WO2024011745A1 - 一种基于油田网电设备的故障预警方法 - Google Patents

一种基于油田网电设备的故障预警方法 Download PDF

Info

Publication number
WO2024011745A1
WO2024011745A1 PCT/CN2022/118701 CN2022118701W WO2024011745A1 WO 2024011745 A1 WO2024011745 A1 WO 2024011745A1 CN 2022118701 W CN2022118701 W CN 2022118701W WO 2024011745 A1 WO2024011745 A1 WO 2024011745A1
Authority
WO
WIPO (PCT)
Prior art keywords
fault
warning
data
interval
historical
Prior art date
Application number
PCT/CN2022/118701
Other languages
English (en)
French (fr)
Inventor
杨华
柴世奇
覃鸿
王超
Original Assignee
江苏辰午节能科技股份有限公司
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 江苏辰午节能科技股份有限公司 filed Critical 江苏辰午节能科技股份有限公司
Publication of WO2024011745A1 publication Critical patent/WO2024011745A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Definitions

  • the invention relates to the field of computer fault early warning methods, and in particular, to a fault early warning method for oil field network power equipment.
  • Equipment failure early warning technology has a beneficial effect on extending the service life of oilfield power grid equipment, and can provide early warning of poor equipment status before a failure occurs.
  • Oil drilling sites are usually in inaccessible outdoor places. At present, oil drilling often uses grid-connected drive motors as the main power drive equipment to extract oil.
  • On-site grid power conditions often face long wiring distances, unstable grid voltage, and large loads. , large harmonics and other problems. Long-term operation under such working conditions will lead to heavy loads on grid power equipment and prone to failure and loss. Therefore, it is very necessary and urgent to design an efficient fault early warning method based on oilfield grid power equipment. .
  • the purpose of the present invention is to provide a fault early warning method based on oilfield power grid equipment in view of the shortcomings of the existing technology.
  • a fault early warning method based on oilfield power grid equipment including the following steps:
  • step (1) includes the following sub-steps:
  • (1.1) Define parameters: According to the fault warning requirements of oilfield power grid equipment, extract the core data set G e for fault prediction, define the value of each element in the core data set G e when a fault occurs as G en , and define each element
  • the real-time data of the element is defined as G rn
  • the safe interval of each element in the core data is defined as G sn
  • the number of fault occurrences is defined as N
  • the number of times is defined as G bn
  • the early warning accuracy is defined as um
  • the interval where the fault warning is triggered is defined as G umn
  • the probability of each element in the model triggering a fault is defined as G tn , n is 1, 2, 3, 4... sequence;
  • the step (2) is specifically: counting the number of historical faults into N, and clearing the number G bn of the element values G en in the core data exceeding the safe interval G sn in the N faults.
  • the value of the fault warning trigger interval G umn is obtained by multiplying the upper and lower limits of the safety interval G sn by the warning accuracy um.
  • the value of the failure probability G tn of each element is given by G bn Divided by the number of historical faults, a fault warning model based on actual historical faults is obtained.
  • the beneficial effect of the present invention is that by constructing a fault warning algorithm and model for oilfield power grid equipment, the invention can effectively protect the oilfield power grid equipment, predict the dangerous behavior of the equipment in advance, and extend the service life of the equipment.
  • the effect is better than the existing prediction method based on fault threshold, and it has the characteristics of strong versatility and convenient construction.
  • Figure 1 is an algorithm block diagram of the present invention's fault early warning method based on oilfield power grid equipment.
  • This invention analyzes the historical fault data of oilfield network power equipment, builds a reasonable fault warning model through a fault early warning algorithm, compares the real-time data of oilfield network power equipment with the model, and makes a fault early warning for data that is not consistent with the model. .
  • the present invention proposes a fault early warning method based on oilfield power network equipment, which includes the following steps:
  • the core data set G e for fault prediction is extracted, and the value of each element in the core data set G e when a fault occurs is defined as G en (n is 1.2.3.4... sequence of%), define the real-time data of each element as G rn (n is the sequence of 1.2.3.4..., corresponding to the sequence of G en ), and define the safe interval of each element in the core data as G sn (n is the sequence of 1.2.3.4..., corresponding to the sequence of G en ), define the number of fault occurrences as N, and define the number of times that each element value G en in the core data exceeds the safe interval G sn in N faults as G bn (n is the sequence of 1.2.3.4..., corresponding to the sequence of G en ), the early warning accuracy is defined as um, and the interval triggered by the fault warning is defined as G umn (n is the sequence of 1.2.3.4..., which corresponds to the sequence
  • Fault warning model training Import the historical fault data of oilfield power grid equipment into the fault data training algorithm in step 1 to generate a probabilistic fault prediction model.
  • the number of historical faults is counted as N.
  • the number of times G bn in which the element value G en in the core data exceeds the safe interval G sn is cleared.
  • the value of the fault warning triggering interval G umn is the upper and lower limits of the safety interval G sn multiplied by the warning accuracy um.
  • the value of the failure probability G tn of each element is obtained by dividing G bn by the number of historical faults. From this, we get the basis A fault warning model generated from actual historical faults.
  • Fault warning The fault warning is verified through the implementation data.
  • the real-time data of the oil field power grid equipment is imported into the fault model generated by the historical fault data G en in step 2, and compared, the fault is predicted and warned:
  • G rn exceeds the safety interval G sn
  • G tn um
  • G umn G umn
  • G tn ⁇ um is accumulated, divide the accumulated value by the number of accumulations to get m , if (m+1)/2>0.5, a fault warning is triggered.
  • the fault early warning algorithm based on oilfield power grid equipment proposed by the present invention has higher prediction accuracy and can more effectively protect oilfield power grid equipment.
  • this prediction algorithm has the problem of being unable to accurately measure elements in the model whose probability of triggering a fault is less than the prediction accuracy in actual predictions. Fuzzy algorithms can only be used to reduce the probability of false triggering and improve prediction accuracy as much as possible.
  • the implementation example of the present invention is implemented on a machine equipped with Intel Core i7-3770 central processor, NVidia GTX760 graphics processor and 32GB memory.
  • a computer program is written to use historical fault data to derive a fault prediction model.
  • the real-time data of the equipment is brought into the fault model for comparison.
  • the fault warning accuracy is reached, the fault is triggered.
  • this method can more accurately predict equipment failures and issue early warnings when failures occur, which greatly reduces the failure rate of equipment and extends the service life of equipment.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种基于油田网电设备的故障预警方法，首先基于油田网电设备故障预警要求，搭建适用于油田网电的概率型数据训练算法；然后将油田网电设备的历史故障数据导入到概率型数据训练算法，生成概率型故障预测模型；最后将油田网电设备的实时数据导入到概率型故障预测模型中，进行比对，对故障做出预测及预警。在原有故障阈值判断的基础上，提出了历史故障模型搭建的新思路，针对设备故障发生的可能性做出了更为有效的预警机制，优于现有方法，具有通用性强，搭建便捷等优点。

Description

一种基于油田网电设备的故障预警方法 技术领域
本发明涉及计算机故障预警方法领域，尤其涉及一种油田网电设备的故障预警方法。
背景技术
设备故障预警技术在油田网电设备的使用中对延长设备的使用寿命有着增益作用，可以在故障发生之前就对设备的不良状态做出预警。石油钻井现场通常在户外人迹罕至的地方，现阶段石油钻井常采用网电驱动电机作为主要的动力驱动设备开采石油，现场的网电工况经常面临线路架线距离远、电网电压不稳定、负载大、谐波大等问题，长期在这种工况下运行，导致网电设备负荷大、易发生故障和损耗，因此设计一种基于油田网电设备的高效的故障预警方法是十分必要和迫切的。
以往只针对数据故障阈值的预测方法，通常只针对两类数据：一是仪表内的保护和故障标志位，常以开关量的形式被采集和判断；二是针对电流、电压、谐波等参数的阈值；这两种方法各有缺陷，针对保护和故障标值位的方式，通常只能检测到故障的发生，而无法对故障做出提前的预测；而针对参数阈值的预警方式，预警的依据往往是来自设备的额定参数计算而来，但是油田网电设备的工况十分复杂，设备损耗较大，很多设备在工作一段时间后，已经无法在设计的额定区间内正常工作，通过参数阈值进行预警的方式在油田网电设备中有一定的局限性。
发明内容
本发明的目的在于针对现有技术的不足，提供一种基于油田网电设备的故障预警方法。
本发明的目的是通过以下技术方案来实现的：一种基于油田网电设备的故障预警方法，包括如下步骤：
(1)故障数据训练算法搭建：基于油田网电设备故障预警要求，搭建适用于油田网电的概率型数据训练算法；
(2)故障预警模型训练：将油田网电设备的历史故障数据导入到步骤(1)搭建的概率型数据训练算法中，生成概率型故障预测模型；
(3)故障预警：将油田网电设备的实时数据导入到步骤(2)生成的概率型故障预测模型中，进行比对，对故障做出预测及预警。
进一步地，所述步骤(1)包括以下子步骤：
(1.1)定义参数：根据油田网电设备的故障预警要求，抽离出故障预测的核心数据集合G e， 将故障发生时核心数据集合G e中的各元素的值定义为G en，将各元素的实时数据定义为G rn，将核心数据中各元素的安全区间定义为G sn，将故障发生次数定义为N，将N次故障中核心数据中各元素值G en超出安全区间G sn的次数定义为G bn，将预警精度定义为um，将故障预警触发的区间定义为G umn，将模型中各元素触发故障的概率定义为G tn，n为1、2、3、4……的序列；
(1.2)通过步骤(1.1)中定义的参数，得到历史故障数据的以下算法：G bn初始值为0，当G en超出G sn时，G bn+1；G umn的区间为G sn的上下限值乘um，G umn＝G sn*um；G tn＝G bn/N。
进一步地，所述步骤(2)具体为：将历史故障的次数计入N，N次故障中核心数据中个元素值G en超出安全区间G sn的次数G bn清零，当核心数据的历史数据G en超出安全区间G sn时，次数G bn+1；故障预警触发区间G umn的值为安全区间G sn的上下限乘预警精度um得到，各元素故障发生概率G tn的值由G bn除以历史故障次数得到，由此得到根据实际发生的历史故障产生的故障预警模型。
进一步地，所述步骤(3)具体为：当G tn>um时，G rn超出G umn即触发故障预警；当G tn<um或G tn＝um时，设累加参数n＝0，G rn超出G umn时累加，将累加值n除以G tn≤um的数据次数得到m，若(m+1)/2>0.5则触发故障预警。
本发明的有益效果是，本发明通过构建油田网电设备的故障预警算法和模型，可以有效地对油田网电设备做出保护，提前预测设备的危险行为，延长设备的使用寿命。效果优于现有的基于故障阈值进行预测的方法，具有通用性强，搭建便捷等特点。
附图说明
图1是本发明基于油田网电设备的故障预警方法的算法框图。
具体实施方式
本发明对油田网电设备的历史故障数据进行分析，通过故障预警算法搭建出合理的故障预警模型，在将油田网电设备的实时数据与模型比对，对与模型不服的数据做出故障预警。
本发明提出了一种基于油田网电设备的故障预警方法，包括如下步骤：
1.故障数据训练算法搭建：基于油田网电设备故障预警要求，搭建适用于油田网电的概率型数据训练算法；具体包括以下子步骤：
1.1.根据油田网电设备的故障预警要求，抽离出故障预测的核心数据集合G e，将故障发生时核心数据集合G e中的各元素的值定义为G en(n为1.2.3.4……的序列)，将各元素的实时数据定义为G rn(n为1.2.3.4……的序列，与G en的序列对应)，将核心数据中各元素的安全区间定义为G sn(n为1.2.3.4……的序列，与G en的序列对应)，将故障发生次数定义为N，将N次故障中核心数据中各元素值G en超出安全区间G sn的次数定义为G bn(n为1.2.3.4……的序列， 与G en的序列对应)，将预警精度定义为um，将故障预警触发的区间定义为G umn(n为1.2.3.4……的序列，与G en的序列对应)，将模型中各元素触发故障的概率定义为G tn(n为1.2.3.4……的序列，与G en的序列对应)。
1.2.通过步骤1.1)中定义的参数，得到历史故障数据的以下算法：G bn初始值为0，当G en超出G sn时G bn+1；G umn的区间为G sn的上下限值乘um，G umn＝G sn*um；G tn＝G bn/N。
2.故障预警模型训练：将油田网电设备的历史故障数据导入到步骤1中故障数据训练算法中，生成概率型故障预测模型。
将历史故障的次数计入N，N次故障中核心数据中个元素值G en超出安全区间G sn的次数G bn清零，当核心数据的历史数据G en超出安全区间G sn时，次数G bn+1；故障预警触发区间G umn的值为安全区间G sn的上下限乘预警精度um得到，各元素故障发生概率G tn的值由G bn除以历史故障次数得到，由此便得到根据实际发生的历史故障产生的故障预警模型。
3.故障预警：故障预警通过实施数据进行验证，将油田网电设备的实时数据导入到步骤2中由历史故障数据G en生成的故障模型中，进行比对，对故障做出预测及预警：当实时数据G rn超出安全区间G sn时触发故障预警；当G tn>um时，若G rn超出G umn时触发故障预警；当G tn<um时累加，将累加值除以累加次数得到m，若(m+1)/2>0.5则触发故障预警。
相较于以往只针对数据故障阈值的预测方法，本发明提出的基于油田网电设备的故障预警算法具有更高的预测精度，可以更为有效的保护油田网电设备。但该预测算法对于模型中触发故障的概率小于预测精度的元素，在实际的预测中存在无法精准衡量的问题，只能采用模糊算法，尽可能的减小误触发概率和提升预测精度。
实施实例
在一台配备Intel Core i7-3770中央处理器，NVidia GTX760图形处理器及32GB内存的机器上实现本发明的实施实例。根据图1所示的算法，编写出计算机程序，用历史故障数据得出故障预测模型，在将设备的实时数据带入故障模型中进行比对，到到达故障预警精度时便触发故障。与现有方法相比，本方法可以更为精准的预测设备的故障，在故障发生便发出预警，极大的降低了设备的故障率，延长了设备的使用寿命。

Claims (4)

  1. 一种基于油田网电设备的故障预警方法，其特征在于，包括如下步骤：
    (1)故障数据训练算法搭建：基于油田网电设备故障预警要求，搭建适用于油田网电的概率型数据训练算法；
    (2)故障预警模型训练：将油田网电设备的历史故障数据导入到步骤(1)搭建的概率型数据训练算法中，生成概率型故障预测模型；
    (3)故障预警：将油田网电设备的实时数据导入到步骤(2)生成的概率型故障预测模型中，进行比对，对故障做出预测及预警。
  2. 根据权利要求1所述的方法，其特征在于，所述步骤(1)包括以下子步骤：
    (1.1)定义参数：根据油田网电设备的故障预警要求，抽离出故障预测的核心数据集合G e，将故障发生时核心数据集合G e中的各元素的值定义为G en，将各元素的实时数据定义为G rn，将核心数据中各元素的安全区间定义为G sn，将故障发生次数定义为N，将N次故障中核心数据中各元素值G en超出安全区间G sn的次数定义为G bn，将预警精度定义为um，将故障预警触发的区间定义为G umn，将模型中各元素触发故障的概率定义为G tn，n为1、2、3、4……的序列；
    (1.2)通过步骤(1.1)中定义的参数，得到历史故障数据的以下算法：G bn初始值为0，当G en超出G sn时，G bn+1；G umn的区间为G sn的上下限值乘um，G umn＝G sn*um；G tn＝G bn/N。
  3. 根据权利要求1所述的方法，其特征在于，所述步骤(2)具体为：将历史故障的次数计入N，N次故障中核心数据中个元素值G en超出安全区间G sn的次数G bn清零，当核心数据的历史数据G en超出安全区间G sn时，次数G bn+1；故障预警触发区间G umn的值为安全区间G sn的上下限乘预警精度um得到，各元素故障发生概率G tn的值由G bn除以历史故障次数得到，由此得到根据实际发生的历史故障产生的故障预警模型。
  4. 根据权利要求1所述的方法，其特征在于，所述步骤(3)具体为：当G tn>um时，G rn超出G umn即触发故障预警；当G tn<um或G tn＝um时，设累加参数n＝0，G rn超出G umn时累加，将累加值n除以G tn<um的数据次数得到m，若(m+1)/2>0.5则触发故障预警。
PCT/CN2022/118701 2022-07-11 2022-09-14 一种基于油田网电设备的故障预警方法 WO2024011745A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210811681.4 2022-07-11
CN202210811681.4A CN115236453A (zh) 2022-07-11 2022-07-11 一种基于油田网电设备的故障预警方法

Publications (1)

Publication Number Publication Date
WO2024011745A1 true WO2024011745A1 (zh) 2024-01-18

Family

ID=83670706

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/118701 WO2024011745A1 (zh) 2022-07-11 2022-09-14 一种基于油田网电设备的故障预警方法

Country Status (2)

Country Link
CN (1) CN115236453A (zh)
WO (1) WO2024011745A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11955782B1 (en) 2022-11-01 2024-04-09 Typhon Technology Solutions (U.S.), Llc System and method for fracturing of underground formations using electric grid power

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103714401A (zh) * 2014-01-21 2014-04-09 国家电网公司 一种电网配变故障次数预测的方法
US20190087247A1 (en) * 2015-04-24 2019-03-21 Senslytics Corporation Methods and systems correlating hypotheses outcomes using relevance scoring for intuition based forewarning
CN112257988A (zh) * 2020-09-29 2021-01-22 中广核工程有限公司 核电厂复杂事故特征识别及风险预警系统和方法
CN112836436A (zh) * 2021-03-01 2021-05-25 云南电网有限责任公司电力科学研究院 一种基于概率图模型的配电网线路风险量化预测方法
CN114444714A (zh) * 2021-12-31 2022-05-06 上海恕笃智能科技有限公司 一种基于快速自主学习的电厂设备实时故障预警方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103714401A (zh) * 2014-01-21 2014-04-09 国家电网公司 一种电网配变故障次数预测的方法
US20190087247A1 (en) * 2015-04-24 2019-03-21 Senslytics Corporation Methods and systems correlating hypotheses outcomes using relevance scoring for intuition based forewarning
CN112257988A (zh) * 2020-09-29 2021-01-22 中广核工程有限公司 核电厂复杂事故特征识别及风险预警系统和方法
CN112836436A (zh) * 2021-03-01 2021-05-25 云南电网有限责任公司电力科学研究院 一种基于概率图模型的配电网线路风险量化预测方法
CN114444714A (zh) * 2021-12-31 2022-05-06 上海恕笃智能科技有限公司 一种基于快速自主学习的电厂设备实时故障预警方法

Also Published As

Publication number Publication date
CN115236453A (zh) 2022-10-25

Similar Documents

Publication Publication Date Title
CN110596486B (zh) 一种充电桩智能预警运维的方法和系统
CN110210701A (zh) 一种电网设备风险感知方法
WO2024011745A1 (zh) 一种基于油田网电设备的故障预警方法
CN104299367A (zh) 一种滑坡灾害多级综合监测预警方法
US20210278464A1 (en) Condition assessment method and device for an outdoor post-mounted vacuum switch
CN104966147A (zh) 一种考虑基态和事故态的电网运行风险分析方法
CN102508046B (zh) 一种户外电气设备实时雷击预警方法及装置
CN102721924A (zh) 风力发电机组的故障预警方法
CN102324068A (zh) 电力二次设备风险评估方法及其系统
CN103324992A (zh) 基于马尔可夫和熵权模糊综合评价的变压器风险预测方法
CN103914951A (zh) 基于gis电网自然灾害数据信息挖掘与预警响应分析方法
CN103226651A (zh) 基于相似度统计的风电机组状态评估预警方法和系统
CN103903196A (zh) 一种考虑设备老化因素的电网调度运行实时风险评估方法
CN112986868B (zh) 一种变压器状态监测方法
CN104346691B (zh) 一种基于故障树分析的电力企业现场作业风险分析方法
CN112713649A (zh) 一种基于极限学习机的电力设备剩余寿命预测方法
CN103630776A (zh) 一种电力设备故障概率评估方法
CN101158916A (zh) 一种数据库性能监控方法
KR102316516B1 (ko) 분포도를 통한 기기의 예지 보전방법
JP7296525B2 (ja) 分布図を通じた機器の予知保全方法
CN115037046A (zh) 一种电力二次设备运行状态分析检测系统
CN105404962B (zh) 电力系统继电保护装置的改造评估方法及其生成系统
CN105023198A (zh) 一种基于网络规则约束的发电厂数据异常辨识方法
CN103020431A (zh) 油浸式变压器状态评价的建模方法及其评价系统
CN114465217B (zh) 一种基于高频隔离的防雷方法及装置

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: 22950856

Country of ref document: EP

Kind code of ref document: A1