CN112196784A - ELM-based high-pressure diaphragm pump health state estimation system and method - Google Patents
ELM-based high-pressure diaphragm pump health state estimation system and method Download PDFInfo
- Publication number
- CN112196784A CN112196784A CN202010984240.5A CN202010984240A CN112196784A CN 112196784 A CN112196784 A CN 112196784A CN 202010984240 A CN202010984240 A CN 202010984240A CN 112196784 A CN112196784 A CN 112196784A
- Authority
- CN
- China
- Prior art keywords
- diaphragm pump
- pressure diaphragm
- sensor
- state
- elm
- 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.)
- Granted
Links
- 230000036541 health Effects 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 title claims abstract description 13
- 230000005540 biological transmission Effects 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 230000004927 fusion Effects 0.000 claims abstract description 11
- 230000001133 acceleration Effects 0.000 claims abstract description 10
- 230000015556 catabolic process Effects 0.000 claims abstract description 8
- 238000006731 degradation reaction Methods 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims description 10
- 238000007635 classification algorithm Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000010521 absorption reaction Methods 0.000 claims 1
- 230000003862 health status Effects 0.000 claims 1
- 238000004949 mass spectrometry Methods 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000003745 diagnosis Methods 0.000 abstract description 2
- 238000005299 abrasion Methods 0.000 description 12
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000009776 industrial production Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006854 communication Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Mechanical Engineering (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Measuring Fluid Pressure (AREA)
Abstract
The invention relates to a system and a method for estimating the health state of a high-pressure diaphragm pump based on ELM, belonging to the technical field of fault diagnosis. The method comprises the steps of firstly transmitting vibration signals, temperature signals, acoustic emission signals, leakage quantity signals and pressure signals of the high-pressure diaphragm pump, which are respectively acquired by an acceleration sensor, a temperature sensor, an acoustic emission sensor, a leakage flow sensor and a pressure sensor, to a PC (personal computer) end analysis module through a wireless WIFI (wireless fidelity) transmission module through a signal acquisition module, performing data fusion by adopting a weighted fusion algorithm, fitting a performance degradation index capable of comprehensively reflecting fault information of the high-pressure diaphragm pump, and establishing a health condition estimation model of the high-pressure diaphragm pump through improved ELM (electronic engineering model) so as to estimate the running state of the high-pressure diaphragm pump. The method has the advantages of safety, convenience, high efficiency, high accuracy and the like, and is suitable for estimating the health state of the high-pressure diaphragm pump of various devices.
Description
Technical Field
The invention relates to a system and a method for estimating the health state of a high-pressure diaphragm pump based on ELM, belonging to the technical field of fault diagnosis.
Background
The high-pressure diaphragm pump plays a vital role in an ore pulp pipeline conveying system, can obviously improve the operation efficiency while bringing economic benefits to production enterprises, and is widely applied to various fields. In actual operation, the operational health of high pressure diaphragm pumps is a significant concern. If the valve body is replaced according to the minimum full period, a large number of available valve bodies cannot be fully utilized, so that the production cost is increased, the economic benefit is reduced, and the operation period of the high-pressure diaphragm pump is shortened. The cost of manual inspection is high, so that the occurrence of faults is difficult to completely avoid, and the faults are difficult to forecast in advance.
The running state of the high-pressure diaphragm pump is remotely monitored and estimated, so that the fault of the high-pressure diaphragm pump can be effectively diagnosed on line, and the fault is prevented.
Disclosure of Invention
The invention aims to solve the technical problem of providing a health state estimation system and method of a high-pressure diaphragm pump based on ELM, which are used for solving the problem of a series of production safety accidents such as valve chamber breakdown and the like caused by valve body abrasion in the severe working environment of the high-pressure diaphragm pump, realize effective prediction of the running state of the high-pressure diaphragm pump and have strong practicability and engineering significance.
The technical scheme of the invention is as follows: the utility model provides a high pressure diaphragm pump health state pre-estimate system based on ELM, includes signal acquisition module, wireless WIFI transmission module and PC end analysis module.
The signal acquisition module comprises acceleration sensor, temperature sensor, acoustic emission sensor, reveal flow sensor, pressure sensor and STM32f103 chip.
The wireless WIFI transmission module consists of an ESP8266 and a wireless router.
The signal acquisition module is connected the vibration signal, temperature signal, acoustic emission signal, the volume of revealing signal and the pressure signal of the high pressure diaphragm pump that acceleration sensor, temperature sensor, acoustic emission sensor, reveal flow sensor and pressure sensor gathered respectively with the serial ports of singlechip STM33f103 chip and wireless WIFI transmission module ESP8266, realizes the conversion between serial ports and the WIFI to realize the wireless transmission of data. The wireless router uploads the data to the cloud end through the Internet, the cloud end downloads the data to the PC end analysis module through the Internet and the router, the PC end analysis module performs data fusion on the received data by adopting a weighted fusion algorithm to fit a performance degradation index capable of comprehensively reflecting fault information of the high-pressure diaphragm pump, and a health condition estimation model of the high-pressure diaphragm pump is established through the improved ELM so as to estimate the running state of the high-pressure diaphragm pump, so that the high-pressure diaphragm pump is judged to be in a normal state, a light fault state, a severe fault state or a failure state.
The acceleration sensor and the temperature sensor in the signal acquisition module are axially arranged on the high-pressure diaphragm pump, the acoustic emission sensor is radially arranged on the high-pressure diaphragm pump, the leakage flow sensor and the pressure sensor are vertically arranged on the high-pressure diaphragm pump, and three acquisition channels are adopted to acquire data in parallel. The leakage flow of the high-pressure diaphragm pump is a main characteristic parameter reflecting the abrasion condition of the pump, is related to the output pressure of the pump, the real-time discharge capacity of the pump, the temperature and the rotating speed, the influence of the factors is comprehensively considered, and 5 types of sensors are selected to respectively carry out signal acquisition on the running state of the high-pressure diaphragm pump.
The normal state, the light fault state, the severe fault state or the failure state of the high-pressure diaphragm pump can be judged by the abrasion degree of the diaphragm pump, and the low fault state is defined when the abrasion degree is 0.1-5 mm. And when the abrasion degree is 5mm-10mm, the state is a severe fault state. The wear was in a failure state when the wear was more than 10 mm. When the data acquired in real time on line are analyzed, when the abrasion degree exceeds 5mm, the data are judged to be in a serious fault state. And further analyzing whether the state belongs to a failure state. When the fault belongs to a severe fault state or a failure state, the system interface can automatically alarm, and the system alarms to remind field workers to remove the fault in time, so that the normal operation of the industrial production equipment under the condition of the maximum utilization rate is ensured.
The use method of the health state estimation system of the high-pressure diaphragm pump based on the ELM comprises the following specific steps:
step 1: collecting the full-period life data of the high-pressure diaphragm pump, extracting multi-dimensional characteristic parameters of the original data, and reducing the dimensions through LDA to obtain an initial data sample serving as a training sample set.
Step 2: clustering modeling is carried out on a training sample set by utilizing an improved ELM classification algorithm, and nodes R (X, C) of hidden layers of each state are obtained through trainingj) Cluster center C ofjThe clustering centers of the normal state, the light fault state, the heavy fault state and the failure state are respectively marked as C0,C1,C2,C3。
Calculating the membership degree of each clustering center, and respectively recording as U0,U1,U2,U3。
Wherein X ═ X1,x2,...,xn]TFor an input training sample set, selecting an RBF function as an activation function of a hidden layer node, and recording as:
the network output is:
wherein Y (X) represents the output of the network, ω, when the input is XjFor the weight from the jth node of the hidden layer to the output layer, X ═ X1,x2,...,xn]TAs input vectors to the network, CjCluster center of the jth hidden node of ELM, deltajAnd expanding the width of the jth hidden node of the RBF network. DeltajThe smaller the extent of the expansion of the radial basis function, the less likely it is to indicate such a failure. FIG. 1 is a modified ELM model.
Step 3: and carrying out multi-dimensional characteristic parameter extraction on the data acquired by the plurality of sensors on site in real time after weighting fusion. And taking a data sample obtained after dimensionality reduction through LDA as a test sample set.
Step 4: and calculating the membership degree of each clustering center by using the clustering model obtained in Step2 by using the improved ELM on the test sample set.
Step 5: and judging the fault type of which stage the fault of the high-pressure diaphragm pump belongs to according to the membership value of each clustering center calculated by Step4, and realizing the estimation of the health state of the high-pressure diaphragm pump.
The invention has the beneficial effects that:
1. the invention provides a method for predicting the running state of a high-pressure diaphragm pump by using an improved ELM to establish a clustering model for a training sample set, using an online collected sample as a test sample for testing, and realizing the prediction of the running state of the high-pressure diaphragm pump by the testing set according to the classification fault type of the training model.
2. According to the signal acquisition module provided by the invention, five sensors are adopted to realize wireless transmission of data on five generated signals through the wireless WIFI transmission module, the obtained data is subjected to weighted data fusion, a performance degradation index capable of comprehensively reflecting fault information of the high-pressure diaphragm pump is fitted, and the life health state of the high-pressure diaphragm pump can be effectively estimated by utilizing the index, so that faults can be found in advance, and accidents are prevented.
Drawings
FIG. 1 is a block diagram of an improved ELM of the present invention;
FIG. 2 is a schematic diagram of the system of the present invention;
FIG. 3 is a flow chart of the steps of the method of the present invention.
Detailed Description
The invention is further described with reference to the following drawings and detailed description.
As shown in fig. 2, a system for estimating the health state of a high-pressure diaphragm pump based on an ELM includes a signal acquisition module, a wireless WIFI transmission module and a PC end analysis module.
The signal acquisition module comprises acceleration sensor, temperature sensor, acoustic emission sensor, reveal flow sensor, pressure sensor and STM32f103 chip.
The wireless WIFI transmission module consists of an ESP8266 and a wireless router and is used for transmitting signals acquired by the multiple sensors to the PC end analysis module in a WIFI mode.
And the PC end analysis module is used for carrying out data weighting fusion on the data acquired by the multiple sensors to obtain a performance degradation index capable of comprehensively reflecting the fault information of the high-pressure diaphragm pump, and finally, the index is utilized to realize the prediction of the running state of the high-pressure diaphragm pump by adopting improved ELM.
The signal acquisition module is connected the vibration signal, temperature signal, acoustic emission signal, the volume of revealing signal and the pressure signal of the high pressure diaphragm pump that acceleration sensor, temperature sensor, acoustic emission sensor, reveal flow sensor and pressure sensor gathered respectively with the serial ports of singlechip STM33f103 chip and wireless WIFI transmission module ESP8266, realizes the conversion between serial ports and the WIFI to realize the wireless transmission of data. The wireless router uploads the data to the cloud end through the Internet, the cloud end downloads the data to the PC end analysis module through the Internet and the router, the PC end analysis module performs data fusion on the received data by adopting a weighted fusion algorithm to fit a performance degradation index capable of comprehensively reflecting fault information of the high-pressure diaphragm pump, and a health condition estimation model of the high-pressure diaphragm pump is established through the improved ELM so as to estimate the running state of the high-pressure diaphragm pump, so that the high-pressure diaphragm pump is judged to be in a normal state, a light fault state, a severe fault state or a failure state.
The acceleration sensor and the temperature sensor in the signal acquisition module are axially arranged on the high-pressure diaphragm pump, the acoustic emission sensor is radially arranged on the high-pressure diaphragm pump, the leakage flow sensor and the pressure sensor are vertically arranged on the high-pressure diaphragm pump, and three acquisition channels are adopted to acquire data in parallel. The leakage flow of the high-pressure diaphragm pump is a main characteristic parameter reflecting the abrasion condition of the pump, is related to the output pressure of the pump, the real-time discharge capacity of the pump, the temperature and the rotating speed, the influence of the factors is comprehensively considered, and 5 types of sensors are selected to respectively carry out signal acquisition on the running state of the high-pressure diaphragm pump.
The normal state, the light fault state, the severe fault state or the failure state of the high-pressure diaphragm pump can be judged by the abrasion degree of the diaphragm pump, and the low fault state is defined when the abrasion degree is 0.1-5 mm. And when the abrasion degree is 5mm-10mm, the state is a severe fault state. The wear was in a failure state when the wear was more than 10 mm. When the data acquired in real time on line are analyzed, when the abrasion degree exceeds 5mm, the data are judged to be in a serious fault state. And further analyzing whether the state belongs to a failure state. When the fault belongs to a severe fault state or a failure state, the system interface can automatically alarm, and the system alarms to remind field workers to remove the fault in time, so that the normal operation of the industrial production equipment under the condition of the maximum utilization rate is ensured.
The PC terminal analysis module is a PC.
The system signal communication process of the invention: firstly, an acceleration sensor, a temperature sensor, an acoustic emission sensor, a leakage flow sensor, a pressure sensor, an STM32f103 microcontroller, a wireless WIFI transmission module ESP8266 and a PC are respectively connected to corresponding sockets, and then a network mode of the WIFI wireless transmission module, a server IP address and a server port are configured. The STM32f103 microcontroller is connected with the wireless WIFI transmission module in a TCP/IP mode, and the PC starts a wireless hotspot and waits for the WIFI module to connect the hotspot. The system sends data to a PC through TCP/IP, and receives the collected vibration signal, temperature signal, acoustic emission signal, leakage flow signal and pressure signal data.
The method is executed by a PC, and when data analysis is carried out, the online acquired data is used as a test sample set and is used as the input of the ELM, and the health state of the high-pressure diaphragm pump is predicted through a trained model. When a fault occurs, selecting the data of the corresponding group to calculate the clustering center, calculating the value of the corresponding membership degree according to the value of each clustering center, judging the fault type of the fault according to the membership degree, judging the fault to be in a serious fault state or a failure state when the abrasion degree exceeds 5mm, automatically alarming by the system, and reminding field workers of timely troubleshooting by the system alarm, thereby ensuring that the industrial production equipment normally operates under the condition of the maximum utilization rate.
As shown in fig. 3, the method for using the system for estimating the health state of the high-pressure diaphragm pump based on the ELM specifically comprises the following steps:
step 1: collecting the full-period life data of the high-pressure diaphragm pump, extracting multi-dimensional characteristic parameters of the original data, and reducing the dimensions through LDA to obtain an initial data sample serving as a training sample set.
Step 2: and carrying out cluster modeling on the training set samples by using an improved ELM classification algorithm. The activation function of the improved ELM hidden layer node is selected as an RBF function and is marked as:
the network output is:
wherein X is ═ X1,x2,...,xn]TIs a training sample set of inputs, Y (X) represents the output of the network when the input is X, ωjIs the jthWeight of hidden layer node to output layer, CjCluster center of the jth hidden node of ELM, deltajFor the extension width, delta, of the jth hidden node of the RBF networkjThe smaller the extent of the expansion of the radial basis function, the less likely it is to indicate such a failure.
Fig. 1 shows an improved ELM model, and the improved ELM classification algorithm is as follows:
step2.1: data center C for determining hidden layer nodes of RBF neural networkj. Clustering training samples by using a K-means clustering algorithm, and calculating Euclidean distance d between the current input vector X of the RBF neural network and each neuronw：
Step2.2: taking the neuron with the minimum Euclidean distance as a data center of each hidden layer node of the RBF neural network, and marking as CjAnd (j is 1, 2, …, n), wherein n is the number of fault types of the high-pressure diaphragm pump.
Step2.3: calculating the expansion width delta of hidden nodes of RBF neural networkj. Calculating Euclidean distances from all sample points in each class to each fault class clustering center, taking the minimum value of the Euclidean distances of each sample point as the minimum distance from the sample point to the class to which the sample point belongs, wherein the minimum distance is the distance from all sample points in the class to the clustering center to which the sample point belongs, taking the maximum value of the distances as the radius of the clustering center of the fault class sample to which the sample point belongs, and the maximum value of the distances is the expansion width delta of the jth hidden layer node of the respective neuron RBF networkjAnd (j is 1, 2, …, n), wherein n is the number of fault types of the high-pressure diaphragm pump.
Step2.4: finally, the number n of clustering categories determined by a K-means clustering algorithm is used as the number of hidden nodes of the RBF network, namely the number of fault categories of the high-pressure diaphragm pump; clustering center C determined by each neuronjThe cluster center is used as a hidden node of the RBF neural network; will cluster radius deltajExpansion width delta serving as jth hidden layer node of RBF network of each neuronj(j＝1，2,…,n)。
Step2.5: and training the obtained training sample X of the high-pressure diaphragm pump as the input of the improved ELM classifier to obtain a training model.
Step 3: weighting and fusing signals acquired by five sensors on site in real time, fitting a performance degradation index data set capable of comprehensively reflecting fault information of the high-pressure diaphragm pump, and performing multi-dimensional feature extraction on the obtained data set; and taking a data sample obtained after dimensionality reduction through LDA as a test sample set.
Step 4: and training the test sample set by using the improved ELM through the clustering model obtained in Step2 to obtain the clustering centers of all the running states of the high-pressure diaphragm pump, and calculating the membership degrees of all the clustering centers.
Step 5: and judging the fault type of which stage the running state of the high-pressure diaphragm pump belongs to according to the membership value of each running state clustering center calculated by Step4, thereby realizing the estimation of the health state of the high-pressure diaphragm pump.
While the present invention has been described in detail with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, and various changes can be made without departing from the spirit and scope of the present invention.
Claims (2)
1. The utility model provides a high-pressure diaphragm pump health state pre-estimation system based on ELM which characterized in that: the wireless WIFI transmission system comprises a signal acquisition module, a wireless WIFI transmission module and a PC (personal computer) end analysis module;
the signal acquisition module consists of an acceleration sensor, a temperature sensor, an acoustic emission sensor, a leakage flow sensor, a pressure sensor and an STM32f103 chip;
the wireless WIFI transmission module consists of an ESP8266 and a wireless router;
the signal acquisition module is with acceleration sensor, a temperature sensor, the acoustic emission sensor, reveal the vibration signal of the high-pressure diaphragm pump that flow sensor and pressure sensor gathered respectively, the temperature signal, the acoustic emission signal, leakage quantity signal and pressure signal are connected with the serial ports of wireless WIFI transmission module ESP8266 by the serial ports of singlechip STM33f103 chip, upload to the high in the clouds again by wireless router through the Internet, pass through the Internet by the high in the clouds, the router is down to PC end analysis module on, PC end analysis module adopts the weighted fusion algorithm with the data received to carry out data fusion, fit out a performance degradation index that can comprehensively reflect high-pressure diaphragm pump fault information, and establish high-pressure diaphragm pump health status through modified ELM and predict the model, thereby predict the running state of high-pressure diaphragm pump.
2. The method of using an ELM-based high-pressure diaphragm pump state of health estimation system of claim 1, wherein:
step 1: collecting full-period life data of the high-pressure diaphragm pump, extracting multi-dimensional characteristic parameters of original data, and reducing dimensions through LDA (laser direct absorption) to obtain an initial data sample serving as a training sample set;
step 2: clustering modeling is carried out on a training sample set by utilizing an improved ELM classification algorithm, and nodes R (X, C) of hidden layers of each state are obtained through trainingj) Cluster center C ofjThe clustering centers of the normal state, the light fault state, the heavy fault state and the failure state are respectively marked as C0,C1,C2,C3；
Calculating the membership degree of each clustering center, and respectively recording as U0,U1,U2,U3；
Wherein X ═ X1,x2,...,xn]TFor an input training sample set, selecting an RBF function as an activation function of a hidden layer node, and recording as:
the network output is:
wherein Y (X) represents an input ofX time, output of the network, omegajFor the weight from the jth node of the hidden layer to the output layer, X ═ X1,x2,...,xn]TAs input vectors to the network, CjCluster center of the jth hidden node of ELM, deltajExpanding the width of a jth hidden node of the RBF network;
step 3: carrying out multi-dimensional characteristic parameter extraction on data acquired and weighted and fused in real time by a plurality of sensors on site; taking a data sample obtained after LDA dimension reduction as a test sample set;
step 4: calculating the membership degree of each clustering center by using an improved ELM (element-free mass spectrometry) through a clustering model obtained in Step2 for the test sample set;
step 5: and judging the fault type of which stage the fault of the high-pressure diaphragm pump belongs to according to the membership value of each clustering center calculated by Step4, and realizing the estimation of the health state of the high-pressure diaphragm pump.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010984240.5A CN112196784B (en) | 2020-09-18 | 2020-09-18 | ELM-based high-pressure diaphragm pump health state estimation system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010984240.5A CN112196784B (en) | 2020-09-18 | 2020-09-18 | ELM-based high-pressure diaphragm pump health state estimation system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112196784A true CN112196784A (en) | 2021-01-08 |
| CN112196784B CN112196784B (en) | 2021-11-26 |
Family
ID=74015465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010984240.5A Active CN112196784B (en) | 2020-09-18 | 2020-09-18 | ELM-based high-pressure diaphragm pump health state estimation system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112196784B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113074940A (en) * | 2021-03-18 | 2021-07-06 | 昆明理工大学 | Rolling bearing health state estimation system and method based on OS-ELM |
| CN114882680A (en) * | 2022-04-19 | 2022-08-09 | 深圳闪回科技有限公司 | Remote terminal management system and method |
| CN115095535A (en) * | 2022-06-17 | 2022-09-23 | 长沙昌佳自动化设备有限公司 | Industrial pump operation multi-parameter detector |
| CN116906306A (en) * | 2023-07-05 | 2023-10-20 | 山东亿宁环保科技有限公司 | Vacuum pump integrated control system based on convolutional neural network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150152861A1 (en) * | 2009-06-11 | 2015-06-04 | Eaton Corporation | Fault detection and mitigation in hybrid drive system |
| CN106989004A (en) * | 2017-04-14 | 2017-07-28 | 北京理工大学 | A kind of electro-hydraulic pump diagnostic system and method |
| CN107605721A (en) * | 2017-07-24 | 2018-01-19 | 昆明理工大学 | A kind of ore slurry pipeline high-pressure diaphragm pump health status Prediction System and method |
| CN108393744A (en) * | 2018-04-11 | 2018-08-14 | 温州大学 | A kind of more sensor monitoring methods of cutting tool state |
| CN111220373A (en) * | 2020-03-18 | 2020-06-02 | 北京工业大学 | Method for diagnosing faults of centrifugal pump rotor system |
-
2020
- 2020-09-18 CN CN202010984240.5A patent/CN112196784B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150152861A1 (en) * | 2009-06-11 | 2015-06-04 | Eaton Corporation | Fault detection and mitigation in hybrid drive system |
| CN106989004A (en) * | 2017-04-14 | 2017-07-28 | 北京理工大学 | A kind of electro-hydraulic pump diagnostic system and method |
| CN107605721A (en) * | 2017-07-24 | 2018-01-19 | 昆明理工大学 | A kind of ore slurry pipeline high-pressure diaphragm pump health status Prediction System and method |
| CN108393744A (en) * | 2018-04-11 | 2018-08-14 | 温州大学 | A kind of more sensor monitoring methods of cutting tool state |
| CN111220373A (en) * | 2020-03-18 | 2020-06-02 | 北京工业大学 | Method for diagnosing faults of centrifugal pump rotor system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113074940A (en) * | 2021-03-18 | 2021-07-06 | 昆明理工大学 | Rolling bearing health state estimation system and method based on OS-ELM |
| CN114882680A (en) * | 2022-04-19 | 2022-08-09 | 深圳闪回科技有限公司 | Remote terminal management system and method |
| CN115095535A (en) * | 2022-06-17 | 2022-09-23 | 长沙昌佳自动化设备有限公司 | Industrial pump operation multi-parameter detector |
| CN116906306A (en) * | 2023-07-05 | 2023-10-20 | 山东亿宁环保科技有限公司 | Vacuum pump integrated control system based on convolutional neural network |
| CN116906306B (en) * | 2023-07-05 | 2024-02-13 | 山东亿宁环保科技有限公司 | Vacuum pump integrated control system based on convolutional neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112196784B (en) | 2021-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112196784B (en) | ELM-based high-pressure diaphragm pump health state estimation system and method | |
| CN111596604B (en) | Intelligent fault diagnosis and self-healing control system and method for engineering equipment based on digital twinning | |
| CN109465676B (en) | Tool life prediction method | |
| Fan et al. | A framework for knowledge discovery in massive building automation data and its application in building diagnostics | |
| CN113762329A (en) | Method and system for constructing state prediction model of large rolling mill | |
| CN107403279B (en) | Oil transfer pump working condition self-adaptive state early warning system and method | |
| CN110942137A (en) | Power grid information operation and maintenance monitoring method based on deep learning | |
| CN105574604B (en) | Monitoring pre-judging analysis system for power grid operation event | |
| CN106656669B (en) | A kind of device parameter abnormality detection system and method based on threshold adaptive setting | |
| CN109711664B (en) | Power transmission and transformation equipment health assessment system based on big data | |
| CN112734977B (en) | Equipment risk early warning system and algorithm based on Internet of things | |
| CN117156442B (en) | Cloud data security protection method and system based on 5G network | |
| CN114266289A (en) | Complex equipment health state assessment method | |
| WO2023044770A1 (en) | Dry pump downtime early warning method and apparatus, electronic device, storage medium, and program | |
| CN116538092B (en) | Compressor on-line monitoring and diagnosing method, device, equipment and storage medium | |
| CN116599857B (en) | Digital twin application system suitable for multiple scenes of Internet of things | |
| CN117421994A (en) | Edge application health monitoring method and system | |
| CN117076869A (en) | Time-frequency domain fusion fault diagnosis method and system for rotary machine | |
| CN115037559B (en) | Data safety monitoring system based on flow, electronic equipment and storage medium | |
| CN117675230A (en) | Knowledge-graph-based oil well data integrity identification method | |
| CN113253113A (en) | Method for predicting consistency fault of power battery | |
| Banerjee et al. | Application of IoT in industry 4.0 for predictive analytics | |
| Chen et al. | A multimode anomaly detection method based on oc-elm for aircraft engine system | |
| Zhao et al. | Deep Learning Anomaly Detection Based on Hierarchical Status-Connection Features in Networked Control Systems. | |
| CN117520993B (en) | Electric drive axle test system and method thereof |
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 |