Method, device and equipment for detecting activity of radioactive liquid

Abstract

The invention discloses a radioactive liquid activity detection method, device and equipment, and relates to the field of nuclear energy. The method comprises the following steps: acquiring a to-be-detected water sample, and respectively inputting the to-be-detected water sample into a detector to obtain a detector original pulse signal and a detector fast signal corresponding to the to-be-detected water sample; performing FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new energy spectrum corresponding to the to-be-detected water sample; obtaining a preprocessed theoretical energy spectrum template, and inputting the theoretical energy spectrum template and the new energy spectrum into the peak decomposition network model to obtain a characteristic peak area list of the target nuclide; and obtaining an efficiency curve of the detector, and obtaining an activity detection result of the target nuclear element in the to-be-detected water sample based on the characteristic peak area list of the target nuclide and the efficiency curve. According to the method, the final result is determined by combining the theoretical energy spectrum template, the original pulse signal of the detector and the fast signal of the detector, so that the spectrum unfolding error of the overlapping peak can be reduced, and the recognition accuracy of the mixed system is effectively improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of nuclear energy, and particularly relate to a method, device and equipment for detecting activity of radioactive liquid. BACKGROUND

[0002] During operation of a nuclear power plant, a certain amount of radioactive waste liquid is generated, such as steam generator blowdown water equipment cooling water, auxiliary steam system condensate water, and discharge flow. According to requirements of safety operation and supervision departments of a nuclear power plant, online sampling continuous activity monitoring and analysis need to be performed. At present, a NaI detector with energy spectrum analysis capability is usually used for activity detection. The NaI detector has characteristics such as long-term work, simple structure, and excellent radiation resistance. During detection, γ rays emitted by the liquid to be detected are incident into a NaI crystal, the crystal is excited to emit visible light, the visible light is converted into an electric signal by a rear-end photomultiplier tube, the electric signal is amplified and output, and the rear-end electronics outputs pulse signals, which are collected, analyzed and calculated to give volume specific activity information of the liquid to be detected.

[0003] However, research shows that the existing method is seriously interfered by overlapping peaks, the weak peak recognition rate is reduced, the activity calculation error is large, and the safety disposal efficiency of nuclear waste liquid is seriously restricted. Therefore, there is an urgent need for a new method for detecting activity of radioactive liquid to solve the problems in the prior art. SUMMARY

[0004] The purpose of the present application is to at least solve one of the above technical defects.

[0005] In one aspect, the embodiments of the present application provide a method for detecting activity of radioactive liquid, which comprises: acquiring a water sample to be detected, and inputting the water sample to be detected into a detector to obtain a detector original pulse signal and a detector fast signal corresponding to the water sample to be detected, the detector original pulse signal being a signal obtained by inputting the water sample to be detected into an HPGe detector, and the detector fast signal being a signal obtained by inputting the water sample to be detected into a LaBr3 detector; performing FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new energy spectrum corresponding to the water sample to be detected; acquiring a preprocessed theoretical energy spectrum template, inputting the theoretical energy spectrum template and the new energy spectrum into a peak decomposition network model to obtain a characteristic peak area list of a target nuclide; acquiring an efficiency curve of the detector, and obtaining an activity detection result of the target nuclear element in the water sample to be detected based on the characteristic peak area list of the target nuclide and the efficiency curve.

[0006] Optionally, the FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain the new energy spectrum corresponding to the water sample to be detected comprises: The signal preprocessing is performed on the detector raw pulse signal and the detector fast signal respectively to obtain a timestamp corresponding to the detector raw pulse signal and a timestamp corresponding to the detector fast signal; The time difference between the timestamp corresponding to the detector raw pulse signal and the timestamp corresponding to the detector fast signal is determined based on the FPGA; The energy spectrum corresponding to the water sample to be measured is obtained through energy spectrum construction processing based on the timestamp and the detector raw pulse signal.

[0007] Optionally, the signal preprocessing is performed on the detector raw pulse signal and the detector fast signal respectively to obtain a timestamp corresponding to the detector raw pulse signal and a timestamp corresponding to the detector fast signal, including: The detector raw pulse signal is converted through constant ratio timing to obtain a timestamp corresponding to the detector raw pulse signal; The timestamp corresponding to the detector fast signal is obtained through analog-to-digital conversion processing when the detector fast signal is at a peak value.

[0008] Optionally, the energy spectrum corresponding to the water sample to be measured is obtained through energy spectrum construction processing based on the timestamp and the detector raw pulse signal, including: The timestamp threshold is obtained, and at least one coincidence event is determined according to the relationship between the timestamp threshold and the timestamp; The amplitude value of the detector raw pulse signal corresponding to each coincidence event is obtained, and the amplitude value of the detector raw pulse signal corresponding to each coincidence event is accumulated into the corresponding energy channel to obtain the new energy spectrum corresponding to the water sample to be measured.

[0009] Optionally, the theoretical energy spectrum template and the new energy spectrum are input into the peak decomposition network model to obtain a list of nuclide characteristic peak areas, including: The new energy spectrum is normalized to obtain a normalized energy spectrum, and the weights corresponding to the normalized energy spectrum and the theoretical energy spectrum template are obtained; According to the weights corresponding to the normalized energy spectrum and the theoretical energy spectrum template, the normalized energy spectrum and the theoretical energy spectrum template are spliced to obtain a fusion input feature; The fusion input feature is input into the peak decomposition network model to obtain a list of nuclide characteristic peak areas.

[0010] Optionally, the peak decomposition network model is a Transformer neural network, and the Transformer neural network includes an encoder structure, a parallel attention head, and at least one attention head of an overlapping area, the attention head of the overlapping area is activated in the Transformer preset layer, and the key value projection matrix is shared with the non-overlapping area attention head.

[0011] Optionally, the theoretical energy spectrum template is obtained through the following way: obtain input data, the input data including parameters for describing physical and geometric characteristics of a water sample to be measured and nuclides included in the water sample to be measured and decay characteristics; perform particle transport simulation based on the input data through a Geant4 toolkit to obtain an original energy spectrum; obtain a preset Gaussian response function, and perform convolution processing on the Gaussian response function and the original energy spectrum to obtain a theoretical energy spectrum template.

[0012] Optionally, based on the list of characteristic peak areas of the target nuclide and the efficiency curve, an activity detection result of the target nuclear element in the water sample to be measured is obtained, including: obtain a pre-stored background database, and determine a net peak area according to the background database and the list of characteristic peak areas of the nuclide; perform linear interpolation processing on the efficiency curve and a preset characteristic energy of the target nuclide to obtain an efficiency value of a target energy point; obtain the activity detection result of the target nuclear element in the water sample to be measured according to the net peak area and the efficiency value of the target energy point.

[0013] On the other hand, an embodiment of the present application provides a detection device for activity of radioactive liquid, comprising: a signal obtaining module, configured to obtain a water sample to be measured, and input the water sample to be measured into a detector to obtain a detector original pulse signal corresponding to the water sample to be measured and a detector fast signal, the detector original pulse signal being a signal obtained by inputting the water sample to be measured into an HPGe detector, and the detector fast signal being a signal obtained by inputting the water sample to be measured into a LaBr3 detector; a spectrum constructing module, configured to perform FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new spectrum corresponding to the water sample to be measured; an area list determining module, configured to obtain a preprocessed theoretical energy spectrum template, input the theoretical energy spectrum template and the new spectrum into a peak decomposition network model to obtain a list of characteristic peak areas of a target nuclide; a detection result determining module, configured to obtain an efficiency curve of the detector, and obtain an activity detection result of a target nuclear element in the water sample to be measured based on the list of characteristic peak areas of the target nuclide and the efficiency curve.

[0014] Optionally, when the spectrum constructing module performs FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new spectrum corresponding to the water sample to be measured, the spectrum constructing module is specifically configured to: perform signal preprocessing on the detector original pulse signal and the detector fast signal respectively to obtain a time stamp corresponding to the detector original pulse signal and a time stamp corresponding to the detector fast signal; determine a time difference between the timestamp corresponding to the original pulse signal of the detector and the timestamp corresponding to the fast signal of the detector based on the FPGA; perform energy spectrum construction processing based on the timestamp and the original pulse signal of the detector to obtain a new energy spectrum corresponding to the water sample to be measured.

[0015] Optionally, when the energy spectrum construction module respectively performs signal preprocessing on the original pulse signal of the detector and the fast signal of the detector to obtain the timestamp corresponding to the original pulse signal of the detector and the timestamp corresponding to the fast signal of the detector, it is specifically used for: convert the original pulse signal of the detector by constant ratio timing to obtain the timestamp corresponding to the original pulse signal of the detector; obtain the timestamp corresponding to the fast signal of the detector by analog-to-digital conversion processing when the fast signal of the detector is at a peak value.

[0016] Optionally, when the energy spectrum construction module performs energy spectrum construction processing based on the timestamp and the original pulse signal of the detector to obtain a new energy spectrum corresponding to the water sample to be measured, it is specifically used for: obtain at least one coincidence event according to the relationship between the timestamp threshold and the timestamp; obtain the amplitude value of the original pulse signal of the detector corresponding to each coincidence event, and accumulate the amplitude value of the original pulse signal of the detector corresponding to each coincidence event into the corresponding energy channel to obtain a new energy spectrum corresponding to the water sample to be measured.

[0017] Optionally, when the area list determination module inputs the theoretical energy spectrum template and the new energy spectrum into the peak decomposition network model to obtain the nuclide characteristic peak area list, it is specifically used for: perform normalization processing on the new energy spectrum to obtain a normalized energy spectrum, and obtain weights corresponding to the normalized energy spectrum and the theoretical energy spectrum template, respectively; splice the normalized energy spectrum and the theoretical energy spectrum template according to the weights corresponding to the normalized energy spectrum and the theoretical energy spectrum template, respectively, to obtain a fusion input feature; input the fusion input feature into the peak decomposition network model to obtain the nuclide characteristic peak area list.

[0018] Optionally, the peak decomposition network model is a Transformer neural network, the Transformer neural network includes an encoder structure, a parallel attention head, and at least one attention head of an overlapping area, the attention head of the overlapping area is activated in a Transformer preset layer and shares a key value projection matrix with the non-overlapping area attention head.

[0019] Optionally, the theoretical energy spectrum template is obtained by the following way: Obtain input data, the input data including parameters for describing physical and geometric characteristics of the water sample to be measured and nuclides included in the water sample to be measured and decay characteristics; Perform particle transport simulation by means of the Geant4 toolkit according to the input data to obtain an original energy spectrum; Obtain a preset Gaussian response function, and perform convolution processing on the Gaussian response function and the original energy spectrum to obtain a theoretical energy spectrum template.

[0020] Optionally, when the activity detection result of the target nuclear element in the water sample to be measured is obtained based on the characteristic peak area list of the target nuclide and the efficiency curve, the detection result determination module is specifically used for: Obtain a pre-stored background database, and determine a net peak area according to the background database and the nuclide characteristic peak area list; Perform linear interpolation processing according to the efficiency curve and a preset characteristic energy of the target nuclide to obtain an efficiency value of a target energy point; Obtain the activity detection result of the target nuclear element in the water sample to be measured according to the net peak area and the efficiency value of the target energy point.

[0021] In another aspect, an electronic device is provided, including a processor and a memory: The memory is configured to store machine-readable instructions, which, when executed by the processor, cause the processor to perform any one of the methods for detecting the activity of a radioactive liquid.

[0022] The technical solutions provided in the embodiments of the present application have at least the following beneficial effects: In the embodiments of the present application, the detector original pulse signal and the detector fast signal corresponding to the water sample to be measured can be obtained based on the detector, and then the final result can be determined by combining the theoretical energy spectrum template and the detector original pulse signal and the detector fast signal. At this time, the overlap peak spectrum error can be greatly reduced, and the identification accuracy of the mixed system can be effectively improved. At the same time, the theoretical energy spectrum template and the new energy spectrum are input into the peak decomposition network model to obtain the characteristic peak area list of the target nuclide, and then the activity detection result of the target nuclear element in the water sample to be measured is obtained based on the characteristic peak area list of the target nuclide and the efficiency curve. Since the peak decomposition network model is trained based on the loss function combined with physical constraints, the single sample analysis time can be reduced, the power consumption can be reduced, the online monitoring demand of the nuclear power plant can be met, and the operation and maintenance cost can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating any inventive labor.

[0024] Figure 1 A schematic flowchart illustrating a method for detecting the activity of a radioactive liquid provided in an embodiment of this application; Figure 2 A schematic diagram of a device for detecting the activity of a radioactive liquid provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0025] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting the invention.

[0026] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0028] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0029] Specifically, such as Figure 1 As shown, the method may include: In step S101, the water sample to be measured is obtained, and the water sample to be measured is input into a detector to obtain a detector original pulse signal corresponding to the water sample to be measured and a detector fast signal corresponding to the water sample to be measured. The detector original pulse signal is a signal obtained by inputting the water sample to be measured into an HPGe (High-Purity Germanium) detector. The detector fast signal is a signal obtained by inputting the water sample to be measured into a LaBr3 (Lanthanum Bromide) detector.

[0030] Optionally, when the detector original pulse signal corresponding to the water sample to be measured is obtained, the water sample to be measured can be input into the HPGe detector. At this time, the nuclides in the water sample to be measured decay and emit gamma photons that pass through the container wall (or the water sample matrix). Part of the photons enter the HPGe crystal to cause photoelectric effect (or Compton scattering). At this time, an electron-hole pair is generated, and then the charge is integrated to form a pulse (i.e., the detector original pulse signal). The pulse height of the detector original pulse signal is proportional to the energy of the gamma photon, which represents the identification of the nuclide species. The pulse count rate is proportional to the activity concentration of the nuclide.

[0031] Meanwhile, the water sample to be measured also needs to be input into the LaBr3 detector. At this time, the nuclides in the water sample to be measured decay and emit gamma photons. The LaBr3 crystal in the LaBr3 detector captures the photons, excites Ce³ + emits light, and is converted into a nanosecond-level electrical pulse by a PMT (Photomultiplier Tube), i.e., the detector fast signal corresponding to the water sample to be measured. The time stamp of the detector fast signal marks the absolute time of the gamma event, and has a fast response characteristic, which can capture the cascade gamma time correlation.

[0032] In step S102, FPGA (Field-Programmable Gate Array) time window coincidence processing is performed based on the detector original pulse signal and the detector fast signal to obtain a new energy spectrum corresponding to the water sample to be measured.

[0033] Optionally, after the detector original pulse signal and the detector fast signal are obtained, the detector original pulse signal and the detector fast signal can be subjected to FPGA time window coincidence processing to obtain a new energy spectrum corresponding to the water sample to be measured.

[0034] In an optional embodiment of the present application, the FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new energy spectrum corresponding to the water sample to be measured includes: The detector original pulse signal and the detector fast signal are respectively subjected to signal preprocessing to obtain a time stamp corresponding to the detector original pulse signal and a time stamp corresponding to the detector fast signal. determine a time difference between the time stamp corresponding to the original pulse signal of the detector and the time stamp corresponding to the fast signal of the detector based on the FPGA; perform energy spectrum construction processing based on the time stamp and the original pulse signal of the detector to obtain a new energy spectrum corresponding to the water sample to be measured.

[0035] Optionally, the obtained original pulse of the HPGe detector is a microsecond-level slow pulse, the data format is a digitized waveform, and the specific key parameters contained include pulse height, gamma ray energy, and pulse count rate. The fast signal of the detector is a nanosecond-level fast pulse, and the data format is a time stamp sequence. Based on this, the original pulse signal of the detector and the fast signal of the detector can be respectively preprocessed to obtain the time stamp corresponding to the original pulse signal of the detector and the time stamp corresponding to the fast signal of the detector.

[0036] In an optional embodiment of the present application, the original pulse signal of the detector and the fast signal of the detector are respectively preprocessed to obtain the time stamp corresponding to the original pulse signal of the detector and the time stamp corresponding to the fast signal of the detector, including: convert the original pulse signal of the detector by constant fraction timing to obtain the time stamp corresponding to the original pulse signal of the detector; obtain the time stamp corresponding to the fast signal of the detector by analog-digital conversion processing when the fast signal of the detector is at a peak value.

[0037] Optionally, for the original pulse signal of the detector, since the signal is usually an amplified and shaped electrical pulse signal, the amplitude thereof is proportional to the gamma ray energy, but the pulse rise and fall is slow (microsecond level). At this time, the pulse can be converted into an accurate time mark (time stamp) by constant fraction timing (CFD) technology to obtain the time stamp corresponding to the original pulse signal of the detector, and then the influence of amplitude change on timing accuracy is eliminated. For the fast signal of the detector, since the LaBr3 crystal has a fast decay time, the signal pulse width output thereby is very narrow, and it is more suitable for accurate time marking. Therefore, the analog-digital conversion processing can be performed when the fast signal of the detector is at a peak value (i.e., a "data ready" signal is generated), and the time stamp at this time is recorded to obtain the time stamp corresponding to the fast signal of the detector.

[0038] Further, the FPGA internal logic can compare the time stamp corresponding to the original pulse signal of the detector and the time stamp corresponding to the fast signal of the detector, and determine the time difference (i.e., the time stamp) therebetween. Then, energy spectrum construction processing is performed based on the time stamp and the original pulse signal of the detector to obtain a new energy spectrum corresponding to the water sample to be measured.

[0039] In an optional embodiment of the present application, the energy spectrum construction processing is performed based on the time stamp and the original pulse signal of the detector to obtain a new energy spectrum corresponding to the water sample to be measured, including: acquire a timestamp threshold, and determine at least one coincidence event according to a relationship between the timestamp threshold and the timestamp; acquire an amplitude value of a detector raw pulse signal corresponding to each coincidence event, and accumulate the amplitude value of the detector raw pulse signal corresponding to each coincidence event into a corresponding energy channel to obtain a new energy spectrum corresponding to the water sample to be measured.

[0040] Optionally, a predetermined timestamp threshold can be acquired, and then the determined timestamp is compared with the timestamp threshold to determine at least one coincidence event. For example, if the determined timestamp Δt ≤ timestamp threshold 5ns, it is considered that the two signals (i.e. the detector raw pulse signal and the detector fast signal) come from the same γ event (or cascade γ event), that is, it is identified as a coincidence event.

[0041] Further, a part of the determined coincidence events are random coincidences (i.e. two unrelated events happen at the same time), at this time, the random coincidence background needs to be deducted from the coincidence events, and the random coincidence count rate can be calculated by the following formula:

[0042] wherein, R random is the random coincidence count rate, τ is the window width, R LaBr3 is the single-sided count rate of the LaBr3 detector, R HPGe is the single-sided count rate of the HPGe detector.

[0043] Correspondingly, for each coincidence event, the amplitude of the detector raw pulse signal (i.e. the HPGe detector pulse) can be read (the amplitude represents the energy information), and accumulated into a corresponding energy channel to form a new energy spectrum. At this time, the new energy spectrum formed only contains coincidence events, and non-coincidence events are excluded, so the peak-to-com continuum ratio (the ratio of peak height to background continuum) of the new energy spectrum will be greatly improved, and the overlapping peaks are separated.

[0044] In step S103, a preprocessed theoretical energy spectrum template is acquired, and the theoretical energy spectrum template and the new energy spectrum are input into a peak decomposition network model to obtain a characteristic peak area list of the target nuclide.

[0045] Optionally, a preprocessed theoretical energy spectrum template can be acquired, and then the theoretical energy spectrum template and the new energy spectrum are input into a peak decomposition network model to obtain a characteristic peak area list of the target nuclide.

[0046] In an optional embodiment of the present application, the peak decomposition network model is a Transformer (a sequence conversion architecture based on a self-attention mechanism) neural network, the Transformer neural network includes an encoder structure, parallel attention heads, and at least one overlapping-zone attention head, the overlapping-zone attention head is activated at a preset layer of the Transformer and shares a key-value projection matrix with a non-overlapping-zone attention head.

[0047] In the Transformer neural network, the encoder structure included therein can be a 12-layer encoder structure, the parallel attention heads can be 8 parallel attention heads, the overlapping-zone attention head can be an attention head dedicated to a 662-800 keV overlapping zone, and the overlapping-zone attention head satisfies being activated at the 9th-12th layers of the Transformer, a weight distribution ratio, and sharing a key-value projection matrix with the non-overlapping-zone attention head.

[0048] Optionally, a physical constraint term is introduced into a loss function used when training the Transformer neural network, so that the model is no longer disturbed by the physical environment, thereby ensuring that the obtained list of nuclide characteristic peak areas is more accurate. The loss function is represented by the following formula: L= α· L MSE + β· L Energy + γ· L FWHM wherein, α 、 β and γ are weights, L MSE is a mean square error loss, L Energy is an energy conservation constraint, L FWHM is a peak width constraint.

[0049] In an optional embodiment of the present application, the theoretical energy spectrum template and the new energy spectrum are input into the peak decomposition network model to obtain the list of nuclide characteristic peak areas, including: The new energy spectrum is normalized to obtain a normalized energy spectrum, and weights corresponding to the normalized energy spectrum and the theoretical energy spectrum template are obtained; The normalized energy spectrum and the theoretical energy spectrum template are spliced according to the weights corresponding to the normalized energy spectrum and the theoretical energy spectrum template to obtain fused input features; The fused input features are input into the peak decomposition network model to obtain the list of nuclide characteristic peak areas.

[0050] Optionally, for the obtained new energy spectrum, the following normalization processing can be performed on the new energy spectrum to obtain a normalized energy spectrum.

[0051]

[0052] wherein, is the normalized spectrum, is the new spectrum, wherein is the mean of the background counts in the 300-1500 keV energy interval, is the standard deviation of the 300-1500 keV energy interval.

[0053] Further, an enhancement weight coefficient greater than 1 is applied to the overlapping channel, which specifically includes the weight corresponding to the normalized spectrum and the theoretical spectrum template respectively, and then the normalized spectrum and the theoretical spectrum template are spliced according to the weight corresponding to the weight in proportion to the fusion input, and input into the peak decomposition network model to obtain the list of nuclide characteristic peak areas.

[0054] In an optional embodiment of the present application, the theoretical spectrum template is obtained by the following way: obtaining input data, the input data including parameters for describing the physical and geometric characteristics of the water sample to be tested and the nuclides included in the water sample to be tested and the decay characteristics; performing particle transport simulation by Geant4 toolkit according to the input data to obtain an original spectrum; obtaining a preset Gaussian response function, and performing convolution processing on the Gaussian response function and the original spectrum to obtain a theoretical spectrum template.

[0055] Optionally, the parameters for describing the physical and geometric characteristics of the water sample to be tested can include liquid density, liquid composition (for example, water, organic solvent, acid, etc., and the atomic ratio of each element), the material, thickness and geometric shape of the container, the sample volume and geometric size, and the position of the detector relative to the container. The nuclides included in the water sample to be tested and the decay characteristics can include the name of the nuclide, the energy of the gamma ray emitted by each nuclide and its emission probability, and the cascade relationship of the nuclides emitting cascade gamma rays.

[0056] Further, the particle transport simulation by Geant4 toolkit according to the input data can be performed in the following way to obtain the original spectrum: 1. Construct a geometric model: define the range of the simulation space (for example, a cube with a side length of 1 m), then create a container geometry according to the base parameters, fill the liquid inside the created container geometry, and finally create a geometric model of the HPGe detector, which specifically includes the crystal size and is placed at a specified position.

[0057] 2. Set material properties: select or customize materials from the Geant4 built-in material library, which specifically includes liquid materials, container materials, and detector materials.

[0058] 3. Define the physical process: select the electromagnetic interaction process, photoelectric effect, Compton scattering, electron pair effect, and the transport process of secondary electrons generated by these processes (ionization, bremsstrahlung, etc.) to set the cutoff energy (for example, 1 keV for gamma rays and 10 keV for electrons).

[0059] 4. Set the particle source: define the radiation source, source type, particle type, energy and emission mode, and emission direction according to the nuclide library.

[0060] 5. Set data acquisition: record energy deposition within the detector volume, use "G4PSEnergyDeposit" (energy deposition) or "G4PSTrackLength" (track length) to record the gamma ray energy spectrum within the detector sensitive volume (i.e. the gamma photon energy spectrum entering the detector, but usually we are interested in the deposited energy), and set the number of channels of the energy spectrum (for example, 4096 channels, corresponding to 0-3000 keV).

[0061] 6. Run simulation: simulate a specified number of events (for example, 10^7 decay events), each event may contain multiple gamma photons (such as Co-60 emitting two gamma photons per decay).

[0062] 7. Generate the original energy deposition spectrum: after the simulation is completed, the energy deposition histogram within the detector (i.e. the original energy spectrum) is obtained, which already contains statistical fluctuations and geometric effects of the detector.

[0063] Further, since the energy resolution of the actual detector will cause the peak of the single-energy gamma ray to be broadened into a Gaussian distribution, a pre-set Gaussian response function can be convolved with the original energy spectrum to finally obtain a theoretical energy spectrum template. The theoretical energy spectrum template is a one-dimensional array with a length equal to the number of energy channels, and each element represents the count rate of the corresponding energy channel, which specifically includes the characteristic photopeak, Compton continuum, backscatter peak, and annihilation peak.

[0064] Step S104, obtaining the efficiency curve of the detector, and based on the characteristic peak area list of the target nuclide and the efficiency curve, obtaining the activity detection result of the target nuclear element in the water sample to be measured.

[0065] Wherein, the efficiency curve of the detector is obtained by pre-calibration with a standard source, which describes the detection efficiency of the detector at different energies, usually in the form of a function or a lookup table. Optionally, based on the characteristic peak area list of the target nuclide obtained in the foregoing and the obtained efficiency curve, the activity detection result of the target nuclear element in the water sample to be measured can be obtained.

[0066] In the optional embodiment of the present application, based on the characteristic peak area list of the target nuclide and the efficiency curve, the activity detection result of the target nuclear element in the water sample to be measured is obtained, comprising: obtain a pre-stored background database, and determine a net peak area according to the background database and a list of characteristic peak areas of the nuclide; perform linear interpolation processing on the characteristic energy of the target nuclide according to the efficiency curve and the preset characteristic energy, to obtain an efficiency value of the target energy point; obtain the activity detection result of the target nuclear element in the water sample to be detected according to the net peak area and the efficiency value of the target energy point.

[0067] Optionally, the background count can be matched according to the pre-stored background database (for example, the 662 keV background count is 120 counts), and then the difference between the list of characteristic peak areas of the nuclide and the background count is determined, which is the net peak area.

[0068] Further, the linear interpolation formula is used to perform linear interpolation processing on the characteristic energy of the target nuclide in the efficiency curve, to obtain the efficiency value of the target energy point, and then the activity detection result of the target nuclear element in the water sample to be detected is determined based on the following formula: A = C net / ( ε × T × V) wherein, A is the activity detection result, C net is the net peak area, ε is the interpolation efficiency, T is the collection time, V is the sample volume.

[0069] In the embodiments of the present application, the detector original pulse signal and the detector fast signal corresponding to the water sample to be detected can be obtained based on the detector, and then the final result is determined by combining the theoretical energy spectrum template and the detector original pulse signal and the detector fast signal, which can greatly reduce the overlap peak spectrum error and effectively improve the identification accuracy of the mixed system. At the same time, the theoretical energy spectrum template and the new energy spectrum are input into the peak decomposition network model to obtain the list of characteristic peak areas of the target nuclide, and then the activity detection result of the target nuclear element in the water sample to be detected is obtained based on the list of characteristic peak areas of the target nuclide and the efficiency curve. Since the peak decomposition network model is trained based on the loss function combined with physical constraints, the single sample analysis time can be reduced, the power consumption can be reduced, the online monitoring demand of the nuclear power plant can be met, and the operation and maintenance cost can be reduced.

[0070] The embodiments of the present application provide a radioactive liquid activity detection device, as shown in Figure 2 The device can include a signal acquisition module 201, an energy spectrum construction module 202, an area list determination module 203, and a detection result determination module 204, wherein, The signal acquisition module is configured to acquire the water sample to be tested and input the water sample to be tested into the detectors to obtain detector original pulse signals and detector fast signals corresponding to the water sample to be tested, the detector original pulse signals being signals obtained by inputting the water sample to be tested into the HPGe detector, and the detector fast signals being signals obtained by inputting the water sample to be tested into the LaBr3 detector. The spectrum construction module is configured to perform FPGA time window coincidence processing based on the detector original pulse signals and the detector fast signals to obtain a new spectrum corresponding to the water sample to be tested. The area list determination module is configured to acquire the preprocessed theoretical spectrum template, input the theoretical spectrum template and the new spectrum into the peak decomposition network model, and obtain a characteristic peak area list of the target nuclide. The detection result determination module is configured to acquire the efficiency curve of the detector, and obtain the activity detection result of the target nuclear element in the water sample to be tested based on the characteristic peak area list of the target nuclide and the efficiency curve.

[0071] Optionally, when the signal acquisition module inputs the water sample to be tested into the detectors to obtain the detector original pulse signals and the detector fast signals corresponding to the water sample to be tested, the signal acquisition module is specifically configured to: input the water sample to be tested into the HPGe detector to obtain the detector original pulse signals corresponding to the water sample to be tested; input the water sample to be tested into the LaBr3 detector to obtain the detector fast signals corresponding to the water sample to be tested.

[0072] Optionally, when the spectrum construction module performs FPGA time window coincidence processing based on the detector original pulse signals and the detector fast signals to obtain the new spectrum corresponding to the water sample to be tested, the spectrum construction module is specifically configured to: perform signal preprocessing on the detector original pulse signals and the detector fast signals respectively to obtain time stamps corresponding to the detector original pulse signals and time stamps corresponding to the detector fast signals; determine, based on the FPGA, a time difference between the time stamps corresponding to the detector original pulse signals and the time stamps corresponding to the detector fast signals; perform spectrum construction processing based on the time stamps and the detector original pulse signals to obtain the new spectrum corresponding to the water sample to be tested.

[0073] Optionally, when the spectrum construction module performs signal preprocessing on the detector original pulse signals and the detector fast signals respectively to obtain the time stamps corresponding to the detector original pulse signals and the time stamps corresponding to the detector fast signals, the spectrum construction module is specifically configured to: convert the detector original pulse signals by constant ratio timing to obtain the time stamps corresponding to the detector original pulse signals; perform analog-to-digital conversion processing when the detector fast signals are at a peak value to obtain the time stamps corresponding to the detector fast signals.

[0074] Optionally, when the spectrum construction module performs spectrum construction processing based on the timestamp and the detector raw pulse signal to obtain the new spectrum corresponding to the water sample to be measured, the spectrum construction module is specifically configured to: obtain a timestamp threshold, and determine at least one coincidence event according to a relationship between the timestamp threshold and the timestamp; obtain an amplitude value of the detector raw pulse signal corresponding to each coincidence event, and accumulate the amplitude value of the detector raw pulse signal corresponding to each coincidence event into a corresponding energy channel to obtain the new spectrum corresponding to the water sample to be measured.

[0075] Optionally, when the area list determination module inputs the theoretical spectrum template and the new spectrum into the peak decomposition network model to obtain the nuclide characteristic peak area list, the area list determination module is specifically configured to: perform normalization processing on the new spectrum to obtain a normalized spectrum, and obtain weights corresponding to the normalized spectrum and the theoretical spectrum template respectively; splice the normalized spectrum and the theoretical spectrum template according to the weights corresponding to the normalized spectrum and the theoretical spectrum template respectively to obtain a fusion input feature; input the fusion input feature into the peak decomposition network model to obtain the nuclide characteristic peak area list.

[0076] Optionally, the peak decomposition network model is a Transformer neural network, the Transformer neural network includes an encoder structure, parallel attention heads, and at least one attention head of an overlapping area, the attention head of the overlapping area is activated at a Transformer preset layer, and the attention head of the overlapping area shares a key-value projection matrix with the non-overlapping area attention head.

[0077] Optionally, the theoretical spectrum template is obtained by the following method: obtain input data, the input data including parameters for describing physical and geometric characteristics of the water sample to be measured and nuclides included in the water sample to be measured and decay characteristics; perform particle transport simulation on the input data by using a Geant4 toolkit to obtain an original spectrum; obtain a preset Gaussian response function, and perform convolution processing on the Gaussian response function and the original spectrum to obtain the theoretical spectrum template.

[0078] Optionally, when the detection result determination module obtains the activity detection result of the target nuclear element in the water sample to be measured based on the characteristic peak area list of the target nuclide and the efficiency curve, the detection result determination module is specifically configured to: obtain a pre-stored background database, and determine a net peak area according to the background database and the nuclide characteristic peak area list; perform linear interpolation processing on the efficiency curve and a preset characteristic energy of the target nuclide to obtain a target energy point efficiency value; According to the net peak area and the target energy point efficiency value, the activity detection result of the target nuclear element in the water sample to be detected is obtained.

[0079] The detection device for the activity of the radioactive liquid in the embodiment can perform the detection method for the activity of the radioactive liquid shown in the embodiment, and the implementation principle is similar, which will not be described here.

[0080] The electronic device provided in the embodiment includes a processor and a memory. The memory is configured to store machine-readable instructions. When the instructions are executed by the processor, the processor performs a detection method for the activity of a radioactive liquid.

[0081] The electronic device provided in the embodiment includes a processor and a memory. The memory is configured to store machine-readable instructions. When the instructions are executed by the processor, the processor performs a detection method for the activity of a radioactive liquid. Figure 3 As shown in the figure, Figure 3 The electronic device shown in the figure includes a processor 2001 and a memory 2003. The processor 2001 and the memory 2003 are connected, for example, through a bus 2002. Optionally, the electronic device 2000 can also include a transceiver 2004. It should be noted that in actual application, the transceiver 2004 is not limited to one, and the structure of the electronic device 2000 does not constitute a limitation on the embodiments of the present application.

[0082] The processor 2001 can be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure. The processor 2001 can also be a combination of computing functions, such as one or more microprocessor combinations, DSP and microprocessor combinations, etc.

[0083] The bus 2002 can include a channel for transmitting information between the above-mentioned components. The bus 2002 can be a PCI bus or an EISA bus, etc. The bus 2002 can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 3 In the figure, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0084] The memory 2003 can be a ROM or other types of static storage devices that can store static information and instructions, a RAM or other types of dynamic storage devices that can store information and instructions, an EEPROM, a CD-ROM or other optical disc storage, an optical disc storage (including a compact disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but not limited to this.

[0085] The memory 2003 is configured to store application codes for implementing the scheme of the present application, and the processor 2001 is configured to control the execution of the application codes stored in the memory 2003. The processor 2001 is configured to execute the application codes stored in the memory 2003 to implement the following functions. Figure 2 The detection device for radioactive liquid activity provided by the embodiment shown in the application.

[0086] It should be understood that, although each step in the flowchart of the accompanying drawings is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0087] The above only describes some embodiments of the present application. It should be pointed out that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method of detecting the activity of a radioactive liquid, characterized in that, The method comprises the following steps: acquiring a to-be-tested water sample, and inputting the to-be-tested water sample into a detector to obtain a detector original pulse signal and a detector fast signal corresponding to the to-be-tested water sample, wherein the detector original pulse signal is a signal obtained by inputting the to-be-tested water sample into an HPGe detector, and the detector fast signal is a signal obtained by inputting the to-be-tested water sample into a LaBr3 detector; performing FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new energy spectrum corresponding to the to-be-tested water sample; acquiring a pretreated theoretical energy spectrum template, inputting the theoretical energy spectrum template and the new energy spectrum into a peak decomposition network model to obtain a characteristic peak area list of a target nuclide; acquiring an efficiency curve of the detector, and obtaining an activity detection result of a target nuclear element in the to-be-tested water sample based on the characteristic peak area list of the target nuclide and the efficiency curve.

2. The method of claim 1, wherein, The FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain the new energy spectrum corresponding to the to-be-tested water sample comprises the following steps: respectively performing signal preprocessing on the detector original pulse signal and the detector fast signal to obtain a time stamp corresponding to the detector original pulse signal and a time stamp corresponding to the detector fast signal; determining a time difference between the time stamp corresponding to the detector original pulse signal and the time stamp corresponding to the detector fast signal based on FPGA; performing energy spectrum construction processing based on the time stamp and the detector original pulse signal to obtain the new energy spectrum corresponding to the to-be-tested water sample.

3. The method of claim 2, wherein, The signal preprocessing on the detector original pulse signal and the detector fast signal to obtain the time stamp corresponding to the detector original pulse signal and the time stamp corresponding to the detector fast signal comprises the following steps: converting the detector original pulse signal by constant ratio timing to obtain the time stamp corresponding to the detector original pulse signal; obtaining the time stamp corresponding to the detector fast signal by analog-digital conversion processing when the detector fast signal is at a peak value.

4. The method of claim 2, wherein, The energy spectrum construction processing based on the time stamp and the detector original pulse signal to obtain the new energy spectrum corresponding to the to-be-tested water sample comprises the following steps: acquiring a time stamp threshold, and determining at least one coincidence event according to a relationship between the time stamp threshold and the time stamp; acquiring an amplitude value of the detector original pulse signal corresponding to each coincidence event, and adding the amplitude value of the detector original pulse signal corresponding to each coincidence event to a corresponding energy channel to obtain the new energy spectrum corresponding to the to-be-tested water sample.

5. The method of claim 1, wherein, The inputting of the theoretical energy spectrum template and the new energy spectrum into the peak decomposition network model to obtain the nuclide characteristic peak area list comprises the following steps: performing normalization processing on the new energy spectrum to obtain a normalized energy spectrum, and acquiring respective weights of the normalized energy spectrum and the theoretical energy spectrum template; splicing the normalized energy spectrum and the theoretical energy spectrum template according to the respective weights of the normalized energy spectrum and the theoretical energy spectrum template to obtain fused input features. Input the fusion input feature into the peak decomposition network model to obtain a list of nuclide characteristic peak areas.

6. The method of claim 1, wherein, The peak decomposition network model is a Transformer neural network, and the Transformer neural network includes an encoder structure, parallel attention heads, and at least one attention head of an overlapping area, the attention head of the overlapping area is activated at a Transformer preset layer, and shares a key-value projection matrix with a non-overlapping area attention head.

7. The method of claim 1, wherein, The theoretical energy spectrum template is obtained by the following method: Obtain input data, the input data including parameters for describing physical and geometric characteristics of the water sample to be tested and nuclides included in the water sample to be tested and decay characteristics; According to the input data, a particle transport simulation is performed by a Geant4 toolkit to obtain an original energy spectrum; Obtain a preset Gaussian response function, and perform convolution processing on the Gaussian response function and the original energy spectrum to obtain the theoretical energy spectrum template.

8. The method of claim 8, wherein, Based on the list of characteristic peak areas of the target nuclide and the efficiency curve, the activity detection result of the target nuclear element in the water sample to be tested is obtained, including: Obtain a pre-stored background database, and determine a net peak area according to the background database and the list of nuclide characteristic peak areas; According to the efficiency curve and the preset characteristic energy of the target nuclide, linear interpolation processing is performed to obtain a target energy point efficiency value; According to the net peak area and the target energy point efficiency value, the activity detection result of the target nuclear element in the water sample to be tested is obtained.

9. A device for detecting the activity of a radioactive liquid, characterized in that it comprises: It includes: The signal acquisition module is used for acquiring a water sample to be tested, and inputting the water sample to be tested into a detector to obtain a detector original pulse signal and a detector fast signal corresponding to the water sample to be tested, the detector original pulse signal is a signal obtained by inputting the water sample to be tested into an HPGe detector, and the detector fast signal is a signal obtained by inputting the water sample to be tested into a LaBr3 detector; The energy spectrum construction module is used for performing FPGA time window coincidence processing based on the detector original pulse signal and the detector fast signal to obtain a new energy spectrum corresponding to the water sample to be tested; The area list determination module is used for obtaining a preprocessed theoretical energy spectrum template, inputting the theoretical energy spectrum template and the new energy spectrum into a peak decomposition network model to obtain a list of characteristic peak areas of a target nuclide; The detection result determination module is used for obtaining an efficiency curve of the detector, and based on the list of characteristic peak areas of the target nuclide and the efficiency curve, an activity detection result of a target nuclear element in the water sample to be tested is obtained.

10. An electronic device, comprising: It includes a processor and a memory: The memory is configured to store a computer program, the computer program being executed by the processor, so that the processor executes the method of any one of claims 1-8. The memory is configured to store a computer program, the computer program being executed by the processor, so that the processor executes the method of any one of claims 1-8.

Images