CN114358172A - Nuclear reactor fault classification method, apparatus, computer device, and storage medium - Google Patents
Nuclear reactor fault classification method, apparatus, computer device, and storage medium Download PDFInfo
- Publication number
- CN114358172A CN114358172A CN202111653107.2A CN202111653107A CN114358172A CN 114358172 A CN114358172 A CN 114358172A CN 202111653107 A CN202111653107 A CN 202111653107A CN 114358172 A CN114358172 A CN 114358172A
- Authority
- CN
- China
- Prior art keywords
- state
- data
- nuclear reactor
- fault classification
- different sampling
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005070 sampling Methods 0.000 claims abstract description 204
- 239000011159 matrix material Substances 0.000 claims abstract description 44
- 238000004590 computer program Methods 0.000 claims abstract description 37
- 238000013145 classification model Methods 0.000 claims abstract description 22
- 238000005259 measurement Methods 0.000 claims description 125
- 239000002245 particle Substances 0.000 claims description 56
- 230000004927 fusion Effects 0.000 claims description 26
- 238000004088 simulation Methods 0.000 claims description 26
- 238000004422 calculation algorithm Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 7
- 230000000737 periodic effect Effects 0.000 claims description 5
- 238000003745 diagnosis Methods 0.000 description 22
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000009835 boiling Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005253 cladding Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000009257 reactivity Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present application relates to a nuclear reactor fault classification method, apparatus, computer device, storage medium and computer program product. The method comprises the following steps: acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions; generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions; inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result. By adopting the method, the nuclear reactor fault classification result with high precision can be obtained.
Description
Technical Field
The present application relates to the field of nuclear reactor fault diagnosis technologies, and in particular, to a nuclear reactor fault classification method, apparatus, computer device, storage medium, and computer program product.
Background
As a technology related to measurement instruments and control, industrial big data, artificial intelligence, reliability, quality engineering and other multidisciplinary intersections, fault diagnosis is currently widely applied to aerospace, rail transit, energy equipment, intelligent plants and the like. In particular in the field of nuclear reactor systems, nuclear reactor systems are complex dynamic systems with strict safety constraint requirements that make early diagnosis of all types of faults difficult and critical. The key safety objective is to correctly and accurately identify the state when any abnormality/fault occurs, and to assist the operator in making the correct decision, thereby improving the safety level of the reactors. In consideration of the catastrophic consequences of a nuclear accident, which may be caused by nuclear reactor conditions, a diagnostic system that classifies the normal operating deviations of a nuclear power plant can help operators to focus on the most important monitoring, decision-making and control, reducing the decision-making error rate and workload of the operators. Therefore, it is necessary to investigate how to improve the diagnostic accuracy of the type of nuclear reactor fault.
Early fault diagnosis systems based on expert system technology were significantly limited. The conventional nuclear reactor accident diagnosis system adopts an expert system technology, namely, task support such as state monitoring, fault diagnosis, operation and maintenance decision and the like is provided through logical judgment of yes and no of a series of processes of fault diagnosis of a nuclear reactor system. Since the judgment method based on the rule is adopted to judge the determined symptom, the system mechanism is not deeply known, and the condition except the rule cannot be processed. Some early systems based on conventional fuzzy logic rules have difficulty overcoming their validity problems, although they can "guess" for new situations outside the rules, with a great deal of uncertainty, coupled with the lack of expert knowledge base. In the last two decades, new fault diagnosis methods have been widely studied and applied in nuclear power systems. The existing fault diagnosis method is extremely sensitive to measurement noise, overfitting or out-of-range reasoning occurs when the fault does not belong to the training or knowledge storage type of the fault, the fault diagnosis accuracy is influenced because the fault diagnosis method has no response capability of representing unknown, the deep learning robustness is low under the conditions of uncertain disturbance or poor data quality, and the existing learning-based method, particularly a machine learning method based on shallow characterization, cannot fully meet the accuracy requirement of fault classification diagnosis.
It can be seen that the technical scheme of the prior art has the problem of low fault classification precision.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a nuclear reactor fault classification method, apparatus, computer device, computer readable storage medium, and computer program product with higher accuracy.
In a first aspect, the present application provides a nuclear reactor fault classification method. The method comprises the following steps:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data;
forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result;
and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In one embodiment, obtaining optimal estimation data of the state of the nuclear reactor at different sampling moments comprises:
acquiring state data of a nuclear reactor at different sampling moments;
and fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the status data includes:
high precision simulation state data, real-time measurement state data, and periodic measurement state data.
In one embodiment, the obtaining the optimal state estimation data of the nuclear reactor at different sampling moments by fusing the state data at different sampling moments through a state optimal estimation fusion algorithm includes:
acquiring the best estimation data of the state of the nuclear reactor at the previous moment at different sampling moments and the confidence corresponding to the best estimation data of the state at the previous moment;
obtaining state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient corresponding to the state optimal estimation data at the previous moment;
acquiring state measurement data of a nuclear reactor at different sampling moments;
and fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, obtaining the state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence degrees corresponding to the state optimal estimation data at the previous moment comprises:
acquiring state data of the particles in the nuclear reactor at the previous moment at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient of the state optimal estimation data;
inputting the state data of the particles at the previous moment of different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments;
and obtaining state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at the different sampling moments.
In one embodiment, the obtaining the optimal state estimation data of the nuclear reactor at different sampling moments by fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm includes:
obtaining confidence degrees corresponding to state prediction data of the nuclear reactor at different sampling moments;
acquiring state data of particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data;
obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data;
acquiring actual state measurement data of the nuclear reactor at different sampling moments;
obtaining residual coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data;
obtaining a gain coefficient of the state measurement data according to the confidence coefficient corresponding to the estimated state measurement data;
and according to the gain coefficient and the residual coefficient, fusing the state prediction data and the state measurement data to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In a second aspect, the present application further provides a nuclear reactor fault classification apparatus. The device comprises:
the data acquisition module is used for acquiring the state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
the data processing module is used for generating a state optimal estimation sequence with multiple dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
the fault classification module is used for inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data;
forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result;
and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data;
forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result;
and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data;
forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result;
and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
According to the nuclear reactor fault classification method, the device, the computer equipment, the storage medium and the computer program product, on one hand, accurate fault classification results can be obtained by acquiring the best state estimation data of the nuclear reactor at different sampling moments and integrating the data at different sampling moments, and the precision of fault classification diagnosis is greatly improved; and secondly, inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the fault classification result, and then inputting the obtained state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model, so that different fault classification results are collected, the accuracy of the finally output nuclear reactor fault classification result and the confidence coefficient corresponding to the fault classification result can be ensured, and the high-precision nuclear reactor fault classification result can be obtained.
Drawings
FIG. 1 is a diagram of an environment in which a method for fault classification for a nuclear reactor according to an embodiment may be implemented;
FIG. 2 is a schematic flow diagram of a method for fault classification of a nuclear reactor according to an embodiment;
FIG. 3 is a schematic flow chart illustrating fusion of state data to obtain state-optimized estimation data according to an embodiment;
FIG. 4 is a schematic illustration of nuclear reactor state data fusion in one embodiment;
FIG. 5 is a schematic flow chart illustrating the substep of S380 in another embodiment;
FIG. 6 is a block diagram of an apparatus for a method of fault classification for a nuclear reactor according to an embodiment;
FIG. 7 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The nuclear reactor fault classification method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104, or may be located on the cloud or other network server. The server 104 obtains state optimal estimation data of the terminal 102 at different sampling moments, generates state optimal estimation sequences of multiple dimensions based on a preset period according to the obtained state optimal estimation data, forms a state estimation matrix, inputs the state optimal estimation matrix into a preset fault classification model to obtain a state fault classification result and a corresponding confidence coefficient, and inputs the obtained state fault classification result and the corresponding confidence coefficient into a preset self-learning model to obtain a fault classification result and a confidence coefficient of the nuclear reactor. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices and portable wearable devices, and the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart car-mounted devices, and the like. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, and the like. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers.
In one embodiment, as shown in fig. 2, a nuclear reactor fault classification method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
s100, acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions.
Wherein the state-optimal estimation data is state estimation data that is closest to a true state inside the nuclear reactor. Specifically, the nuclear reactor comprises a plurality of particles, each particle presents different states at different sampling moments, the states of all the particles in the reactor can be used for representing the states of the whole reactor at the sampling moments, fusion processing is carried out according to the state data of all the particles in the nuclear reactor at the different sampling moments, the state optimal estimation data of the nuclear reactor at the different sampling moments can be obtained, and each sampling moment has one state optimal estimation data. Meanwhile, it should be noted that the optimal estimation data of the state of the nuclear reactor and the state data of the particles are parameter data including a plurality of dimensions, such as: the nuclear reactor safety monitoring system comprises a plurality of parameter data such as reactivity coefficients, deviation from nucleate boiling, maximum linear power density, maximum fuel rod and cladding temperature and the like which are very relevant to the safety operation of a nuclear reactor, wherein the parameter data of a plurality of dimensions jointly form state optimal estimation data or state data of a certain particle.
And S200, generating a state optimal estimation sequence with multiple dimensions based on a preset period according to the state optimal estimation data.
The state optimal estimation sequence is a sequence into which state optimal estimation data at different sampling moments are divided according to a preset period. Specifically, according to a preset period, combining the state optimal data at different sampling moments according to the sequence of each moment, and correspondingly obtaining the state optimal estimation sequences of multiple dimensions. For example, the best state estimation data of the nuclear reactor at 1s, 2s, 3s, 4s, and 5s is obtained, and a best state estimation sequence with three dimensions of {3s, 4s, 5s }, {2s, 3s, 4s }, {1s, 2s, and 3s } can be obtained by presetting each 3s as a period.
And S300, forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions.
The state estimation matrix refers to a set of data of multiple dimensions in the state optimal estimation data at each sampling time in the same state optimal estimation sequence. Specifically, the diagnosis of the accident/fault type is performed from the current state of the nuclear reactor, which is the mainstream technology of many fault diagnoses at present. However, if only the data at the current moment is available, the historical change trend of the data is ignored, so that some accidents caused by extreme evolution of the working conditions of the reactor cannot be effectively diagnosed, and the opportunity of effective intervention of the operation of the reactor is missed. Therefore, in order to ensure the accuracy of the obtained fault classification result, it is necessary to effectively perform fusion diagnosis on the data at the historical sampling time. The importance of the data at different historical sampling moments relative to the data at the current sampling moment is a decreasing effect, the data at different historical sampling moments form a matrix, typically, the nuclear reactor state with 1 second as an update frequency, the period can be preset to be between 10 seconds and several minutes, and according to the comprehensive balance such as the trend of the nuclear reactor working condition change and the matrix processing speed, it should be noted that the data in the state matrix needs to be normalized, that is:
where P is the state variable estimate at the sampling instant, and Pmin and Pmax represent the maximum and minimum values of the state physical quantity under all accident conditions, respectively. The description will be given by taking the 5 th s as the current sampling time as an example: the method comprises the steps of obtaining three multidimensional state optimal estimation sequence sequences of {3s, 4s, 5s }, {2s, 3s, 4s }, {1s, 2s, 3s }, selecting the state optimal estimation sequence of {3s, 4s, 5s }, and explaining, wherein a nuclear reactor has state optimal estimation data at each moment in the state optimal estimation sequence, extracting 100 particles in the nuclear reactor at each moment as samples to obtain the state data of 100 particles at each moment, and obtaining 300 state data in the state optimal estimation sequence of {3s, 4s, 5s }, wherein the 300 state data jointly form a state estimation matrix 1. Similarly, 300 state data can be obtained from the other two sequences {2s, 3s, 4s }, {1s, 2s, 3s }, respectively, to form a state estimation matrix 2 and a state optimal estimation matrix 3.
And S400, inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result.
The preset fault classification model is a model of a preset neural network with a fault classification function; confidence is the probability that an overall parameter value falls within a certain region of the sample statistics when the overall parameter is estimated by sampling. Specifically, the state of the nuclear reactor changes in real time, and the state monitoring data also changes continuously, so in order to ensure the accuracy of the state fault classification result, all state data acquired at different sampling moments need to be used as the input of a preset fault classification model, deep learning is performed in a deep residual shrinkage network in the preset fault classification model, the state fault classification result and the confidence coefficient corresponding to the state fault classification result can be effectively identified according to multi-dimensional state data in a state estimation matrix, and thus, preliminary fault classification diagnosis is realized. Similarly, the 5 th time is taken as the current time to explain, the state estimation matrix 1, the state estimation matrix 2 and the state estimation matrix 3 which are obtained in the 5 th time are sequentially input into a deep residual shrinkage network in a preset fault classification model for deep learning, and the fault classification result 1, the fault classification result 2, the fault classification result 3 and the confidence corresponding to each fault classification result are respectively output. For further explanation, the deep residual shrinkage network in the preset fault classification model is described in the following: compared with the common deep learning method, the deep residual shrinkage network has the advantages that: the method is more suitable for feature extraction of the noise-containing data. Because the depth residual shrinking network adopts soft thresholding as a nonlinear layer in the structure, the method is a core step of signal noise reduction, and is equivalent to integrating noise reduction into a depth neural network as a trainable step.
S500, inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
The preset self-learning model is a model with self-learning common functions and comprising various types of neural networks. Specifically, different state optimal estimation matrices are from different state optimal estimation sequences, and state fault classification results obtained according to the different state optimal estimation matrices may not be consistent with each other, which is a result of comprehensive superposition due to measurement uncertainty, prediction result uncertainty, uncertainty of a deep neural network diagnosis model, and the like. In order to consider the difference of different measurement results, a self-learning model is constructed based on the idea of integrated learning stacki ng. The self-learning model is a secondary learning model in essence, and can be a machine learning model with strong random effect such as random forest, and can also be a learning model formed by a forward artificial neural network. In the self-learning model, different types of nuclear reactor fault classification diagnosis models are placed in a frame based on an integrated learning method, advantage complementation between different models is achieved, a previously obtained state fault classification result and a confidence coefficient corresponding to the state fault classification result are input into the pre-constructed self-learning model, a stacking technology is adopted, a plurality of previously obtained state fault classification results and confidence coefficients corresponding to the state fault classification results are used as basic diagnosis results, and after the plurality of state fault classification results and the confidence coefficients corresponding to the state fault classification results are stacked and combined under large sample space variation or large uncertainty, the obtained fault classification results and the confidence coefficients corresponding to the fault classification results of the nuclear reactor are used as final diagnosis results, for example: inputting the obtained fault classification result 1, the obtained fault classification result 2 and the obtained fault classification result 3 into a preset self-learning model, wherein the preset self-learning model may comprise a deep residual error network 1, a deep residual error network 2 or more other deep neural networks, and outputting a final fault classification result of the nuclear reactor at the current moment and the confidence coefficient of the fault classification result.
According to the nuclear reactor fault classification method, on one hand, accurate fault classification results can be obtained by acquiring the best state estimation data of the nuclear reactor at different sampling moments and integrating the data at different sampling moments, and the precision of fault classification diagnosis is greatly improved; and secondly, inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the fault classification result, and then inputting the obtained state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model, so that different fault classification results are collected, the accuracy of the finally output nuclear reactor fault classification result and the confidence coefficient corresponding to the fault classification result can be ensured, and the high-precision nuclear reactor fault classification result can be obtained.
In one embodiment, obtaining state-best-estimate data for a nuclear reactor at different sampling instants comprises:
step 1, acquiring state data of a nuclear reactor at different sampling moments;
and 2, fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
The state optimal estimation fusion algorithm is an algorithm for converting state optimal estimation for solving a plurality of measurement data into estimation values of posterior probability for satisfying state values to be measured according to the posterior estimation theory. Specifically, state data of the nuclear reactor at different sampling moments and different angles and different update periods are acquired, the state data are mainly divided into simulation state data and measurement state data, and the simulation state data and the measurement state data are fused through a state optimal estimation fusion algorithm to obtain state optimal estimation data.
In this embodiment, the optimal state estimation data obtained by fusing the state data of the nuclear reactor at different sampling moments through the optimal state estimation algorithm is extracted from the state data at different angles and different update periods, and compared with single state data, the optimal state estimation data can more accurately estimate the state of the nuclear reactor system at each sampling moment, thereby ensuring that accurate fault classification results can be obtained subsequently to a certain extent, and improving the precision of fault classification diagnosis.
In one embodiment, the status data includes:
high precision simulation state data, real-time measurement state data, and periodic measurement state data.
Wherein the high-precision simulation state data is data generated by nuclear energy design professional software. The design software is established based on a physical equation of a first principle and is subjected to complete test, verification and confirmation. These software provide high-precision calculation results of nuclear energy systems under various conditions to ensure the safety of the designed nuclear reactor; the real-time simulation state data is data obtained by a real-time simulation model, and the real-time simulation model is used for reducing, simplifying and equivalence on a fine model of high-precision simulation so as to realize real-time simulation of the reactor within the acceptable precision of engineering, such as coarser spatial grid division, conversion of a three-dimensional model into a point stack model and the like. The real-time measurement state data is composed of thousands of actual measurement hardware signals in a nuclear energy system and is refreshed at a certain frequency (such as 0.1s), and the comparison analysis of the real-time measurement data and a setting value (or curve) is used as a signal for early warning, control and protection to realize the operation action at a specific component or assembly level. The periodic measurement state data is experimental measurement state data of higher precision periodically developed under given experimental conditions. Specifically, the state data of the four different angles are acquired in different modes at each sampling moment, and the state of the nuclear reactor system can be represented to a certain extent. Comparing the four data can judge whether the nuclear reactor system has faults, the position and the reason of the faults, and the like, because the four parameters comprise hidden variables (such as reactivity coefficient, deviation of nucleate boiling, maximum linear power density, maximum fuel rod and cladding temperature) which cannot be directly measured but are very relevant to safe operation, and the like.
In this embodiment, the state data of the nuclear reactor includes high-precision simulation state data, real-time measurement state data, and periodic measurement state data, and the accuracy of the state optimal estimation data obtained by fusing the state data of the above four different angles and different update periods is higher, so that the precision of the finally obtained nuclear reactor fault classification result can be further ensured.
In one embodiment, as shown in fig. 3, the obtaining the optimal state estimation data of the nuclear reactor at different sampling times by fusing the state data at different sampling times through a state optimal estimation fusion algorithm includes:
s320, acquiring the best estimation data of the state of the nuclear reactor at the previous moment at different sampling moments and the confidence corresponding to the best estimation data of the state at the previous moment;
s340, obtaining state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient corresponding to the state optimal estimation data at the previous moment;
s360, acquiring state measurement data of the nuclear reactor at different sampling moments;
and S380, fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments.
Specifically, as shown in fig. 4, based on the optimal estimation of the state of the nuclear reactor at the time T0 before each sampling time, prediction data and confidence (prior probability) of the nuclear reactor state at each sampling time T1 can be obtained based on high-precision simulation or real-time simulation software. After the measured data of the nuclear reactor state at the time T1 is obtained, it is necessary to perform fusion calculation of the measured data and the predicted data according to the uncertainty of the measurement and the prior confidence of the predicted data, and obtain the state optimal estimated data and the uncertainty (posterior probability) thereof at each sampling time T1. Then starting at time T1, the state posteriori at time T2 is estimated until the entire reactor operation is tracked.
In this embodiment, the state prediction data at the sampling time is obtained according to the state optimal data at the previous time of the sampling time, and the obtained state prediction data is fused with the obtained state measurement data, so that the efficiency of data processing can be greatly improved, more accurate state optimal estimation data can be obtained, and the precision of the final nuclear reaction final fault classification result can be ensured.
In one embodiment, obtaining the state prediction data of the nuclear reactor at different sampling moments according to the state best estimation data at the previous moment and the confidence degrees corresponding to the state best estimation data at the previous moment comprises:
step 1, acquiring previous-time state data of particles in a nuclear reactor at different sampling moments according to the previous-time state optimal estimation data and confidence coefficients of the state optimal estimation data;
step 2, inputting the state data of the particles at the previous moment of different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments;
and 3, obtaining state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at different sampling moments.
The state prediction data is predicted from the previous moment and is the state data of the current moment. In particular, the optimal estimate of the state of the nuclear reactor at a time t preceding a given sampling time is stAnd its confidence (covariance matrix) PtThe average error e of prediction of a known simulation prediction model (including a high-precision simulation model or a real-time simulation model)mAnd extracting N particles in the nuclear reactor on the premise of confidence coefficient (covariance matrix) Q, and acquiring state data of each particle at the previous moment of each sampling momentWhere φ is a multivariate Gaussian-distributed sampling function, i represents the ith particle, and N particles characterizing the prediction error of the nuclear reactor state are constructedWherein phi also represents a sampling function of multivariate Gaussian distribution, sequentially inputting the acquired state data of each particle at the previous moment of the sampling moment into a preset simulation prediction model, and outputting the state prediction data of each particle at each sampling momentFor predicting the state of each particle at each sampling instant, where F represents the accuracy of the high or low accuracy simulation model. Predicting data based on the state of the extracted particles at each sampling timeObtaining state prediction data of the whole nuclear reactor at each sampling timeFor predicting the state of the nuclear reactor at the sampling moment, the following formula can be used for calculation:
in this embodiment, the state estimation data of the particles in the nuclear reactor at the sampling time can be quickly obtained by inputting the state estimation data of the particles in the nuclear reactor at the time before the sampling time into the preset simulation prediction model, and the state prediction data of the particles in the nuclear reactor can be collected to obtain accurate state prediction data of the nuclear reactor at the same sampling time.
In one embodiment, as shown in fig. 5, S380 includes:
s382, obtaining confidence degrees corresponding to state prediction data of the nuclear reactor at different sampling moments;
s384, acquiring the state data of the particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data;
s386, obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data;
s388, acquiring actual state measurement data of the nuclear reactor at different sampling moments;
s390, obtaining residual error coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data;
s392, obtaining a gain coefficient of the state measurement data according to the confidence corresponding to the estimated state measurement data;
and S394, fusing the state prediction data and the state measurement data according to the gain coefficient and the residual error coefficient to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
The estimated state measurement data is data of a detector measuring the state of the nuclear reactor, which is obtained by calculation through a correlation formula and is used for estimating the state measurement data of the detector; the actual state measurement data is actual state measurement data actually measured by a detector; the residual coefficient is a coefficient for characterizing an error between the estimated state measurement data and the actual state measurement data. Specifically, the state prediction data obtained according to the previous embodimentThe confidence degree corresponding to the state prediction data can be calculated by the following formula
After obtaining the confidence degree of the state prediction data of the nuclear reactor at the sampling time and the corresponding state prediction data, extracting N particles in the nuclear reactor through Gaussian sampling, and obtaining the state data of the N particles at the sampling timeThen the obtainedMapping the state space into a measurement space to obtainWhere h is the observation function of the detector. According to obtainingEstimating the state data of the nuclear reactor at the sampling time t +1 measured by the detector by the following formula, namely estimating the state measurement dataAnd a confidence level St+1:
Wherein,Rmesthe measurement uncertainty of the detector is present, and since different types of detectors or the same type of detectors deployed at different positions may cause variation in measurement accuracy of different detectors, the measurement uncertainty is present. Acquiring actual state measurement data of a nuclear reactor at a sampling moment, which is actually measured by a detectorBy measuring data of actual state to be acquiredAnd the estimated state measurement data obtained by calculationComparing to obtain the residual error of the state measurement dataAccording to the confidence degree corresponding to the estimated state measurement dataCalculating a gain coefficient of the state measurement data:according to the gain coefficient and the residual coefficient of the state measurement data, the fusion of the state prediction data and the state measurement data can be realized, so that the state optimal estimation data s of the nuclear reactor at the sampling moment can be obtainedt+1And confidence degree P corresponding to the state best estimation datat+1And s andt+1and Pt+1The correction may be based on the residual coefficients and the gain coefficients, i.e.:
in this embodiment, the gain coefficient and the optimal state estimation data s of the state measurement data are obtained according to the residual coefficient of the state measurement data obtained by comparing the estimated state measurement data and the actual state measurement data of the nuclear reactor and the confidence coefficient corresponding to the estimated state measurement datat+1And confidence degree P corresponding to the state best estimation datat+1The method can be used for correcting based on the residual coefficient and the gain coefficient, and can ensure the accuracy of the state optimal estimation data and the confidence coefficient thereof when the measurement state data is irregularly changed, thereby ensuring that the accurate fault classification result can be finally obtained.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a nuclear reactor fault classification device for realizing the nuclear reactor fault classification method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so specific limitations in one or more embodiments of the nuclear reactor fault classification device provided below can be referred to the limitations on the nuclear reactor fault classification method in the foregoing, and details are not repeated herein.
In one embodiment, as shown in fig. 6, there is provided a nuclear reactor fault classification apparatus comprising: an acquire data module 100, a process data module 200, and a fault classification module 300, wherein:
an obtaining data module 100, configured to obtain optimal state estimation data of the nuclear reactor at different sampling moments, where the optimal state estimation data includes data of multiple dimensions;
the data processing module 200 is used for generating a state optimal estimation sequence with multiple dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
the fault classification module 300 is configured to input the state estimation matrix into a preset fault classification model, so as to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
On one hand, the nuclear reactor fault classification device can obtain accurate fault classification results by acquiring the best state estimation data of the nuclear reactor at different sampling moments and integrating the data at different sampling moments, so that the precision of fault classification diagnosis is greatly improved; and secondly, inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the fault classification result, and then inputting the obtained state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model, so that different fault classification results are collected, the accuracy of the finally output nuclear reactor fault classification result and the confidence coefficient corresponding to the fault classification result can be ensured, and the high-precision nuclear reactor fault classification result can be obtained.
In one embodiment, the data acquisition module 100 is further configured to acquire status data of the nuclear reactor at different sampling times; and fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the data obtaining module 100 is further configured to obtain the best estimation data of the state of the nuclear reactor at the previous time at different sampling times and the confidence degrees corresponding to the best estimation data of the state at the previous time; obtaining state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient corresponding to the state optimal estimation data at the previous moment; acquiring state measurement data of a nuclear reactor at different sampling moments; and fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the data obtaining module 100 is further configured to obtain the state data of the particles in the nuclear reactor at the previous time at different sampling times according to the confidence of the state optimal estimation data and the state optimal estimation data at the previous time; and inputting the state data of the particles at the previous moment at different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments to obtain the state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at different sampling moments.
In one embodiment, the data obtaining module 100 is further configured to obtain confidence levels corresponding to state prediction data of the nuclear reactor at different sampling moments; acquiring state data of particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data; obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data; acquiring actual state measurement data of the nuclear reactor at different sampling moments; obtaining residual coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data; and according to the confidence coefficient corresponding to the estimated state measurement data, obtaining a gain coefficient of the state measurement data, and fusing the state prediction data and the state measurement data according to the gain coefficient and the residual coefficient to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
The various modules of the nuclear reactor fault classification apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is for storing nuclear reactor fault classification data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executable by a processor to implement a nuclear reactor fault classification method.
Those skilled in the art will appreciate that the architecture shown in fig. 7 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions; generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions; inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
acquiring state data of a nuclear reactor at different sampling moments; and fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
acquiring the best estimation data of the state of the nuclear reactor at the previous moment at different sampling moments and the confidence corresponding to the best estimation data of the state at the previous moment; according to the state optimal estimation data at the previous moment and the confidence degree corresponding to the state optimal estimation data at the previous moment, state prediction data of the nuclear reactor at different sampling moments are obtained, state measurement data of the nuclear reactor at different sampling moments are obtained, and the state prediction data and the state measurement data are fused through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
acquiring state data of the particles in the nuclear reactor at the previous moment at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient of the state optimal estimation data; inputting the state data of the particles at the previous moment of different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments; and obtaining state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at the different sampling moments.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
obtaining confidence degrees corresponding to state prediction data of the nuclear reactor at different sampling moments; acquiring state data of particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data; obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data; acquiring actual state measurement data of the nuclear reactor at different sampling moments; obtaining residual coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data; obtaining a gain coefficient of the state measurement data according to the confidence coefficient corresponding to the estimated state measurement data; and according to the gain coefficient and the residual coefficient, fusing the state prediction data and the state measurement data to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions; generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions; inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring state data of a nuclear reactor at different sampling moments; and fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring the best estimation data of the state of the nuclear reactor at the previous moment at different sampling moments and the confidence corresponding to the best estimation data of the state at the previous moment; obtaining state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient corresponding to the state optimal estimation data at the previous moment; acquiring state measurement data of a nuclear reactor at different sampling moments; and fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring state data of the particles in the nuclear reactor at the previous moment at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient of the state optimal estimation data; and inputting the state data of the particles at the previous moment at different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments to obtain the state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at different sampling moments.
In one embodiment, the computer program when executed by the processor further performs the steps of:
obtaining confidence degrees corresponding to state prediction data of the nuclear reactor at different sampling moments; acquiring state data of particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data; obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data; acquiring actual state measurement data of the nuclear reactor at different sampling moments; obtaining residual coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data; and according to the confidence coefficient corresponding to the estimated state measurement data, obtaining a gain coefficient of the state measurement data, and fusing the state prediction data and the state measurement data according to the gain coefficient and the residual coefficient to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:
acquiring state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions; generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions; inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring state data of a nuclear reactor at different sampling moments; and fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring the best estimation data of the state of the nuclear reactor at the previous moment at different sampling moments and the confidence corresponding to the best estimation data of the state at the previous moment; obtaining state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient corresponding to the state optimal estimation data at the previous moment; acquiring state measurement data of a nuclear reactor at different sampling moments; and fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring state data of the particles in the nuclear reactor at the previous moment at different sampling moments according to the state optimal estimation data at the previous moment and the confidence coefficient of the state optimal estimation data; and inputting the state data of the particles at the previous moment at different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments to obtain the state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at different sampling moments.
In one embodiment, the computer program when executed by the processor further performs the steps of:
obtaining confidence degrees corresponding to state prediction data of the nuclear reactor at different sampling moments; acquiring state data of particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data; obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data; acquiring actual state measurement data of the nuclear reactor at different sampling moments; obtaining residual coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data; and according to the confidence coefficient corresponding to the estimated state measurement data, obtaining a gain coefficient of the state measurement data, and fusing the state prediction data and the state measurement data according to the gain coefficient and the residual coefficient to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
It should be noted that, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A method of nuclear reactor fault classification, the method comprising:
acquiring state optimal estimation data of a nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
generating a state optimal estimation sequence with a plurality of dimensions based on a preset period according to the state optimal estimation data;
forming a state estimation matrix according to the state optimal estimation sequences of the multiple dimensions;
inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result;
and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
2. The method of claim 1, wherein the obtaining state-optimal estimation data for the nuclear reactor at different sampling instants comprises:
acquiring state data of the nuclear reactor at different sampling moments;
and fusing the state data at different sampling moments through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
3. The method of claim 2, wherein the status data comprises:
high precision simulation state data, real-time measurement state data, and periodic measurement state data.
4. The method of claim 2, wherein the fusing the state data at different sampling times through a state optimal estimation fusion algorithm to obtain the state optimal estimation data of the nuclear reactor at different sampling times comprises:
acquiring the best estimation data of the state of the nuclear reactor at the previous moment at different sampling moments and the confidence degree corresponding to the best estimation data of the state at the previous moment;
obtaining state prediction data of the nuclear reactor at different sampling moments according to the state optimal estimation data at the previous moment and the confidence degree corresponding to the state optimal estimation data at the previous moment;
acquiring state measurement data of the nuclear reactor at different sampling moments;
and fusing the state prediction data and the state measurement data through a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments.
5. The method of claim 4, wherein obtaining the predicted state data of the nuclear reactor at different sampling times according to the best state estimation data at the previous time and the confidence degrees corresponding to the best state estimation data at the previous time comprises:
acquiring state data of the particles in the nuclear reactor at the previous moment at different sampling moments according to the state best estimation data of the previous moment and the confidence coefficient of the state best estimation data;
inputting the state data of the particles at the previous moment of different sampling moments into a preset simulation prediction model, and outputting the state prediction data of the particles at different sampling moments;
and obtaining the state prediction data of the nuclear reactor at different sampling moments according to the state prediction data of the particles at different sampling moments.
6. The method of claim 4, wherein the fusing the state prediction data and the state measurement data by a state optimal estimation fusion algorithm to obtain state optimal estimation data of the nuclear reactor at different sampling moments comprises:
obtaining confidence degrees corresponding to state prediction data of the nuclear reactor at different sampling moments;
acquiring state data of the particles in the nuclear reactor at different sampling moments according to the state prediction data and the confidence degrees corresponding to the state prediction data;
obtaining estimated state measurement data of the nuclear reactor at different sampling moments and confidence degrees corresponding to the estimated state measurement data according to the state data;
acquiring actual state measurement data of the nuclear reactor at different sampling moments;
obtaining residual error coefficients of the state measurement data of the nuclear reactor at different sampling moments according to the actual state measurement data and the estimated state measurement data;
obtaining a gain coefficient of the state measurement data according to the confidence corresponding to the estimated state measurement data;
and according to the gain coefficient and the residual error coefficient, fusing the state prediction data and the state measurement data to obtain the state optimal estimation data of the nuclear reactor at different sampling moments.
7. A nuclear reactor fault classification apparatus, the apparatus comprising:
the data acquisition module is used for acquiring the state optimal estimation data of the nuclear reactor at different sampling moments, wherein the state optimal estimation data comprises data of multiple dimensions;
the data processing module is used for generating a state optimal estimation sequence with multiple dimensions based on a preset period according to the state optimal estimation data; forming a state estimation matrix according to the state optimal estimation sequences of multiple dimensions;
the fault classification module is used for inputting the state estimation matrix into a preset fault classification model to obtain a state fault classification result and a confidence coefficient corresponding to the state fault classification result; and inputting the state fault classification result and the confidence coefficient corresponding to the state fault classification result into a preset self-learning model to obtain the fault classification result of the nuclear reactor and the confidence coefficient corresponding to the fault classification result.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 6 when executed by a processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111653107.2A CN114358172A (en) | 2021-12-30 | 2021-12-30 | Nuclear reactor fault classification method, apparatus, computer device, and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111653107.2A CN114358172A (en) | 2021-12-30 | 2021-12-30 | Nuclear reactor fault classification method, apparatus, computer device, and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114358172A true CN114358172A (en) | 2022-04-15 |
Family
ID=81103317
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111653107.2A Pending CN114358172A (en) | 2021-12-30 | 2021-12-30 | Nuclear reactor fault classification method, apparatus, computer device, and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114358172A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116430817A (en) * | 2023-04-26 | 2023-07-14 | 同心县启胜新能源科技有限公司 | Data acquisition processing method and system applied to photovoltaic module production system |
-
2021
- 2021-12-30 CN CN202111653107.2A patent/CN114358172A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116430817A (en) * | 2023-04-26 | 2023-07-14 | 同心县启胜新能源科技有限公司 | Data acquisition processing method and system applied to photovoltaic module production system |
CN116430817B (en) * | 2023-04-26 | 2023-09-29 | 同心县启胜新能源科技有限公司 | Data acquisition processing method and system applied to photovoltaic module production system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Qin et al. | Macroscopic–microscopic attention in LSTM networks based on fusion features for gear remaining life prediction | |
JP2022500745A (en) | Computer implementation methods, computer program products and systems for anomaly detection and / or predictive maintenance | |
Dai et al. | A reliability evaluation model of rolling bearings based on WKN-BiGRU and Wiener process | |
CN110119854A (en) | Voltage-stablizer water level prediction method based on cost-sensitive LSTM Recognition with Recurrent Neural Network | |
Tolo et al. | Robust on-line diagnosis tool for the early accident detection in nuclear power plants | |
Yong-kuo et al. | A cascade intelligent fault diagnostic technique for nuclear power plants | |
Kim et al. | Reactor vessel water level estimation during severe accidents using cascaded fuzzy neural networks | |
Zhu et al. | A coupled model for dam foundation seepage behavior monitoring and forecasting based on variational mode decomposition and improved temporal convolutional network | |
CN114358172A (en) | Nuclear reactor fault classification method, apparatus, computer device, and storage medium | |
Racheal et al. | Evaluation of optimized machine learning models for nuclear reactor accident prediction | |
CN114003422A (en) | Host anomaly detection method, computer device, and storage medium | |
Amin et al. | Probabilistic monitoring of nuclear plants using R-vine copula | |
CN117195044A (en) | Fault analysis method, device, computer equipment, storage medium and program product | |
Ma et al. | A GRU network-based approach for steam drum water level predictions | |
WO2024087404A1 (en) | Nuclear reactor fault determination method, apparatus, device, storage medium, and product | |
Ding et al. | Deep Forest‐Based Fault Diagnosis Method for Chemical Process | |
CN116610973A (en) | Sensor fault monitoring and failure information reconstruction method and system | |
CN115859783A (en) | Multi-fidelity network construction method and device for nuclear reactor simulation test | |
CN114298413A (en) | Hydroelectric generating set runout trend prediction method | |
Najar et al. | Comparative Machine Learning Study for Estimating Peak Cladding Temperature in AP1000 Under LOFW | |
Guohua et al. | Distributed fault diagnosis framework for nuclear power plants | |
Chen et al. | Research on the prediction of LOCA condition in nuclear power plants based on GRU recurrent neural network and its variants | |
Zhao et al. | Knowledge-Informed Uncertainty-Aware Machine Learning for Time Series Forecasting of Dynamical Engineered Systems | |
CN117786586B (en) | Self-locking nut detection system and method thereof | |
Pastore | Implementation of two computational techniques for the detection of abnormal conditions in safety-critical systems: an application to the simulator of a nuclear fission plant |
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 |