CN113221435A - Sensor screening method and device and sensor data reconstruction method and system - Google Patents
Sensor screening method and device and sensor data reconstruction method and system Download PDFInfo
- Publication number
- CN113221435A CN113221435A CN202110183121.4A CN202110183121A CN113221435A CN 113221435 A CN113221435 A CN 113221435A CN 202110183121 A CN202110183121 A CN 202110183121A CN 113221435 A CN113221435 A CN 113221435A
- Authority
- CN
- China
- Prior art keywords
- sensor
- failure
- sensors
- screening
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012216 screening Methods 0.000 title claims abstract description 159
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005516 engineering process Methods 0.000 claims abstract description 36
- 238000012544 monitoring process Methods 0.000 claims description 41
- 230000006870 function Effects 0.000 claims description 25
- 238000013507 mapping Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 14
- 230000004044 response Effects 0.000 claims description 12
- 238000005259 measurement Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000011084 recovery Methods 0.000 claims description 8
- 239000007788 liquid Substances 0.000 claims description 7
- 230000004907 flux Effects 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 description 8
- 238000013145 classification model Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 239000002826 coolant Substances 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 230000008054 signal transmission Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000035899 viability Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000012896 Statistical algorithm Methods 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- KGBXLFKZBHKPEV-UHFFFAOYSA-N boric acid Chemical compound OB(O)O KGBXLFKZBHKPEV-UHFFFAOYSA-N 0.000 description 1
- 239000004327 boric acid Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000010230 functional analysis Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Analysis (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Software Systems (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Operations Research (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a sensor screening method based on a sensor data reconstruction technology, which comprises the following steps: and screening out a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensors, and/or the importance of the sensors. Further, a screening device of the sensor based on the sensor data reconstruction technology, a sensor data reconstruction method and a system are also provided. The method can screen out a reasonable sensor application range which needs data reconstruction, so as to meet the limited computing resources of a data computing and analyzing system and the safety of a nuclear power plant.
Description
Technical Field
The invention belongs to the technical field of nuclear power equipment detection, and particularly relates to a sensor screening method and device based on a sensor data reconstruction technology, and a sensor data reconstruction method and system.
Background
The sensor data reconstruction technology is used for intelligently monitoring a nuclear power plant sensor on line and replacing a sensor signal with a fault so that a nuclear power unit has automatic recovery capability after the sensor fault in a certain time period.
The sensor data reconstruction technology plays an important role in reducing the unplanned shutdown probability and reducing the construction, operation and maintenance costs of the unit. However, there are many sensors in the nuclear power plant, and there are differences in the support and influence degree of each sensor on the unit operation, and the computing resources of the data computing and analyzing system are limited, so how to reasonably determine the range of the sensors in the nuclear power plant that need the data reconstruction technology is an urgent problem to be solved.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a sensor screening method and device, a sensor data reconstruction method and system based on a sensor data reconstruction technology, aiming at the defects in the prior art, so that a reasonable sensor application range needing data reconstruction can be screened out to meet the limited computing resources of a data computing and analyzing system and the safety of a nuclear power plant.
In a first aspect, an embodiment of the present invention provides a sensor screening method based on a sensor data reconstruction technique, including: and screening a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensors, and/or the importance of the sensors.
Preferably, the screening of the target sensor set from the alternative sensors according to sensor data feasibility, and/or sensor failure probability, and/or sensor importance comprises: screening out a first sensor set from alternative sensors in a default value range according to the feasibility of the sensor data; selecting sensors with failure probability larger than a failure threshold value from the first sensor set to obtain a second sensor set; and screening the target sensor set from the second sensor set according to the importance of the sensors.
Preferably, the screening out the first set of sensors from the candidate sensors within the default value range according to the sensor data feasibility includes: screening out sensors accessed to the distributed control system from the alternative sensors in the default value range to obtain a third sensor set; judging whether the response time of the sensors in the third sensor set is smaller than a first preset threshold value or not, and screening out the sensors of which the judgment result is that the response time is smaller than the first preset threshold value to obtain a fourth sensor set; and judging whether the packet loss rate of data acquisition of the sensors in the fourth sensor set is smaller than a second preset threshold value or not, and screening out the sensors with the judgment results that the packet loss rate is smaller than the second preset threshold value to obtain a first sensor set.
Preferably, the screening out the sensors of which the packet loss rate is smaller than the second preset threshold as the determination result, and obtaining the first sensor set includes: screening out the sensors with the judgment results that the packet loss rate is smaller than a second preset threshold value from the fourth sensor set to obtain a fifth sensor set; screening out a sensor with a measuring object of any one of temperature, pressure, flow, liquid level, concentration and neutron flux from the fifth sensor set to obtain a sixth sensor set; and judging whether the sensors in the sixth sensor set can collect historical operating data above a preset working condition and a preset collection frequency, wherein the quantity of the historical operating data is larger than a third preset threshold value, screening out the sensors of which the judgment result is that the historical operating data above the preset working condition and the preset collection frequency can be collected, and the quantity of the historical operating data is larger than the third preset threshold value, and thus obtaining the first sensor set.
Preferably, the selecting a sensor with a failure probability greater than a failure threshold from the first sensor set to obtain a second sensor set specifically includes: classifying the sensors in the first sensor set according to the measurement principle to obtain WiA type sensor, i is more than or equal to 1 and less than or equal to 14; obtaining WiN failure modes of the type sensor and failure probabilities corresponding to the N failure modes, wherein N is a positive integer; summing the failure probabilities corresponding to the N failure modes to obtain WiTotal probability of failure of type sensor; selecting W with total failure probability greater than failure thresholdiType sensor, and obtaining W from the first set of sensorsiAnd obtaining a second sensor set by the sensors corresponding to the type sensors.
Preferably, the obtaining WiThe N failure modes and the corresponding failure probabilities of the N failure modes of the type sensor specifically include: collecting W in the first set of sensorsiHistorical operating data of the sensor corresponding to the type sensor; according to preset frequency pairs of collected WiCarrying out frequency thinning on historical operating data of the type sensor; performing data cutting on the historical operating data subjected to frequency thinning according to preset duration to obtain WiSample set C of type sensorsi(ii) a From the sample set CiIn obtaining WiFailure sample set F of type sensori(ii) a According to N failure modes to WiFailure sample set F of type sensoriAll the subsamples in (1)This FikClassifying to obtain WiN failure modes of the type sensor comprise numerical value missing, numerical value jamming, numerical value mutation and numerical value drifting; calculating WiObtaining the failure probability of N failure modes of the type sensor to obtain WiThe N failure modes of the type sensor correspond to failure probabilities.
Preferably, the slave sample set CiIn obtaining WiFailure sample set F of type sensoriThe method specifically comprises the following steps: s1, from sample set CiM subsamples C without failure dataijTraining as a training sample to obtain a first state monitoring model, wherein j is 1, 2, n, and M is less than n, and the state monitoring model is used for judging whether the sample is a failure sample; s2, all the subsamples CijInputting the data into a first state monitoring model to obtain all subsamples CijThe result of the judgment of whether the device is invalid or not; s3, screening out the subsample C corresponding to the failure judgment resultijObtaining a sample set Zi(ii) a S4, collecting the samples ZiTraining the first state monitoring model as a training sample to obtain a second state monitoring model, and emptying a sample set Zi(ii) a S5, iteratively executing the steps S2-S4 h times to obtain an h +2 state monitoring model, wherein h is a positive integer; s6, all the subsamples CijInputting the data into an h +2 state monitoring model to obtain all subsamples CijThe result of the judgment of whether the device is invalid or not; s7, screening the subsample C corresponding to the judgment result of yes failureijTo obtain WiFailure sample set F of type sensori。
Preferably, the pairs W according to N failure modesiFailure sample set F of type sensoriAll samples F in (1)ikClassifying to obtain WiThe N failure modes of the type sensor specifically include: calculating WiFailure sample set F of type sensoriAll subsamples F in (1)ikThe mean, the variance, and the mean of the slopes of two adjacent data points; if the mean value is smaller than the fourth preset threshold, the subsample FikThe failure mode of (a) is a numerical value missing; if the variance is less than the fifth thresholdSetting the threshold value, the subsample FikThe failure mode of (a) is numerical jamming; if the average value of the slopes of two adjacent data points is greater than a sixth preset threshold, the subsample FikThe failure mode of (a) is numerical mutation; otherwise, subsample FikThe failure mode of (a) is numerical drift; statistical failure sample set FiSub-samples F corresponding to the N failure modesikIs given as the number of WiN failure modes of the type sensor. The calculation WiObtaining the failure probability of N failure modes of the type sensor to obtain WiThe corresponding failure probability of N failure modes of the type sensor specifically comprises: according to failure sample set FiSub-samples F corresponding to the N failure modesikNumber of and sample set CiNeutron sample CijAnd respectively calculating the failure probability corresponding to the N failure modes according to the ratio of the total number.
Preferably, the screening the target sensor set from the second sensor set according to the sensor importance specifically includes: acquiring important control functions of each sensor in the second sensor set to construct a first mapping table, wherein the first mapping table comprises sensor names, the important control functions and mapping relations between the sensor names and the important control functions; acquiring the aftereffects of each important control function in the first mapping table under the condition of failure according to engineering experience, wherein the consequences comprise serious consequences and general consequences; and screening out the sensors corresponding to the serious consequences of the important control functions in the first mapping table under the failure condition to obtain a target sensor set.
In a second aspect, an embodiment of the present invention further provides a sensor data reconstruction method, where a target sensor set is screened out by using the screening method in the first aspect, and signals of all sensors in the target sensor set are monitored online; when any sensor in the target sensor set has a fault signal, the fault signal is replaced, so that the nuclear motor group has the automatic recovery capability after the sensor fault within the preset time.
In a third aspect, an embodiment of the present invention further provides a sensor screening apparatus based on a sensor data reconstruction technique, including a screening module. And the screening module is used for screening out a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensors, and/or the importance of the sensors. The screening module comprises a first screening unit, a second screening unit and a third screening unit. The system comprises a first screening unit, a second screening unit and a third screening unit, wherein the first screening unit is used for screening a first sensor set from alternative sensors in a default value range according to the feasibility of sensor data, the second screening unit is connected with the first screening unit and used for selecting the sensors with failure probability larger than a failure threshold value from the first sensor set to obtain a second sensor set, and the third screening unit is connected with the second screening unit and used for screening the target sensor set from the second sensor set according to the importance of the sensors.
Preferably, the second screening unit comprises a classification component, an acquisition component, a calculation component and a screening component. A classification component connected with the first screening unit and used for classifying the sensors in the first sensor set according to the measurement principle to obtain WiType sensor, i is more than or equal to 1 and less than or equal to 14. An acquisition component connected with the classification component for acquiring WiN failure modes of the type sensor and corresponding failure probabilities of the N failure modes, wherein N is a positive integer. The calculation component is connected with the acquisition component and is used for summing the failure probabilities corresponding to the N failure modes to obtain WiTotal probability of failure of type sensor. A screening component connected with the computing component and the first screening unit and used for selecting W with the total failure probability larger than the failure threshold valueiType sensor, and obtaining W from the first set of sensorsiAnd obtaining a second sensor set by the sensors corresponding to the type sensors.
In a fourth aspect, an embodiment of the present invention further provides a sensor data reconstruction system, including the screening device of the sensor based on the sensor data reconstruction technology in the third aspect, and a monitoring module and a replacement module. The sensor screening device based on the sensor data reconstruction technology is used for screening out a target sensor set. And the monitoring module is connected with the screening device of the sensor based on the sensor data reconstruction technology and used for monitoring a target sensor set screened by the screening device of the sensor based on the sensor data reconstruction technology on line and outputting a monitoring result. And the replacing module is connected with the monitoring module and used for replacing the fault signal when the fault signal appears in the monitoring result so as to enable the nuclear motor group to have the automatic recovery capability after the sensor fault within the preset time.
According to the sensor data reconstruction technology-based screening method and device and the sensor data reconstruction method and system, a target sensor set is screened from alternative sensors according to the feasibility of sensor data and/or the failure probability of the sensors and/or the importance of the sensors, and the target sensor set is used as a sensor set in the application range of the data reconstruction technology. The application of the data reconstruction technology is carried out according to the sensor set screened out according to the conditions, so that the limited computing resources of a data computing and analyzing system and the safety of a nuclear power plant can be met.
Drawings
FIG. 1: the invention is a flow chart of a sensor screening method based on a sensor data reconstruction technology in embodiment 1;
FIG. 2: the structure diagram of the screening device of the sensor based on the sensor data reconstruction technology in the embodiment 3 of the invention;
FIG. 3: a structure diagram of a second screening unit of the screening apparatus for a sensor based on the sensor data reconstruction technique according to embodiment 3 of the present invention;
FIG. 4: is a structural diagram of a sensor data reconstruction system according to embodiment 4 of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention is further described in detail below with reference to the accompanying drawings and examples.
Example 1:
as shown in fig. 1, the present embodiment provides a sensor screening method based on a sensor data reconstruction technique, including: and screening out a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensor, and/or the importance of the sensor. The sequence of the sensor screening according to any multiple screening conditions in the three screening conditions is not unique, and the screened target sensor set can be suitable for a sensor data reconstruction technology so as to meet the requirements of limited computing resources of a data computing and analyzing system and the safety of a nuclear power plant.
Optionally, the screening out the target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the probability of the sensor failure, and/or the importance of the sensor comprises:
Optionally, the step 101 specifically includes steps S11-S13:
and step S11, screening out the sensors accessed to the distributed control system from the alternative sensors in the default value range to obtain a third sensor set.
Step S12, determining whether the response time of the sensors in the third sensor set is smaller than a first preset threshold, and screening out the sensors whose determination result is that the response time is smaller than the first preset threshold, to obtain a fourth sensor set.
Step S13, determining whether the packet loss rate of data acquisition of the sensors in the fourth sensor set is smaller than a second preset threshold, and screening out the sensors whose determination result is that the packet loss rate is smaller than the second preset threshold, to obtain a first sensor set.
Optionally, step S13 specifically includes steps S131 to S133:
step 131, selecting the sensors with the judgment result that the packet loss rate is smaller than a second preset threshold value from the fourth sensor set, and obtaining a fifth sensor set.
Step S132, a sensor whose measurement object is any one of temperature, pressure, flow rate, liquid level, concentration, and neutron flux is screened from the fifth sensor set, and a sixth sensor set is obtained.
Step S133, determining whether the sensors in the sixth sensor set can collect historical operating data above a preset operating condition and a preset collection frequency, and the number of the historical operating data is greater than a third preset threshold, and screening out the sensors whose determination result is that the historical operating data above the preset operating condition and the preset collection frequency can be collected, and the number of the historical operating data is greater than the third preset threshold, thereby obtaining the first sensor set.
In this embodiment, an RCV system (chemical and volumetric system) is taken as an example for explanation, and according to a nuclear power engineering document, sensors in an original default value range of the RCV system are obtained and used as alternative sensors, and 34 sensors are counted, as shown in table 1:
TABLE 1
The sensor data viability of the 34 sensors in table 1 was analyzed, wherein the sensor data viability includes sensor data transmission availability and sensor operational data availability.
First, whether 34 sensors in table 1 all satisfy the sensor data transmission availability (i.e., the signal transmission requirement) is analyzed, and the determination conditions are as follows: (1) whether the sensor is accessed to a nuclear power plant DCS system or not; (2) whether the response time from sensor data acquisition to output is less than a threshold X1(ii) a (3) Whether the packet loss rate of sensor data acquisition is less than a threshold value X2。
If a certain sensor cannot meet the signal transmission requirement, removing the sensor from the alternative sensors; if a certain sensor simultaneously meets the three judgment conditions, the sensor is continuously kept in the alternative sensor. In this embodiment, all the 34 sensors in table 1 obtained through analysis are connected to the DCS system, and the response time of each sensor is smaller than the threshold X1And the packet loss rate of data acquisition of each sensor is less than a threshold value X2Therefore, 34 sensors remain in the alternative sensors.
Secondly, analyzing the availability of the sensor operation data (namely whether the historical operation data is sufficient or not), wherein the historical operation data comprises normal operation data and abnormal operation data, and judging conditions are as follows: (1) whether the object measured by the sensor is the slowly-varying data with the inertia characteristic of the fluid process system of the nuclear power plant comprises the following steps: temperature, pressure, flow, liquid level, concentration, neutron flux 6 types; (2) whether the quantity of data points of normal historical data and abnormal historical data which are collected by a sensor under the working condition of more than 20 percent of Pn and have the collection frequency of 1 time in 1 second or more is greater than a given threshold value X3。
If a certain sensor cannot meet the availability requirement of the sensor operation data, removing the sensor from the alternative sensors; if a certain sensor simultaneously meets the two judgment conditions, the sensor is continuously kept in the alternative sensor. In this embodiment, the 34 sensors in table 1 obtained through analysis are four types, i.e., temperature, pressure, flow and liquid level, and all the 34 sensors can collect historical operating data under the working condition of more than 20% Pn and with the acquisition frequency of 1 time in 1 second, and the number of the historical operating data is greater than the threshold value X3Therefore, 34 sensors in table 1 are all retained in the candidate sensors, so that the first sensor set includes 34 sensors in table 1. It should be noted that, since the sensor data reconstruction technique can be applied only after the stack is started to 20% full power, the preset operating condition in this implementation is set to 20% Pn (20% full power). In addition, the sampling frequency of DCS in this embodiment is 1 time in 1 second, and if the data sampling frequency is too low, the accuracy of the analysis result will be affected, so the preset sampling frequency in this embodiment is set to 1 time in 1 second.
And 102, selecting the sensors with the failure probability larger than a failure threshold value from the first sensor set to obtain a second sensor set.
Optionally, the step 102 specifically includes steps S21-S24:
step S21, classifying the sensors in the first sensor set according to the measurement principle to obtain WiType sensor, i is more than or equal to 1 and less than or equal to 14.
Generally, the classification is based on the type of parameter measured by the sensor,is classified as QdClass d 1, 2. As shown in table 2, the measurement parameter types have 6 types, including: the temperature, pressure, flow, level, concentration, neutron flux are in 6 categories. For each parameter type, classification was performed according to the measurement principle, and there were 14 classes, as shown in table 2 below.
TABLE 2
In this embodiment, 34 sensors in the first sensor set are classified according to the measurement principle types shown in table 2 to obtain WiThe number of the type sensors is 4, and the type sensors are thermocouple temperature sensors W1Differential pressure sensor W5Dynamic pressure generating flow sensor W6Differential pressure liquid level sensor W11. The classification results are shown in table 3 below:
TABLE 3
Step S22, obtaining WiN failure modes of the type sensor and corresponding failure probabilities of the N failure modes, wherein N is a positive integer.
Optionally, obtaining WiThe N failure modes of the type sensor and the failure probabilities corresponding to the N failure modes specifically include: collecting W in the first set of sensorsiHistorical operating data of the sensor corresponding to the type sensor; according to preset frequency pairs of collected WiCarrying out frequency thinning on historical operating data of the type sensor; performing data cutting on historical operating data subjected to frequency thinning according to preset duration to obtain WiSample set C of type sensorsi(ii) a From the sample set CiIn obtaining WiType sensingFailure sample set F of devicei(ii) a According to N failure modes to WiFailure sample set F of type sensoriAll subsamples F in (1)ikClassifying to obtain WiN failure modes of the type sensor comprise numerical value missing, numerical value clamping stagnation, numerical value mutation and numerical value drifting; calculating WiObtaining the failure probability of N failure modes of the type sensor to obtain WiThe N failure modes of the type sensor correspond to a failure probability.
Wherein, from the sample set CiIn obtaining WiFailure sample set F of type sensoriThe method specifically comprises the following steps:
s1, from sample set CiM sub-samples C without failure data are selectedijTraining as a training sample to obtain a first state monitoring model, wherein j is 1, 2, and n, and M is less than n, and the state monitoring model is used for judging whether the sample is a failure sample;
s2, all the subsamples CijInputting the data into a first state monitoring model to obtain all sub-samples CijThe result of the judgment of whether the device is invalid or not;
s3, screening out the subsample C corresponding to the failure judgment resultijObtaining a sample set Zi；
S4, collecting the samples ZiTraining a first state monitoring model as a training sample to obtain a second state monitoring model, and emptying a sample set Zi；
S5, iteratively executing the steps S2-S4 h times to obtain an h +2 state monitoring model, wherein h is a positive integer;
s6, all the subsamples CijInputting the data into a h +2 state monitoring model to obtain all subsamples CijThe result of the judgment of whether the device is invalid or not;
s7, screening the subsample C corresponding to the judgment result of yes failureijTo obtain WiFailure sample set F for type sensori。
In this embodiment, W is collected in sequenceiThe sensors corresponding to the type sensors (i.e. 34 sensors in the first set of sensors)Sensor), thinning the frequency of the data according to the frequency of 1 time in 1 second, taking the data of 1 minute as a sample, namely cutting the data after thinning the frequency according to the preset time length (1 minute) to obtain a sample set C of which each sample comprises 60 data pointsi. For example, in the present embodiment, each W is obtained separatelyiType sensor 600000 data points. In which the flow sensor W is actuated by dynamic pressure6The type is taken as an example, all dynamic pressure emission flow data collected by the RCV system from the DCS are frequency of 1 time in 1 second, so that frequency thinning is not needed, collected historical operation data are cut, and a sample set C is obtained6A 10000 × 60 data matrix, in which the subsamples C6j1, 2, 10000. Manually selecting 400 sub-sample data (namely a 400 x 60 data matrix) without obvious failure from a 10000 x 60 data matrix, substituting the sub-sample data into a kernel regression classification model for training (0 is output by the model to be a normal sample; 1 is output by the model to be a failure sample), and obtaining a trained kernel regression classification model; then, 10000 × 60 data matrix is substituted into the trained kernel regression classification model, and each group of 1 × 60 states (output is 10000 × 1 matrix; 0 represents normal sample, 1 represents failure sample) are output as shown below:
then, a normal sample output by the model in the 10000 × 60 data matrix is used as a training sample to retrain the kernel regression classification model, and a final kernel regression classification model is obtained through iteration; substituting 10000 × 60 data matrix into the final kernel regression classification model, and outputting each group of 1 × 60 state (output is 10000 × 1 matrix; 0 represents normal sample, and 1 represents failure sample); combining all 1 × 60 subsamples output as failure samples into one 3213 × 60 failure sample set F6(3213 is the number of failed subsamples of dynamic pressure trigger flow sensor), thereby obtaining W6Failure sample set F of type sensor6. In addition, W is also obtained according to the above method1、W5、W11The corresponding failure sample sets of the type sensors are respectively F1、F5、F11。
Wherein W is paired according to N failure modesiFailure sample set F of type sensoriAll samples F in (1)ikClassifying to obtain WiThe N failure modes of the type sensor specifically include: calculating WiFailure sample set F of type sensoriAll subsamples F in (1)ikAverage value, variance, and average value of slopes of two adjacent data points; if the mean value is less than the fourth preset threshold, the subsample FikThe failure mode of (a) is a numerical value missing; if the variance is less than a fifth preset threshold, the subsample FikThe failure mode of (a) is numerical jamming; if the average value of the slopes of two adjacent data points is greater than a sixth preset threshold, the subsample FikThe failure mode of (a) is numerical sudden change; otherwise, subsample FikThe failure mode of (a) is numerical drift; statistical failure sample set FiSub-samples F corresponding to the N failure modesikIs given as the number of WiN failure modes of the type sensor.
In this embodiment, 4W are each pairediCarrying out mode classification on the failure sample set of the type sensor to obtain WiThe number of subsamples corresponding to the 4 failure modes of the type sensor. Flow sensor W is triggered by dynamic pressure6The types are as follows: 3213 × 60 failure sample set F is calculated6In each subsample F6k(1 × 60), mean, variance, mean of slopes of two adjacent data points. Dynamic pressure generating flow sensor W obtained through calculation63213 failed subsamples of the type sensor are respectively 104 missing values, 178 stuck values, 189 abrupt changes of the values and 2742 drifting values.
Wherein W is calculatediObtaining the failure probability of N failure modes of the type sensor to obtain WiThe corresponding failure probabilities of the N failure modes of the type sensor specifically include: according to the failure sample set FiSub-samples F corresponding to the N failure modesikNumber of and sample set CiNeutron sample CijAnd respectively calculating the failure probability corresponding to the N failure modes according to the ratio of the total number.
In this example, 4 Ws are calculated respectivelyiThe 4 failure modes of the type sensor correspond to a failure probability. If FiIf the number of the data of the failure mode a is more than 0, the failure probability p is calculated by adopting a statistical methodia(ii) a If FiThe number of data of failure mode a of (2) is 0, and the failure probability is temporarily set to 0. Flow sensor W is triggered by dynamic pressure6The types are as follows: respectively substituting 104 missing values, 178 stuck values, 189 abrupt changes and 2742 shifted values into a statistical algorithm to calculate the failure probability of each failure mode, for example, F6Failure probability p of failure mode (missing value)61Is 104/10000.
Step S23, summing the failure probabilities corresponding to the N failure modes to obtain WiTotal probability of failure for type sensor.
Step S24, selecting W with total failure probability greater than failure thresholdiType sensor, and obtaining W from the first set of sensorsiAnd obtaining a second sensor set by the sensors corresponding to the type sensors.
In this example, 4 Ws are calculated respectivelyiTotal probability of failure p of type sensori＝∑a＝1，2...4pia. If a certain WiIf the total probability of failure of the type sensors is greater than the failure threshold, then W is retained in the first set of sensorsiAll sensors corresponding to the type sensors; otherwise, delete the W in the first set of sensorsiAll sensors corresponding to the type sensor. 4W are obtained by calculationiThe total probability of failure of a type sensor is not extremely low, so 4 WsiAll 34 sensors corresponding to the type sensors are reserved to form a second sensor set.
And 103, screening out a target sensor set from the second sensor set according to the importance of the sensors.
Optionally, the step 103 specifically includes steps S31-S33:
step S31, obtaining the important control function of each sensor in the second sensor set to construct a first mapping table, where the first mapping table includes the name of the sensor, the important control function, and the mapping relationship between the two.
And step S32, obtaining the consequences of each important control function in the first mapping table under the condition of failure according to engineering experience, wherein the consequences comprise serious consequences and general consequences.
And step S33, screening out the sensors corresponding to the serious consequences of the important control functions in the first mapping table under the condition of failure, and obtaining a target sensor set.
In this embodiment, according to a "functional analysis" file of the nuclear power plant, an important control function list of each sensor in the second sensor set is obtained:
Bi1、Bi2、...Bij、...、Binwherein B isijIndicating the j important control function of the i sensor. For each important control function BijAnalyzing the important control function B according to engineering experienceijThe possible consequences of a failure, including serious and general ones, the number s of serious consequences is calculatedijWherein s isijIndicating the jth severe consequence of the ith sensor. Serious consequences include two major areas:
(1) serious consequences of two types of working conditions occur, which specifically include:
the set of control rods is lifted out of control at reactor subcritical;
the set of control rods is lifted out of control during the operation of the reactor;
one or a group of control rod assemblies is improperly controlled in position or dropped;
uncontrolled dilution of boric acid;
partial loss of reactor coolant flow;
start of a shutdown loop;
total loss of load and/or turbine shutdown;
loss of primary feed water;
the steam generator main water supply system is not operating normally;
excessive load increase (at full power);
short reactor coolant system depressurization;
one safety valve of the second loop is opened by mistake;
false start of the safety injection system.
(2) Leading to serious consequences of violation of the operational constraints (LCO) of the nuclear power plant operational specifications.
The number s of serious consequences of all important control functions of the ith sensor in the second sensor seti＝∑j＝1，2，...，nSij. If si is greater than or equal to 1, the ith sensor is regarded as a key sensor signal and is kept in the second sensor set; otherwise, the ith sensor is deleted from the second set of sensors. In this embodiment, the mapping relationship between the 34 sensors in the second sensor set and the important control function is shown in table 4:
TABLE 4
For each important control function in table 4, according to engineering experience, possible consequences caused by the failure of the important control function are analyzed, and finally, 15 sensors with serious consequences caused by the failure are screened out as shown in table 5, so that a target sensor set is formed. The application of the data reconstruction technique may then be performed on the 15 sensors in the target sensor set.
TABLE 5
The sensor screening method based on the sensor data reconstruction technology provided by the embodiment has the following beneficial effects:
(1) the economic efficiency is as follows: the feasibility of the nuclear power plant sensor data reconstruction technology is improved, sensors with infeasible technology, low failure probability and serious consequences caused by failure are screened out, and limited computing resources are preferentially allocated to an important sensor data reconstruction function, so that the unit computing cost is saved, and the economy is improved; the targeted screening of the sensors can improve the performance of a subsequent sensor data reconstruction model, so that the subsequent reconstruction model can enable a fault sensor to have a strong automatic recovery function, and a unit can continuously keep running at full power. Therefore, the system can replace the traditional 'periodic test', and can uniformly maintain pertinence when the shutdown is planned, thereby reducing the operation and maintenance cost.
(2) Safety: the collection and classification work of sensor failure data is beneficial to building a model during sensor data reconstruction, the performance of the model can be improved, the reconstructed model can accurately repair the faults of the nuclear power plant sensor, and therefore the misoperation or the operation rejection of the power plant system triggered by the automatic action based on the wrong unit state signals are avoided, and the operation reliability of the sensor is improved. Limited computing resources can be provided for important sensors, the response time of sensor reconstruction is shortened, and serious consequences caused by false unit state signal triggering automatic action maloperation or failure operation due to delayed response when a power plant system sensor fails are avoided, so that the operation reliability of the sensors is improved.
Example 2:
the embodiment provides a sensor data reconstruction method applied to a nuclear power plant, and the method comprises the following steps:
step 201, screening out a target sensor set by using the sensor screening method based on the sensor data reconstruction technology described in embodiment 1.
In step 202, the signals of all sensors in the target sensor set are monitored online.
And 203, when any sensor in the target sensor set has a fault signal, replacing the fault signal so that the nuclear power generating unit has the automatic recovery capability after the sensor fails within a preset time.
Example 3:
as shown in fig. 2, the present embodiment provides a sensor screening apparatus based on a sensor data reconstruction technique, which includes a screening module 3. The screening module 3 is used for screening out a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensors, and/or the importance of the sensors.
Optionally, the screening module 3 comprises a first screening unit 31, a second screening unit 32 and a third screening unit 33.
A first screening unit 31, configured to screen out a first set of sensors from the candidate sensors within the default value range according to the sensor data feasibility. And the second screening unit 32 is connected with the first screening unit 31 and is used for selecting the sensors with the failure probability greater than the failure threshold value from the first sensor set to obtain a second sensor set. And a third screening unit 33 connected to the second screening unit 32, for screening the target sensor set from the second sensor set according to the sensor importance.
Optionally, the first screening unit comprises a judging component. The judging component is used for screening out the sensors accessed to the distributed control system from the alternative sensors in the default value range to obtain a third sensor set, judging whether the response time length of the sensors in the third sensor set is smaller than a first preset threshold value, screening out the sensors with the judgment result that the response time length is smaller than the first preset threshold value to obtain a fourth sensor set, judging whether the packet loss rate of data acquisition of the sensors in the fourth sensor set is smaller than a second preset threshold value, screening out the sensors with the judgment result that the packet loss rate is smaller than the second preset threshold value from the fourth sensor set to obtain a fifth sensor set, screening out the sensors with the measurement objects of any one of temperature, pressure, flow, liquid level, concentration and neutron flux from the fifth sensor set to obtain a sixth sensor set, and the sensor is also used for judging whether the sensors in the sixth sensor set can collect historical operating data above a preset working condition and a preset acquisition frequency, the quantity of the historical operating data is greater than a third preset threshold value, and screening the sensors of which the judgment result is that the historical operating data above the preset working condition and the preset acquisition frequency can be collected and the quantity of the historical operating data is greater than the third preset threshold value, so that the first sensor set is obtained.
Optionally, as shown in fig. 3, the second screening unit 32 comprises a classification component 321, an acquisition component 322, a calculation component 323 and a screening component 324.
A classification component 321 connected to the first screening unit 31 for classifying the sensors in the first sensor set according to the measurement principle to obtain WiType sensor, i is more than or equal to 1 and less than or equal to 14.
An obtaining component 322 coupled to the classifying component 321 for obtaining WiN failure modes of the type sensor and corresponding failure probabilities of the N failure modes, wherein N is a positive integer.
A calculating component 323 connected to the obtaining component 322 for summing the failure probabilities corresponding to the N failure modes to obtain WiTotal probability of failure of type sensor.
A screening component 324 connected to the calculating component 323 and the first screening unit 31 for selecting W with the total failure probability larger than the failure thresholdiType sensor, and obtaining W from the first set of sensorsiAnd obtaining a second sensor set by the sensors corresponding to the type sensors.
Optionally, the acquisition component is for collecting W in the first set of sensorsiHistorical operating data of the sensors corresponding to the type sensors and according to the preset frequency, the collected WiFrequency thinning is carried out on historical operating data of the type sensor, data cutting is carried out on the historical operating data after the frequency thinning according to preset duration, and W is obtainediSample set C for type sensoriAlso for deriving from the sample set CiIn obtaining WiFailure sample set F of type sensoriAccording to N failure modes to WiFailure sample set F of type sensoriAll subsamples F in (1)ikClassifying to obtain WiN losses of type sensorA number N of failure modes including numerical absence, numerical stuck, numerical mutation and numerical drift, and, for calculating WiObtaining the failure probability of N failure modes of the type sensor to obtain WiThe N failure modes of the type sensor correspond to a failure probability.
It should be noted that the functions of each module, unit, and component in the screening apparatus of the sensor based on the sensor data reconstruction technology according to this embodiment are described as in embodiment 1.
Example 4:
as shown in fig. 4, the present embodiment provides a sensor data reconstruction system, which includes the sensor screening apparatus based on the sensor data reconstruction technology described in embodiment 3, and a monitoring module 41 and a replacement module 42.
The sensor screening device based on the sensor data reconstruction technology is used for screening out a target sensor set.
And the monitoring module 41 is connected with the screening device of the sensor based on the sensor data reconstruction technology, and is used for performing online monitoring on the target sensor set screened by the screening device of the sensor based on the sensor data reconstruction technology and outputting a monitoring result.
And the replacing module 42 is connected with the monitoring module 41 and is used for replacing the fault signal when the fault signal occurs in the monitoring result, so that the nuclear power generating unit has the automatic recovery capability after the sensor fault within the preset time.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (13)
1. A sensor screening method based on a sensor data reconstruction technology is characterized by comprising the following steps:
and screening out a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensors, and/or the importance of the sensors.
2. The method for screening sensors based on sensor data reconstruction technology as claimed in claim 1, wherein screening a target sensor set from candidate sensors according to sensor data feasibility, and/or sensor failure probability, and/or sensor importance comprises:
screening out a first sensor set from alternative sensors in a default value range according to the feasibility of the sensor data;
selecting sensors with failure probability larger than a failure threshold value from the first sensor set to obtain a second sensor set;
and screening the target sensor set from the second sensor set according to the importance of the sensors.
3. The method for screening sensors based on sensor data reconstruction technology according to claim 2, wherein the screening a first set of sensors from candidate sensors within a default value range according to sensor data feasibility comprises:
screening out sensors accessed to the distributed control system from the alternative sensors in the default value range to obtain a third sensor set;
judging whether the response time of the sensors in the third sensor set is smaller than a first preset threshold value or not, and screening out the sensors of which the judgment results are that the response time is smaller than the first preset threshold value to obtain a fourth sensor set;
and judging whether the packet loss rate of data acquisition of the sensors in the fourth sensor set is smaller than a second preset threshold value, and screening out the sensors with the judgment results that the packet loss rate is smaller than the second preset threshold value to obtain a first sensor set.
4. The method for screening the sensors based on the sensor data reconstruction technology according to claim 3, wherein the screening out of the sensors with the packet loss rate smaller than the second preset threshold as the determination result to obtain the first sensor set comprises:
screening out the sensors with the judgment results that the packet loss rate is smaller than a second preset threshold value from the fourth sensor set to obtain a fifth sensor set;
screening out a sensor with a measuring object of any one of temperature, pressure, flow, liquid level, concentration and neutron flux from the fifth sensor set to obtain a sixth sensor set;
and judging whether the sensors in the sixth sensor set can collect historical operating data above a preset working condition and a preset acquisition frequency, wherein the quantity of the historical operating data is greater than a third preset threshold value, screening out the sensors of which the judgment result is that the historical operating data above the preset working condition and the preset acquisition frequency can be collected, and the quantity of the historical operating data is greater than the third preset threshold value, so as to obtain the first sensor set.
5. The sensor data reconstruction technology-based sensor screening method according to claim 4, wherein the step of selecting the sensor with the failure probability greater than the failure threshold from the first sensor set to obtain a second sensor set specifically comprises:
classifying the sensors in the first sensor set according to the measurement principle to obtain WiA type sensor, i is more than or equal to 1 and less than or equal to 14;
obtaining WiN failure modes of the type sensor and corresponding failure probabilities of the N failure modes, wherein N is a positive integer;
summing the failure probabilities corresponding to the N failure modes to obtain WiTotal probability of failure of type sensor;
selecting W with total failure probability greater than failure thresholdiType sensor, and obtaining W from the first set of sensorsiAnd obtaining a second sensor set by the sensors corresponding to the type sensors.
6. The method of claim 5, wherein the sensor data reconstruction technique is based on a sensor screening methodSaid obtaining WiThe N failure modes and the corresponding failure probabilities of the N failure modes of the type sensor specifically include:
collecting W in the first set of sensorsiHistorical operating data of the sensor corresponding to the type sensor;
according to preset frequency pairs of collected WiCarrying out frequency thinning on historical operating data of the type sensor;
performing data cutting on historical operating data subjected to frequency thinning according to preset duration to obtain WiSample set C of type sensorsi；
From the sample set CiIn obtaining WiFailure sample set F of type sensori；
According to N failure modes to WiFailure sample set F of type sensoriAll subsamples F in (1)ikClassifying to obtain WiN failure modes of the type sensor, wherein the N failure modes comprise numerical value missing, numerical value jamming, numerical value mutation and numerical value drifting;
calculating WiObtaining the failure probability of N failure modes of the type sensor to obtain WiThe N failure modes of the type sensor correspond to a failure probability.
7. The method of claim 6, wherein the screening of the sensor is performed from a sample set CiIn obtaining WiFailure sample set F of type sensoriThe method specifically comprises the following steps:
s1, from sample set CiM sub-samples C without failure data are selectedijTraining as a training sample to obtain a first state monitoring model, wherein j is 1, 2, n, and M is less than n, and the state monitoring model is used for judging whether the sample is a failure sample;
s2, all the subsamples CijInputting the data into a first state monitoring model to obtain all subsamples CijThe result of the judgment of whether the device is invalid or not;
s3, screening the sub-sample corresponding to the failure-negative judgment resultThis CijObtaining a sample set Zi；
S4, collecting the samples ZiTraining the first state monitoring model as a training sample to obtain a second state monitoring model, and emptying a sample set Zi；
S5, iteratively executing the steps S2-S4 h times to obtain an h +2 state monitoring model, wherein h is a positive integer;
s6, all the subsamples CijInputting the data into a h +2 state monitoring model to obtain all subsamples CijThe result of the judgment of whether the device is invalid or not;
s7, screening the subsample C corresponding to the judgment result of yes failureijTo obtain WiFailure sample set F of type sensori。
8. The method of claim 7, wherein the sensor data reconstruction technique based sensor screening method,
the pair W according to N failure modesiFailure sample set F of type sensoriAll samples F in (1)ikClassifying to obtain WiThe N failure modes of the type sensor specifically include:
calculating WiFailure sample set F of type sensoriAll subsamples F in (1)ikThe mean, the variance, and the mean of the slopes of two adjacent data points;
if the mean value is less than the fourth preset threshold, the subsample FikThe failure mode of (a) is a numerical value missing;
if the variance is less than a fifth preset threshold, the subsample FikThe failure mode of (a) is numerical jamming;
if the average value of the slopes of two adjacent data points is greater than a sixth preset threshold, the subsample FikThe failure mode of (a) is numerical mutation;
otherwise, subsample FikThe failure mode of (a) is numerical drift;
statistical failure sample set FiSub-samples F corresponding to the N failure modesikIs given as the number of WiN failure modes of the type sensor,
the calculation WiObtaining the failure probability of N failure modes of the type sensor to obtain WiThe corresponding failure probabilities of the N failure modes of the type sensor specifically include:
according to failure sample set FiSub-samples F corresponding to the N failure modesikNumber of and sample set CiNeutron sample CijAnd respectively calculating the failure probability corresponding to the N failure modes according to the ratio of the total number.
9. The method for screening sensors based on sensor data reconstruction technology according to claim 8, wherein the screening the target sensor set from the second sensor set according to sensor importance specifically includes:
acquiring important control functions of all sensors in the second sensor set to construct a first mapping table, wherein the first mapping table comprises sensor names, the important control functions and mapping relations between the sensor names and the important control functions;
obtaining the consequences of each important control function in the first mapping table under the condition of failure according to engineering experience, wherein the consequences comprise serious consequences and general consequences;
and screening out the sensors corresponding to the serious consequences of the important control functions in the first mapping table under the failure condition to obtain a target sensor set.
10. A sensor data reconstruction method, wherein a target sensor set is selected by the screening method according to any one of claims 1 to 9,
monitoring signals of all sensors in a target sensor set on line;
when any sensor in the target sensor set has a fault signal, the fault signal is replaced, so that the nuclear power generating unit has the automatic recovery capability after the sensor fault within the preset time.
11. The screening device of the sensor based on the sensor data reconstruction technology is characterized by comprising a screening module,
a screening module for screening a target sensor set from the alternative sensors according to the feasibility of the sensor data, and/or the failure probability of the sensor, and/or the importance of the sensor,
the screening module comprises a first screening unit, a second screening unit and a third screening unit,
a first screening unit for screening out a first set of sensors from the alternative sensors within the default value range according to the sensor data feasibility,
the second screening unit is connected with the first screening unit and used for selecting the sensors with the failure probability larger than the failure threshold value from the first sensor set to obtain a second sensor set,
and the third screening unit is connected with the second screening unit and is used for screening the target sensor set from the second sensor set according to the importance of the sensors.
12. The sensor-based screening apparatus of claim 11, wherein the second screening unit comprises a classification component, an acquisition component, a calculation component and a screening component,
a classification component connected with the first screening unit and used for classifying the sensors in the first sensor set according to the measurement principle to obtain WiA type sensor, i is more than or equal to 1 and less than or equal to 14,
an acquisition component connected with the classification component for acquiring WiN failure modes of the type sensor and the corresponding failure probabilities of the N failure modes, wherein N is a positive integer,
a calculating component connected with the obtaining component and used for summing the failure probabilities corresponding to the N failure modes to obtain WiThe total probability of failure of the type sensor,
a screening component connected with the computing component and the first screening unit and used for selecting W with the total failure probability larger than the failure threshold valueiType sensor, and obtaining W from the first set of sensorsiThe sensor corresponding to the type sensor obtains a second transmissionA set of sensors.
13. A sensor data reconstruction system comprising the sensor based sensor data reconstruction technique of claim 12, a screening device, and a monitoring module and a replacement module,
a sensor screening device based on sensor data reconstruction technology, which is used for screening out a target sensor set,
the monitoring module is connected with the screening device of the sensor based on the sensor data reconstruction technology and used for carrying out on-line monitoring on the target sensor set screened by the screening device of the sensor based on the sensor data reconstruction technology and outputting a monitoring result,
and the replacing module is connected with the monitoring module and used for replacing the fault signal when the fault signal appears in the monitoring result so that the nuclear power unit has the automatic recovery capability after the sensor fault within the preset time.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110183121.4A CN113221435A (en) | 2021-02-08 | 2021-02-08 | Sensor screening method and device and sensor data reconstruction method and system |
| PCT/CN2021/141567 WO2022166466A1 (en) | 2021-02-08 | 2021-12-27 | Sensor screening method and apparatus and sensor data reconstruction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110183121.4A CN113221435A (en) | 2021-02-08 | 2021-02-08 | Sensor screening method and device and sensor data reconstruction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113221435A true CN113221435A (en) | 2021-08-06 |
Family
ID=77084852
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110183121.4A Pending CN113221435A (en) | 2021-02-08 | 2021-02-08 | Sensor screening method and device and sensor data reconstruction method and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113221435A (en) |
| WO (1) | WO2022166466A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469151A (en) * | 2021-09-03 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Method, device, equipment and medium for screening sensor in industrial manufacturing |
| WO2022166466A1 (en) * | 2021-02-08 | 2022-08-11 | 中国核电工程有限公司 | Sensor screening method and apparatus and sensor data reconstruction method and system |
| CN115201865A (en) * | 2022-07-18 | 2022-10-18 | 广东汇天航空航天科技有限公司 | Fault detection and equipment selection method, device, equipment and storage medium |
| WO2023088355A1 (en) * | 2021-11-17 | 2023-05-25 | 中国核电工程有限公司 | Intelligent data reconstruction method and system for sensor |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116150455B (en) * | 2023-04-17 | 2023-07-18 | 华能信息技术有限公司 | Heterogeneous data analysis method |
| CN117648213B (en) * | 2024-01-30 | 2024-05-07 | 杭银消费金融股份有限公司 | Data transmission path model building method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184395A (en) * | 2015-08-26 | 2015-12-23 | 华北电力科学研究院有限责任公司 | Initial parameter determination method for thermal power generating unit comprising afterheat utilizing system |
| CN106651065A (en) * | 2015-10-29 | 2017-05-10 | 苏州热工研究院有限公司 | Method for screening components of interest in lifetime management of nuclear power plant in operation |
| CN109799269A (en) * | 2019-01-24 | 2019-05-24 | 山东工商学院 | Electronic nose gas sensor array optimization method based on behavioral characteristics different degree |
| CN110087207A (en) * | 2019-05-05 | 2019-08-02 | 江南大学 | Wireless sensor network missing data method for reconstructing |
| CN112162545A (en) * | 2020-10-21 | 2021-01-01 | 长安大学 | Automobile fault diagnosis method and system |
| CN112183999A (en) * | 2020-09-24 | 2021-01-05 | 国网山东省电力公司电力科学研究院 | Power distribution main equipment sensor reliability evaluation index feature extraction method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103513621B (en) * | 2012-06-28 | 2017-11-17 | 中国核电工程有限公司 | A kind of method for analyzing technological parameter default value of digital control system of nuclear power plant |
| CN104390657B (en) * | 2014-11-05 | 2017-12-12 | 浙江大学 | A kind of Generator Unit Operating Parameters measurement sensor fault diagnosis method and system |
| US11162824B2 (en) * | 2019-01-24 | 2021-11-02 | Vital Connect, Inc. | System and method of screening biosensors |
| CN111780800B (en) * | 2020-07-09 | 2021-10-29 | 哈尔滨工程大学 | Method and system for monitoring, isolating and reconstructing sensor fault |
| CN113221435A (en) * | 2021-02-08 | 2021-08-06 | 中国核电工程有限公司 | Sensor screening method and device and sensor data reconstruction method and system |
-
2021
- 2021-02-08 CN CN202110183121.4A patent/CN113221435A/en active Pending
- 2021-12-27 WO PCT/CN2021/141567 patent/WO2022166466A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184395A (en) * | 2015-08-26 | 2015-12-23 | 华北电力科学研究院有限责任公司 | Initial parameter determination method for thermal power generating unit comprising afterheat utilizing system |
| CN106651065A (en) * | 2015-10-29 | 2017-05-10 | 苏州热工研究院有限公司 | Method for screening components of interest in lifetime management of nuclear power plant in operation |
| CN109799269A (en) * | 2019-01-24 | 2019-05-24 | 山东工商学院 | Electronic nose gas sensor array optimization method based on behavioral characteristics different degree |
| CN110087207A (en) * | 2019-05-05 | 2019-08-02 | 江南大学 | Wireless sensor network missing data method for reconstructing |
| CN112183999A (en) * | 2020-09-24 | 2021-01-05 | 国网山东省电力公司电力科学研究院 | Power distribution main equipment sensor reliability evaluation index feature extraction method |
| CN112162545A (en) * | 2020-10-21 | 2021-01-01 | 长安大学 | Automobile fault diagnosis method and system |
Non-Patent Citations (5)
| Title |
|---|
| 吴健宏;任玉洁;: "化工仪表设备失效率数据库研究", 广东化工, no. 09, 15 May 2018 (2018-05-15) * |
| 周鸣籁: "传感器失效的防护和检测", 自动化与仪表, vol. 7, no. 3, 31 December 1992 (1992-12-31), pages 13 - 16 * |
| 李季, 关艳霞: "传感器可靠性筛选方法的研究", 仪表技术与传感器, no. 11, 30 November 1996 (1996-11-30), pages 10 - 13 * |
| 蔡开龙 等: "基于改进LS2SVM的航空发动机传感器故障诊断与自适应重构控制", 航空动力学报, vol. 23, no. 6, 15 June 2008 (2008-06-15), pages 1118 - 1126 * |
| 邢鹏 等: "基于可拓学理论的传感器选型优化研究", 兵器装备工程学报, vol. 41, no. 9, 30 September 2020 (2020-09-30), pages 75 - 79 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022166466A1 (en) * | 2021-02-08 | 2022-08-11 | 中国核电工程有限公司 | Sensor screening method and apparatus and sensor data reconstruction method and system |
| CN113469151A (en) * | 2021-09-03 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Method, device, equipment and medium for screening sensor in industrial manufacturing |
| CN113469151B (en) * | 2021-09-03 | 2022-02-15 | 深圳市信润富联数字科技有限公司 | Method, device, equipment and medium for screening sensor in industrial manufacturing |
| WO2023088355A1 (en) * | 2021-11-17 | 2023-05-25 | 中国核电工程有限公司 | Intelligent data reconstruction method and system for sensor |
| CN115201865A (en) * | 2022-07-18 | 2022-10-18 | 广东汇天航空航天科技有限公司 | Fault detection and equipment selection method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022166466A1 (en) | 2022-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113221435A (en) | Sensor screening method and device and sensor data reconstruction method and system | |
| CN101770219B (en) | Knowledge acquisition method of fault diagnosis knowledge library of turn-milling combined machine tool | |
| CN102460529B (en) | Device abnormality monitoring method and system | |
| CN112749509B (en) | Intelligent substation fault diagnosis method based on LSTM neural network | |
| CN112817280A (en) | Implementation method for intelligent monitoring alarm system of thermal power plant | |
| CN111412579A (en) | Air conditioning unit fault type diagnosis method and system based on big data | |
| US20230162484A1 (en) | Apparatus and method for generating learning data for artificial intelligence model | |
| CN110687473B (en) | Fault positioning method and system for relay protection test of intelligent substation | |
| CN116308304B (en) | New energy intelligent operation and maintenance method and system based on meta learning concept drift detection | |
| CN111664083A (en) | Nuclear power main pump fault diagnosis method based on Bayesian network | |
| CN111784537B (en) | Power distribution network state parameter monitoring method and device and electronic equipment | |
| CN117060409A (en) | Automatic detection and analysis method and system for power line running state | |
| CN115453267A (en) | Fault diagnosis system for electric power information system | |
| CN115081647A (en) | Industrial intelligent instrument fault pre-diagnosis method based on Bayesian network model | |
| CN117289085A (en) | Multi-line fault analysis and diagnosis method and system | |
| CN115037603A (en) | Diagnosis evaluation method, device and system of electricity consumption information acquisition equipment | |
| CN114519382A (en) | Nuclear power plant key operation parameter extraction and abnormity monitoring method | |
| CN114487705A (en) | Power grid equipment fault positioning detection method | |
| KR20220089853A (en) | Method for Failure prediction and prognostics and health management of renewable energy generation facilities using machine learning technology | |
| CN105302476B (en) | A kind of reliability data online acquisition for nuclear power plant equipment analyzes storage system and its storage method | |
| CN115034094B (en) | Prediction method and system for operation state of metal processing machine tool | |
| CN115982578A (en) | Fault positioning method and device, electronic equipment and storage medium | |
| CN114283956A (en) | Intelligent data reconstruction method and system for sensor | |
| CN115310693A (en) | Comprehensive energy data cleaning method, system, equipment and medium | |
| CN114442543A (en) | Computer monitoring method suitable for early warning of hydropower station fault |
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 |