CN110008245B - Method suitable for searching equipment fault early warning model time period - Google Patents
Method suitable for searching equipment fault early warning model time period Download PDFInfo
- Publication number
- CN110008245B CN110008245B CN201910276801.3A CN201910276801A CN110008245B CN 110008245 B CN110008245 B CN 110008245B CN 201910276801 A CN201910276801 A CN 201910276801A CN 110008245 B CN110008245 B CN 110008245B
- Authority
- CN
- China
- Prior art keywords
- data
- time period
- conditions
- screening
- interpolation
- 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 20
- 238000012216 screening Methods 0.000 claims abstract description 39
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 230000014509 gene expression Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The method suitable for searching the equipment fault early-warning model time period comprises the following steps of; (1) When a user uses the software, obtaining a final required time period through twice screening, firstly, primarily screening the time period meeting the conditions according to the primary screening conditions, and then, precisely screening the time period meeting the conditions according to the model measuring points and a certain calculation rule; (2) The user uses the software to execute the request according to the operation, the searched time period meeting the conditions is returned to the execution result, the calculation result can be derived, and the time period not meeting the conditions is recorded by the log.
Description
Technical Field
The invention relates to the technical field of power plant informatization, in particular to a method suitable for searching a time period of an equipment fault early warning model.
Background
The equipment fault early warning system is different from the traditional parameter modeling mode, and is characterized in that a special data modeling technology based on a similarity principle is used, a real-time dynamic model of equipment is quickly created by automatically collecting massive real-time/historical data of the equipment to perform similarity analysis, a similarity curve reflecting the state of the equipment is generated aiming at the equipment in real time, and real-time dynamic intelligent early warning of early equipment faults of the equipment fault early warning system is realized. Through the implementation of the equipment fault early warning system, a user can establish an early fault monitoring early warning system of the running state of main production equipment in the whole enterprise and a factory-level equipment performance management system platform, so that the potential accident occurrence probability of equipment is greatly reduced, the unplanned downtime of key equipment is shortened, the reliability and the utilization rate of the equipment are improved, and the maintenance running cost of the equipment is reduced.
The modeling principle of the equipment fault early warning system is to generate a process object from a set of historical data, wherein the historical data used for generating an equipment model can meet the following requirements:
1. covering a long enough run time;
2. each set of data expresses a normal state of the device object;
3. the simultaneity of the variables in each set of sample values is satisfied, and the sample values must be uniform time samples.
For example, a device object is modeled, the number of model measurement points (variables) included in the device model is 12, and the device object is sampled every 1 minute for 168 hours, so that 10080 sample value groups are obtained, and the whole of the sample value groups constitutes historical data of the device modeling.
Each set of sampling values represents an operating state of the device object, and the historical data is a set of normal states in the actual operation of the device object. The function of the model generator is to extract the state points from the state sets that are most representative of the characteristics of the device object, and then construct the device model using these state points.
When equipment fault early warning modeling is carried out, it is important to add historical data of a normal state and a corresponding time period into the model. In order to reduce false alarm missing, as many time periods as possible reflecting the normal operation of the equipment need to be added into the model. In order to select a proper time period, in the past, abnormal time periods such as data jump or numerical value bad are shielded by searching through a method of manually analyzing the trend of a model measuring point, so that the workload is huge and the efficiency is low, and therefore, the development of a device fault early-warning model time period searching software is necessary.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a method suitable for searching a device fault early warning model time period, so as to avoid the defects of time waste, labor waste and error easiness in manually searching the model time period and greatly improve the working efficiency.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
the method suitable for searching the equipment fault early-warning model time period comprises the following steps of;
(1) The primary screening steps are as follows:
the user obtains a final required time period through twice screening, and firstly, the initial time period meeting the conditions is screened according to the initial screening conditions; specifically, taking an hour as a minimum statistical unit, automatically collecting historical interpolation data at equal time intervals within a range from a designated starting time to an ending time, judging whether the historical data meets the conditions by using initial screening conditions, if each interpolation data meets the conditions in the hour, the hour data is valid, if 1 interpolation data is invalid, the hour data is invalid, and finally recording an initial time period meeting the conditions to form a list, wherein the list can be manually adjusted (added, modified or deleted in time period); the initial time period may be exported as an excel file; alternatively, the step may be performed by introducing an initial period of time without performing the automatic acquisition process;
(2) The accurate screening steps are as follows:
on the basis of obtaining an initial time period, sequentially analyzing time interval interpolation data of all measuring points of the equipment model, obtaining a final time period of accurate screening according to a certain calculation rule, deriving the final time period into an excel file, recording unconditional time periods by using a log, specifically, sequentially carrying out time interval interpolation on the data and the like according to different measuring points of the equipment model in a preliminary screening time period, screening the time period according to conditions such as data quality and maximum judgment, removing the unconditional time period, and finally obtaining a common time period result of all the measuring points meeting the conditions.
The precise screening conditions are as follows: 1) The maximum condition, namely the upper and lower limit condition, takes an hour as the minimum statistical unit, the data interpolation data of the data within one hour is within the range defined by the upper and lower limits, the range cannot be exceeded, the condition is met if the range is not exceeded, and the condition is not met if the range is exceeded; 2) The quality condition, namely, the data quality condition of each measuring point is obtained through interpolation at equal time intervals, if the data quality is good, the condition is met, otherwise, the condition is not met; 3) If the data obtained by interpolation of the data at equal time intervals within one hour is unchanged, eliminating the time period of the hour and the adjacent hours before and after the hour; 4) If the number of samples is less than 3, taking the hour as the minimum statistical unit, if the number of sample data stored in the hour is less than 3, the hour is eliminated.
The model measuring point (a): modeling an equipment model, wherein the model comprises measuring points capable of reflecting the running state of the equipment;
(b) Primary screening conditions: the logic expression is used to express that the requirement condition is met, for example, when the equipment is in normal operation, the current of the equipment is larger than a certain value, and if the current is lower than the certain value, the equipment is stopped.
(c) Time interval: is interpolation time interval data, and data is interpolated from historical data once every interval, and time units are minutes.
(d) Sample value: the data actually stored by the database is not necessarily data of equal time intervals.
The software can take historical sample data, interpolation data and statistical data of different databases in different time periods.
The software can perform preliminary screening according to manually selected starting time, ending time, time interval and screening conditions with a certain format, and statistics is performed by an interpolation method.
The software can accurately screen the historical time periods meeting the conditions according to a certain calculation rule, and the screening is carried out by interpolation or statistic value methods in the screening, wherein the adopted screening conditions are whether the number of samples is less than 3, the dead pixel time periods are removed, the maximum value is judged, the statistic value is or interpolation is selected.
The software can export time periods and historical interpolation data to query in an excel file format.
The invention has the beneficial effects that:
the invention judges the needed historical time period by software, thereby avoiding time consuming, labor consuming and error prone manual implementation.
The method is used for judging the time period of the equipment fault early-warning model, the reliability and the accuracy work efficiency are enhanced, and the use effect of the program is better by matching with equipment fault early-warning system software.
Drawings
Fig. 1 is a software program flow.
FIG. 2 is a working condition collection main interface.
Fig. 3 is a schematic diagram of an automatic acquisition window, which can be filled with start time, end time, time interval, and condition information.
FIG. 4 is a schematic diagram of a final screening that may be performed after a verification site is imported into the verification window.
Fig. 5 is a schematic diagram of a result/log/export window, in which screening results can be checked, reject logs are checked, log records are recorded under the condition that the conditions are not met, and the screened time period results are exported to an excel file.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings.
As shown in fig. 1: the method for searching the time period of the equipment fault early warning model comprises the steps of taking historical sample data, time interval interpolation data and statistical data of different time periods of different databases, taking the data, judging the number of samples of the data in given conditions by taking hours as a unit, and rejecting the time period when the number of the samples is too small.
1. The folder containing the software is copied to the proper position of the hard disk without installing the software.
2. The following are determined in the notepad open devicewarncondition. Exe. Config file:
description: 1) Verifying that the maximum number is 50 measuring points; 2) Verifying the upper limit of the parameter, wherein the default value is 0.8, and the upper limit = hour average value + default value ×hour average value; 3) Verifying the upper limit of the parameter, wherein the default value is 0.2, and the lower limit = hour average value-default value is an hour average value; 4) Whether the data quality is judged according to the statistical value is executed according to different databases, and the default value is False (judging the data quality by adopting an interpolation mode); 5) And connecting the real-time databases, and filling in the connection character strings according to the connection contents of different databases.
3. Double-click DeviceWarningCondition. Exe opening program is shown in FIG. 2, click automatic acquisition is shown in FIG. 3, and the selection of start time, end time, time interval (which is interpolation time interval data, and how often data is interpolated from historical data, time units are minutes), and initial conditions (supporting logic expressions). In the method, the historical data is extracted at equal time intervals within a range from a specified starting time to an ending time by taking an hour as a minimum statistical unit, whether the historical data meets the given condition is judged, if each interpolation data meets the condition in the hour, the hour data is valid, and if 1 interpolation data is invalid, the hour data is invalid. Finally, the time period meeting the conditions is recorded, and a list is formed in fig. 2. The addition or deletion of time periods may be manually screened in fig. 2. The time period may be exported as an excel file. Alternatively, this step may be accomplished by importing a time period without performing an automatic acquisition process. Clicking the next step after the step is completed to enter the next link.
4. In fig. 4, after adding the measuring points in a "lead-in" manner or adding the measuring points in a "+" single point, selecting or not selecting a period of time for eliminating dead points, clicking "check", and entering accurate screening. The method comprises the steps of sequentially carrying out data interpolation according to different measuring points of a model in a preliminary screening time period, carrying out time period final screening according to conditions such as data quality, maximum value judgment and the like, removing the time period which does not meet the conditions, and finally obtaining a common time period result of all the measuring points meeting the conditions. The conditions in the procedure were: 1) The maximum condition, namely the upper and lower limit conditions defined in the step 2; 2) The quality condition, namely the data quality condition of each measuring point is obtained through interpolation, if the data quality is good, the condition is met, otherwise, the condition is not met; 3) The data not refreshing condition, namely if the data obtained by interpolation of the data in one hour is unchanged, eliminating the time period of the hour and the adjacent hours before and after the hour; 4) If the number of samples is less than 3, taking the hour as the minimum statistical unit, if the number of sample data stored in the hour is less than 3, the hour is eliminated. .
5. In fig. 5, the screening results can be viewed; checking a rejection log, wherein the log records the condition that the condition is not satisfied; and exporting the filtered time period result to an excel file.
Claims (1)
1. The method for searching the equipment fault early-warning model time period is characterized by comprising the following steps of;
(1) The primary screening steps are as follows:
screening an initial time period meeting the conditions according to the initial screening conditions, taking an hour as a minimum statistical unit, automatically collecting historical interpolation data at equal time intervals in a range from appointed starting time to ending time, judging whether the historical data meet the conditions or not by using the initial screening conditions, if each interpolation data meet the conditions in the hour, enabling the hour data to be effective, and if 1 interpolation data are invalid, enabling the hour data to be invalid, and finally recording the initial time period meeting the conditions to form a list;
(2) The accurate screening steps are as follows:
sequentially analyzing the interpolation data of all measuring points of the equipment model at equal time intervals on the basis of obtaining an initial time period, obtaining a final time period accurately screened according to a certain calculation rule, exporting the final time period into an excel file, and recording the time period which does not meet the conditions by using a log;
the precise screening conditions are as follows: 1) The maximum condition, namely the upper and lower limit condition, takes an hour as the minimum statistical unit, the data interpolation data of the data within one hour is within the range defined by the upper and lower limits, the range cannot be exceeded, the condition is met if the range is not exceeded, and the condition is not met if the range is exceeded; 2) The quality condition, namely, the data quality condition of each measuring point is obtained through interpolation at equal time intervals, if the data quality is good, the condition is met, otherwise, the condition is not met; 3) If the data obtained by interpolation of the data at equal time intervals within one hour is unchanged, eliminating the time period of the hour and the adjacent hours before and after the hour; 4) If the number of samples is less than 3, taking the hour as the minimum statistical unit, and if the number of sample data stored in the hour is less than 3, rejecting the hour;
the software can accurately screen the historical time periods meeting the conditions according to a certain calculation rule, and the screening is carried out by an interpolation or statistical value method in the screening, wherein the adopted screening conditions are whether the number of samples is less than 3, the dead pixel time periods are removed, the maximum value is judged, the statistical value is selected or the interpolation is selected;
(a) Model measuring point: modeling an equipment model, wherein the model comprises measuring points capable of reflecting the running state of the equipment;
(b) Primary screening conditions: the logic expression is adopted to express that the requirement condition is met, the current of the equipment is larger than a certain value when the equipment normally operates, and if the current is lower than the certain value, the equipment stops operating;
(c) Time interval: interpolation time interval data is obtained, the data is interpolated and extracted from the historical data once every interval, and the time unit is minutes;
(d) Sample value: the data actually stored in the database is not necessarily the data of the same time interval;
the software can take historical sample data, interpolation data and statistical data of different databases in different time periods;
the software can perform preliminary screening according to manually selected starting time, ending time, time interval and screening conditions with a certain format, and statistics is performed by an interpolation method;
the software can export the time period and the historical interpolation data for inquiring in an excel file format;
and in the preliminary screening time period, sequentially carrying out data time interval interpolation according to different measuring points of the equipment model, carrying out time period screening according to the data quality and the maximum judgment conditions, removing the time periods which do not meet the conditions, and finally obtaining the common time period result of all the measuring points which meet the conditions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910276801.3A CN110008245B (en) | 2019-04-08 | 2019-04-08 | Method suitable for searching equipment fault early warning model time period |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910276801.3A CN110008245B (en) | 2019-04-08 | 2019-04-08 | Method suitable for searching equipment fault early warning model time period |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110008245A CN110008245A (en) | 2019-07-12 |
| CN110008245B true CN110008245B (en) | 2024-01-23 |
Family
ID=67170224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910276801.3A Active CN110008245B (en) | 2019-04-08 | 2019-04-08 | Method suitable for searching equipment fault early warning model time period |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110008245B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113568961B (en) * | 2021-09-26 | 2021-12-14 | 西安热工研究院有限公司 | Curve generation method and equipment for selecting measuring points through time marks and integrating alarm information |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101968758A (en) * | 2010-10-28 | 2011-02-09 | 西北工业大学 | Multiple data point threshold detection method |
| CN102270271A (en) * | 2011-05-03 | 2011-12-07 | 北京中瑞泰科技有限公司 | Equipment failure early warning and optimizing method and system based on similarity curve |
| CN106020924A (en) * | 2016-06-07 | 2016-10-12 | 中国电建集团贵阳勘测设计研究院有限公司 | Method for extracting temperature time process line in concrete temperature control optical fiber temperature measurement host |
| CN106504366A (en) * | 2016-10-31 | 2017-03-15 | 四川岭傲信息技术有限公司 | A kind of railway locomotive wind regime data analysing method and intelligent management system |
| CN109472846A (en) * | 2018-12-27 | 2019-03-15 | 燕山大学 | The method for obtaining Bode diagram with MATLAB processing frequency sweep data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10729382B2 (en) * | 2016-12-19 | 2020-08-04 | Mitsubishi Electric Research Laboratories, Inc. | Methods and systems to predict a state of the machine using time series data of the machine |
-
2019
- 2019-04-08 CN CN201910276801.3A patent/CN110008245B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101968758A (en) * | 2010-10-28 | 2011-02-09 | 西北工业大学 | Multiple data point threshold detection method |
| CN102270271A (en) * | 2011-05-03 | 2011-12-07 | 北京中瑞泰科技有限公司 | Equipment failure early warning and optimizing method and system based on similarity curve |
| CN106020924A (en) * | 2016-06-07 | 2016-10-12 | 中国电建集团贵阳勘测设计研究院有限公司 | Method for extracting temperature time process line in concrete temperature control optical fiber temperature measurement host |
| CN106504366A (en) * | 2016-10-31 | 2017-03-15 | 四川岭傲信息技术有限公司 | A kind of railway locomotive wind regime data analysing method and intelligent management system |
| CN109472846A (en) * | 2018-12-27 | 2019-03-15 | 燕山大学 | The method for obtaining Bode diagram with MATLAB processing frequency sweep data |
Non-Patent Citations (4)
| Title |
|---|
| Research on abnormal search method of monitoring parameters for power plant equipment based on cluster analysis;G.K. Zhang etc.;《2012 IEEE 11th International Conference on Signal Processing》;第2331-2336页 * |
| 基于多源传感数据相关性分析的电厂设备故障检测方法;柴政 等;《计算机与数字工程》;第382-388页 * |
| 基于设备故障相关参数筛选模型建立风电机组故障预警体系;陈晓清 等;《风能》;第92-97页 * |
| 涂扬举 等.《智慧企业 框架与实践 第2版》.北京：经济日报出版社,2018,第311-316段. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110008245A (en) | 2019-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101989087B (en) | On-line real-time failure monitoring and diagnosing system device for industrial processing of residual oil | |
| CN106383916B (en) | Data processing method based on predictive maintenance of industrial equipment | |
| CN112181960B (en) | Intelligent operation and maintenance framework system based on AIOps | |
| AU2022204116A1 (en) | Verification method for electrical grid measurement data | |
| CN111459698A (en) | Database cluster fault self-healing method and device | |
| CN111027615A (en) | Middleware fault early warning method and system based on machine learning | |
| CN117172556B (en) | Construction risk early warning method and system for bridge engineering | |
| CN112148602B (en) | Source code security analysis method based on history optimization feature intelligent learning | |
| CN116992346A (en) | Enterprise production data processing system based on artificial intelligence big data analysis | |
| CN110008245B (en) | Method suitable for searching equipment fault early warning model time period | |
| CN117291575A (en) | Equipment maintenance method, equipment maintenance device, computer equipment and storage medium | |
| CN113051161A (en) | API misuse detection method based on historical code change information | |
| CN116737549A (en) | Time sequence database stability test method | |
| CN111190944A (en) | Data mining method and system | |
| CN115794624A (en) | Closed-loop testing method and system for single automation device of intelligent substation | |
| CN115357637A (en) | Method for extracting sample data of specific working condition of industrial equipment | |
| CN114528314A (en) | Engineering construction project supervision system and method | |
| CN110298585B (en) | Hierarchical automatic auditing method for monitoring information of substation equipment | |
| CN113887862A (en) | Energy metering service data analysis method and system | |
| CN113836203A (en) | Network data diagnosis detection analysis system | |
| CN117076454B (en) | Engineering quality acceptance form data structured storage method and system | |
| CN117971605B (en) | Automatic log information collection method and system based on database abnormality | |
| RU2809254C1 (en) | Method and system for monitoring automated systems | |
| RU2809254C9 (en) | Method and system for monitoring automated systems | |
| CN118094234B (en) | Automatic data labeling method and device based on multi-source power data |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |