CN111488262A - Process health system evaluation method based on cloud model convolution theory - Google Patents
Process health system evaluation method based on cloud model convolution theory Download PDFInfo
- Publication number
- CN111488262A CN111488262A CN202010194042.9A CN202010194042A CN111488262A CN 111488262 A CN111488262 A CN 111488262A CN 202010194042 A CN202010194042 A CN 202010194042A CN 111488262 A CN111488262 A CN 111488262A
- Authority
- CN
- China
- Prior art keywords
- cloud
- sensor
- health
- subsystem
- single sensor
- 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
- 230000036541 health Effects 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000008569 process Effects 0.000 title claims abstract description 14
- 238000011156 evaluation Methods 0.000 title abstract description 27
- 238000005259 measurement Methods 0.000 claims description 27
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000011158 quantitative evaluation Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a process health system evaluation method based on a cloud model convolution theory. The method can effectively realize quantitative evaluation of the working state of the system, enables the health state of the system to be visualized, provides quantitative theoretical support for state prediction of the system, improves the precision of evaluation indexes through a cloud model convolution theory, effectively reduces the probability of system failure occurrence, and greatly improves the reliability of the system.
Description
Technical Field
The invention relates to a process health system evaluation method based on a cloud model convolution theory, which is used for quantitatively evaluating the working state of a system with multiple sensors or measuring points, so that the working state of the system is more visual and reliable, and belongs to the technical field of health management decision of process systems.
Background
Because the complex system comprises the table hole measuring point, the output value of the complex system comprises dozens of or even hundreds of parameters, and the complex system works in a complex or severe environment for a long time, the probability of the fault is greatly increased. After long-term work, when the system works, the faults of the system are mainly concentrated on key sensors, such as an angle sensor, a torque sensor, a Hall current sensor, a position sensor and the like. Once these sensors fail, they will directly affect the performance of the system, even if they fail severely. Therefore, it is necessary to know the operating states of these sensors to effectively reduce the probability of system failure and greatly improve the reliability of the system, and it is necessary to perform a health diagnosis and evaluation technique for the system, key components, sensors, and the like.
At present, most of researches are based on a shallow learning method for the complex system, and key information in the system is often ignored during evaluation, so that the problem of low precision occurs during evaluation, and the evaluation index is deviated, so that the research of a system health evaluation method based on a big data depth theory needs to be further researched.
Disclosure of Invention
Compared with the traditional sensor health assessment model, the process health assessment method based on the cloud model convolution theory can effectively achieve preliminary feature extraction of each sensor through the cloud model theory, achieves preliminary health assessment, performs data fusion through a convolution neural network, obtains health reliability, achieves quantitative health assessment, enables the working state of the system to be visualized and visualized, and improves the reliability of the system.
The invention is realized by the following technical scheme.
The invention provides a process health system evaluation method based on a cloud model convolution theory, which comprises the following steps:
①, acquiring analog quantity acquired by each measuring point of the system through a plurality of sensors, and converting the analog quantity into digital quantity through a data acquisition card to realize data acquisition;
② preprocessing the collected data to obtain the mean and variance of the sample;
③ setting a cloud measurement function of the sensor according to the mapping relation between the output data of the sensor and the health state of the sensor;
④, constructing a single sensor cloud measurement matrix and a subsystem cloud measurement matrix;
⑤ calculating the weight distribution of different points and different sensors for each sensor in the system;
⑥ determining cloud membership degree parameters of the single sensor through the cloud measurement matrix of the single sensor and the weight distribution of different points at different times;
⑦ the health credibility of the single sensor and the subsystem is obtained by using the convolution neural network, the cloud membership parameter of the single sensor and the cloud membership parameter of the subsystem, and the lowest health credibility of the single sensor and the subsystem is used as the final health credibility of the system.
In step ②, the data preprocessing includes noise reduction filtering and normalization of the data.
In constructing the cloud measure function at step ③, the eigenvalue assignment is performed using a normal function.
In the step ⑤, when calculating the weights at different time points, the weight calculation method used is an analytic hierarchy process.
The invention has the beneficial effects that: the health state of each sensor and each system is quantitatively described, the cloud model theory and the convolution nerve theory are applied, the precision of evaluation indexes is improved, the working state of the system can be accurately evaluated, theoretical support is provided for the health prediction of the system, the maintainability of the system is effectively improved, and the reliability of the system is improved.
Drawings
FIG. 1 is a flow chart of the cloud model based convolution theory system and the calculation of health confidence level of each sensor according to the present invention;
FIG. 2 is a schematic diagram of a system based on a cloud model convolution theory and a calculation of health credibility of each sensor according to the present invention;
FIG. 3 is a comparison of a cloud model convolution theory system of the present invention with a gray theory system;
FIG. 4 is a diagram of the health assessment results of a single moment sensor of the cloud model convolution theory based system of the present invention.
Detailed Description
The technical solution of the present invention is further described below, but the scope of the claimed invention is not limited to the described.
As shown in fig. 1, a process health system evaluation method based on a cloud model convolution theory includes the following steps:
①, acquiring analog quantity acquired by each measuring point of the system through a plurality of sensors, and converting the analog quantity into digital quantity through a data acquisition card to realize data acquisition;
② preprocessing the collected data to obtain the mean and variance of the sample;
③ setting a cloud measurement function of the sensor according to the mapping relation between the output data of the sensor and the health state of the sensor;
④, constructing a single sensor cloud measurement matrix and a subsystem cloud measurement matrix;
⑤ calculating the weight distribution of different points and different sensors for each sensor in the system;
⑥ determining cloud membership degree parameters of the single sensor through the cloud measurement matrix of the single sensor and the weight distribution of different points at different times;
⑦ the health credibility of the single sensor and the subsystem is obtained by using the convolution neural network, the cloud membership parameter of the single sensor and the cloud membership parameter of the subsystem, and the lowest health credibility of the single sensor and the subsystem is used as the final health credibility of the system.
In step ②, the data preprocessing includes noise reduction filtering and normalization of the data.
In constructing the cloud measure function at step ③, the eigenvalue assignment is performed using a normal function.
In the step ⑤, when calculating the weights at different time points, the weight calculation method used is an analytic hierarchy process.
Example 1
As described above, a process health system evaluation method based on a cloud model convolution theory aims to qualitatively obtain the health reliability of a system and each component thereof, so as to quantitatively evaluate the working state of the system, as shown in fig. 1, and specifically includes the following steps:
the method comprises the following steps: acquiring analog quantity of each measuring point of the system through a sensor, and converting the analog quantity into digital quantity through a data acquisition card to realize data acquisition;
step two: preprocessing the acquired data, including noise reduction filtering and normalization of the data, and acquiring the mean value, the variance and the like of a sample;
step three: setting a cloud measurement function of the sensor according to the mapping relation between the output data of the sensor and the health state of the sensor;
in the third step, a normal distribution function is adopted as a cloud measurement function, and cloud measurement parameters and output values of the normal distribution function are shown in fig. 2;
further, in the detection of the sensor, generally, when the output value of the sensor exceeds 1 time of the variance, the state is regarded as a sub-healthy state, when the output value exceeds 3 times of the variance, the state is regarded as a fault edge state, and when the output value exceeds 5 times, the fault is considered to be a fault; the four cloud measurement functions are shown in equations (1) to (4):
fHS(x)＝exp[-||x-μ||2/22](1)
health reliability and health status are shown in table 1:
TABLE 1
Step four: constructing a single sensor cloud measurement matrix according to the cloud measurement function in the step three;
specifically, in step four, at time point j being 1,2, …, m, the cloud measurement evaluation matrix for the ith sensor is recorded as: CMEj＝(cmeijk)m×n(i ═ 1,2, …, m; k ═ 1,2, …, n), the format is shown in formula 5:
wherein i represents the serial number of the sensitive unit to be calculated, j represents the position of the time sequence of the data, k represents the serial number of the cloud evaluation index set, cmeijkThe method comprises the steps that at a time point j, when an evaluation set index is k, a cloud measurement function value corresponding to a sensitive unit i is obtained;
step five: constructing a subsystem cloud measurement matrix according to the cloud measurement function in the step three;
in step five, for the multifunctional sensor, the cloud measurement evaluation matrix corresponding to the jth time point is recorded as: CMEj＝(cmeijk)m×n(i ═ 1,2, …, m; k ═ 1,2, …, n), as shown in formula 6:
wherein j represents the time point sequence, i represents the corresponding sensitive unit serial number, k represents the serial number of the evaluation index set, cmeijkThe method comprises the steps that at a time point j, when an evaluation set index is k, a cloud measurement function value corresponding to a sensitive unit i is obtained;
step six: carrying out weight calculation method AHP method on each sensor in the system to realize weight distribution at different time points;
preferably, the AHP method in step six is divided into the following four steps:
step 6.1, calculating a time sequence deviation vector of the sensor i: deviation d at time jij＝|xij-μijL, where μijThe best estimated value of the data at the moment; deviation d when the actual output value is far from the optimum estimated valueijThe larger. With a timing offset vector of di＝[di1,di2,…,dim]；
Step 6.2, constructing pairwise-comparison health evaluation matrix BCM by using deviation valuesi: comparing the deviation values of any two time points, and judging the deviation between the two time points so as to determine the importance degree of the health state of the sensitive unit between the two time points;
step 6.3 of solving health evaluation matrix BCMiAnd the eigenvector α, and selecting the largest eigenvalue λmaxAnd corresponding feature vector αmax；
BCMiα＝λα (8)
Step 6.4, consistency check is carried out, whether the non-consistency of the matrix can be accepted or not is judged, specifically, a consistency Index (consistency Index, CI) is calculated, as shown in a formula (9), and then a corresponding average Random consistency Index (Random Index, RI) is found out;
further, table 2 shows that the RI of 1000 experimental results when n is less than 15 is taken, the consistency Ratio (Consist Ratio, CR) is calculated, when CR is less than 0.1, the matrix is determined to meet the random consistency index, and conversely, if CR is greater than or equal to 0.1, the matrix a is determined not to meet the random consistency index, and the matrix needs to be adjusted until CR is less than 0.1, as shown in equation (10).
TABLE 2
Step seven: determining cloud membership parameters of a single sensor by applying the cloud measurement matrix of the sensor obtained in the fourth step and the weight calculated in the sixth step;
seventhly, acquiring cloud membership parameters through a formula (11);
CMEVsensor＝WAHP×CMEi(11)
step eight: determining cloud membership parameters of the subsystems by applying the system cloud measurement matrix obtained in the fifth step and the weights calculated in the seventh step;
step nine: respectively calculating the health credibility of each sensor and each subsystem by using a convolutional neural network;
further, in the ninth step, the cloud membership parameter of the sensor used as the evaluation index is the cloud membership parameter CMEV obtained in the eighth stepsensor＝[crdA1crdA2crdA3crdA4]；
Further, in the ninth step, the subsystem cloud membership parameter used as the evaluation index is the cloud membership parameter CMEV obtained in the ninth stepsubsystem＝[crdA1crdA2crdA3crdA4]；
Further, in the ninth step, the acquired cloud membership parameter is transformed by the convolutional neural network to acquire the health reliability, and a schematic diagram thereof is shown in fig. 3.
HRD＝f(crdA1,crdA2,crdA3,crdA4) (12)
Step ten: and taking the lowest health reliability of the subsystem as the final health reliability of the system.
Example 2
The invention takes a steering engine system health state evaluation method as an example, and the system comprises four main parts such as an angle sensor (A type sensor), a torque sensor (B type sensor), a Hall current sensor (C type sensor), a position sensor (D type sensor) and the like which are respectively used as four subsystems.
The test was carried out by experimental verification using 400 groups of samples (100 groups for each state), the test results are shown in fig. 4, and the evaluation accuracy is shown in table 3:
TABLE 3
The evaluation accuracy rate is increased from the original 87.25% to 97.25%, the invention takes the health reliability of calculating single moment as an example, and the result is shown in fig. 4, wherein the curve on the graph represents the health reliability of the measuring point at each moment.
Claims (4)
1. A process health system assessment method based on a cloud model convolution theory is characterized in that: the method comprises the following steps:
① acquiring analog quantity collected by each measuring point of the system via a sensor, and converting the analog quantity into digital quantity via a data acquisition card to realize data acquisition;
② preprocessing the collected data to obtain the mean and variance of the sample;
③ setting a cloud measurement function of the sensor according to the mapping relation between the output data of the sensor and the health state of the sensor;
④, constructing a single sensor cloud measurement matrix and a subsystem cloud measurement matrix;
⑤ calculating the weight distribution of different points and different sensors for each sensor in the system;
⑥ determining cloud membership degree parameters of the single sensor through the cloud measurement matrix of the single sensor and the weight distribution of different points at different times;
⑦ the health credibility of the single sensor and the subsystem is obtained by using the convolution neural network, the cloud membership parameter of the single sensor and the cloud membership parameter of the subsystem, and the lowest health credibility of the single sensor and the subsystem is used as the final health credibility of the system.
2. The method of claim 1, wherein the step ② of pre-processing the data includes de-noising, filtering, and normalizing the data.
3. The method for evaluating the health system of a process based on the cloud model convolution theory as claimed in claim 1, wherein the distribution of the eigenvalues is implemented using a normal function when the cloud measurement function is constructed in step ③.
4. The method of claim 1, wherein the weight calculation method is an analytic hierarchy process when the weights at different time points are calculated in step ⑤.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010194042.9A CN111488262A (en) | 2020-03-19 | 2020-03-19 | Process health system evaluation method based on cloud model convolution theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010194042.9A CN111488262A (en) | 2020-03-19 | 2020-03-19 | Process health system evaluation method based on cloud model convolution theory |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111488262A true CN111488262A (en) | 2020-08-04 |
Family
ID=71812589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010194042.9A Pending CN111488262A (en) | 2020-03-19 | 2020-03-19 | Process health system evaluation method based on cloud model convolution theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111488262A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117010060A (en) * | 2023-07-19 | 2023-11-07 | 重庆交通大学 | Concrete dam health diagnosis method and system based on improved D-S evidence theory |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109669849A (en) * | 2018-12-04 | 2019-04-23 | 哈尔滨工业大学 | It is a kind of based on the complication system health state evaluation method for not knowing depth theory |
-
2020
- 2020-03-19 CN CN202010194042.9A patent/CN111488262A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109669849A (en) * | 2018-12-04 | 2019-04-23 | 哈尔滨工业大学 | It is a kind of based on the complication system health state evaluation method for not knowing depth theory |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117010060A (en) * | 2023-07-19 | 2023-11-07 | 重庆交通大学 | Concrete dam health diagnosis method and system based on improved D-S evidence theory |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110533167B (en) | Fault diagnosis method and system for electric valve actuating mechanism | |
| CN109710983B (en) | Diesel engine cylinder layered fault diagnosis method based on key performance indexes | |
| CN113188794B (en) | Gearbox fault diagnosis method and device based on improved PSO-BP neural network | |
| CN108090515B (en) | Data fusion-based environment grade evaluation method | |
| CN109669849B (en) | Complex system health state assessment method based on uncertain depth theory | |
| CN113159162A (en) | Fault diagnosis method and system based on information fusion and grey correlation | |
| CN114266289A (en) | Complex equipment health state assessment method | |
| CN111488262A (en) | Process health system evaluation method based on cloud model convolution theory | |
| CN110220885B (en) | Comprehensive judgment method for wear state of mechanical equipment | |
| CN111007390A (en) | Analog circuit fault diagnosis model based on algebraic method | |
| CN102289718A (en) | Method for softly measuring height of mud layer in settlement process of red mud | |
| CN111507374A (en) | Power grid mass data anomaly detection method based on random matrix theory | |
| CN115828114A (en) | Energy consumption abnormity detection method for aluminum profile extruder | |
| CN115841446A (en) | Rapid detection system and method for fault risk of solar heat collection field | |
| CN114814707A (en) | Intelligent ammeter stress error analysis method, equipment, terminal and readable medium | |
| CN113807431A (en) | Intelligent spindle state evaluation method and system based on multi-source information fusion | |
| CN112621381A (en) | Intelligent health state evaluation method and device for machine tool feeding system | |
| CN117909668B (en) | Bearing fault diagnosis method and system based on convolutional neural network | |
| CN112116003B (en) | Self-adaptive dynamic fault diagnosis method and system for complex sensor network | |
| CN117688496B (en) | Abnormality diagnosis method, system and equipment for satellite telemetry multidimensional time sequence data | |
| CN113486742B (en) | Fault identification method, device and system and computer readable storage medium | |
| CN117272055B (en) | Electric energy meter abnormality detection method and device based on filtering enhancement self-encoder | |
| CN117055537B (en) | Motorcycle electric control system fault diagnosis method based on Internet of things | |
| CN114611865A (en) | Instrument control platform-based health assessment method for industrial control station instrument | |
| CN114372344A (en) | Quantitative identification method for subsynchronous oscillation damping characteristic influence factors |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200804 |
|
| RJ01 | Rejection of invention patent application after publication |