CN115358265B - Method for detecting faults of ultra-low head water lifting system - Google Patents
Method for detecting faults of ultra-low head water lifting system Download PDFInfo
- Publication number
- CN115358265B CN115358265B CN202210979108.4A CN202210979108A CN115358265B CN 115358265 B CN115358265 B CN 115358265B CN 202210979108 A CN202210979108 A CN 202210979108A CN 115358265 B CN115358265 B CN 115358265B
- Authority
- CN
- China
- Prior art keywords
- firefly
- neural network
- lifting system
- adaptive
- self
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 19
- 241000254158 Lampyridae Species 0.000 claims abstract description 75
- 238000013528 artificial neural network Methods 0.000 claims abstract description 40
- 238000003745 diagnosis Methods 0.000 claims abstract description 24
- 238000010845 search algorithm Methods 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 6
- 238000010521 absorption reaction Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims 1
- 238000003062 neural network model Methods 0.000 abstract 2
- 238000002922 simulated annealing Methods 0.000 abstract 1
- 238000001514 detection method Methods 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a method for detecting faults of an ultra-low head water lifting system, which is characterized by comprising the following steps of: the method comprises the following steps: extracting feature numbers of different fault signals of the water lifting system, and establishing a fault set; secondly, the light intensity change and the attractive force of the firefly algorithm are introduced into the optimizing process of the optimal solution, and proper control parameter values are adaptively selected according to the quality of the solution. Meanwhile, the convergence accuracy of the algorithm is further improved by using a simulated annealing algorithm. Thirdly, searching an optimal weight threshold parameter of the BP neural network by adopting a self-adaptive firefly algorithm, and then establishing an ultra-low head water lifting system fault diagnosis model; compared with a neural network model and a firefly search neural network model, the method can remarkably improve the efficiency and accuracy of fault positioning of the ultra-low water head water lifting system.
Description
Technical Field
The invention relates to the technical field of fault detection, in particular to a method for detecting faults of an ultra-low head water lifting system.
Background
The natural energy water lifting system utilizes the water flow fall to generate high-pressure air, and water is pressed to a high place through the high-pressure air. During the operation of the system, faults can be generated due to the influence of internal and external factors, and different vibration signals can be formed at the conveying pipeline by various faults. The difference between various vibration signals is small, and the vibration signals need to be effectively analyzed so as to accurately judge different faults.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a method for detecting faults of an ultra-low head water lifting system.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a method for detecting faults of an ultra-low head water lifting system comprises the following steps:
step 1, collecting vibration signals corresponding to a plurality of fault modes of an ultra-low head water lifting system, and encoding the fault modes to be used as output samples;
Step 2, extracting characteristic values corresponding to all vibration signals, and dividing the characteristic values into two groups by adopting a random selection mode, wherein the characteristic values are respectively used as a training set and a test set of the BP neural network;
Step 3, determining the topological structure of the BP neural network according to the output error minimum principle of the BP neural network;
Step 4, encoding weight threshold parameters of the BP neural network into solution vectors of the self-adaptive firefly algorithm, embedding the light intensity and the attractive force of the self-adaptive firefly search algorithm into the solution process, obtaining an optimal solution of the self-adaptive firefly search algorithm through searching, and taking the optimal solution as the optimal weight threshold parameters of the BP neural network;
Step 5, endowing the optimal weight threshold value parameter obtained in the step 4 to the BP neural network, and carrying out learning training on the BP neural network with the optimal weight threshold value parameter by adopting the training set obtained in the step 2 to obtain a diagnosis model of the self-adaptive firefly search neural network;
step 6, performing fault diagnosis on the test set obtained in the step 2 by using a self-adaptive firefly search neural network diagnosis model, outputting a diagnosis result, and comparing the diagnosis result with the output sample obtained in the step 1;
And 7, collecting vibration signals of the water lifting system every 30s to 60s, and judging the fault type of the water lifting system according to the self-adaptive firefly search neural network diagnosis model.
Preferably, the step 4 specifically includes the following steps:
Step 4.1, initializing parameters of a self-adaptive firefly search algorithm;
The basic parameters of the self-adaptive firefly search algorithm comprise the number n of fireflies, the maximum attraction degree beta max, the light intensity absorption coefficient gamma, the step factor alpha and the maximum iteration frequency Tmax;
Step 4.2, randomly selecting weight threshold parameters of a fault diagnosis model of the BP neural network, and encoding the weight threshold parameters of the BP neural network into solution vectors of a self-adaptive firefly search algorithm;
Step 4.3, initializing the problem, namely converting an individual into fireflies, setting a brightness function, and initializing all parameters beta max, gamma and alpha;
Step 4.4, calculating the light intensity value of each firefly according to the position of the firefly, wherein the higher the light intensity value is, the higher the firefly brightness is;
Wherein the light intensity value is calculated according to the following rule:
step 4.5, determining the distance between every two fireflies: the Cartesian distance of any two fireflies i and j in space coordinates is:
step 4.6, calculating the attraction degree of surrounding fireflies: since the attractive force of fireflies is proportional to the light intensity seen by neighboring fireflies, the calculation is performed by the following rule:
where β 0 is the attractive force at r=0;
step 4.7, for each firefly, finding out the firefly individual with the highest attraction degree, and updating the moving position; when one firefly is attracted by another brighter firefly, the law of motion is as follows:
Where α is the step factor and rand is a random number generator uniformly distributed in [0,1 ].
And 4.8, recalculating the light intensity of each firefly, and ending if iteration is completed or the intensity reaches the requirement.
And 4.9, taking the optimal position as an optimal weight threshold parameter of the BP neural network.
Preferably, in step 2, the characteristic value selects the amplitude of the spectral component of the vibration signal as follows: <0.5f 0、f0、2f0、3f0、>3f0, where f 0 is the fundamental frequency.
Preferably, in step 3, the output error E is:
Where N is the number of samples in the training set, zi is the actual output value of the network for the ith sample, and O i is the expected output value for the ith sample.
Compared with the prior art, the invention has the beneficial effects that: the collected vibration signals are analyzed and calculated through establishing a model, and the vibration signals are corresponding to the fault modes, so that the fault detection efficiency and the accuracy of the water lifting system can be remarkably improved.
Detailed Description
Embodiments of the present invention are described in detail below.
A method for detecting faults of an ultra-low head water lifting system comprises the following steps:
Step 1, collecting vibration signals corresponding to a plurality of fault modes of an ultra-low head water lifting system, and encoding the fault modes to be used as output samples, wherein the fault modes comprise leakage of an air inlet pipeline, abnormal vibration of a gas collecting pipe pipeline, leakage of the gas collecting pipe and the like, and meanwhile, collecting vibration signals without faults;
Step 2, extracting the characteristic values corresponding to the vibration signals collected in the step 1, and dividing the characteristic values into two groups by adopting a random selection mode to respectively serve as a training set and a test set of the BP neural network;
Step 3, determining the topological structure of the BP neural network according to the output error minimum principle of the BP neural network;
Step 4, encoding weight threshold parameters of the BP neural network into solution vectors of the self-adaptive firefly algorithm, embedding the light intensity and the attractive force of the self-adaptive firefly search algorithm into the solution process, obtaining an optimal solution of the self-adaptive firefly search algorithm through searching, and taking the optimal solution as the optimal weight threshold parameters of the BP neural network;
The specific steps are 4.1, initializing parameters of a self-adaptive firefly search algorithm;
The basic parameters of the self-adaptive firefly search algorithm comprise the number n of fireflies, the maximum attraction degree beta max, the light intensity absorption coefficient gamma, the step factor alpha and the maximum iteration frequency Tmax;
4.2, randomly selecting weight threshold parameters of a fault diagnosis model of the BP neural network, and encoding the weight threshold parameters of the BP neural network into a solution vector of a self-adaptive firefly search algorithm;
4.3, initializing the problem, converting an individual into fireflies, setting a brightness function, and initializing all parameters beta max, gamma and alpha;
4.4, calculating the light intensity value of each firefly according to the position of the firefly, wherein the higher the light intensity value is, the higher the firefly brightness is;
Wherein the light intensity value is calculated according to the following rule:
4.5, determining the distance between every two fireflies: the Cartesian distance of any two fireflies i and j in space coordinates is:
4.6, calculating the attraction degree of surrounding fireflies: since the attractive force of fireflies is proportional to the light intensity seen by neighboring fireflies, the calculation is performed by the following rule:
where β 0 is the attractive force at r=0;
4.7, for each firefly, finding out the firefly individual with the highest attraction degree, and updating the moving position; when one firefly is attracted by another brighter firefly, the law of motion is as follows:
Where α is the step factor and rand is a random number generator uniformly distributed in [0,1 ].
4.8, Recalculating the light intensity of each firefly, and ending if iteration is completed or the intensity reaches the requirement.
And 4.9, taking the optimal position as an optimal weight threshold parameter of the BP neural network.
And 5, endowing the BP neural network with the optimal weight threshold parameter obtained in the step 4, and performing learning training on the BP neural network with the optimal weight threshold parameter by adopting the training set obtained in the step 2 to obtain a diagnosis model of the self-adaptive firefly search neural network.
And 6, performing fault diagnosis on the test set obtained in the step 2 by using the self-adaptive firefly search neural network diagnosis model, outputting a diagnosis result, and comparing the diagnosis result with the output sample obtained in the step 1 to form a detection standard.
And 7, collecting vibration signals of the water lifting system every 30s to 120s, or collecting vibration signals of the water lifting system every 60s, calculating a diagnosis result through a self-adaptive firefly search neural network diagnosis model, and comparing the diagnosis result with the detection standard in the step 6 to judge the fault type in the water lifting system.
Preferably, in step 2, the characteristic value selects the amplitude of the spectral component of the vibration signal as follows: <0.5f 0、f0、2f0、3f0、>3f0, where f 0 is the fundamental frequency.
In step 3, the output error E is:
Where N is the number of samples in the training set, zi is the actual output value of the network for the ith sample, and O i is the expected output value for the ith sample.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the present invention may occur to one skilled in the art without departing from the principles of the present invention and are intended to be within the scope of the present invention.
Claims (3)
1. A method for detecting faults of an ultra-low head water lifting system is characterized by comprising the following steps of: the method comprises the following steps:
step 1, collecting vibration signals corresponding to a plurality of fault modes of an ultra-low head water lifting system, and encoding the fault modes to be used as output samples;
Step 2, extracting characteristic values corresponding to all vibration signals, and dividing the characteristic values into two groups by adopting a random selection mode, wherein the characteristic values are respectively used as a training set and a test set of the BP neural network;
Step 3, determining the topological structure of the BP neural network according to the output error minimum principle of the BP neural network;
Step 4, encoding weight threshold parameters of the BP neural network into solution vectors of the self-adaptive firefly algorithm, embedding the light intensity and the attractive force of the self-adaptive firefly search algorithm into the solution process, obtaining an optimal solution of the self-adaptive firefly search algorithm through searching, and taking the optimal solution as the optimal weight threshold parameters of the BP neural network;
the step 4 specifically comprises the following steps:
Step 4.1, initializing parameters of a self-adaptive firefly search algorithm;
The basic parameters of the self-adaptive firefly search algorithm comprise the number n of fireflies, the maximum attraction degree beta max, the light intensity absorption coefficient gamma, the step factor alpha and the maximum iteration frequency Tmax;
Step 4.2, randomly selecting weight threshold parameters of a fault diagnosis model of the BP neural network, and encoding the weight threshold parameters of the BP neural network into solution vectors of a self-adaptive firefly search algorithm;
Step 4.3, initializing the problem, namely converting an individual into fireflies, setting a brightness function, and initializing all parameters beta max, gamma and alpha;
Step 4.4, calculating the light intensity value of each firefly according to the position of the firefly, wherein the higher the light intensity value is, the higher the firefly brightness is;
Wherein the light intensity value is calculated according to the following rule:
step 4.5, determining the distance between every two fireflies: the Cartesian distance of any two fireflies i and j in space coordinates is:
step 4.6, calculating the attraction degree of surrounding fireflies: since the attractive force of fireflies is proportional to the light intensity seen by neighboring fireflies, the calculation is performed by the following rule:
where β 0 is the attractive force at r=0;
step 4.7, for each firefly, finding out the firefly individual with the highest attraction degree, and updating the moving position; when one firefly is attracted by another brighter firefly, the law of motion is as follows:
where α is a step factor and rand is a random number generator uniformly distributed in [0,1 ];
Step 4.8, recalculating the light intensity of each firefly, and ending if iteration is completed or the intensity reaches the requirement;
step 4.9, taking the optimal position as an optimal weight threshold parameter of the BP neural network;
Step 5, endowing the optimal weight threshold value parameter obtained in the step 4 to the BP neural network, and carrying out learning training on the BP neural network with the optimal weight threshold value parameter by adopting the training set obtained in the step 2 to obtain a diagnosis model of the self-adaptive firefly search neural network;
step 6, performing fault diagnosis on the test set obtained in the step 2 by using a self-adaptive firefly search neural network diagnosis model, outputting a diagnosis result, and comparing the diagnosis result with the output sample obtained in the step 1;
And 7, collecting vibration signals of the water lifting system every 30s to 60s, and judging the fault type of the water lifting system according to the self-adaptive firefly search neural network diagnosis model.
2. The method for detecting the fault of the ultra-low head water lifting system according to claim 1, wherein the method comprises the following steps: in step 2, the characteristic value selects the amplitude of the frequency spectrum component of the vibration signal as follows: <0.5f 0、f0、2f0、3f0、>3f0, where f 0 is the fundamental frequency.
3. The method for detecting the fault of the ultra-low head water lifting system according to claim 1, wherein the method comprises the following steps: in step 3, the output error E is:
Where N is the number of samples in the training set, zi is the actual output value of the network for the ith sample, and O i is the expected output value for the ith sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210979108.4A CN115358265B (en) | 2022-08-16 | 2022-08-16 | Method for detecting faults of ultra-low head water lifting system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210979108.4A CN115358265B (en) | 2022-08-16 | 2022-08-16 | Method for detecting faults of ultra-low head water lifting system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115358265A CN115358265A (en) | 2022-11-18 |
| CN115358265B true CN115358265B (en) | 2024-04-30 |
Family
ID=84033037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210979108.4A Active CN115358265B (en) | 2022-08-16 | 2022-08-16 | Method for detecting faults of ultra-low head water lifting system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115358265B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108596212A (en) * | 2018-03-29 | 2018-09-28 | 红河学院 | Based on the Diagnosis Method of Transformer Faults for improving cuckoo chess game optimization neural network |
| CN109444740A (en) * | 2018-11-14 | 2019-03-08 | 湖南大学 | A kind of the malfunction intellectual monitoring and diagnostic method of Wind turbines |
| CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
| CN110307981A (en) * | 2019-06-17 | 2019-10-08 | 西安工程大学 | Method for Bearing Fault Diagnosis based on PNN-IFA |
| CN110363277A (en) * | 2019-07-15 | 2019-10-22 | 南京工业大学 | Method for diagnosing fault of power transformer and system based on improved glowworm swarm algorithm optimization probabilistic neural network |
| CN114528883A (en) * | 2022-02-24 | 2022-05-24 | 河北工业大学 | Wind power converter IGBT fault identification method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116348158A (en) * | 2020-11-09 | 2023-06-27 | 海德鲁西昂公司 | System, apparatus and method for pump motor fault protection |
-
2022
- 2022-08-16 CN CN202210979108.4A patent/CN115358265B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
| CN108596212A (en) * | 2018-03-29 | 2018-09-28 | 红河学院 | Based on the Diagnosis Method of Transformer Faults for improving cuckoo chess game optimization neural network |
| CN109444740A (en) * | 2018-11-14 | 2019-03-08 | 湖南大学 | A kind of the malfunction intellectual monitoring and diagnostic method of Wind turbines |
| CN110307981A (en) * | 2019-06-17 | 2019-10-08 | 西安工程大学 | Method for Bearing Fault Diagnosis based on PNN-IFA |
| CN110363277A (en) * | 2019-07-15 | 2019-10-22 | 南京工业大学 | Method for diagnosing fault of power transformer and system based on improved glowworm swarm algorithm optimization probabilistic neural network |
| CN114528883A (en) * | 2022-02-24 | 2022-05-24 | 河北工业大学 | Wind power converter IGBT fault identification method |
Non-Patent Citations (2)
| Title |
|---|
| Diagnostic method based on the analysis of the vibration and acoustic emission energy for emergency diesel generators in nuclear plants;Jorge Arroyo et al;《Applied Acoustics》;第74卷(第4期);502-508 * |
| 基于CEEMDAN和混合灰狼算法优化SVM的水电机组故障诊断方法;杨彤等;《水电能源科学》;第40卷(第3期);195-198 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115358265A (en) | 2022-11-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108596212B (en) | Transformer fault diagnosis method based on improved cuckoo search optimization neural network | |
| CN111507990B (en) | Tunnel surface defect segmentation method based on deep learning | |
| CN109035233B (en) | Visual attention network system and workpiece surface defect detection method | |
| CN110176001B (en) | Grad-CAM algorithm-based high-speed rail contact net insulator damage accurate positioning method | |
| CN113695713B (en) | On-line monitoring method and device for welding quality of water heater liner | |
| CN110985310B (en) | Wind driven generator blade fault monitoring method and device based on acoustic sensor array | |
| CN109325263A (en) | Truss bridge damage position neural network based and damage extent identification method | |
| CN111833310B (en) | Surface defect classification method based on neural network architecture search | |
| CN111914705A (en) | Signal generation method and device for improving health state evaluation accuracy of reactor | |
| CN111624522A (en) | Ant colony optimization-based RBF neural network control transformer fault diagnosis method | |
| CN113239991A (en) | Flame image oxygen concentration prediction method based on regression generation countermeasure network | |
| CN114444620A (en) | Indicator diagram fault diagnosis method based on generating type antagonistic neural network | |
| CN115358265B (en) | Method for detecting faults of ultra-low head water lifting system | |
| CN115222983A (en) | Cable damage detection method and system | |
| CN113763364B (en) | Image defect detection method based on convolutional neural network | |
| CN115163424A (en) | Wind turbine generator gearbox oil temperature fault detection method and system based on neural network | |
| CN117113066B (en) | Transmission line insulator defect detection method based on computer vision | |
| CN114839492A (en) | Method and device for identifying GIS partial discharge type based on MOBILE NETV3 | |
| CN110956112B (en) | Novel high-reliability slewing bearing service life assessment method | |
| CN116051496A (en) | Real-time sewer defect detection method | |
| CN115577744A (en) | Intelligent prediction method for oil and gas pipeline leakage fire risk | |
| CN113361209B (en) | Quantitative analysis method for magnetic anomaly of surface defects of high-temperature alloy | |
| CN114330141A (en) | Fan main shaft bearing service life prediction method based on GRU (generalized regression Unit) super-parameter optimization | |
| CN112183745A (en) | High-voltage cable partial discharge mode identification method based on particle swarm algorithm and DBN | |
| CN112161785A (en) | Ocean engineering structure micro-damage judgment method |
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 |