CN117034109A - Engine oil abrasive grain analysis method and system based on segmentation threshold and computer readable storage medium - Google Patents
Engine oil abrasive grain analysis method and system based on segmentation threshold and computer readable storage medium Download PDFInfo
- Publication number
- CN117034109A CN117034109A CN202310968076.2A CN202310968076A CN117034109A CN 117034109 A CN117034109 A CN 117034109A CN 202310968076 A CN202310968076 A CN 202310968076A CN 117034109 A CN117034109 A CN 117034109A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- oil
- lubricating oil
- data
- engine
- 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.)
- Pending
Links
- 239000006061 abrasive grain Substances 0.000 title claims abstract description 32
- 239000010705 motor oil Substances 0.000 title claims abstract description 28
- 238000004458 analytical method Methods 0.000 title claims abstract description 27
- 230000011218 segmentation Effects 0.000 title claims abstract description 17
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 238000012544 monitoring process Methods 0.000 claims abstract description 176
- 239000010687 lubricating oil Substances 0.000 claims abstract description 128
- 239000002245 particle Substances 0.000 claims abstract description 112
- 239000003921 oil Substances 0.000 claims abstract description 90
- 238000009826 distribution Methods 0.000 claims abstract description 58
- 238000000034 method Methods 0.000 claims abstract description 42
- 238000007781 pre-processing Methods 0.000 claims abstract description 14
- 230000002159 abnormal effect Effects 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 7
- 230000004069 differentiation Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 claims 1
- 238000007405 data analysis Methods 0.000 abstract description 2
- 238000012423 maintenance Methods 0.000 abstract description 2
- 238000012545 processing Methods 0.000 description 10
- 238000005299 abrasion Methods 0.000 description 8
- 230000036541 health Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 239000007788 liquid Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000001050 lubricating effect Effects 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
- G06F18/256—Fusion techniques of classification results, e.g. of results related to same input data of results relating to different input data, e.g. multimodal recognition
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/182—Level alarms, e.g. alarms responsive to variables exceeding a threshold
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Lubrication Details And Ventilation Of Internal Combustion Engines (AREA)
Abstract
The invention belongs to the technical field of intelligent maintenance and big data analysis, and provides an engine oil abrasive grain analysis method, system and computer readable storage medium based on a segmentation threshold; the method comprises the following steps: step S1: preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data; step S2: dividing the preprocessed lubricating oil data into N different monitoring interval sections according to different lubricating oil time or engine time respectively; step S3: selecting an oil monitoring index, fitting numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type of the oil monitoring index; step S4: based on the confidence, setting thresholds of the lubricating oil monitoring indexes according to different monitoring interval sections; step S5: and carrying out differential early warning on the overrun of abrasive particles in different monitoring intervals according to the lubricating oil monitoring index and the threshold value. The invention fully considers the difference of the abrasive particle concentration of the engine in different states, and the monitoring and early warning method is more rigorous and scientific.
Description
Technical Field
The invention belongs to the technical field of intelligent maintenance and big data analysis, and particularly relates to an engine oil abrasive grain analysis method, system and computer readable storage medium based on a segmentation threshold value.
Background
The lubricating oil is taken as the 'blood' of the machine and contains a great amount of information such as the lubrication state, the abrasion state and the like of the equipment, so that the oil monitoring method has become an important means for monitoring the health state of important equipment. The abrasion direct product abrasive particles carried in the lubricating oil become important research objects, and the monitoring analysis of the abrasive particles in the lubricating oil has important significance for daily monitoring and maintaining of the engine state.
In the prior art, the method for monitoring and analyzing the lubricating oil abrasive particles mainly detects the concentration, the form and the like of the abrasive particles in the lubricating oil by means of a spectrometer, a ferrograph and other instruments, and monitors and pre-warns the abnormal lubricating oil abrasive particles by means of numerical analysis, image recognition and the like. In addition, because the oil liquid sensor is arranged in part of engine equipment, the voltage signal in the lubricating oil can be monitored based on the oil liquid sensor and is processed by utilizing a signal processing means, so that the indirect monitoring of the oil liquid abrasive particles is realized.
The existing oil abrasive particle monitoring method mainly has the following defects: (1) The existing monitoring and analyzing method is highly dependent on special equipment, such as a ferrograph, a comprehensive detecting instrument and the like, and the monitoring index of oil abrasive particles is relatively complex; (2) Part of researches are based on deep learning, and simulation data with engine fault labels are used for supervised learning training, so that monitoring of engine faults based on abrasive particle concentration data is realized, but in practice, the fault data of the engine are very few, and the accuracy of a model in practical application and the reliability of daily health management monitoring are limited; (3) Some institutions set warning values of daily monitoring and early warning according to the abrasive particle concentration when an obvious fault occurs in an engine, the technical means cannot meet the requirement of daily health monitoring of equipment, and early warning cannot be performed at the early stage of equipment abrasion.
Disclosure of Invention
The invention aims to provide an engine oil abrasive grain analysis method based on a segmentation threshold value, which aims to solve the technical problems existing in the prior art.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
the engine oil abrasive grain analysis method based on the segmentation threshold comprises the following steps:
step S1: preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data;
step S2: dividing the pretreated lubricating oil data into N different monitoring intervals according to different lubricating oil time or engine time, wherein N is more than or equal to 2;
step S3: selecting an oil monitoring index, fitting numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type of the oil monitoring index;
step S4: based on the confidence, setting thresholds of the lubricating oil monitoring indexes in different monitoring interval sections according to the different monitoring interval sections;
step S5: and carrying out differential early warning on the overrun of abrasive particles in different monitoring intervals according to the lubricating oil monitoring index and the threshold value.
In one embodiment, further comprising step S6: and (3) optimizing a threshold value, and adjusting the threshold value of the lubricating oil monitoring index in a confidence degree adjusting mode according to the early warning result.
Further, the preprocessing in step S1 includes: (1) Discarding the data and marking the data as abnormal if the engine time or the oil time cannot be obtained analytically or the obtained engine time or the oil time is wrong; (2) For the same aircraft and engine, measurements were made multiple times a day, leaving only the latest piece of data.
Further, the lubricating oil monitoring index comprises a lubricating oil abrasive particle element concentration and a lubricating oil abrasive particle element concentration increasing rate.
Further, the increase rate of the element concentration of the lubricating oil abrasive particles is calculated by selecting K pieces of data of the element concentration of the lubricating oil abrasive particles adjacent to the current data point, wherein the K value is a set value.
Further, the method for setting the threshold in step S4 specifically includes the following steps: and determining the upper score of distribution under a specific confidence coefficient according to the distribution type, and taking the upper score as a threshold value for exceeding the value of the lubricating oil monitoring index.
Further, the early warning in step S5 includes: (1) When the two indexes of the element concentration of the lubricating oil abrasive particles and the increase rate of the element concentration of the lubricating oil abrasive particles are abnormal, judging that the monitoring of the engine oil is abnormal; (2) And when one index of the concentration of the lubricating oil abrasive particle element and the increase rate of the concentration of the lubricating oil abrasive particle element is abnormal, judging that the monitoring of the engine oil is abnormal.
Further, the oil abrasive particle data refers to abrasive particle concentration data obtained by detecting abrasive particles of engine lubricating oil sampled periodically through an in-oil abrasive particle monitoring instrument.
In order to achieve the above object, the present invention further provides a system for implementing the engine oil abrasive grain analysis method based on the segmentation threshold, which comprises:
and a data preprocessing module: preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data;
the monitoring interval dividing module: dividing the pretreated lubricating oil data into N different monitoring intervals according to different lubricating oil time or engine time, wherein N is more than or equal to 2;
the lubricating oil monitoring index selection module: selecting an oil monitoring index, fitting numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type of the oil monitoring index;
the lubricating oil monitoring index threshold value setting module: based on the confidence, setting thresholds of the lubricating oil monitoring indexes in different monitoring interval sections according to the different monitoring interval sections;
and the differentiation early warning module is used for: and carrying out differential early warning on the overrun of abrasive particles in different monitoring intervals according to the lubricating oil monitoring index and the threshold value.
In order to achieve the above object, the present invention further provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the engine oil abrasive grain analysis method based on the segment threshold.
Compared with the prior art, the invention has the following beneficial effects:
(1) According to the invention, based on oil abrasive particle data, different monitoring intervals are divided for the oil data according to different oil time or engine time, abrasive particle concentration thresholds are set in different monitoring intervals in a targeted manner, the difference of abrasive particle concentrations of the engine in different states is fully considered, and the monitoring and early warning method is more rigorous and scientific.
(2) The method only needs abrasive particle concentration data, has low functional requirements on the detecting instrument, has simple and understandable principle and simple calculation method, and has stronger practical popularization significance especially for basic units with poor aircraft oil analysis conditions.
(3) According to the invention, the index distribution is monitored by fitting, so that the threshold is calculated based on the set confidence coefficient, and compared with a fixed threshold set based on a statistical result in a traditional method, the threshold provided by the invention can be adjusted according to the distribution type of data fitting and the confidence coefficient set by a user, so that the method is more flexible.
Drawings
FIG. 1 shows the numerical distribution and fitting distribution of the concentration of Fe element when the lubricating oil time is 0-200 h.
FIG. 2 shows the numerical distribution and fitting distribution of the concentration of Fe element at the time of 200-400h of lubricating oil.
FIG. 3 shows the distribution of the concentration values and the fitting distribution of Fe element at the time of lubricating oil of 400-600 h.
FIG. 4 is a numerical distribution and a fitting distribution of the increase rate of the concentration of Fe element at the engine time of 0 to 500 hours.
FIG. 5 shows the numerical distribution and the fitting distribution of the increase rate of the concentration of Fe element at the engine time of 500-1000 hours.
FIG. 6 shows the numerical distribution and fitting distribution of the Fe element concentration increase rate at the engine time of 1000-1500 hours.
FIG. 7 shows the numerical distribution and fitting distribution of the Fe element concentration increase rate at 1500-2000h (i.e. above 1500 h) of engine time.
Fig. 8 is a flow chart of the present invention.
Fig. 9 is a schematic block diagram of the present invention-embodiment 2.
Detailed Description
The present invention will be further described in detail with reference to examples so as to enable those skilled in the art to more clearly understand and understand the present invention. It should be understood that the following specific embodiments are only for explaining the present invention, and it is convenient to understand that the technical solutions provided by the present invention are not limited to the technical solutions provided by the following embodiments, and the technical solutions provided by the embodiments should not limit the protection scope of the present invention.
Example 1
As shown in fig. 8, the embodiment provides an engine oil abrasive grain analysis method based on a segmentation threshold, which is based on the data of the concentration of the abrasive grains of the lubricating oil, divides the data of the lubricating oil into different monitoring intervals according to different lubricating oil time or engine time, sets the threshold of the concentration of the abrasive grains in the different monitoring intervals in a targeted manner, and realizes monitoring and early warning on abnormal conditions of the engine through monitoring of the abrasive grains of the lubricating oil through monitoring and analysis of the concentration of the abrasive grains in the oil.
In this embodiment, the specific content of the engine oil abrasive grain analysis method based on the segmentation threshold value is as follows:
1. preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data
The oil abrasive particle data refers to abrasive particle concentration data obtained by detecting abrasive particles of engine lubricating oil sampled and collected periodically through an in-oil abrasive particle monitoring instrument (such as a spectrometer and the like), and belongs to the existing data.
In this embodiment, the main data preprocessing method includes: (1) For the engine time or the oil time which cannot be obtained by analysis or the obtained engine time (oil time) is wrong, discarding the data during analysis, and marking the oil data as abnormal so as to remind a data acquisition personnel to recalibrate the data; (2) The same aircraft and engine are measured multiple times in a day, and only the latest piece of data is left. It should be noted that, because the purpose of oil monitoring is to find the abnormal value in the oil abrasive particle data, the reasonable value range of the numerical value is not set during the data preprocessing, so as to avoid deleting the important abnormal value by mistake.
For example: the data of the lubricating oil abrasive particles of the batch are derived from a spectrometer of a certain model, and fields such as engine time, lubricating oil time, airplane model, concentration values of different elements and the like in the data derived by the spectrometer are stored in the form of spliced texts, so that firstly, the data which cannot be normally analyzed or repeatedly occur are analyzed according to the format of the derived text of the spectrometer.
2. Dividing the pretreated lubricating oil data into N different monitoring intervals according to different lubricating oil time or engine time, wherein N is more than or equal to 2
The concentration of abrasive particles in the lubricating oil is often affected by a number of factors, including: the longer the engine is used, the more the wear of devices is accumulated along with time, the more the number of abrasive particles is generated, and each oil change can cause larger fluctuation of the concentration of the abrasive particles in the lubricating oil, meanwhile, the concentration of the abrasive particles in the lubricating oil can be correspondingly changed along with accumulation of the lubricating oil time and deposition of the abrasive particles because the abrasive particles are suspended in the lubricating oil.
The applicant finds that the abrasive particle concentration in the lubricating oil has great difference in different engine time or lubricating oil time, and when monitoring and analyzing the lubricating oil abrasive particle concentration data, the influence of the engine time length and the lubricating oil time length corresponding to the abrasive particle data must be considered.
Compared with the prior art, the method adopts the technical means of not distinguishing the difference of the oil abrasive particle concentration under different engine states, and fully considers that the abrasive particle data on different lubricating oil times (or engine times) have obvious differences, so that when the abrasive particle data is monitored, the lubricating oil times (or engine times) to which the abrasive particle data belong are required to be distinguished for independent monitoring analysis. Obviously, it is impossible to set a monitoring standard for each lubricating oil time, so that the lubricating oil time (or engine time) needs to be divided into discrete different sections, the abrasive particle data of the different sections are analyzed for numerical variability, and based on the different sections, a monitoring threshold standard is set for each different section independently, so as to realize differential monitoring and early warning.
Therefore, the present embodiment divides the oil data into different sections according to different oil time and engine time. For example: considering the effect of the oil time on the oil data, since the oil sampling monitoring service typically samples every 50 hours, the oil time is about 600 hours at the maximum, the oil time interval is preferably an integer multiple of 50 hours (e.g., 50 hours, 100 hours, 200 hours, etc.), for example: the maximum oil-lubricating time in the existing data is 600 hours, and the interval size can be set to be 200 hours, so that the data is divided into: the lubricating oil time is in three monitoring intervals of 0-200h,200-400h and 400-600 h. All existing oil sampling data are divided according to the selected oil time interval.
When the interval is divided, the interval can be divided according to the lubricating oil time or the engine time. For example: considering the effect of engine time on the oil data, since the oil is sampled every 50 hours and the engine time is up to about 2000 hours, the engine time interval is equally well selected to be an integer multiple of 50 hours (e.g., 200 hours, 500 hours, 1000 hours, etc.), and the oil data is partitioned at selected engine time intervals, for example: in the existing data, the maximum engine time is 2000 hours, and the settable interval size is 500 hours, so that the data is divided into: the lubricating oil time is in four monitoring intervals of 0-500h, 500-1000h, 1000-1500h and 1500-2000 h.
It should be noted that the size and number of the divided sections can be flexibly adjusted, different section dividing modes, and the section threshold values of the monitoring indexes determined in the subsequent steps will also have differences, so that the flexible adjustment can be performed according to the analysis requirements and effects 。
3. Selecting an oil monitoring index, fitting the numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type
In the oil abrasive particle concentration data, the concentration ratio of each key element is contained, different metal elements represent different abrasion types and abrasion positions, different abrasive particle element concentrations represent the change of the abrasive particle quantity in the oil, and the change of the abrasion degree is indirectly represented; meanwhile, according to the principle of device abrasion, the device abrasion is increased, and the concentration increase rate of abrasive particles in oil liquid is increased.
In combination with the design principle of the invention, two indexes are selected as the lubricating oil monitoring indexes in the embodiment: the concentration of the element of the lubricating oil abrasive particles and the increase rate of the concentration of the element of the lubricating oil abrasive particles are used as key monitoring indexes of the lubricating oil, and the introduction of the index of the increase rate of the concentration of the element of the lubricating oil abrasive particles has the advantages that the abrasive particle concentration data is easily interfered by multiple factors, only the concentration of the abrasive particles is considered, large deviation is easily caused, and after the index is introduced, the early warning accuracy can be improved.
The concentration of the abrasive particle elements is the concentration value of the key elements, and the key elements refer to the important focused abrasive particle elements, such as Fe element, cu element, cr element and the like, which are determined according to the analysis requirements in daily oil quality monitoring service, and the key elements can be flexibly adjusted according to the monitoring requirements, so that the method is not particularly limited; the abrasive grain element concentration increase rate is calculated by selecting K abrasive grain element concentration data adjacent to the current data point, the K value is a set value, and the value is set and adjusted according to the size of the data amount, and is not particularly limited herein.
After the oil monitoring index is selected, the numerical distribution of the monitoring index values of all the oil data in each monitoring interval is fitted on the basis of the division of the monitoring intervals. Fitting the numerical distribution, namely firstly counting the numerical distribution of the monitoring index value, then fitting the statistical distribution by using the existing common distribution function, and solving the parameters of the distribution so as to determine the distribution type of the monitoring index value; typically, numerical statistics and distribution fitting are often accomplished by tools such as distfit functions in the python library or automated by tool software such as matlab.
For example: aiming at the monitoring index of the element concentration of the lubricating oil abrasive particles, the influence of the lubricating oil time is mainly considered, the original abrasive particle data are divided according to different lubricating oil time intervals (namely three monitoring interval sections of 0-200h,200-400h and 400-600 h), the element concentration of the lubricating oil abrasive particles is independently and statistically analyzed on each lubricating oil time interval, and the numerical distribution and fitting distribution of the monitoring index of the element concentration of the lubricating oil abrasive particles on each interval are shown in figures 1, 2 and 3. Figures 1, 2 and 3 show the numerical distribution of the lubricating oil abrasive particle monitoring index 'Fe element concentration' and the fitting distribution thereof in different intervals of 0-200 hours, 200-400 hours and 400-600 hours respectively. As noted in fig. 1, 2, and 3, the distribution with the smallest fitting error is an exponential distribution, and a lognetwork distribution, respectively, it can be seen that the numerical distribution and the fitting distribution of the lubricating oil abrasive particle data in different time intervals have differences, thereby further verifying the necessity of partition monitoring of the data.
For another example: for the monitoring index of the concentration increase rate of the element of the lubricating oil abrasive particles, the influence of the time of the lubricating engine is mainly considered, the original abrasive particle data are divided according to different engine time intervals (namely four monitoring interval sections of 0-500h, 500-1000h, 1000-1500h and more than 1500-2000 h), the concentration increase rate of the element of the lubricating oil abrasive particles is independently and statistically analyzed on each engine time interval, and the numerical distribution and fitting distribution of the monitoring index are shown in the following figures 4, 5, 6 and 7.
4. Based on the confidence, according to different monitoring interval sections, setting the threshold value of the lubricating oil monitoring index in the monitoring interval section
The prior art adopts the technical means that: the setting process of the threshold value is based on experience or simple numerical statistics, and mainly has the technical problems that the setting method is not scientific enough, the set fixed threshold value is highly dependent on experience, and can not be dynamically adjusted along with the change of equipment state and the accumulation of historical data, so that the long-term and accurate monitoring of oil abrasive particles is not facilitated; in contrast, in this embodiment, based on the distribution type determined by fitting the monitoring index value, the upper score of the distribution under the specific confidence coefficient is determined and used as the threshold value for exceeding the monitoring index value. The more the historical lubricating oil data is, the more accurate the result of data fitting distribution in each interval is, the more accurate the threshold value is determined, and the more reliable the early warning result is.
It should be noted that, the higher the confidence is, the more strict the threshold is set, which is easy to cause higher false alarm rate of monitoring and early warning, whereas the lower the confidence is, the higher false alarm rate of monitoring and early warning is easy to cause. The confidence value can be set according to the monitoring requirement. Further, for different abrasive particle monitoring elements (such as Fe, cu, cr, etc.), different abrasive particle monitoring indexes (such as abrasive particle element concentration, abrasive particle element concentration increasing rate), in practice, the differential setting can be performed in combination with specific monitoring needs and analysis emphasis. The method for determining the quantiles on a certain distribution according to the set confidence is a conventional statistical analysis method, and is usually automatically calculated by a statistical tool, and is not described herein.
For example: setting the confidence coefficient to be 0.005, and solving the upper score of the distribution, which is the oil monitoring index threshold value of the data in each interval section, according to the distribution obtained by fitting in the third step and the data of each interval section based on the confidence coefficient. Taking Fe element and Cu element as examples, the monitoring threshold values of the "concentration of the lubricating oil abrasive particle element" are shown in table 1:
table 1 monitoring thresholds of the slip monitoring index "concentration of slip abrasive element" over different slip time intervals
As can be seen from table 1, compared with a single fixed threshold determined by experience and fault data, the method provided by the embodiment gives the oil monitoring threshold under different oil time, the early warning is more flexible and more specific, the numerical value is generally lower than the warning threshold given by the fault data in the traditional method, the early warning prompt can be given at the early stage of the abnormal engine, and the timeliness of daily health monitoring of equipment is improved.
Similarly, for the abrasive grain concentration increase rate, the confidence is set to 0.01, and the monitoring threshold value of the "lubricating oil abrasive grain element concentration increase rate" can be obtained is shown in table 2:
TABLE 2 monitoring thresholds for the slip monitoring index "slip abrasive element concentration increase Rate" over different engine time intervals
As can be seen from table 2, compared with the conventional rule-of-thumb threshold, the present embodiment provides a monitoring threshold for the increase rate of the concentration of the abrasive particles in the lubricating oil at different engine times (i.e., different engine state phases), and the early warning is more flexible and timely.
5. According to the lubricating oil monitoring index and the threshold value, carrying out differential early warning on overrun of abrasive particles in different monitoring interval sections
Different time intervals correspond to different states of the engine and the lubricating oil, and the abnormal performance of the abrasive particle concentration data is different, so that when the technical means provided by the embodiment is used for early warning by utilizing the lubricating oil abrasive particle data, differential early warning is required to be carried out for different time intervals.
According to the embodiment, a monitoring threshold corresponding to each monitoring interval is calculated according to each monitoring interval, so that refined early warning of lubricating oil monitoring is realized, and the early warning condition comprises: (1) When the two indexes of the element concentration of the lubricating oil abrasive particles and the increase rate of the element concentration of the lubricating oil abrasive particles are abnormal, judging that the monitoring of the engine oil is abnormal; (2) And when one index of the concentration of the lubricating oil abrasive particle element and the increase rate of the concentration of the lubricating oil abrasive particle element is abnormal, judging that the monitoring of the engine oil is abnormal. In another embodiment, pre-warning according to weight coefficients may also be employed, such as: and setting different importance weight coefficients for the two indexes, giving a comprehensive early warning score through weighted summation, giving corresponding grade early warning when the comprehensive early warning score exceeds a certain range, and the like. Preferably, in order to find an abnormality of the engine as early as possible by the oil abrasive grains, the present embodiment selects the above-described (2) type of early warning method.
For example: in combination with the above example, two oil abrasive particle monitoring indexes, namely, the "concentration of the element of the lubricating oil abrasive particle" and the "increase rate of the concentration of the element of the lubricating oil abrasive particle" and the corresponding alarm threshold values thereof, are determined through the previous steps. In a certain oil spot check of a certain flight base, the element concentration of Fe element is detected to be 7.2, the element concentration increase rate is 1.72, and according to the conventional fixed threshold rule (see table 1 and table 2) of the base, the two indexes of the oil spot check result are not exceeded, but the engine time (405 hours) and the oil time (382 hours) of the oil spot check are smaller (namely, the oil spot check belongs to the early stage of the engine and is new oil just changed) in consideration of the oil spot check in the spot check, and according to the method provided by the invention, the two indexes in the spot check result exceed the alarm threshold value of the Fe element concentration in the current engine time and the oil time state, so that early warning is sent out. Compared with the traditional fixed threshold rule, the method can more accurately monitor the lubricating oil abrasive particles and early warn the health of the engine, and timely find the potential safety hazard of the engine.
It should be noted that the present invention focuses on providing a method for monitoring and analyzing oil abrasive particles, and reflecting an analysis method with operability, namely, by establishing an abrasive particle monitoring index and setting a monitoring threshold value, providing an early warning prompt of abrasive particle overrun; however, how to judge the overall anomaly behind the anomaly of the single index belongs to specific business rules and business experiences, and can be selected by combining specific application scenes and business requirements (such as early warning strength, false alarm tolerance, missed alarm tolerance and the like), and the method is not particularly limited.
6. Threshold optimization
In actual early warning execution, early warning results can have false alarm rate and missed alarm rate of early warning, the false alarm rate and missed alarm rate are counted, and early warning requirements (such as false alarm tolerance, missed alarm tolerance and the like) are combined, and the threshold value of a monitoring index is adjusted by adjusting the confidence level, so that oil monitoring early warning is continuously improved, and the early warning accuracy is improved; the higher the confidence is, the more strict the threshold is set, which is easy to cause higher false alarm rate of monitoring and early warning, otherwise, the lower the confidence is, the higher false alarm rate of monitoring and early warning is easy to cause. Taking the alarm omission rate as an example, if the alarm omission rate is higher, the threshold value is set to be stricter, and the confidence level can be lowered; taking the false alarm rate as an example, if the false alarm rate is higher, the threshold value is loosely set, and the confidence level can be increased.
Example 2
As shown in fig. 9, the present embodiment provides a system for implementing the engine oil abrasive grain analysis method based on the segmentation threshold provided in embodiment 1, specifically, the system includes:
and a data preprocessing module: preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data;
the monitoring interval dividing module: dividing the pretreated lubricating oil data into N different monitoring intervals according to different lubricating oil time or engine time, wherein N is more than or equal to 2;
the lubricating oil monitoring index selection module: selecting an oil monitoring index, fitting numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type of the oil monitoring index;
the lubricating oil monitoring index threshold value setting module: based on the confidence, setting thresholds of the lubricating oil monitoring indexes in different monitoring interval sections according to the different monitoring interval sections;
and the differentiation early warning module is used for: and carrying out differential early warning on the overrun of abrasive particles in different monitoring intervals according to the lubricating oil monitoring index and the threshold value.
In a still further preferred scheme, the system further comprises a threshold optimization module, and the threshold of the lubricating oil monitoring index is adjusted by adjusting the confidence coefficient according to the early warning result; specifically, the false alarm rate and the false alarm rate of the early warning can be counted according to the early warning result, and the early warning requirement is combined, and the threshold value of the monitoring index is adjusted by adjusting the confidence level, so that the aim of continuously improving the oil monitoring early warning is fulfilled.
The working principles of the above modules are in one-to-one correspondence with the method provided in embodiment 1, and thus are not described herein.
It should be noted that, it should be understood that the division of the modules of the above system is merely a division of a logic function, and may be fully or partially integrated into a physical entity or may be physically separated. And these modules may all be implemented in software in the form of calls by the processing element; or can be realized in hardware; the method can also be realized in a form of calling software by a processing element, and the method can be realized in a form of hardware by a part of modules. For example, a module may be a processing element that is set up separately, may be implemented in a chip of an apparatus, may be stored in a memory of the apparatus in the form of program codes, may be called by a processing element of the apparatus and perform functions of a module, and may be implemented similarly. In addition, all or part of the modules can be integrated together or can be independently implemented. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in a software form.
For example, the modules above may be one or more integrated circuits configured to implement the methods above, such as: one or more specific integrated circuits, or one or more microprocessors, or one or more field programmable gate arrays, etc. For another example, when a module above is implemented in the form of a processing element scheduler code, the processing element may be a general purpose processor, such as a central processing unit or other processor that may invoke the program code. For another example, the modules may be integrated together and implemented in a system-on-chip form.
Example 3
The present embodiment provides a computer-readable storage medium having stored thereon a computer program that is executed by a processor to implement the engine oil abrasive grain analysis method based on the segment threshold provided in embodiment 1. Those of ordinary skill in the art will appreciate that: all or part of the steps of implementing the method provided in embodiment 1 may be implemented by hardware associated with a computer program, where the computer program may be stored in a computer readable storage medium, and when executed, the program performs steps including the method provided in embodiment 1; and the storage medium includes: various media that can store program code, such as ROM, RAM, magnetic or optical disks.
The foregoing is a preferred embodiment of the present invention. It should be noted that those skilled in the art may make several modifications without departing from the design principles and technical solutions of the present invention, and these modifications should also be considered as the protection scope of the present invention.
Claims (10)
1. The engine oil abrasive grain analysis method based on the segmentation threshold is characterized by comprising the following steps of:
step S1: preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data;
step S2: dividing the pretreated lubricating oil data into N different monitoring intervals according to different lubricating oil time or engine time, wherein N is more than or equal to 2;
step S3: selecting an oil monitoring index, fitting numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type of the oil monitoring index;
step S4: based on the confidence, setting thresholds of the lubricating oil monitoring indexes in different monitoring interval sections according to the different monitoring interval sections;
step S5: and carrying out differential early warning on the overrun of abrasive particles in different monitoring intervals according to the lubricating oil monitoring index and the threshold value.
2. The method for analyzing engine oil abrasive grains based on the segment threshold according to claim 1, further comprising step S6: and (3) optimizing a threshold value, and adjusting the threshold value of the lubricating oil monitoring index in a confidence degree adjusting mode according to the early warning result.
3. The engine oil abrasive grain analysis method based on the segmentation threshold according to claim 1 or 2, characterized in that: the preprocessing in step S1 includes: (1) Discarding the data and marking the data as abnormal if the engine time or the oil time cannot be obtained analytically or the obtained engine time or the oil time is wrong; (2) For the same aircraft and engine, measurements were made multiple times a day, leaving only the latest piece of data.
4. The method for analyzing engine oil abrasive particles based on the segmentation threshold according to claim 3, wherein the method comprises the following steps of: the lubricating oil monitoring index comprises a lubricating oil abrasive particle element concentration and a lubricating oil abrasive particle element concentration increasing rate.
5. The method for analyzing engine oil abrasive particles based on the segmentation threshold according to claim 4, wherein the method comprises the following steps of: the increase rate of the element concentration of the lubricating oil abrasive particles is calculated by selecting K pieces of data of the element concentration of the lubricating oil abrasive particles adjacent to the current data point, wherein the K value is a set value.
6. The method for analyzing engine oil abrasive particles based on the segmentation threshold according to claim 5, wherein the method comprises the following steps of: the method for setting the threshold in step S4 specifically includes the following steps: and determining the upper score of distribution under a specific confidence coefficient according to the distribution type, and taking the upper score as a threshold value for exceeding the value of the lubricating oil monitoring index.
7. The method for analyzing engine oil abrasive grains based on the segment threshold according to claim 6, wherein the early warning in step S5 includes: (1) When the two indexes of the element concentration of the lubricating oil abrasive particles and the increase rate of the element concentration of the lubricating oil abrasive particles are abnormal, judging that the monitoring of the engine oil is abnormal; (2) And when one index of the concentration of the lubricating oil abrasive particle element and the increase rate of the concentration of the lubricating oil abrasive particle element is abnormal, judging that the monitoring of the engine oil is abnormal.
8. The engine oil abrasive grain analysis method based on the segmentation threshold according to claim 7, wherein the oil abrasive grain data refers to abrasive grain concentration data obtained by conducting abrasive grain detection on engine lubricating oil sampled periodically through an in-oil abrasive grain monitoring instrument.
9. A system for implementing the method for analyzing engine oil abrasive grains based on the segmentation threshold according to any one of claims 1 to 8, comprising:
and a data preprocessing module: preprocessing oil abrasive particle data to obtain preprocessed lubricating oil data;
the monitoring interval dividing module: dividing the pretreated lubricating oil data into N different monitoring intervals according to different lubricating oil time or engine time, wherein N is more than or equal to 2;
the lubricating oil monitoring index selection module: selecting an oil monitoring index, fitting numerical distribution of the oil monitoring index in each monitoring interval section, and determining the distribution type of the oil monitoring index;
the lubricating oil monitoring index threshold value setting module: based on the confidence, setting thresholds of the lubricating oil monitoring indexes in different monitoring interval sections according to the different monitoring interval sections;
and the differentiation early warning module is used for: and carrying out differential early warning on the overrun of abrasive particles in different monitoring intervals according to the lubricating oil monitoring index and the threshold value.
10. A computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the segment threshold based engine oil abrasive grain analysis method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310968076.2A CN117034109A (en) | 2023-08-03 | 2023-08-03 | Engine oil abrasive grain analysis method and system based on segmentation threshold and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310968076.2A CN117034109A (en) | 2023-08-03 | 2023-08-03 | Engine oil abrasive grain analysis method and system based on segmentation threshold and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117034109A true CN117034109A (en) | 2023-11-10 |
Family
ID=88629171
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310968076.2A Pending CN117034109A (en) | 2023-08-03 | 2023-08-03 | Engine oil abrasive grain analysis method and system based on segmentation threshold and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117034109A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110631624A (en) * | 2019-09-04 | 2019-12-31 | 精英数智科技股份有限公司 | Method, device and system for identifying abnormal operation data of mine sensor |
| CN110926809A (en) * | 2019-06-19 | 2020-03-27 | 山西大学 | Big data analysis-based wind turbine generator transmission chain fault early warning method |
| CN112946469A (en) * | 2020-11-26 | 2021-06-11 | 华能澜沧江水电股份有限公司 | Monitoring method for self-adaptive dynamic alarm threshold of hydraulic generator |
| CN114220189A (en) * | 2021-12-15 | 2022-03-22 | 震坤行工业超市(上海)有限公司 | Monitoring method, prediction system, electronic device and storage medium |
| CN115563180A (en) * | 2022-09-19 | 2023-01-03 | 烽台科技（北京）有限公司 | Dynamic threshold generation method, device, equipment and storage medium |
| CN116416764A (en) * | 2021-12-31 | 2023-07-11 | 北京国双科技有限公司 | Alarm threshold generation method and device, electronic equipment and storage medium |
-
2023
- 2023-08-03 CN CN202310968076.2A patent/CN117034109A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110926809A (en) * | 2019-06-19 | 2020-03-27 | 山西大学 | Big data analysis-based wind turbine generator transmission chain fault early warning method |
| CN110631624A (en) * | 2019-09-04 | 2019-12-31 | 精英数智科技股份有限公司 | Method, device and system for identifying abnormal operation data of mine sensor |
| CN112946469A (en) * | 2020-11-26 | 2021-06-11 | 华能澜沧江水电股份有限公司 | Monitoring method for self-adaptive dynamic alarm threshold of hydraulic generator |
| CN114220189A (en) * | 2021-12-15 | 2022-03-22 | 震坤行工业超市(上海)有限公司 | Monitoring method, prediction system, electronic device and storage medium |
| CN116416764A (en) * | 2021-12-31 | 2023-07-11 | 北京国双科技有限公司 | Alarm threshold generation method and device, electronic equipment and storage medium |
| CN115563180A (en) * | 2022-09-19 | 2023-01-03 | 烽台科技（北京）有限公司 | Dynamic threshold generation method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 王广伟: "发动机滑油信息综合与面向任务维修决策", 《中国优秀硕博士学位论文全文数据库（硕士） 信息科技辑》, no. 02, 15 June 2003 (2003-06-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110347116B (en) | Machine tool state monitoring system and monitoring method based on operation data flow | |
| CN109186813A (en) | A kind of temperature sensor self-checking unit and method | |
| CN109816031B (en) | Transformer state evaluation clustering analysis method based on data imbalance measurement | |
| CN107679734A (en) | It is a kind of to be used for the method and system without label data classification prediction | |
| CN113344133B (en) | Method and system for detecting abnormal fluctuation of time sequence behaviors | |
| CN112129928B (en) | Method, computing device and computer storage medium for detecting state of lubricating oil | |
| KR20160114666A (en) | Monitoring device, monitoring method, and program | |
| CN115858794B (en) | Abnormal log data identification method for network operation safety monitoring | |
| CN115524002B (en) | Operation state early warning method, system and storage medium of power plant rotating equipment | |
| CN115410342B (en) | Landslide hazard intelligent early warning method based on real-time monitoring of crack meter | |
| CN116680113B (en) | Equipment detection implementation control system | |
| CN115033463A (en) | Method, device, equipment and storage medium for determining system exception type | |
| CN112720071A (en) | Cutter real-time state monitoring index construction method based on intelligent fusion of multi-energy domain signals | |
| CN107463963A (en) | A kind of Fault Classification and device | |
| CN113551766B (en) | Mechanical equipment working state detection method based on artificial intelligence | |
| CN117170303B (en) | PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction | |
| CN207992717U (en) | A kind of gate of hydropower station on-line condition monitoring system | |
| Chen et al. | THE APPLICATION OF MULTINOMIAL CONTROL CHARTS FOR INSPECTION ERROR. | |
| CN106355314A (en) | A Vehicle Engine Air Valve Quality Streaming Data Processing Method and Device | |
| CN117034109A (en) | Engine oil abrasive grain analysis method and system based on segmentation threshold and computer readable storage medium | |
| CN111274949A (en) | Structural analysis-based blood disease white blood cell scatter diagram similarity analysis method | |
| CN116383645A (en) | Intelligent system health degree monitoring and evaluating method based on anomaly detection | |
| CN115378000A (en) | Power distribution network operation state evaluation method based on interval two-type fuzzy clustering analysis | |
| CN113129530B (en) | Intelligent security electronic fence alarm system based on Internet of things and machine vision | |
| CN114597886A (en) | Power distribution network operation state evaluation method based on interval type two fuzzy clustering analysis |
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 |