WO2021223515A1 - 一种监测数据模态改变的方法、装置、设备及存储介质 - Google Patents

一种监测数据模态改变的方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2021223515A1
WO2021223515A1 PCT/CN2021/081021 CN2021081021W WO2021223515A1 WO 2021223515 A1 WO2021223515 A1 WO 2021223515A1 CN 2021081021 W CN2021081021 W CN 2021081021W WO 2021223515 A1 WO2021223515 A1 WO 2021223515A1
Authority
WO
WIPO (PCT)
Prior art keywords
change point
performance data
historical performance
preset
preset historical
Prior art date
Application number
PCT/CN2021/081021
Other languages
English (en)
French (fr)
Inventor
胡文迪
刘建伟
韩静
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2021223515A1 publication Critical patent/WO2021223515A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Definitions

  • This application relates to the field of communication technology, and in particular to a method, device, equipment, and storage medium for monitoring data modal changes.
  • wireless network systems need to ensure continuous and stable operation to meet the communication needs of society in daily life, business, and public services.
  • wireless communication equipment developers have designed many types of performance index data to facilitate network operation and maintenance personnel to monitor the network components and the status of the network.
  • performance index data To present, due to the slow trend change of some performance data, it is difficult to effectively detect the long-term change of the trend through the threshold method. If the algorithm of change point detection is used, it will be affected by the trend and cause false detection.
  • embodiments of the present application provide a method, device, device, and storage medium for monitoring data modal changes.
  • the embodiment of the present application provides a method for monitoring data modal changes, including: obtaining preset historical performance data;
  • the preset historical performance data is bi-periodic data, perform de-periodical preprocessing on the preset historical performance data; perform change point detection on the preset historical performance data to determine the preset historical performance data According to the preset historical performance data in the left and right segments of each of the change points, the trend consistency determination is performed on the change points; wherein the preset historical performance data is divided by the change points Points are divided; the change points that do not meet the trend consistency condition are filtered.
  • An embodiment of the present application provides a device for monitoring data modal changes, including: a data acquisition module configured to acquire preset historical performance data; a de-periodical processing module configured to provide dual Periodic data, which performs de-periodical preprocessing on the preset historical performance data; a change point determination module, is configured to perform change point detection on the preset historical performance data, and determine the value of the preset historical performance data. Change point; a trend consistency determination module configured to perform trend consistency determination on the change point based on the preset historical performance data at the left and right ends of each change point; wherein the preset historical performance data The change point is used as the dividing point for segmentation; the change point filtering module is configured to filter the change points that do not meet the trend consistency condition.
  • An embodiment of the present application provides a device including a processor and a memory; wherein the processor is configured to execute a program stored in the memory to implement any method in the embodiments of the present application.
  • the embodiment of the present application provides a storage medium storing a computer program, where the computer program implements any one of the methods in the embodiments of the present application when the computer program is executed by a processor.
  • the method, device, device, and storage medium for monitoring data modal changes provided in the embodiments of the present application use de-periodical preprocessing of preset historical performance data, detect change points, and identify trends, eliminating data cycles The impact of performance and trend on data modal monitoring, to achieve accurate judgment of modal changes of performance indicators.
  • Figure 1 is a schematic diagram of an application scenario of an embodiment of the application
  • FIG. 2 is a flowchart of a method for monitoring data modal changes provided by an embodiment of the application
  • FIG. 3 is a schematic structural diagram of a device for monitoring data modal changes provided by an embodiment of the application.
  • Figure 4 is a schematic diagram of a device provided by an embodiment of the application.
  • FIG. 1 is a schematic diagram of an application scenario of an embodiment of the present invention, that is, a Long Term Evolution (LTE) wireless network system.
  • LTE Long Term Evolution
  • a typical LTE wireless network system includes multiple base stations, and a base station is logically divided into several cells.
  • Some key performance indicators such as radio resource control (Radio Resource Control, RRC) call drop rate, RRC connection establishment success rate, and uplink and downlink average traffic, reflect the operating status of the network system. Long-term trend or modal monitoring of these performance index data can help operation and maintenance personnel to manage the network.
  • KPI Key Performance Indicator
  • Fig. 2 shows a method for monitoring data modal changes according to an embodiment of the present application, including:
  • S23 Perform change point detection on the preset historical performance data, and determine the change point in the preset historical performance data
  • the preset historical performance data may be historical performance data reflecting the operating state of the wireless network system within a preset time period. These historical performance data are offline data, and change points can be detected through these offline data, and the recorded historical performance data often contains noise interference.
  • the preset time period includes multiple weeks of historical performance data. Set the time period to 30 days.
  • obtaining preset historical performance data includes: obtaining historical performance data within a preset time period; and using a preset filter to perform denoising processing on the historical performance data to obtain the preset historical performance data.
  • a preset filter may be used to perform denoising processing on historical performance data.
  • a hammer filter is used to perform denoising processing on historical performance data to obtain preset historical performance data.
  • performing de-periodical preprocessing on the preset historical performance data includes:
  • DFT Discrete Fourier Transform
  • preset historical performance data is bi-periodic data, calculate the average value of the preset historical performance data on specific days in multiple weeks;
  • the DFT algorithm is used to perform periodic data analysis on the preset historical performance data to perform dual-period detection. If the detection is a dual-period (day period and weekly period) signal, calculate the preset history within a specific day of multiple weeks The average value of performance data.
  • the so-called specific days of the week are Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, and Sunday.
  • the average value of the preset historical performance data for each day of the week is calculated.
  • the preset historical performance data includes multiple weeks of historical performance data, that is, the average value of the preset historical performance data for all Mondays and the average value of the preset historical performance data for all Tuesdays are calculated until the preset history for all Sundays is calculated Mean value of performance data.
  • the average value of the specific day of the corresponding week is subtracted from the preset historical performance data of the specific day of each week. That is, the average value of the preset historical performance data of Monday is subtracted from the preset historical performance data of each Monday, the average value of the preset historical performance data of Tuesday is subtracted from the preset historical performance data of each Tuesday, and so on. , Complete the operation of subtracting the average value of the corresponding specific day of the week from all the preset historical performance data. Since people's work and life in a week vary greatly between working days and rest days, and there are obvious differences in activities at different times of the day, the operation of the wireless network will also be affected by this, which is reflected in the preset historical performance data. De-periodic preprocessing of the preset historical performance data can reduce the influence of rest day data on trend determination.
  • the detecting the change point of the preset historical performance data and determining the change point in the preset historical performance data includes:
  • Scroll reading the preset historical performance data with a preset time step gradually perform change point detection of the preset historical performance data, and record the detected change point and the corresponding change point time;
  • the change point at this change point time is retained, and the remaining change points are filtered.
  • the preset historical performance data can be scrolled by the preset time step S to detect the change point, and record the time of the change point.
  • the change point of the statistics record only when the change point detected in the T step meets a certain number of conditions, the change point is regarded as the real change point in the T step time period, otherwise it is filtered out.
  • one day is set as the preset time step, and the preset number of steps is 7 steps, that is, a 7-day cycle is used for rolling detection.
  • the 7-day cycle is met, the current change point at the same change point is counted
  • the number of change points is greater than or equal to the number of first preset change points. That is, if the number of change points at the same change point time is less than the first preset number of change points, the change point at this change point time will be filtered out.
  • the number of first preset change points may be 3 or 4.
  • the determining the trend consistency of the change points according to the preset historical performance data of the left and right segments of each of the change points includes:
  • the linear regression mean square error of the unit length statistic in the left segment of the change point and the linear regression mean square error of the unit length statistic in the right segment of the change point are both less than the first threshold, and the linear regression of the unit length statistic across the left and right segments of the change point is both
  • the square error is less than the double of the largest of the linear regression mean square error of the unit length statistic of the left segment of the change point and the linear regression mean square error of the unit length statistic of the right segment of the change point, and it is determined that the change point meets the trend consistency condition.
  • the preset historical performance data is divided by taking the change point as the dividing point, so that segmented preset historical performance data will be obtained.
  • the preset historical performance data of the left segment of the current change point is the preset historical performance data from the adjacent change point on the left to the current change point
  • the preset historical performance data of the right segment of the current change point is the adjacent change from the right
  • the preset historical performance data between the point and the current change point For each change point, calculate the linear regression (Liner regression, LR) average of the unit length statistic, respectively Square error (mean square error, MSE)lr_mse:
  • y j is the value on the mean line
  • Is the value on the regression fitted line
  • len(bkp 2 -bkp 1 ) is the distance of the horizontal axis of the current fitted line.
  • the change point is judged to be a pseudo change point caused by trend influence, and it is filtered.
  • the first threshold TH is an empirically set threshold, such as 0.01.
  • the detecting the change point of the preset historical performance data and determining the change point in the preset historical performance data includes:
  • Scroll reading the preset historical performance data with a preset time step gradually perform change point detection of the preset historical performance data, and record the detected change point and the corresponding change point time;
  • the change point at this change point time is retained, and the remaining change points are filtered.
  • the preset historical performance data can be scrolled by the preset time step S to detect the change point, and record the time of the change point.
  • the change point of the statistics record only when the change point detected in the T step meets a certain number of conditions, the change point is regarded as the real change point in the T step time period, otherwise it is filtered out.
  • one day is set as the preset time step, and the preset number of steps is 7 steps, that is, a 7-day cycle is used for rolling detection.
  • the 7-day cycle is met, the current change point at the same change point is counted
  • the number of change points is greater than or equal to the second preset number of change points.
  • the number of second preset change points may be 3 or 4.
  • the determining the trend consistency of the change points according to the preset historical performance data of the left and right segments of each of the change points includes:
  • linear regression mean square error of the unit length statistic of the left segment of the change point and the unit length statistic linear regression mean square error of the unit length statistic of the right segment of the change point are both less than the second threshold, and the linear regression of the unit length statistic across the left and right segments of the change point is both
  • the square error is less than the double of the largest of the linear regression mean square error of the unit length statistic of the left segment of the change point and the linear regression mean square error of the unit length statistic of the right segment of the change point, and it is determined that the change point meets the trend consistency condition.
  • y j is the original data value
  • len(bkp 2 -bkp 1 ) is the distance of the horizontal axis of the current fitted line.
  • bkp 1 is the current change point
  • bkp 2 is the first change point from the left of the current change point
  • Bkp 1 is the first change point from the right of the current change point
  • bkp 2 is the first change point from the current change point
  • bkp 1 is the first change point from the right of the current change point
  • bkp 1 is the first change point from the right of the current change point.
  • the second threshold TH is an empirical setting.
  • Set threshold such as 0.02.
  • Fig. 3 shows a device for monitoring data modal changes according to an embodiment of the present application, including:
  • the data acquisition module 31 is configured to acquire preset historical performance data
  • the de-periodical processing module 32 is configured to perform de-periodical preprocessing on the preset historical performance data if the preset historical performance data is bi-periodic data;
  • the change point determination module 33 is configured to perform change point detection on the preset historical performance data, and determine the change point in the preset historical performance data;
  • the trend consistency determination module 34 is configured to determine the trend consistency of the change points according to the preset historical performance data in the left and right segments of each of the change points; wherein, the preset historical performance data is based on the preset historical performance data.
  • the change point is the dividing point for division;
  • the change point filtering module 35 is configured to filter the change points that do not meet the trend consistency condition.
  • the data acquisition module 31 includes:
  • the historical performance data obtaining unit is configured to obtain historical performance data within a preset time period
  • the noise filtering unit is configured to use a preset filter to perform denoising processing on the performance history data to obtain the preset history performance data.
  • the de-periodic processing module 32 includes:
  • the periodic data determining unit is configured to perform periodic data analysis on the preset historical performance data based on the Discrete Fourier Transform DFT algorithm to determine whether the preset historical performance data is dual periodic data; wherein, the dual period
  • the sexual data is the data of the day period and the week period;
  • the data average value calculation unit for specific days of the week is configured to calculate the average value of the preset historical performance data on specific days of multiple weeks if the preset historical performance data is bi-periodic data;
  • the data de-period unit is configured to subtract the corresponding average value from the preset historical performance data on a specific day of each week.
  • the change point determination module 33 is specifically set to:
  • Scroll reading the preset historical performance data with a preset time step gradually perform change point detection of the preset historical performance data, and record the detected change point and the corresponding change point time;
  • the change point at this change point time is retained, and the remaining change points are filtered.
  • the trend consistency determination module 34 is specifically configured to:
  • the linear regression mean square error of the unit length statistic in the left segment of the change point and the linear regression mean square error of the unit length statistic in the right segment of the change point are both less than the first threshold, and the linear regression of the unit length statistic across the left and right segments of the change point is both
  • the square error is less than the double of the largest of the linear regression mean square error of the unit length statistic of the left segment of the change point and the linear regression mean square error of the unit length statistic of the right segment of the change point, and it is determined that the change point meets the trend consistency condition.
  • the change point determination module 33 is specifically set to:
  • Scroll reading the preset historical performance data with a preset time step gradually perform change point detection of the preset historical performance data, and record the detected change point and the corresponding change point time;
  • the change point at this change point time is retained, and the remaining change points are filtered.
  • the trend consistency determination module 34 is specifically configured to:
  • linear regression mean square error of the unit length statistic of the left segment of the change point and the unit length statistic linear regression mean square error of the unit length statistic of the right segment of the change point are both less than the second threshold, and the linear regression of the unit length statistic across the left and right segments of the change point is both
  • the square error is less than the double of the largest of the linear regression mean square error of the unit length statistic of the left segment of the change point and the linear regression mean square error of the unit length statistic of the right segment of the change point, and it is determined that the change point meets the trend consistency condition.
  • an embodiment of the present application provides a device, the device includes a processor 41 and a memory 42;
  • the processor 41 is configured to execute a program stored in the memory 42 to implement any method in the embodiments of the present application.
  • the embodiment of the present application provides a storage medium that stores a computer program, and when the computer program is executed by a processor, any one of the methods in the embodiments of the present application is implemented.
  • the present application provides a method, device, equipment, and storage medium for monitoring data modal changes, which are used to monitor the modal changes of performance indicators and accurately judge the modal changes of performance indicator data.
  • user terminal encompasses any suitable type of wireless user equipment, such as a mobile phone, a portable data processing device, a portable web browser, or a vehicle-mounted mobile station.
  • the various embodiments of the present application can be implemented in hardware or dedicated circuits, software, logic or any combination thereof.
  • some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device, although the present application is not limited thereto.
  • Computer program instructions can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code written in any combination of one or more programming languages or Object code.
  • ISA instruction set architecture
  • the block diagram of any logic flow in the drawings of the present application may represent program steps, or may represent interconnected logic circuits, modules, and functions, or may represent a combination of program steps and logic circuits, modules, and functions.
  • the computer program can be stored on the memory.
  • the memory can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as but not limited to read only memory (ROM), random access memory (RAM), optical storage devices and systems (digital multi-function optical discs) DVD or CD) etc.
  • Computer-readable media may include non-transitory storage media.
  • the data processor can be any type suitable for the local technical environment, such as but not limited to general-purpose computers, special-purpose computers, microprocessors, digital signal processors (DSP), application-specific integrated circuits (ASIC), programmable logic devices (FGPA) And processors based on multi-core processor architecture.
  • DSP digital signal processors
  • ASIC application-specific integrated circuits
  • FGPA programmable logic devices

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Debugging And Monitoring (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

一种监测数据模态改变的方法、装置、设备及存储介质，该方法包括：获取预设历史性能数据(S21)；若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理(S22)；对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点(S23)；根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定(S24)；其中，所述预设历史性能数据以所述改变点为分割点进行分割；将不满足趋势一致性条件的所述改变点予以过滤(S25)。

Description

一种监测数据模态改变的方法、装置、设备及存储介质 技术领域
本申请涉及通讯技术领域，具体涉及一种监测数据模态改变的方法、装置、设备及存储介质。
背景技术
作为信息时代重要的基础设施，无线网络系统需保证持续稳定的运行，以满足社会日常、商业及公共服务等各方面的通讯需要。为此，无线通信设备开发人员设计了众多种类的性能指标数据，方便网络运维人员监控网络各组件以及网络的状态。目前，由于部分性能数据的趋势变化缓慢，难以有效地通过阈值方法检测出趋势的长期变化。如果使用改变点检测的算法，又会被趋势影响产生误检。
发明内容
有鉴于此，本申请实施例提供一种监测数据模态改变的方法、装置、设备及存储介质。
本申请实施例提供了一种监测数据模态改变的方法，包括：获取预设历史性能数据；
若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理；对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点；根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定；其中，所述预设历史性能数据以所述改变点为分割点进行分割；将不满足趋势一致性条件的所述改变点予以过滤。
本申请实施例提供一种监测数据模态改变的装置，包括：数据获取模块，被设置成获取预设历史性能数据；去周期性处理模块，被设置成若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理；改变点确定模块，被设置成对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点；趋势一致性判定模块，被设置成根据每个所述改变点左右两端的所述预设历史性能数据，对所述改变点进行趋势一致性判定；其中，所述预设历史性能数据以所述改变点为分割点进行分割；改变点过滤模块，被设置成将不满足趋势一致性条件的所述改变点予以过滤。
本申请实施例提供一种设备，包括处理器以及存储器；其中，所述处理器被设置成执行存储器中存储的程序，以实现本申请实施例中的任意一种方法。
本申请实施例提供了一种存储介质，存储有计算机程序，其中，所述计算机程序被处理器执行时实现本申请实施例中的任意一种方法。
本申请实施例所提供的监测数据模态改变的方法、装置、设备及存储介质，通过对预设历史性能数据的去周期性预处理，检测改变点，并进行趋势的识别，排除了数据周期性以及趋势对数据模态监测的影响，实现准确判断性能指标模态变化。
附图说明
图1为本申请的实施例运用场景示意图；
图2为本申请实施例提供的监测数据模态改变的方法的流程图；
图3为本申请实施例提供的监测数据模态改变的装置的结构示意图；
图4为本申请实施例提供的设备的示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚明白，下文中将结合附图对本申请的实施例进行详细说明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。
图1是本发明实施例的一个运用场景示意图，即长期演进(Long Term Evolution，LTE)无线网络系统。如图1所示，一个典型的LTE无线网络系统包含有多个基站，一个基站又按逻辑划分成几个小区。每个小区下有多个手机用户接入，进行无线通话、上网、看视频等移动通信活动。一些关键性能指标(Key Performance Indicator，KPI)，例如无线资源控制(Radio Resource Control，RRC)掉话率、RRC连接建立成功率和上下行平均流量等，反映了网络系统的运行状态。通过对这些性能指标数据进行长期趋势或模态监测，可以帮助运维人员对网络进行管理。
图2示出根据本申请实施例的监测数据模态改变的方法，包括：
S21、获取预设历史性能数据；
S22、若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理；
S23、对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点；
S24、根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定；其中，所述预设历史性能数据以所述改变点为分割点进行分割；
S25、将不满足趋势一致性条件的所述改变点予以过滤。
其中，预设历史性能数据可以是预设时间段内，反映无线网络系统的运行状态的历史性能数据。这些历史性能数据为离线数据，通过这些离线数据可以进行检测改变点，而被记录下的历史性能数据中往往会存在噪声干扰，预设时间段包括多周的历史性能数据，示例性地，预设时间段为30天。在一种实现方式中，获取预设历史性能数据，包括：获取 预设时间段内的历史性能数据；采用预设滤波器对所述历史性能数据进行去噪处理，得到所述预设历史性能数据。可以采用预设滤波器对历史性能数据进行去噪处理，在一些实例中，采用hamper filter对历史性能数据进行去噪处理，得到预设历史性能数据。
在一种实现方式中，所述若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理，包括：
基于离散傅立叶变换(Discrete Fourier transform，DFT)算法对所述预设历史性能数据进行周期性数据分析，确定所述预设历史性能数据是否为双周期性数据；其中，所述双周期性数据为天周期和周周期的数据；
若所述预设历史性能数据为双周期性数据，计算出多个周特定天内的所述预设历史性能数据的均值；
将每个周特定天内的所述预设历史性能数据减去对应的所述均值。
其中，对预设历史性能数据利用DFT算法进行周期性数据分析，来进行双周期检测，如果检测为双周期(天周期和周周期)信号，计算出多个周特定天内的所述预设历史性能数据的均值，所谓周特定天就是周一、周二、周三、周四、周五、周六和周日，计算出在一周中每一天的预设历史性能数据的均值。预设历史性能数据中包括多周的历史性能数据，也就是，计算全部周一的预设历史性能数据的均值，全部周二的预设历史性能数据的均值，直到计算出全部周日的预设历史性能数据的均值。然后，在每个周特定天的预设历史性能数据中减去对应周特定天的均值。也就是，每个周一的预设历史性能数据中减去周一的预设历史性能数据的均值，每个周二的预设历史性能数据中减去周二的预设历史性能数据的均值，以此类推，完成所有预设历史性能数据中减去对应周特定天的均值的操作。由于人们每周的工作生活在工作日和休息日差别较大，每天中不同时段的活动情况也有明显区别，那么，无线网络的运行也会受此影响，体现在预设历史性能数据上。对所述预设历史性能数据进行去周期性预处理，可以减少休息日数据对趋势判定的影响。
在一种实现方式中，所述对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点，包括：
以预设时间步长滚动读取所述预设历史性能数据，逐步进行所述预设历史性能数据的改变点检测，记录检测到的所述改变点和对应的改变点时刻；
若所述改变点检测的步数达到预设步数，将所述预设步数内检测出的同一所述改变点时刻的改变点数量与第一预设改变点数量进行比较；
若同一所述改变点时刻的所述改变点数量大于等于所述第一预设改变点数量，将这一所述改变点时刻的所述改变点予以保留，其余所述改变点予以过滤。
其中，可以以预设时间步长S滚动读取预设历史性能数据进行改变点检测，并记录改变点时刻。当经过预设步数T后，统计记录的改变点，只有当T步内检测出的改变点满足 一定数量条件，才将此改变点作为T步时间段内真正的改变点，否则滤除。本实现方式中设定以一天为预设时间步长，预设步数为7步，也就是以7天为一个周期滚动检测，当满足7天周期时，统计当前相同改变点时刻的改变点数量大于等于第一预设改变点数量的改变点。也就是，同一改变点时刻的改变点数量如果小于第一预设改变点数量，这个改变点时刻的改变点就会被滤除。示例性地，第一预设改变点数量可以为3或4。
在一种实现方式中，所述根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定，包括：
以所述改变点为分割点，对所述预设历史性能数据进行分段；
计算每段所述预设历史性能数据的移动均值，其中，周期为天周期数据数；
对每个所述改变点的改变点左段移动均值、改变点右段移动均值和跨改变点左右段移动均值的均值线分别求线性回归，计算对应的单位长度统计量线性回归均方误差；
若改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差均小于第一阈值，且跨改变点左右段的单位长度统计量线性回归均方误差小于改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差中最大者的双倍，确定所述改变点满足所述趋势一致性条件。
其中，所述预设历史性能数据以所述改变点为分割点进行分割，这样会得到分段的预设历史性能数据。当前改变点左段的预设历史性能数据是从左侧相邻改变点到当前改变点之间的预设历史性能数据，当前改变点右段的预设历史性能数据是从右侧相邻改变点到当前改变点之间的预设历史性能数据。对每个改变点，分别对改变点左段移动均值、改变点右段移动均值及跨改变点左右段移动均值的均值线求线性回归，计算单位长度统计量线性回归(Liner regression，LR)均方误差(mean square error，MSE)lr_mse:
Figure PCTCN2021081021-appb-000001
其中，y j为均值线上的值，
Figure PCTCN2021081021-appb-000002
为回归拟合线上的值，len(bkp 2-bkp 1)为当前拟合线的横轴的距离。具体来说，当计算改变点左段lr_mse_left时，bkp 1为当前改变点，bkp 2为当前改变点左起的第一个改变点；当计算改变点右段lr_mse_rigiht时，bkp 2是当前改变点，bkp 1为当前改变点右起的第一个改变点；当计算跨改变点左右段lr_mse_cross时，bkp 2为当前改变点做起第一个改变点，bkp 1为当前改变点右起的第一个改变点。
对以上所述的改变点的趋势一致性进行判定，具体如下：
当满足lr_mse_left<TH and lr_mse_rigiht<TH and lr_mse_cross<2*MAX(lr_mse_left,lr_mse_rigiht)时，判断该改变点为因趋势影响产生的伪改变点，对其进行过滤。其中，第一阈值TH为经验设定的阈值，例如0.01。
计算每段预设历史性能数据的移动均值，并根据改变点左段移动均值、改变点右段移动均值和跨改变点左右段移动均值的单位长度统计量线性回归均方误差判断改变点的趋 势一致性，对于数据改变趋势的幅度持续增加的情况，过滤改变点的效果更佳，数据模态改变的判断更准确。
在一种实现方式中，所述对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点，包括：
以预设时间步长滚动读取所述预设历史性能数据，逐步进行所述预设历史性能数据的改变点检测，记录检测到的所述改变点和对应的改变点时刻；
若所述改变点检测的步数达到预设步数，将所述预设步数内检测出的同一所述改变点时刻的连续改变点数量与第二预设改变点数量进行比较；
若同一所述改变点时刻的所述连续改变点数量大于等于所述第二预设改变点数量，将这一所述改变点时刻的所述改变点予以保留，其余所述改变点予以过滤。
其中，可以以预设时间步长S滚动读取预设历史性能数据进行改变点检测，并记录改变点时刻。当经过预设步数T后，统计记录的改变点，只有当T步内检测出的改变点满足一定数量条件，才将此改变点作为T步时间段内真正的改变点，否则滤除。本实现方式中设定以一天为预设时间步长，预设步数为7步，也就是以7天为一个周期滚动检测，当满足7天周期时，统计当前相同改变点时刻的改变点数量大于等于第二预设改变点数量的改变点。也就是，判断同一改变点时刻的改变点是否连续出现，并且连续出现的次数大于等于第二预设改变点数量，如果是确定为改变点，否则过滤。示例性地，第二预设改变点数量可以为3或4。若所述预设步数内检测出同一改变点时刻连续出现改变点，且连续的改变点的数量大于等于第二预设改变点数量，保留这一改变点时刻连续出现改变点；若所述预设步数内未检测出同一改变点时刻连续出现改变点，滤除这一改变点时刻非连续出现的改变点；或者，预设步数内检测出同一改变点时刻连续出现改变点，但连续出现的改变点的数量小于第二预设改变点数量，滤除这一改变点时刻连续出现的改变点。
在一种实现方式中，所述根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定，包括：
以所述改变点为分割点，对所述预设历史性能数据进行分段；
对每个所述改变点的改变点左段数据、改变点右段数据和跨改变点左右段数据的数据线分别求线性回归，计算对应的单位长度统计量线性回归均方误差；
若改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差均小于第二阈值，且跨改变点左右段的单位长度统计量线性回归均方误差小于改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差中最大者的双倍，确定所述改变点满足所述趋势一致性条件。
对各改变点，分别对改变点左段数据、改变点右段数据及跨改变点左右段数据求线性回归线，计算单位长度统计量线性回归均方误差lr_mse:
Figure PCTCN2021081021-appb-000003
其中，y j为原数据值，
Figure PCTCN2021081021-appb-000004
为回归拟合线上的值，len(bkp 2-bkp 1)为当前拟合线的横轴的距离。具体来说，当计算改变点左段lr_mse_left时，bkp 1为当前改变点，bkp 2为当前改变点左起的第一个改变点；当计算改变点右段lr_mse_rigiht时，bkp 2是当前改变点，bkp 1为当前改变点右起的第一个改变点；当计算跨改变点左右段lr_mse_cross时，bkp 2为当前改变点做起第一个改变点，bkp 1为当前改变点右起的第一个改变点。
对以上所述的改变点的趋势一致性进行判定，具体如下：
当满足lr_mse_left<TH and lr_mse_rigiht<TH and lr_mse_cross<2*MAX(lr_mse_left,lr_mse_rigiht)时，判断该改变点为因趋势影响产生的伪改变点，对其进行过滤，其中，第二阈值TH为经验设定的阈值，例如0.02。
根据改变点左段数据、改变点右段数据和跨改变点左右段数据的单位长度统计量线性回归均方误差判断改变点的趋势一致性，对于数据改变趋势的幅度保存稳定的情况，过滤改变点的效果更佳，数据模态改变的判断更准确。
图3示出根据本申请实施例的监测数据模态改变的装置，包括：
数据获取模块31，被设置成获取预设历史性能数据；
去周期性处理模块32，被设置成若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理；
改变点确定模块33，被设置成对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点；
趋势一致性判定模块34，被设置成根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定；其中，所述预设历史性能数据以所述改变点为分割点进行分割；
改变点过滤模块35，被设置成将不满足趋势一致性条件的所述改变点予以过滤。
在一种实施方式中，数据获取模块31，包括：
历史性能数据获取单元，被设置成获取预设时间段内的历史性能数据；
滤噪单元，被设置成采用预设滤波器对所述性能历史数据进行去噪处理，得到所述预设历史性能数据。
在一种实施方式中，去周期性处理模块32包括：
周期性数据确定单元，被设置成基于离散傅立叶变换DFT算法对所述预设历史性能数据进行周期性数据分析，确定所述预设历史性能数据是否为双周期性数据；其中，所述双周期性数据为天周期和周周期的数据；
周特定天数据均值计算单元，被设置成若所述预设历史性能数据为双周期性数据，计算出多个周特定天内的所述预设历史性能数据的均值；
数据去周期单元，被设置成将每个周特定天内的所述预设历史性能数据减去对应的所述均值。
在一种实现方式中，改变点确定模块33具体被设置成：
以预设时间步长滚动读取所述预设历史性能数据，逐步进行所述预设历史性能数据的改变点检测，记录检测到的所述改变点和对应的改变点时刻；
若所述改变点检测的步数达到预设步数，将所述预设步数内检测出的同一所述改变点时刻的改变点数量与第一预设改变点数量进行比较；
若同一所述改变点时刻的所述改变点数量大于等于所述第一预设改变点数量，将这一所述改变点时刻的所述改变点予以保留，其余所述改变点予以过滤。
在一种实现方式中，趋势一致性判定模块34具体被设置成：
以所述改变点为分割点，对所述预设历史性能数据进行分段；
计算每段所述预设历史性能数据的移动均值，其中，周期为天周期数据数；
对每个所述改变点的改变点左段移动均值、改变点右段移动均值和跨改变点左右段移动均值的均值线分别求线性回归，计算对应的单位长度统计量线性回归均方误差；
若改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差均小于第一阈值，且跨改变点左右段的单位长度统计量线性回归均方误差小于改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差中最大者的双倍，确定所述改变点满足所述趋势一致性条件。
在一种实现方式中，改变点确定模块33具体被设置成：
以预设时间步长滚动读取所述预设历史性能数据，逐步进行所述预设历史性能数据的改变点检测，记录检测到的所述改变点和对应的改变点时刻；
若所述改变点检测的步数达到预设步数，将所述预设步数内检测出的同一所述改变点时刻的连续改变点数量与第二预设改变点数量进行比较；
若同一所述改变点时刻的所述连续改变点数量大于等于所述第二预设改变点数量，将这一所述改变点时刻的所述改变点予以保留，其余所述改变点予以过滤。
在一种实现方式中，趋势一致性判定模块34具体被设置成：
以所述改变点为分割点，对所述预设历史性能数据进行分段；
对每个所述改变点的改变点左段数据、改变点右段数据和跨改变点左右段数据的数据线分别求线性回归，计算对应的单位长度统计量线性回归均方误差；
若改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差均小于第二阈值，且跨改变点左右段的单位长度统计量线性回归均方误差小于改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差中最大者的双倍，确定所述改变点满足所述趋势一致性条件。
参照图4，本申请实施例提供一种设备，所述设备包括处理器41以及存储器42；
所述处理器41被设置成执行存储器42中存储的程序，以实现本申请实施例中的任意一种方法。
本申请实施例提供了一种存储介质，所述存储介质存储有计算机程序，所述计算机程序被处理器执行时实现本申请实施例中的任意一种方法。
本申请提供一种监测数据模态改变的方法、装置、设备及存储介质，用以实现监测性能指标模态变化，并准确判断性能指标数据模态变化。
以上所述，仅为本申请的示例性实施例而已，并非用于限定本申请的保护范围。
本领域内的技术人员应明白，术语用户终端涵盖任何适合类型的无线用户设备，例如移动电话、便携数据处理装置、便携网络浏览器或车载移动台。
一般来说，本申请的多种实施例可以在硬件或专用电路、软件、逻辑或其任何组合中实现。例如，一些方面可以被实现在硬件中，而其它方面可以被实现在可以被控制器、微处理器或其它计算装置执行的固件或软件中，尽管本申请不限于此。
本申请的实施例可以通过移动装置的数据处理器执行计算机程序指令来实现，例如在处理器实体中，或者通过硬件，或者通过软件和硬件的组合。计算机程序指令可以是汇编指令、指令集架构(ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码。
本申请附图中的任何逻辑流程的框图可以表示程序步骤，或者可以表示相互连接的逻辑电路、模块和功能，或者可以表示程序步骤与逻辑电路、模块和功能的组合。计算机程序可以存储在存储器上。存储器可以具有任何适合于本地技术环境的类型并且可以使用任何适合的数据存储技术实现，例如但不限于只读存储器(ROM)、随机访问存储器(RAM)、光存储器装置和系统(数码多功能光碟DVD或CD光盘)等。计算机可读介质可以包括非瞬时性存储介质。数据处理器可以是任何适合于本地技术环境的类型，例如但不限于通用计算机、专用计算机、微处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、可编程逻辑器件(FGPA)以及基于多核处理器架构的处理器。
通过示范性和非限制性的示例，上文已提供了对本申请的示范实施例的详细描述。但结合附图和权利要求来考虑，对以上实施例的多种修改和调整对本领域技术人员来说是显而易见的，但不偏离本发明的范围。因此，本发明的恰当范围将根据权利要求确定。

Claims (10)

  1. 一种监测数据模态改变的方法，包括：
    获取预设历史性能数据；
    若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理；
    对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点；
    根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定；其中，所述预设历史性能数据以所述改变点为分割点进行分割；
    将不满足趋势一致性条件的所述改变点予以过滤。
  2. 根据权利要求1所述的方法，其中，所述获取预设历史性能数据，包括：
    获取预设时间段内的历史性能数据；
    采用预设滤波器对所述性能历史数据进行去噪处理，得到所述预设历史性能数据。
  3. 根据权利要求1所述的方法，其中，所述若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理，包括：
    基于离散傅立叶变换DFT算法对所述预设历史性能数据进行周期性数据分析，确定所述预设历史性能数据是否为双周期性数据；其中，所述双周期性数据为天周期和周周期的数据；
    若所述预设历史性能数据为双周期性数据，计算出多个周特定天内的所述预设历史性能数据的均值；
    将每个周特定天内的所述预设历史性能数据减去对应的所述均值。
  4. 根据权利要求1所述的方法，其中，所述对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点，包括：
    以预设时间步长滚动读取所述预设历史性能数据，逐步进行所述预设历史性能数据的改变点检测，记录检测到的所述改变点和对应的改变点时刻；
    若所述改变点检测的步数达到预设步数，将所述预设步数内检测出的同一所述改变点时刻的改变点数量与第一预设改变点数量进行比较；
    若同一所述改变点时刻的所述改变点数量大于等于所述第一预设改变点数量，将这一所述改变点时刻的所述改变点予以保留，其余所述改变点予以过滤。
  5. 根据权利要求4所述的方法，其中，所述根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定，包括：
    以所述改变点为分割点，对所述预设历史性能数据进行分段；
    计算每段所述预设历史性能数据的移动均值，其中，周期为天周期数据数；
    对每个所述改变点的改变点左段移动均值、改变点右段移动均值和跨改变点左右段移动均值的均值线分别求线性回归，计算对应的单位长度统计量线性回归均方误差；
    若改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差均小于第一阈值，且跨改变点左右段的单位长度统计量线性回归均方误差小于改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差中最大者的双倍，确定所述改变点满足所述趋势一致性条件。
  6. 根据权利要求1所述的方法，其中，所述对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点，包括：
    以预设时间步长滚动读取所述预设历史性能数据，逐步进行所述预设历史性能数据的改变点检测，记录检测到的所述改变点和对应的改变点时刻；
    若所述改变点检测的步数达到预设步数，将所述预设步数内检测出的同一所述改变点时刻的连续改变点数量与第二预设改变点数量进行比较；
    若同一所述改变点时刻的所述连续改变点数量大于等于所述第二预设改变点数量，将这一所述改变点时刻的所述改变点予以保留，其余所述改变点予以过滤。
  7. 根据权利要求6所述的方法，其中，所述根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定，包括：
    以所述改变点为分割点，对所述预设历史性能数据进行分段；
    对每个所述改变点的改变点左段数据、改变点右段数据和跨改变点左右段数据的数据线分别求线性回归，计算对应的单位长度统计量线性回归均方误差；
    若改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差均小于第二阈值，且跨改变点左右段的单位长度统计量线性回归均方误差小于改变点左段的单位长度统计量线性回归均方误差和改变点右段的单位长度统计量线性回归均方误差中最大者的双倍，确定所述改变点满足所述趋势一致性条件。
  8. 一种监测数据模态改变的装置，包括：
    数据获取模块，被设置成获取预设历史性能数据；
    去周期性处理模块，被设置成若所述预设历史性能数据为双周期性数据，对所述预设历史性能数据进行去周期性预处理；
    改变点确定模块，被设置成对所述预设历史性能数据进行改变点检测，确定所述预设历史性能数据中的改变点；
    趋势一致性判定模块，被设置成根据每个所述改变点左右段的所述预设历史性能数据，对所述改变点进行趋势一致性判定；其中，所述预设历史性能数据以所述改变点为分割点 进行分割；
    改变点过滤模块，被设置成将不满足趋势一致性条件的所述改变点予以过滤。
  9. 一种设备，包括处理器以及存储器；其中，
    所述处理器被设置成执行存储器中存储的程序，以实现权利要求1-7任一项所述的方法。
  10. 一种存储介质，存储有计算机程序，其中，所述计算机程序被处理器执行时实现权利要求1-7任一项所述的方法。
PCT/CN2021/081021 2020-05-06 2021-03-16 一种监测数据模态改变的方法、装置、设备及存储介质 WO2021223515A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010373317.5 2020-05-06
CN202010373317.5A CN113627696A (zh) 2020-05-06 2020-05-06 一种监测数据模态改变的方法、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2021223515A1 true WO2021223515A1 (zh) 2021-11-11

Family

ID=78376609

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/081021 WO2021223515A1 (zh) 2020-05-06 2021-03-16 一种监测数据模态改变的方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN113627696A (zh)
WO (1) WO2021223515A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102346745A (zh) * 2010-08-02 2012-02-08 阿里巴巴集团控股有限公司 单词的用户行为数的预测方法和装置
WO2013035344A1 (ja) * 2011-09-08 2013-03-14 国立大学法人東京工業大学 時系列データの変化点検出方法及びプログラム、未来の時系列データ値の確率密度分布予測方法及びプログラム
CN103678514A (zh) * 2013-11-26 2014-03-26 安徽科大讯飞信息科技股份有限公司 一种业务趋势预测方法及系统
US9692674B1 (en) * 2013-12-30 2017-06-27 Google Inc. Non-parametric change point detection
CN108120796A (zh) * 2017-11-20 2018-06-05 太原鹏跃电子科技有限公司 一种检测铁路蓄电池co32-时ph值突变点的测定计算方法
CN109828548A (zh) * 2019-01-17 2019-05-31 西安交通大学 基于时间序列变化突变点检测的性能退化特征评估方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102346745A (zh) * 2010-08-02 2012-02-08 阿里巴巴集团控股有限公司 单词的用户行为数的预测方法和装置
WO2013035344A1 (ja) * 2011-09-08 2013-03-14 国立大学法人東京工業大学 時系列データの変化点検出方法及びプログラム、未来の時系列データ値の確率密度分布予測方法及びプログラム
CN103678514A (zh) * 2013-11-26 2014-03-26 安徽科大讯飞信息科技股份有限公司 一种业务趋势预测方法及系统
US9692674B1 (en) * 2013-12-30 2017-06-27 Google Inc. Non-parametric change point detection
CN108120796A (zh) * 2017-11-20 2018-06-05 太原鹏跃电子科技有限公司 一种检测铁路蓄电池co32-时ph值突变点的测定计算方法
CN109828548A (zh) * 2019-01-17 2019-05-31 西安交通大学 基于时间序列变化突变点检测的性能退化特征评估方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI, ZHEHAO: "Non-Stationary Identification of Runoff Series and Driving Mechanism Analysis in Wuding River Basin", CHINA MASTER’S THESES FULL-TEXT DATABASE, no. 1, 15 January 2020 (2020-01-15), pages 1 - 107, XP055864478, ISSN: 1674-0246 *

Also Published As

Publication number Publication date
CN113627696A (zh) 2021-11-09

Similar Documents

Publication Publication Date Title
JP7451479B2 (ja) コンピューティング・デバイスについてのシステム性能およびイベント・データを収集、追跡、および記憶するためのシステムおよび方法
CN107871190B (zh) 一种业务指标监控方法及装置
US10558544B2 (en) Multiple modeling paradigm for predictive analytics
ES2950039T3 (es) Identificación automatizada del estado de un dispositivo y modificación dinámica resultante de las operaciones del dispositivo
US10379717B2 (en) Device based visualization and analysis of multivariate data
US20180285320A1 (en) User-level kqi anomaly detection using markov chain model
US20180039895A1 (en) Data predicting method and apparatus
US11652699B2 (en) Computerized system and method for an improved self organizing network
CN110889597A (zh) 业务时序指标异常检测方法及装置
WO2021223515A1 (zh) 一种监测数据模态改变的方法、装置、设备及存储介质
CN114116128B (zh) 容器实例的故障诊断方法、装置、设备和存储介质
CN107566187B (zh) 一种sla违例监测方法、装置和系统
CN113297039A (zh) 数据监控方法及装置
US20230333958A1 (en) Diagnosis and troubleshooting of devices using multiple layer architecture
CN110351531B (zh) 视频大数据云平台视频质量诊断服务方法
EP3944590A1 (en) Apparatus and a method for processing time series data
US20230231778A1 (en) Computerized system and method for an improved self organizing network
CN116701491A (zh) 运行指标静态阈值确定方法、装置、存储介质及处理器
CN112200461A (zh) 关键路径的预测方法及装置、电子设备、计算机存储介质
WO2015119607A1 (en) Resource management
CN111367774A (zh) 检测方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21800767

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 05.04.2023

122 Ep: pct application non-entry in european phase

Ref document number: 21800767

Country of ref document: EP

Kind code of ref document: A1