CN113806969B - Compressor unit health prediction method based on time domain data correlation modeling - Google Patents
Compressor unit health prediction method based on time domain data correlation modeling Download PDFInfo
- Publication number
- CN113806969B CN113806969B CN202111244835.8A CN202111244835A CN113806969B CN 113806969 B CN113806969 B CN 113806969B CN 202111244835 A CN202111244835 A CN 202111244835A CN 113806969 B CN113806969 B CN 113806969B
- Authority
- CN
- China
- Prior art keywords
- data
- fault
- compressor unit
- state
- model
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Abstract
The invention discloses a compressor set health prediction method based on time domain data correlation modeling, which relates to the technical field of pipe network operation assembly detection and aims to solve the problem that early warning is difficult to provide at the early stage of equipment failure in a mode of in-process and after-process analysis, and the technical scheme is characterized by comprising the following steps of: establishing a model, namely establishing a state data perception model and establishing a fault mechanism association model; collecting normal parameters, obtaining a normal working condition state health degree reference value through a state data perception model, and recording the change rule of the fault under different fault conditions through a fault mechanism association model; performing comparative analysis of data extraction, and performing multiple analysis on a state data perception model and a fault mechanism association model to obtain an average value; and (6) alarming data abnormity. The effects of prediction diagnosis, early warning and timely reminding of the faults of the compressor unit equipment are achieved.
Description
Technical Field
The invention relates to the technical field of pipe network operation assembly detection, in particular to a compressor set health prediction method based on time domain data correlation modeling.
Background
The safe and normal operation of the compressor unit equipment relates to normal and stable operation of pipe network conveying, in the early stage of the compressor unit equipment failure, as the failure characteristics are relatively unobvious, the influence on the normal operation of the pipe network is small, the change of the equipment in the early stage of the equipment failure is difficult to monitor by the existing technical means, the change is often ignored easily, when the failure develops to a more serious stage, the damage of the equipment is often already generated when the alarm is generated and monitored, except that a large amount of manpower and material resources are consumed for maintaining the equipment, the equipment can possibly stop working at any time to further influence the stable operation of the pipe network, therefore, before the equipment failure occurs, the potential failure of the equipment is found and early warning is performed in advance, so that timely and proper manual operation intervention is performed in advance, the expansion of the failure and the occurrence of accidents are prevented, the safe production of the equipment and the pipe network is ensured, and the method has great practical significance.
At present, the research on the equipment fault diagnosis technology of the domestic compressor group mainly adopts a method for researching a fault mechanism, when the fault characteristics are obvious, the fault is often developed to a certain stage, and a data model is introduced into equipment state monitoring at home and abroad to quantitatively perceive the health state of equipment, identify early symptoms of the fault and provide early warning, wherein the mode driven by the data model obtains a lot of results in the fields of equipment health management and early warning.
The above prior art solutions have the following drawbacks: the mode of trouble thing and after analysis, hardly realize carrying out whole and quantitative health assessment to compressor unit equipment running state, also hardly provide early warning at the equipment initial stage that breaks down, the mode of data model drive also has the early warning not qualitative well, and the early warning wrong report is many, the shortcoming that the practicality is not strong.
Disclosure of Invention
The invention aims to provide a compressor unit health prediction method based on time domain data correlation modeling, which greatly improves the technical level of fault diagnosis of the existing compressor unit and early gives out early warning of faults.
In order to realize the purpose, the invention provides the following technical scheme:
a compressor set health prediction method based on time domain data correlation modeling comprises the following steps:
s1: establishing a model, namely establishing a 'state data perception model' for judging the whole working state through normal working condition data in massive industrial time domain operation data of a compressor set, establishing a 'failure mechanism association model' for comparing and analyzing the change of the failure state of a compressor unit through a typical failure rule of the compressor set, and analyzing and observing the whole normal operation state;
s2: normal parameter collection, namely, calculating and analyzing normal state modeling data of compressor unit equipment through a state data sensing model to obtain a normal working condition state health degree reference value, and a fault mechanism association model obtains a change mechanism of a typical fault through a parameter change rule of typical fault characteristics and records the change rule of the fault under different fault conditions, so that error types can be conveniently and integrally analyzed;
s3: the method comprises the steps of performing contrastive analysis of data extraction, establishing compressor unit equipment by using a compressor unit safety state analysis method of a state data perception model and a fault mechanism correlation model, comparing integrally obtained data, performing calculation analysis on compressor unit real-time sampling data to realize real-time state health quantitative evaluation of the compressor unit equipment, performing prediction diagnosis and early warning prompt on potential faults of the compressor unit by combining a fault mechanism correlation model, performing correlation sequencing on measuring points causing fault warning, and performing multiple analysis and averaging in the process of integral contrastive analysis to ensure the contrastive accuracy of integral data;
s4: and (3) data abnormity warning, namely comparing the safety state analysis condition of the compressor set of the state data perception model and the fault mechanism association model with the normal operation condition, if abnormity occurs, performing online real-time calculation on the data of the fault parameters of the compressor set equipment at the moment by using the fault mechanism association model, and performing specific typical fault prediction diagnosis and early warning prompt on the compressor set equipment through concept calculation of the fault model, so that early warning on the fault risk of the compressor set integrally is facilitated, and the compressor set is overhauled before the integral state is serious.
By adopting the technical scheme, the normal operation state and abnormal operation state of the compressor are recorded by using the state data perception model and the fault mechanism association model, and the recorded data is compared with the actual data, so that the operation state of the compressor can be analyzed and compared integrally.
Further, the normal working condition data in the massive industrial time domain operation data of the compressor unit equipment in S1 is data in which real-time values of all parameters of the compressor unit equipment are within a normal range on the same time axis for a period of time.
By adopting the technical scheme, all parameters on the same time axis in a period of time can be used as normal comparison data, different data on the same time axis also represent different equipment running conditions, and the condition of equipment failure can be judged according to different data conditions during comparison.
Further, the normal working condition data in the massive industrial time domain operation data of the compressor unit equipment in S1 is screened to remove abnormal and interference data, a state group of the compressor unit equipment under the normal working condition is generated, and a "state data sensing model" reflecting the actual operation rule of the compressor unit equipment is automatically created according to the feature distribution of all parameters in the state group.
Through adopting above-mentioned technical scheme, the whole body detects the contrast result and can produce the influence by improper and interference data, so need calculate improper and interference data and substitute whole detection numerical value and get rid of, guarantee the accuracy of whole detection.
Further, the "fault mechanism association model" in S2 is a unit fault mechanism association model that is created by selecting characteristic parameter change rules of different typical faults and building a fault rule combination by using typical fault rules of compressor unit equipment, and combining characteristic parameter change rules of multiple typical faults of a compressor unit.
By adopting the technical scheme, the fault mechanism correlation model is used for recording fault combination and independent fault condition data under different conditions, and the condition of the running compressor is compared with the fault request, so that the overall fault type can be conveniently and integrally and rapidly analyzed, and the overall maintenance is convenient.
Further, the modeling data of the fault mechanism association model and the state data perception model in the step S2 are historical normal working condition time domain data of the compressor unit equipment and historical time domain data of each typical fault parameter.
By adopting the technical scheme, the fault mechanism correlation model and the state data perception model need to record data under various conditions, and the overall record is historical normal working condition time domain data and historical time domain data of each typical fault parameter, so that common faults can be conveniently and integrally identified.
Furthermore, the 'state data perception model' in the S2 is dimensionless data with a health degree reference value of 0 to 1.
By adopting the technical scheme, the state data perception model is provided with the reference value, so that the whole body can be conveniently judged on the normal reference value, the whole body can be ensured to be well judged, and the whole body can be conveniently judged on the normal operation.
Further, the contrastive analysis in S3 is divided into two groups, one contrastive analysis in S3 adopts the multi-period data change condition under the same state abnormal condition, and the other contrastive analysis in S3 adopts the data change condition of different parts in the same period.
Through adopting above-mentioned technical scheme, contrast the detection data that wholly produce to guarantee wholly can detect the compressor running state, the detection data that wholly produces simultaneously carries out many times contrast, prevents that external influence from making data produce the change, guarantees the accuracy that wholly detects.
Further, in the S4, the data abnormal alarm adopts a corresponding speaker alarm device and a warning light alarm device.
Through adopting above-mentioned technical scheme, speaker alarm equipment and warning light alarm equipment can remind the staff in vision and sense of hearing, guarantee whole good warning effect, make things convenient for the staff to overhaul in time.
In summary, the beneficial technical effects of the invention are as follows:
1. the method comprises the steps of establishing a state data perception model and a fault mechanism association model by adopting a normal data and fault parameter data change rule of compressor unit equipment, then carrying out calculation analysis on real-time data through the state data perception model to obtain the similarity degree of the current equipment running state and the historical normal working condition, and finally carrying out calculation analysis on the association parameter data of state transaction by using the fault mechanism association model to realize the effects of prediction diagnosis and early warning of the compressor unit equipment fault;
2. the scheme of more reasonably analyzing and prompting the early fault of the compressor unit equipment by adopting a state data perception model and a fault mechanism correlation model solves the problem that the mode always depends on rules, expert experience judgment and after-treatment, and realizes the effects of quick detection and repeated detection;
3. compared analysis and data abnormity alarm of data extraction are adopted, so that monitoring and analysis of typical faults of compressor unit equipment on site are simpler and more reliable, early warning prompt is carried out on the typical faults of the compressor unit equipment, the faults can be timely solved in a sprouting state, normal and stable operation of the compressor unit equipment and a natural gas conveying pipe network is facilitated, and a timely reminding effect is generated.
Drawings
FIG. 1 is a schematic diagram of the working procedure of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
Referring to fig. 1, a method for predicting health of a compressor unit based on time domain data association modeling includes the following steps:
s1: establishing a model, establishing a 'state data perception model' for judging the whole working state through normal working condition data in mass industrial time domain running data of a compressor set, establishing a 'failure mechanism correlation model' for comparing and analyzing the change of the failure state of the compressor set through a typical failure rule of the compressor set, analyzing and observing the whole common running state, wherein the normal working condition data in the mass industrial time domain running data of the compressor set is data in which the real-time values of all parameters of the compressor set are in a normal range on the same time axis in a period of time, and the normal working condition data in the mass industrial time domain running data of the compressor set are screened to remove abnormal and interference data to generate a state set under the normal working condition of the compressor set, and automatically establishing a 'state data perception model' reflecting the actual running rule of the compressor set according to the characteristic distribution of all parameters in the state set;
s2: collecting normal parameters, calculating and analyzing normal state modeling data of compressor unit equipment through a state data perception model to obtain a normal working condition state health degree reference value, obtaining a change mechanism of a typical fault through a parameter change rule of typical fault characteristics by a fault mechanism association model, recording the change rule of the fault under different fault conditions, and conveniently analyzing the error type integrally, wherein the fault mechanism association model is a unit fault mechanism association model which is established by selecting the change rule of the characteristic parameters of different typical faults to establish a fault rule combination by utilizing a typical fault rule of the compressor unit equipment, integrating the change rules of the characteristic parameters of various typical faults of the compressor unit to jointly establish a unit fault mechanism association model, modeling data of the fault mechanism association model and the state data perception model are non-dimensionalization data of which the health degree reference value is 0 to 1 and are historical normal working condition time domain data of the compressor unit equipment and historical time domain data of various typical fault parameters;
s3: the method comprises the steps of performing contrastive analysis of data extraction, establishing compressor unit equipment by using a compressor unit safety state analysis method of a state data perception model and a fault mechanism correlation model, comparing integrally obtained data, performing calculation analysis on real-time sampled data of a compressor unit to realize real-time state health quantitative evaluation of the compressor unit equipment, performing prediction diagnosis and early warning prompt on potential faults of the compressor unit by combining the fault mechanism correlation model, performing correlation sequencing on measurement points causing fault warning, performing multiple analysis and averaging in the process of integral contrastive analysis to ensure the contrast accuracy of integral data, dividing the contrastive analysis into two groups, performing contrastive analysis on multi-period data change conditions under the condition that the same state is abnormal, and performing contrastive analysis on the relative data change conditions of different parts in the same period;
s4: and (3) data abnormity warning, namely comparing the safety state analysis condition of the compressor set of the state data perception model and the fault mechanism association model with the normal operation condition, if abnormity occurs, performing online real-time calculation on the data of fault parameters of the compressor set equipment at the moment by using the fault mechanism association model, performing specific typical fault prediction diagnosis and warning prompt on the compressor set equipment through concept calculation of the fault model, and being beneficial to performing warning on the fault risk of the compressor as a whole and performing maintenance before the integral situation is serious, wherein the data abnormity warning adopts corresponding loudspeaker warning equipment and warning lamp warning equipment.
The working principle is as follows: establishing a state data perception model by using time domain data of historical normal working conditions of compressor unit equipment, establishing a fault mechanism correlation model by using a typical fault mode of the compressor unit equipment, wherein the state data perception model covers a section of sampling values capable of reflecting the running time of the compressor unit equipment under each working condition, each group of data of the state data perception model can express a normal running state of the compressor unit equipment, and the state data perception model satisfies the simultaneity of each variable parameter in each group of sampling values, namely the sampling values of each parameter at the same historical moment are required, performing calculation analysis on historical data for modeling by using the established state data perception model to obtain a health degree reference value of the normal working condition, performing calculation analysis on typical fault rules for modeling by using the established fault mechanism correlation model to obtain a fault characteristic parameter variation trend, after the state data perception model and the fault mechanism correlation model are established, performing back calculation on the modeled data by using the state data perception model to obtain a state data correlation model judgment reference value, performing calculation analysis on the parameter variation law of typical fault characteristics by using the time domain data perception model to obtain a variation mechanism of the typical fault mechanism, and obtaining a state data of the compression mechanism correlation model, and once the state of the compressor unit before performing the fault mechanism calculation analysis on the historical fault mechanism is lower than the current state of the compressor unit, and the fault mechanism, performing the fault mechanism calculation model, and obtaining a fault mechanism analysis on the fault mechanism of the historical state of the compressor unit. And then, calculating and analyzing the characteristic data of the compressor unit equipment typical faults mainly represented by the abnormal operation time period by using a fault mechanism correlation model, and carrying out health prediction and early warning prompt on the compressor unit equipment so as to carry out health prediction diagnosis and early warning on the compressor unit equipment.
The embodiments of the present invention are preferred embodiments of the present invention, and the scope of the present invention is not limited by these embodiments, so: all equivalent changes made according to the structure, shape and principle of the invention are covered by the protection scope of the invention.
Claims (8)
1. A compressor set health prediction method based on time domain data correlation modeling is characterized by comprising the following steps: the method comprises the following steps:
s1: establishing a model, namely establishing a state data perception model for judging the whole working state through normal working condition data in massive industrial time domain operation data of a compressor set, establishing a fault mechanism association model for comparing and analyzing the change of the fault state condition of a compressor unit through a typical fault rule of the compressor set, and analyzing and observing the whole normal operation state;
s2: normal parameter collection, namely, calculating and analyzing normal state modeling data of compressor unit equipment through a state data perception model to obtain a normal working condition state health degree reference value, and a fault mechanism correlation model obtains a change mechanism of a typical fault through a parameter change rule of typical fault characteristics and records the change rule of the fault under different fault conditions, so that the error type can be conveniently analyzed integrally;
s3: the method comprises the steps of performing contrastive analysis of data extraction, establishing compressor unit equipment by using a compressor unit safety state analysis method of a state data perception model and a fault mechanism correlation model, comparing integrally obtained data, performing calculation analysis on real-time sampled data of a compressor unit to realize real-time state health quantitative evaluation of the compressor unit equipment, performing prediction diagnosis and early warning prompt on potential faults of the compressor unit by combining the fault mechanism correlation model, performing correlation sequencing on measuring points causing fault warning, and performing multiple analysis and averaging in the process of integral contrastive analysis to ensure the accuracy of integral data comparison;
s4: and (3) data abnormity warning, namely comparing the safety state analysis condition of the compressor set of the state data perception model and the fault mechanism correlation model with the normal operation condition, if abnormity occurs, performing online real-time calculation on the data of fault parameters of the compressor set equipment at the moment by using the fault mechanism correlation model, and performing specific typical fault prediction diagnosis and early warning prompt on the compressor set equipment through concept calculation of the fault model, so that early warning on the fault risk of the compressor set is facilitated integrally and overhaul is performed before the integral state is serious.
2. The method for predicting the health of the compressor unit based on the time-domain data correlation modeling as claimed in claim 1, wherein: and the normal working condition data in the massive industrial time domain operation data of the compressor unit equipment in the S1 are data in which the real-time values of all parameters of the compressor unit equipment are in a normal range on the same time axis in a period of time.
3. The method for predicting the health of a compressor unit based on time domain data correlation modeling according to claim 2, wherein: and (2) screening normal working condition data in the massive industrial time domain operation data of the compressor unit equipment in the S1 to remove abnormal and interference data, generating a state group of the compressor unit equipment under the normal working condition, and automatically creating a 'state data perception model' reflecting the actual operation rule of the compressor unit equipment according to the characteristic distribution of all parameters in the state group.
4. The method for predicting the health of a compressor unit based on time domain data correlation modeling according to claim 1, wherein: the 'fault mechanism association model' in the S2 is a unit fault mechanism association model which is established by utilizing typical fault rules of compressor unit equipment, selecting characteristic parameter change rules of different typical faults to establish a fault rule combination, and integrating the characteristic parameter change rules of various typical faults of the compressor unit.
5. The method for predicting the health of a compressor unit based on time domain data correlation modeling according to claim 1, wherein: and modeling data of the fault mechanism association model and the state data perception model in the S2 are historical normal working condition time domain data of the compressor unit equipment and historical time domain data of each typical fault parameter.
6. The method for predicting the health of a compressor unit based on time domain data correlation modeling according to claim 5, wherein: and in the S2, the health degree reference value of the state data perception model is dimensionless data of 0 to 1.
7. The method for predicting the health of a compressor unit based on time domain data correlation modeling according to claim 1, wherein: the contrastive analysis in the S3 is divided into two groups, one contrastive analysis in the S3 adopts the multi-time-period data change condition under the condition of the same state abnormity, and the other contrastive analysis in the S3 adopts the data change condition of different parts in the same time period.
8. The method for predicting the health of a compressor unit based on time domain data correlation modeling according to claim 1, wherein: and in the S4, corresponding speaker alarm equipment and warning light alarm equipment are adopted for data abnormity alarm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111244835.8A CN113806969B (en) | 2021-10-26 | 2021-10-26 | Compressor unit health prediction method based on time domain data correlation modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111244835.8A CN113806969B (en) | 2021-10-26 | 2021-10-26 | Compressor unit health prediction method based on time domain data correlation modeling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113806969A CN113806969A (en) | 2021-12-17 |
| CN113806969B true CN113806969B (en) | 2022-12-27 |
Family
ID=78898288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111244835.8A Active CN113806969B (en) | 2021-10-26 | 2021-10-26 | Compressor unit health prediction method based on time domain data correlation modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113806969B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116756666B (en) * | 2023-08-15 | 2023-11-24 | 广东石油化工学院 | Multi-source data fusion type health state assessment method for centrifugal compressor |
| CN117319179A (en) * | 2023-11-28 | 2023-12-29 | 中南大学 | Equipment abnormality monitoring method and system based on mechanism model and industrial Internet of things |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102270271B (en) * | 2011-05-03 | 2014-03-19 | 北京中瑞泰科技有限公司 | Equipment failure early warning and optimizing method and system based on similarity curve |
| CN106406295B (en) * | 2016-12-02 | 2019-02-26 | 南京康尼机电股份有限公司 | Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state |
| CN109240244B (en) * | 2018-10-26 | 2020-11-20 | 云达世纪（北京）科技有限公司 | Data-driven equipment running state health degree analysis method and system |
-
2021
- 2021-10-26 CN CN202111244835.8A patent/CN113806969B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113806969A (en) | 2021-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113806969B (en) | Compressor unit health prediction method based on time domain data correlation modeling | |
| CN110647133B (en) | Rail transit equipment state detection maintenance method and system | |
| CN104573850A (en) | Method for evaluating state of thermal power plant equipment | |
| CN103198147B (en) | The differentiation of automatic monitoring abnormal data and processing method | |
| CN112561736A (en) | Fault diagnosis system and method for relay protection device of intelligent substation | |
| CN112925279A (en) | Fault comprehensive analysis system based on MES system | |
| CN104425046B (en) | Method and system for quantizing operation security indexes of nuclear power station | |
| CN113562562A (en) | Elevator safety early warning monitoring system and judgment method thereof | |
| CN115566804B (en) | Electric power monitoring system based on distributed optical fiber sensing technology | |
| CN113177646B (en) | Power distribution equipment online monitoring method and system based on self-adaptive edge proxy | |
| CN112903815A (en) | Monitoring method and monitoring system for bridge expansion joint state | |
| CN113902241A (en) | Power grid equipment maintenance strategy system and method based on comprehensive state evaluation | |
| CN114781476A (en) | Fault analysis system and method for measuring equipment | |
| CN115248402A (en) | Oil immersed transformer on-line monitoring system and method | |
| CN116383633A (en) | Method and system for detecting faults of machine-made sand vibrating screen through multi-factor comprehensive analysis | |
| CN113982850B (en) | Fan comprehensive health analysis method and system integrating high-low frequency signals | |
| CN110866695A (en) | Auxiliary maintenance decision method based on association of equipment running state and maintenance work | |
| CN107506832B (en) | Hidden danger mining method for assisting monitoring tour | |
| CN113418731A (en) | Online fault diagnosis method for cigarette making machine set | |
| CN112580858A (en) | Equipment parameter prediction analysis method and system | |
| CN116611953A (en) | Electric energy meter production and manufacturing real-time data management method and system based on Internet of things | |
| CN115630284A (en) | Air compressor fault analysis and service life prediction system considering service period | |
| CN113487189A (en) | Petrochemical equipment fault probability risk assessment system and assessment method | |
| US11047833B2 (en) | Method for automatic determination of trend in graphic analysis of turbomachines | |
| CN117589444B (en) | Wind driven generator gear box fault diagnosis method based on federal learning |
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 |