CN117990221A - Automatic bus temperature measurement inspection method and system based on RFID - Google Patents
Automatic bus temperature measurement inspection method and system based on RFID Download PDFInfo
- Publication number
- CN117990221A CN117990221A CN202410404699.1A CN202410404699A CN117990221A CN 117990221 A CN117990221 A CN 117990221A CN 202410404699 A CN202410404699 A CN 202410404699A CN 117990221 A CN117990221 A CN 117990221A
- Authority
- CN
- China
- Prior art keywords
- temperature
- value
- time
- inspection
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 96
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000009529 body temperature measurement Methods 0.000 title claims abstract description 63
- 238000012549 training Methods 0.000 claims abstract description 47
- 238000012360 testing method Methods 0.000 claims abstract description 22
- 230000008859 change Effects 0.000 claims description 39
- 230000011218 segmentation Effects 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 12
- 238000012544 monitoring process Methods 0.000 claims description 11
- 235000019580 granularity Nutrition 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 102100037651 AP-2 complex subunit sigma Human genes 0.000 claims description 6
- 101000806914 Homo sapiens AP-2 complex subunit sigma Proteins 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000007613 environmental effect Effects 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 abstract description 2
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 239000013307 optical fiber Substances 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisions for transferring data to distant stations, e.g. from a sensing device
- G06K17/0029—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisions for transferring data to distant stations, e.g. from a sensing device the arrangement being specially adapted for wireless interrogation of grouped or bundled articles tagged with wireless record carriers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K1/00—Details of thermometers not specially adapted for particular types of thermometer
- G01K1/02—Means for indicating or recording specially adapted for thermometers
- G01K1/024—Means for indicating or recording specially adapted for thermometers for remote indication
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K1/00—Details of thermometers not specially adapted for particular types of thermometer
- G01K1/02—Means for indicating or recording specially adapted for thermometers
- G01K1/026—Means for indicating or recording specially adapted for thermometers arrangements for monitoring a plurality of temperatures, e.g. by multiplexing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K1/00—Details of thermometers not specially adapted for particular types of thermometer
- G01K1/14—Supports; Fastening devices; Arrangements for mounting thermometers in particular locations
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K7/00—Measuring temperature based on the use of electric or magnetic elements directly sensitive to heat ; Power supply therefor, e.g. using thermoelectric elements
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K7/00—Measuring temperature based on the use of electric or magnetic elements directly sensitive to heat ; Power supply therefor, e.g. using thermoelectric elements
- G01K7/42—Circuits effecting compensation of thermal inertia; Circuits for predicting the stationary value of a temperature
- G01K7/427—Temperature calculation based on spatial modeling, e.g. spatial inter- or extrapolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses an automatic bus temperature measurement inspection method and system based on RFID, wherein the method comprises the following steps: passive RFID ultra-high frequency temperature measurement tags are installed at a plurality of designated positions of the bus duct, and the real-time monitored bus duct temperature values are stored in the cloud platform; training the historical temperature of the bus duct stored in the cloud platform by adopting a DeepAR time sequence prediction model to obtain a trained model, predicting a test value of future time granularity by using DARM, adding the temperature measuring points meeting the early warning condition into a routing inspection task list, and using an RFID reader to walk to a signal designated range according to the sequence of the routing inspection task list by a routing inspection personnel to measure the temperature of the current temperature measuring points on the bus duct to obtain a corresponding temperature value. According to the invention, the DeepAR time sequence prediction model is adopted to train the historical temperature of the bus duct stored in the cloud platform in a downlink mode, so that an optimal prediction model is obtained, the temperature of the bus and the reason of high-temperature abnormality can be accurately known in the later temperature prediction, and the prediction accuracy is improved.
Description
Technical Field
The invention relates to the technical field of bus temperature monitoring, in particular to an automatic bus temperature measurement inspection method and system based on RFID.
Background
The bus is an important component device in a power transmission and distribution system, the bus duct is a device for transmitting high current and high power, the running safety of the bus duct determines the overall running safety, the running safety and the running stability of the bus duct are the basis for ensuring the stable running of the power transmission and distribution system, and the bus temperature rise index is the most critical index for judging the running state of the bus duct.
Abnormal heat release is caused by aging, oxidation, or excessive contact resistance. If the detection cannot be timely carried out, busbar fusion welding can be caused, and even serious consequences such as electric fire and the like caused by cabinet explosion can be caused, so that huge casualties and property loss are caused. Therefore, the bus temperature is necessary to be monitored and pre-warned in real time, and potential safety hazards are eliminated in time.
At present, the temperature monitoring of the bus duct mainly comprises the following technologies:
(1) Infrared temperature measurement technology: the integrated circuit is in cable communication, needs cabling and temperature measurement space, cannot realize the integrated integration of bus equipment and temperature on-line detection, and can cause temperature measurement failure on a large-area line if the line breaks down.
(2) The fiber bragg grating temperature measurement technology comprises the following steps: the optical fiber is adopted to transmit signals, the optical fiber is easy to be subjected to surrounding environment and electromagnetic interference, more optical fibers are required to be laid, the optical fiber wiring difficulty is high, the cost is high, and the temperature measurement failure on a large-area circuit can be caused if the circuit breaks down.
(3) Wireless temperature measurement technology: the wireless signal is greatly influenced by the obstacle, so that the signal is easy to attenuate; the wireless temperature measurement can not realize the function of collecting data in real time, and potential safety hazards exist.
All the three methods are hardware control, and data analysis on collected historical data of related buses is not considered, so that fault influence factors of buses cannot be accurately obtained, and effective prediction on the bus data cannot be accurately and timely carried out.
The invention patent document with the publication number of CN108151897B discloses a bus temperature measurement system and a bus temperature measurement method, wherein a temperature monitoring module, a signal transmitting and receiving module and a monitoring end are adopted to process the received temperature, and the processing process is simpler: when the temperature value exceeds the first warning value, a first warning signal is output, and obviously, the simple comparison cannot meet various changes of the bus temperature.
The invention patent document with the publication number of CN 116754097A discloses a bus temperature measurement system capable of automatically distributing addresses, which realizes equipment scanning, automatic address distribution and topology updating through cooperation among equipment, completes full-automatic address distribution, does not need manual setting, avoids the instability of the whole communication connection system caused by input errors, and considers the problem of unstable connection caused by the error of the communication addresses.
The invention patent with the publication number of CN117213650A discloses a bus temperature monitoring system and a bus temperature monitoring method based on UHF RFID, which specifically describe how to paste UHF RFID electronic identification tags on corresponding monitoring positions of buses, but do not give a specific scheme how to store and analyze temperature data of the buses running in real time, so that uninterrupted monitoring of the bus duct is realized, and the burden of operation and maintenance personnel of the bus duct is greatly reduced.
Disclosure of Invention
The invention aims to: in order to overcome the defects in the prior art, the invention provides an automatic bus temperature measurement inspection method based on RFID, which comprises the following steps:
S1, installing passive RFID ultrahigh frequency temperature measurement tags at a plurality of designated positions of a bus duct, and storing real-time monitored bus duct temperature values into a cloud platform, wherein the designated positions are temperature measurement points;
S2, training a bus duct historical temperature value stored in the cloud platform by adopting a DeepAR time sequence prediction model to obtain a trained optimal model, and marking the optimal model as a time sequence prediction model DARM;
S3, setting early warning parameters and inspection interval duration D1 on the cloud platform, predicting test values of 3 time granularity in the future by using the time sequence prediction model DARM, analyzing and processing the early warning parameters and the test values, and adding temperature measuring points meeting early warning conditions into an inspection task list every interval duration D1;
s4, the patrol personnel uses the handheld RFID reader-writer to walk to the designated range of the signal according to the sequence of the patrol task list to perform wireless temperature measurement on the temperature measuring points on the bus duct, so that corresponding temperature values are obtained, and the temperature values are recorded and stored in a cloud database.
Further, the method comprises the steps of: the method further comprises the steps of:
S5, setting a high-temperature alarm threshold R21 and an ultrahigh-temperature alarm threshold R22 on the cloud platform; a temperature change alarm threshold R23 and a temperature change alarm threshold R24; a temperature rise alarm threshold R25, a temperature rise alarm threshold R26, a temperature rise speed alarm threshold R27, and a key object inspection interval D2, wherein D2< D1, R22> R21, R24> R23, and R26> R25;
If the temperature value obtained by current inspection is greater than or equal to R21 or the temperature change value is greater than or equal to R23, yellow warning and short message warning are performed, wherein the temperature change value=the temperature value measured at the current time point of the temperature measuring point-the temperature value measured at the last time point of the temperature measuring point, namely the temperature value of the last temperature measuring time point, because the temperature change value is the temperature difference value between the current time point and the last time point of the same temperature measuring point.
Further, the method comprises the steps of: if the temperature rise value obtained by current inspection is greater than or equal to R24 or the temperature rise value is greater than or equal to R25, orange warning and short message warning are carried out, and the temperature measuring point is added into a key attention list, wherein Wen Shengzhi =temperature value of the temperature measuring point-environmental temperature value;
If the current temperature value is obtained through inspection, the current temperature rise change rate ts is obtained through a temperature rise speed calculation formula, when ts is more than or equal to R27, orange warning is carried out, a short message is given out, and the temperature measuring point is added into a focused attention list.
Further, the method comprises the steps of: if the temperature value obtained by current inspection is greater than or equal to R22 or the temperature rise value is greater than or equal to R26, red early warning and immediate voice alarm are performed, meanwhile, the relay protection device is linked to realize tripping protection and short message alarm, and meanwhile, the current acquired data and alarm event are stored in a cloud database;
and adding the temperature measuring points into the patrol task list at each interval D2 of the temperature measuring points of the focus attention list.
Further, the method comprises the steps of: step S2, including step S21, is to process the historical temperature value of the bus duct stored in the cloud platform to obtain a covariate value of DeepAR time series prediction model, and specifically includes:
S211, selecting an hourly historical temperature value of the bus duct within two years as a training data set, and marking as The temperature measurement point i is shown from time 1 to time/>Temperature value of (2);
s212, the bus duct is arranged from the front of the current moment T The historical temperature value of the hour is taken as a prediction data set and is recorded asWhich represents the time of the temperature measurement point i/>A temperature value up to time T;
S213 takes every 8 hours as a sliding window, and the training is iterated step by step until 7 continuous whole days are reached, namely, the training sample number batch size of 21 training samples is trained;
S214, taking the label position, the equipment type, the label using time length, the maximum temperature value in 24 hours, the average temperature value in 24 hours, the current year, the current month and the current week of the temperature measuring point as covariates, and recording as All covariate values from time 1 to time T for temperature measurement point i are shown.
Further, the method comprises the steps of: the step S2 further includes a step S22: training the historical temperature value of the bus duct stored in the cloud platform by adopting DeepAR time sequence prediction model, and specifically comprising the following steps:
s221, obtaining peak-valley time intervals of temperature change values by using a Fisher optimal segmentation method according to real-time temperature values 24 hours before a predicted time point, and recording the temperature values 24 hours before the predicted time point as I.e. monitoring the temperature value once per minute, the total variance is/>, in the Fisher optimal segmentation method，/>Is the average value of the temperature values of the first 24 hours,；
S222, dividing the temperature data set formed by the temperature values into 2 groups, namely, classifying the temperature data set into 2 groups, and then，
Wherein,Is the temperature mean of the kth class,/>For the ith temperature value in the kth class, recordIs the sum of squares of the deviations in the group;
Definition of the definition ，/>The sum of squares of the deviations within the smallest group, i.e. when/>, with a classification number of 2Minimum time, correspond to
The segmentation method of (2) is an optimal segmentation method; assuming a division into 2 groups, it can be understood that the temperature value of one day is found to be a division point, the smallestThe corresponding division point is the optimal division point.
S223, the two time periods obtained after the optimal segmentation are called load time periods, and the load time periods are used as covariates to be added into DeepAR time sequence prediction models to start training;
If the division method is divided into 3 groups according to the above, namely two optimal division points, 13:00 and 19:30 are obtained, and then the three groups are named as a load rising period, a high load working period and a load falling period.
Setting an expected temperature error target as RMSE, performing training test by utilizing DeepAR time sequence prediction model according to the data, stopping training when the model prediction error does not exceed a preset threshold RMSE, and recording the prediction error value asWhere n is the total number of samples,/>Is the temperature prediction value of the i-th sample,And if the model error value exceeds a preset threshold RMSE, adjusting the training parameters of the neural network model or adding training samples, and retraining the model until the prediction error value meets the preset threshold.
Further, the method comprises the steps of: step S22 further includes:
s224, grouping the temperature data sets into 3 groups, and calculating according to the method of the step S222 to obtain the sum of squares of the deviations in the minimum group with the classification number of 3 ; Then adding the time period obtained after the optimal segmentation as a covariate to a DeepAR model to start training, and calculating according to the method of the step S223 to obtain a prediction error value/>If the model error value exceeds a preset threshold value RMSE, adjusting the training parameters of the neural network model or adding training samples, and retraining the model until the prediction error value meets the preset threshold value;
s225, sequentially dividing the temperature dataset into 4-6 groups, respectively calculating the corresponding minimum intra-group dispersion square sum and the prediction error value, and training the corresponding model to obtain DeepAR models under the corresponding groups.
At this time, S226, the model synthesis target attenuation rate is recorded asWherein, the method comprises the steps of, wherein,，/>，/>Get/>The corresponding classification number k at the maximum is the optimal classification number, and the model corresponding to the optimal classification number is a time sequence prediction model DARM;
S227 updates DARMs with the latest temperature data every half an hour, and updates optimal classification numbers and their DARMs with the latest temperature data every 1 week.
Model training is performed for each class number, and then the calculation is performed for each class numberFinally, find a most reasonable number of classifications as the final number of classifications, namely use/>If the value of (a) is increased, the description model error is reduced, if the value of (a) is unchanged or the description model error is reduced, the value of (b) corresponds to the largest one/>The model and the classification number of (c) are optimal.
Further, the method comprises the steps of: in the step S3, the early warning parameters and the test values are analyzed and processed, and the time length D1 of each interval of the temperature measuring points meeting the conditions is added to the inspection task list, which specifically includes:
The early warning parameters include: the inspection trigger temperature early warning value R11, the inspection trigger temperature deviation early warning value R110, the inspection trigger temperature change early warning value R12 and the inspection trigger temperature rise early warning value R13, wherein the test values comprise: predicting temperature values of 3 time granularities in the future by using a time sequence prediction model DARM, namely, predicting time steps in the future, predicting 3 steps backwards, and corresponding variance values sigma1, sigma2 and sigma3, calculating corresponding temperature change predicted values tv2 and tv3, namely, tv2 = t2-t1, tv3 = t3-t2, and recording the last environmental temperature value as t0, thereby calculating corresponding temperature rise predicted values tr1, tr2 and tr3, namely, subtracting the environmental temperature value from the bus temperature value at the same time point;
The analysis processing process comprises the following steps:
When max (t 1, t2, t 3) is not less than R11 or max (sigma 1, sigma2, sigma 3) is not less than R110 or max (tv 2, tv 3) is not less than R12 or max (tr 1, tr2, tr 3) is not less than R13, adding the current temperature measuring point into the inspection task list, eliminating the red label after the inspection is finished, and adding the corresponding temperature measuring point into the inspection task list at intervals D1.
On the other hand, the invention also provides an automatic bus temperature measurement inspection system based on RFID, which comprises a cloud platform, a cloud database, a processor, a passive RFID ultrahigh frequency temperature measurement tag and a handheld RFID reader-writer, wherein the passive RFID ultrahigh frequency temperature measurement tag is arranged at a designated position of a bus duct, the cloud platform is used for storing a bus duct temperature value monitored by the passive RFID ultrahigh frequency temperature measurement tag in real time, and the designated position is a temperature measurement point;
The processor acquires a historical bus duct temperature value stored in the cloud platform and trains by adopting a DeepAR time sequence prediction model to obtain a trained optimal model, the trained optimal model is recorded as a time sequence prediction model DARM, the DARM is updated by reusing the latest temperature data every half an hour, and the optimal classification number and the DARM are updated by reusing the latest temperature data every 1 week;
Setting early warning parameters and inspection interval duration D1 on the cloud platform, predicting test values of 3 time granularity in the future by using the time sequence prediction model DARM, analyzing and processing the early warning parameters and the test values, and adding temperature measuring points meeting the conditions into an inspection task list at intervals duration D1;
And the patrol personnel uses the handheld RFID reader-writer to walk to the appointed range of the signal according to the sequence of the patrol task list to wirelessly measure the temperature of the current temperature measuring point on the bus duct, so as to obtain a corresponding temperature value, and the temperature value record is stored in a cloud database.
Further, the method comprises the steps of: the system further comprises:
The cloud platform sets a high-temperature alarm threshold R21 and an ultra-high-temperature alarm threshold R22; a temperature change alarm threshold R23 and a temperature change alarm threshold R24; a temperature rise alarm threshold R25, a temperature rise alarm threshold R26, a temperature rise speed alarm threshold R27 and a key object inspection interval D2, and satisfies D2< D1, R22> R21, R24> R23 and R26> R25;
And if the temperature value obtained by current inspection is greater than or equal to R21 or the temperature change value is greater than or equal to R23, yellow warning and short message warning are carried out, wherein the temperature change value=the temperature value measured at the current time point of the temperature measuring point-the temperature value measured at the last time point of the temperature measuring point.
The beneficial effects are that: compared with the prior art, the invention has the following advantages:
(1) According to the invention, the DeepAR time sequence prediction model is adopted to train the historical temperature of the bus duct stored in the cloud platform in a downlink mode to obtain the optimal DeepAR time sequence prediction model, so that the temperature of the bus and the reason of high-temperature abnormality can be accurately known in the later temperature prediction, and the prediction accuracy is improved;
(2) According to the invention, the acquired temperature dataset is subjected to load time-sharing optimal division, and an optimal DeepAR time sequence prediction model is obtained according to grouping, so that the model is increased with covariates, and therefore, the characteristic rule can be trained, the model can be correspondingly trained according to the temperature change trend of the bus duct, and the temperature characteristic of the bus duct is more met, so that the prediction accuracy of the trained model is higher.
(3) The invention adopts an RFID matched temperature measuring tag and a mode of using a handheld RFID reader-writer in a matching way to provide bus duct safety service based on robot/manual inspection, and the bus duct safety service comprises the steps of information inquiry of a bus duct outside a sight distance, numerical value and abnormal alarm of the bus duct temperature, automatic generation of temperature inspection records and the like. The invention overcomes the condition limitations of no data, no record, invisible shielding, difficult maintenance and the like faced by the traditional temperature sensing paste temperature measurement mode, and ensures the stable operation of bus duct equipment with low cost.
Drawings
FIG. 1 is a flow chart of an automatic inspection method for measuring bus temperature based on RFID according to an embodiment of the invention;
FIG. 2 is a flow chart of a method for model training according to an embodiment of the invention;
FIG. 3 is a flow chart of a method for performing hierarchical early warning on bus temperature according to another embodiment of the invention;
FIG. 4 is a flowchart of an automatic inspection method for measuring temperature of a bus based on RFID according to another embodiment of the present invention;
FIG. 5 is a flow chart of an automatic inspection method for measuring temperature of a bus based on RFID according to another embodiment of the invention;
Fig. 6 is a graph showing actual temperature variation trend of a bus duct according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, the invention provides an automatic inspection method for measuring bus temperature based on RFID, which comprises the following steps:
s1, installing passive RFID ultrahigh frequency temperature measurement tags at a plurality of designated positions of the bus duct, and storing the bus duct temperature values monitored in real time into a cloud platform, wherein the designated positions are temperature measurement points.
The passive RFID ultrahigh frequency temperature measurement tag consists of a coupling element and a chip, and each tag has unique electronic codes and has the functions of temperature measurement, identification, RFID communication and the like. In the embodiment, the temperature measuring label matched with the DQC-BSMX100 can bear the high temperature exceeding 220 ℃, and the specific parameters are as follows:
the temperature measurement mode is as follows: UHF RFID passive wireless;
temperature range: -40-150 ℃;
Temperature error: (+ -) (standard reading X1% +1 ℃);
resolution ratio: less than 0.1 ℃;
measurement repeatability Requirement (RSD): less than 1%;
reading distance: not less than 7m (at room temperature);
sampling period: less than or equal to 20ms (at room temperature);
Working frequency band: 902MHz to 928MHz;
service life is as follows: >10 years (at room temperature);
Temperature resistance: -40-200 ℃;
Protection grade: IP66.
The following steps S2 and S3 are model training and temperature prediction steps, and step S4 is to compare actual temperature measurement of each temperature measurement point with the model predicted temperature according to the abnormal temperature measurement points obtained by the inspection personnel, and update the model again to improve the accuracy of the model, and on the other hand, the inspection personnel can also inspect the site situation to prevent accidents.
S2, training the historical temperature value of the bus duct stored in the cloud platform by adopting a DeepAR time sequence prediction model to obtain a trained optimal model, and marking the optimal model as a time sequence prediction model DARM.
As shown in fig. 2, step S2 includes:
step S21: processing the historical temperature value of the bus duct stored in the cloud platform to obtain a covariate value of DeepAR time sequence prediction model, specifically comprising:
S211, selecting an hourly historical temperature value of the bus duct within two years as a training data set, and marking as The temperature measurement point i is shown from time 1 to time/>Temperature value of (2); for example, from 1 st 2022 to 31 nd 2023, 12 th.
S212, the bus duct is arranged from the front of the current moment TThe historical temperature value of the hour is taken as a prediction data set and is recorded asWhich represents the time of the temperature measurement point i/>A temperature value up to time T;
s213 takes every 8 hours as a sliding window, and training is iterated step by step until 7 continuous whole days are reached, namely 21 batch size are trained altogether;
In this embodiment, the historical temperature data of the bus duct from the current (e.g., 2024, 1) for the first 8 hours is recorded as a predictive data set The time/>, of the temperature measurement point i is shownA temperature value up to time T; and training is iterated stepwise every 8 time periods as a sliding window, until 24 of 7 consecutive whole days (e.g. 2024 1 month 7 days): 00, i.e. only 8 time steps are predicted, unit: for an hour, a total of 21 batch sizes were trained. The missing values are filled with the temperature average of the three periods before and after the missing values.
S214, taking the label position of the temperature measuring point, the equipment type, the label using time length (unit: quarter), the maximum temperature value in 24 hours, the average temperature value in 24 hours, the current year, the current month and the current week (working day and non-working day) as covariates, and recording asAll covariate values from time 1 to time T for temperature measurement point i are shown.
Step S22: simultaneously training temperature time sequences of all temperature measuring points by using a Fisher optimal segmentation method and DeepAR time sequence prediction models, and specifically comprising the following steps:
s221, obtaining peak-valley time intervals of temperature change values by using a Fisher optimal segmentation method according to real-time temperature values 24 hours before a predicted time point, and recording the temperature values 24 hours before the predicted time point as I.e. monitoring the temperature value once per minute, the total variance is/>, in the Fisher optimal segmentation method，/>Is the average value of the temperature values of the first 24 hours,；
S222, dividing the temperature data set formed by the temperature values into 2 groups, namely, classifying the temperature data set into 2 groups, and thenWherein/>Is the temperature mean of the kth class,/>For the ith temperature value in the kth class, recordIs the sum of squares of the deviations in the group;
Definition of the definition ，/>The sum of squares of the deviations within the smallest group, i.e. when/>, with a classification number of 2At the minimum, the corresponding segmentation method is the optimal segmentation method; by minimizing/>And obtaining an optimal division method, so that the difference among similar samples is minimum, and the difference among samples of each category is maximum.
S223, the time period obtained after the optimal segmentation is called a load time period, and the load time period is used as a covariate to be added into DeepAR time sequence prediction model to start training; for example, a load-up period (8:00-19:00), a load-down period (19:30-8:00).
Setting an expected temperature error target as RMSE, performing training test by utilizing DeepAR time sequence prediction model according to the data, stopping training when the model prediction error does not exceed a preset threshold RMSE, and recording the prediction error value asWhere n is the total number of samples,/>Is the temperature predictive value of the ith sample,/>And if the model error value exceeds a preset threshold RMSE, adjusting the training parameters of the neural network model or adding training samples, and retraining the model until the prediction error value meets the preset threshold.
S224, dividing the temperature data set into 3 groups, namely classifying the temperature data set into 3 groups, and calculating the least squares sum of the deviations in the groups with the classifying number of 3 according to the method of the step S222; Then adding the time period obtained after the optimal segmentation as a covariate to a DeepAR model to start training, and calculating according to the method of the step S223 to obtain a prediction error value/>; Training the model until the prediction error value meets a preset threshold value;
The optimal division is divided into 3 groups according to the above-mentioned dividing method, namely two optimal dividing points, and the obtained 13:00 and 19:30 are named as load rising period, high load working period and load falling period.
S225, sequentially dividing the temperature data set into 4-6 groups, and respectively calculating the corresponding minimum intra-group dispersion square sum and the prediction error value; and training the model until the prediction error value meets a preset threshold value.
At this time, S226, the model synthesis target attenuation rate is recorded asWherein, the method comprises the steps of, wherein,，/>，/>Get/>The maximum corresponding classification number k is the optimal classification number, and the model corresponding to the optimal classification number is a time sequence prediction model DARM.
S227 updates DARMs with the latest temperature data every half an hour, and updates the optimal classification count and its DARMs with the latest temperature data every week.
As shown in fig. 6, which is a graph of actual temperature trend of the bus duct in a certain month, the dividing time period can be 4 segments, and the dividing time periods are respectively as follows: (00:00-8:00), (8:00-13:00), (13:00-19:00), (19:00-00:00) and according to actual temperature, most of the segmentation time periods cannot exceed 6 and are larger than 2, so that the classification range set by the method is 2-6, the classification aim is that the time sequence prediction model is more in accordance with the characteristics of the bus duct temperature, the rule with the optimal segmentation point can be obtained through training, namely, the model is increased by covariates, the characteristic rule can be trained, and the bus duct temperature is predicted by the model more accurately.
S3, setting early warning parameters and inspection interval duration D1 on the cloud platform, predicting test values of 3 time granularity in the future by using the time sequence prediction model DARM, analyzing and processing the early warning parameters and the test values, and adding temperature measuring points meeting the conditions into an inspection task list every interval duration D1.
In step S3, analyzing and processing the early warning parameter and the test value, and adding the time length D1 of each interval of the temperature measuring points meeting the condition to the inspection task list, which specifically includes:
The early warning parameters include: the inspection trigger temperature early warning value R11, the inspection trigger temperature deviation early warning value R110, the inspection trigger temperature change early warning value R12 and the inspection trigger temperature rise early warning value R13, wherein the test values comprise: predicting temperature values of 3 time granularities in the future by using a time sequence prediction model DARM, recording the temperature values as t1, t2 and t3 and corresponding variance values sigma1, sigma2 and sigma3, calculating corresponding temperature change predicted values tv2 and tv3, recording the last environmental temperature value as t0, and calculating corresponding temperature rise predicted values tr1, tr2 and tr3;
The analysis processing process comprises the following steps:
When max (t 1, t2, t 3) is not less than R11 or max (sigma 1, sigma2, sigma 3) is not less than R110 or max (tv 2, tv 3) is not less than R12 or max (tr 1, tr2, tr 3) is not less than R13, adding the current temperature measuring point into the inspection task list, eliminating the red label after the inspection is finished, and adding the corresponding temperature measuring point into the inspection task list at intervals D1.
S4, the patrol personnel uses a handheld RFID reader-writer to scan the terminal against the RFID temperature measuring label of the room temperature measuring point (such as a factory building gate) to obtain an environment temperature value, then walks to the signal appointed range according to the sequence of the patrol task list to carry out wireless temperature measurement on the temperature measuring points on the bus duct one by one, so that a corresponding temperature value is obtained, and the temperature value record is stored in a cloud database.
Wherein, handheld RFID reader: the RFID tag wireless reading function is provided, and all parameters are shown in the following table:
After the handheld RFID reader-writer obtains the bus duct temperature value returned by the RFID temperature measurement tag, the data record is stored in the cloud database.
Further, in another embodiment of the present invention, in order to further improve the prediction accuracy, the method further includes performing a hierarchical early warning on the bus temperature, as shown in fig. 3, the method includes the following steps:
S5, setting a high-temperature alarm threshold R21 and an ultrahigh-temperature alarm threshold R22 on the cloud platform; a temperature change alarm threshold R23 and a temperature change alarm threshold R24; a temperature rise alarm threshold R25, a temperature rise alarm threshold R26, a temperature rise speed alarm threshold R27, and a key object inspection interval D2, wherein D2< D1, R22> R21, R24> R23, and R26> R25;
And if the temperature value obtained by current inspection is greater than or equal to R21 or the temperature change value is greater than or equal to R23, yellow warning and short message warning are carried out, wherein the temperature change value=the temperature value measured at the current time point of the temperature measuring point-the temperature value measured at the last time point of the temperature measuring point.
If the temperature rise value obtained by current inspection is greater than or equal to R24 or the temperature rise value is greater than or equal to R25, orange warning and short message warning are carried out, and the temperature measuring point is added into a key attention list, wherein Wen Shengzhi =temperature value of the temperature measuring point-environmental temperature value;
If the current temperature value is obtained through inspection, the current temperature rise change rate ts is obtained through a temperature rise speed calculation formula, when ts is more than or equal to R27, orange warning is carried out, a short message is given out, and the temperature measuring point is added into a focused attention list.
If the temperature value obtained by current inspection is greater than or equal to R22 or the temperature rise value is greater than or equal to R26, red early warning and immediate voice alarm are performed, meanwhile, the relay protection device is linked to realize tripping protection and short message alarm, and meanwhile, the current acquired data and alarm event are stored in a cloud database;
and adding the temperature measuring points into the patrol task list at each interval D2 of the temperature measuring points of the focus attention list.
As shown in fig. 4, the inspection method according to another embodiment of the present invention further includes:
And S6, the cloud platform uses the temperature data and the alarm record stored in the inspection to carry out data statistics analysis and graph trend analysis and comparison. The method mainly provides services such as temperature early warning/alarming condition setting, inspection record inquiring, real-time temperature displaying, historical temperature comparing and curve analyzing, overtemperature alarming, temperature rise alarming, alarming confirmation, important attention list displaying, alarming record inquiring and the like, and supports displaying the state of a temperature measuring point in a bus model mode; the temperature measurement point details are used for checking temperature values, wen Shengzhi and the like.
As shown in fig. 5, the inspection method according to another embodiment of the present invention further includes:
S7, optionally, when no patrol personnel of the factory building take care of, an automatic patrol robot (with an RFID reader) can be arranged for patrol, the robot moves to an automatic matching label in a specified range in sequence according to a patrol task list, stays for 10 seconds, leaves until temperature data are read, acquires the temperature data, and uploads and stores the temperature data to a cloud platform. And meanwhile, the alarm related calculation rule is utilized to automatically alarm, so that the steps S5-S6 are completed.
On the other hand, the invention also provides an automatic bus temperature measurement inspection system based on RFID, which comprises a cloud platform, a cloud database, a processor, a passive RFID ultrahigh frequency temperature measurement tag and a handheld RFID reader-writer, wherein the passive RFID ultrahigh frequency temperature measurement tag is arranged at a designated position of a bus duct, the cloud platform is used for storing a bus duct temperature value monitored by the passive RFID ultrahigh frequency temperature measurement tag in real time, and the designated position is a temperature measuring point;
the processor acquires a historical bus duct temperature value stored in the cloud platform, trains by adopting a DeepAR time sequence prediction model, obtains a trained optimal model, and marks the optimal model as a time sequence prediction model DARM;
Setting early warning parameters and inspection interval duration D1 on the cloud platform, predicting test values of 3 time granularity in the future by using the time sequence prediction model DARM, analyzing and processing the early warning parameters and the test values, and adding temperature measuring points meeting the conditions into an inspection task list at intervals duration D1;
And the patrol personnel uses the handheld RFID reader-writer to walk to the appointed range of the signal according to the sequence of the patrol task list to wirelessly measure the temperature of the current temperature measuring point on the bus duct, so as to obtain a corresponding temperature value, and the temperature value record is stored in a cloud database.
The cloud platform is also provided with a high-temperature alarm threshold R21 and an ultra-high-temperature alarm threshold R22; a temperature change alarm threshold R23 and a temperature change alarm threshold R24; a temperature rise alarm threshold R25, a temperature rise alarm threshold R26, a temperature rise speed alarm threshold R27, and a key object inspection interval D2, wherein D2< D1, R22> R21, R24> R23, and R26> R25;
And if the temperature value obtained by current inspection is greater than or equal to R21 or the temperature change value is greater than or equal to R23, yellow warning and short message warning are carried out, wherein the temperature change value=the temperature value measured at the current time point of the temperature measuring point-the temperature value measured at the last time point of the temperature measuring point.
If the temperature rise value obtained by current inspection is greater than or equal to R24 or the temperature rise value is greater than or equal to R25, orange warning and short message warning are carried out, and the temperature measuring point is added into a key attention list, wherein Wen Shengzhi =temperature value of the temperature measuring point-environmental temperature value;
If the current temperature value is obtained through inspection, the current temperature rise change rate ts is obtained through a temperature rise speed calculation formula, when ts is more than or equal to R27, orange warning is carried out, a short message is given out, and the temperature measuring point is added into a focused attention list.
If the temperature value obtained by current inspection is greater than or equal to R22 or the temperature rise value is greater than or equal to R26, red early warning and immediate voice alarm are performed, meanwhile, the relay protection device is linked to realize tripping protection and short message alarm, and meanwhile, the current acquired data and alarm event are stored in a cloud database;
and adding the temperature measuring points into the patrol task list at each interval D2 of the temperature measuring points of the focus attention list.
And the cloud platform performs data statistical analysis and graph trend analysis and comparison by using the temperature data and the alarm record which are stored in the inspection. The method mainly provides services such as temperature early warning/alarming condition setting, inspection record inquiring, real-time temperature displaying, historical temperature comparing and curve analyzing, overtemperature alarming, temperature rise alarming, alarming confirmation, important attention list displaying, alarming record inquiring and the like, and supports displaying the state of a temperature measuring point in a bus model mode; the temperature measurement point details are used for checking temperature values, wen Shengzhi and the like.
Optionally, when no personnel of patrolling and examining in the factory building cares, can set up and patrol and examine the robot automatically, this patrol and examine the robot and have RFID read write line, and the robot is according to patrol and examine task list and remove automatic matching label in the appointed within range in proper order, stay 10 seconds, only leave until reading the temperature data, acquire temperature data and upload and store to the high in the clouds. And simultaneously, the alarm is automatically given out by utilizing the alarm correlation calculation rule.
Other technical features of the automatic bus temperature measurement inspection system based on RFID are similar to those of the automatic bus temperature measurement inspection method based on RFID, and are not described in detail herein.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims and the equivalents thereof, the present invention is also intended to include such modifications and variations.
Claims (10)
1. An automatic bus temperature measurement inspection method based on RFID is characterized by comprising the following steps: the method comprises the following steps:
S1, installing passive RFID ultrahigh frequency temperature measurement tags at a plurality of designated positions of a bus duct, and storing real-time monitored bus duct temperature values into a cloud platform, wherein the designated positions are temperature measurement points;
S2, training a bus duct historical temperature value stored in the cloud platform by adopting a DeepAR time sequence prediction model to obtain a trained optimal model, and marking the optimal model as a time sequence prediction model DARM;
S3, setting early warning parameters and inspection interval duration D1 on the cloud platform, predicting test values of 3 time granularity in the future by using the time sequence prediction model DARM, analyzing and processing the early warning parameters and the test values, and adding temperature measuring points meeting early warning conditions into an inspection task list every interval duration D1;
s4, the patrol personnel uses the handheld RFID reader-writer to walk to the designated range of the signal according to the sequence of the patrol task list to perform wireless temperature measurement on the temperature measuring points on the bus duct, so that corresponding temperature values are obtained, and the temperature values are recorded and stored in a cloud database.
2. The automatic inspection method for measuring the temperature of a bus based on RFID according to claim 1, wherein the method comprises the following steps: the method further comprises the steps of:
S5, setting a high-temperature alarm threshold R21 and an ultrahigh-temperature alarm threshold R22 on the cloud platform; a temperature change alarm threshold R23 and a temperature change alarm threshold R24; a temperature rise alarm threshold R25, a temperature rise alarm threshold R26, a temperature rise speed alarm threshold R27, and a key object inspection interval D2, wherein D2< D1, R22> R21, R24> R23, and R26> R25;
And if the temperature value obtained by current inspection is greater than or equal to R21 or the temperature change value is greater than or equal to R23, yellow warning and short message warning are carried out, wherein the temperature change value=the temperature value measured at the current time point of the temperature measuring point-the temperature value measured at the last time point of the temperature measuring point.
3. The automatic inspection method for measuring the temperature of the bus based on the RFID according to claim 2, wherein the method comprises the following steps: if the temperature rise value obtained by current inspection is greater than or equal to R24 or the temperature rise value is greater than or equal to R25, orange warning and short message warning are carried out, and the temperature measuring point is added into a key attention list, wherein Wen Shengzhi =temperature value of the temperature measuring point-environmental temperature value;
If the current temperature value is obtained through inspection, the current temperature rise change rate ts is obtained through a temperature rise speed calculation formula, when ts is more than or equal to R27, orange warning is carried out, a short message is given out, and the temperature measuring point is added into a focused attention list.
4. The automatic inspection method for measuring the temperature of a bus based on RFID according to claim 3, wherein the method comprises the following steps: if the temperature value obtained by current inspection is greater than or equal to R22 or the temperature rise value is greater than or equal to R26, red early warning and immediate voice alarm are performed, meanwhile, the relay protection device is linked to realize tripping protection and short message alarm, and meanwhile, the current collected data and alarm event are stored in a cloud database;
and adding the temperature measuring points into the patrol task list at each interval D2 of the temperature measuring points of the focus attention list.
5. The automatic inspection method for measuring temperature of bus bar based on RFID according to any one of claims 1-4, wherein: the step S2 includes a step S21: processing the historical temperature value of the bus duct stored in the cloud platform to obtain a covariate value of DeepAR time sequence prediction model, specifically comprising:
S211, selecting an hourly historical temperature value of the bus duct within two years as a training data set, and marking as The temperature measurement point i is shown from time 1 to time/>Temperature value of (2); s212, bus duct is formed from front/>, of current moment TThe historical temperature value of an hour is taken as a predicted data set and is recorded as/>Which represents the time of the temperature measurement point i/>A temperature value up to time T;
S213 takes every 8 hours as a sliding window, and the training is iterated step by step until 7 continuous whole days are reached, namely, the number of training samples is 21 for the first time;
S214, taking the label position, the equipment type, the label using time length, the maximum temperature value in 24 hours, the average temperature value in 24 hours, the current year, the current month and the current week of the temperature measuring point as covariates, and recording as All covariate values from time 1 to time T for temperature measurement point i are shown.
6. The automatic inspection method for measuring the temperature of the bus based on the RFID according to claim 5, wherein the method comprises the following steps: the step S2 further includes a step S22: training the historical temperature value of the bus duct stored in the cloud platform by adopting DeepAR time sequence prediction model, and specifically comprising the following steps:
s221, obtaining peak-valley time intervals of temperature change values by using a Fisher optimal segmentation method according to real-time temperature values 24 hours before a predicted time point, and recording the temperature values 24 hours before the predicted time point as I.e. monitoring the temperature value once per minute, the total variance is/>, in the Fisher optimal segmentation method，/>Is the average value of the temperature values of the first 24 hours,；
S222, dividing the temperature data set formed by the temperature values into 2 groups, namely, classifying the temperature data set into 2 groups, and then; Wherein,Is the temperature mean of the kth class,/>For the ith temperature value in the kth class, recordIs the sum of squares of the deviations in the group;
Definition of the definition Wherein/>The sum of squares of the deviations within the smallest group, i.e. when/>, with a classification number of 2At the minimum, the corresponding segmentation method is the optimal segmentation method;
s223, the two time periods obtained after the optimal segmentation are called load time periods, and the load time periods are used as covariates to be added into DeepAR time sequence prediction models to start training;
setting an expected temperature error target as RMSE, performing training test by utilizing DeepAR time sequence prediction model according to the data, stopping training when the model prediction error does not exceed a preset threshold RMSE, and recording the prediction error value as Where n is the total number of samples,/>Is the temperature predictive value of the ith sample,/>And if the model error value exceeds a preset threshold RMSE, adjusting the training parameters of the neural network model or increasing the number of training samples, and retraining the model until the prediction error value meets the preset threshold.
7. The automatic inspection method for measuring the temperature of the bus based on the RFID according to claim 6, wherein the method comprises the following steps: step S22 further includes:
s224, grouping the temperature data sets into 3 groups, and calculating according to the method of the step S222 to obtain the sum of squares of the deviations in the minimum group with the classification number of 3 ; Then adding the time period obtained after the optimal segmentation as a covariate to a DeepAR model to start training, and calculating according to the method of the step S223 to obtain a prediction error value/>If the model error value exceeds a preset threshold value RMSE, adjusting the training parameters of the neural network model or adding training samples, and retraining the model until the prediction error value meets the preset threshold value;
S225, sequentially dividing the temperature dataset into 4-6 groups, respectively calculating the corresponding minimum intra-group dispersion square sum and the prediction error value, and training a corresponding model to obtain DeepAR models under the corresponding groups;
At this time, S226, the model synthesis target attenuation rate is recorded as Wherein, the method comprises the steps of, wherein,，/>，/>Get/>The corresponding classification number k at the maximum is the optimal classification number, and the model corresponding to the optimal classification number is a time sequence prediction model DARM;
S227 updates DARMs with the latest temperature data every half an hour, and updates the optimal classification count and its DARMs with the latest temperature data every week.
8. The automatic inspection method for measuring temperature of bus bar based on RFID according to any one of claims 1-4, wherein: in the step S3, the early warning parameters and the test values are analyzed and processed, and the time length D1 of each interval of the temperature measuring points meeting the early warning conditions is added to the inspection task list, which specifically includes:
The early warning parameters include: the inspection trigger temperature early warning value R11, the inspection trigger temperature deviation early warning value R110, the inspection trigger temperature change early warning value R12 and the inspection trigger temperature rise early warning value R13, wherein the test values comprise: predicting temperature values of 3 time granularities in the future by using a time sequence prediction model DARM, recording the temperature values as t1, t2 and t3 and corresponding variance values sigma1, sigma2 and sigma3, calculating corresponding temperature change predicted values tv2 and tv3, recording the last environmental temperature value as t0, and calculating corresponding temperature rise predicted values tr1, tr2 and tr3, namely subtracting the environmental temperature value from the bus temperature value at the same time point;
wherein the analysis process comprises:
When max (t 1, t2, t 3) is not less than R11 or max (sigma 1, sigma2, sigma 3) is not less than R110 or max (tv 2, tv 3) is not less than R12 or max (tr 1, tr2, tr 3) is not less than R13, adding the current temperature measuring point into the inspection task list, eliminating the red label after the inspection is finished, and adding the corresponding temperature measuring point into the inspection task list at intervals D1.
9. An automatic bus temperature measurement inspection system based on RFID, which is characterized in that: the system comprises a cloud platform, a cloud database, a processor, a passive RFID ultra-high frequency temperature measurement tag and a handheld RFID reader-writer, wherein the passive RFID ultra-high frequency temperature measurement tag is arranged at a designated position of a bus duct, the cloud platform is used for storing a bus duct temperature value monitored by the passive RFID ultra-high frequency temperature measurement tag in real time, and the designated position is a temperature measurement point;
the processor acquires a historical bus duct temperature value stored in the cloud platform, trains by adopting a DeepAR time sequence prediction model, obtains a trained optimal model, and marks the optimal model as a time sequence prediction model DARM;
Setting early warning parameters and inspection interval duration D1 on the cloud platform, predicting test values of 3 time granularity in the future by using the time sequence prediction model DARM, analyzing and processing the early warning parameters and the test values, and adding the temperature measuring points meeting early warning conditions into an inspection task list at intervals duration D1;
And the patrol personnel uses the handheld RFID reader-writer to walk to the signal appointed range according to the sequence of the patrol task list to wirelessly measure the temperature of the temperature measuring points on the bus duct, so as to obtain the corresponding temperature value, and the temperature value is recorded and stored in a cloud database.
10. The RFID-based bus bar temperature measurement automatic inspection system of claim 9, wherein: the system further comprises:
The cloud platform sets a high-temperature alarm threshold R21 and an ultra-high-temperature alarm threshold R22; a temperature change alarm threshold R23 and a temperature change alarm threshold R24; a temperature rise alarm threshold R25, a temperature rise alarm threshold R26, a temperature rise speed alarm threshold R27 and a key object inspection interval D2, and satisfies D2< D1, R22> R21, R24> R23 and R26> R25;
And if the temperature value obtained by current inspection is greater than or equal to R21 or the temperature change value is greater than or equal to R23, yellow warning and short message warning are carried out, wherein the temperature change value=the temperature value measured at the current time point of the temperature measuring point-the temperature value measured at the last time point of the temperature measuring point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410404699.1A CN117990221B (en) | 2024-04-07 | 2024-04-07 | Automatic bus temperature measurement inspection method and system based on RFID |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410404699.1A CN117990221B (en) | 2024-04-07 | 2024-04-07 | Automatic bus temperature measurement inspection method and system based on RFID |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117990221A true CN117990221A (en) | 2024-05-07 |
| CN117990221B CN117990221B (en) | 2024-07-16 |
Family
ID=90896240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410404699.1A Active CN117990221B (en) | 2024-04-07 | 2024-04-07 | Automatic bus temperature measurement inspection method and system based on RFID |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117990221B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108151897A (en) * | 2017-12-29 | 2018-06-12 | 南通市盈思仪器科技有限公司 | Busbar temp measuring system and temp measuring method |
| US20210064689A1 (en) * | 2019-08-27 | 2021-03-04 | Nec Laboratories America, Inc. | Unsupervised multivariate time series trend detection for group behavior analysis |
| CN115361525A (en) * | 2022-08-15 | 2022-11-18 | 浙江省一建建设集团有限公司 | Monitoring system and method for inspection operation of overhead hoisting machinery |
| CN116307507A (en) * | 2023-02-01 | 2023-06-23 | 华能国际电力股份有限公司上海石洞口第二电厂 | Multi-target particle swarm coal blending method based on superheater wall temperature prediction |
| CN116698227A (en) * | 2023-07-13 | 2023-09-05 | 中冶南方工程技术有限公司 | Online monitoring and early warning method and system for temperature of extrusion roll shafts of pickling process section |
| CN116754097A (en) * | 2023-06-14 | 2023-09-15 | 南京大全自动化科技有限公司 | Automatic bus temperature measurement system of distribution address |
| CN117213650A (en) * | 2023-08-31 | 2023-12-12 | 镇江西门子母线有限公司 | Bus temperature monitoring system and method based on UHF RFID |
| CN117609861A (en) * | 2023-11-29 | 2024-02-27 | 国网上海市电力公司 | BP neural network algorithm and temperature measurement system capable of being used for temperature prediction of electrical equipment of transformer substation |
| CN117761416A (en) * | 2023-07-27 | 2024-03-26 | 南京信息职业技术学院 | Bus duct on-line monitoring system |
-
2024
- 2024-04-07 CN CN202410404699.1A patent/CN117990221B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108151897A (en) * | 2017-12-29 | 2018-06-12 | 南通市盈思仪器科技有限公司 | Busbar temp measuring system and temp measuring method |
| US20210064689A1 (en) * | 2019-08-27 | 2021-03-04 | Nec Laboratories America, Inc. | Unsupervised multivariate time series trend detection for group behavior analysis |
| CN115361525A (en) * | 2022-08-15 | 2022-11-18 | 浙江省一建建设集团有限公司 | Monitoring system and method for inspection operation of overhead hoisting machinery |
| CN116307507A (en) * | 2023-02-01 | 2023-06-23 | 华能国际电力股份有限公司上海石洞口第二电厂 | Multi-target particle swarm coal blending method based on superheater wall temperature prediction |
| CN116754097A (en) * | 2023-06-14 | 2023-09-15 | 南京大全自动化科技有限公司 | Automatic bus temperature measurement system of distribution address |
| CN116698227A (en) * | 2023-07-13 | 2023-09-05 | 中冶南方工程技术有限公司 | Online monitoring and early warning method and system for temperature of extrusion roll shafts of pickling process section |
| CN117761416A (en) * | 2023-07-27 | 2024-03-26 | 南京信息职业技术学院 | Bus duct on-line monitoring system |
| CN117213650A (en) * | 2023-08-31 | 2023-12-12 | 镇江西门子母线有限公司 | Bus temperature monitoring system and method based on UHF RFID |
| CN117609861A (en) * | 2023-11-29 | 2024-02-27 | 国网上海市电力公司 | BP neural network algorithm and temperature measurement system capable of being used for temperature prediction of electrical equipment of transformer substation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117990221B (en) | 2024-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115329812B (en) | Bridge infrastructure anomaly monitoring method based on artificial intelligence | |
| CN105261079A (en) | Portable intelligent electric power inspection device | |
| CN115395646B (en) | Intelligent operation and maintenance system of digital twin traction substation | |
| CN113763667B (en) | Fire disaster early warning and state monitoring device and method based on 5G edge calculation | |
| CN110690699B (en) | Transformer substation intelligent detection system based on ubiquitous power Internet of things | |
| CN115060312A (en) | Building material safety monitoring system based on artificial intelligence | |
| CN116755411A (en) | Industrial equipment data acquisition method and device | |
| CN116896166A (en) | Digital twin panoramic management and control system and method for urban cable | |
| CN108093210A (en) | A kind of transformer oil level warning system and its alarm method | |
| CN117990221B (en) | Automatic bus temperature measurement inspection method and system based on RFID | |
| CN116754022B (en) | On-line detection emergency early warning method and system for cable tunnel | |
| CN117791876A (en) | Substation equipment operation state monitoring method and abnormal control system | |
| CN116562672A (en) | Inspection work quality evaluation method and system | |
| CN115361061A (en) | Optical fiber fault monitoring method | |
| CN114063601A (en) | Equipment state diagnosis system and method based on artificial intelligence | |
| CN117932278B (en) | Smart city environment-friendly monitoring system and method | |
| CN118536817B (en) | Cable channel monitoring system based on sensor network | |
| CN116896167B (en) | Power transmission line dynamic capacity-increasing monitoring and early warning method based on artificial intelligence | |
| CN118175538B (en) | Industrial sewage treatment intelligent remote control system based on wireless transmission | |
| CN111738382B (en) | Asset management method and asset management system based on RFID | |
| CN116773238B (en) | Fault monitoring method and system based on industrial data | |
| Kim et al. | A study on fire prediction method using air quality measurement sensors of smart indoor parking lot | |
| CN117053860A (en) | Comprehensive intelligent online monitoring method for high-voltage motor | |
| CN118494571A (en) | Fault risk prediction system for curtain door of train platform | |
| CN118095751A (en) | Internet-based mining enterprise production state supervision system |
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 |