A time series data generation method and device, and a storage medium

Abstract

The application discloses a time series data generation method, device and storage medium, and is used for improving the simulation accuracy and efficiency of non-stationary time series data. The time series data generation method disclosed by the application comprises the following steps: performing hardware configuration; performing specific configuration according to the hardware configuration; constructing a time series data generation system according to the specific configuration; and generating time series data according to the time series data generation system. The application also provides a time series data generation device and a storage medium. The technical scheme can improve the simulation accuracy and efficiency of non-stationary time series data.

Description

Technical Field

[0001] This application relates to the field of computing technology, and in particular to a method, apparatus and storage medium for generating time-series data. Background Technology

[0002] In the exploration of time series data, traditional time series simulation techniques can be used to compensate for insufficient data due to objective reasons. Insufficient data in deep learning can lead to underfitting of models, resulting in ineffective classification. Simulation methods can obtain sufficient time series data, but there are subtle differences between real and simulated data. Currently, non-stationary time series data is the most challenging type in time series data exploration because its changes are almost unpredictable. For large-scale industrial production, the simulation of non-stationary time series data generation requires processing of this data. In certain specific scenarios, it is necessary to design a device that continuously generates data 24 hours a day, producing the required data within a specified timeframe; however, existing simulation techniques suffer from low accuracy and efficiency, failing to meet these requirements. Summary of the Invention

[0003] To address the aforementioned technical problems, embodiments of this application provide a time-series data generation method, apparatus, and storage medium to improve the simulation accuracy and efficiency of non-stationary time-series data.

[0004] In a first aspect, an embodiment of this application provides a time-series data generation method, comprising:

[0005] Perform hardware configuration;

[0006] Configure the hardware according to the aforementioned hardware configuration;

[0007] Construct a time-series data generation system based on the specific configuration described above;

[0008] The time series data is generated by the time series data generation system.

[0009] Preferably, the process before hardware configuration further includes:

[0010] Configure a program to generate S types of time-series data in M ​​virtual machines;

[0011] Where M is a positive integer preset according to requirements, and S is an integer greater than or equal to 5.

[0012] Preferably, the S types of time-series data include:

[0013] S equals 5;

[0014] The S types of time-series data are as follows:

[0015] Non-stationary timing data of high-voltage direct current discharge (HVDC);

[0016] Non-stationary time-series data RV of stray current interference in subway;

[0017] Outlier-related time series data OPV;

[0018] Non-stationary time-series data NV due to natural discharge interference;

[0019] Non-stationary timing data of high-voltage AC discharge (HVAC).

[0020] Furthermore, the hardware configuration includes:

[0021] Configure a distributed virtual machine management system;

[0022] The distributed virtual machine management system includes:

[0023] A computing cluster deployed using N hosts;

[0024] Where N is a positive integer preset according to requirements.

[0025] Preferably, the computing cluster includes:

[0026] There are M virtual machines, each virtual machine is a test stub, and the test stub can generate one of S types of time series data.

[0027] Preferably, the specific configuration based on the hardware configuration includes:

[0028] Virtual machines are built on the N hosts;

[0029] Each virtual machine is packaged using a preset operating system;

[0030] Each of the aforementioned virtual machines is deployed with a program for generating S types of time-series data;

[0031] The networking method between the N hosts is NET mode, and the physical connection is implemented using bridged mode;

[0032] Each virtual machine is allocated maximum resources and configured for concurrent execution.

[0033] Preferably, the step of constructing the time-series data generation system according to the specific configuration includes:

[0034] Confirm the status of the physical machines, which include N servers, each of which is an independent network unit and each has independent access permissions;

[0035] Deploy an operating system for each of the servers, and install and start the cloud computing management platform;

[0036] Confirm that the maximum virtual machine resources configured on each server are within the capacity of the physical machine;

[0037] The program for generating S types of time-series data is copied to the N servers respectively;

[0038] M virtual machines are established according to the cloud computing management platform. Each virtual machine includes one of the S types of time series data generation programs. The number of virtual machines containing each type of time series data is determined according to the probability of the occurrence of the corresponding time series data.

[0039] The M virtual machines are pulled by the N servers, each server represents a region, and the number of virtual machines pulled by each server is determined according to a preset rule.

[0040] Run the time-series data generation program in the M virtual environments;

[0041] The cloud computing management platform is OpenStack.

[0042] Preferably, the generation of time-series data based on the time-series data generation system includes:

[0043] Run the timing data generation program in the M virtual environments to generate simulated timing data, wherein the simulated timing data includes:

[0044] HVDC Time Series Dataset

[0045] RV time series dataset

[0046] OPV time series dataset

[0047] NV Time Series Dataset

[0048] HVAC Time Series Dataset

[0049] Where v1 is the number of HVDC time series data, v2 is the number of RV time series data, v3 is the number of OPV time series data, v4 is the number of NV time series data, and v5 is the number of HVAC time series data.

[0050] Using the method provided by this invention, the OpenStack virtual machine technology is first used to operate the system. Virtual machines replace test stubs, and each virtual machine is responsible for generating time series data for one or more scenarios. As the system runs simultaneously, time series data is obtained as needed, resulting in large-scale time series data. With the orderly operation of virtual machines, the recursive patterns between each type of data can be traced, laying the foundation for statistical analysis of the inherent patterns in the data.

[0051] Secondly, embodiments of this application also provide a time-series data generation apparatus, comprising:

[0052] The hardware configuration module is configured to perform hardware configuration.

[0053] The specific configuration module is configured to perform specific configurations based on the hardware configuration.

[0054] The system construction module is configured to build a time-series data generation system according to the specific configuration.

[0055] The timing generation module is configured to generate timing data based on the timing data generation system.

[0056] Thirdly, embodiments of this application also provide a timing data generation apparatus, including: a memory, a processor, and a user interface;

[0057] The memory is used to store computer programs;

[0058] The user interface is used to interact with the user;

[0059] The processor is used to read the computer program in the memory, and when the processor executes the computer program, it implements the timing data generation method provided by the present invention.

[0060] Fourthly, embodiments of this application also provide a processor-readable storage medium storing a computer program, wherein the processor executes the computer program to implement the timing data generation method provided by the present invention.

[0061] The advantages of this invention are: using the method of this invention makes the generation of time series data more systematic, integrated, large-scale, and professional, with high generation efficiency, and allows for customized production of large-scale time series data that meets production conditions. Modular management is more convenient, and the application of time series data is more extensive, which can improve the simulation accuracy and efficiency of non-stationary time series data. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 This is a schematic diagram of a time-series data generation method provided in an embodiment of this application;

[0064] Figure 2 This is a schematic diagram of a time-series data generation system provided in an embodiment of this application;

[0065] Figure 3 This is a schematic diagram illustrating the steps of building a time-series data generation system provided in an embodiment of this application;

[0066] Figure 4 A schematic diagram of a timing data generation device provided in an embodiment of this application;

[0067] Figure 5 This is a schematic diagram of another time-series data generation device provided in an embodiment of this application. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0069] The following are explanations of some of the words that appear in the text:

[0070] 1. In the embodiments of this invention, the term "and / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the preceding and following associated objects have an "or" relationship.

[0071] 2. In the embodiments of this application, the term "multiple" refers to two or more, and other quantifiers are similar.

[0072] 3. OpenStack, an open-source cloud computing management platform.

[0073] The non-stationary time series data used in this application include:

[0074] 1. HVDC, an abbreviation for Higher Voltage DC, describes non-stationary time-series data of high-voltage DC discharge. Its characteristic is that the voltage will show regular sudden increases or decreases over a certain period of time, indicating that the outer coating is missing at this time.

[0075] 2. RV, short for Random Voltage, describes non-stationary time-series data of stray current interference in subways. It is characterized by large fluctuations. It is caused by the West-East Gas Pipeline network around the city receiving discharges from the subway tracks to the ground, and there happens to be pipelines nearby. It is an accidental discharge.

[0076] 3. OPV, short for One Point Voltage, describes outlier-related time-series data. It is characterized by large fluctuations at a single point and is a time-series data waveform that occurs when equipment fails.

[0077] 4. NV, short for Nature Voltage, describes non-stationary time-series data of natural discharge interference, characterized by drastic fluctuations at single or multiple points. It is caused by natural discharge striking the ground near the pipe, and the probability of occurrence is extremely low.

[0078] 5. HVAC, short for High Voltage AC, describes non-stationary time-series data of high-voltage alternating current discharge, characterized by periodic and severe fluctuations over a period of time. It is caused by discharges in high-voltage line networks near pipelines and occurs on certain fixed lines.

[0079] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0080] It should be noted that the order in which the embodiments of this application are presented only represents the chronological order of the embodiments and does not represent the superiority or inferiority of the technical solutions provided by the embodiments.

[0081] Example 1

[0082] See Figure 1 The present application provides a schematic diagram of a time-series data generation method, as shown in the embodiment. Figure 1 As shown, the method includes steps S101 to S104:

[0083] S101. Perform hardware configuration;

[0084] As a preferred example, this step involves hardware configuration, including:

[0085] Configure a distributed virtual machine management system;

[0086] The distributed virtual machine management system includes:

[0087] A computing cluster deployed using N hosts;

[0088] Where N is a positive integer preset according to requirements.

[0089] In this invention, a computing cluster deployed using N hosts may include M virtual machines, each virtual machine serving as a test stub. Each test stub can generate one of S types of time-series data. Here, M is a preset positive integer, and S is an integer greater than or equal to 5. For example, depending on the requirements, M can be set to 2^11, and S can be set to 5.

[0090] As a preferred example, when S is 5, the S types of time-series data are as follows:

[0091] Non-stationary timing data of high-voltage direct current discharge (HVDC);

[0092] Non-stationary time-series data RV of stray current interference in subway;

[0093] Outlier-related time series data OPV;

[0094] Non-stationary time-series data NV due to natural discharge interference;

[0095] Non-stationary timing data of high-voltage AC discharge (HVAC).

[0096] When implementing the method of this invention, the platform used to configure the distributed virtual machine management system can be selected as needed, and this invention does not impose any limitations. The following explanation uses the configuration of the distributed virtual machine management system using OpenStack as an example to illustrate this step.

[0097] The OpenStack distributed system uses CentOS 8.0 as its foundation, deploying a computing cluster using seven hosts—specifically, a 7-host, 7-virtual-machine (7VM) cluster—to establish the main architecture. The 7VM cluster consists of seven hosts, each hosting seven virtual machines (VMs). OpenStack's MariaDB container accesses a MySQL database, and the IP address of each VM is stored in the MySQL database. Each VM represents a test stub, and the networking within each physical machine uses host-based mode, while the servers are bridged. This simulates the entire project's operation. Load balancing technology, implemented through elastic scaling, reduces the hardware load on the OpenStack system and ultimately yields a complete set of simulated time-series data.

[0098] As another preferred example, the process of configuring the hardware further includes:

[0099] Configure S types of time-series data generation programs in M ​​virtual machines; where M is a positive integer preset according to requirements, and S is an integer greater than or equal to 5.

[0100] As another preferred example, preliminary work is also included before hardware configuration. Taking the above example of configuring a distributed virtual machine management system using OpenStack, with the number of virtual machines M equal to 2^11 and the number of time-series data types S being 5, the preliminary work includes:

[0101] This method utilizes virtual machines created on multiple control nodes, each representing a test stub and operating an independent operating system. Real-time simulation of time-series data is achieved using virtual machines created with OpenStack. Simulation programs for five types of time-series data (HVDC, HVAC, OPV, RV, NV, etc.) are embedded into each test stub according to its location. These programs are automatically executed upon startup using the yum command. Finally, the simulated time-series data generated by all 221 virtual machines are aggregated to obtain a simulated time-series dataset. In the OpenStack distributed time-series data simulation method, time-series data simulation is the foundation of OpenStack's distributed technology. It is integrated into OpenStack and simulates time-series data on each virtual machine node within OpenStack to simulate all time-series data for a day or even several days in the entire project, as well as the changes in each node, simulating the evolution process and actual changes. OpenStack is deployed on a computer using cloud computing virtualization technology, with virtual machine management technology serving as each test stub node. Efficient time-series data simulation can be achieved through OpenStack.

[0102] For example, applying the method of this invention to the West-East Gas Pipeline, the specific process of deploying the OpenStack system on OpenStack involves configuring each virtual machine with a time-series data framework designed using the aforementioned five time-series data simulation techniques to simulate the entire process of potential changes during pipeline transportation. 211 devices are deployed across the country to monitor for leakage on the pipeline's outer wall. Using the method of this invention to monitor for paint coating leakage on the outer wall of the West-East Gas Pipeline, an energized device can be set up every 3-5 km to detect potential changes; abnormal potential changes indicate a pipeline fault. The 211 devices determine whether the pipeline paint coating has peeled off based on changes in the points of contact. In this method, HVDC represents the potential change pattern when the coating peels off. HVAC indicates DC discharge in the soil caused by leakage from nearby power poles; RV indicates random voltage caused by dry soil and air; NV indicates natural discharge encountered by the pipeline, such as lightning; and OPV indicates stray current interference from the subway, as some sections of the pipeline are built under subway lines. The 211 test devices are replaced by 211 virtual machines deployed in different ways. Nationwide, the West-East Gas Pipeline project comprises seven regions. A 7-control, 7-computing cluster is deployed on seven resource-rich physical machines. Each control node represents a region, and the virtual machines running on each control point act as pipeline monitoring devices within that region. The transmission of potential change information between different virtual machines relies on OpenStack's local area network configured in host mode for unified management. This enables data transfer between virtual machines and simulates changes in pipeline potential data. Each virtual machine is configured with the probability of the aforementioned potential anomalies occurring at that point, and automatically generates corresponding time-series data in chronological order.

[0103] It should be noted that the above parameter configuration is only an example of the application of the method of the present invention on the West-East Gas Pipeline. If used in other scenarios, the number of virtual machines can be customized according to different requirements.

[0104] S102. Configure the hardware according to the hardware configuration.

[0105] As a preferred example, the specific configuration based on the hardware configuration includes:

[0106] Virtual machines are built on the N hosts;

[0107] Each virtual machine is packaged using a preset operating system;

[0108] Each of the aforementioned virtual machines is deployed with a program for generating S types of time-series data;

[0109] The networking method between the N hosts is NET mode, and the physical connection is implemented using bridged mode;

[0110] Each virtual machine is allocated maximum resources and configured for concurrent execution.

[0111] The following example uses the configuration of a distributed virtual machine management system using OpenStack to illustrate this step.

[0112] 1. Hardware Foundation of the OpenStack System: Using multi-node OpenStack virtual machine management technology, virtualized virtual machines are built on seven physical machines running CentOS 8.0. Each virtual machine represents a test stub. Each VM is encapsulated within a Linux system, and each Linux system loads basic Python modules, injecting the five timing generation programs mentioned above. The virtual machines are run using OpenStack. The networking method for each physical machine is NET mode. Because isolation and security are not a primary concern, appropriately reducing the isolation requirements of OpenStack can improve its operational efficiency.

[0113] 2. OpenStack system management mode: The VirtualBox virtualization platform is used for unified management on the servers. Connections between physical machines are bridged, sharing network resources. For example, on the West-East Gas Pipeline, because each physical machine represents one of the seven major regions, the connection method uses a bridged, independent physical machine mode. The host machine controls one network interface card (NIC), and multiple different virtual machines are virtualized based on the host NIC.

[0114] 3. OpenStack Configuration on the OpenStack Server: Based on the CentOS 8.0 operating system, a computing cluster is deployed using 7 high-performance hosts, resulting in 7 high-performance nodes. This mode effectively prevents physical machine downtime, making the virtualization platform more stable. MariaDB is used to improve MySQL service. The IP address of each virtual machine is stored in the MySQL database; then, virtual machines are built, with each VM representing a test stub. Resources are allocated to the virtual machines on each physical machine to the maximum extent possible to meet their needs, and they are configured for concurrent execution.

[0115] S103. Construct a time-series data generation system according to the specific configuration;

[0116] As a preferred example, the construction of the time-series data generation system according to the specific configuration includes:

[0117] Confirm the status of the physical machines, which include N servers, each of which is an independent network unit and each has independent access permissions;

[0118] Deploy an operating system for each of the servers, and install and start the cloud computing management platform;

[0119] Confirm that the maximum virtual machine resources configured on each server are within the capacity of the physical machine;

[0120] The program for generating S types of time-series data is copied to the N servers respectively;

[0121] M virtual machines are established according to the cloud computing management platform. Each virtual machine includes one of the S types of time series data generation programs. The number of virtual machines containing each type of time series data is determined according to the probability of the occurrence of the corresponding time series data.

[0122] The M virtual machines are pulled by the N servers, each server represents a region, and the number of virtual machines pulled by each server is determined according to a preset rule.

[0123] Run the time-series data generation program in the M virtual environments.

[0124] The following example illustrates this step using OpenStack to configure a distributed virtual machine management system:

[0125] like Figure 2 The diagram shows the composition of a time-series data generation system.

[0126] Five kernel functions were designed for five fault types using time-series data simulation technology. A corresponding framework was then constructed based on directional probability constraints. In the distributed architecture, each module uses seven different servers to simulate 211 test cases occurring simultaneously in different areas. For example, to approximate real-world conditions, data was obtained based on two years of real-world data and fault occurrence rates: HVDC fault occurrence rate 3%, OPV fault occurrence rate 3%, HVAC fault occurrence rate 4%, NV fault occurrence rate 0.01%, and RV fault occurrence rate 2%. Each of the five fault types has its own independent probabilistic directional constraint framework and corresponding kernel function. Deploying the probabilities of the five fault types across seven servers aims to simulate a real time-series environment and achieve real-time concurrency. For example… Figure 2 In this example, taking the application of the method of this invention to the West-East Gas Pipeline, Control01 represents the server for the Jiangsu-Zhejiang-Shanghai route; Control02 represents the server for the Guangzhou-Shenzhen route; Control03 represents the server for the Yinchuan route; Control04 represents the server for the Changsha route; Control05 represents the server for the Henan route; Control06 represents the server for the Hubei route; and Control07 represents the server for the Xinjiang route. OpenStack configures these parameters into the corresponding time-series data generation algorithm program on each virtual machine.

[0127] The construction process of the time series data generation system is as follows:

[0128] S301. Confirm the physical machine status;

[0129] The physical machine includes N servers, each server is an independent network unit, and each server has independent access permissions;

[0130] For example, seven servers are used, each corresponding to a different node. The networking method between the nodes adopts host mode to ensure that each node is an independent network unit and that each node has independent access permissions.

[0131] S302. Deploy an operating system for each of the servers, and install and start the cloud computing management platform;

[0132] For example, install the CentOS 8.0 operating system on each node. After installation, deploy the OpenStack-U version. In the IP address configuration, the two servers correspond to two Control01 IPs, and the other five physical machines are configured with five Control02 to Control07 IPs. After configuring the OpenStack IPs, proceed with the OpenStack installation. After installation, start OpenStack.

[0133] S303. Confirm that the maximum virtual machine resources configured on each server are within the capacity of the physical machine;

[0134] For example, the intermediate message transmission component is controlled by the MySQL database. The elastic scaling and load balancing of virtual machines ensure that the maximum number of virtual machines pulled by each node is within the capacity of the physical machine. If it cannot be handled, the virtual machine service of that node will be started by other physical machines without affecting its normal operation.

[0135] S304. Copy the generation program for the S types of time series data to the N servers respectively;

[0136] For example, five different virtual machines with data generation code are copied to seven physical machines. Using these five virtual machines as the standard, passwordless operation is configured through a scripting language, enabling passwordless communication between OpenStack and the virtual machines. These five virtual machines can generate five different time-series data sequences: non-stationary time-series data of high-voltage direct current discharge (HVDC), non-stationary time-series data of subway stray current interference (RV), outlier-related time-series data of OPV, non-stationary time-series data of spontaneous discharge interference (NV), and non-stationary time-series data of high-voltage alternating current discharge (HVAC).

[0137] S305. Establish M virtual machines according to the cloud computing management platform. Each virtual machine includes one of the S types of time series data generation programs. The number of virtual machines containing each type of time series data in the M virtual machines is determined according to the probability of the occurrence of the corresponding time series data.

[0138] For example, using kubectl in OpenStack to control the launching of virtual machines, 211 virtual machines are launched. Each of these 211 virtual machines is one of the five basic virtual machines mentioned above, selected according to the required quantity. The quantity of each type of virtual machine is allocated based on the probability of its occurrence.

[0139] S306. The M virtual machines are pulled by the N servers, each server represents a region, and the number of virtual machines pulled by each server is determined according to a preset rule.

[0140] For example, 211 virtual machines are pulled as needed by 7 physical machines. These 7 physical machines correspond to 7 different regions, and the number of test stubs varies from region to region. Therefore, the number of virtual machines pulled from the 7 physical machines will also vary.

[0141] S307. Run the time-series data generation program in the M virtual systems.

[0142] For example, all virtual machines are started to run their data generation code. The code then uses sh script commands to return the generated time-series data to the hard disk under Cinder-Volume for storage. This completes the process of generating large-scale time-series data.

[0143] S104. The timing data generation system generates timing data based on the timing data.

[0144] Preferably, the timing data generation program in the M virtual systems is run to generate simulated timing data, wherein the simulated timing data includes:

[0145] HVDC Time Series Dataset

[0146] RV time series dataset

[0147] OPV time series dataset

[0148] NV Time Series Dataset

[0149] HVAC Time Series Dataset

[0150] Where v1 is the number of HVDC time series data, v2 is the number of RV time series data, v3 is the number of OPV time series data, v4 is the number of NV time series data, and v5 is the number of HVAC time series data.

[0151] Using the method provided by this invention, OpenStack virtual machine technology is first employed to replace test stubs. Each virtual machine is responsible for generating time-series data for one or more scenarios. As the system runs simultaneously, time-series data is obtained on demand, resulting in large-scale time-series data. Furthermore, the orderly operation of the virtual machines reveals the recursive patterns between each data type, laying the foundation for statistical analysis of the data's inherent regularities. The OpenStack system generates massive amounts of non-stationary time-series data such as HVDC, HVAC, OPV, and NV, which can meet the needs of deep learning training. The use of simulated data can save more than half of the data cleaning time, which is of great value in training and tuning parameters for deep learning models. In addition, the method of this invention makes time-series data generation more systematic, integrated, scalable, and professional, with high generation efficiency. It allows for customized production of large-scale time-series data to meet production conditions, and modular management is more convenient, leading to wider applications of time-series data.

[0152] Example 2

[0153] Based on the same inventive concept, embodiments of the present invention also provide a time-series data generation apparatus, such as... Figure 4 As shown, the device includes:

[0154] Hardware configuration module 401 is configured to perform hardware configuration.

[0155] The specific configuration module 402 is configured to perform specific configurations based on the hardware configuration.

[0156] System construction module 403 is configured to construct a time-series data generation system according to the specific configuration;

[0157] The timing generation module 404 is configured to generate timing data based on the timing data generation system.

[0158] As a preferred example, the hardware configuration module 401 is also configured to be used for

[0159] Configure a distributed virtual machine management system;

[0160] The distributed virtual machine management system includes:

[0161] A computing cluster deployed using N hosts;

[0162] Where N is a positive integer preset according to requirements.

[0163] The computing cluster comprises M virtual machines, each virtual machine serving as a test stub, and each test stub can generate one of S types of time-series data. A program for generating the S types of time-series data is configured in each of the M virtual machines.

[0164] As a preferred example, the S types of time-series data include:

[0165] S equals 5;

[0166] The S types of time-series data are as follows:

[0167] Non-stationary timing data of high-voltage direct current discharge (HVDC);

[0168] Non-stationary time-series data RV of stray current interference in subway;

[0169] Outlier-related time series data OPV;

[0170] Non-stationary time-series data NV due to natural discharge interference;

[0171] Non-stationary timing data of high-voltage AC discharge (HVAC).

[0172] As a preferred example, the specific configuration module 402 is configured to perform specific configurations based on the hardware configuration:

[0173] Virtual machines are built on the N hosts;

[0174] Each virtual machine is packaged using a preset operating system;

[0175] Each of the aforementioned virtual machines is deployed with a program for generating S types of time-series data;

[0176] The networking method between the N hosts is NET mode, and the physical connection is implemented using bridged mode;

[0177] Each virtual machine is allocated maximum resources and configured for concurrent execution.

[0178] As a preferred example, system building module 403 is configured to build a time-series data generation system according to the specific configuration:

[0179] Confirm the status of the physical machines, which include N servers, each of which is an independent network unit and each has independent access permissions;

[0180] Deploy an operating system for each of the servers, and install and start the cloud computing management platform;

[0181] Confirm that the maximum virtual machine resources configured on each server are within the capacity of the physical machine;

[0182] The program for generating S types of time-series data is copied to the N servers respectively;

[0183] M virtual machines are established according to the cloud computing management platform. Each virtual machine includes one of the S types of time series data generation programs. The number of virtual machines containing each type of time series data is determined according to the probability of the occurrence of the corresponding time series data.

[0184] The M virtual machines are pulled by the N servers, each server represents a region, and the number of virtual machines pulled by each server is determined according to a preset rule.

[0185] Run the time-series data generation program in the M virtual environments.

[0186] As a preferred example, the timing generation module 404 generates timing data including:

[0187] Run the timing data generation program in the M virtual environments to generate simulated timing data, wherein the simulated timing data includes:

[0188] HVDC Time Series Dataset

[0189] RV time series dataset

[0190] OPV time series dataset

[0191] NV Time Series Dataset

[0192] HVAC Time Series Dataset

[0193] Where v1 is the number of HVDC time series data, v2 is the number of RV time series data, v3 is the number of OPV time series data, v4 is the number of NV time series data, and v5 is the number of HVAC time series data.

[0194] It should be noted that the hardware configuration module 401 provided in this embodiment can realize all the functions included in step S101 in embodiment one, solve the same technical problems, and achieve the same technical effects, which will not be repeated here;

[0195] It should be noted that the specific configuration module 402 provided in this embodiment can realize all the functions included in step S102 in embodiment one, solve the same technical problems, and achieve the same technical effects, and will not be described again here;

[0196] It should be noted that the system construction module 403 provided in this embodiment can realize all the functions included in step S103 in embodiment one, solve the same technical problems, and achieve the same technical effects, which will not be repeated here;

[0197] It should be noted that the timing generation module 404 provided in this embodiment can realize all the functions included in step S104 in embodiment one, solve the same technical problems, and achieve the same technical effects, which will not be repeated here;

[0198] It should be noted that the device provided in Embodiment 2 and the method provided in Embodiment 1 belong to the same inventive concept, solve the same technical problem, and achieve the same technical effect. The device provided in Embodiment 2 can implement all the methods in Embodiment 1, and the similarities will not be repeated.

[0199] Example 3

[0200] Based on the same inventive concept, embodiments of the present invention also provide a time-series data generation apparatus, such as... Figure 5 As shown, the device includes:

[0201] It includes a memory 502, a processor 501, and a user interface 503;

[0202] The memory 502 is used to store computer programs;

[0203] The user interface 503 is used to interact with the user.

[0204] The processor 501 is used to read the computer program in the memory 502, and when the processor 501 executes the computer program, it implements:

[0205] Perform hardware configuration;

[0206] Configure the hardware according to the aforementioned hardware configuration;

[0207] Construct a time-series data generation system based on the specific configuration described above;

[0208] The time series data is generated by the time series data generation system.

[0209] Among them, Figure 5In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (processor 501) and memory (memory 502). The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides the interface. Processor 501 is responsible for managing the bus architecture and general processing, and memory 502 can store data used by processor 501 during operation.

[0210] Processor 501 can be a CPU, ASIC, FPGA or CPLD, and processor 501 can also adopt a multi-core architecture.

[0211] When processor 501 executes the computer program stored in memory 502, it implements any of the timing data generation methods in Embodiment 1.

[0212] It should be noted that the device provided in Embodiment 3 and the method provided in Embodiment 1 belong to the same inventive concept, solve the same technical problem, and achieve the same technical effect. The device provided in Embodiment 3 can implement all the methods in Embodiment 1, and the similarities will not be repeated.

[0213] This application also proposes a processor-readable storage medium. This processor-readable storage medium stores a computer program, and when the processor executes the computer program, it implements any of the timing data generation methods in Embodiment 1.

[0214] It should be noted that the division of units in the embodiments of this application is illustrative and only represents one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units.

[0215] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0216] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0217] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0218] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for generating time series data, the method comprising: Comprising: carrying out hardware configuration; carrying out specific configuration according to the hardware configuration; building a time series data generation system according to the specific configuration; generating time series data according to the time series data generation system; the carrying out hardware configuration further comprising: configuring S kinds of time series data generation programs in M virtual machines; wherein M is a positive integer preset according to demand, and S is an integer greater than or equal to 5; wherein the S kinds of time series data include: S equals 5; the S kinds of time series data are respectively: non-stationary time series data HVDC of high-voltage direct current discharge; non-stationary time series data RV of subway stray current interference; outlier related time series data OPV; non-stationary time series data NV of natural discharge interference; non-stationary time series data HVAC of high-voltage alternating current discharge; the carrying out hardware configuration includes: configuring a distributed virtual machine management system; the distributed virtual machine management system includes: a computing cluster deployed using N hosts; wherein N is a positive integer preset according to demand; the computing cluster includes: M virtual machines, each virtual machine being a test stub, and the test stub being capable of generating one of the S kinds of time series data; the building a time series data generation system according to the specific configuration includes: confirming the state of a physical machine, the physical machine including N servers, each server being an independent network unit, and each server having independent access authority; deploying an operating system for each of the servers, installing and starting a cloud computing management platform; confirming that the maximum virtual machine resources configured for each server are within the bearing range of the physical machine; copying the S kinds of time series data generation programs to the N servers respectively; establishing M virtual machines according to the cloud computing management platform, each virtual machine including one of the S kinds of time series data generation programs, and the number of virtual machines containing each kind of time series data in the M virtual machines being determined according to the probability of occurrence of the corresponding time series data; the M virtual machines are pulled by the N servers, each server representing a region, and the number of virtual machines pulled by each server being determined according to a preset rule; running the time series data generation programs in the M virtual machines; the cloud computing management platform is Openstack.

2. The method of claim 1, wherein, the carrying out specific configuration according to the hardware configuration includes: building virtualized virtual machines on the N hosts; each of the virtual machines is encapsulated through a preset operating system; each of the virtual machines is deployed with S kinds of time series data generation programs; the networking mode between the N hosts is a NET mode, and physical connection is achieved using a bridging mode; the resource allocation of each of the virtual machines is maximum configuration, and is configured to be executed concurrently.

3. The method of claim 1, wherein, the generating time series data according to the time series data generation system includes: running the time series data generation programs in the M virtual machines to generate simulated time series data, wherein the simulated time series data includes: HVDC time series dataset ​ RV time series dataset ; OPV time series dataset ; NV time series dataset ; HVAC time series dataset ; wherein v1 is the number of HVDC time series data, v2 is the number of RV time series data, v3 is the number of OPV time series data, v4 is the number of NV time series data, and v5 is the number of HVAC time series data.

4. A time-series data generation apparatus characterized by comprising: Comprising: a hardware configuration module configured to carry out hardware configuration; a specific configuration module configured to configure specifically according to the hardware configuration; a system construction module configured to construct a time series data generation system according to the specific configuration; a time series generation module configured to generate time series data according to the time series data generation system; the hardware configuration further comprises: configuring generation procedures of S kinds of time series data in M virtual machines; wherein M is a positive integer preset according to requirements, and S is an integer greater than or equal to 5; wherein the S kinds of time series data include: S equals 5; the S kinds of time series data are respectively: non-stationary time series data HVDC of high-voltage direct current discharge; non-stationary time series data RV of subway stray current interference; outlier-related time series data OPV; non-stationary time series data NV of natural discharge interference; non-stationary time series data HVAC of high-voltage alternating current discharge; the hardware configuration comprises: configuring a distributed virtual machine management system; the distributed virtual machine management system comprises: a computing cluster deployed using N hosts; wherein N is a positive integer preset according to requirements; the computing cluster comprises: M virtual machines, each of which is a test stub, and each test stub can generate one of the S kinds of time series data; constructing the time series data generation system according to the specific configuration comprises: confirming the state of a physical machine, which includes N servers, each of which is an independent network unit and has independent access authority; deploying an operating system for each server, installing and starting a cloud computing management platform; confirming that the maximum virtual machine resource configured for each server is within the bearing range of the physical machine; copying the generation procedures of the S kinds of time series data to the N servers respectively; establishing M virtual machines according to the cloud computing management platform, each of which includes one of the generation procedures of the S kinds of time series data, and the number of virtual machines containing each kind of time series data in the M virtual machines is determined according to the probability of occurrence of the corresponding time series data; the M virtual machines are pulled by the N servers, each of which represents a region, and the number of virtual machines pulled by each server is determined according to a preset rule; running the time series data generation procedures in the M virtual machines; the cloud computing management platform is Openstack.

5. A time-series data generation apparatus characterized by comprising: comprises a memory, a processor and a user interface; the memory is used to store a computer program; the user interface is used to realize interaction with a user; the processor is used to read the computer program in the memory, and the processor realizes the time series data generation method of any one of claims 1 to 3 when executing the computer program.

6. A processor-readable storage medium, characterized in that, the processor readable storage medium stores a computer program, and the processor realizes the time series data generation method of any one of claims 1 to 3 when executing the computer program.

Images