CN111914490B - Pump station unit state evaluation method based on depth convolution random forest self-coding - Google Patents
Pump station unit state evaluation method based on depth convolution random forest self-coding Download PDFInfo
- Publication number
- CN111914490B CN111914490B CN202010895695.XA CN202010895695A CN111914490B CN 111914490 B CN111914490 B CN 111914490B CN 202010895695 A CN202010895695 A CN 202010895695A CN 111914490 B CN111914490 B CN 111914490B
- Authority
- CN
- China
- Prior art keywords
- waveform
- vibration
- sample
- pump station
- coding
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- 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]
Abstract
The invention discloses a pump station unit state evaluation method based on depth convolution random forest self-coding, which comprises the steps of collecting key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large-axis swing degree waveform and all waveform data of a pump station unit in a whole life cycle, so as to obtain a historical waveform health sample library; selecting information of each measuring point in the healthy state period of the pump station unit from a historical waveform healthy sample library as a historical normal data sample; and the like. The advantages are that: based on-line monitoring data, the machine is not required to be stopped and the measuring points are not required to be manually arranged, so that the economic benefit of the power plant is increased, the multi-source data of the machine set is utilized for modeling, hidden features are automatically mined, and the machine set is suitable for the machine sets with different forms and running time.
Description
Technical Field
The invention relates to the technical field of pump station state evaluation, in particular to a pump station unit state evaluation method based on depth convolution random forest self-coding.
Background
Along with the continuous construction of the south-to-north water transfer project, the pump station unit is an important power device of the water transfer project, and the reliability of the pump station unit directly influences the safe and economic operation of the water transfer project. Meanwhile, because the water demand of the user side is continuously changeable, the adjusting frequency and the depth are greatly improved compared with the prior art, and the pump station unit works under severe conditions such as high temperature, high speed and the like and the influence of various random factors, and various faults of the unit are easy to occur. The performance of each component gradually deteriorates with the increase of the service time, resulting in the occurrence of a malignant accident. Therefore, real-time accurate mastering of the running state of the unit is urgently needed, and accurate and scientific decision basis is made for the running of the unit.
Aiming at the mechanical performance state evaluation of the pump station unit, china mostly adopts vibration measurement results according to key points of unit components to evaluate, and the health condition of the unit is analyzed through performance tests. Because the pump station unit is influenced by a plurality of factors such as water, machine, electricity and the like, a plurality of performance tests are adopted for state evaluation, and the running stability state of the pump station unit is represented by abnormality or exceeding of vibration and water pressure pulsation. However, conventional health assessment requires manual work such as sensor placement, data acquisition, and manual data processing analysis, and manual testing typically requires downtime or is scheduled for maintenance, subject to construction period constraints, and causes some economic loss. Meanwhile, corresponding judging standards for the operation health of pump station units are lacking at home and abroad at present, and are generally evaluated according to related regulations. Because the standard or the regulation often only prescribes a certain limit value, the unit types can be various in the actual operation process of the power station, and meanwhile, the health state of the unit can not be accurately and reasonably measured by the evaluation standard under a single limit value because the operation condition of the unit changes frequently. Meanwhile, the performance degradation of the unit can only be reflected in the trend change of the performance index along with time, the condition that the monitoring value exceeds the limit value does not occur yet, but the unit operation is obviously abnormal at the moment. Therefore, the method for evaluating the limit value under the current single working condition point is simply adopted, so that the health state of the running state of the unit cannot be effectively evaluated.
With the continuous development of industrial internet technology, real-time online evaluation and residual life estimation of a unit are possible through combination of multi-source online information and a data driving model. However, conventional life estimation algorithms often require the assumption of failure curves, such as exponential, linear, and extrapolation of health indicators by means of a hypothetical model to predict unit residual life. Because the pump station unit is frequently regulated, the assumption is often far from the actual situation, on one hand, the failure curve often does not follow a fixed shape only, and is difficult to express by a simple trend model; on the other hand, the time to failure tends not to be the same. Meanwhile, the defects of sensor noise, small sample number and the like are accompanied, so that the sensor is difficult to use in an actual system.
Disclosure of Invention
The invention aims to provide a pump station unit state evaluation method based on depth convolution random forest self-coding, so as to solve the problems in the prior art.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a pump station unit state evaluation method based on depth convolution random forest self-coding comprises the following steps,
s1, collecting key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large axis swing degree waveform and waveform data of the pump station unit in a whole life cycle so as to obtain a historical waveform health sample library;
s2, selecting information of each measuring point in the pump station unit health state period from the historical waveform health sample library as a historical normal data sample;
s3, selecting whole period data of each waveform according to the key phase signal from the historical normal data sample, and forming a two-dimensional matrix with a horizontal axis as a time point and a vertical axis as a channel number according to the corresponding time scale;
s4, extracting features of the two-dimensional matrix through a deep convolutional neural network to obtain sample features;
s5, inputting sample features into a random forest self-coding model for further compression to obtain a health feature sample set;
s6, monitoring a pump station unit on line, and acquiring key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large-axis swing degree waveform and waveform data of the pump station unit in the whole period in real time to obtain a waveform sample to be tested;
s7, executing the step S2 to the step S5 on the waveform sample to be tested to obtain a characteristic sample set to be tested;
s8, performing Euclidean distance calculation on the feature sample set to be tested and the health feature sample set to obtain a distance deviation degree;
s9, evaluating the health state of the pump station unit by using the distance deviation degree.
Preferably, the deep convolutional neural network comprises a forward convolutional model and a backward convolutional model, wherein the forward convolutional model adopts 5 convolutional-pooling layers, and convolution kernels are 128×3, 64×3, 32×3, 16×3 and 8*3 in sequence; the deconvolution model adopts 5 convolution-pooling layers, and convolution kernels are 8*3, 16×3, 32×3, 64×3 and 128×3 in sequence; and the forward convolution model and the deconvolution model perform feature extraction on the two-dimensional matrix to obtain sample features.
Preferably, the random forest self-coding model adopts a tree integration model of a random forest to construct a self-coding process; the random forest self-coding model comprises forward coding and reverse coding; after the forward code receives the sample feature, the sample feature is sent to each root node of the tree in the integration, and after the sample feature traverses the root nodes of all the trees, a first result vector is returned; the reverse coding utilizes the characteristic that each leaf node in the tree corresponds to only one path from the root to the leaf, a rule base is established to acquire each path from the leaf node to the root node of the tree, a plurality of rules in the rule base are simplified according to the logic that the range of each constraint in the rules can not be expanded any more, and finally, the mapping between each leaf node to the root node of all the trees is acquired, namely, a second result vector mapped with the first result vector is acquired, and the second result vector is a health feature sample set.
The beneficial effects of the invention are as follows: 1. the failure trend of the unit is obtained by analyzing a plurality of full life cycle examples from the beginning to the failure of the unit, and when a new failure trend of the unit occurs, the most probable trend change can be calculated through a model and the residual life is estimated. 2. According to the invention, inherent fault characteristics of the unit can be automatically mined through massive data, the health state of the unit is represented by modeling the data of the normal behavior of the unit and abstracting the implicit characteristics of the extracted data step by step, after the model is built, the health degree of the unit can be measured by calculating the similarity between the current sensor data and the normal behavior state, and more accurate residual life is obtained through an extrapolation algorithm. 3. The state evaluation method is based on-line monitoring data, does not need to stop and manually arrange measuring points, thereby greatly increasing the economic benefit of a power plant, simultaneously can comprehensively utilize multi-source data of the unit to model, automatically excavate hidden characteristics of the unit, does not need to be manually judged by an expert, is suitable for units in different forms and different running times, and has self-adaption and objectivity.
Drawings
FIG. 1 is a schematic flow chart of an evaluation method in an embodiment of the invention;
fig. 2 is a backward decoding map of a tree node in an embodiment of the invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the detailed description is presented by way of example only and is not intended to limit the invention.
Example 1
As shown in fig. 1, in this embodiment, a method for evaluating a state of a pump station set based on a depth convolution random forest self-coding is provided, including the following steps,
s1, collecting key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large axis swing degree waveform and waveform data of the pump station unit in a whole life cycle so as to obtain a historical waveform health sample library;
s2, selecting information of each measuring point in the pump station unit health state period from the historical waveform health sample library as a historical normal data sample;
s3, selecting whole period data of each waveform according to the key phase signal from the historical normal data sample, and forming a two-dimensional matrix with a horizontal axis as a time point and a vertical axis as a channel number according to the corresponding time scale;
s4, extracting features of the two-dimensional matrix through a deep convolutional neural network to obtain sample features;
s5, inputting sample features into a random forest self-coding model for further compression to obtain a health feature sample set;
s6, monitoring a pump station unit on line, and acquiring key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large-axis swing degree waveform and waveform data of the pump station unit in the whole period in real time to obtain a waveform sample to be tested;
s7, executing the step S2 to the step S5 on the waveform sample to be tested to obtain a characteristic sample set to be tested;
s8, performing Euclidean distance calculation on the feature sample set to be tested and the health feature sample set to obtain a distance deviation degree;
s9, evaluating the health state of the pump station unit by using the distance deviation degree.
In this embodiment, in the two-dimensional matrix, the horizontal axis includes 4096 time points; the vertical axis includes 10 channel numbers.
In this embodiment, the deep convolutional neural network includes a forward convolutional model and a backward convolutional model, where the forward convolutional model adopts 5 convolutional-pooling layers, and convolution kernels are 128×3, 64×3, 32×3, 16×3 and 8*3 in sequence; the deconvolution model adopts 5 convolution-pooling layers, and convolution kernels are 8*3, 16×3, 32×3, 64×3 and 128×3 in sequence; and the forward convolution model and the deconvolution model perform feature extraction on the two-dimensional matrix to obtain sample features.
In this embodiment, a 5-layer convolution-pooling layer is used for construction, and the convolution kernels of the positive convolution models are (128×3), (64×3), (32×3), (16×3), and (8*3), respectively. The deconvolution model also requires deconvolution according to deconvolution kernels of (8*3), (16×3), (32×3), (64×3), (128×3). Modeling training is carried out between forward convolution and deconvolution by adopting a random forest self-coding model.
And performing a random forest de-coding process on the extracted characteristic samples to obtain root node sequences, taking the root node sequences as deconvolution inputs, and performing deconvolution according to deconvolution kernels of (8*3), (16 x 3), (32 x 3), (64 x 3) and (128 x 3) to obtain deconvolution output.
And calculating errors between the original data (positive convolution input) and the deconvolution output, and performing repeated loop iteration training to obtain the minimum errors, wherein the minimum errors indicate that the neural network model and the random forest self-coding model are successfully trained, and the neural network model and the random forest self-coding model can be used for testing samples to be tested.
In this embodiment, 1024-dimensional sample features are generated after the convolution-pooling operation.
In this embodiment, the sample features are input into a random forest self-coding model for further compression, and 100-dimensional features are obtained, namely a health feature sample set.
In this embodiment, the random forest self-coding model is a tree integration model of a random forest to construct a self-coding process; the random forest self-coding model comprises forward coding and reverse coding; after the forward code receives the sample feature, the sample feature is sent to each root node of the tree in the integration, and after the sample feature traverses the root nodes of all the trees, a first result vector is returned; the reverse coding utilizes the characteristic that each leaf node in the tree corresponds to only one path from the root to the leaf, a rule base is established to acquire each path from the leaf node to the root node of the tree, a plurality of rules in the rule base are simplified according to the logic that the range of each constraint in the rules can not be expanded any more, and finally, the mapping between each leaf node to the root node of all the trees is acquired, namely, a second result vector mapped with the first result vector is acquired, and the second result vector is a health feature sample set.
In this embodiment, the Euclidean distance calculation is performed between the feature sample set to be tested and the health feature sample set to obtain a distance deviation degree, which is specifically calculated as follows,
wherein d 12 To be treatedTesting the degree of deviation of the distance between the feature sample set and the health feature sample set; x is x 1 And y 1 Features representative of a healthy sample, x 2 And y 2 Is a sample feature to be tested.
In the embodiment, the distance deviation degree is used as a state evaluation index of the pump station unit, and the specific evaluation mode is that the built-in feature dimension is set to be 50 dimensions, and the 50-dimensional feature vector, namely the health feature sample set, is obtained by taking the health sample of the historical pump station unit into the model; when the online data of the pump station unit is acquired, the calculation is also carried out, the feature sample set to be tested is acquired, the deviation degree calculation and normalization are carried out on the feature sample set to be tested and the health feature sample set, the result value of 1 represents complete failure, the result value of 0 represents health, the comprehensive health degree of the pump station unit can be estimated through the change of the result value from 0 to 1, and particularly, when the health degree shows trend rising, the occurrence of a certain fault is indicated.
Example two
In this embodiment, an axial pump station set of a certain river guiding, jihuai engineering is taken as an example, and the implementation process of the evaluation method of the present invention is specifically described.
(1) Firstly, the rated rotation speed and the structural form of the unit are studied, an implementation object adopts two-guide deployment (an upper guide bearing and a water guide bearing), the implementation object is supported by only adopting an upper frame, and meanwhile, the rotation speed of the unit is 1200r/min, and the calculated main frequency is about 20Hz, so that the speed type sensor is adopted for acquisition. The sensor is arranged on the pump station set as shown in the following table:
(2) The waveforms are sampled in equal period, and the waveforms of all parts are collected and stored by adopting a sampling rate of more than 3000Hz according to the fragrance law.
After the online waveforms are stored, expert experience is combined, historical waveforms are subjected to expert marking, normal samples are collected to form a historical waveform health sample library, waveform data of a positive period are captured through the 0-1 state of key phase signals, and the waveforms are aligned.
(3) And selecting the whole period data of each waveform according to the key phase signals, and forming a two-dimensional matrix with the horizontal axis as 4096 time points (based on the actual sampling rate) and the vertical axis as the channel number (10) according to the corresponding time scale.
(4) The construction was performed using 5 convolution-pooling layers, with convolution kernels of (128 x 3), (64 x 3), (32 x 3), (16 x 3), respectively (8*3), and the resulting outputs were unwrapped to generate features. The deconvolution model also requires deconvolution according to deconvolution kernels of (8*3), (16×3), (32×3) (64×3), (128×3). Modeling training is performed between convolution and deconvolution by adopting random forest self-coding.
(5) And extracting features through a deep convolutional neural network, and generating 1024-dimensional sample features after convolutional-pooling operation. And (3) bringing the sample features into a random forest self-coding model, and further compressing to obtain 100-dimensional features serving as a health feature sample set.
The random forest self-coding adopts a tree integration model of the random forest to construct a self-coding process, the self-coding process is divided into a forward coding part and a backward coding part, the forward coding process receives input data and then sends the data to each root node of the tree in the integration, and when the data traverses leaf nodes of all the trees, the process returns a result vector. The process from the root to the leaf node is forward encoding. And (3) backward decoding, namely establishing a rule base by utilizing the characteristic that each leaf node in the tree corresponds to a unique path from the root to the leaf, finding each path from the leaf node to the root node of the tree, simplifying a plurality of rules according to the maximum complete rule, namely the logic that the range of each constraint in the rule can not be expanded any more, finally obtaining the mapping from each leaf node to the root node, and realizing the backward decoding.
The symbol rule corresponding to the backward decoding is as follows:
RULE 1 :(x 1 ≥0)∧(x 2 ≥1.5)∧(x 3 ＝＝RED)
according to the rule maximum complete rule, i.e. the rule can not be expanded any more in a constraint range, if expanded, a conflict is generated.
(6) And (3) after the sample is obtained through on-line monitoring, obtaining a feature sample set to be tested according to the steps (2) - (3), and carrying out Euclidean distance calculation with the health feature sample set to obtain a distance deviation degree as a health index.
The distance deviation degree is as follows:
by adopting the technical scheme disclosed by the invention, the following beneficial effects are obtained:
the invention provides a pump station unit state evaluation method based on deep convolution random forest self-coding, which is characterized in that a unit failure trend is obtained by analyzing a plurality of full life cycle examples from the beginning to the failure of a unit, and when a new failure trend occurs to the unit, the most probable trend change can be calculated through a model and the residual life can be estimated. The deep learning can automatically mine inherent fault characteristics of the unit through massive data, the health state of the unit is represented by modeling the data of normal behaviors of the unit and abstracting the implicit characteristics of the data step by step, after the model is built, the health degree of the unit can be measured by calculating the similarity between the current sensor data and the normal behavior state, and more accurate residual life is obtained through an extrapolation algorithm. The state evaluation method is based on-line monitoring data, and does not need to stop and manually arrange measuring points, so that the economic benefit of the power plant is greatly increased, meanwhile, the multi-source data of the unit can be comprehensively utilized for modeling, the hidden characteristics of the unit are automatically mined, manual judgment by an expert is not needed, and the method is suitable for units in different forms and different running times, and has self-adaption and objectivity.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which is also intended to be covered by the present invention.
Claims (3)
1. A pump station unit state evaluation method based on depth convolution random forest self-coding is characterized by comprising the following steps of: comprises the following steps of the method,
s1, collecting key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large axis swing degree waveform and waveform data of the pump station unit in a whole life cycle so as to obtain a historical waveform health sample library;
s2, selecting information of each measuring point in the pump station unit health state period from the historical waveform health sample library as a historical normal data sample;
s3, selecting whole period data of each waveform according to the key phase signal from the historical normal data sample, and forming a two-dimensional matrix with a horizontal axis as a time point and a vertical axis as a channel number according to the corresponding time scale;
s4, extracting features of the two-dimensional matrix through a deep convolutional neural network to obtain sample features;
s5, inputting sample features into a random forest self-coding model for further compression to obtain a health feature sample set;
s6, monitoring a pump station unit on line, and acquiring key phase signals of an upward vibration X-direction waveform, an upward vibration Y-direction vibration waveform, a water vibration X-direction vibration waveform, a water vibration Y-direction vibration waveform, an upper frame X-direction vibration waveform, an upper frame Y-direction vibration waveform, an upper frame vertical vibration waveform, an impeller shell X-direction vibration, an impeller shell Y-direction vibration, a large-axis swing degree waveform and waveform data of the pump station unit in the whole period in real time to obtain a waveform sample to be tested;
s7, executing the step S2 to the step S5 on the waveform sample to be tested to obtain a characteristic sample set to be tested;
s8, performing Euclidean distance calculation on the feature sample set to be tested and the health feature sample set to obtain a distance deviation degree;
s9, evaluating the health state of the pump station unit by using the distance deviation degree.
2. The pump station set state evaluation method based on depth convolution random forest self-coding according to claim 1, wherein the method comprises the following steps of: the deep convolution neural network comprises a forward convolution model and a backward convolution model, wherein the forward convolution model adopts 5 convolution-pooling layers, and convolution kernels are 128 x 3, 64 x 3, 32 x 3, 16 x 3 and 8*3 in sequence; the deconvolution model adopts 5 convolution-pooling layers, and convolution kernels are 8*3, 16×3, 32×3, 64×3 and 128×3 in sequence; and the forward convolution model and the deconvolution model perform feature extraction on the two-dimensional matrix to obtain sample features.
3. The pump station set state evaluation method based on depth convolution random forest self-coding according to claim 1, wherein the method comprises the following steps of: the random forest self-coding model adopts a tree integration model of a random forest to construct a self-coding process; the random forest self-coding model comprises forward coding and reverse coding; after the forward code receives the sample feature, the sample feature is sent to each root node of the tree in the integration, and after the sample feature traverses the root nodes of all the trees, a first result vector is returned; the reverse coding utilizes the characteristic that each leaf node in the tree corresponds to only one path from the root to the leaf, a rule base is established to acquire each path from the leaf node to the root node of the tree, a plurality of rules in the rule base are simplified according to the logic that the range of each constraint in the rules can not be expanded any more, and finally, the mapping between each leaf node to the root node of all the trees is acquired, namely, a second result vector mapped with the first result vector is acquired, and the second result vector is a health feature sample set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010895695.XA CN111914490B (en) | 2020-08-31 | 2020-08-31 | Pump station unit state evaluation method based on depth convolution random forest self-coding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010895695.XA CN111914490B (en) | 2020-08-31 | 2020-08-31 | Pump station unit state evaluation method based on depth convolution random forest self-coding |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111914490A CN111914490A (en) | 2020-11-10 |
| CN111914490B true CN111914490B (en) | 2023-10-13 |
Family
ID=73267657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010895695.XA Active CN111914490B (en) | 2020-08-31 | 2020-08-31 | Pump station unit state evaluation method based on depth convolution random forest self-coding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111914490B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112539183B (en) * | 2020-11-25 | 2022-06-17 | 南京河海南自水电自动化有限公司 | Fault diagnosis method and system for water pump shut-off device failure |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7013223B1 (en) * | 2002-09-25 | 2006-03-14 | The Board Of Trustees Of The University Of Illinois | Method and apparatus for analyzing performance of a hydraulic pump |
| CN108241873A (en) * | 2018-01-10 | 2018-07-03 | 中国水利水电科学研究院 | A kind of intelligent failure diagnosis method towards pumping plant main equipment |
| CN108869145A (en) * | 2018-04-26 | 2018-11-23 | 中国水利水电科学研究院 | Pumping plant unit diagnostic method based on compound characteristics index and depth limit learning machine |
| CN109063782A (en) * | 2018-08-16 | 2018-12-21 | 中国水利水电科学研究院 | A kind of adaptive pumping plant intelligent fault diagnosis method |
| CN109102032A (en) * | 2018-09-03 | 2018-12-28 | 中国水利水电科学研究院 | A kind of pumping plant unit diagnostic method based on depth forest and oneself coding |
| WO2019153388A1 (en) * | 2018-02-12 | 2019-08-15 | 大连理工大学 | Power spectral entropy random forest-based aeroengine rolling bearing fault diagnosis method |
-
2020
- 2020-08-31 CN CN202010895695.XA patent/CN111914490B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7013223B1 (en) * | 2002-09-25 | 2006-03-14 | The Board Of Trustees Of The University Of Illinois | Method and apparatus for analyzing performance of a hydraulic pump |
| CN108241873A (en) * | 2018-01-10 | 2018-07-03 | 中国水利水电科学研究院 | A kind of intelligent failure diagnosis method towards pumping plant main equipment |
| WO2019153388A1 (en) * | 2018-02-12 | 2019-08-15 | 大连理工大学 | Power spectral entropy random forest-based aeroengine rolling bearing fault diagnosis method |
| CN108869145A (en) * | 2018-04-26 | 2018-11-23 | 中国水利水电科学研究院 | Pumping plant unit diagnostic method based on compound characteristics index and depth limit learning machine |
| CN109063782A (en) * | 2018-08-16 | 2018-12-21 | 中国水利水电科学研究院 | A kind of adaptive pumping plant intelligent fault diagnosis method |
| CN109102032A (en) * | 2018-09-03 | 2018-12-28 | 中国水利水电科学研究院 | A kind of pumping plant unit diagnostic method based on depth forest and oneself coding |
Non-Patent Citations (6)
| Title |
|---|
| Degradation Assessment and Remaining Useful Life Prognostic of Centrifugal Pump using Multi-Sensor Process Monitoring;Taotao Zhou等;《2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling (APARM)》;第1-4页 * |
| 基于AdaBoost算法的水质监测数据清洗方;侯锋等;《水电站机电技术》;第46卷(第5期);第109-111+126页 * |
| 基于稀疏自编码支持向量机泵站故障检测;殷振兴;《水科学与工程技术》(第06期);第52-55页 * |
| 基于自编码支持向量机泵站故障识别;洪健;《中国水利学会2018学术年会论文集第五分册》(第6期);全文 * |
| 梯级泵站输水系统预测控制方法研究;郑和震等;《人民黄河》;第40卷(第2期);140-143页 * |
| 调水工程水位数据清洗及预测模型研究;薛萍;《中国优秀硕士学位论文全文数据库》;工程科技Ⅱ辑 C037-99 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111914490A (en) | 2020-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109141881B (en) | Rotary machine health assessment method of deep self-coding network | |
| CN107423414B (en) | Information transfer model-based process industry complex electromechanical system fault tracing method | |
| CN111401584B (en) | Automatic equipment fault diagnosis method and system | |
| CN111337237B (en) | Equipment fault diagnosis method and system | |
| CN113570138B (en) | Method and device for predicting residual service life of equipment of time convolution network | |
| CN110399986B (en) | Generating method of pump station unit fault diagnosis system | |
| CN112305388B (en) | On-line monitoring and diagnosing method for insulation partial discharge faults of generator stator winding | |
| CN111209934A (en) | Fan fault prediction and alarm method and system | |
| CN115614292B (en) | Vibration monitoring device and method for vertical water pump unit | |
| CN111504647A (en) | AR-MSET-based performance degradation evaluation method for rolling bearing | |
| CN111914490B (en) | Pump station unit state evaluation method based on depth convolution random forest self-coding | |
| CN111637045A (en) | Fault diagnosis method for air compressor of ocean platform | |
| CN113435759A (en) | Primary equipment risk intelligent evaluation method based on deep learning | |
| CN117074961A (en) | Variation self-encoder power battery abnormality detection method, system, device and storage medium | |
| Liang et al. | Multibranch and multiscale dynamic convolutional network for small sample fault diagnosis of rotating machinery | |
| CN114021620A (en) | Electrical submersible pump fault diagnosis method based on BP neural network feature extraction | |
| CN117076915A (en) | Intelligent fault attribution analysis method and system for FPSO crude oil process system | |
| CN112308391A (en) | Real-time monitoring and anomaly detection method for equipment state based on neural network | |
| CN111143934A (en) | Structural deformation prediction method based on time convolution network | |
| CN114781450B (en) | Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN | |
| CN114239708B (en) | Combustion engine abnormity detection method based on quality control chart theory | |
| CN114330141A (en) | Fan main shaft bearing service life prediction method based on GRU (generalized regression Unit) super-parameter optimization | |
| CN114118720A (en) | Nuclear power plant circulating water pump operation state comprehensive evaluation method and device, electronic equipment and storage medium | |
| CN110441081B (en) | Intelligent diagnosis method and intelligent diagnosis system for faults of rotating machinery | |
| CN114417704A (en) | Wind turbine generator health assessment method based on improved stack type self-coding |
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 |