapping based SCADA flexible alarm configuration method of claim 1, wherein: The specific steps of writing the alarm rule set into the SCADA alarm rule configuration interface and archiving the alarm rules written successfully are as follows: According to the production unit identification code and the rule identification code, a rule publishing task queue is constructed, the alarm rules are written into the SCADA alarm rule configuration interface in the order of the rule publishing task queue, and after a successful response is returned by the interface, the corresponding rule identification code is recorded as the real-time alarm rule version of the current production unit; After the alarm rule publishing is completed, the corresponding alarm rule identification code, alarm threshold, version timestamp and rule publishing task queue processing result are extracted, and are written into a historical rule archive to form an alarm rule version record; and the rule version number of this time is returned to the alarm rule configuration unit in the form of a message structure body, so that the alarm rule can be managed traceably.

10. A SCADA flexible alarm configuration system based on multi-dimensional mapping, characterized in that: Comprise: The data acquisition and preprocessing module, the product name and specification identification binding module, the alarm parameter flexible mapping module and the alarm rule configuration publishing module, wherein: The data acquisition and preprocessing module is configured to collect industrial alarm mapping data in real time, and perform time synchronization, noise suppression, abnormality elimination, missing interpolation and normalization processing on the collected data to obtain preprocessed industrial alarm mapping data; The product name and specification identification binding module is configured to construct a product name and specification mapping table based on the preprocessed industrial alarm mapping data, and establish a window to evaluate the consistency of the working conditions in the current production unit, and determine whether to trigger a mapping segmentation process to optimize and verify the product name and specification mapping table; The alarm parameter flexible mapping module is configured to establish an alarm parameter loading index based on the product name and specification mapping table, retrieve an alarm parameter template of the current production unit, perform correction on original alarm thresholds in the template, generate an alarm parameter set and perform continuous and legal verification; The alarm rule configuration publishing module is configured to construct an alarm rule candidate set according to the alarm parameter loading index, evaluate the conflict degree of multiple alarm candidate rules corresponding to the same industrial alarm mapping data, eliminate the conflicting rules to obtain an alarm rule set, write the alarm rule set into the SCADA alarm rule configuration interface, and archive the alarm rules written successfully.

Images