LU500843B1 - A monitoring system and prediction method of temperature for generator carbon brush based on an infrared image - Google Patents
A monitoring system and prediction method of temperature for generator carbon brush based on an infrared image Download PDFInfo
- Publication number
- LU500843B1 LU500843B1 LU500843A LU500843A LU500843B1 LU 500843 B1 LU500843 B1 LU 500843B1 LU 500843 A LU500843 A LU 500843A LU 500843 A LU500843 A LU 500843A LU 500843 B1 LU500843 B1 LU 500843B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- temperature
- carbon brush
- generator
- infrared image
- data
- Prior art date
Links
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 87
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 87
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012544 monitoring process Methods 0.000 title claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 7
- 210000002569 neuron Anatomy 0.000 claims description 12
- 230000005284 excitation Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000004088 simulation Methods 0.000 claims description 8
- 238000009529 body temperature measurement Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 238000010438 heat treatment Methods 0.000 claims description 6
- 230000008901 benefit Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 5
- 238000009826 distribution Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 3
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 claims description 3
- 238000013439 planning Methods 0.000 claims description 3
- 238000013523 data management Methods 0.000 abstract description 5
- 238000007405 data analysis Methods 0.000 abstract description 4
- 230000002159 abnormal effect Effects 0.000 abstract description 3
- 238000011161 development Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 abstract description 2
- 238000013480 data collection Methods 0.000 abstract description 2
- 238000013500 data storage Methods 0.000 abstract description 2
- 238000001816 cooling Methods 0.000 abstract 1
- 230000010354 integration Effects 0.000 abstract 1
- 238000011897 real-time detection Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000003331 infrared imaging Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J2005/0077—Imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J5/02—Constructional details
- G01J5/025—Interfacing a pyrometer to an external device or network; User interface
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/84—Greenhouse gas [GHG] management systems
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention discloses a monitoring system and prediction method of temperature for generator carbon brush based on an infrared image. In view of the abnormal temperature rise of the generator carbon brush and its connected slip ring in the diversion process due to the high current and abnormal cooling device in the large generator carbon brush, which affects the normal operation of the equipment, a remote real-time detection, data analysis and prediction method is designed. The invention develops an on-line monitoring system for the temperature of the generator carbon brush, which is adapted to multiple platforms and cooperates with 4G network to realize remote temperature monitoring for multi device. The data is connected to the cloud server to realize centralized data storage, collection and management by using the cloud data management system. By using the actual carbon brush temperature data of the power plant collected by the carbon brush temperature monitoring system, the development trend of carbon brush temperature is predicted and analyzed by LSTM-BP. At the same time, based on the integration of infrared image characteristic information and carbon brush temperature information, the prediction accuracy of carbon brush temperature is improved by combination model, and the change of carbon brush temperature is predicted through data analysis.
Description
A MONITORING SYSTEM AND PREDICTION METHOD OF Se
FIELD OF THE INVENTION The invention relates to the field of municipal public facilities, in particular to a monitoring system and prediction method of temperature for generator carbon brush based on an infrared image.
BACKGROUND OF THE RELATED ART Power is related to the national economy and the people's livelihood. In 2018, the capacity of thermal motor assembly units in China reached 1.1 billion kw, and thermal power units accounted for 70% of the total capacity. It is expected that by the end of 2019, china's installed power generation capacity is expected to exceed 2 billion kw for the first time, and the total installed capacity will rank first in the world. With the increasing installed capacity, the safety of the system has been paid close attention. As one of the three main engines of thermal power units, the equipment reliability of generators has a significant impact on the operation of units. Among them, the excitation system is the key to the safe and stable operation of generator units.
As an important part of the excitation system of synchronous generator, carbon brush and slip ring are also the weak links of generator operation. The fault statistics of the generator show that: Carbon brush and slip ring faults are multiple faults,in case of carbon brush runout fault, the contact resistance increases and the current distribution unevenness between carbon brushes increases, when the current in the carbon brush exceeds 80A, a series of problems such as unstable excitation voltage and current and system power fluctuation will occur. The thermal effect of the current may further burn the carbon brush, resulting in abnormal increase in the temperature of the generator carbon brush and slip ring, in serious cases, it will even cause ring 1 fire, which will burn out the slip ring of the generator. Therefore, the quality of carbon 7500849 brush directly affects the production safety of generator, the research of a set of carbon brush temperature monitoring system is of great significance to improve the economic benefits of power plant.
At present, there are mainly two methods to measure the carbon brush temperature at home and abroad, namely "point temperature gun temperature measurement" and "infrared temperature sensor temperature measurement", the temperature measurement method of point temperature gun has cumbersome steps, low efficiency and easy to make mistakes. At the same time, the on-site environment is dangerous. Infrared temperature sensor temperature measurement method is a non-contact measurement, which eliminates complex wiring and cumbersome manual inspection, its disadvantage is that most single point or matrix sensors are used for temperature measurement, with limited coverage, iow pixels, iess data collection, and poor accuracy, which can not meet the needs of large power plants. Therefore, a set of temperature monitoring system can be developed through infrared thermal imager, which can not only infrared imaging, high pixels and large coverage area, but also collect and process temperature data.
Therefore, a set of temperature monitoring system can be developed through infrared thermal imager, which can not only infrared imaging, high pixels and large coverage area, but also collect and process temperature data. Therefore, a set of temperature monitoring system can be developed through infrared thermal imager, which can not only infrared imaging, high pixels and large coverage area, but also collect and process temperature data. The power plant generally draws the carbon brush temperature curve, judges the change of carbon brush temperature through the trend, and predicts the overall future trend of the whole event through the analysis of historical data of a certain period. However, the error of this method is large, and only the general trend of temperature can be obtained, which can not do a good quantitative analysis. Through the prediction of generator carbon brush temperature, it can effectively reduce the unplanned shutdown accident of the unit caused by the fault of generator collector ring and carbon brush,At the same time, the analysis can 2 provide guiding significance for the operation and maintenance personnel to analyze 7500849 the operation state of carbon brushes and the load of excitation system in the power plant.
SUMMARY OF THE INVENTION The technical problem to be solved by the invention is to provide a monitoring system and prediction method of temperature for generator carbon brush based on an infrared image, which can realize multi equipment remote real-time monitoring, database data storage, temperature data analysis and prediction, on the basis of improving the detection efficiency, it can also provide guiding significance for the operation and maintenance personnel to analyze the operation state of carbon brushes and the load of excitation system in the power plant.
In order to achieve the above purpose, the invention adopts the following technical scheme: The temperature monitoring system of generator carbon brush based on infrared image includes infrared thermal imager, 4G network, cloud server and temperature monitoring system of carbon brush; The infrared thermal imager is an infrared detector with uncooled focal plane, the pixel value specification is 160 * 120, each infrared image contains 19200 pixels, and the temperature measurement range is - 20° C ~ 150 °C.
The 4G network and the cloud server access the data of the infrared thermal imager to the network through the industrial router loaded with 4G IOT card, which wirelessly connect with the cloud server through a fixed IP address and thermal imager number.
The temperature monitoring system of carbon brush includes upper computer software and database management software MYSQL developed based on MFC, which can remotely and real-time display the temperature infrared image measured on the industrial site on the industrial computer, and also has the functions of over temperature alarm, data recording, image recording, temperature curve drawing, 3 temperature parameter correction, fault diagnosis and so on. 7500849 The developed mobile terminal detection system is based on the software used by mobile devices such as mobile phones on the Android Studio development platform, and uses JAVA programming to wirelessly connect with the data management system with a fixed IP address.
Based on the above principle, the invention provides a temperature prediction method of generator carbon brush based on infrared image, which comprises the following steps: Step 1: Randomly select the generator brush temperature data for a period of time based on the experimental data of the power plant.
Step 2: Process the infrared image in the data, use Sobel operator to detect the edge and extract the carbon brush contour in the infrared image, and fill in the missing value by using the corresponding relationship between the temperature data and the infrared image area of the heating carbon brush.
Step 3: Use the combination model, give full play to the time series prediction advantages of LSTM, and combine the nonlinear characteristics of BP neural network to construct the LSTM-BP combination model.
Step 4: Use LSTM-BP combined model to predict the trend of generator carbon brush temperature, analyze the operation status of carbon brush and the load status of excitation system through the temperature trend, and verify the prediction accuracy from MAE, MAPE and RMSE.
Further, the temperature experimental data of generator carbon brush described in Step 1 is the temperature data of generator carbon brush collected by the power plant for 11 consecutive days, a total of 539, in which 485 training sets numtrain account for 90%, and 54 test sets numtest account for 10%.
Further, the contour extraction described in Step 2 is characterized in that the following operations are required: 1) Image binarization; 2) Use bwlabel to find the connected area, 4 connected refers to (if the position of pixels is up, down, left or right adjacent to other pixels, they are considered to be connected); 3) The tag value of bwing image is stored in the 4
L array (that is, it is marked in the L matrix after it is determined to be connected); 4) 500845 Select the region and only keep the region of interest; 5) Using region props to calculate the area of the reserved area.
After contour extraction, the current image is captured locally in the generator brush temperature online analysis system, through the corresponding relationship between the temperature data and the infrared image area of the heating carbon brush, the missing value can be filled, and a good effect can be achieved.
Further, the LSTM-BP combination model described in step 3 adopts the series combination method to determine the weight of LSTM model and BP model through the method of linear programming, and the parameter settings are as follows: LSTM network model: The input layer and the output layer are one neuron respectively, and the LSTM structural unit of the hidden layer is set as three layers with 60 ‚200 and 60 neurons respectively. The maximum number of iterations is 400. When the error is less than 10 — 5, the loop will jump out.
BP network model: The input layer and output layer are single neurons, by constantly adjusting the parameter trial and error method, it is determined that the BP neural network is a double hidden layer,there are 8 and 4 neurons respectively, the learning coefficient is 0.01, and the error control rate is 1x10 — 5, the maximum training times is 5000, the training function trainlm, and the test relative error distribution interval [0.01,0.08]; Further, in Step 4, the LSTM-BP combined model is used to predict the temperature trend of generator carbon brush, compare the results with the actually collected temperature curve and the results predicted by LSTM model, and then use the evaluation index to verify the prediction results, the specific steps are as follows: The evaluation indicators used are: 1 & _ MAE =y 2h, | a RMSE = X 3) Nia (2):
N _ LU500843 Na (3); Where “is the real temperature value of generator carbon brush / ° C; Yı is the predicted generator brush temperature / °C; N is the number of test sample sets.
Using matlab simulation, this method can well predict the temperature trend of generator carbon brush, and most of the prediction errors are stable within 0.3 °C, by predicting the temperature of generator carbon brush and analyzing the operation status of carbon brush and the load status of excitation system through temperature trend, at the same time, the combined model can provide a reasonable reference for related types of data prediction and decision planning.
BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a system structure composition diagram of the invention; Fig. 2 is the software flow chart of the carbon brush temperature analysis system of the invention; Fig. 3 is the field acquired images of the generator of the invention and the results of Sobel contour extraction (experiment); Fig. 4 is the system block diagram of the LSTM-BP combination model of the invention, and the combination model adopts the series combination mode; Fig. 5 is a comparison diagram (simulation) of LSTM-BP prediction results, LSTM prediction results and actual collected temperature data of the invention; Fig. 6 is the prediction network error (simulation) of the invention; Fig. 7 is the prediction relative error (simulation) of the invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In order to make the object, technical scheme and advantages of the invention more clear, the invention is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.
As shown in Fig. 1, the temperature monitoring system of generator carbon brush 6 based on infrared image includes two infrared thermal imagers, 4G industrial router, HU500843 GPRS / 4G base station, cloud server, carbon brush temperature analysis system, data management system and mobile terminal monitoring system; Two infrared thermal imagers are installed on both sides of the generator to ensure that their detection range includes the main parts of the generator slip ring and the whole range of the carbon brush; The data of the thermal imager is transmitted to the industrial 4G router through twisted pair, and the real-time measurement data image is transmitted to the cloud server through the 4G / GPRS base station, figure 2 is the software flow chart of carbon brush temperature analysis system, the upper computer carbon brush temperature analysis system developed on the remote industrial control computer obtains the temperature and image data on the cloud server for infrared thermal imaging and data analysis, in cooperation with MYSQL database, the upper computer can also realize the functions of over temperature alarm, data recording, image recording, temperature curve drawing, temperature parameter correction, fault diagnosis and so on; Mobile terminal monitoring system is a software used by mobile devices such as mobile phones based on Android Studio development platform. It uses JAVA programming to wirelessly connect with the data management system with a fixed IP address.
Based on the above principle, the invention provides a temperature prediction method of generator carbon brush based on infrared image based on infrared image, which comprises the following steps: Step 1: Randomly select the generator brush temperature data for a period of time based on the experimental data of the power plant.
The temperature experimental data of generator carbon brush is the temperature data of generator carbon brush collected by the power plant for 11 consecutive days, a total of 539, in which 485 training sets numtrain account for 90%, and 54 test sets numtest account for 10%.
Step 2: Process the infrared image in the data, use Sobel operator to detect the 7 edge and extract the carbon brush contour in the infrared image, and fill in the missing 7500849 value by using the corresponding relationship between the temperature data and the infrared image area of the heating carbon brush.
Contour extraction requires the following operations: 1) Image binarization; 2) Use bwlabel to find the connected area, 4 connected refers to (if the position of pixels is up, down, left or right adjacent to other pixels, they are considered to be connected); 3) The tag value of bwing image is stored in the L array (that is, it is marked in the L matrix after it is determined to be connected); 4) Select the region and only keep the region of interest; 5) Using region props to calculate the area of the reserved area. Figure 3 is the field acquired images of the generator and the results based on Sobel contour extraction.
After contour extraction, the current image is captured locally in the generator brush temperature online analysis system, through the corresponding relationship between the temperature data and the infrared image area of the heating carbon brush, the missing value can be filled, and a good effect can be achieved.
Step 3: Use the combination model, give full play to the time series prediction advantages of LSTM, and combine the nonlinear characteristics of BP neural network to construct the LSTM-BP combination model.
Figure 4 is the system block diagram of LSTM-BP combination model the combination model adopts series combination mode, the weights of LSTM model and BP model are determined by linear programming, the parameter settings are as follows: LSTM network model: The input layer and the output layer are one neuron respectively, and the LSTM structural unit of the hidden layer is set as three layers with 60 ,200 and 60 neurons respectively. The maximum number of iterations is 400. When the error is less than 10 — 5, the loop will jump out.
BP network model: The input layer and output layer are single neurons, by constantly adjusting the parameter trial and error method, it is determined that the BP neural network is a double hidden layer,there are 8 and 4 neurons respectively, the learning coefficient is 0.01, and the error control rate is 1x10 — 5, the maximum 8 training times is 5000, the training function trainlm, and the test relative error 500845 distribution interval [0.01,0.08]; Step 4: Use LSTM-BP combined model to predict the trend of generator carbon brush temperature, analyze the operation status of carbon brush and the load status of excitation system through the temperature trend, and verify the prediction accuracy from MAE, MAPE and RMSE.
Further, in Step 4, the LSTM-BP combined model is used to predict the temperature trend of generator carbon brush, compare the results with the actually collected temperature curve and the results predicted by LSTM model, and then use the evaluation index to verify the prediction results. The evaluation indexes used are: 1 & MAE =—>"|y, -5;| N= (1); RMSE = X 3) Nia (2); NT (3) ; Where ** is the real temperature value of generator carbon brush /° C; Yı is the predicted generator brush temperature / °C; N is the number of test sample sets. In order to verify the prediction effect of the above combined model, fill the data in step 1 with the missing value using the method in step 2, and then use the LSTM-BP combined model established in step 3 to predict the temperature trend, using matlab simulation: Figure 5 is the comparison between LSTM-BP prediction results, LSTM prediction results and actual collected temperature data. The front section of the curve is training data and the rear section is prediction data Figure 6 is the prediction network error Figure 7 is the relative error of prediction Table 1 shows the performance evaluation (simulation) of LSTM network and LSTM-BP network in predicting the temperature of generator carbon brush 9
Table 1 The simulation results show that this method can well predict the temperature trend of generator carbon brush, and most of the prediction errors are stable within
0.3 ° C. by predicting the temperature of generator carbon brush, the operation status of carbon brush and the load status of excitation system are analyzed through the temperature trend.
At the same time, the combined model can provide a reasonable reference for related types of data prediction and decision planning.
The above is a preferred embodiment of the invention. According to the teaching of the invention, the changes, modifications, substitutions and modifications of the embodiment still belong to the protection scope of the invention without departing from the principle and spirit of the invention.
Claims (6)
1. À temperature monitoring system of generator carbon brush based on infrared image is characterized in that it includes infrared thermal imager, 4G network, cloud server and temperature monitoring system of carbon brush; The infrared thermal imager is an infrared detector with uncooled focal plane, the pixel value specification is 160 * 120, each infrared image contains 19200 pixels, and the temperature measurement range is - 20° C— 150° C; The 4G network and the cloud server access the data of the infrared thermal imager to the network through the industrial router loaded with 4G IOT card, which wirelessly connect with the cloud server through a fixed IP address and thermal imager number; The temperature monitoring system of carbon brush includes upper computer software and database management software MYSQL developed based on MFC, which can remotely and real-time display the temperature infrared image measured on the industrial site on the industrial computer, also has the functions of over temperature alarm, data recording, image recording, temperature curve drawing, temperature parameter correction and fault diagnosis.
2. A temperature prediction method of generator carbon brush based on infrared image, which is characterized by comprising the following steps: Step 1: Based on the experimental data of the power plant, randomly select the temperature data of the generator carbon brush for a period of time; Step 2: Process the infrared image in the data, use Sobel operator to detect the edge and extract the carbon brush contour in the infrared image, and fill in the missing value by using the corresponding relationship between the temperature data and the infrared image area of the heating carbon brush; Step 3: Use the combination model, give full play to the time series prediction advantages of LSTM, and combine the nonlinear characteristics of BP neural network to construct the LSTM-BP combination model; Step 4: Use LSTM-BP combined model to predict the trend of generator carbon brush temperature, analyze the operation status of carbon brush and the load status of 11 excitation system through the temperature trend, and verify the prediction accuracy 7500849 from MAE, MAPE and RMSE.
3. The temperature prediction method of generator carbon brush based on infrared image according to Claim 2, which is characterized in that the temperature experimental data of generator carbon brush described in Step 1 is the temperature data of generator carbon brush collected by the power plant for 11 consecutive days, a total of 539, in which 485 training sets numtrain account for 90%, and 54 test sets numtest account for 10%.
4. The temperature prediction method of generator carbon brush based on infrared image according to Claim 2, which is characterized in that the contour extraction described in Step 2 requires the following operations: 1) Image binarization; 2) Use bwlabel to find the connected area, 4 connected refers to (if the position of pixels is up, down, left or right adjacent to other pixels, they are considered to be connected); 3) The tag value of bwing image is stored in the L array (that is, it is marked in the L matrix after it is determined to be connected); 4) Select the region and only keep the region of interest; 5) Using region props to calculate the area of the reserved area; After contour extraction, the current image is captured locally in the generator brush temperature online analysis system, through the corresponding relationship between the temperature data and the infrared image area of the heating carbon brush, the missing value can be filled, and a good effect can be achieved.
5. The temperature prediction method of generator carbon brush based on infrared image according to Claim 2, which is characterized in that the LSTM-BP combination model described in step 3 adopts the series combination method to determine the weight of LSTM model and BP model through the method of linear programming, and the parameter settings are as follows: LSTM network model: The input layer and the output layer are one neuron respectively, and the LSTM structural unit of the hidden layer is set as three layers with 60 ‚200 and 60 neurons respectively. The maximum number of iterations is 400. When the error is less than 10 — 5, the loop will jump out.
12
BP network model: The input layer and output layer are single neurons, by 7500849 constantly adjusting the parameter trial and error method, it is determined that the BP neural network is a double hidden layer,there are 8 and 4 neurons respectively, the learning coefficient is 0.01, and the error control rate is 1x10 — 5, the maximum training times is 5000, the training function trainlm, and the test relative error distribution interval [0.01,0.08].
6. The temperature prediction method of generator carbon brush based on infrared image according to Claim 2, which is characterized in that in Step 4, the LSTM-BP combined model is used to predict the temperature trend of generator carbon brush, compare the results with the actually collected temperature curve and the results predicted by LSTM model, and then use the evaluation index to verify the prediction results, the specific steps are as follows: The evaluation indicators used are: 1 & MAE =--D |v | i=l (1); RMSE = X 3) Nia (2); NT (3) ; Where iis the real temperature value of generator carbon brush / °C; Vis the predicted generator brush temperature /°C; N is the number of test sample sets, Simulation experiments show that this method can well predict the temperature trend of generator carbon brush, and most of the prediction errors are stable within 0.3 °C, by predicting the temperature of generator carbon brush and analyzing the operation status of carbon brush and the load status of excitation system through temperature trend, at the same time, the combined model can provide a reasonable reference for related types of data prediction and decision planning.
13
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| LU500843A LU500843B1 (en) | 2021-11-10 | 2021-11-10 | A monitoring system and prediction method of temperature for generator carbon brush based on an infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| LU500843A LU500843B1 (en) | 2021-11-10 | 2021-11-10 | A monitoring system and prediction method of temperature for generator carbon brush based on an infrared image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| LU500843B1 true LU500843B1 (en) | 2022-05-10 |
Family
ID=81607298
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| LU500843A LU500843B1 (en) | 2021-11-10 | 2021-11-10 | A monitoring system and prediction method of temperature for generator carbon brush based on an infrared image |
Country Status (1)
| Country | Link |
|---|---|
| LU (1) | LU500843B1 (en) |
-
2021
- 2021-11-10 LU LU500843A patent/LU500843B1/en active IP Right Grant
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114640173B (en) | Early warning model of transformer and generator based on many characteristic quantities | |
| CN103793853B (en) | Condition of Overhead Transmission Lines Based appraisal procedure based on two-way Bayesian network | |
| CN109781268B (en) | Switch cabinet interior key part temperature monitoring system based on low-cost infrared hot-vision technology | |
| CN106655522A (en) | Master station system suitable for operation and maintenance management of secondary equipment of power grid | |
| CN106777984A (en) | A kind of method that photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm | |
| CN106199305A (en) | Underground coal mine electric power system dry-type transformer insulation health state evaluation method | |
| CN102393929B (en) | State overhauling control method of distribution network equipment state assessment system | |
| CN106384186A (en) | Distributed new energy grid-connected power quality monitoring method and system | |
| CN113836816A (en) | Generator carbon brush temperature monitoring system based on infrared image and temperature prediction method | |
| CN113820017A (en) | Generator carbon brush temperature monitoring system based on infrared image and temperature prediction method | |
| CN108199891B (en) | Cps network attack identification method based on artificial neural network multi-angle comprehensive decision | |
| CN113763667A (en) | Fire early warning and state monitoring device and method based on 5G edge calculation | |
| CN111850609A (en) | Aluminum electrolysis control system based on digital twinning | |
| CN105305391A (en) | Protection action evaluation method | |
| CN103368261A (en) | On-line monitoring and inspection system for transformer substation | |
| CN107153915A (en) | A kind of transformer station does not change the method and system of telemetry automatic detection | |
| CN117614137A (en) | Power distribution network optimization system based on multi-source data fusion | |
| CN116029699A (en) | Power information system based on data twinning and operation and maintenance method | |
| CN118035814A (en) | State fault early warning method and system based on digital twin field data analysis equipment | |
| CN108093210A (en) | A kind of transformer oil level warning system and its alarm method | |
| LU500843B1 (en) | A monitoring system and prediction method of temperature for generator carbon brush based on an infrared image | |
| CN105652157B (en) | Method for analyzing health state of power distribution network based on traveling wave electric quantity | |
| CN107505049A (en) | A kind of transformer station's infrared detection method and system | |
| CN110262313A (en) | Power Material key point remote supervision system based on technology of Internet of things | |
| Zhao et al. | [Retracted] Working Condition Monitoring System of Substation Robot Based on Video Monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FG | Patent granted |
Effective date: 20220510 |