CN108107360B - Motor fault identification method and system - Google Patents
Motor fault identification method and system Download PDFInfo
- Publication number
- CN108107360B CN108107360B CN201711269578.7A CN201711269578A CN108107360B CN 108107360 B CN108107360 B CN 108107360B CN 201711269578 A CN201711269578 A CN 201711269578A CN 108107360 B CN108107360 B CN 108107360B
- Authority
- CN
- China
- Prior art keywords
- fault
- motor
- parameters
- identification
- probability
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 84
- 238000012544 monitoring process Methods 0.000 claims abstract description 88
- 230000007246 mechanism Effects 0.000 claims abstract description 41
- 238000007726 management method Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000012423 maintenance Methods 0.000 claims description 79
- 208000024891 symptom Diseases 0.000 claims description 51
- 230000002159 abnormal effect Effects 0.000 claims description 38
- 238000004458 analytical method Methods 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 23
- 230000001364 causal effect Effects 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 8
- 238000012546 transfer Methods 0.000 claims 1
- 238000003745 diagnosis Methods 0.000 abstract description 28
- 230000008439 repair process Effects 0.000 abstract description 11
- 238000013021 overheating Methods 0.000 description 35
- 238000010586 diagram Methods 0.000 description 13
- 238000004804 winding Methods 0.000 description 7
- 230000005856 abnormality Effects 0.000 description 6
- 230000017525 heat dissipation Effects 0.000 description 6
- 238000007689 inspection Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 230000015556 catabolic process Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 238000009411 base construction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000005461 lubrication Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 239000010687 lubricating oil Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003921 oil Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
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
- G01R31/34—Testing dynamo-electric machines
- G01R31/343—Testing dynamo-electric machines in operation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
Description
技术领域technical field
本发明涉及电机故障处理技术领域,特别是涉及一种电机故障辨识方法与系统。The invention relates to the technical field of motor fault processing, in particular to a motor fault identification method and system.
背景技术Background technique
电机的各个主要功能部件(包括定子、转子、铁心、气隙、轴承组件)之间存在着紧密的电磁和机械联系,它们的状态互相影响。某部件的损坏极可能导致其他部件的异常,而某部件的异常可能由多个原因引起。电机的故障辨识遵循电机的故障特性。所有故障的发生都会引起相应故障征兆的出现,征兆出现可通过相关参数的变化来判断,而故障辨识的任务可以包括判断征兆参数是否异常,以及当出现参数异常后,如何确定是哪些故障引起的。There are close electromagnetic and mechanical connections between the main functional components of the motor (including stator, rotor, iron core, air gap, bearing assembly), and their states affect each other. The damage of one part is very likely to cause the abnormality of other parts, and the abnormality of one part may be caused by many reasons. The fault identification of the motor follows the fault characteristics of the motor. The occurrence of all faults will cause the occurrence of corresponding fault symptoms. The occurrence of symptoms can be judged by the changes of relevant parameters. The task of fault identification can include judging whether the symptom parameters are abnormal, and how to determine which faults are caused when the parameters are abnormal. .
为了确定故障发生与具体征兆参数的关系,围绕具体故障(如断条、绝缘击穿、匝间短路、轴承损坏等)的征兆提取和故障分析技术不断提出,随着监测手段的丰富,诊断方法的提升,越来越多的电机故障可以被较精确地表达,例如传统的电机故障的逻辑诊断方法,为了刻画电机故障和征兆之间的多对多关系,通过布尔矩阵实现这种关联关系表达,利用矩阵的逻辑运算进行诊断。In order to determine the relationship between fault occurrence and specific symptom parameters, symptom extraction and fault analysis techniques for specific faults (such as broken bars, insulation breakdown, inter-turn short circuit, bearing damage, etc.) have been continuously proposed. With the enrichment of monitoring methods, diagnostic methods With the improvement of , more and more motor faults can be expressed more accurately, such as the traditional logical diagnosis method of motor faults, in order to describe the many-to-many relationship between motor faults and symptoms, the Boolean matrix is used to express this relationship. , using matrix logic operations for diagnosis.
在实现过程中,发明人发现传统技术中至少存在如下问题:传统电机故障的逻辑诊断方法中几百个征兆和几十个故障模式,易形成巨大的稀疏矩阵;即传统技术只能围绕单个故障模式的征兆识别和诊断,或多传感器的融合诊断,在实际应用中,可扩展性和实用性较差。During the implementation process, the inventor found that there are at least the following problems in the traditional technology: in the traditional motor fault logic diagnosis method, there are hundreds of symptoms and dozens of failure modes, which are easy to form a huge sparse matrix; that is, the traditional technology can only focus on a single fault Pattern recognition and diagnosis, or multi-sensor fusion diagnosis, have poor scalability and practicability in practical applications.
发明内容SUMMARY OF THE INVENTION
基于此,有必要针对传统故障辨识方法可扩展性和实用性较差的问题,提供一种电机故障辨识方法与系统。Based on this, it is necessary to provide a motor fault identification method and system for the problems of poor scalability and practicability of traditional fault identification methods.
为了实现上述目的,一方面,本发明实施例提供了一种电机故障辨识方法,包括以下步骤:In order to achieve the above purpose, on the one hand, an embodiment of the present invention provides a method for identifying a motor fault, including the following steps:
对运行中的电机进行信号采集监测,得到电机的信号监测参数,处理信号监测参数,得到电机的性能跟踪参数;Perform signal acquisition and monitoring on the running motor, obtain the signal monitoring parameters of the motor, process the signal monitoring parameters, and obtain the performance tracking parameters of the motor;
根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率;故障知识库包括故障关联规则和故障维修案例数据;预设故障辨识规则包括故障判定规则和故障推理规则;故障判定规则包括将信号监测参数、性能跟踪参数,分别与预设征兆判据库中各征兆记录进行逐条匹配,得到电机发生故障时各故障判定发生概率;故障推理规则包括将根据故障关联规则处理各故障判定发生概率、得到的故障关联推理结果,与将信号监测参数和性能跟踪参数分别与故障维修案例数据进行比对、得到的案例推理结果,进行加权求平均,得到各故障概率;According to the preset fault identification rules, the signal monitoring parameters and performance tracking parameters are processed based on the fault knowledge base to obtain the probability of each fault when the motor fails; the fault knowledge base includes fault association rules and fault maintenance case data; the preset fault identification rules include faults Judgment rules and fault reasoning rules; fault judgment rules include matching the signal monitoring parameters and performance tracking parameters with each symptom record in the preset symptom criterion database one by one to obtain the probability of occurrence of each fault judgment when the motor fails; fault reasoning rules Including the probability of occurrence of each fault determined according to the fault association rules, the obtained fault correlation reasoning results, and the case reasoning results obtained by comparing the signal monitoring parameters and performance tracking parameters with the fault maintenance case data respectively, weighted and averaged, Get the probability of each failure;
对各故障概率进行大小排序,得到电机的故障辨识结果。The probability of each fault is sorted by size, and the fault identification result of the motor is obtained.
在其中一个实施例中,信号监测参数包括电机的电参数、各部件温度参数、工作参数以及状态参数;性能跟踪参数包括电机功率、电机效率、最小转矩、最大转矩、堵转转矩、堵转电流以及温升;In one embodiment, the signal monitoring parameters include electrical parameters of the motor, temperature parameters of various components, working parameters and state parameters; the performance tracking parameters include motor power, motor efficiency, minimum torque, maximum torque, stall torque, Stall current and temperature rise;
在根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率的步骤之前还包括步骤:Before the step of processing the signal monitoring parameters and the performance tracking parameters based on the fault knowledge base according to the preset fault identification rules, and obtaining the probability of each fault when the motor fails, the steps further include:
构建电机的故障基础信息库;故障基础信息库包括故障属性信息;Build the fault basic information database of the motor; the fault basic information database includes the fault attribute information;
根据故障属性信息,构建故障知识库。Build a fault knowledge base according to fault attribute information.
在其中一个实施例中,故障关联规则为因果关系链;In one of the embodiments, the fault association rule is a causal relationship chain;
构建故障基础信息库的步骤包括:The steps to build a fault base information base include:
根据电机故障特征逐级进行故障对象划分,并引入具备电机故障特征的虚拟结构,得到电机的故障对象划分结构模型;According to the fault characteristics of the motor, the fault objects are divided step by step, and the virtual structure with the fault characteristics of the motor is introduced to obtain the structure model of the fault object division of the motor;
根据故障对象划分结构模型进行故障模式分析,确定电机的各潜在故障;并根据各潜在故障,生成故障属性信息;According to the fault object division structure model, the failure mode analysis is carried out, and the potential faults of the motor are determined; and the fault attribute information is generated according to the potential faults;
构建故障知识库的步骤包括:The steps to build a fault knowledge base include:
根据故障树分析,得到故障属性信息中各故障的因果关系链;According to the fault tree analysis, the causal relationship chain of each fault in the fault attribute information is obtained;
根据故障对象划分结构模型、信号监测参数和性能跟踪参数,生成故障维修案例数据。According to the fault object division structure model, signal monitoring parameters and performance tracking parameters, the fault maintenance case data is generated.
在其中一个实施例中,故障关联推理结果包括完整故障机理结果和并发故障结果;In one of the embodiments, the fault correlation reasoning result includes a complete fault mechanism result and a concurrent fault result;
根据故障推理规则,处理故障判定发生概率的步骤包括:According to the fault inference rules, the steps of processing the probability of fault determination include:
根据故障关联规则,在确认各故障判定发生概率对应的各故障处于同一条故障关联链路上时,通过概率计算得到完整故障机理结果;According to the fault association rule, when it is confirmed that each fault corresponding to the probability of occurrence of each fault is on the same fault-related link, the complete fault mechanism result is obtained through probability calculation;
或or
在确认各故障判定发生概率对应的各故障处于不同的故障关联链路上时,通过概率计算得到并发故障结果。When it is confirmed that each fault corresponding to the occurrence probability of each fault judgment is on a different fault-related link, the concurrent fault result is obtained through probability calculation.
在其中一个实施例中,在根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率的步骤之前还包括步骤:In one of the embodiments, before the step of processing the signal monitoring parameters and performance tracking parameters based on the fault knowledge base according to the preset fault identification rule, and obtaining the probability of each fault when the motor fails, the steps further include:
在监测到信号监测参数、性能跟踪参数超过相应报警值时,发出异常报警;When it is detected that the signal monitoring parameters and performance tracking parameters exceed the corresponding alarm values, an abnormal alarm is issued;
在得到电机的故障辨识结果的步骤之后还包括步骤:After the step of obtaining the fault identification result of the motor, it also includes the following steps:
根据故障辨识结果,确定维修计划;According to the fault identification results, determine the maintenance plan;
根据故障辨识结果和维修计划,更新故障维修案例数据。According to the fault identification result and maintenance plan, update the fault maintenance case data.
一方面,本发明实施例还提供了一种电机故障辨识系统,包括:On the one hand, an embodiment of the present invention also provides a motor fault identification system, including:
状态监测单元,用于对运行中的电机进行信号采集监测,得到电机的信号监测参数;The state monitoring unit is used to collect and monitor the signal of the motor in operation, and obtain the signal monitoring parameters of the motor;
性能跟踪单元,用于处理信号监测参数,得到电机的性能跟踪参数;The performance tracking unit is used to process the signal monitoring parameters and obtain the performance tracking parameters of the motor;
故障辨识单元,用于根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率;故障知识库包括故障关联规则和故障维修案例数据;预设故障辨识规则包括故障判定规则和故障推理规则;故障判定规则包括将信号监测参数、性能跟踪参数,分别与预设征兆判据库中各征兆记录进行逐条匹配,得到电机发生故障时各故障判定发生概率;故障推理规则包括将根据故障关联规则处理各故障判定发生概率、得到的故障关联推理结果,与将信号监测参数和性能跟踪参数分别与故障维修案例数据进行比对、得到的案例推理结果,进行加权求平均,得到各故障概率;The fault identification unit is used to process signal monitoring parameters and performance tracking parameters based on the fault knowledge base according to the preset fault identification rules, and obtain each fault probability when the motor fails; the fault knowledge base includes fault association rules and fault maintenance case data; It is assumed that the fault identification rules include fault judgment rules and fault inference rules; the fault judgment rules include matching the signal monitoring parameters and performance tracking parameters with each symptom record in the preset symptom criterion database one by one, so as to obtain each fault judgment when the motor fails. Occurrence probability; fault inference rules include the probability of occurrence of each fault determined by processing each fault according to the fault correlation rule, the obtained fault correlation inference results, and the case inference results obtained by comparing the signal monitoring parameters and performance tracking parameters with the fault maintenance case data respectively. , perform weighted averaging to obtain the probability of each failure;
排序单元,用于对各故障概率进行大小排序,得到电机的故障辨识结果。在其中一个实施例中,The sorting unit is used for sorting the probability of each fault to obtain the fault identification result of the motor. In one of the embodiments,
在其中一个实施例中,信号监测参数包括电机的电参数、各部件温度参数、工作参数以及状态参数;性能跟踪参数包括电机功率、电机效率、最小转矩、最大转矩、堵转转矩、堵转电流以及温升;In one embodiment, the signal monitoring parameters include electrical parameters of the motor, temperature parameters of various components, working parameters and state parameters; the performance tracking parameters include motor power, motor efficiency, minimum torque, maximum torque, stall torque, Stall current and temperature rise;
还包括:Also includes:
故障基础信息管理单元,用于构建电机的故障基础信息库;故障基础信息库包括故障属性信息;The fault basic information management unit is used to construct the fault basic information database of the motor; the fault basic information database includes the fault attribute information;
故障知识库单元,用于根据故障属性信息,构建故障知识库。The fault knowledge base unit is used to construct a fault knowledge base according to the fault attribute information.
在其中一个实施例中,故障关联规则为因果关系链;In one of the embodiments, the fault association rule is a causal relationship chain;
故障基础信息管理单元包括:The fault basic information management unit includes:
故障对象划分模块,用于根据电机故障特征逐级进行故障对象划分,并引入具备电机故障特征的虚拟结构,得到电机的故障对象划分结构模型;The fault object division module is used to divide the fault objects step by step according to the fault characteristics of the motor, and introduce a virtual structure with the fault characteristics of the motor to obtain the fault object division structure model of the motor;
故障模式分析模块,用于根据故障对象划分结构模型进行故障模式分析,确定电机的各潜在故障;The failure mode analysis module is used to analyze the failure mode according to the fault object division structure model, and determine the potential faults of the motor;
故障属性信息管理模块,用于根据各潜在故障,生成故障属性信息;The fault attribute information management module is used to generate fault attribute information according to each potential fault;
故障知识库单元包括:Fault knowledge base units include:
故障关联规则模块,用于根据故障树分析,得到故障属性信息中各故障的因果关系链;The fault association rule module is used to obtain the causal relationship chain of each fault in the fault attribute information according to the fault tree analysis;
故障维修案例库模块,用于根据故障对象划分结构模型、信号监测参数和性能跟踪参数,生成故障维修案例数据。The fault maintenance case library module is used to divide the structural model, signal monitoring parameters and performance tracking parameters according to the fault object, and generate fault maintenance case data.
一方面,本发明实施例提供了一种计算机设备,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,处理器执行程序时实现上述电机故障辨识方法各实施例的步骤。On the one hand, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the program, the above embodiments of the motor fault identification method are implemented. step.
另一方面,本发明实施例还提供了一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时实现上述电机故障辨识方法各实施例的步骤。On the other hand, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, implements the steps of each embodiment of the above-mentioned motor fault identification method.
上述技术方案中的一个技术方案具有如下优点和有益效果:A technical scheme in the above-mentioned technical scheme has the following advantages and beneficial effects:
对运行中的电机进行信号采集与监测,得到电机的信号监测参数,进而对电机进行性能跟踪,通过预设故障辨识规则,基于完整的故障知识库处理相关参数,进而从底层实现故障管理、性能跟踪、判据管理与征兆管理,同时采用与故障辨识相协调的故障推理机制,基于故障维修案例数据,得到能够辅助维修人员进行确诊和检修的故障辨识结论。本发明实施例提供了一种可不断扩展的故障辨识机制,结合案例、标准和经验形成故障知识管理方法,以及利用这些知识进行故障辨识推理;本发明实施例基于电机故障知识管理和故障辨识机制,具有较强的实用性和可扩展性,对于电机运行的在线故障诊断和离线故障定位,以及快速修复系统故障,提出了综合解决方案。Collect and monitor the signal of the motor in operation, obtain the signal monitoring parameters of the motor, and then track the performance of the motor. Through the preset fault identification rules, the relevant parameters are processed based on the complete fault knowledge base, and then the fault management and performance are realized from the bottom layer. Tracking, criterion management and symptom management, and at the same time adopt a fault reasoning mechanism coordinated with fault identification, based on fault maintenance case data, to obtain fault identification conclusions that can assist maintenance personnel in diagnosis and maintenance. The embodiment of the present invention provides a fault identification mechanism that can be continuously expanded, forms a fault knowledge management method by combining cases, standards and experience, and uses these knowledge to perform fault identification and reasoning; the embodiment of the present invention is based on the motor fault knowledge management and fault identification mechanism. , has strong practicability and scalability, and proposes a comprehensive solution for online fault diagnosis and offline fault location of motor operation, as well as rapid repair of system faults.
附图说明Description of drawings
图1为本发明电机故障辨识方法实施例1的流程示意图;1 is a schematic flowchart of
图2为本发明电机故障辨识方法实施例中各环节架构示意图;2 is a schematic diagram of the structure of each link in the embodiment of the motor fault identification method of the present invention;
图3为本发明电机故障辨识方法实施例中故障对象划分示意图;3 is a schematic diagram of the division of fault objects in the embodiment of the motor fault identification method of the present invention;
图4为本发明电机故障辨识方法实施例中故障模式分析示意图;4 is a schematic diagram of failure mode analysis in an embodiment of a motor fault identification method according to the present invention;
图5为本发明电机故障辨识方法实施例中故障关联规则示意图;5 is a schematic diagram of a fault association rule in an embodiment of the motor fault identification method of the present invention;
图6为本发明电机故障辨识方法实施例中故障辨识过程的流程示意图;6 is a schematic flowchart of a fault identification process in an embodiment of a motor fault identification method according to the present invention;
图7为本发明电机故障辨识系统实施例1的结构示意图。FIG. 7 is a schematic structural diagram of
具体实施方式Detailed ways
为了便于理解本发明,下面将参照相关附图对本发明进行更全面的描述。附图中给出了本发明的首选实施例。但是,本发明可以以许多不同的形式来实现,并不限于本文所描述的实施例。相反地,提供这些实施例的目的是使对本发明的公开内容更加透彻全面。In order to facilitate understanding of the present invention, the present invention will be described more fully hereinafter with reference to the related drawings. Preferred embodiments of the invention are shown in the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
除非另有定义,本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的,不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terms used herein in the description of the present invention are for the purpose of describing specific embodiments only, and are not intended to limit the present invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
本发明电机故障辨识方法与系统各实施例中涉及的专业术语解释及定义:Explanation and definitions of technical terms involved in the embodiments of the motor fault identification method and system of the present invention:
电机:是把电能转换成机械能的一种设备。一般是利用通电线圈(也就是定子绕组)产生旋转磁场并作用于转子形成磁电动力旋转扭矩。电动机按使用电源不同分为直流电动机和交流电动机,电力系统中的电动机大部分是交流电机,可以是同步电机或者是电机(电机定子磁场转速与转子旋转转速不保持同步速)。电动机主要由定子与转子组成,通电导线在磁场中受力运动的方向跟电流方向和磁感线(磁场方向)方向有关。电动机工作原理是磁场对电流受力的作用,使电动机转动。Motor: A device that converts electrical energy into mechanical energy. Generally, the energized coil (that is, the stator winding) is used to generate a rotating magnetic field and act on the rotor to form a magneto-electric rotational torque. Motors are divided into DC motors and AC motors according to different power sources. Most of the motors in the power system are AC motors, which can be synchronous motors or motors (the stator magnetic field speed of the motor and the rotor rotation speed do not maintain a synchronous speed). The motor is mainly composed of a stator and a rotor, and the direction of the energized wire in the magnetic field is related to the direction of the current and the direction of the magnetic field line (magnetic field direction). The working principle of the motor is that the force of the magnetic field on the current causes the motor to rotate.
故障辨识:通过设备所有可获得的当前和历史运行信息,判断系统状态,识别系统异常,定位故障原因。Fault identification: Judging system status, identifying system abnormalities, and locating fault causes through all available current and historical operating information of the equipment.
知识管理:故障知识是一切与故障辨识有关的产品设计、标准和经验等相关的格式化信息。Knowledge management: Fault knowledge is all formatted information related to product design, standards and experience related to fault identification.
推理机制:从众多故障知识中推理出故障辨识结果的逻辑和流程。Reasoning mechanism: The logic and process of inferring fault identification results from numerous fault knowledge.
本发明电机故障辨识方法实施例1:
为了解决传统故障辨识方法可扩展性和实用性较差的问题,本发明提供了一种电机故障辨识方法实施例1;图1为本发明电机故障辨识方法实施例1的流程示意图;如图1所示,可以包括以下步骤:In order to solve the problems of poor scalability and practicability of the traditional fault identification method, the present invention provides a first embodiment of a motor fault identification method; FIG. 1 is a schematic flowchart of the first embodiment of the motor fault identification method of the present invention; FIG. 1 shown, may include the following steps:
步骤S110:对运行中的电机进行信号采集监测,得到电机的信号监测参数,处理信号监测参数,得到电机的性能跟踪参数;Step S110 : collecting and monitoring signals of the motor in operation, obtaining signal monitoring parameters of the motor, processing the signal monitoring parameters, and obtaining performance tracking parameters of the motor;
具体而言,本发明实施例中的状态监测是故障辨识的来源依据,可以包括信号采集方案和状态跟踪方案两部分;电机的信号采集方案可以采集与电机故障相关的主要参数,而在应用现场的电机,出于成本的考虑用户往往不会要求在电机内部安装振动和温度等传感器,工业现场也不一定具备完善的监测手段。Specifically, the state monitoring in the embodiment of the present invention is the source basis for fault identification, and may include two parts: a signal acquisition scheme and a state tracking scheme; the motor signal acquisition scheme can collect the main parameters related to the motor fault, and in the application site Because of cost considerations, users often do not require vibration and temperature sensors to be installed inside the motor, and industrial sites do not necessarily have complete monitoring methods.
本发明实施例中故障辨识的有效性和全面性,很大程度取决于状态监测数据的全面性。性能跟踪方案中,考虑电机关键性能的变化情况,电机性能参数是电机是否异常的重要表征,不同故障发生会导致不同的性能衰退现象。而上述性能跟踪参数,可通过所述信号采集方案中获得的相关参数计算而得。The effectiveness and comprehensiveness of the fault identification in the embodiment of the present invention largely depends on the comprehensiveness of the condition monitoring data. In the performance tracking scheme, the changes in the key performance of the motor are considered. The motor performance parameters are an important indicator of whether the motor is abnormal. Different faults will lead to different performance degradation phenomena. The above-mentioned performance tracking parameters can be calculated from the relevant parameters obtained in the signal acquisition scheme.
需要说明的是,由于所述信号采集方案受限于电机应用现场的监测手段,若计算过程的相关参数未进行监测,则该性能无法跟踪。It should be noted that, since the signal acquisition scheme is limited by the monitoring means at the motor application site, if the relevant parameters of the calculation process are not monitored, the performance cannot be tracked.
步骤S120:根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率;故障知识库包括故障关联规则和故障维修案例数据;预设故障辨识规则包括故障判定规则和故障推理规则;故障判定规则包括将信号监测参数、性能跟踪参数,分别与预设征兆判据库中各征兆记录进行逐条匹配,得到电机发生故障时各故障判定发生概率;故障推理规则包括将根据故障关联规则处理各故障判定发生概率、得到的故障关联推理结果,与将信号监测参数和性能跟踪参数分别与故障维修案例数据进行比对、得到的案例推理结果,进行加权求平均,得到各故障概率;Step S120: According to the preset fault identification rules, the signal monitoring parameters and performance tracking parameters are processed based on the fault knowledge base to obtain each fault probability when the motor fails; the fault knowledge base includes fault association rules and fault maintenance case data; preset fault identification The rules include fault judgment rules and fault inference rules; the fault judgment rules include matching the signal monitoring parameters and performance tracking parameters with each symptom record in the preset symptom criterion database one by one to obtain the probability of occurrence of each fault judgment when the motor fails; The fault inference rules include the probability of occurrence of each fault determined according to the fault association rules, the obtained fault correlation inference results, and the case inference results obtained by comparing the signal monitoring parameters and performance tracking parameters with the fault maintenance case data, and weighting them. Take the average to get the probability of each failure;
具体而言,故障知识库是完成本发明实施例故障辨识的故障知识,可以包括故障关联规则和故障维修案例数据;其中,故障关联规则可以指故障发生的因果关系链,采用自顶向下的分析方法,从故障现象逐层找到其故障原因。例如,电机过热可能由定子过热、转子过热、轴承过热、散热异常等原因引起,定子过热等又可能由定转子相擦等引起,如此递推,建立故障基础信息中所有故障的因果关系图。进一步的,可以基于上述因果关系图,逐个故障分析引起该故障的所有故障原因的发生频率。Specifically, the fault knowledge base is the fault knowledge for completing the fault identification in the embodiment of the present invention, and may include fault association rules and fault maintenance case data; wherein, the fault association rules may refer to the causal relationship chain of fault occurrence, and a top-down method is adopted. Analytical method to find the cause of the failure layer by layer from the failure phenomenon. For example, motor overheating may be caused by stator overheating, rotor overheating, bearing overheating, abnormal heat dissipation, etc., and stator overheating may be caused by stator and rotor rubbing. Further, based on the above-mentioned causal relationship diagram, the frequency of occurrence of all failure causes causing the failure can be analyzed one by one.
本发明实施例中的故障维修案例数据可以包括:故障现象、工况描述、异常参数、确诊故障、更换备件以及维修措施;而故障辨识机制(规则)解决通过监测参数,结合故障知识库,如何进行故障定位并给出维修方案的问题。将信号监测参数和性能跟踪参数与预设征兆判据库逐条匹配,预设征兆判据库包括故障及发生故障对应的征兆,每条故障及其征兆记录可在后期由领域专家进行进一步确认。The fault maintenance case data in the embodiment of the present invention may include: fault phenomenon, working condition description, abnormal parameters, fault diagnosis, replacement of spare parts, and maintenance measures; and the fault identification mechanism (rule) solves the problem by monitoring parameters, combined with fault knowledge base, how to Locating faults and giving maintenance solutions. Match the signal monitoring parameters and performance tracking parameters with the preset symptom criterion database one by one. The preset symptom criterion database includes faults and the symptoms corresponding to the faults. Each fault and its symptom records can be further confirmed by domain experts in the later stage.
预设征兆判据库是假设故障发生,判断可能引起的监测参数的变化;而故障判定是逆向的,根据监测的参数变化判断哪些故障发生了,由于一个故障可能对应多个参数变化,而一个参数变化也可能由多个故障引起,因此故障判定只能通过征兆匹配数目进行大致的故障发生概率的计算。The preset symptom criterion database is to assume that a fault occurs, and to judge the changes of monitoring parameters that may be caused; while the fault judgment is inverse, and judges which faults have occurred according to the changes of the monitored parameters. Because one fault may correspond to multiple parameter changes, and one Parameter changes may also be caused by multiple faults, so the fault determination can only be calculated roughly by the number of symptom matching.
因此,本发明实施例根据故障判定规则,将各实测参数与上述征兆判据库进行逐条匹配,每条记录中,一个故障对应多个征兆,征兆被匹配的数量越多,该故障发生的概率越大。以匹配征兆数与总征兆数的比值作为该故障的发生概率;例如,当前监测到m个征兆;预设征兆判据库中第X条记录有n个征兆;如果n中有k个被匹配,则k/n就是该故障的发生概率。同时,随着监测数据和故障数据的积累,预设征兆判据库是可扩展的,征兆数据既可以人工更新,也可以通过从历史数据中分析修正得到。Therefore, in the embodiment of the present invention, according to the fault determination rule, each measured parameter is matched with the above symptom criterion database one by one. In each record, one fault corresponds to multiple symptoms. bigger. The ratio of the number of matching symptoms to the total number of symptoms is used as the probability of occurrence of the fault; for example, m symptoms are currently monitored; there are n symptoms in the X-th record in the preset symptom criterion database; if k out of n are matched , then k/n is the probability of occurrence of the fault. At the same time, with the accumulation of monitoring data and fault data, the preset symptom criterion database is expandable, and symptom data can be updated manually or obtained through analysis and correction from historical data.
此外,根据故障判定规则可以得出电机出现了哪些故障,但无法定位出全部的故障原因以及完整的故障机理。因此,本发明实施例提出了故障推理规则,将上述故障判定获得的故障及概率信息代入故障关联规则,通过故障概率计算得到故障关联推理结果。In addition, according to the fault determination rules, the faults of the motor can be obtained, but all the fault causes and complete fault mechanisms cannot be located. Therefore, the embodiment of the present invention proposes a fault inference rule, and substitutes the fault and probability information obtained by the above fault determination into the fault association rule, and obtains the fault correlation inference result through the calculation of the fault probability.
同时,故障推理规则还包括将实时监测参数与故障维修案例数据进行比对,得到案例推理结果;进而将故障关联规则推理结果和案例推理结果进行加权求平均,得到各故障概率。At the same time, the fault reasoning rule also includes comparing the real-time monitoring parameters with the fault maintenance case data to obtain the case reasoning result; and then weighted and averaging the fault association rule reasoning result and the case reasoning result to obtain each fault probability.
步骤S130:对各故障概率进行大小排序,得到电机的故障辨识结果;Step S130: Ranking the probability of each failure to obtain a fault identification result of the motor;
具体而言,可以将上述故障推理机制获得的故障按概率大小进行排序,取前预设个数,例如10个,相关字段包括有助于维修人员确诊检修相关的完整信息;进一步形成故障辨识结果。Specifically, the faults obtained by the above-mentioned fault reasoning mechanism can be sorted by probability, and the pre-set number, such as 10, can be selected, and the relevant fields include complete information that is helpful for maintenance personnel to diagnose and repair; further form fault identification results. .
综上,本发明电机故障辨识方法实施例1,对电机进行监测和跟踪;根据监测参数或性能参数的正常范围,当实测参数超过该范围,根据异常参数表征,初步判定发生了什么故障或哪几个故障;根据初步判定的征兆故障,根据之前建立的规则关联和故障记录案例等进行推理,得出是哪些内部隐藏的故障或外部故障,导致电机发生故障。根据推理获得的计算概率,从大到小排列故障原因,获得故障辨识结果,从而制定维修计划。To sum up, the first embodiment of the motor fault identification method of the present invention monitors and tracks the motor; according to the normal range of the monitoring parameters or performance parameters, when the measured parameters exceed the range, according to the abnormal parameter characterization, it is preliminarily determined what fault or which fault occurred. Several faults; based on the preliminary judgment of symptom faults, inferences based on previously established rule associations and fault record cases, etc., to obtain which internal hidden faults or external faults caused the motor to fail. According to the calculated probabilities obtained by reasoning, the causes of the faults are arranged in descending order, and the fault identification results are obtained, so as to formulate the maintenance plan.
本发明实施例提供了一种可不断扩展的故障辨识机制,结合案例、标准和经验形成故障知识管理方法,以及利用这些知识进行故障辨识推理;本发明实施例基于电机故障知识管理和故障辨识机制,具有较强的实用性和可扩展性,对于电机运行的在线故障诊断和离线故障定位,以及快速修复系统故障,提出了综合解决方案。The embodiment of the present invention provides a fault identification mechanism that can be continuously expanded, forms a fault knowledge management method by combining cases, standards and experience, and uses these knowledge to perform fault identification and reasoning; the embodiment of the present invention is based on the motor fault knowledge management and fault identification mechanism. , has strong practicability and scalability, and proposes a comprehensive solution for online fault diagnosis and offline fault location of motor operation, as well as rapid repair of system faults.
在一个具体的实施例中,信号监测参数包括电机的电参数、各部件温度参数、工作参数以及状态参数;性能跟踪参数包括电机功率、电机效率、最小转矩、最大转矩、堵转转矩、堵转电流以及温升;In a specific embodiment, the signal monitoring parameters include electrical parameters of the motor, temperature parameters of various components, working parameters and state parameters; performance tracking parameters include motor power, motor efficiency, minimum torque, maximum torque, stall torque , stall current and temperature rise;
在根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率的步骤之前还包括步骤:Before the step of processing the signal monitoring parameters and the performance tracking parameters based on the fault knowledge base according to the preset fault identification rules, and obtaining the probability of each fault when the motor fails, the steps further include:
构建电机的故障基础信息库;故障基础信息库包括故障属性信息;Build the fault basic information database of the motor; the fault basic information database includes the fault attribute information;
根据故障属性信息,构建故障知识库。Build a fault knowledge base according to fault attribute information.
具体而言,本发明实施例中电机的信号采集方案可以采集与电机故障相关的四类主要参数:三相电压、三相电流、频率等电参数;可通过电压电流钳分别实时监测三相有效值、峰值等,针对某些安全性和可靠性要求很高的场合的电机,可通过录波器等定期监测三相电压电流波形;电机外壳、绕组、轴承等温度参数;外壳温度可通过外壳上贴温度传感器获取;轴承温度可通过在靠近轴承的轴承盖上安装温度传感器获取;绕组温度可通过在每相绕组中预埋热电偶(如PT100)进行监测。电机转速、扭矩、持续运行时间等工作参数;可通过电机输出端安装扭矩传感器实时监测。电机噪声、外壳/轴承盖振动等状态参数。通过外壳、轴承座、或者轴承盖等位置安装振动传感器和噪声测试仪进行实时采集。而用于故障辨识的性能跟踪,可以包括跟踪以下参数:电机功率、电机效率、电机转矩、最小转矩、最大转矩、堵转转矩、堵转电流以及温升。Specifically, the signal acquisition scheme of the motor in the embodiment of the present invention can collect four types of main parameters related to the motor fault: three-phase voltage, three-phase current, frequency and other electrical parameters; the three-phase effective parameters can be monitored in real time through voltage and current clamps respectively For some motors with high safety and reliability requirements, the three-phase voltage and current waveforms can be monitored regularly through a wave recorder; the temperature parameters of the motor casing, windings, bearings, etc.; the casing temperature can be passed through the casing. The temperature of the bearing can be obtained by attaching a temperature sensor; the temperature of the bearing can be obtained by installing a temperature sensor on the bearing cover close to the bearing; the temperature of the winding can be monitored by pre-embedding a thermocouple (such as PT100) in each phase winding. Working parameters such as motor speed, torque, continuous running time, etc.; can be monitored in real time by installing a torque sensor at the motor output. Status parameters such as motor noise, housing/bearing cap vibration, etc. Real-time acquisition is performed by installing vibration sensors and noise testers in the housing, bearing housing, or bearing cover. The performance tracking for fault identification can include tracking the following parameters: motor power, motor efficiency, motor torque, minimum torque, maximum torque, stall torque, stall current, and temperature rise.
图2为本发明电机故障辨识方法实施例中各环节架构示意图;如图2所示,本发明电机故障辨识方法各实施例可以包括电机的故障基础信息管理、状态监测方案、故障知识库构建、故障辨识机制和故障辨识结论等环节。首先,通过基本工作,根据设计、标准、经验,将电机的故障知识信息起来;然后,明确电机所处工作环境的状态监测条件,以及所有能监测参数和能跟踪性能;再得出电机内部故障之间的关联规则,收集故障维修记录收集,进行规整;并对电机进行监测和跟踪;同时根据监测参数或性能参数的正常范围,当实测参数超过该范围,则进行异常报警;在异常报警的同时,可根据异常参数表征,初步判定发生了什么故障或哪几个故障;进一步的,可根据初步判定的征兆故障,根据之前建立的规则关联和故障记录案例等进行推理,得出是哪些内部隐藏或外部的故障导致电机出现故障。最后,可根据推理获得的计算概率,从大到小排列故障原因,获得故障辨识结果,从而制定维修计划。FIG. 2 is a schematic diagram of the structure of each link in the embodiment of the motor fault identification method of the present invention; as shown in FIG. 2 , each embodiment of the motor fault identification method of the present invention may include basic information management of motor faults, state monitoring scheme, fault knowledge base construction, Fault identification mechanism and fault identification conclusions and other links. First, through basic work, according to design, standards, and experience, the fault knowledge of the motor is obtained; then, the state monitoring conditions of the working environment of the motor, as well as all the parameters that can be monitored and the performance can be tracked; then the internal fault of the motor is obtained. According to the association rules between them, collect fault maintenance records, and conduct regularization; and monitor and track the motor; at the same time, according to the normal range of monitoring parameters or performance parameters, when the measured parameters exceed the range, an abnormal alarm will be issued; At the same time, it can preliminarily determine what fault or several faults have occurred based on the abnormal parameter representation; further, based on the preliminarily determined symptom fault, inference can be made based on the previously established rule association and fault record cases, etc., to obtain which internal faults Hidden or external faults cause the motor to fail. Finally, according to the calculated probability obtained by reasoning, the fault causes can be arranged from large to small, and the fault identification result can be obtained, so as to formulate a maintenance plan.
在一个具体的实施例中,故障关联规则为因果关系链;In a specific embodiment, the fault association rule is a causal relationship chain;
构建故障基础信息库的步骤包括:The steps to build a fault base information base include:
根据电机故障特征逐级进行故障对象划分,并引入具备电机故障特征的虚拟结构,得到电机的故障对象划分结构模型;According to the fault characteristics of the motor, the fault objects are divided step by step, and the virtual structure with the fault characteristics of the motor is introduced to obtain the structure model of the fault object division of the motor;
根据故障对象划分结构模型进行故障模式分析,确定电机的各潜在故障;并根据各潜在故障,生成故障属性信息;According to the fault object division structure model, the failure mode analysis is carried out, and the potential faults of the motor are determined; and the fault attribute information is generated according to the potential faults;
构建故障知识库的步骤包括:The steps to build a fault knowledge base include:
根据故障树分析,得到故障属性信息中各故障的因果关系链;According to the fault tree analysis, the causal relationship chain of each fault in the fault attribute information is obtained;
根据故障对象划分结构模型、信号监测参数和性能跟踪参数,生成故障维修案例数据。According to the fault object division structure model, signal monitoring parameters and performance tracking parameters, the fault maintenance case data is generated.
具体而言,图3为本发明电机故障辨识方法实施例中故障对象划分示意图;如图3所示,故障对象是故障模式的发生主体,如电机的发热,可能有三个主体,电机外壳、定子绕组和前后轴承。依据故障辨识的目标深度,决定是否进一步分解故障主体,如轴承发热可分为内圈、外圈和滚珠等不同部件的发热。Specifically, FIG. 3 is a schematic diagram of the division of fault objects in the embodiment of the motor fault identification method of the present invention; as shown in FIG. 3 , the fault object is the main body of the failure mode. Windings and front and rear bearings. According to the target depth of fault identification, it is decided whether to further decompose the main body of the fault. For example, the heating of the bearing can be divided into the heating of different components such as the inner ring, the outer ring and the ball.
本发明实施例中故障对象的划分方法将考虑电机所有潜在故障模式的发生主体,优选的,可在传统物料清单(Bill of Material,BOM)基础上,根据电机故障行为特征,逐级进行故障对象划分,主要包括:The method for classifying fault objects in the embodiment of the present invention will consider the occurrence subjects of all potential fault modes of the motor. Preferably, on the basis of a traditional bill of materials (Bill of Material, BOM), fault objects can be classified step by step according to the fault behavior characteristics of the motor division, mainly including:
1、去掉不会或者几乎不发生故障的对象;1. Remove objects that will not or hardly fail;
2、相同型号的不同对象如果在不同位置有不同的故障表现,需要分别列出并标号(BOM里同型号的往往只给出数量);2. If different objects of the same model have different faults in different positions, they need to be listed and labeled separately (the same model in the BOM often only gives the number);
3、引入有电机故障特征的虚拟结构,包括散热系统(风扇、风罩及其键连接组成,任一故障都可能导致电机系统的散热问题)、轴承润滑系统(包括润滑油、轴承盖、密封件等,润滑问题将引起电机异响、振动、温升等问题)、供电系统(电源、接线盒、接线头等,其三相电压和电流特性对电机影响巨大)、气隙(定子和转子之间的缝隙,太大太小或者不均匀都将影响电机性能)等等;虚拟结构的故障是某类功能异常的综合表述,在故障辨识时,功能异常的提炼可大大促进后续故障关联的管理和故障主体的分解。3. Introduce a virtual structure with motor fault characteristics, including heat dissipation system (composed of fans, hoods and their key connections, any fault may cause heat dissipation problems in the motor system), bearing lubrication system (including lubricating oil, bearing caps, seals, etc.) parts, etc., the lubrication problem will cause abnormal noise, vibration, temperature rise and other problems of the motor), power supply system (power supply, junction box, terminal block, etc., whose three-phase voltage and current characteristics have a huge impact on the motor), air gap (the gap between the stator and the rotor) The gap between them, too large, too small or uneven will affect the performance of the motor), etc.; the fault of the virtual structure is a comprehensive expression of a certain type of functional abnormality. During the fault identification, the extraction of the functional abnormality can greatly promote the management of subsequent fault associations. and breakdown of the fault body.
图4为本发明电机故障辨识方法实施例中故障模式分析示意图;如图4所示,故障模式分析方法在故障对象划分结构的基础上(如上图),自顶向下逐个故障对象明确其潜在的故障模式,如电机,具有无法启动、过热、异响、甩油、冒烟、振动等故障模式。通过“故障对象+故障模式”确定了电机内部和外部所有的潜在故障。FIG. 4 is a schematic diagram of the failure mode analysis in the embodiment of the motor fault identification method of the present invention; as shown in FIG. 4 , the failure mode analysis method is based on the fault object division structure (as shown in the above figure), and the potential fault objects are identified one by one from the top to the bottom. The failure mode of the motor, such as the motor, has failure modes such as failure to start, overheating, abnormal noise, oil rejection, smoke, vibration, etc. All potential faults inside and outside the motor are determined by "fault object + fault mode".
进一步的,故障关联规则指故障发生的因果关系链,优选的,可以参考故障树分析(FTA)自顶向下的分析方法,从故障现象逐层找到其故障原因,图5为本发明电机故障辨识方法实施例中故障关联规则示意图;如图5所示。电机过热可能由定子过热、转子过热、轴承过热、散热异常等原因引起,定子过热等又可能由定转子相擦等引起,如此递推,建立故障基础信息中所有故障的因果关系图,如图5,关联关系通过箭头表示。Further, the fault association rule refers to the causal relationship chain of the fault occurrence. Preferably, the top-down analysis method of fault tree analysis (FTA) can be referred to, and the fault cause can be found layer by layer from the fault phenomenon. Figure 5 shows the fault of the motor in the present invention. A schematic diagram of the fault association rule in the embodiment of the identification method; as shown in FIG. 5 . Motor overheating may be caused by stator overheating, rotor overheating, bearing overheating, abnormal heat dissipation, etc., and stator overheating may be caused by stator and rotor rubbing. 5. The relationship is represented by arrows.
针对上述因果关系图中,逐个故障分析引起该故障的所有故障原因的发生频率,如对于电机过热,其故障原因有定子过热、轴承过热、转子过热、散热异常4种,根据故障统计或专家经验,在100次电机过热中,4种原因引起的频次分别为50次、35次、10次和5次。那么,相应的关联关系箭头可再附上相关的频率经验信息。For the above causal relationship diagram, analyze the frequency of all fault causes that cause the fault one by one. For example, for motor overheating, there are four types of failure causes: stator overheating, bearing overheating, rotor overheating, and abnormal heat dissipation. According to fault statistics or expert experience , in 100 times of motor overheating, the frequencies caused by 4 reasons are 50 times, 35 times, 10 times and 5 times respectively. Then, the corresponding relationship arrows can be attached with relevant frequency experience information.
而本发明实施例中的故障维修案例数据可以包括:The fault maintenance case data in the embodiment of the present invention may include:
故障现象:故障表现的描述;Failure phenomenon: description of failure performance;
工况描述:故障发生前后的环境、工况和负载情况;Description of working conditions: the environment, working conditions and load conditions before and after the failure;
异常参数:故障时刻及前一段时间的信号采集和状态跟踪参数值;Abnormal parameters: signal acquisition and state tracking parameter values at the moment of failure and the previous period;
确诊故障:确诊出现问题的故障,可多个;Diagnosing faults: Diagnosing faults with problems, there can be multiple;
更换备件:维修更换的部件;Replacement Spare Parts: Repair and replace parts;
维修措施:维修的具体过程。Maintenance measures: the specific process of maintenance.
在一个具体的实施例中,故障关联推理结果包括完整故障机理结果和并发故障结果;In a specific embodiment, the fault correlation reasoning result includes a complete fault mechanism result and a concurrent fault result;
根据故障推理规则,处理故障判定发生概率的步骤包括:According to the fault inference rules, the steps of processing the probability of fault determination include:
根据故障关联规则,在确认各故障判定发生概率对应的各故障处于同一条故障关联链路上时,通过概率计算得到完整故障机理结果;According to the fault association rule, when it is confirmed that each fault corresponding to the probability of occurrence of each fault is on the same fault-related link, the complete fault mechanism result is obtained through probability calculation;
根据故障关联规则,在确认各故障判定发生概率对应的各故障处于不同的故障关联链路上时,通过概率计算得到并发故障结果。According to the fault association rule, when it is confirmed that the faults corresponding to the occurrence probability of each fault are on different fault-related links, the concurrent fault result is obtained through probability calculation.
具体而言,本发明电机故障辨识方法各实施例的故障推理机制,将上述根据故障判定规则获得的故障及概率信息代入故障关联规则,如判定出“电机—过热”的概率是0.6,那么在图中,“定子—过热”的概率为0.6*0.5=0.3,转子过热概率为0.6*0.35=0.21,依此类推;多个故障链路上的多个概率可以累加,如定转子相擦可能由定子过热或转子过热导致,那么如图5中定转子相擦的故障概率为:0.3*0.1+0.21*0.6=0.156。Specifically, the fault reasoning mechanism of each embodiment of the motor fault identification method of the present invention substitutes the above-mentioned fault and probability information obtained according to the fault determination rule into the fault association rule. If it is determined that the probability of "motor-overheating" is 0.6, then In the figure, the probability of "stator-overheating" is 0.6*0.5=0.3, the probability of rotor overheating is 0.6*0.35=0.21, and so on; multiple probabilities on multiple faulty links can be accumulated, such as the possibility of stator-rotor friction. Caused by the overheating of the stator or the overheating of the rotor, the failure probability of the stator and rotor rubbing against each other as shown in Figure 5 is: 0.3*0.1+0.21*0.6=0.156.
如果上述故障判定获得的多个故障在一条故障关联链路上,那么这条故障链路很可能为此次故障的完整机理;如上述故障判定规则获得“电机—过热”和“转子—动不平衡”具有一定发生概率,那么很可能是转子不平衡导致气息不均匀,导致定转子相擦,导致定子和转子的温升,导致电机发热。该链路上的故障概率通过相同的故障概率计算方法获得,即完整故障机理结果。If the multiple faults obtained by the above fault judgment are on a fault-related link, then this fault link is likely to be the complete mechanism of the fault; for example, the above fault judgment rules obtain "motor-overheating" and "rotor-motionless" "Balance" has a certain probability of occurrence, then it is likely that the rotor is unbalanced, causing uneven air, causing the stator and rotor to rub, causing the temperature of the stator and the rotor to rise, causing the motor to heat up. The failure probability on this link is obtained by the same failure probability calculation method, that is, the complete failure mechanism result.
如果上述故障判定规则获得的故障较多,且不全在一条故障链路上,则可能发生比较复杂的并发故障,按照相同故障概率计算方法获得各故障的发生概率,即并发故障结果;If there are many faults obtained by the above fault determination rules, and not all of them are on one faulty link, more complex concurrent faults may occur, and the probability of occurrence of each fault is obtained according to the same fault probability calculation method, that is, the concurrent fault result;
本发明实施例中的故障推理规则,还包括将实时监测参数与故障维修案例数据中的异常参数进行比对,如果异常参数全部匹配,则将全匹配的案例作为“相似案例”,故障维修案例数据中的确诊故障赋予一个故障概率,故障概率用本案例的异常参数数量与当前异常参数总数量的比值。多个案例的多个故障的概率进行加权求平均;The fault reasoning rule in the embodiment of the present invention further includes comparing the real-time monitoring parameters with the abnormal parameters in the fault maintenance case data. If all the abnormal parameters match, the fully matched cases are regarded as "similar cases", and the fault maintenance The diagnosed failure in the data is assigned a failure probability, and the failure probability is the ratio of the number of abnormal parameters in this case to the total number of current abnormal parameters. The probability of multiple failures in multiple cases is weighted and averaged;
进一步的,本发明实施例中的故障推理规则将故障关联规则推理结果和案例推理结果进行加权求平均,得到各故障概率。Further, in the fault inference rule in the embodiment of the present invention, the inference result of the fault association rule and the case inference result are weighted and averaged to obtain each fault probability.
在一个具体的实施例中,在根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率的步骤之前还包括步骤:In a specific embodiment, before the step of processing the signal monitoring parameters and performance tracking parameters based on the fault knowledge base according to the preset fault identification rules, and obtaining the probability of each fault when the motor fails, the steps further include:
在监测到信号监测参数、性能跟踪参数超过相应报警值时,发出异常报警;When it is detected that the signal monitoring parameters and performance tracking parameters exceed the corresponding alarm values, an abnormal alarm is issued;
在得到电机的故障辨识结果的步骤之后还包括步骤:After the step of obtaining the fault identification result of the motor, it also includes the following steps:
根据故障辨识结果,确定维修计划;According to the fault identification results, determine the maintenance plan;
根据故障辨识结果和维修计划,更新故障维修案例数据。According to the fault identification result and maintenance plan, update the fault maintenance case data.
具体而言,异常报警机制根据异常报警表逐条匹配,任一条超出正常范围的则进行异常报警。同时还可根据确诊和维修措施,更新故障维修案例数据。Specifically, the abnormal alarm mechanism matches items one by one according to the abnormal alarm table, and if any item exceeds the normal range, an abnormal alarm is issued. At the same time, the fault maintenance case data can be updated according to the diagnosis and maintenance measures.
为了进一步详细阐述本发明技术方案,特以应用本发明电机故障辨识方法的电机故障辨识过程为例进行说明,图6为本发明电机故障辨识方法实施例中故障辨识过程的流程示意图,如图6所示,本发明电机故障辨识方法包括电机的故障基础信息管理、状态监测方案、故障知识库构建、故障辨识机制和故障辨识结论等环节。(需要说明的是,图6中的各小图仅仅用于表示辨识流程中的某一过程,并不用于表示具体的数值、数据或内容)In order to further elaborate the technical solution of the present invention, the motor fault identification process using the motor fault identification method of the present invention is taken as an example for illustration. FIG. 6 is a schematic flowchart of the fault identification process in the embodiment of the motor fault identification method of the present invention, as shown in FIG. 6 As shown, the motor fault identification method of the present invention includes the steps of basic information management of motor faults, state monitoring scheme, fault knowledge base construction, fault identification mechanism and fault identification conclusion. (It should be noted that the small graphs in Figure 6 are only used to represent a certain process in the identification process, and are not used to represent specific numerical values, data or content)
(1)故障基础信息管理(1) Fault basic information management
故障基础信息管理为构建故障辨识所需知识库提供基础信息,包括故障对象划分、故障模式分析、故障信息规整等。Fault basic information management provides basic information for building a knowledge base for fault identification, including fault object division, fault mode analysis, and fault information regulation.
(1.1)故障对象划分(参见上文对应内容)(1.1) Division of fault objects (see the corresponding content above)
(1.2)故障模式分析(参见上文对应内容)(1.2) Failure mode analysis (see the corresponding content above)
(1.3)故障属性信息管理(1.3) Fault attribute information management
针对上述电机所有潜在故障,可依据专家经验(人工输入)逐个确定故障属性信息,电机故障属性信息管理包括:For all potential faults of the above-mentioned motors, the fault attribute information can be determined one by one according to expert experience (manual input). The management of motor fault attribute information includes:
发生级别:根据同类电机的故障维修记录中,故障发生频率进行分级定义,如按百次故障发生次数的占比、或按照年故障发生次数进行划分;Occurrence level: According to the fault maintenance records of similar motors, the fault occurrence frequency is classified and defined, for example, according to the proportion of the number of faults per hundred times, or according to the number of fault occurrences per year;
故障类型:分为相关故障和非相关故障,相关故障指间接故障,即由其他故障引起且通过故障原因的修复可消除的故障;非相关故障指直接由本身引起、需要维修/更换当前故障对象才能得以修复的故障。故障类型的划分,对于故障辨识后维修任务的确认具有意义,只有非相关故障才需要维护。Fault type: divided into related faults and non-related faults, related faults refer to indirect faults, that is, faults caused by other faults and can be eliminated by repairing the cause of the fault; non-related faults refer to directly caused by itself and need to be repaired/replaced the current fault object faults that can be repaired. The division of fault types is meaningful for the confirmation of maintenance tasks after fault identification. Only non-related faults need maintenance.
确诊方式:非相关故障必须,即如何确诊该故障是否发生的方式,包括运行排查,即正常运行时可排查;停机排查,即需要停机后进行特定操作才能排查;拆机排查,即停机后还需要拆卸才可排查。Diagnosis method: non-related faults must, that is, how to diagnose whether the fault occurs, including operation inspection, that is, it can be checked during normal operation; shutdown inspection, that is, it needs to perform specific operations after shutdown; dismantling inspection, that is, after shutdown It needs to be disassembled for inspection.
确诊设备:确诊所需要的设备,常用的电机确诊设备有目检、电流表、转速仪、温度记录仪、功率分析仪、激光对中仪、测振仪、电阻仪、耐压仪、动平衡仪、专业检修等。需根据电机应用现场的仪器设备情况具体确定,如现场若有测振仪,可通过“运行排查”方式确诊某些振动问题,若没有测振仪,则只能进行停机排查甚至拆机排查,甚至送专业检修。Diagnosis equipment: equipment required for diagnosis, commonly used motor diagnosis equipment includes visual inspection, ammeter, tachometer, temperature recorder, power analyzer, laser alignment instrument, vibration meter, resistance meter, withstand voltage meter, dynamic balancer , professional maintenance, etc. It needs to be determined according to the equipment conditions of the motor application site. If there is a vibration meter on site, some vibration problems can be diagnosed through the "operation inspection" method. Even send it to a professional for repair.
确诊方法:描述如何利用确诊设备进行故障确诊的具体操作;Diagnosis method: describe the specific operation of how to use the diagnosis equipment to diagnose the fault;
维修措施:确诊故障发生后,如何进行维修维护操作才能排除故障。Maintenance measures: After the fault is diagnosed, how to carry out maintenance and repair operations can eliminate the fault.
具体可如下表表1所示。The details are shown in Table 1 below.
表1-电机故障属性信息管理Table 1 - Motor fault attribute information management
(2)状态监测(2) Condition monitoring
状态监测是故障辨识的来源依据,包括信号采集方案和状态跟踪方案两部分。Condition monitoring is the source basis of fault identification, including signal acquisition scheme and status tracking scheme.
(2.1)信号采集方案(2.1) Signal acquisition scheme
本发明电机的信号采集方案可以采集与电机故障相关的主要参数,如下表表2所示:The signal acquisition scheme of the motor of the present invention can collect the main parameters related to the motor fault, as shown in Table 2 below:
(2.2)状态跟踪方案;(2.2) Status tracking scheme;
在信号采集的基础上,需进一步计算获得电机的各状态参数,如下表表3所示。On the basis of signal acquisition, it is necessary to further calculate and obtain various state parameters of the motor, as shown in Table 3 below.
(3)故障知识库构建(3) Construction of fault knowledge base
故障知识库是完成所述故障辨识的故障知识,包括故障关联规则、性能跟踪记录、异常征兆判据和故障维修案例;The fault knowledge base is the fault knowledge for completing the fault identification, including fault association rules, performance tracking records, abnormal symptom criteria and fault maintenance cases;
表2-电机信号监测参数Table 2 - Motor Signal Monitoring Parameters
表3-电机性能跟踪参数Table 3 - Motor Performance Tracking Parameters
(3.1)故障关联规则库(3.1) Fault association rule base
故障关联规则指故障发生的因果关系链,参考故障树分析(FTA)自顶向下的分析方法,从故障现象逐层找到其故障原因,如下图所示。电机过热可能由定子过热、转子过热、轴承过热、散热异常等原因引起,定子过热等又可能由定转子相擦等引起,如此递推,建立故障基础信息中所有故障的因果关系图The fault association rule refers to the causal relationship chain of fault occurrence. Refer to the top-down analysis method of fault tree analysis (FTA) to find the fault cause layer by layer from the fault phenomenon, as shown in the following figure. Motor overheating may be caused by stator overheating, rotor overheating, bearing overheating, abnormal heat dissipation, etc., and stator overheating may be caused by stator-rotor rubbing, etc. In this way, a causal relationship diagram of all faults in the basic fault information is established.
(3.2)性能跟踪记录库(3.2) Performance tracking library
性能跟踪记录库将实时监测和保存电机实际运行过程中(2.1)状态采集参数和(2.2)性能跟踪参数;The performance tracking record library will monitor and save the (2.1) state acquisition parameters and (2.2) performance tracking parameters in the actual running process of the motor in real time;
(3.3)故障维修案例库(3.3) Fault maintenance case library
故障维修案例数据包括:故障现象:故障表现的描述;工况描述:故障发生前后的环境、工况和负载情况;异常参数:故障时刻及前一段时间的信号采集和状态跟踪参数值,引用(3.2)性能跟踪记录;确诊故障:确诊出现问题的故障,可多个,引用(1.3)故障信息;更换备件:维修更换的部件,引用(1.1)故障对象;维修措施:维修的具体过程。The fault maintenance case data includes: fault phenomenon: description of fault performance; working condition description: environment, working conditions and load conditions before and after the fault occurs; abnormal parameters: signal acquisition and state tracking parameter values at the time of fault and a period of time before, refer to ( 3.2) Performance tracking records; Diagnosis of faults: Diagnose the faults with problems, which can be multiple, and refer to (1.3) fault information; Replace spare parts: Repair and replace parts, refer to (1.1) fault objects; Maintenance measures: The specific process of maintenance.
其中,需要说明的是,故障维修数据库记录了以往的故障维修数据。具体而言,如果要对当前某个电机进行故障辨识,应用本发明可首先将它以往的历史故障维修记录(包括更换的部件)存储起来,现在要做故障辨识的其中一种辨识方案为根据当前的实时参数与案例库中“实施参数”进行“匹配”,匹配到的记录就表示以往发生类似异常时,是什么原因导致,做了什么维修,进而可将这些信息提供给维修人员,同时在根据这些辨识结论进行故障确诊及维修后,把更换的备件,维修过程等信息再录回案例库,这样案例库就多了一条记录。Among them, it should be noted that the fault maintenance database records the past fault maintenance data. Specifically, if the fault identification of a current motor is to be carried out, the present invention can firstly store its past historical fault maintenance records (including replaced parts), and now one of the identification schemes for fault identification is based on The current real-time parameters are "matched" with the "implementation parameters" in the case library, and the matched records indicate what caused similar abnormalities in the past, and what maintenance was done. This information can then be provided to maintenance personnel, and at the same time After the fault diagnosis and maintenance are carried out based on these identification conclusions, the replaced spare parts, maintenance process and other information are recorded back to the case database, so that there is one more record in the case database.
(4)故障辨识机制(4) Fault identification mechanism
故障辨识机制解决通过监测参数,结合故障知识库,如何进行异常报警,故障定位并给出维修方案的问题。The fault identification mechanism solves the problem of how to perform abnormal alarms, locate faults and provide maintenance plans by monitoring parameters and combining with the fault knowledge base.
(4.1)异常报警机制(4.1) Abnormal alarm mechanism
所述异常报警机制根据异常报警表逐条匹配,任一条超出正常范围的则进行异常报警。异常报警表(如下表表4所示)包括:GB/T 1032-2012等相关标准要求,如对温升、振动、噪声级别的要求;电机的具体设计要求,如效率、转矩等各类性能参数的容差上下限;专家经验,标准和设计要求可能相对宽泛,可以根据经验对某些报警值进行调整。The abnormal alarm mechanism matches items one by one according to the abnormal alarm table, and if any item exceeds the normal range, an abnormal alarm is performed. The abnormal alarm table (as shown in Table 4 below) includes: GB/T 1032-2012 and other relevant standard requirements, such as requirements for temperature rise, vibration, and noise level; specific design requirements of the motor, such as efficiency, torque, etc. Upper and lower tolerance limits for performance parameters; expert experience, standards and design requirements may be relatively broad, and some alarm values can be adjusted based on experience.
表4-异常报警表Table 4 - Abnormal alarm table
(4.2)故障判定机制(4.2) Fault determination mechanism
异常报警只是告诉你是否可能出现故障,但不能判断到底哪里出现了故障。Anomaly alarms only tell you if there might be a fault, but not where the fault is.
所述故障判定机制是在异常报警的同时,将(2.1)采集参数和(2.2)性能参数与征兆判据库逐条匹配,征兆判据库由故障及发生故障对应的征兆组成,故障引用所述(1.3),征兆为所述(2.1)和(2.2)。每条故障及其征兆记录可由领域专家审核确认。如下表-表5所示:The fault determination mechanism is to match the (2.1) acquisition parameters and (2.2) performance parameters with the symptom criterion database one by one while the abnormal alarm occurs. The symptom criterion database is composed of the fault and the symptom corresponding to the failure. (1.3), the indications are described in (2.1) and (2.2). Each fault and its symptom records can be reviewed and confirmed by domain experts. As shown in the following table - table 5:
征兆判据库是假设故障发生,判断可能引起的监测参数的变化;而故障判定是逆向的,根据监测的参数变化判断哪些故障发生了,由于一个故障可能对应多个参数变化,而一个参数变化也可能由多个故障引起,因此故障判定只能通过征兆匹配数目进行大致的故障发生概率的计算。The symptom criterion database assumes that a fault occurs, and judges the possible changes in monitoring parameters; the fault judgment is reversed, and judges which faults have occurred according to the changes of the monitored parameters. Because one fault may correspond to multiple parameter changes, and one parameter changes It may also be caused by multiple faults, so the fault judgment can only be used to calculate the approximate fault occurrence probability through the matching number of symptoms.
因此,将各实测参数与上述征兆判据库进行逐条匹配,每条记录中,一个故障对应多个征兆,征兆被匹配的数量越多,该故障发生的概率越大。以匹配征兆数与总征兆数的比值作为该故障的发生概率。具体的,当前监测到m个征兆;征兆库第X条记录有n个征兆;如果n中有k个被匹配,则k/n就是概率。Therefore, each measured parameter is matched with the above symptom criterion database one by one. In each record, one fault corresponds to multiple symptoms. The more the matched symptoms, the greater the probability of the fault. Take the ratio of the number of matching symptoms to the total number of symptoms as the probability of occurrence of the fault. Specifically, m signs are currently monitored; there are n signs in the X record of the sign library; if k out of n are matched, then k/n is the probability.
(4.3)故障推理机制(4.3) Fault reasoning mechanism
上述的故障判定可以得出电机出现了哪些故障,但无法定位出全部的故障原因以及完整的故障机理。因此,所述的故障推理机制,将上述(4.2)故障判定获得的故障及概率信息代入(3.1)故障关联规则,如判定出“电机—过热”的概率是0.6,那么在图中,“定子—过热”的概率为0.6*0.5=0.3,转子过热概率为0.6*0.35=0.21,依此类推;多个故障链路上的多个概率可以累加,如定转子相擦可能由定子过热或转子过热导致,那么如示例图中定转子相擦的故障概率为:0.3*0.1+0.21*0.6=0.156。The above fault determination can determine which faults have occurred in the motor, but cannot locate all the fault causes and the complete fault mechanism. Therefore, the fault reasoning mechanism described above substitutes the fault and probability information obtained from the above (4.2) fault judgment into the fault association rule (3.1), if it is determined that the probability of "motor-overheating" is 0.6, then in the figure, "stator" The probability of "overheating" is 0.6*0.5=0.3, the probability of rotor overheating is 0.6*0.35=0.21, and so on; multiple probabilities on multiple fault links can be accumulated, such as stator-rotor phase friction may be caused by stator overheating or rotor overheating Caused by overheating, the failure probability of stator and rotor rubbing as shown in the example figure is: 0.3*0.1+0.21*0.6=0.156.
如果上述(4.2)故障判定获得的多个故障在一条故障关联链路上,那么这条故障链路很可能为此次故障的完整机理;如上述(4.2)判定获得“电机—过热”和“转子—动不平衡”具有一定发生概率,那么很可能是转子不平衡导致气息不均匀,导致定转子相擦,导致定子和转子的温升,导致电机发热。该链路上的故障概率通过相同的故障概率计算方法获得。If multiple faults obtained by the above (4.2) fault judgment are on a fault-related link, then this fault link is likely to be the complete mechanism of the fault; as in the above (4.2) judgment, the “motor-overheating” and “motor-overheating” are obtained. Rotor-dynamic unbalance" has a certain probability of occurrence, then it is likely that the rotor is unbalanced, causing uneven air, causing the stator and rotor to rub, causing the temperature of the stator and the rotor to rise, causing the motor to heat up. The failure probability on this link is obtained by the same failure probability calculation method.
如果上述(4.2)获得的故障较多,且不全在一条故障链路上,则可能发生比较复杂的并发故障,按照相同故障概率计算方法获得各故障的发生概率;If there are many faults obtained in the above (4.2), and not all of them are on one faulty link, more complex concurrent faults may occur, and the probability of occurrence of each fault is obtained according to the same fault probability calculation method;
故障推理机制,还包括将实时监测参数与(3.4)案例库的异常参数进行比对,如果异常参数全部匹配,则将全匹配的案例作为“相似案例”,案例库(即故障维修案例数据)中的确诊故障赋予一个故障概率,故障概率用本案例的异常参数数量与当前异常参数总数量的比值。多个案例的多个故障的概率进行加权求平均;The fault reasoning mechanism also includes comparing the real-time monitoring parameters with the abnormal parameters of the (3.4) case database. If all the abnormal parameters match, the fully matched cases will be regarded as "similar cases", and the case database (that is, the fault maintenance case data) The diagnosed fault in is assigned a fault probability, and the fault probability is the ratio of the number of abnormal parameters in this case to the total number of current abnormal parameters. The probability of multiple failures in multiple cases is weighted and averaged;
基于故障推理机制(即故障推理规则),将故障关联规则推理结果和案例推理结果进行加权求平均,得到各故障概率。Based on the fault inference mechanism (ie, fault inference rules), the inference results of the fault association rules and the case inference results are weighted and averaged to obtain the probability of each fault.
(5)维修计划(5) Maintenance plan
所述维修计划表将上述故障推理机制获得的故障按概率大小进行排序,取前10个,相关字段包括有助于维修人员确诊检修相关的完整信息;In the maintenance schedule table, the faults obtained by the above-mentioned fault reasoning mechanism are sorted by probability, and the top 10 are selected, and the relevant fields include complete information related to the diagnosis and maintenance of the maintenance personnel;
故障:引用1.3,包括故障对象(1.1)和故障模式(1.2)Failures: Reference 1.3, including failure objects (1.1) and failure modes (1.2)
发生概率:引用(4.3)Probability of Occurrence: Reference (4.3)
故障类型:应用(1.3)Fault Type: Application (1.3)
确诊设备:引用(1.3)Diagnostic Equipment: References (1.3)
确诊方法:引用(1.3)Diagnosis method: reference (1.3)
维修措施:引用(1.3)Maintenance Actions: References (1.3)
表5-征兆判据库Table 5 - Symptom Criterion Database
根据4.2推理,形成故障辨识结果,如下表-表6所示。According to the reasoning in 4.2, the fault identification result is formed, as shown in the following table-table 6.
表6-故障辨识结果Table 6 - Fault Identification Results
最后根据确诊和维修措施,更新故障案例(3.3)。Finally, update the fault case (3.3) according to the diagnosis and maintenance measures.
本发明电机故障辨识方法各实施例中,故障知识是故障辨识的基础依据,本发明提供了一种良好的、可扩展的管理办法,是面向不同电机实现有效故障辨识的基本条件,本发明提出的故障知识管理办法是可扩展的:1.故障基础管理模块,可以根据不同电机的基本结构组成不同、故障表现不同、现场测试手段不同,分别建立不同的故障基础数据库;2.状态监测模块,可根据不同电机应用现场的条件,布置安装不同的传感器,系统在信号采集和性能跟踪项的设置上,也可分别设置;3.在故障知识库模块中,每个电机不同功率大小、不同应用场合的阈值是不一样的,需要根据对象及其应用现场来确定,可更新。In each embodiment of the motor fault identification method of the present invention, fault knowledge is the basic basis for fault identification. The present invention provides a good and scalable management method, which is the basic condition for effective fault identification for different motors. The present invention proposes The fault knowledge management method is scalable: 1. The fault basic management module can establish different fault basic databases according to the basic structure of different motors, different fault performance, and different on-site testing methods; 2. Condition monitoring module, Different sensors can be arranged and installed according to the conditions of different motor application sites, and the system can also be set separately in the setting of signal acquisition and performance tracking items; 3. In the fault knowledge base module, each motor has different power levels and different applications. The thresholds are different for different occasions and need to be determined according to the object and its application site, and can be updated.
而本发明实施例对运行中的电机进行信号采集与监测,得到电机的信号监测参数,进而对电机进行性能跟踪,通过预设故障辨识规则,基于完整的故障知识库处理相关参数,进而从底层实现故障管理、性能跟踪、判据管理与征兆管理,同时采用与故障辨识相协调的故障推理机制,基于故障维修案例数据,得到能够辅助维修人员进行确诊和检修的故障辨识结论。本发明实施例提供了一种可不断扩展的故障辨识机制,结合案例、标准和经验形成故障知识管理方法,以及利用这些知识进行故障辨识推理;本发明实施例基于电机故障知识管理和故障辨识机制,具有较强的实用性和可扩展性,对于电机运行的在线故障诊断和离线故障定位,以及快速修复系统故障,提出了综合解决方案。In the embodiment of the present invention, signals are collected and monitored for the motor in operation, and the signal monitoring parameters of the motor are obtained, and then the performance of the motor is tracked, and the relevant parameters are processed based on the complete fault knowledge base by presetting fault identification rules, and then from the bottom layer. It realizes fault management, performance tracking, criterion management and symptom management. At the same time, a fault reasoning mechanism coordinated with fault identification is adopted. Based on fault maintenance case data, fault identification conclusions that can assist maintenance personnel in diagnosis and maintenance are obtained. The embodiment of the present invention provides a fault identification mechanism that can be continuously expanded, forms a fault knowledge management method by combining cases, standards and experience, and uses these knowledge to perform fault identification and reasoning; the embodiment of the present invention is based on the motor fault knowledge management and fault identification mechanism. , has strong practicability and scalability, and proposes a comprehensive solution for online fault diagnosis and offline fault location of motor operation, as well as rapid repair of system faults.
本发明电机故障辨识系统实施例1:
为了解决传统故障辨识方法可扩展性和实用性较差的问题,本发明提供了一种电机故障辨识系统实施例1;图7为本发明电机故障辨识系统实施例1的结构示意图。如图7所示,可以包括In order to solve the problem of poor scalability and practicability of the traditional fault identification method, the present invention provides a first embodiment of a motor fault identification system; FIG. 7 is a schematic structural diagram of the first embodiment of the motor fault identification system of the present invention. As shown in Figure 7, it can include
状态监测单元710,用于对运行中的电机进行信号采集监测,得到电机的信号监测参数;The
性能跟踪单元720,用于处理信号监测参数,得到电机的性能跟踪参数;A
故障辨识单元730,用于根据预设故障辨识规则,基于故障知识库处理信号监测参数、性能跟踪参数,得到电机发生故障时的各故障概率;故障知识库包括故障关联规则和故障维修案例数据;预设故障辨识规则包括故障判定规则和故障推理规则;故障判定规则包括将信号监测参数、性能跟踪参数,分别与预设征兆判据库中各征兆记录进行逐条匹配,得到电机发生故障时各故障判定发生概率;故障推理规则包括将根据故障关联规则处理各故障判定发生概率、得到的故障关联推理结果,与将信号监测参数和性能跟踪参数分别与故障维修案例数据进行比对、得到的案例推理结果,进行加权求平均,得到各故障概率;The
排序单元740,用于对各故障概率进行大小排序,得到电机的故障辨识结果。在其中一个实施例中,The
在一个具体的实施例中,信号监测参数包括电机的电参数、各部件温度参数、工作参数以及状态参数;性能跟踪参数包括电机功率、电机效率、最小转矩、最大转矩、堵转转矩、堵转电流以及温升;In a specific embodiment, the signal monitoring parameters include electrical parameters of the motor, temperature parameters of various components, working parameters and state parameters; performance tracking parameters include motor power, motor efficiency, minimum torque, maximum torque, stall torque , stall current and temperature rise;
还包括:Also includes:
故障基础信息管理单元750,用于构建电机的故障基础信息库;故障基础信息库包括故障属性信息;A fault basic
故障知识库单元760,用于根据故障属性信息,构建故障知识库。The fault
在一个具体的实施例中,故障关联规则为因果关系链;In a specific embodiment, the fault association rule is a causal relationship chain;
故障基础信息管理单元750包括:The fault basic
故障对象划分模块752,用于根据电机故障特征逐级进行故障对象划分,并引入具备电机故障特征的虚拟结构,得到电机的故障对象划分结构模型;The fault object division module 752 is used to divide the fault objects step by step according to the fault characteristics of the motor, and introduce a virtual structure with the fault characteristics of the motor to obtain the fault object division structure model of the motor;
故障模式分析模块754,用于根据故障对象划分结构模型进行故障模式分析,确定电机的各潜在故障;The failure mode analysis module 754 is configured to analyze the failure mode according to the fault object division structure model, and determine each potential failure of the motor;
故障属性信息管理模块756,用于根据各潜在故障,生成故障属性信息;A fault attribute
故障知识库单元760包括:The fault
故障关联规则模块762,用于根据故障树分析,得到故障属性信息中各故障的因果关系链;The fault association rule module 762 is configured to obtain the causal relationship chain of each fault in the fault attribute information according to the fault tree analysis;
故障维修案例库模块764,用于根据故障对象划分结构模型、信号监测参数和性能跟踪参数,生成故障维修案例数据。The fault maintenance case library module 764 is configured to divide the structural model, signal monitoring parameters and performance tracking parameters according to the fault object, and generate fault maintenance case data.
具体而言,需要说明的是,上述电机故障辨识系统各实施例中的单元模块,能够对应实现上述电机故障辨识方法实施例中的各流程步骤,此处不再重复赘述。Specifically, it should be noted that the unit modules in each embodiment of the above-mentioned motor fault identification system can correspondingly implement each process step in the above-mentioned embodiment of the above-mentioned motor fault identification method, which will not be repeated here.
在一个实施例中,还提供一种计算机设备,该计算机设备包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,其中,处理器执行所述程序时实现如上述各实施例中的任意一种电机故障辨识方法。In one embodiment, a computer device is also provided, the computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements each of the above-mentioned programs when executing the program. Any one of the motor fault identification methods in the embodiments.
该计算机设备,其处理器执行程序时,通过实现如上述各实施例中的任意一种电机故障辨识方法,从而可以结合案例、标准和经验形成故障知识管理方法,以及利用这些知识进行故障辨识推理;本发明实施例基于电机故障知识管理和故障辨识机制,具有较强的实用性和可扩展性,对于电机运行的在线故障诊断和离线故障定位,以及快速修复系统故障,提出了综合解决方案。In the computer equipment, when the processor executes the program, by implementing any one of the motor fault identification methods in the above-mentioned embodiments, a fault knowledge management method can be formed in combination with cases, standards and experience, and the fault identification and reasoning can be carried out by using these knowledge. ; Based on the knowledge management and fault identification mechanism of motor faults, the embodiment of the present invention has strong practicability and scalability, and provides a comprehensive solution for online fault diagnosis and offline fault location of motor operation, as well as rapid repair of system faults.
此外,本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于一非易失性的计算机可读取存储介质中,如本发明实施例中,该程序可存储于计算机系统的存储介质中,并被该计算机系统中的至少一个处理器执行,以实现包括如上述各电机故障辨识方法的实施例的流程。In addition, those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable In the storage medium, as in the embodiment of the present invention, the program may be stored in the storage medium of the computer system, and executed by at least one processor in the computer system, so as to realize the embodiments including the above-mentioned methods for identifying motor faults process.
在一个实施例中,还提供一种存储介质,其上存储有计算机程序,其中,该程序被处理器执行时实现如上述各实施例中的任意一种电机故障辨识方法。其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)或随机存储记忆体(RandomAccess Memory,RAM)等。In one embodiment, a storage medium is also provided, on which a computer program is stored, wherein when the program is executed by a processor, any one of the motor fault identification methods in the foregoing embodiments is implemented. The storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or a random access memory (Random Access Memory, RAM) or the like.
该计算机存储介质,其存储的计算机程序,通过实现包括如上述各电机故障辨识方法的实施例的流程,从而可以结合案例、标准和经验形成故障知识管理方法,以及利用这些知识进行故障辨识推理;本发明实施例基于电机故障知识管理和故障辨识机制,具有较强的实用性和可扩展性,对于电机运行的在线故障诊断和离线故障定位,以及快速修复系统故障,提出了综合解决方案。The computer storage medium, and the computer program stored therein, can form a fault knowledge management method in combination with cases, standards and experience by implementing the processes including the embodiments of the above-mentioned motor fault identification methods, and use these knowledge to perform fault identification and reasoning; Based on the motor fault knowledge management and fault identification mechanism, the embodiments of the present invention have strong practicability and scalability, and provide a comprehensive solution for online fault diagnosis and offline fault location of motor operation, as well as quick repair of system faults.
以上所述实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity, all possible combinations of the technical features in the above-described embodiments are not described. However, as long as there is no contradiction between the combinations of these technical features, All should be regarded as the scope described in this specification.
以上所述实施例仅表达了本发明的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本发明构思的前提下,还可以做出若干变形和改进,这些都属于本发明的保护范围。因此,本发明专利的保护范围应以所附权利要求为准。The above-mentioned embodiments only represent several embodiments of the present invention, and the descriptions thereof are specific and detailed, but should not be construed as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can also be made, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711269578.7A CN108107360B (en) | 2017-12-05 | 2017-12-05 | Motor fault identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711269578.7A CN108107360B (en) | 2017-12-05 | 2017-12-05 | Motor fault identification method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108107360A CN108107360A (en) | 2018-06-01 |
CN108107360B true CN108107360B (en) | 2020-07-17 |
Family
ID=62209025
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711269578.7A Active CN108107360B (en) | 2017-12-05 | 2017-12-05 | Motor fault identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108107360B (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109211555A (en) * | 2018-10-15 | 2019-01-15 | 安徽旭辰达电子科技有限公司 | A kind of holder fault self-diagnosis system |
CN109447187B (en) * | 2018-12-25 | 2021-06-15 | 中南大学 | Motor fault diagnosis method and system |
CN109740770A (en) * | 2018-12-29 | 2019-05-10 | 广州供电局有限公司 | Unit exception processing method, device, computer equipment and storage medium |
CN110247473A (en) * | 2019-03-26 | 2019-09-17 | 国网浙江桐乡市供电有限公司 | A kind of monitoring system exception information shortage probability analysis method |
CN110554316B (en) * | 2019-09-16 | 2021-08-13 | 上海电机系统节能工程技术研究中心有限公司 | A method of motor fault diagnosis |
CN111177485B (en) * | 2019-12-16 | 2023-06-27 | 中建材智慧工业科技有限公司 | Parameter rule matching based equipment fault prediction method, equipment and medium |
CN111408484A (en) * | 2019-12-25 | 2020-07-14 | 西安水牧环境科技有限公司 | A horizontal screw discharge centrifuge |
CN112304608B (en) * | 2020-10-14 | 2021-11-23 | 中车株洲电力机车研究所有限公司 | Fault diagnosis method and system for traction transmission system and related components |
CN112396197A (en) * | 2020-12-18 | 2021-02-23 | 三门核电有限公司 | Equipment fault diagnosis system and method based on information fusion technology |
CN112990493B (en) * | 2021-02-25 | 2023-04-07 | 中冶南方连铸技术工程有限责任公司 | Automatic equipment fault identification method and system |
CN112988714B (en) * | 2021-04-06 | 2023-07-07 | 中车青岛四方机车车辆股份有限公司 | Bearing fault database construction method, analysis method and system |
CN115310503A (en) * | 2021-05-06 | 2022-11-08 | 芜湖美智空调设备有限公司 | Fault diagnosis method, device and computer readable storage medium |
CN115310504A (en) * | 2021-05-06 | 2022-11-08 | 芜湖美智空调设备有限公司 | Air conditioner fault detection method and device and storage medium |
CN113377564B (en) * | 2021-06-08 | 2022-06-14 | 珠海格力电器股份有限公司 | Fault diagnosis method and device, computer equipment and storage medium |
CN114994528A (en) * | 2022-05-17 | 2022-09-02 | 中科航星科技有限公司 | Fault diagnosis and analysis method for permanent magnet synchronous motor of aviation turbojet engine |
CN115860123A (en) * | 2023-03-02 | 2023-03-28 | 哈尔滨电机厂有限责任公司 | Fault diagnosis reasoning and checking method for water turbine |
CN116500441B (en) * | 2023-06-30 | 2023-08-29 | 无锡中基电机制造有限公司 | Motor fault detection and positioning method and system |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102496028B (en) * | 2011-11-14 | 2013-03-20 | 华中科技大学 | Breakdown maintenance and fault analysis method for complicated equipment |
CN105302112A (en) * | 2015-10-23 | 2016-02-03 | 中国电子科技集团公司第十研究所 | Intelligent fault diagnosis system for ICNI system |
CN106919981A (en) * | 2015-12-24 | 2017-07-04 | 北京航天测控技术有限公司 | A kind of knowledge acquisition and management system towards comprehensive diagnos engineering |
CN105590146A (en) * | 2016-02-29 | 2016-05-18 | 上海带来科技有限公司 | Power plant device intelligent prediction overhaul method and power plant device intelligent prediction overhaul system based on big data |
-
2017
- 2017-12-05 CN CN201711269578.7A patent/CN108107360B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN108107360A (en) | 2018-06-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108107360B (en) | Motor fault identification method and system | |
Siddiqui et al. | Health monitoring and fault diagnosis in induction motor-a review | |
Amanuel et al. | Design of Vibration Frequency Method with Fine-Tuned Factor for Fault Detection of Three Phase Induction Motor | |
CN110987166A (en) | Prediction method, device, equipment and storage medium for the health state of rotating machinery | |
Irfan et al. | Condition monitoring of induction motors via instantaneous power analysis | |
Swana et al. | Investigation of combined electrical modalities for fault diagnosis on a wound-rotor induction generator | |
Laayati et al. | Smart energy management system: SCIM diagnosis and failure classification and prediction using energy consumption data | |
Leonidovich et al. | The development and use of diagnostic systems and estimation of residual life in industrial electrical equipment | |
Choudhary et al. | Fault diagnosis of induction motor under varying operating condition | |
Djagarov et al. | Overview of diagnostic methods for rotating electrical machines | |
Babu et al. | Review on various signal processing techniques for predictive maintenance | |
CN117514883A (en) | Large-scale vertical fan vibration large processing method | |
Naveena et al. | Fault diagnosis of VSI fed induction motor drive using fuzzy logic approach | |
Geethanjali et al. | Fault diagnosis of induction motors using motor current signature analysis: A review | |
Pharne et al. | Fault diagnosis of motor using fuzzy logic technique | |
Jose et al. | Induction motor fault diagnosis methods: A comparative study | |
Priyanka et al. | Condition monitoring of induction motors through simulation of bearing fault and air gap eccentricity fault | |
Prasad et al. | Smart Motor Analytics: Real-Time Internal and External Parameters Identification Through IoT | |
Bessous et al. | Detailed study of rolling bearing element faults in rotating electrical Machines using MVSA technique | |
Mustafa et al. | Experimental research on machinery fault simulator (MFS): A review | |
Gundewar et al. | Condition Monitoring and Fault Diagnosis of BLDC Motor in Electric Vehicles Using Artificial Intelligence | |
Al-Leythi et al. | Motor Watch-Motor Fault Signature Analysis | |
Badajena et al. | Realizing On-Chip Digital Twin for Event Tracking in Squirrel Cage Induction Motors and Drives | |
Komatsu et al. | Data generation method for domain adaptation in fault diagnosis using motor current signals | |
Wang | Induction motor bearing fault detection using a sensorless approach |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 511300 No.78, west of Zhucun Avenue, Zhucun street, Zengcheng District, Guangzhou City, Guangdong Province Applicant after: CHINA ELECTRONIC PRODUCT RELIABILITY AND ENVIRONMENTAL TESTING Research Institute Address before: 510610 No. 110 Zhuang Road, Tianhe District, Guangdong, Guangzhou, Dongguan Applicant before: CHINA ELECTRONIC PRODUCT RELIABILITY AND ENVIRONMENTAL TESTING Research Institute |
|
GR01 | Patent grant | ||
GR01 | Patent grant |