A computing engine based industrial alarm system and method

Abstract

The application discloses an industrial alarm system based on a computing engine and an implementation method thereof. The industrial alarm system comprises a business rule management subsystem and a computing engine subsystem. The business rule management subsystem comprises a business rule configuration, a business rule application and a task management module, and is used for customizing a business rule, selecting a rule type, maintaining an alarm item and establishing a connection between the alarm item and a task according to a demand of a user. The computing engine subsystem comprises five modules, i.e., an alarm item synchronization, a real-time data processing, a rule calculation, an alarm storage and a message pushing. The modules work cooperatively to ensure timely synchronization of alarm rules, efficient data processing, accurate rule calculation, safe result storage and instant notification of alarm information. Through the above scheme, the application effectively reduces the coupling between alarm rule calculation and business, enables the computing engine to be independent of the business for distributed dynamic expansion, simultaneously customizes the business rule for the user, and further enhances the flexibility and adaptability of the system.

Description

Technical Field

[0001] This invention relates to the field of industrial automation, and in particular to an industrial alarm system based on a computing engine and its implementation method. Background Technology

[0002] Traditional industrial alarm systems are typically tightly integrated with specific business processes, with alarm calculations often embedded within the business logic. While this approach can meet the needs of specific business scenarios, it also introduces numerous limitations and challenges.

[0003] First, due to the deep coupling between alarm calculation and business logic, the system lacks flexibility and scalability when dealing with complex and ever-changing production environments. When production processes change or new monitoring needs arise, extensive redesign and development of the system are often required, increasing maintenance costs and implementation difficulty.

[0004] Furthermore, traditional alarm systems often fail to function effectively when faced with massive amounts of data and real-time processing requirements. With the development of Industrial Internet of Things (IIoT) technology, the amount of data is growing exponentially, and extracting effective information and responding promptly has become a major challenge for existing systems.

[0005] Therefore, there is an urgent need for an industrial alarm system based on an independent computing engine, which can be independent of specific business logic and achieve independent deployment and centralized management of alarm calculations. Summary of the Invention

[0006] The purpose of this invention is to provide an industrial alarm system and its implementation method based on a computing engine, so as to solve the limitations caused by the deep coupling between alarm calculation and business logic, and to achieve the independence and efficiency of alarm business calculation.

[0007] To achieve the above objectives, the technical solution provided by the present invention is as follows:

[0008] An industrial alarm system based on a computing engine includes a business rule management subsystem and a computing engine subsystem;

[0009] The business rules management subsystem includes:

[0010] The business rules configuration module is used by users to create and publish business rules through the front-end interface, and store the rules in the database;

[0011] The business rules application module is used by users to maintain alarm items and store them in the database based on custom rule types or system built-in rule types through the front-end interface;

[0012] The task management module allows users to create task objects through the front-end interface, then bind alarm items to the task objects by selecting alarm items, establish the association between alarm items and task objects, and store them in the database and data middleware.

[0013] The aforementioned computing engine subsystem includes:

[0014] The alarm item synchronization module is used to monitor changes in the data middleware task version and to synchronize alarm items with custom rules.

[0015] The real-time data processing module is used to collect and preprocess time-series data;

[0016] The rule calculation module is used to calculate alarm rules based on time sequence values ​​and the business rules to which alarm items belong, and output the calculation results.

[0017] The alarm storage module is used to store the calculation results in a database.

[0018] The message push module is used to monitor whether the calculation results data meet the notification conditions. When the notification conditions are met, an alarm message is sent to the relevant personnel.

[0019] This invention uses a user-oriented task management model as the link and a computing engine as the core to decouple business logic and computation. Users connect alarm service configuration data and tasks, and the computing engine proactively preempts tasks to calculate alarm service rules. Furthermore, users can customize the engine's operation based on task parameters and define user-defined business rules to adapt to more complex business scenarios.

[0020] This invention also provides an industrial alarm method based on a computing engine, the steps of which are as follows:

[0021] 1) Users create and publish business rules through the front-end interface and store the rules in the database; the rules include the following information: rule type identifier, rule name, rule content, and publication status;

[0022] 2) Users maintain alarm items through the front-end interface according to custom rule types or system built-in rule types and store them in the database; the alarm items include the following information: business data source, business rule type and parameters, and notification configuration content data;

[0023] 3) Users create task objects through the front-end interface, then select alarm items and bind them to the task objects, establishing an association between alarm items and task objects, and storing the information in the database and data middleware. The task object information includes task execution time, whether it is enabled, and periodic parameters. The data middleware records the mapping relationship between tasks and the engine through a task pool, which is a one-to-one relationship. The task pool records the task identifier and the initial version value corresponding to the task. The database stores the association information between task objects and alarm items. The association information between task objects and alarm items includes a list of task identifiers and corresponding associated alarm item identifiers, which is a one-to-many relationship.

[0024] 4) Monitor data middleware task version changes through the alarm item synchronization module that runs independently within the computing engine subsystem, and perform alarm item synchronization and custom rule synchronization;

[0025] 5) The real-time data processing module, which runs independently within the computing engine subsystem, detects whether the synchronized task is in the execution period. If it is in the execution period, it reads the real-time value and timestamp of the data point from the corresponding data source according to the execution cycle based on the data point configuration information, stores them in the timestamp value object, and pushes them to the real-time value memory queue.

[0026] 6) The rule calculation module, which runs independently within the calculation engine subsystem, checks whether a timestamp value object exists in the real-time value memory queue. If a timestamp value object exists, it matches the real-time value of the corresponding data point according to the system's built-in rules or custom rules corresponding to the alarm item, performs calculations, and saves the calculation result as a calculation result object. This result is then pushed to the alarm storage module through the calculation result memory queue or the data middleware. If the calculation result object is a real-time alarm, the calculation result object data is also saved to the real-time alarm area of ​​the data middleware.

[0027] 7) Determine whether a calculation result object exists in the calculation result memory queue or data middleware through the alarm storage module that runs independently within the calculation engine subsystem. If the calculation result data exists, parse the calculation result data and store it in the database.

[0028] 8) The message push module, which runs independently within the computing engine subsystem, determines whether the real-time alarm zone data in the data middleware matches the notification configuration conditions of the corresponding alarm item. If the notification configuration conditions are met, the message is sent according to the push configuration.

[0029] More specifically:

[0030] 1) Users create and publish business rules through the front-end interface and store the rules in the database;

[0031] Furthermore, users can create rules through the business rule configuration module. The rule information includes: rule identifier, rule name, rule content, and publication status. The rule content is Java code containing user-defined rule logic.

[0032] 2) Users maintain alarm items through the front-end interface according to custom rule types or system built-in rule types, and store them in the database;

[0033] Furthermore, custom rule types are defined by the business rule application module reading the rule type identifiers published by the rule configuration module through a scheduled task, then comparing the local rule type identifiers with the published rule type identifiers to check for any unincluded types. If a type exists, it is bound to the local custom rule type list.

[0034] Furthermore, the system's built-in rule types are the rule type logic already implemented in the computing engine subsystem.

[0035] Furthermore, the alarm item information includes data such as business data source, business rule type and parameters, and notification configuration content.

[0036] 3) Users create task objects through the front-end interface, then select alarm items and bind them to the task objects to establish the association between alarm items and task objects, and store them in the database and data middleware;

[0037] Furthermore, users can create task objects through the task management module. The task management module supports creating task objects that include defining the task execution time, whether to enable or disable, and periodic parameter configurations, and storing them in the database. At the same time, the task identifier and the initial value of the corresponding version are recorded in the task pool in the data middleware.

[0038] Furthermore, the task pool is used to record the mapping relationship between tasks and engines, which is one-to-one.

[0039] Furthermore, the initial version value is a random string identifier (UUID).

[0040] Furthermore, the database storage task management module associates alarm item information with the following: alarm item identifier and task identifier.

[0041] Furthermore, the data middleware stores the task management module and alarm item association information as task identifiers and corresponding associated alarm item list data, with a one-to-many relationship.

[0042] 4) The alarm item synchronization module monitors whether there are any unassigned tasks in the data middleware. If so, it uses a distributed lock to preempt the task. If the task preemption is successful, it binds the task identifier and engine identifier in the task pool to a one-to-one relationship. If the task preemption fails, it tries to preempt the task again. If the task preemption is successful, it monitors whether the engine identifier bound to the task in the data middleware is consistent with its own identifier. If they are inconsistent, it tries to preempt the task again. If they are consistent, it monitors whether the task version in the data middleware is consistent with the version recorded by the engine. If they are consistent, it monitors whether the version records are consistent in the next synchronization cycle. If they are inconsistent, it loads the alarm item configuration information and task configuration parameters stored in the task bound in the data middleware. Then, it reads the execution time and period configuration of the bound task. The task execution time is used to set the execution time period of the real-time data processing module, and the task period is used to set the execution period of the real-time data processing module. Secondly, the system reads the alarm item data bound to the task, parsing the data point configuration information contained in the alarm item data and pushing it to the real-time data processing module; parsing the business rule information contained in the alarm item data and pushing it to the rule calculation module; if custom rules exist, loading the rule content according to the custom rule type and dynamically creating custom rule objects; and parsing the alarm item notification configuration information contained in the alarm item data and pushing it to the message push module. Finally, after synchronization is complete, the system enters the next synchronization cycle to continue monitoring whether the task binding identifier is consistent with the engine identifier.

[0043] 5) The real-time data processing module detects whether the synchronized task is within its execution period. If it is, it reads the real-time value and timestamp of the data point from the corresponding data source according to the data point configuration information and the execution cycle, stores them in a timestamp value object, and pushes them to the real-time value memory queue.

[0044] Furthermore, the format of the timestamp numerical object is: data point identifier, data point real-time value, data point timestamp.

[0045] 6) The rule calculation module checks whether a timestamp value object exists in the real-time value memory queue. If a timestamp value object exists, it matches the real-time value of the corresponding data point according to the system's built-in rules or custom rules corresponding to the alarm item, calculates the result, saves it as a calculation result object, and pushes it to the alarm storage module through the calculation result memory queue or data middleware. If the calculation result object is a real-time alarm, the calculation result object data is also saved to the real-time alarm area of ​​the data middleware.

[0046] Furthermore, the calculation result object format is: corresponding alarm item configuration data, alarm start time, and alarm end time.

[0047] Furthermore, the data format of the real-time alarm zone in the data middleware is a key-value pair that maps one-to-one with the real-time alarm data corresponding to the alarm item identifier.

[0048] 7) The alarm storage module determines whether the calculation result object exists in the memory queue or data middleware. If the calculation result data exists, it parses the calculation result data and stores it in the database.

[0049] 8) The message push module determines whether the real-time alarm zone data in the data middleware matches the notification configuration conditions of the corresponding alarm item, and determines whether the notification configuration conditions are met. If the message push is met, the message is sent according to the push configuration.

[0050] In the above invention, both the business rule management subsystem and the computing engine subsystem can be distributed and dynamically expanded, and each computing engine subsystem will create a unique engine identifier after deployment.

[0051] In the above invention, the database supports MySQL, SQL Server, and Oracle databases.

[0052] In the above invention, the memory queue adopts Java's LinkedBlockingQueue blocking queue.

[0053] The present invention has the following advantages over the prior art:

[0054] 1. Independent Alarm Calculation: By introducing an independent calculation engine, alarm calculation is decoupled from specific business processes, allowing alarm logic to be deployed and managed independently of business logic. This not only improves system flexibility and scalability but also reduces the complexity of system maintenance and upgrades.

[0055] 2. Unified management and optimized alarm strategies: By centrally managing alarm rules and strategies, this invention can effectively reduce false alarms and missed alarms, optimize the configuration and adjustment of alarm strategies, and achieve comprehensive coverage and support for different business scenarios.

[0056] 3. Improve the adaptability and scalability of the system: The alarm system designed in this invention can flexibly adapt to the needs of different industrial environments, support multiple data sources and complex business logic, and can be expanded and upgraded according to actual needs to meet the needs of future business development.

[0057] In summary, by introducing an independent computing engine and applying advanced technologies, this invention significantly improves the performance, reliability, and intelligence level of industrial alarm systems, thereby providing safer and more efficient monitoring and alarm services for industrial production. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the system of the present invention;

[0059] Figure 2 This is a flowchart illustrating the application process of the business rules module.

[0060] Figure 3 This is a flowchart of the task management module;

[0061] Figure 4 This is a schematic diagram of the alarm item synchronization module of the present invention;

[0062] Figure 5 This is a schematic diagram of the real-time data processing module of the present invention;

[0063] Figure 6 This is a schematic diagram of the rule calculation module of the present invention;

[0064] Figure 7 This is a schematic diagram of the message push module of the present invention;

[0065] Figure 8 This is a schematic diagram of the alarm storage module of the present invention. Detailed Implementation

[0066] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0067] like Figure 1 The diagram shows a logical representation of the industrial alarm system based on a computing engine and its implementation method according to the present invention. The specific implementation steps are as follows:

[0068] 1) Users maintain custom rule data according to business needs through the business rule configuration module;

[0069] 2) such as Figure 2 As shown, the business rules application module periodically synchronizes the custom rules published by the business rules module, compares them with the local custom rules type to see if there are any non-existent rule types, and adds them to the local custom rules type if they exist. Users can choose the system's built-in rule type or a custom rule type to maintain according to their business needs, and then configure the data source information, rule parameter information, and notification configuration information to maintain alarm items.

[0070] 3) such as Figure 3 As shown, in the task management module, the user configures task parameters and saves the task parameter configuration information to the Redis middleware and MySQL. The task identifier and the initial version value of the task are recorded in the task pool of the Redis middleware. Then, alarm items are selected and associated with the task, and the task and alarm item association information is stored in MySQL. At the same time, the task alarm item association information and task configuration data are saved in the Redis middleware. The task management module and alarm item association information stored in the Redis middleware is the task identifier and the corresponding list of associated alarm items, which has a one-to-many relationship.

[0071] 4) such as Figure 4 As shown, the alarm item synchronization module of the computing engine subsystem monitors whether there are any unassigned tasks in the Redis middleware. If so, it preempts the task using a distributed lock. If the task preemption is successful, it binds the task identifier and engine identifier in the task pool to a one-to-one relationship. If the task preemption fails, it tries to preempt the task again. If the task preemption is successful, it checks whether the engine identifier bound to the task in the Redis middleware is consistent with its own identifier. If they are inconsistent, it tries to preempt the task again. If they are consistent, it checks whether the task version in the Redis middleware is consistent with the version recorded by the engine. If they are consistent, it checks whether the version records are consistent in the next synchronization cycle. If they are inconsistent, it loads the alarm item configuration information and task configuration parameters stored in the task bound in the Redis middleware. Then, it reads the execution time and period configuration of the bound task, where reading the task execution time is used to set the execution time period of the real-time data processing module, and reading the task period is used to set the execution period of the real-time data processing module. Secondly, the system reads the alarm item data bound to the task, parsing the data point configuration information contained in the alarm item data and pushing it to the real-time data processing module; parsing the business rule information contained in the alarm item data and pushing it to the rule calculation module; if custom rules exist, loading the rule content according to the custom rule type and dynamically creating custom rule objects; and parsing the alarm item notification configuration information contained in the alarm item data and pushing it to the message push module. Finally, after synchronization is complete, the system enters the next synchronization cycle to continue monitoring whether the task binding identifier is consistent with the engine identifier.

[0072] 4) such as Figure 5 As shown, the real-time data processing module detects whether the synchronized task is within its execution period. If it is, it parses the data point configuration information and reads the real-time value and timestamp from the corresponding data source according to the execution cycle, storing them in a timestamp value object and pushing it to the real-time value memory queue.

[0073] 5) such as Figure 6 As shown, the rule calculation module checks whether a timestamp value object exists in the real-time value memory queue. If a timestamp value object exists, it matches the real-time value of the corresponding data point according to the system's built-in rules or custom rules corresponding to the alarm item, performs calculations, and saves the calculation result as a calculation result object. This result is then pushed to the alarm storage module through the calculation result memory queue or Kafka middleware. If the calculation result object is a real-time alarm, the calculation result object data is also saved to the Redis middleware real-time alarm area.

[0074] 9) For example Figure 7As shown, the message push module determines whether the real-time alarm data in the Redis middleware matches the notification configuration conditions of the corresponding alarm item. If the notification configuration conditions are met, a message is sent according to the push configuration.

[0075] 10) such as Figure 8 As shown, the alarm storage module determines whether the calculation result object exists in the memory queue or Kafka middleware. If the calculation result data exists, it parses the calculation result data and stores it in MySQL.

[0076] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be considered within the scope of protection of the present invention.

Claims

1. An industrial alarm method based on a computing engine, characterized in that, Includes the following steps: 1) Users create and publish business rules through the front-end interface, and store the rules in the database; rules This includes the following information: rule type identifier, rule name, rule content, and publication status; 2) Users maintain alarm items through the front-end interface according to custom rule types or system built-in rule types and store them in the database; the alarm items include the following information: business data source, business rule type and parameters, and notification configuration content data; 3) Users create task objects through the front-end interface, and then bind alarm items to the task objects by selecting alarm items, establishing an association between alarm items and task objects, and storing the information in the database and data middleware. Task object information includes task execution time, whether it is enabled, and periodic parameters. The data middleware records the mapping relationship between tasks and the engine through a task pool, which is a one-to-one relationship. The task pool records the task identifier and the initial version value corresponding to the task. The database stores the association information between task objects and alarm items. The association information between task objects and alarm items includes a list of task identifiers and corresponding associated alarm item identifiers, which is a one-to-many relationship. 4) Monitor data middleware task version changes through the alarm item synchronization module that runs independently within the computing engine subsystem, and perform alarm item synchronization and custom rule synchronization; 5) The real-time data processing module, which runs independently within the computing engine subsystem, detects whether the synchronized task is in the execution period. If it is in the execution period, it reads the real-time value and timestamp of the data point from the corresponding data source according to the execution cycle based on the data point configuration information, stores them in the timestamp value object, and pushes them to the real-time value memory queue. 6) The rule calculation module, which runs independently within the calculation engine subsystem, checks whether a timestamp value object exists in the real-time value memory queue. If a timestamp value object exists, it matches the real-time value of the corresponding data point according to the system's built-in rules or custom rules corresponding to the alarm item, performs calculations, and saves the calculation result as a calculation result object. This result is then pushed to the alarm storage module through the calculation result memory queue or the data middleware. If the calculation result object is a real-time alarm, the calculation result object data is also saved to the real-time alarm area of ​​the data middleware. 7) Determine whether a calculation result object exists in the memory queue or data middleware through the alarm storage module that runs independently within the calculation engine subsystem. If the calculation result data exists, parse the calculation result data and store it in the database. 8) The message push module, which runs independently within the computing engine subsystem, determines whether the real-time alarm zone data in the data middleware matches the corresponding alarm item notification configuration conditions. If the notification configuration conditions are met, the message is sent according to the push configuration.

2. An industrial alarm system based on a computing engine that employs the industrial alarm method described in claim 1, characterized in that, This includes a business rules management subsystem and a computing engine subsystem; The business rules management subsystem includes: The business rules configuration module is used by users to create and publish business rules through the front-end interface, and store the rules in the database; The business rules application module is used by users to maintain alarm items and store them in the database based on custom rule types or system built-in rule types through the front-end interface; The task management module allows users to create task objects through the front-end interface, then bind alarm items to the task objects by selecting alarm items, establish the association between alarm items and task objects, and store them in the database and data middleware. The computing engine subsystem includes: The alarm item synchronization module is used to monitor changes in the data middleware task version and to synchronize alarm items with custom rules. The real-time data processing module is used to collect and preprocess time-series data; The rule calculation module is used to calculate alarm rules based on time sequence values ​​and the business rules to which alarm items belong, and output the calculation results. The alarm storage module is used to store the calculation results in a database. The message push module is used to monitor whether the calculation results data meet the notification conditions. When the notification conditions are met, an alarm message is sent to the relevant personnel.

3. The industrial alarm system based on a computing engine according to claim 2, characterized in that, The rules created in the business rule configuration module include the following information: rule type identifier, rule name, rule content, and publication status.

4. The industrial alarm system based on a computing engine according to claim 3, characterized in that, The business rule application module is implemented as follows: A scheduled task reads the rule type identifiers published by the rule configuration module, then compares the local rule type identifier with the published rule type identifiers to check for any unincluded types. If any exist, they are bound to the local custom rule type list. The user selects a custom rule type or a system-built-in rule type, and then maintains the data source, rule parameters, and notification configuration data to create alarm items. The system-built-in rule type is the rule type logic already implemented in the computing engine subsystem. The alarm item includes the following information: business data source, business rule type and parameters, and notification configuration content.

5. The industrial alarm system based on a computing engine according to claim 4, characterized in that, The task management module includes task execution time, whether it is enabled, and periodic parameters for the task object information. Simultaneously, the task identifier and the initial version value corresponding to the task are recorded in the task pool of the data middleware, and the association information between the task object and alarm items is stored in the database. The task pool is used to record the mapping relationship between tasks and the engine, which is one-to-one. The association information between the task object and alarm items includes a list of task identifiers and corresponding associated alarm item identifiers, which is one-to-many.

6. The industrial alarm system based on a computing engine according to claim 5, characterized in that, The alarm item synchronization module is an independently running module within the computing engine subsystem, and its implementation steps are as follows: Step S1: Monitor whether there are any unassigned tasks enabled in the data middleware. If so, preempt the task using a distributed lock. If the task preemption is successful, bind the task identifier and engine identifier in the task pool to a one-to-one relationship and proceed to step S2. If the task preemption fails, proceed to step S1. Step S2: Monitor whether the task's bound engine identifier in the data middleware matches its own identifier. If they do not match, proceed to step S1; if they match, proceed to step S3. Step S3: Monitor whether the version of the data middleware task is consistent with the version recorded by the engine. If they are consistent, proceed to step S2; otherwise, proceed to step S4. Step S4: Load the alarm item configuration information and task configuration parameters stored in the task bound to the data middleware; Step S5: Read the bound task execution time and period configuration, where reading the task execution time is used to set the execution time period of the real-time data processing module, and reading the task period is used to set the execution period of the real-time data processing module; Step S6: Read the alarm item data bound to this task, including parsing the data point configuration information contained in the alarm item data and pushing it to the real-time data processing module; parsing the business rule information contained in the alarm item data and pushing it to the rule calculation module; if there are custom rules, load the rule content according to the custom rule type and dynamically create a custom rule object; parsing the alarm item notification configuration information contained in the alarm item data and pushing it to the message push module. After synchronization is complete, proceed to step S2.

7. The industrial alarm system based on a computing engine according to claim 6, characterized in that, The real-time data processing module is an independently running module within the computing engine subsystem. Its implementation steps are as follows: Detect whether the synchronized task is in the execution time period. If it is in the execution time period, read the real-time value and timestamp of the data point from the corresponding data source according to the execution cycle based on the data point configuration information, store them in a timestamp value object, and push them to the real-time value memory queue. The timestamp value object includes the following information: data point identifier, data point real-time value, and data point timestamp.

8. The industrial alarm system based on a computing engine according to claim 7, characterized in that, The rule calculation module is an independently running module within the calculation engine subsystem. Its implementation steps are as follows: It checks whether a timestamp value object exists in the real-time value memory queue. If a timestamp value object exists, it calculates the corresponding real-time value of the data point according to the system's built-in rules or custom rules corresponding to the alarm item. The calculation result is saved as a calculation result object and pushed to the alarm storage module through the calculation result memory queue or data middleware. If the calculation result object is a real-time alarm, the calculation result object data is also saved to the real-time alarm area of ​​the data middleware. The calculation result object includes the following information: corresponding alarm item configuration data, alarm start time, and alarm end time. The data format of the real-time alarm area of ​​the data middleware is a key-value pair that maps alarm item identifiers to corresponding real-time alarm data.

9. The industrial alarm system based on a computing engine according to claim 8, characterized in that, The alarm storage module is an independently running module within the computing engine subsystem. Its implementation steps are as follows: determine whether there is a computing result object in the computing result memory queue or data middleware. If computing result data exists, parse the computing result data and store it in the database.

10. The industrial alarm system based on a computing engine according to claim 9, characterized in that, The message push module is an independently running module within the computing engine subsystem. Its implementation steps are as follows: determine whether the real-time alarm zone data in the data middleware matches the corresponding alarm item notification configuration conditions, determine whether the notification configuration conditions are met, and if the message push is met, send the message according to the push configuration.

Images