CN114357864A - Phase modulator state evaluation method and evaluation system based on fuzzy reasoning - Google Patents
Phase modulator state evaluation method and evaluation system based on fuzzy reasoning Download PDFInfo
- Publication number
- CN114357864A CN114357864A CN202111553662.8A CN202111553662A CN114357864A CN 114357864 A CN114357864 A CN 114357864A CN 202111553662 A CN202111553662 A CN 202111553662A CN 114357864 A CN114357864 A CN 114357864A
- Authority
- CN
- China
- Prior art keywords
- phase modulator
- state
- factors
- different
- fuzzy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The application discloses a phase modulator state evaluation method and system based on fuzzy reasoning, wherein the method comprises the following steps: screening a characteristic factor set; determining an evaluation set; establishing a membership function; determining influence factors and weights; training a membership function; establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model; and predicting and evaluating the phase modifier state to realize the evaluation and early warning of different operation state types. The method determines the optimal influence factor and weight based on the phase modulator operation stable state curve evaluated by different factors; and each characteristic factor adopts an optimal prediction model to carry out data simulation prediction, and effective evaluation and early warning of different operation state types can be realized by combining the trained membership function.
Description
Technical Field
The invention belongs to the technical field of phase modulator state monitoring, and relates to a phase modulator state evaluation method and system based on fuzzy reasoning.
Background
The phase modulator state is related to characteristics such as electricity, magnetism, vibration, heat, force, acoustics, the complexity of the phase modulator characteristic relation determines that the same fault or operation state has different characteristic quantities, sometimes the same characteristic quantity corresponds to different faults or states (such as vibration increase and many causes), the phase modulator state evaluation mainly researches the relation between internal faults and vibration/acoustics, and the relation between the faults and the vibration characteristics is revealed from various aspects such as waveforms, frequency spectrums, the values of specific frequency components, energy spectrums, fault fingerprints and the like.
The state evaluation should reflect the early state change of the equipment, and the emphasis is not to judge the developed fault. Therefore, useful components (characteristic quantities) in the monitored quantity are very weak (to reflect early abnormality of an object), and for early detection of abnormality of the phase modulator, weak characteristic signals in strong quantity are needed to be analyzed, so that extraction of the weak characteristic signals in the strong interference is also a key point and a difficulty point of phase modulator state evaluation. In addition, most of the existing phase modulator state evaluation methods are based on the current situation evaluation of real-time data, and the effective prediction evaluation of the phase modulator state evaluation is more important for early diagnosis and early warning.
Disclosure of Invention
In order to overcome the defects in the prior art, the application provides a phase modulator state evaluation method and system based on fuzzy reasoning.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a phase modulator state estimation method based on fuzzy inference, the method comprising the steps of:
step 1: screening a characteristic factor set for phase modifier state evaluation according to parameters of the running state of the phase modifier;
step 2: determining an evaluation set according to the operation state type of the phase modifier;
and step 3: acquiring historical operating state data of a phase modulator, and establishing a membership function of each characteristic factor of the phase modulator;
and 4, step 4: determining influence factors and weights by adopting a weighted voting method based on a phase modulator running stable state curve evaluated by different factors;
and 5: acquiring recent historical operating data of the influence factors as a training set, and training a membership function by combining the weights of the influence factors;
step 6: combining fuzzy logic reasoning with a neural network, establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model;
and 7: and (6) predicting and evaluating the phase modifier state by adopting the trained membership function and the data in the step 6, and realizing evaluation and early warning of different operation state types.
Preferably, the characteristic factor set obtained by screening in step 1 comprises current, voltage, temperature and vibration signals of a phase modulator;
in step 2, an evaluation set { stable, unstable and unstable } is determined according to the stable, unstable and unstable operation state types of the phase modulator.
Preferably, step 3 specifically comprises:
step 3.1: acquiring a recent historical operating data set of a phase modulator to be evaluated;
step 3.2: clustering the data set to obtain a clustering center point of each characteristic factor;
step 3.3: obtaining cluster center points of different running states of each characteristic factor according to the running state type corresponding to the data set;
step 3.4: and 3.3, establishing a membership function of each characteristic factor of the phase modulator according to the clustering center point in the step 3.3.
Preferably, a fuzzy C-means clustering method is used for clustering in step 3.2.
Preferably, the membership function of step 3 is a gaussian membership function.
Preferably, in step 4, different characteristic factors are combined to form a plurality of comparison factor sets, stable state curves evaluated by the different comparison factor sets are obtained, the variation trend of the comparison curves is analyzed, sensitive factors in stability judgment are screened as influence factors by adopting a weighted voting method, and the weight of the influence factors is set according to the sensitivity degree.
Preferably, in step 4, the screening determines the influence factors as voltage, temperature and vibration signals, with weights optimally assigned to 0.2, 0.5 and 0.3.
Preferably, in step 6, different neural networks and fuzzy logic reasoning are respectively adopted to combine to establish a fuzzy neural network model, and according to the dynamic prediction result of the photographic camera operation data, data prediction models of different characteristic factors are determined, specifically:
and respectively predicting different characteristic factors of the camera by adopting fuzzy neural network models obtained according to different neural networks, comparing the predicted value with the actual value, calculating a correlation coefficient, a mean value and a variance thereof, and screening and determining data prediction models of different characteristic factors according to the correlation coefficient, the mean value and the variance thereof.
Preferably, the different neural networks include a BP neural network, a parallel genetic algorithm-based neural network, and a hybrid neural network.
The invention also provides a phase modulator state evaluation system based on fuzzy reasoning, which comprises:
the characteristic factor set screening module is used for screening a characteristic factor set used for phase modulator state evaluation according to the parameters of the running state of the reaction phase modulator;
the evaluation set setting module is used for determining an evaluation set according to the running state type of the phase modulator;
the membership function building module is used for acquiring historical operating state data of the phase modulator and building membership functions of various characteristic factors of the phase modulator;
the influence factor and weight setting module is used for determining the influence factor and weight based on a phase modulator running stable state curve evaluated by different factors by adopting a weighted voting method;
the membership function training module is used for acquiring recent historical operating data of the influence factors as a training set and training the membership functions by combining the weights of the influence factors;
the operation data prediction module is used for combining fuzzy logic reasoning with a neural network, establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model;
and the state evaluation module is used for predicting and evaluating the state of the phase modifier by adopting the trained membership function and the data of the operation data prediction module, and realizing evaluation and early warning of different operation state types.
The beneficial effect that this application reached:
the method comprises the steps of determining the optimal influence factor and weight based on a phase modulator running stable state curve evaluated by different factors by adopting a weighted voting method; further acquiring recent historical operating data of the influence factors as a training set, and training a membership function by combining the weight of the influence factors, so that the optimal evaluation effect of the operating stable state of the phase modulator can be realized;
the invention combines fuzzy logic reasoning with various neural networks, establishes a fuzzy neural network model of the phase modulator in the operation process, screens and determines data prediction models of different characteristic factors, adopts an optimal prediction model to carry out data simulation prediction on each characteristic factor, and combines a trained membership function to realize effective evaluation and early warning of different operation state types.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
As shown in fig. 1, the phase modulator state evaluation method based on fuzzy inference of the present invention includes the following steps:
a phase modulator state evaluation method based on fuzzy inference comprises the following steps:
step 1: screening a characteristic factor set for phase modifier state evaluation according to parameters of the running state of the phase modifier;
when the method is implemented specifically, the characteristic factor set obtained by screening comprises current, voltage, temperature and vibration signals of a phase modulator;
step 2: determining an evaluation set according to the operation state type of the phase modifier, specifically:
determining an evaluation set { stable, unstable and unstable } according to the stable, unstable and unstable operation state types of the phase modulator;
and step 3: obtaining historical operating state data of a phase modulator, and establishing a membership function of each characteristic factor of the phase modulator, wherein the method specifically comprises the following steps:
step 3.1: acquiring a recent historical operating data set of a phase modulator to be evaluated;
step 3.2: clustering the data set to obtain a clustering center point of each characteristic factor;
preferably, the clustering is performed using a fuzzy C-means clustering method.
Step 3.3: obtaining cluster center points of different running states of each characteristic factor according to the running state type corresponding to the data set;
step 3.4: and 3.3, establishing a membership function of each characteristic factor of the phase modulator according to the clustering center point in the step 3.3.
Preferably, the membership functions are gaussian membership functions.
And 4, step 4: determining influence factors and weights by adopting a weighted voting method and based on a phase modulator running stable state curve evaluated by different factors, wherein the method specifically comprises the following steps:
combining different characteristic factors to form a plurality of comparison factor sets, acquiring stable state curves evaluated by the different comparison factor sets, analyzing the variation trend of the comparison curves, screening sensitive factors in the aspect of stability judgment as influence factors by adopting a weighted voting method, and setting the weight of the influence factors according to the sensitivity degree.
Through a large number of experiments, the influence factors are screened and determined to be voltage, temperature and vibration signals, the running stable state of the phase modulator can be well reflected, and the optimal weight distribution is 0.2, 0.5 and 0.3.
And 5: acquiring recent historical operating data of the influence factors as a training set, and training a membership function by combining the weights of the influence factors;
step 6: combining fuzzy logic reasoning with a neural network, establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model;
respectively adopting different neural networks and fuzzy logic reasoning to combine to establish a fuzzy neural network model, and determining data prediction models of different characteristic factors according to dynamic prediction results of the photographic camera operation data, wherein the data prediction models specifically comprise:
and respectively predicting different characteristic factors of the camera by adopting fuzzy neural network models obtained according to different neural networks, comparing the predicted value with the actual value, calculating a correlation coefficient, a mean value and a variance thereof, and screening and determining data prediction models of different characteristic factors according to the correlation coefficient, the mean value and the variance thereof.
Preferably, the different neural networks include a BP neural network, a parallel genetic algorithm-based neural network, and a hybrid neural network.
And 7: and (6) predicting and evaluating the phase modifier state by adopting the trained membership function and the data in the step 6, and realizing evaluation and early warning of different operation state types.
A fuzzy inference based phase modulator state estimation system comprising:
the characteristic factor set screening module is used for screening a characteristic factor set used for phase modulator state evaluation according to the parameters of the running state of the reaction phase modulator;
the evaluation set setting module is used for determining an evaluation set according to the running state type of the phase modulator;
the membership function building module is used for acquiring historical operating state data of the phase modulator and building membership functions of various characteristic factors of the phase modulator;
the influence factor and weight setting module is used for determining the influence factor and weight based on a phase modulator running stable state curve evaluated by different factors by adopting a weighted voting method;
the membership function training module is used for acquiring recent historical operating data of the influence factors as a training set and training the membership functions by combining the weights of the influence factors;
the operation data prediction module is used for combining fuzzy logic reasoning with a neural network, establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model;
and the state evaluation module is used for predicting and evaluating the state of the phase modifier by adopting the trained membership function and the data of the operation data prediction module, and realizing evaluation and early warning of different operation state types.
The method comprises the steps of determining the optimal influence factor and weight based on a phase modulator running stable state curve evaluated by different factors by adopting a weighted voting method; further acquiring recent historical operating data of the influence factors as a training set, and training a membership function by combining the weight of the influence factors, so that the optimal evaluation effect of the operating stable state of the phase modulator can be realized;
the invention combines fuzzy logic reasoning with various neural networks, establishes a fuzzy neural network model of the phase modulator in the operation process, screens and determines data prediction models of different characteristic factors, adopts an optimal prediction model to carry out data simulation prediction on each characteristic factor, and combines a trained membership function to realize effective evaluation and early warning of different operation state types.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are merely preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for limiting the scope of the present invention, and on the contrary, any improvement or modification made based on the spirit of the present invention should fall within the scope of the present invention.
Claims (10)
1. A phase modulator state evaluation method based on fuzzy reasoning is characterized in that:
the method comprises the following steps:
step 1: screening a characteristic factor set for phase modifier state evaluation according to parameters of the running state of the phase modifier;
step 2: determining an evaluation set according to the operation state type of the phase modifier;
and step 3: acquiring historical operating state data of a phase modulator, and establishing a membership function of each characteristic factor of the phase modulator;
and 4, step 4: determining influence factors and weights by adopting a weighted voting method based on a phase modulator running stable state curve evaluated by different factors;
and 5: acquiring recent historical operating data of the influence factors as a training set, and training a membership function by combining the weights of the influence factors;
step 6: combining fuzzy logic reasoning with a neural network, establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model;
and 7: and (6) predicting and evaluating the phase modifier state by adopting the trained membership function and the data in the step 6, and realizing evaluation and early warning of different operation state types.
2. The phase modulator state estimation method based on fuzzy inference as claimed in claim 1, characterized in that:
the characteristic factor set obtained by screening in the step 1 comprises current, voltage, temperature and vibration signals of a phase modulator;
in step 2, an evaluation set { stable, unstable and unstable } is determined according to the stable, unstable and unstable operation state types of the phase modulator.
3. The phase modulator state estimation method based on fuzzy inference as claimed in claim 1, characterized in that:
the step 3 specifically comprises the following steps:
step 3.1: acquiring a recent historical operating data set of a phase modulator to be evaluated;
step 3.2: clustering the data set to obtain a clustering center point of each characteristic factor;
step 3.3: obtaining cluster center points of different running states of each characteristic factor according to the running state type corresponding to the data set;
step 3.4: and 3.3, establishing a membership function of each characteristic factor of the phase modulator according to the clustering center point in the step 3.3.
4. A phase modulator state estimation method based on fuzzy inference, according to claim 3, characterized by:
and 3.2, clustering by adopting a fuzzy C-means clustering method.
5. The phase modulator state estimation method based on fuzzy inference as claimed in claim 1, characterized in that:
and 3, the membership function is a Gaussian membership function.
6. The phase modulator state estimation method based on fuzzy inference as claimed in claim 1, characterized in that:
and 4, combining different characteristic factors to form a plurality of comparison factor sets, acquiring stable state curves evaluated by the different comparison factor sets, analyzing the variation trend of the comparison curves, screening sensitive factors in the aspect of stability judgment as influence factors by adopting a weighted voting method, and setting the weight of the influence factors according to the sensitivity degree.
7. The phase modulator state estimation method based on fuzzy inference as claimed in claim 1, characterized in that:
in step 4, the influence factors are screened and determined to be voltage, temperature and vibration signals, and the optimal weight distribution of the influence factors is 0.2, 0.5 and 0.3.
8. The phase modulator state estimation method based on fuzzy inference as claimed in claim 1, characterized in that:
in step 6, different neural networks and fuzzy logic reasoning are respectively adopted to be combined to establish a fuzzy neural network model, and data prediction models of different characteristic factors are determined according to dynamic prediction results of the photographic camera operation data, wherein the data prediction models specifically comprise:
and respectively predicting different characteristic factors of the camera by adopting fuzzy neural network models obtained according to different neural networks, comparing the predicted value with the actual value, calculating a correlation coefficient, a mean value and a variance thereof, and screening and determining data prediction models of different characteristic factors according to the correlation coefficient, the mean value and the variance thereof.
9. The phase modulator state estimation method based on fuzzy inference as claimed in claim 8, wherein:
the different neural networks include a BP neural network, a parallel genetic algorithm based neural network, and a hybrid neural network.
10. A phase modulation machine state estimation system of a phase modulation machine state estimation method based on fuzzy inference according to any one of claims 1 to 9, characterized in that:
the system comprises:
the characteristic factor set screening module is used for screening a characteristic factor set used for phase modulator state evaluation according to the parameters of the running state of the reaction phase modulator;
the evaluation set setting module is used for determining an evaluation set according to the running state type of the phase modulator;
the membership function building module is used for acquiring historical operating state data of the phase modulator and building membership functions of various characteristic factors of the phase modulator;
the influence factor and weight setting module is used for determining the influence factor and weight based on a phase modulator running stable state curve evaluated by different factors by adopting a weighted voting method;
the membership function training module is used for acquiring recent historical operating data of the influence factors as a training set and training the membership functions by combining the weights of the influence factors;
the operation data prediction module is used for combining fuzzy logic reasoning with a neural network, establishing a fuzzy neural network model of the phase modulator in the operation process, and simulating and predicting data of different characteristic factors in the camera operation process by adopting the fuzzy neural network model;
and the state evaluation module is used for predicting and evaluating the state of the phase modifier by adopting the trained membership function and the data of the operation data prediction module, and realizing evaluation and early warning of different operation state types.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2020115124661 | 2020-12-19 | ||
| CN202011512466 | 2020-12-19 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114357864A true CN114357864A (en) | 2022-04-15 |
Family
ID=81100133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111553662.8A Pending CN114357864A (en) | 2020-12-19 | 2021-12-17 | Phase modulator state evaluation method and evaluation system based on fuzzy reasoning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114357864A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117272233A (en) * | 2023-11-21 | 2023-12-22 | 中国汽车技术研究中心有限公司 | Diesel engine emission prediction method, apparatus, and storage medium |
-
2021
- 2021-12-17 CN CN202111553662.8A patent/CN114357864A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117272233A (en) * | 2023-11-21 | 2023-12-22 | 中国汽车技术研究中心有限公司 | Diesel engine emission prediction method, apparatus, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108731923B (en) | Fault detection method and device for rotary mechanical equipment | |
| CN111596604A (en) | Intelligent fault diagnosis and self-healing control system and method for engineering equipment based on digital twinning | |
| CN109145516B (en) | Analog circuit fault identification method based on improved extreme learning machine | |
| CN109739214B (en) | Method for detecting intermittent faults in industrial process | |
| CN1981297A (en) | Data compressing device and method, data analyzing device and method, and data managing system | |
| CN110889111A (en) | Power grid virtual data injection attack detection method based on deep belief network | |
| CN103020591A (en) | Medium scale crowd abnormal behavior detection method based on causal network analysis | |
| CN114499979B (en) | SDN abnormal flow cooperative detection method based on federal learning | |
| CN113125987A (en) | Novel hybrid lithium ion battery health state prediction method | |
| CN107976934A (en) | A kind of oil truck oil and gas leakage speed intelligent early-warning system based on wireless sensor network | |
| CN114357864A (en) | Phase modulator state evaluation method and evaluation system based on fuzzy reasoning | |
| Lu et al. | Joint training of a predictor network and a generative adversarial network for time series forecasting: A case study of bearing prognostics | |
| CN111695452A (en) | Parallel reactor internal aging degree evaluation method based on RBF neural network | |
| CN117375237A (en) | Substation operation and maintenance method and system based on digital twin technology | |
| CN115758083A (en) | Motor bearing fault diagnosis method based on time domain and time-frequency domain fusion | |
| CN117113729A (en) | Digital twinning-based power equipment online state monitoring system | |
| CN115098962A (en) | Method for predicting residual life of mechanical equipment in degradation state based on hidden half Markov model | |
| CN112232370A (en) | Fault analysis and prediction method for engine | |
| Song et al. | Intelligent diagnosis method for machinery by sequential auto-reorganization of histogram | |
| CN114841076A (en) | Power battery production process fluctuation abnormity detection method based on space-time diagram model | |
| CN114896861A (en) | Rolling bearing residual life prediction method based on square root volume Kalman filtering | |
| CN117093852A (en) | Early abnormality monitoring model and method for industrial robot | |
| CN110956112B (en) | Novel high-reliability slewing bearing service life assessment method | |
| CN116094758B (en) | Large-scale network flow acquisition method and system | |
| CN116699400A (en) | Generator rotor short-circuit fault monitoring system, method and readable storage medium |
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 |