CN113077915A - Rod position detector and control rod displacement measuring method - Google Patents
Rod position detector and control rod displacement measuring method Download PDFInfo
- Publication number
- CN113077915A CN113077915A CN202110326229.4A CN202110326229A CN113077915A CN 113077915 A CN113077915 A CN 113077915A CN 202110326229 A CN202110326229 A CN 202110326229A CN 113077915 A CN113077915 A CN 113077915A
- Authority
- CN
- China
- Prior art keywords
- displacement
- coil
- control rod
- position detector
- neural network
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
- G21C—NUCLEAR REACTORS
- G21C17/00—Monitoring; Testing ; Maintaining
- G21C17/10—Structural combination of fuel element, control rod, reactor core, or moderator structure with sensitive instruments, e.g. for measuring radioactivity, strain
- G21C17/12—Sensitive element forming part of control element
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E30/00—Energy generation of nuclear origin
- Y02E30/30—Nuclear fission reactors
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Physiology (AREA)
- Measurement Of Length, Angles, Or The Like Using Electric Or Magnetic Means (AREA)
- Plasma & Fusion (AREA)
- High Energy & Nuclear Physics (AREA)
- Genetics & Genomics (AREA)
Abstract
The invention discloses a rod position detector and a control rod displacement measuring method, wherein the rod position detector comprises a coil framework, a detection coil, a first compensation coil and a second compensation coil; the detection coil is wound on the outer wall of the whole coil framework; one end of the first compensation coil is connected with one end of the detection coil; one end of the second compensation coil is connected with the other end of the detection coil. The control rod displacement measurement method includes acquiring the temperature of the environment where the control rod to be measured is located, measuring the ratio of the voltage of the rod position detector to the resistance voltage when the control rod to be measured moves, and inputting the temperature of the environment where the control rod to be measured is located and the ratio into the displacement neural network model to obtain the displacement of the control rod to be measured. The rod position detector is high in sensitivity and low in failure rate, the control rod displacement measuring method can accurately measure the displacement of the control rod, and the rod position detector is high in reliability and high in precision.
Description
Technical Field
The invention relates to the field of control rods, in particular to a rod position detector and a control rod displacement measuring method.
Background
The control rod is mainly composed of neutron absorption components, is one of core components of the nuclear reactor, is also the only movable component in the core structure, and is mainly used for completing the startup, power and temperature regulation, normal shutdown and the like of the nuclear reactor. Control rods play a crucial role in the safety and normal operation of nuclear reactors, and shutdown maintenance of nuclear power plants is required once a rod position control system is out of control or fails, which affects the power supply plan and the economic income of the nuclear power plants.
The rod position detector is generally positioned at the top of the pressure vessel, the pressure vessel is an important barrier for ensuring the safety of the reactor, direct interfaces on the pressure vessel are required to be reduced as much as possible in order to ensure the integrity of the boundary of the pressure vessel, and in addition, the rod position detector works in a high-temperature, high-pressure and high-radiation environment, so the rod position measurement is usually carried out by an indirect method instead of a direct method, and the measurement difficulty is increased.
Inductive rod position detectors commonly used in most pressurized water reactor nuclear power plants are used to measure control rod position primarily through changes in induced voltage in the coil. The rod position detector is installed outside the stroke casing pipe, its detecting coil is formed by combining several groups of short coils according to a certain order, when a control rod driving rod made of magnetic conductive material moves in the stroke casing pipe, the coil of the rod position detector will be penetrated by the driving rod, when a certain coil is penetrated, the coil inductance is increased, the voltage corresponding to the coil will be changed, and these coil voltages are compared and shaped, and then several bit binary codes are outputted to represent the rod position, so that said inductive rod position detector also can be called as coding type rod position detector. At present, the detection precision of an encoding type rod position detector used by an AP1000 nuclear power station is high, the theoretical precision reaches +/-3 steps (15.875 mm in each mechanical step), and the actual precision is +/-5 steps, namely +/-79.375 mm.
The coded rod position detector has the problems of excessive coils, low coil utilization rate, more welding spots among the coils, more connecting cables, more occupied electrical penetration pieces of a containment vessel and the like, so that the rod position detector has frequent failure, low reliability and low detection precision, and the displacement error of a control rod measured by the rod position detector is large.
Disclosure of Invention
In view of the above, an object of the present invention is to overcome the defects in the prior art, and provide a rod position detector and a control rod displacement measurement method, where the rod position detector has high sensitivity and low failure rate, and the control rod displacement measurement method can accurately measure the displacement of a control rod, and has strong reliability and high precision.
A rod position detector comprises a coil framework, a detection coil, a first compensation coil and a second compensation coil;
the detection coil is wound on the outer wall of the whole coil framework;
one end of the first compensation coil is connected with one end of the detection coil, and the first compensation coil is wound at one end of the detection coil;
one end of the second compensation coil is connected with the other end of the detection coil, and the second compensation coil is wound at the other end of the detection coil;
the detection coils are in a plurality of layers, and the winding directions of the coils of each layer in the detection coils are consistent; and the winding directions of the first compensation coil and the second compensation coil are consistent with the winding direction of the detection coil.
Further, the detection coil is 2 layers.
Further, the first compensation coil and the second compensation coil are both 2 layers.
Further, the detection coil comprises a first detection coil and a second detection coil, and the first detection coil and the second detection coil are wound on the outer wall of the whole coil framework in parallel;
the first compensation coil comprises a compensation coil A and a compensation coil B, and the compensation coil A and the compensation coil B are wound at one end of the detection coil in parallel; one end of the compensation coil A is connected with one end of the first detection coil, and the compensation coil A is wound at one end of the first detection coil; one end of the compensation coil B is connected with one end of the second detection coil, and the compensation coil B is wound at one end of the second detection coil;
the second compensation coil comprises a compensation coil C and a compensation coil D, and the compensation coil C and the compensation coil D are wound at the other end of the detection coil in parallel; one end of the compensation coil C is connected with the other end of the first detection coil, and the compensation coil C is wound at the other end of the first detection coil; one end of the compensation coil D is connected with the other end of the second detection coil, and the compensation coil D is wound at the other end of the second detection coil.
A control rod displacement measurement method based on a rod position detector comprises the following steps:
s1, constructing a displacement neural network model;
s2, constructing a displacement detection circuit; the displacement detection circuit comprises a rod position detector and a reference resistor; one end of the rod position detector is connected with one pole of the power supply, the other end of the rod position detector is connected with one end of the reference resistor, and the other end of the reference resistor is connected with the other pole of the power supply;
s3, acquiring the temperature of the environment where the control rod is located, measuring the actual displacement of the control rod, inputting the temperature of the environment where the control rod is located, the ratio of the voltage of a rod position detector in a displacement detection circuit to the voltage of a reference resistor into a displacement neural network model for training, and after iteration setting times, enabling the error between the output displacement and the actual displacement of the displacement neural network model to be smaller than a set threshold value;
and S4, acquiring the temperature of the environment where the control rod to be detected is located, measuring the ratio of the rod position detector voltage to the reference resistor voltage in the displacement detection circuit when the control rod to be detected moves, and inputting the temperature of the environment where the control rod to be detected is located and the ratio into a displacement neural network model to obtain the displacement of the control rod to be detected.
Further, step S3 specifically includes:
s31, acquiring the temperature of the environment where the control rod is located, measuring the actual displacement of the control rod at the temperature and measuring the ratio of the rod position detector voltage in the displacement detection circuit corresponding to the actual displacement to the reference resistor voltage;
s32, respectively carrying out normalization processing on the temperature, the actual displacement and the ratio to obtain the normalized temperature, the normalized actual displacement and the normalized ratio;
s33, initializing the displacement neural network model to obtain an initialized displacement neural network model;
s34, inputting the normalized temperature and the normalized ratio as samples into an initialized displacement neural network model, and outputting the predicted displacement of a control rod;
s35, judging whether the error between the predicted displacement of the control rod and the normalized actual displacement is smaller than a set threshold value or not, and if so, ending the operation; if not, go to step S36;
the error between the predicted displacement of the control rod and the normalized actual displacement is as follows:
wherein e is the error between the predicted displacement of the control rod and the normalized actual displacement; o is3kmIs the predicted displacement of the control rod; xmIs the normalized actual displacement of the control rod; k is the number of neuron in output layer in the displacement neural network model; k is the number of output layer neurons in the displacement neural network model; m is the serial number of the sample in the displacement neural network model; m is the number of samples in the displacement neural network model;
s36, judging whether the number of times of executing the step S34 is less than the set number of times UNIf yes, updating the initialized displacement neural network model, and returning to execute the step S34; if not, the process is ended.
Further, in step S31, a ratio of the rod position detector voltage to the reference resistance voltage in the displacement detection circuit corresponding to the actual displacement is measured according to the following steps:
s311, performing full-wave rectification processing on the rod position detector voltage and the reference resistor voltage respectively to obtain a sine half-wave signal of the rod position detector and a sine half-wave signal of the reference resistor;
s312, low-pass filtering is carried out on the sine half-wave signal of the rod position detector and the sine half-wave signal of the reference resistor respectively to obtain a direct-current voltage signal of the rod position detector and a direct-current voltage signal of the reference resistor;
s313, dividing the direct current voltage signal of the rod position detector and the direct current voltage signal of the reference resistor to obtain a ratio signal of the voltage of the rod position detector and the voltage of the reference resistor;
s314, filtering the ratio signal to enable the amplitude of the ratio signal to meet a set threshold value, and taking the value of the ratio signal meeting the set threshold value as the ratio of the rod position detector voltage to the reference resistor voltage.
Further, in step S33, initializing the displacement neural network model, specifically including:
s331, carrying out random assignment on the neural parameters of the displacement neural network model to obtain initial values of the neural parameters; the neural parameters comprise weight values, translation factors and scaling factors;
s332, taking each parameter in the neural parameters as a gene, and arranging the genes to form a chromosome;
s333, repeatedly executing the steps S331-S332N times to obtain N chromosomes, and using the N chromosomes as a group;
s334, determining the fitness sequence of the population (f)1,f2,...,fi,...,fN) (ii) a Wherein f isiFitness of the ith chromosome;
s335, acquiring the maximum fitness in the fitness sequence of the group, judging whether the maximum fitness is larger than a set fitness threshold value, and if so, taking the neural parameter corresponding to the chromosome with the maximum fitness as an initial parameter of the displacement neural network model; if not, go to step S336;
S336. judging whether the number of times of executing the step S334 is less than the set number VNIf so, selecting and/or hybridizing and/or mutating the population to obtain a processed population, and returning to execute S334; if not, the process is ended.
Further, in step S334, a fitness sequence (f) of the population is determined1,f2,...,fi,...,fN) The method specifically comprises the following steps:
s3341, setting the neural network parameters corresponding to the ith chromosome in the population into a displacement neural network model, inputting the normalized temperature and the normalized ratio as samples into the displacement neural network model, and outputting the predicted displacement of a control rod;
s3342, calculating fitness f of ith chromosomei:
Wherein, O3kmIs the predicted displacement of the control rod; xmIs the actual displacement of the control rod; k is the number of neuron in output layer in the displacement neural network model; k is the number of output layer neurons in the displacement neural network model; m is the serial number of the sample in the displacement neural network model; m is the number of samples in the displacement neural network model;
s3343, analogizing according to the steps S3341-S3342 to obtain a group fitness sequence (f)1,f2,...,fi,...,fN)。
Further, the ratio of the rod position detector voltage to the reference resistance voltage is:
wherein, UBFor alternating voltage of rod position detector, UAIs the alternating voltage of the reference resistor, R1Is the resistance value of the reference resistor, R2The resistance value of a reference resistor of the detection coil, f being an AC constant current sourceThe frequency, L, is the inductance of the detection coil.
The invention has the beneficial effects that: according to the rod position detector and the control rod displacement measuring method, the rod position detector is high in sensitivity and low in failure rate, the control rod displacement measuring method can accurately measure the displacement of the control rod, and the control rod displacement measuring method is high in reliability and high in precision.
Drawings
The invention is further described below with reference to the following figures and examples:
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a schematic diagram of a displacement detection circuit according to the present invention;
FIG. 3 is a schematic diagram of a voltage signal processing flow according to the present invention;
FIG. 4 is a diagram of a shift neural network model architecture of the present invention;
FIG. 5 is a schematic view of the rod position detector of the present invention;
FIG. 6 is a schematic diagram of the control rod displacement measurement circuitry of the present invention;
wherein, in FIG. 5, 1-coil armature; 21-a first compensation coil; 22-a second compensation coil; 3-a detection coil; 4-control rod.
Detailed Description
The invention is further described with reference to the accompanying drawings, in which:
a rod position detector of the present invention, as shown in fig. 5, includes a bobbin 1, a detection coil 3, a first compensation coil 21 and a second compensation coil 22;
the detection coil 3 is wound on the outer wall of the whole coil framework 1;
one end of the first compensation coil 21 is connected with one end of the detection coil 3, and the first compensation coil 21 is wound at one end of the detection coil 3;
one end of the second compensation coil 22 is connected with the other end of the detection coil 3, and the second compensation coil 22 is wound at the other end of the detection coil 3;
the detection coils 3 are in a plurality of layers, the winding directions of the coils in each layer of the detection coils 3 are consistent, and the directions of the formed magnetic fields are the same, so that the magnetic field strengthening effect is achieved; the winding directions of the first compensation coil 21 and the second compensation coil 22 are consistent with the winding direction of the detection coil 3, so that the directions of the magnetic fields formed by the first compensation coil 21 and the first compensation coil 22 are the same as the direction of the magnetic field formed by the detection coil 3, and the function of enhancing the magnetic field at the end part of the detection coil 3 is achieved.
The coil framework 1 is made of a non-magnetic-conductive material, the detection coil 3, the first compensation coil 21 and the second compensation coil 22 are enameled wires, and the first compensation coil 21 and the detection coil 3 and the second compensation coil 22 and the detection coil 3 are connected in series in a welding mode;
the rod position detector is only provided with the primary coil and does not contain the secondary coil through the structure, the coil structure is simple, the number of cable joints and the number of electrical penetrating pieces of the containment shell are less, the reliability is enhanced, the fault rate is reduced, meanwhile, the linear measurement of the full detection stroke of the rod position detector is realized through compensating the magnetic field at the end part of the detection coil, the coil utilization rate is improved, and the detection blind area is eliminated.
In this embodiment, the detection coil 3 has 2 layers, which improves the inductance sensitivity.
In this embodiment, the first compensation coil 21 and the second compensation coil 22 are 2 layers, and each layer has 15 turns, so as to compensate the magnetic field at the end of the detection coil.
In this embodiment, the detection coil 3 includes a first detection coil and a second detection coil, and the first detection coil and the second detection coil are wound on the outer wall of the whole coil frame 1 in parallel;
the first compensation coil 21 comprises a compensation coil A and a compensation coil B, and the compensation coil A and the compensation coil B are wound at one end of the detection coil 3 in parallel; one end of the compensation coil A is connected with one end of the first detection coil, and the compensation coil A is wound at one end of the first detection coil; one end of the compensation coil B is connected with one end of the second detection coil, and the compensation coil B is wound at one end of the second detection coil;
the second compensation coil 22 comprises a compensation coil C and a compensation coil D, and the compensation coil C and the compensation coil D are wound on the other end of the detection coil 3 in parallel; one end of the compensation coil C is connected with the other end of the first detection coil, and the compensation coil C is wound at the other end of the first detection coil; one end of the compensation coil D is connected with the other end of the second detection coil, and the compensation coil D is wound at the other end of the second detection coil.
Through the structure, the rod position detector comprises the first detection coil, the compensation coil A and the compensation coil C, and further comprises the second detection coil, the compensation coil B and the compensation coil D, namely the rod position detector is provided with two sets of coils, when one set of the coil fails, the other set of the coil can be used as a standby scheme, and therefore the rod position detector is more stable and reliable.
It should be noted that the other ends of the compensation coils are connected according to the usage of the rod position detector, and are not described herein again.
A control rod displacement measuring method based on a rod position detector, as shown in fig. 1, includes the following steps:
s1, constructing a displacement neural network model; as shown in fig. 4, the displacement neural network model includes an input layer, a hidden layer, and an output layer; the input layer has 2 neurons, the hidden layer has 5 neurons, and the output layer has 1 neuron;
s2, constructing a displacement detection circuit; as shown in fig. 2, the displacement detection circuit includes a rod position detector and a reference resistor; the other end of the compensation coil A in the rod position detector is connected with one pole of a power supply, the other end of the compensation coil C in the rod position detector is connected with one end of a reference resistor, and the other end of the reference resistor is connected with the other pole of the power supply; wherein, the power supply is an alternating current constant current source;
s3, acquiring the temperature of the environment where the control rod is located, measuring the actual displacement of the control rod, inputting the temperature of the environment where the control rod is located, the ratio of the voltage of a rod position detector in a displacement detection circuit to the voltage of a reference resistor into a displacement neural network model for training, and after iteration setting times, enabling the error between the output displacement and the actual displacement of the displacement neural network model to be smaller than a set threshold value; wherein, the training of the displacement neural network model in the step S3 is processed by Matlab;
and S4, acquiring the temperature of the environment where the control rod to be detected is located, measuring the ratio of the rod position detector voltage to the reference resistor voltage in the displacement detection circuit when the control rod to be detected moves, and inputting the temperature of the environment where the control rod to be detected is located and the ratio into a displacement neural network model to obtain the displacement of the control rod to be detected.
In this embodiment, the step S3 specifically includes:
s31, selecting an initial temperature T according to the practical application condition1And at a temperature T1Based on 2 ℃ as step length, gradually increasing the ambient temperature to the maximum working temperature TnObtaining n temperature values; and under n temperature values, the control rods are processed as follows:
at a certain ambient temperature TiNext, moving the control rod once by taking 5mm as a step length, and recording the actual displacement of the control rod once, so as to obtain a plurality of actual displacements of the control rod; at a certain actual displacement X of the control rodmMeasuring the ratio of the rod position detector voltage to the reference resistance voltage in the displacement detection circuit for n times, and taking the average value of n ratios as the actual displacement XmRatio U of corresponding rod position detector voltage to reference resistance voltageoutThen the actual displacement X of the control rod can be obtainedmRatio U of corresponding rod position detector voltage to reference resistance voltageoutFurther obtaining the ratio of the rod position detector voltage and the reference resistance voltage corresponding to a plurality of actual displacements of the control rod respectively;
finally, n groups of data corresponding to n different temperature values can be obtained, wherein each group of data is temperature T, and a plurality of actual displacement sets (X) of the control rod measured under the temperature T1、X2、…、Xm…) and the ratio (U) of the rod position detector voltage to the reference resistance voltage corresponding to each of the plurality of actual displacementsout-1、Uout-2、…、Uout-m…), in the embodiment, the value of n is 10;
s32, normalizing the temperature and the actual displacement in the 10 groups of data and the ratio of the voltage of the rod position detector to the voltage of the reference resistor to obtain the normalized temperature, the normalized actual displacement and the normalized ratio; in the 10 groups of data, each group of data has a temperature value, and n temperature values are calculated in total, and the n temperature values are normalized to obtain normalized temperature values; similarly, all the actual displacements in the n groups of data are normalized to obtain normalized actual displacements, and similarly, the ratio of the normalized rod position detector voltage to the reference resistance voltage can also be obtained.
S33, initializing the displacement neural network model to obtain an initialized displacement neural network model;
s34, inputting the normalized temperature and the normalized ratio as samples into an initialized displacement neural network model, and outputting the predicted displacement of a control rod;
s35, judging whether the error between the predicted displacement of the control rod and the normalized actual displacement is smaller than a set threshold value or not, and if so, ending the operation; if not, go to step S36;
s36, judging whether the number of times of executing the step S34 is less than the set number of times UNIf yes, updating the initialized displacement neural network model, and returning to execute the step S34; if not, the process is ended. Wherein the number of times UNThe setting is carried out according to the actual application condition.
In this embodiment, in step S31, as shown in fig. 3, the ratio of the rod position detector voltage to the reference resistance voltage in the displacement detection circuit corresponding to the actual displacement is measured according to the following steps:
s311, respectively aligning the rod position detector voltage UBAnd a reference resistance voltage UAFull-wave rectification is carried out to obtain a sine half-wave signal U of the rod position detectorB1Sinusoidal half-wave signal U with reference resistanceA1;
S312, respectively aligning sine half-wave signals U of the rod position detectorB1Sinusoidal half-wave signal U with reference resistanceA1Low-pass filtering to obtain DC for rod position detectorPressure signal UB2DC voltage signal U with reference resistanceA2(ii) a Wherein the low-pass filtering is second-order low-pass filtering;
s313, direct current voltage signal U of rod position detectorB2DC voltage signal U with reference resistanceA2Dividing to obtain a ratio signal U of the rod position detector voltage and the reference resistance voltageB2/UA2;
S314, comparing the ratio signal UB2/UA2Filtering is carried out to enable a ratio signal UB2/UA2Satisfies a set threshold value, and will satisfy a ratio signal U of the set threshold valueB2/UA2The value of (D) is used as the ratio U of the rod position detector voltage to the reference resistance voltageout。
In this embodiment, in step S33, initializing the shift neural network model specifically includes:
s331, carrying out random assignment on the neural parameters of the displacement neural network model by using a random algorithm to obtain initial values of the neural parameters; the neural parameters comprise weight values, translation factors and scaling factors; wherein the weight comprises a weight ω between the ith neuron in the input layer of the neural network and the jth neuron in the hidden layerijAnd the weight value omega between the jth neuron in the hidden layer of the neural network and the kth neuron in the output layerjkThe number of the weights between the input layer and the hidden layer is 10, and the number of the weights between the hidden layer and the output layer is 5; the translation factor and the scaling factor are both parameters in the hidden layer, each neuron in the hidden layer corresponds to one translation factor and one scaling factor, and the number of the translation factors and the number of the scaling factors in the hidden layer are respectively 5; the random algorithm adopts the prior art and is not described herein again;
s332, taking each parameter in the neural parameters as a gene, and arranging the genes to form a chromosome; wherein the neural parameters comprise 15 weight values and 10 factors, the total number of the weight values and the total number of the factors is 25, each parameter corresponds to one gene, and a chromosome is obtained by arranging the 25 genes; the arrangement can adopt a random ordering rule;
s333, repeatedly executing the steps S331-S332N times to obtain N chromosomes, and using the N chromosomes as a group; wherein, the times N are set according to the actual application condition;
s334, determining the fitness sequence of the population (f)1,f2,...,fi,...,fN) (ii) a Wherein f isiFitness of the ith chromosome;
s335, acquiring the maximum fitness in the fitness sequence of the group, judging whether the maximum fitness is larger than a set fitness threshold value, and if so, taking the neural parameter corresponding to the chromosome with the maximum fitness as an initial parameter of the displacement neural network model; if not, go to step S336; the fitness threshold is set according to the actual application condition;
s336, judging whether the frequency of executing the step S334 is less than the set frequency VNIf so, selecting and/or hybridizing and/or mutating the population to obtain a processed population, and returning to execute S334; if not, the process is ended. The times V are set according to the actual application condition;
in this embodiment, in step S334, the fitness sequence (f) of the population is determined1,f2,...,fi,...,fN) The method specifically comprises the following steps:
s3341, setting the neural parameters corresponding to the ith chromosome in the population into a displacement neural network model, inputting the normalized temperature and voltage ratio as samples into the displacement neural network model, and outputting the predicted displacement of a control rod;
s3342, calculating fitness f of ith chromosomei:
Wherein, O3kmIs the predicted displacement of the control rod; xmIs the actual displacement of the control rod; k is the number of neuron in output layer in the displacement neural network model; k is output layer nerve in displacement neural network modelThe number of the elements, wherein K is 1; m is the serial number of the sample in the displacement neural network model; m is the number of samples in the displacement neural network model, and M is 10;
s3343, analogizing according to the steps S3341-S3342 to obtain a group fitness sequence (f)1,f2,...,fi,...,fN)。
In this embodiment, in step S35, the error between the predicted displacement of the control rod and the normalized actual displacement is:
wherein e is the error between the predicted displacement of the control rod and the normalized actual displacement; o is3kmIs the predicted displacement of the control rod; xmIs the normalized actual displacement of the control rod; k is the number of neuron in output layer in the displacement neural network model; k is the number of output layer neurons in the displacement neural network model, and is 1; m is the serial number of the sample in the displacement neural network model; and M is the number of samples in the displacement neural network model, and the M is 10.
In this embodiment, the ratio of the rod position detector voltage to the reference resistor voltage is:
wherein, UBTaking alternating current voltages at two ends of a detection coil as the voltage of the rod position detector during actual measurement; u shapeAIs the alternating voltage of the reference resistor, R1Is the resistance value of the reference resistor, R2Is the resistance value of the detection coil, f is the frequency of the AC constant current source, and L is the inductance of the detection coil.
In this embodiment, the step S4 specifically includes: measuring the temperature of the environment where the control rod to be measured is located by using a temperature sensor; and constructing a signal processing circuit which comprises a rectifying circuit, a second-order low-pass filter circuit, a division circuit, a conditioning circuit, an analog-to-digital conversion circuit, a DSP processing circuit, a digital-to-analog conversion circuit and a 4-20mA conversion circuit. And the DSP processing circuit is embedded with a displacement neural network model.
As shown in FIG. 6, the voltage U of the reference resistor in the displacement detection circuit is measuredAVoltage U of the rod position detectorBMake UBAnd UAFull-wave rectification is carried out through a rectification circuit to respectively obtain sine half-wave signals UB1And a half-wave signal UA1Then filtering the signal by a second-order low-pass filter circuit to obtain a DC voltage signal U with smaller ripple wavesB2And a DC voltage signal UA2(ii) a Then two direct current signals are input into a division circuit to be divided to obtain UB2/UA2Then, the U is put inB2/UA2Inputting the signal into a conditioning circuit for filtering to enable UB2/UA2The amplitude of the analog-to-digital conversion circuit meets the input voltage requirement of the analog-to-digital conversion circuit; the analog-to-digital conversion circuit converts an analog signal UB2/UA2The digital signal is converted into a 16-bit digital signal, the 16-bit digital signal is transmitted to a DSP processing circuit by using an SPI communication protocol, the temperature of the environment where the control rod to be detected is located is input into the DSP processing circuit, the DSP processing circuit can output the displacement of the control rod to be detected through a trained displacement neural network model embedded in the DSP processing circuit, and the DSP processing circuit transmits the digital signal representing the displacement to a 16-bit digital-to-analog conversion circuit by using the SPI communication protocol;
the digital-to-analog conversion circuit is used for completing conversion from the digital signal to an analog signal to obtain a converted voltage signal representing displacement, the voltage signal is input into the 4-20mA conversion circuit, the 4-20mA conversion circuit outputs a current signal representing displacement, and finally the current signal is output and displayed on a human-computer display interface.
Finally, the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be covered in the claims of the present invention.
Claims (10)
1. A rod position detector, characterized by: the device comprises a coil framework, a detection coil, a first compensation coil and a second compensation coil;
the detection coil is wound on the outer wall of the whole coil framework;
one end of the first compensation coil is connected with one end of the detection coil, and the first compensation coil is wound at one end of the detection coil;
one end of the second compensation coil is connected with the other end of the detection coil, and the second compensation coil is wound at the other end of the detection coil;
the detection coils are in a plurality of layers, and the winding directions of the coils of each layer in the detection coils are consistent; and the winding directions of the first compensation coil and the second compensation coil are consistent with the winding direction of the detection coil.
2. The rod position detector of claim 1, wherein: the detection coil is 2 layers.
3. The rod position detector of claim 1, wherein: the first compensation coil and the second compensation coil are both 2 layers.
4. The rod position detector of claim 1, wherein: the detection coil comprises a first detection coil and a second detection coil, and the first detection coil and the second detection coil are wound on the outer wall of the whole coil framework in parallel;
the first compensation coil comprises a compensation coil A and a compensation coil B, and the compensation coil A and the compensation coil B are wound at one end of the detection coil in parallel; one end of the compensation coil A is connected with one end of the first detection coil, and the compensation coil A is wound at one end of the first detection coil; one end of the compensation coil B is connected with one end of the second detection coil, and the compensation coil B is wound at one end of the second detection coil;
the second compensation coil comprises a compensation coil C and a compensation coil D, and the compensation coil C and the compensation coil D are wound at the other end of the detection coil in parallel; one end of the compensation coil C is connected with the other end of the first detection coil, and the compensation coil C is wound at the other end of the first detection coil; one end of the compensation coil D is connected with the other end of the second detection coil, and the compensation coil D is wound at the other end of the second detection coil.
5. A control rod displacement measuring method according to any one of claims 1 to 4, wherein: the method comprises the following steps:
s1, constructing a displacement neural network model;
s2, constructing a displacement detection circuit; the displacement detection circuit comprises a rod position detector and a reference resistor; one end of the rod position detector is connected with one pole of the power supply, the other end of the rod position detector is connected with one end of the reference resistor, and the other end of the reference resistor is connected with the other pole of the power supply;
s3, acquiring the temperature of the environment where the control rod is located, measuring the actual displacement of the control rod, inputting the temperature of the environment where the control rod is located, the ratio of the voltage of a rod position detector in a displacement detection circuit to the voltage of a reference resistor into a displacement neural network model for training, and after iteration setting times, enabling the error between the output displacement and the actual displacement of the displacement neural network model to be smaller than a set threshold value;
and S4, acquiring the temperature of the environment where the control rod to be detected is located, measuring the ratio of the rod position detector voltage to the reference resistor voltage in the displacement detection circuit when the control rod to be detected moves, and inputting the temperature of the environment where the control rod to be detected is located and the ratio into a displacement neural network model to obtain the displacement of the control rod to be detected.
6. The control rod displacement measurement method as set forth in claim 5, wherein: the step S3 specifically includes:
s31, acquiring the temperature of the environment where the control rod is located, measuring the actual displacement of the control rod at the temperature and measuring the ratio of the rod position detector voltage in the displacement detection circuit corresponding to the actual displacement to the reference resistor voltage;
s32, respectively carrying out normalization processing on the temperature, the actual displacement and the ratio to obtain the normalized temperature, the normalized actual displacement and the normalized ratio;
s33, initializing the displacement neural network model to obtain an initialized displacement neural network model;
s34, inputting the normalized temperature and the normalized ratio as samples into an initialized displacement neural network model, and outputting the predicted displacement of a control rod;
s35, judging whether the error between the predicted displacement of the control rod and the normalized actual displacement is smaller than a set threshold value or not, and if so, ending the operation; if not, go to step S36;
the error between the predicted displacement of the control rod and the normalized actual displacement is as follows:
wherein e is the error between the predicted displacement of the control rod and the normalized actual displacement; o is3kmIs the predicted displacement of the control rod; xmIs the normalized actual displacement of the control rod; k is the number of neuron in output layer in the displacement neural network model; k is the number of output layer neurons in the displacement neural network model; m is the serial number of the sample in the displacement neural network model; m is the number of samples in the displacement neural network model;
s36, judging whether the number of times of executing the step S34 is less than the set number of times UNIf yes, updating the initialized displacement neural network model, and returning to execute the step S34; if not, the process is ended.
7. The control rod displacement measurement method as set forth in claim 6, wherein: in step S31, a ratio of the rod position detector voltage to the reference resistance voltage in the displacement detection circuit corresponding to the actual displacement is measured according to the following steps:
s311, performing full-wave rectification processing on the rod position detector voltage and the reference resistor voltage respectively to obtain a sine half-wave signal of the rod position detector and a sine half-wave signal of the reference resistor;
s312, low-pass filtering is carried out on the sine half-wave signal of the rod position detector and the sine half-wave signal of the reference resistor respectively to obtain a direct-current voltage signal of the rod position detector and a direct-current voltage signal of the reference resistor;
s313, dividing the direct current voltage signal of the rod position detector and the direct current voltage signal of the reference resistor to obtain a ratio signal of the voltage of the rod position detector and the voltage of the reference resistor;
s314, filtering the ratio signal to enable the amplitude of the ratio signal to meet a set threshold value, and taking the value of the ratio signal meeting the set threshold value as the ratio of the rod position detector voltage to the reference resistor voltage.
8. The control rod displacement measurement method as set forth in claim 6, wherein: in step S33, initializing the shift neural network model, specifically including:
s331, carrying out random assignment on the neural parameters of the displacement neural network model to obtain initial values of the neural parameters; the neural parameters comprise weight values, translation factors and scaling factors;
s332, taking each parameter in the neural parameters as a gene, and arranging the genes to form a chromosome;
s333, repeatedly executing the steps S331-S332N times to obtain N chromosomes, and using the N chromosomes as a group;
s334, determining the fitness sequence of the population (f)1,f2,...,fi,...,fN) (ii) a Wherein f isiFitness of the ith chromosome;
s335, acquiring the maximum fitness in the fitness sequence of the group, judging whether the maximum fitness is larger than a set fitness threshold value, and if so, taking the neural parameter corresponding to the chromosome with the maximum fitness as an initial parameter of the displacement neural network model; if not, go to step S336;
s336, judging whether the frequency of executing the step S334 is less than the set frequency VNIf so, selecting and/or hybridizing and/or mutating the population to obtain a processed population, and returning to execute S334; if not, the process is ended.
9. The control rod displacement measurement method as set forth in claim 8, wherein: in step S334, a fitness sequence (f) of the population is determined1,f2,...,fi,...,fN) The method specifically comprises the following steps:
s3341, setting the neural network parameters corresponding to the ith chromosome in the population into a displacement neural network model, inputting the normalized temperature and the normalized ratio as samples into the displacement neural network model, and outputting the predicted displacement of a control rod;
s3342 calculating fitness f of ith chromosomei:
Wherein, O3kmIs the predicted displacement of the control rod; xmIs the actual displacement of the control rod; k is the number of neuron in output layer in the displacement neural network model; k is the number of output layer neurons in the displacement neural network model; m is the serial number of the sample in the displacement neural network model; m is the number of samples in the displacement neural network model;
s3343, analogizing according to the steps S3341-S3342 to obtain a group fitness sequence (f)1,f2,...,fi,...,fN)。
10. The control rod displacement measurement method as set forth in claim 5, wherein: the ratio of the rod position detector voltage to the reference resistance voltage is as follows:
wherein, UBFor alternating voltage of rod position detector, UAIs the alternating voltage of the reference resistor, R1Is the resistance value of the reference resistor, R2Is the reference resistance value of the detection coil, f is the frequency of the AC constant current source, and L is the inductance of the detection coil.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110326229.4A CN113077915A (en) | 2021-03-26 | 2021-03-26 | Rod position detector and control rod displacement measuring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110326229.4A CN113077915A (en) | 2021-03-26 | 2021-03-26 | Rod position detector and control rod displacement measuring method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113077915A true CN113077915A (en) | 2021-07-06 |
Family
ID=76610687
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110326229.4A Withdrawn CN113077915A (en) | 2021-03-26 | 2021-03-26 | Rod position detector and control rod displacement measuring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113077915A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114355047A (en) * | 2022-03-11 | 2022-04-15 | 华龙国际核电技术有限公司 | Rod position detector coil testing method and device and electronic equipment |
CN117558472A (en) * | 2024-01-11 | 2024-02-13 | 深圳大学 | Nuclear reactor cooling system and cooling control method thereof |
-
2021
- 2021-03-26 CN CN202110326229.4A patent/CN113077915A/en not_active Withdrawn
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114355047A (en) * | 2022-03-11 | 2022-04-15 | 华龙国际核电技术有限公司 | Rod position detector coil testing method and device and electronic equipment |
CN114355047B (en) * | 2022-03-11 | 2022-05-24 | 华龙国际核电技术有限公司 | Rod position detector coil testing method and device and electronic equipment |
CN117558472A (en) * | 2024-01-11 | 2024-02-13 | 深圳大学 | Nuclear reactor cooling system and cooling control method thereof |
CN117558472B (en) * | 2024-01-11 | 2024-03-15 | 深圳大学 | Nuclear reactor cooling system and cooling control method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113077915A (en) | Rod position detector and control rod displacement measuring method | |
CN110609200B (en) | Power distribution network earth fault protection method based on fuzzy metric fusion criterion | |
CN108614032B (en) | Concrete internal steel bar nondestructive testing system based on improved neural network and control method | |
CN103886923B (en) | System and method for linearly measuring position of control rod | |
CN111160628A (en) | Air pollutant concentration prediction method based on CNN and double-attention seq2seq | |
CN106448768A (en) | Nuclear power plant control rod position measuring system and method | |
CN102411999B (en) | Rod position detector of nuclear reactor control rod | |
CN109188502A (en) | A kind of beam transport network method for detecting abnormality and device based on self-encoding encoder | |
JPH0367561B2 (en) | ||
Hu et al. | Review on sensors to measure control rod position for nuclear reactor | |
CN107731328A (en) | A kind of Gray code double precision control rod location detection methods | |
CN116304819A (en) | Nuclear reactor operation condition judging method based on LeNet-5 algorithm | |
CN112924813B (en) | Power distribution network short-circuit fault monitoring method and device based on electrical data | |
Lai et al. | A novel scale recognition method for pointer meters adapted to different types and shapes | |
CN108344891A (en) | A kind of high accuracy rectangular current coil of length and width certainty ratio | |
CN116973703A (en) | Acoustic diagnosis method for discharge fault and abnormal operation state of dry type air-core reactor | |
CN103871523A (en) | Nuclear power plant control rod position measurement method | |
CN114062997B (en) | Electric energy meter verification method, system and device | |
CN112611309B (en) | Accurate measurement method for control rod position | |
CN114545113A (en) | Insulator health condition diagnosis method based on memory computing architecture | |
CN114969402A (en) | Vector geographic information acquisition method based on remote sensing image | |
Yueyuan et al. | Combination coding control rod position-indicating system | |
CN113344037A (en) | Cable insulation layer and sheath parameter measuring method and measuring device | |
CN107479110A (en) | A kind of earthquake wave detector test system and method | |
CN114878674A (en) | Transformer winding defect diagnosis method based on comprehensive characteristics of winding stress and magnetic leakage parameter of fusion algorithm |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210706 |