A method for object relationship association of early warning events based on a dynamic ontology business model

Abstract

The application discloses a kind of object relationship association methods of early warning event based on dynamic ontology business model implementation.The method improves the dynamic adaptive capacity of early warning management of power grid system.The method is specifically as follows: by executing the alarm step W of association,firstly, the fault object pointed to by the target object in the relationship graph and the alarm event object associated with the fault object are identified;Then find the associated object in the target object relationship graph that has a fault association relationship with the target object;Finally, the alarm relationship from the associated object to the alarm event object is established;By executing the object relationship association step R, for the alarm event object not directly associated with the target object, i.e. isolated alarm object, the association relationship between the isolated alarm object and the above-mentioned fault object is judged, and if there is an association between the two, the isolated alarm object is used as the early warning event of the target object and the associated object, and the early warning relationship from the isolated alarm object to the fault object is established.

Description

Technical Field

[0001] This invention mainly relates to the field of semantic fusion business model technology, and in particular to an object relationship association method for early warning events based on a dynamic ontology business model. Background Technology

[0002] In recent years, with the rapid development of the power industry, the number of distribution network devices in the power grid system has become increasingly large. The traditional mode of the power grid system is to conduct regular inspections and maintenance on these distribution network devices. However, the inspection process requires long-term power outages and also poses safety hazards such as missed inspections due to differences in the capabilities of maintenance personnel. To ensure power safety, the power grid system has now fully launched dynamic monitoring of the operating status of distribution network equipment. After acquiring data from various instruments and sensors that monitor the operating parameters of distribution network equipment at the power field, the power grid system processes this real-time data and cached historical data through a series of processing rules in the power grid system's information processing system. This results in the output of the operating status of the distribution network equipment and the issuance of alarms for abnormal conditions of the distribution network equipment. The current problem is that, with the rapid development of information technology, the real-time data generated by the power grid system is too massive. Existing data processing systems struggle to discover and extract meaningful data in a timely manner when faced with such a large amount of real-time data. They need to import all the data into a database and then use statistical analysis tools to analyze it. Although some meaningful data can be obtained, real-time performance cannot be guaranteed, and timely processing of some meaningful data is not possible. Furthermore, due to the limitations of system administrators' experience in the available statistical analysis tools, the existing power grid system suffers from inconvenient data relationship processing in operation and maintenance management. For example, setting early warning relationships for newly added distribution network equipment requires separate manual configuration. Summary of the Invention

[0003] The inventors envisioned a dynamic ontology semantic fusion platform as an open, autonomous semantic fusion and visualization exploration application platform that utilizes big data analytics and knowledge graph technologies to meet the needs of data analysis, semantic fusion, and business exploration. By employing dynamic ontology semantic modeling methods, multi-source heterogeneous data fusion, and complex event definitions, it can dynamically simulate unobservable aspects of the power grid system, enabling adaptive operation control of the power grid system. This effectively enhances the dynamic adaptive capability of the power grid system in complex scenarios, supporting its flexible, efficient, and reliable operation.

[0004] The technical problem to be solved by this invention is how to improve the dynamic adaptive capability of early warning management in power grid systems.

[0005] To address the aforementioned technical problems, this invention provides a method for associating object relationships in early warning events based on a dynamic ontology business model, comprising the following alarm association steps W:

[0006] Step W1: Obtain the relationship graph of the target object;

[0007] Step W2: Identify the target object in the relationship graph as pointing to: the fault object and the alarm event object associated with and matched with the fault object;

[0008] Step W3: Identify the associated objects in the relationship graph that have a fault association relationship with the target object. The fault association relationship means that the target object and the associated object point to the same fault object.

[0009] Step W4: Establish an alarm relationship from the associated object to the alarm event object;

[0010] The object relationship association method for this early warning event also includes the following object relationship association step R:

[0011] Step R1: Obtain other alarm event objects in the target object's relationship graph that are not directly related to the target object, and define them as isolated alarm objects;

[0012] Step R2: Determine whether there is an association between the isolated alarm object and the faulty object;

[0013] Step R3: If the judgment result of step R2 is yes, then the isolated alarm object is used as the warning event of both the target object and the associated object in the associated warning step W, and a warning relationship is established from the isolated alarm object to the fault object.

[0014] Furthermore, in step W3, the fact that the target object and the associated object point to the same fault object means that the fault object includes faults that occur simultaneously in both the target object and the associated object.

[0015] Furthermore, in step W3, the target object and the associated object pointing to the same fault object means that the target object and the associated object have respectively experienced faults with a causal relationship within a preset time period.

[0016] Furthermore, in step W3, the associated object is a direct associated object of the target object.

[0017] Furthermore, in step R1 of the object relationship association step R, the isolated alarm object refers to an alarm event object that is indirectly associated with the target object through one or more non-alarm event objects.

[0018] Furthermore, the number P of non-alarm event objects sandwiched between the isolated alarm object and the alarm event object does not exceed a preset N.

[0019] Furthermore, the value of N is determined based on the computing power of the dynamic ontology business model or the preset computing time.

[0020] Furthermore, the number Q of objects sandwiched between the isolated alarm object and the fault object is less than the number P of non-alarm event objects sandwiched between the isolated alarm object and the alarm event object.

[0021] Furthermore, all objects sandwiched between the isolated alarm object and the faulty object are associated with the target object.

[0022] Furthermore, in step R3 of the object relationship association step R, an early warning relationship is established between the isolated alarm object and the target object and associated object in the fault object.

[0023] In this embodiment of the power system, after establishing an instance of the target object on the dynamic ontology business model, the dynamic ontology business model is updated according to the relationship and data of the newly added distribution network equipment to obtain the current relationship graph of the target object. The object relationship association method for early warning events implemented based on the above dynamic ontology business model includes the association alarm step W and the object relationship association step R. In this embodiment, after adding distribution network equipment to the power system, the administrator only needs to update the dynamic ontology business model with the added distribution network equipment, and then execute the task of automatically obtaining model early warning relationships for the selected target equipment. The system will then use the newly added distribution network equipment as the associated equipment of the target equipment according to the pre-recorded program, and execute the object relationship association method for early warning events based on the dynamic ontology business model. By executing the associated alarm step W, the alarm relationship between the associated object and the alarm event object is obtained. Then, by executing the object relationship association step R, other alarm events (i.e., isolated alarm objects) that are not directly associated with the target object in the target object relationship graph are obtained, and an early warning relationship is established from the isolated alarm object to the fault object of the newly added distribution network equipment. This is equivalent to directly migrating the alarm events of the target object to the newly added distribution network equipment as the associated object. The administrator does not need to manually identify and set the early warning relationship of the newly added distribution network equipment from the real-time collected data, which improves the dynamic adaptive capability of the early warning management of the power grid system. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0025] Figure 1 This is a flowchart of the object relationship association method for early warning events based on a dynamic ontology business model provided by the present invention;

[0026] Figure 2 This is a schematic diagram of the system structure of the object relationship association method for early warning events based on the dynamic ontology business model provided by the present invention. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to specific embodiments.

[0028] In this embodiment, the dynamic ontology semantic fusion platform is deployed on a server. The server integrates ETL tools, supports multi-source heterogeneous data acquisition, and the administrator logs into the dynamic ontology semantic fusion platform on the server client. After establishing a dynamic ontology business model related to the power system's distribution network equipment, data acquisition operations are performed in the intelligent access module of the data management function to construct business instances. When distribution network equipment is added or changed in the system, the administrator configures the early warning relationship for the updated system, especially for newly added distribution network equipment (such as switching equipment associated with transformers), using the object relationship association method for early warning events implemented by the dynamic ontology business model detailed below. The specific process is as follows: Figure 1 As shown below, the specific steps of the associated alarm step W and the object relationship association step R of this method will be explained in detail below.

[0029] This embodiment obtains the alarm relationship between associated objects and alarm event objects by executing the alarm association step W. Specifically, the alarm association step W includes the following steps: Step W1, obtaining the relationship graph of the target object; Step W2, identifying the fault object and the alarm event object that is associated with and matched with the fault object in the relationship graph; Step W3, identifying the associated objects in the relationship graph that have a fault association relationship with the target object, where the fault association relationship means that the target object and the associated object point to the same fault object; Step W4, establishing the alarm relationship from the associated object to the alarm event object.

[0030] like Figure 2 For the target equipment, the transformer, its directly related distribution network equipment includes switchgear. Both the switchgear and the transformer are linked to the same fault object (circuit breaker). In practice, if a fault occurs simultaneously in both the target and related equipment—for example, when faults corresponding to different switchgear items in the switchgear's attributes occur, and the transformer itself also experiences a corresponding fault—then the system considers this fault to be a fault object. Alternatively, if the target and related equipment experience causally related faults within a preset time period—for example, a circuit breaker failure in the switchgear leads to a circuit breaker failure in the transformer itself—then the system also considers this fault to be a fault object.

[0031] After associating the associated object with the alarm event object through the aforementioned alarm association step W, the other alarm events of the target object are directly migrated to the newly added distribution network device that is used as the associated object through the object relationship association step R detailed below. Specifically, the object relationship association step R is as follows: Step R1, obtain other alarm event objects in the target object relationship graph that are not directly associated with the target object, and define them as isolated alarm objects; Step R2, determine whether there is an association relationship from the isolated alarm object to the fault object; Step R3, if the determination result of step R2 is yes, then the isolated alarm object is used as the warning event of both the target object and the associated object in the alarm association step W, and a warning relationship from the isolated alarm object to the fault object is established.

[0032] In step R1 of the object relationship association step R, the isolated alarm object refers to an alarm event object that is indirectly associated with the target object through one or more non-alarm event objects. The number P of non-alarm event objects sandwiched between the isolated alarm object and the alarm event object is determined in advance based on the computing power of the dynamic ontology business model or the preset computing time. In this embodiment, the upper limit N is 3. Figure 2 The alarm event generated by the weather factor object is used as the isolated alarm object. This alarm event is indirectly associated with the target equipment transformer through the two layers of event relationships of weather factor and regional location. That is, the number P of non-alarm event objects sandwiched between the isolated alarm object and the alarm event object is 2. The system confirms that the value of the number P2 is less than the upper limit N value of 3, and confirms that the alarm event generated by the weather factor object is used as the isolated alarm object.

[0033] The isolated alarm object's "alarm event" attributes include various alarm trigger conditions, such as strong wind warnings, lightning warnings, and heavy rain warnings. The strong wind warning trigger condition is associated with tree-falling accidents, which in turn are associated with faults in transformer-related objects. The number of objects (tree-falling accidents) sandwiched between the isolated alarm object and the fault object (Q) is 1. The system determines that if Q is less than the number of non-alarm event objects (P) sandwiched between the isolated alarm object and the alarm event object (2), then the isolated alarm event is indeed associated with a fault object. This constraint reduces computation in complex events.

[0034] The system verifies whether all objects sandwiched between the isolated alarm object and the fault object (tree collapse accident in this embodiment) are associated with the target object. The tree collapse accident is indirectly associated with the target object transformer through environmental factor objects and regional location. That is, the system determines that the object sandwiched in the middle is indeed associated with the target object transformer. The above restrictions reduce the impact of irrelevant events and objects on the target object in complex events.

[0035] In step R3 of the object relationship association step R, early warning relationships are established between the isolated alarm object and the target object and associated objects in the fault object. In the object relationship association step R, alarm events caused by weather factors are used as early warning conditions for the target object transformer. The switching equipment, as the associated device in the association alarm step W with the target object transformer, can directly transfer this alarm relationship of the target object transformer to the associated device, and synchronously use the alarm events caused by weather factors as its early warning conditions.

[0036] In this embodiment of the power system, after establishing an instance of the target object on the dynamic ontology business model, the dynamic ontology business model is updated according to the relationship and data of the newly added distribution network equipment to obtain the current relationship graph of the target object. The object relationship association method for early warning events implemented based on the above dynamic ontology business model includes the association alarm step W and the object relationship association step R. In this embodiment, after adding distribution network equipment to the power system, the administrator only needs to update the dynamic ontology business model with the added distribution network equipment, and then execute the task of automatically obtaining model early warning relationships for the selected target equipment. The system will then use the newly added distribution network equipment as the associated equipment of the target equipment according to the pre-recorded program, and execute the object relationship association method for early warning events based on the dynamic ontology business model. By executing the associated alarm step W, the alarm relationship between the associated object and the alarm event object is obtained. Then, by executing the object relationship association step R, other alarm events (i.e., isolated alarm objects) that are not directly associated with the target object in the target object relationship graph are obtained, and an early warning relationship is established from the isolated alarm object to the fault object of the newly added distribution network equipment. This is equivalent to directly migrating the alarm events of the target object to the newly added distribution network equipment as the associated object. The administrator does not need to manually identify and set the early warning relationship of the newly added distribution network equipment from the real-time collected data, which improves the dynamic adaptive capability of the early warning management of the power grid system.

[0037] The above description is merely an embodiment of the present invention and does not limit the scope of patent protection. Any non-substantial changes or substitutions made by those skilled in the art based on the present invention will still fall within the scope of patent protection.

Claims

1. A method for associating object relationships in early warning events based on a dynamic ontology business model, characterized in that, This includes the following associated alarm steps W: Step W1: Obtain the relationship graph of the target object; Step W2: Identify the target object in the relationship graph as pointing to: the fault object and the alarm event object associated with and matched with the fault object; Step W3: Identify the associated objects in the relationship graph that have a fault association relationship with the target object. The fault association relationship means that the target object and the associated object point to the same fault object. Step W4: Establish an alarm relationship from the associated object to the alarm event object; It also includes the following object relationship association step R: Step R1: Obtain other alarm event objects in the relationship graph of the target object that are not directly related to the target object, and define them as isolated alarm objects; Step R2: Determine whether there is an association between the isolated alarm object and the faulty object; Step R3: If the judgment result of step R2 is yes, then the isolated alarm object is used as the early warning event of both the target object and the associated object in the associated alarm step W, and an early warning relationship is established from the isolated alarm object to the fault object.

2. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 1, characterized in that, In step W3, the fact that the target object and the associated object point to the same fault object means that the fault object includes the fault that occurs simultaneously in both the target object and the associated object.

3. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 1, characterized in that, In step W3, the target object and the associated object pointing to the same fault object means that the target object and the associated object have respectively experienced faults with a causal relationship within a preset time period.

4. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 1, characterized in that, In step W3, the associated object is the direct associated object of the target object.

5. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 1, characterized in that, In step R1 of the object relationship association step R, the isolated alarm object refers to the alarm event object that is indirectly associated with the target object through one or more non-alarm event objects.

6. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 5, characterized in that, The number P of non-alarm event objects sandwiched between the isolated alarm object and the alarm event object does not exceed a preset number N.

7. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 6, characterized in that, The value of N is determined based on the computing power of the dynamic ontology business model or the preset computing time.

8. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 6, characterized in that, The number Q of objects sandwiched between the isolated alarm object and the fault object is less than the number P of non-alarm event objects sandwiched between the isolated alarm object and the alarm event object.

9. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 8, characterized in that, All objects sandwiched between the isolated alarm object and the faulty object are associated with the target object.

10. The object relationship association method for early warning events based on a dynamic ontology business model as described in claim 2 or 3, characterized in that, in In step R3 of the object relationship association step R, an early warning relationship is established between the isolated alarm object and the target object and associated object in the fault object.

Images