Power network anomaly detection method

Abstract

The invention discloses a power network anomaly detection method, which belongs to the technical field of power networks, and is characterized in that a center node acquires queuing operations of all equipment nodes in an operation transmission module and calculates danger coefficients of all the queuing operations, and the operations are divided into two stages through the design, so that the task load of the center node for processing the operations is greatly reduced, and the operation efficiency is improved. Therefore, the processing time of the central node is shortened, an information transmission mode is designed, an information transmission strategy is formulated, the central node performs allocation operation on available hardware equipment of all equipment nodes in the operation transmission module according to the characteristics of the information transmission strategy, a calculation unit is constructed, the available hardware equipment is fully utilized, and the processing efficiency of the central node is improved. And the overall efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power networks, and in particular to a method for detecting anomalies in a power network. Background Art

[0002] The power grid must analyze and monitor information in real time, be able to respond to abnormal situations promptly, and reduce the probability of failures.

[0003] The traditional anomaly detection method is to hand over a large amount of network data to a central node for processing. However, as the amount and complexity of network information increases, the difficulty and time of anomaly detection will increase greatly. This model has become difficult to cope with the current power network.

[0004] Under the above model, many hardware devices may be idle at certain times, while other resources are overloaded due to overload operation. This unbalanced allocation system will also lead to reduced overall efficiency and increased operating costs. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, the present invention provides a method for detecting anomalies in a power network.

[0006] The present invention adopts the following technical solution:

[0007] A method for detecting anomalies in a power network comprises the following steps:

[0008] S1. The central node obtains the queued jobs of all device nodes in the job transmission module and calculates the risk coefficients of all queued jobs;

[0009] S2. The central node adds all queued jobs to queued job group 1 or queued job group 2 according to the risk coefficient of all queued jobs. At the same time, queued job group 1 is retained in the central node, and queued job group 2 is transmitted to the edge node.

[0010] S3, the central node designs information transmission mode 1 according to the queueing jobs in the queueing job group 1, and formulates information transmission strategy 1 for the queueing jobs in the queueing job group 1 according to the information transmission mode 1, and transmits the information transmission strategy 1 to the central node;

[0011] S4, the edge node sets up information transmission strategy 2 for the queued jobs in the queued job group 2 according to the previously set information transmission mode 2, and transmits the information transmission strategy 2 to the central node at the same time;

[0012] S5. The central node allocates the available hardware devices of all device nodes in the job transmission module according to the characteristics of the information transmission strategy 1 and the information transmission strategy 2, and constructs the computing unit 1 and the computing unit 2.

[0013] S6. The computing unit 1 processes the queued jobs in the information transmission strategy 1 according to the information transmission strategy 1, and the computing unit 2 processes the queued jobs in the information transmission strategy 2 according to the information transmission strategy 2.

[0014] According to the power network anomaly detection method, in each job transmission module, all device nodes transmit the queued jobs to the central node. After the central node obtains the queued jobs, it calculates the risk factor of the queued jobs based on the characteristics of the queued jobs, the characteristics of the information contained in the queued jobs, and the characteristics of the device nodes of the queued jobs.

[0015] According to the power network anomaly detection method, the central node calculates the risk factor of the queue operation according to the following formula:

[0016]

[0017] Where D m is the risk factor of the queue operation with order m; D mi is the characteristic of the information of order j contained in the queued job of order m; E j is the weight of the characteristics of the operation with order j on the risk factor; N is the number of characteristics of the operation; the constant H is the correction factor; R a is the weight of the information characteristics contained in the queuing operation on the risk factor; T m is the characteristic of queuing operations; R b is the weight of the characteristics of the queuing operation on the risk factor; F m is the characteristic of the device node of the queued job with order m; R c It is the weight of the characteristics of the device nodes of the queued jobs on the risk factor.

[0018] According to the power network anomaly detection method, the central node adds queued jobs with risk factors higher than a threshold value to queued job group one, and adds queued jobs with risk factors lower than the threshold value to queued job group two.

[0019] According to the power network anomaly detection method, after the central node obtains the queuing operation group 1, the central node calculates the information transmission selection standard according to the characteristics of each queuing operation in the queuing operation group 1.

[0020] According to the power network anomaly detection method described above, the information transmission method selection standard S is calculated according to the following formula: t ,

[0021]

[0022] Among them, Z aiis the characteristic of the queued job with order i in the queued job group 1; I is the number of queued jobs in the queued job group 1; G a It is the weight of the characteristics of the queuing operation on the selection criteria of the information transmission method; is the optimization standard set by the central node for the characteristics of the queued job with the order n in the queued job group 1, Z an Any queued job selected from the queued job group 1, and the optimization criteria for the central node setting are obtained by analyzing the historical data set; G b It is the weight of the optimization criteria for setting the central node on the selection criteria for the information transmission method.

[0023] According to the described power network anomaly detection method, the central node selects information transmission method one from the previously established information transmission method set according to the selection criteria of the information transmission method, and then establishes information transmission strategy one for the queueing jobs in the queueing job group one according to the information transmission method one, and transmits the information transmission strategy one to the central node at the same time.

[0024] According to the described method for detecting anomalies in a power network, after the central node obtains the information transmission strategy one in the central node and the information transmission strategy two delivered by the edge node, the corresponding expected standards for hardware devices are calculated according to the characteristics of the information transmission strategy one and the characteristics of the information transmission strategy two. At the same time, according to the respective proportions of the expected standards for hardware devices of the information transmission strategy one and the expected standards for hardware devices of the information transmission strategy two, the available hardware devices of all device nodes in the job transmission module are constructed into computing units one and two.

[0025] According to the power network anomaly detection method, the expected standard V of the hardware device of the information transmission mode of the order α is calculated according to the following formula: α ,

[0026]

[0027] In the formula, Ω αy is the importance of the queued job with the order y in the information transmission mode with the order α; Y is the number of queued jobs in the information transmission mode with the order α; is the number of hardware devices with order x that are expected by queued jobs with order y in the information transmission mode with order α; αyx is the weight coefficient of the hardware device with order x expected by the queued job with order y in the information transmission mode with order α; X is the number of categories of hardware devices expected by the queued job with order y in the information transmission mode with order α.

[0028] According to the described power network anomaly detection method, after the idle hardware devices of all device nodes in the central node job transmission module are constructed into computing unit one and computing unit two, computing unit one processes the queued jobs in information transmission strategy one according to information transmission strategy one, and computing unit two processes the queued jobs in information transmission strategy two according to information transmission strategy two.

[0029] Compared with the prior art, the beneficial effects of the present invention include at least:

[0030] A method for calculating the risk factor was designed, and the jobs were divided into two levels. Jobs with high fault levels and urgent processing were executed in the central node, and the others were executed in the edge nodes. This greatly reduced the workload of the central node to process jobs, thereby reducing the processing time of the central node.

[0031] An information transmission strategy is designed to fully utilize available hardware devices and improve overall efficiency. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention is described clearly and completely below. The embodiments described in this application are only embodiments of a part of the present invention, not all embodiments. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the protection scope of the present invention.

[0033] This application provides a method for detecting anomalies in a power network, the steps of which are as follows:

[0034] S1. The central node obtains the queued jobs of all device nodes in the job transmission module and calculates the risk coefficients of all queued jobs;

[0035] In each job transmission module, all device nodes transmit the queued jobs to the central node. After the central node obtains the queued jobs, it calculates the risk factor of the queued jobs based on the characteristics of the queued jobs, the characteristics of the information contained in the queued jobs, and the characteristics of the device nodes of the queued jobs.

[0036] The central node calculates the risk factor of the queue operation according to the following formula:

[0037]

[0038] Where D m is the risk factor of the queue operation with order m; D mj is the characteristic of the information of order j contained in the queued job of order m; E j is the weight of the characteristics of the operation with order j on the risk factor; N is the number of characteristics of the operation; the constant H is the correction factor; Ra is the weight of the information characteristics contained in the queuing operation on the risk factor; T m is the characteristic of queuing operations; R b is the weight of the characteristics of the queuing operation on the risk factor; F m is the characteristic of the device node of the queued job with order m; R c It is the weight of the characteristics of the device nodes of the queued jobs on the risk factor.

[0039] S2. The central node adds all queued jobs to queued job group 1 or queued job group 2 according to the risk coefficient of all queued jobs. At the same time, queued job group 1 is retained in the central node, and queued job group 2 is transmitted to the edge node.

[0040] Because there are many device nodes and all device nodes generate the same number of queued jobs, the central node receives the same number of queued jobs. After the central node calculates the risk coefficients of all queued jobs, it determines whether the queued jobs should be added to queued job group 1 or queued job group 2 according to the risk coefficients of all queued jobs. The central node adds queued jobs with risk coefficients higher than the threshold to queued job group 1, and adds queued jobs with risk coefficients lower than the threshold to queued job group 2.

[0041] S3, the central node designs information transmission mode 1 according to the queueing jobs in the queueing job group 1, and formulates information transmission strategy 1 for the queueing jobs in the queueing job group 1 according to the information transmission mode 1, and transmits the information transmission strategy 1 to the central node;

[0042] After the central node obtains the queued job group 1, the central node calculates the information transmission selection standard according to the characteristics of each queued job in the queued job group 1.

[0043] The information transmission method selection standard S is calculated according to the following formula t ,

[0044]

[0045] Among them, Z ai is the characteristic of the queued job with order i in the queued job group 1; I is the number of queued jobs in the queued job group 1; G a It is the weight of the characteristics of the queuing operation on the selection criteria of the information transmission method; is the optimization standard set by the central node for the characteristics of the queued job with the order n in the queued job group 1, Z an Any queued job selected from the queued job group 1, and the optimization criteria for the central node setting are obtained by analyzing the historical data set; G bIt is the weight of the optimization criteria for setting the central node on the selection criteria for the information transmission method.

[0046] The central node selects information transmission method one from the previously established information transmission method set according to the selection criteria of the information transmission method, and then establishes information transmission strategy one for the queueing jobs in the queueing job group one according to information transmission method one, and transmits information transmission strategy one to the central node at the same time.

[0047] S4, the edge node sets up information transmission strategy 2 for the queued jobs in the queued job group 2 according to the previously set information transmission mode 2, and transmits the information transmission strategy 2 to the central node at the same time;

[0048] Because the edge node processes the job information with a low risk factor, after the edge node obtains the queued job group 2, it directly establishes the information transmission strategy 2 for the queued jobs in the second queued job set according to the information transmission method 2, and transmits the information transmission strategy 2 to the central node at the same time.

[0049] S5. The central node allocates the available hardware devices of all device nodes in the job transmission module according to the characteristics of the information transmission strategy 1 and the information transmission strategy 2, and constructs the computing unit 1 and the computing unit 2.

[0050] After the central node obtains the information transmission strategy one in the central node and the information transmission strategy two delivered by the edge node, it calculates the corresponding expected standards for hardware devices according to the characteristics of information transmission strategy one and the characteristics of information transmission strategy two. At the same time, according to the respective proportions of the expected standards for hardware devices of information transmission strategy one and the expected standards for hardware devices of information transmission strategy two, the available hardware devices of all device nodes in the job transmission module are constructed into computing units one and two.

[0051] According to the following formula, the hardware device expected standard V of the information transmission method with order α is calculated α ,

[0052]

[0053] In the formula, Ω αy is the importance of the queued job with the order y in the information transmission mode with the order α; Y is the number of queued jobs in the information transmission mode with the order α; is the number of hardware devices with order x that are expected by queued jobs with order y in the information transmission mode with order α; αyx is the weight coefficient of the hardware device with order x expected by the queued job with order y in the information transmission mode with order α; X is the number of categories of hardware devices expected by the queued job with order y in the information transmission mode with order α.

[0054] S6, computing unit 1 processes the queued jobs in information transmission strategy 1 according to information transmission strategy 1, and computing unit 2 processes the queued jobs in information transmission strategy 2 according to information transmission strategy 2;

[0055] After the idle hardware devices of all device nodes in the central node job transmission module are constructed into computing unit one and computing unit two, computing unit one processes the queued jobs in information transmission strategy one according to information transmission strategy one, and computing unit two processes the queued jobs in information transmission strategy two according to information transmission strategy two.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for detecting anomalies in a power network, characterized in that: The following steps are involved: S1. The central node obtains the queued jobs of all device nodes in the job transmission module and calculates the risk coefficients of all queued jobs; S2. The central node adds all queued jobs to queued job group 1 or queued job group 2 according to the risk coefficient of all queued jobs. At the same time, queued job group 1 is retained in the central node, and queued job group 2 is transmitted to the edge node. S3, the central node designs information transmission mode 1 according to the queueing jobs in the queueing job group 1, and formulates information transmission strategy 1 for the queueing jobs in the queueing job group 1 according to the information transmission mode 1, and transmits the information transmission strategy 1 to the central node; S4, the edge node sets up information transmission strategy 2 for the queued jobs in the queued job group 2 according to the previously set information transmission mode 2, and transmits the information transmission strategy 2 to the central node at the same time; S5. The central node allocates the available hardware devices of all device nodes in the job transmission module according to the characteristics of the information transmission strategy 1 and the information transmission strategy 2, and constructs the computing unit 1 and the computing unit 2. S6. The computing unit 1 processes the queued jobs in the information transmission strategy 1 according to the information transmission strategy 1, and the computing unit 2 processes the queued jobs in the information transmission strategy 2 according to the information transmission strategy 2.

2. A method for detecting anomalies in a power network according to claim 1, characterized in that: In each job transmission module, all device nodes transmit the queued jobs to the central node. After the central node obtains the queued jobs, it calculates the risk factor of the queued jobs based on the characteristics of the queued jobs, the characteristics of the information contained in the queued jobs, and the characteristics of the device nodes of the queued jobs.

3. A method for detecting anomalies in a power network according to claim 2, characterized in that: The central node calculates the risk factor of the queue operation according to the following formula: Where D m is the risk factor of the queue operation with order m; D mj is the characteristic of the information of order j contained in the queued job of order m; E j is the weight of the characteristics of the operation with order j on the risk factor; N is the number of characteristics of the operation; the constant H is the correction factor; R a is the weight of the information characteristics contained in the queuing operation on the risk factor; T m is the characteristic of queuing operations; R b is the weight of the characteristics of the queuing operation on the risk factor; F m is the characteristic of the device node of the queued job of order m; R c It is the weight of the characteristics of the device nodes of the queued jobs on the risk factor.

4. A method for detecting anomalies in a power network according to claim 1, characterized in that: The central node adds the queued jobs with risk factors higher than the threshold to the queued job group 1, and adds the queued jobs with risk factors lower than the threshold to the queued job group 2.

5. A method for detecting anomalies in a power network according to claim 1, characterized in that: After the central node obtains the queued job group 1, the central node calculates the information transmission selection standard according to the characteristics of each queued job in the queued job group 1.

6. A method for detecting anomalies in a power network according to claim 5, characterized in that: The information transmission method selection standard S is calculated according to the following formula t , Among them, Z ai is the characteristic of the queued job with order i in the queued job group 1; I is the number of queued jobs in the queued job group 1; G a It is the weight of the characteristics of the queuing operation on the selection criteria of the information transmission method; is the optimization standard set by the central node for the characteristics of the queued job with the order n in the queued job group 1, Z an Any queued job selected from the queued job group 1, and the optimization criteria for the central node setting are obtained by analyzing the historical data set; G b It is the weight of the optimization criteria for setting the central node on the selection criteria for the information transmission method.

7. A method for detecting anomalies in a power network according to claim 1, characterized in that: The central node selects information transmission method one from the previously established information transmission method set according to the selection criteria of the information transmission method, and then establishes information transmission strategy one for the queueing jobs in the queueing job group one according to information transmission method one, and transmits information transmission strategy one to the central node at the same time.

8. A method for detecting anomalies in a power network according to claim 1, characterized in that: After the central node obtains the information transmission strategy one in the central node and the information transmission strategy two delivered by the edge node, it calculates the corresponding expected standards for hardware devices according to the characteristics of information transmission strategy one and the characteristics of information transmission strategy two. At the same time, according to the respective proportions of the expected standards for hardware devices of information transmission strategy one and the expected standards for hardware devices of information transmission strategy two, the available hardware devices of all device nodes in the job transmission module are constructed into computing units one and two.

9. A method for detecting anomalies in a power network according to claim 8, characterized in that: According to the following formula, the hardware device expected standard V of the information transmission method with order α is calculated α , In the formula, Ω αy is the importance of the queued job with the order y in the information transmission mode with the order α; Y is the number of queued jobs in the information transmission mode with the order α; is the number of hardware devices with order x that are expected by queued jobs with order y in the information transmission mode with order α; αyx is the weight coefficient of the hardware device with order x expected by the queued job with order y in the information transmission mode with order α; X is the number of categories of hardware devices expected by the queued job with order y in the information transmission mode with order α.

10. The method for detecting anomalies in a power network according to claim 1, characterized in that: After the idle hardware devices of all device nodes in the central node job transmission module are constructed into computing unit one and computing unit two, computing unit one processes the queued jobs in information transmission strategy one according to information transmission strategy one, and computing unit two processes the queued jobs in information transmission strategy two according to information transmission strategy two.

Images