Radiation monitoring method and device

Abstract

The invention provides a radiation monitoring method and device, and relates to the field of radioactive contamination identification. The method comprises the following steps: acquiring gamma radiation data, and converting the gamma radiation data into current pulse signal data; performing amplification processing on the current pulse signal data, and extracting pulse waveform characteristics in the current pulse signal after the amplification processing; grading the current pulse signal into pulse counting data of different amplitude grades based on the pulse waveform characteristics; inputting the pulse counting data into a reference parameter comparison model, and judging whether artificial gamma radiation source data exists in the gamma radiation data or not based on the reference parameter comparison model; and if it is confirmed that the artificial gamma radiation source data exists in the gamma radiation data, carrying out classified early warning on the pollution event in a preset mode. The radiation monitoring device solves the problems that when existing portable or fixed radiation monitoring equipment faces a high-energy low-dose artificial radioactive source with remarkable background fluctuation, the situation of recognition failure or false alarm is prone to occurring, and the overall recognition accuracy is low.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of radioactive contamination identification, and in particular to a radiation monitoring method and device. BACKGROUND

[0002] In recent years, with the continuous expansion of radioactive applications such as nuclear energy utilization, medical isotopes, industrial flaw detection and irradiation processing, the mobility and concealment of artificial radioactive sources in urban and industrial areas have significantly increased, and radiation monitoring technology has gradually shifted from static dose evaluation to dynamic spatiotemporal monitoring and intelligent anomaly identification. The traditional radiation detection method relying on total dose rate measurement and alarm threshold setting cannot meet the fine monitoring needs of high population density, complex environmental disturbance and weak radioactive source identification. The radiation identification method based on energy spectrum structure modeling and statistical feature analysis, combined with pulse signal processing, probability discrimination and multi-dimensional anomaly detection strategy, has become an important development direction of environmental radiation monitoring technology, and has the technical prospect of enhancing anomaly identification sensitivity, reducing false alarm rate and realizing pollution event typing response.

[0003] However, the existing portable or fixed radiation monitoring equipment generally uses a total count rate threshold discrimination model, which only roughly accumulates the total number of all gamma events in a unit time, ignores the energy structure distribution corresponding to different amplitude levels, and lacks the ability to deconstruct the internal structure of the energy spectrum, so it is difficult to fully exploit the structural characteristics of gamma radiation at different energy levels. Especially when facing significant background fluctuations or high-energy low-dose artificial radioactive sources, it is easy to fail to identify or misreport, and the overall identification accuracy is low.

[0004] Therefore, there is an urgent need for a radiation monitoring method and device. SUMMARY

[0005] The present application provides a radiation monitoring method and device, which solves the problem that the existing portable or fixed radiation monitoring equipment is prone to identification failure or false reporting when facing significant background fluctuations or high-energy low-dose artificial radioactive sources, and has low overall identification accuracy.

[0006] In a first aspect of the present application, a radiation monitoring method is provided, the method comprising: acquiring gamma radiation data and converting the gamma radiation data into current pulse signal data; amplifying the current pulse signal data and extracting pulse waveform features in the current pulse signal data after amplification; based on the pulse waveform features, the current pulse signal is classified into pulse count data of different amplitude levels; the pulse count data is input into a reference parameter comparison model, and it is judged whether there is artificial gamma source data in the gamma radiation data based on the reference parameter comparison model; if it is confirmed that there is artificial gamma source data in the gamma radiation data, the pollution event is classified and warned through a preset mode.

[0007] Optionally, the gamma radiation data is acquired and converted into current pulse signal data, specifically including: absorbing the gamma radiation data in the irradiation area by a pre-set plastic scintillator to generate a scintillation light signal related to the gamma energy; transmitting the scintillation light signal to a photomultiplier tube, which amplifies the photoelectrons in the scintillation light signal by multiple stages under stable high voltage conditions based on a multiplier electrode and outputs current pulse signal data.

[0008] Optionally, the current pulse signal data is amplified and the pulse waveform feature in the amplified current pulse signal is extracted: the current pulse signal data is amplified based on a preamplifier circuit; the amplified current pulse signal data is input into a multi-stage comparator circuit, which processes the amplified current pulse signal data by grading according to a set amplitude threshold; the integral values of the graded current pulse signal data at different stages are extracted by an integral window method; the pulse discrimination parameters are calculated based on the integral values at different stages, and the pulse waveform feature is represented based on the pulse discrimination parameters.

[0009] Optionally, the current pulse signal is graded into pulse count data of different amplitude levels based on the pulse waveform feature, specifically including: inputting the pulse waveform feature into a pre-set multi-stage amplitude comparator, and outputting the response result of the current pulse signal under a pre-set condition based on the multi-stage amplitude comparator, the pre-set condition being that a plurality of amplitude thresholds corresponding to a plurality of gamma energy ranges are set, wherein one gamma energy range corresponds to one amplitude threshold; based on the response result of the current pulse signal in each gamma energy range, the number of pulses falling into each amplitude threshold is counted; based on the number of pulses falling into each amplitude threshold, the graded pulse count data corresponding to each amplitude threshold is obtained.

[0010] Optionally, before the pulse count data is input into the reference parameter comparison model and it is determined whether there is artificial gamma radiation source data in the gamma radiation data based on the reference parameter comparison model, the reference parameter comparison model needs to be constructed, specifically including: under the condition of no artificial gamma radiation source interference, acquiring low threshold integral count rate and high threshold integral count rate under different amplitude levels; calculating the ratio between the low threshold integral count rate and the high threshold integral count rate as the initial reference parameter; acquiring historical pulse count data in a continuous sliding time window, and constructing a time-dependent background fluctuation feature vector based on dynamic statistical indicators, the dynamic statistical indicators including count rate mean statistical indicators, variance statistical indicators, slope statistical indicators and skewness statistical indicators; based on the initial reference parameter and the background fluctuation feature vector, a statistical feature space under the condition of no artificial gamma radiation source interference is constructed by a natural background gamma radiation spectrum distribution; the reference parameter comparison model is constructed based on the statistical feature space.

[0011] Optionally, the pulse count data is input into a reference parameter comparison model, and whether the artificial gamma radiation source data exists in the gamma radiation data is determined based on the reference parameter comparison model, specifically including: obtaining the pulse count data under different amplitude levels in the current sliding time window, and constructing a multi-dimensional statistical vector; calculating the Mahalanobis distance of the multi-dimensional statistical vector relative to the initial reference parameter; mapping the Mahalanobis distance into a statistical significance probability value, and determining whether the statistical significance probability value is greater than a preset determination threshold; if the statistical significance probability value is greater than the preset determination threshold, it is confirmed that the artificial gamma radiation source data exists in the gamma radiation data.

[0012] Optionally, if it is confirmed that the artificial gamma radiation source data exists in the gamma radiation data, the pollution event is classified and warned through a preset manner, specifically including: if it is confirmed that the artificial gamma radiation source data exists in the gamma radiation data, the KL divergence between the amplitude normalized distribution in the current sliding window and the historical background distribution is obtained; a joint deviation function is constructed based on the KL divergence, and the judgment result is supplemented and verified through the joint deviation function; if the verification result corresponding to the supplementary verification satisfies a preset deviation condition, the pollution event is classified and warned through a preset manner, and the preset manner includes a location classification manner based on the spatial correlation relationship between the pollution trigger source and the measuring point, a spectral feature classification manner based on the proportion of the sudden increase component in different amplitude levels, and a time domain burst classification manner based on the change trend of the joint deviation function in the continuous sliding window.

[0013] In a second aspect of the present application, a radiation monitoring device is provided, the device comprising an acquisition module and a processing module, wherein, The acquisition module is configured to acquire gamma radiation data and convert the gamma radiation data into current pulse signal data.

[0014] The processing module is configured to amplify the current pulse signal data and extract the pulse waveform features in the current pulse signal after amplification; divide the current pulse signal into pulse count data of different amplitude levels based on the pulse waveform features; input the pulse count data into a reference parameter comparison model, and determine whether artificial gamma radiation source data exists in the gamma radiation data based on the reference parameter comparison model; and if it is confirmed that the artificial gamma radiation source data exists in the gamma radiation data, classify and warn the pollution event through a preset manner.

[0015] In a third aspect of the present application, an electronic device is provided, comprising a processor, a memory, a user interface and a network interface, the memory is configured to store instructions, the user interface and the network interface are configured to communicate with other devices, and the processor is configured to execute the instructions stored in the memory to enable the electronic device to perform the method of any one of the above.

[0016] In a fourth aspect of the present application, a computer-readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to perform the method of any one of the above.

[0017] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Obtain gamma radiation data and convert the gamma radiation data into current pulse signal data; amplify the current pulse signal data and extract the pulse waveform features in the amplified current pulse signal; based on the pulse waveform features, classify the current pulse signal into pulse count data of different amplitude levels; input the pulse count data into a reference parameter comparison model and determine whether there is artificial gamma radiation source data in the gamma radiation data based on the reference parameter comparison model; if it is confirmed that there is artificial gamma radiation source data in the gamma radiation data, classify and warn the pollution event through a preset manner, thereby realizing dynamic deconstruction of the gamma energy spectrum structure under background fluctuation conditions and perturbation response, so that the system can not only accurately identify the artificial gamma radiation source signal with high energy and low dose, but also maintain high identification sensitivity and false alarm suppression capability in a high background radiation interference or multi-point measurement environment, further improving the reliability of pollution event response, the accuracy of classification and the intelligent level of early warning. The technical solution effectively breaks through the traditional low-dimensional identification strategy centered on total dose threshold, has the ability to structurally identify and time series trend track in a complex radiation environment, and significantly improves the risk perception depth and early warning control capability of the radiation monitoring system.

[0018] 2. The pre-set plastic scintillator absorbs the gamma radiation data in the irradiation area to generate a scintillation light signal related to the gamma energy; the scintillation light signal is transmitted to a photomultiplier tube, and the photomultiplier tube amplifies the photoelectrons in the scintillation light signal by multiple stages based on the multiplication electrode under stable high voltage conditions and outputs current pulse signal data, thereby realizing preliminary quantitative conversion of gamma radiation energy information, so that the subsequent electric signal processing unit can perform signal amplification, discrimination and statistics on the basis of preserving the original gamma energy spectrum amplitude structure, ensuring that the system has a complete physical quantity conversion chain and high-fidelity energy spectrum maintenance capability, and laying an accurate and reliable data foundation for pulse waveform feature extraction and subsequent amplitude level classification.

[0019] 3. Under the condition of no artificial gamma source interference, low threshold integral count rate and high threshold integral count rate under different amplitude levels are obtained; the ratio between the low threshold integral count rate and the high threshold integral count rate is calculated as an initial reference parameter; historical pulse count data in a continuous sliding time window is obtained, and a time-dependent background fluctuation feature vector is constructed based on dynamic statistical indicators, including count rate mean statistical indicators, variance statistical indicators, slope statistical indicators, and skewness statistical indicators; based on the initial reference parameter and the background fluctuation feature vector, a statistical feature space under the condition of no artificial gamma source interference is constructed through the natural background gamma radiation spectrum distribution; a reference parameter comparison model is constructed based on the statistical feature space, so as to realize dynamic modeling and quantitative description of background fluctuation states under different measuring points and different time stages, improve the adaptability of the model to time-varying background and spatial heterogeneous background conditions, enable the subsequent identification algorithm to perform anomaly identification based on the background adaptive threshold domain, and significantly enhance the response resolution and statistical robustness of the system to weak artificial gamma source disturbance in a high background and multi-interference environment. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a flowchart of a radiation monitoring method provided by an embodiment of the present application; Figure 2 is a module schematic diagram of a radiation monitoring device provided by an embodiment of the present application; Figure 3 is a structural schematic diagram of an electronic device provided by an embodiment of the present application.

[0021] Legend: 21, acquisition module; 22, processing module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0022] In order for those skilled in the art to better understand the technical solutions in the specification, the technical solutions in the specification will be clearly and completely described below in conjunction with the drawings in the embodiments of the specification. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments.

[0023] The terms used in the following embodiments of the present application are only for the purpose of describing the specific embodiments of the present application, and are not intended to be limiting to the present application. As used in the specification of the present application, the singular expression "one", "a", "said", "the above", "the", and "this" are intended to also include the plural expression, unless there is clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application means and includes any or all possible combinations of one or more listed items.

[0024] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0025] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0026] Please refer to Figure 1 The diagram shows a schematic flowchart of a radiation monitoring method provided in an embodiment of this application. The flowchart mainly includes the following steps: S101 to S105.

[0027] Step S101: Acquire gamma radiation data and convert the gamma radiation data into current pulse signal data.

[0028] Specifically, based on the deployed mobile gamma radiation monitoring equipment, multiple high-sensitivity plastic scintillators positioned on both sides of the monitoring channel receive gamma radiation particles within the irradiation area. Under the influence of gamma photons of different energies, the plastic scintillators generate nonlinear intensity scintillating light related to their energy. Using an optical coupling structure, this scintillating light is guided along a paraxial path into photomultiplier tubes positioned at appropriate locations. The photomultiplier tubes release initial photoelectrons from the incident photocathode, and then, under a stable high-voltage distribution, a cascaded secondary emission process is completed by multiple stages of dynodes, resulting in an exponential increase in the number of photoelectrons. The final anode output of the booster tube is a pulse current signal formed by the total charge generated by the preceding multiplication chain. This signal reflects the energy contribution of gamma particles in the scintillator based on gamma energy deposition. To ensure that the current pulse signal has time and energy discriminability, an active voltage divider circuit is set in the photomultiplication channel to stabilize the voltage distribution of each stage. Damping resistors and low-pass filters are introduced to reduce interference caused by high-frequency noise and photoelectric drift, ensuring the linearity of the pulse output response and amplitude stability. This results in current pulse signal data with physically measurable properties, providing a stable input basis for subsequent amplification and waveform discrimination.

[0029] In one possible implementation, step S101 further includes: absorbing energy from γ radiation data within the irradiation area using a pre-set plastic scintillator to generate a scintillation light signal related to γ ​​energy; transmitting the scintillation light signal to a photomultiplier tube, which, under stable high voltage conditions, amplifies the photoelectrons in the scintillation light signal through multiple stages based on the multiplication electrode and outputs current pulse signal data.

[0030] Specifically, a plastic scintillator with a volume of not less than 3L is used to construct the gamma-ray detection unit, the plastic scintillator has a high Compton scattering response efficiency for gamma radiation in the energy range of 30 keV to 3 MeV, can realize energy deposition and trigger the excited state of organic molecules after the gamma photon enters, and thus emit a scintillation light signal with a wavelength in the visible region; in order to enhance the collection efficiency of the scintillation light, the scintillator is coated with a high reflectivity diffuse reflection layer outside and is optically sealed, and the scintillation light is collimated and coupled to the input end of the photomultiplier tube arranged in alignment with the optical axis thereof through the light guide window; the photomultiplier tube is composed of a quartz or glass window body with high transmittance, and is internally provided with a photokathode and a 10-stage or more dynode structure, a stable bias voltage is applied by a high-voltage power supply, and an exponential gradient potential difference is constructed between the stages based on an active voltage division circuit; after the incident scintillation light irradiates the photokathode, initial photoelectrons are released, the photoelectrons experience accelerated impact and trigger a secondary electron multiplication process between the stages, and finally a considerable amount of electron beam is collected at the final-stage anode; the electron beam is released as a transient charge signal in the sub-nanosecond time scale, is sent to the output end through a coupling capacitor to form a current pulse; the amplitude of the current pulse is positively correlated with the initial gamma photon deposition energy, and the pulse front edge is strictly synchronized with the scintillation light generation process; in order to ensure that the output signal has high stability and linearity, a low-pass filter is further arranged in the voltage division circuit of the photomultiplier tube to suppress the high-frequency power supply ripple, and a damping resistor is connected to the final-stage dynode to suppress the overshoot oscillation, thereby forming a current pulse signal data with stable dynamic response, high time resolution and quantifiable voltage amplitude.

[0031] In step S102, the current pulse signal data is amplified and processed, and the pulse waveform features in the current pulse signal after the amplification processing are extracted.

[0032] Specifically, the current pulse signal output in step S101 is input to a preamplification circuit for signal amplitude enhancement, the preamplification circuit including a wideband low-noise amplifier and an impedance matching module, to improve the signal-to-noise ratio of the pulse signal and suppress weak interference, and to ensure that the pulse response amplitudes corresponding to different energy level gamma events remain distinguishable; then the amplified analog electrical signal is transmitted to a high-speed sampling channel, and the waveform of the current pulse is captured under high sampling rate conditions to obtain a complete digital sequence of the pulse voltage varying with time; on the basis of the digital sequence, multiple waveform parameters of the pulse, such as rise time, fall time, peak amplitude, half-peak width, and front-to-back slope ratio, are extracted as structural feature vectors for subsequent waveform recognition; in the feature extraction process, the pulse starting position is automatically identified and the background baseline is corrected in real time, a fixed time window is used to intercept and amplitude normalization is used to process to ensure consistency between the parameter dimensions, and when abnormal waveforms exist, the mismatched forms are automatically recorded for statistical rejection and false alarm correction; the acquisition of waveform features provides basic data support for subsequent multi-threshold classification, PSD discrimination, and energy spectrum classification, and constitutes the core expression of determining the physical characteristics of gamma events.

[0033] In a possible implementation, step S102 further includes: amplifying the current pulse signal data based on the preamplification circuit; inputting the amplified current pulse signal data to a multi-stage comparator circuit, and performing grading processing on the amplified current pulse signal data according to a set amplitude threshold; extracting integral values of the graded current pulse signal data at different stages by an integral window method; calculating pulse discrimination parameters based on the integral values at different stages, and representing pulse waveform features based on the pulse discrimination parameters.

[0034] Specifically, the current pulse signal is introduced into a multi-stage high-speed analog comparator network after low-distortion enhancement by the preamplification circuit, the comparator performs real-time amplitude division on the input pulse signal according to a plurality of preset amplitude thresholds, each comparator channel corresponds to an energy level window, and can output a logic signal indicating whether the current pulse exceeds the set threshold of the channel; on this basis, integral window analysis is performed on the amplified pulse signal, the pulse time domain is divided into a rising edge stage, a peak neighborhood stage, and a falling tail stage, and segmented integral operations are respectively performed to extract the cumulative values of the energy at each stage, wherein the start and end boundaries of the integral region are locked by a high-speed trigger mechanism to ensure that the integral windows of different pulses remain consistent on the relative time axis; discrimination parameters are constructed using the integral values at each stage, the discrimination parameters are selected as the ratio of the integral value in the falling stage to the integral value in the whole stage , and are represented as

[0035] wherein represents the integral value in the falling stage, The PSD parameter represents the total integral value of the entire pulse. As a structural waveform indicator, it can reflect the differences in pulse morphology corresponding to different gamma events or radiation source types. For example, natural background gamma and artificial source gamma exhibit different energy deposition paths and fast / slow components in the scintillation response, which leads to the separability of the PSD distribution. By statistically modeling a large number of historical pulse PSD values, a reference distribution can be formed for subsequent type determination. The PSD extraction mechanism, combined with a multi-threshold structure, enables the identification of different radiation components from the pulse morphology level.

[0036] Step S103: Based on the pulse waveform characteristics, the current pulse signal is divided into pulse count data of different amplitude levels.

[0037] Specifically, based on the pulse waveform features extracted in step S102, especially the amplitude information related to the pulse peak amplitude and integral energy, the amplified current pulse signal is classified and statistically analyzed within each sliding time window according to the set amplitude classification standard. Several energy threshold intervals are preset, with each amplitude level corresponding to a pulse energy segment. When the maximum amplitude or total integral value of a single pulse signal falls into a specific interval, the pulse is recorded in the counting slot of the corresponding amplitude level. While the sliding time window is running continuously, the number of pulses corresponding to each amplitude level is accumulated in real time, ultimately forming a statistical vector containing multiple graded counts. To improve statistical accuracy, a partial overlapping area is introduced when the window boundaries alternate to avoid the impact of cross-boundary pulse truncation on the counting and classification results. At the same time, an edge fitting mechanism is introduced to handle the boundary judgment of abnormal amplitude pulses. This grading operation discretizes continuous amplitude information into a countable classification structure, forming the input basis for subsequent statistical discrimination, ensuring that various statistical deviation measures are calculated and judged on a unified data structure.

[0038] In one possible implementation, step S103 further includes: inputting pulse waveform features into a preset multi-level amplitude comparator, and based on the multi-level amplitude comparator, outputting the response result corresponding to the current pulse signal under preset conditions, wherein the preset conditions are to set multiple amplitude thresholds corresponding to multiple γ energy ranges, and one γ energy range corresponds to one amplitude threshold; based on the response result corresponding to the current pulse signal in each γ energy range, counting the number of pulses falling into each amplitude threshold; and based on the number of pulses falling into each amplitude threshold, obtaining the corresponding pulse count data for each grade.

[0039] Specifically, multiple amplitude comparator links are configured at the end of the amplified signal channel, and each comparator is configured with a set of upper and lower limit amplitude thresholds. These correspond to the preset γ energy grading ranges. Used to determine whether a single current pulse signal belongs to the first... One γ energy range; within the current sliding time window For each amplified pulse signal, if its peak amplitude satisfies:

[0040] then the pulse is attributed to the energy response event of the th level, and is added to the corresponding count register once.

[0041] The number of pulses falling into the interval is counted to form the multi-level pulse count vector :

[0042] wherein represents the number of pulses accumulated in the th amplitude threshold interval within the sliding time window .

[0043] To further optimize the classification accuracy, the amplitude comparator has high-gain contrast window recognition capability, which can accurately divide the high-density pulse stream with a resolution of no less than 1 mV amplitude, and suppresses the boundary crossing misjudgment through a dynamic mismatch suppression mechanism.

[0044] In addition, to reduce the impact of classification drift caused by transient noise disturbance, a short-time sliding average mechanism is introduced for each classification pulse set in the statistical process, and different count weight adjustment functions are introduced for the low-energy region and the high-energy region, forming a parameter-weighted sub-level count output :

[0045] wherein is the energy response weight factor corresponding to the th pulse count, which is dynamically set according to the gamma spectrum response characteristics, so that the statistical counts of different energy regions have controllable contribution ratios in the subsequent discrimination model, enhancing the sensitivity to the interference spectrum structure of artificial gamma radiation sources.

[0046] If the time processing thread determines that the periodic task object does not satisfy the current scheduling condition, i.e., the current timestamp is not on the scheduling period boundary, no change operation is performed on the task object, its state field remains unchanged, and it is retained in the thread local task state table or buffer, and the judgment is triggered again when the next time event arrives. This mechanism ensures that the scheduling behavior of the task object strictly depends on the period configuration, and is not triggered by non-periodic time events, enhancing the stability and predictability of the scheduling mechanism.

[0047] Step S104, input the pulse count data into the reference parameter comparison model, and judge whether there is artificial gamma source data in the gamma radiation data based on the reference parameter comparison model.

[0048] Specifically, in the process of continuous operation, the pulse count data of each amplitude level in the current time period is obtained by a sliding time window mechanism to form a statistical vector, which is consistent in structure with the background feature data established in the reference parameter comparison model, ensuring the structural isomorphism and model compatibility of data input; without involving complex discrimination calculation mechanism, the ratio relationship, proportion distribution and sudden increase index related to the pulse count of each amplitude level are used as the input variables of the model preliminary judgment, for example, based on the count ratio of low energy area and high energy area, the total count change rate and the local growth amplitude of a certain energy area, multiple simple judgment conditions are constructed; the above judgment conditions take the historical background mean, statistical fluctuation interval and experience judgment threshold recorded in the reference parameters as the comparison standard to form the preliminary interval discrimination logic, when any index deviates from the reference value beyond the set range, the early warning sign of the existence of artificial source is triggered; in addition, in order to ensure that the model is adaptable to different background conditions of measuring points, an independent reference parameter set is set for each measuring point in the initialization stage, and the reference parameter is updated regularly based on the time migration strategy, ensuring that the model discrimination basis has timeliness and locality; this preliminary judgment step does not depend on high-dimensional features or complex discrimination functions, but forms a rapid identification of gamma radiation data mutation through parameter comparison, providing traceable input basis for subsequent multi-dimensional deviation verification.

[0049] In one possible implementation, step S104 further includes: under the environmental condition without artificial gamma source interference, obtaining low threshold integral count rate and high threshold integral count rate under different amplitude levels; calculating the ratio between the low threshold integral count rate and the high threshold integral count rate as the initial reference parameter; obtaining the historical pulse count data in the continuous sliding time window, and constructing a time-related background fluctuation feature vector based on dynamic statistical indicators, including count rate mean statistical indicators, variance statistical indicators, slope statistical indicators and skewness statistical indicators; based on the initial reference parameter and the background fluctuation feature vector, a statistical feature space under the environmental condition without artificial gamma source interference is constructed through the natural background gamma radiation spectrum distribution; the reference parameter comparison model is constructed based on the statistical feature space.

[0050] Specifically, under the natural background gamma radiation condition, a continuous pulse count data sequence is collected through multiple sliding time windows, wherein the natural background gamma radiation condition is specifically: the persistent natural radiation field formed by the gamma photons released in the decay process of crust radioactive elements (such as uranium series, thorium series, potassium-40, etc.) in a spatial environment without any artificial gamma source interference, which has the characteristics of stable spatial distribution, continuous energy spectrum structure and slow time fluctuation; a low energy threshold value with a high energy threshold corresponding to the respective count rate with in the window period The average integrated count rate is calculated as follows:

[0051] wherein denotes the time-averaged value of the low-threshold integrated count rate, denotes the time-averaged value of the high-threshold integrated count rate, denotes the starting point in time of the current sliding time window, denotes the duration length of the sliding time window, based on which the initial reference parameter is defined as and the ratio:

[0052] Subsequently, the respective amplitude-level pulse count values corresponding to each window are extracted in a plurality of consecutive time windows for which the statistical dimensions are constructed including:

[0053] wherein denotes the average pulse count rate of the i th amplitude-level pulse count sequence over n consecutive sliding time windows, denotes the time center point corresponding to the j th sliding time window, used to construct the time series trend analysis variable, denotes the total number of sliding time windows, is the standard deviation of the i th amplitude-level pulse count sequence over time windows, denotes the linear change trend (i.e. the regression slope) of the i th amplitude-level pulse count sequence on the time axis, denotes the arithmetic mean of all , and denotes the skewness of the i th amplitude-level pulse count sequence. The above plurality of statistical quantities are constructed as the underlying volatility feature vector

[0054] wherein is the average pulse count rate of the amplitude-level pulse count sequence of the lowest energy region over n consecutive sliding time windows, For the lowest energy level pulse counting sequence in The standard deviation within a time window The linear variation trend of the pulse counting sequence in the lowest energy region on the time axis. The skewness of the amplitude level pulse counting sequence in the lowest energy region. For the highest energy level pulse counting sequence in n Average pulse count rate within a continuous sliding time window For the highest energy level pulse counting sequence in The standard deviation within a time window The linear variation trend of the highest energy level pulse counting sequence on the time axis. The skewness of the amplitude level pulse counting sequence in the lowest energy region. The skewness of the amplitude-level pulse count sequence in the highest energy region is used. When constructing the statistical feature space, the amplitude-normalized distribution of the natural background gamma radiation spectrum is introduced. Combined with the above reference parameters With the background fluctuation eigenvector These features together constitute a typical set of unmanned source points in the statistical feature space. Furthermore, maximum likelihood estimation is used to model the covariance of these features, denoted as the covariance matrix. The reference parameter comparison model is constructed using a high-dimensional Mahalanobis distance function. :

[0055] in For the subsequent input of observed feature vectors, This model is used to determine whether an object deviates from the constructed background spatial boundary, thereby establishing a basis for the existence discrimination of artificial gamma-ray sources. The comparison model possesses statistical adaptability and time-updateability, adapting to changes in natural background noise under different monitoring environments, forming a dynamically responding background criterion system.

[0056] In one possible implementation, step S104 further includes: acquiring pulse count data at different amplitude levels within the current sliding time window and constructing a multidimensional statistical vector; calculating the Mahalanobis distance of the multidimensional statistical vector relative to the initial reference parameter; mapping the Mahalanobis distance to a statistical significance probability value and determining whether the statistical significance probability value is greater than a preset judgment threshold; if the statistical significance probability value is greater than the preset judgment threshold, then confirming the presence of artificial gamma radiation source data in the gamma radiation data.

[0057] Specifically, in each sliding time window Within this context, the pulse counts at each amplitude level at the current moment are statistically analyzed to form an observation statistical vector. :

[0058] The observation vector needs to be amplitude normalized and mean-variance standardized before statistical analysis to eliminate the interference of the absolute order of magnitude difference between different energy levels. The standardized vector is denoted as :

[0059] The multi-dimensional statistical mean vector and covariance matrix established in the initial background stage are used to calculate the Mahalanobis distance of the standardized observation vector :

[0060] This distance measures the deviation of the current gamma count structure from the background statistical structure. Under the assumption of multi-dimensional normal distribution, the square value of the Mahalanobis distance follows a chi-square distribution with degrees of freedom , so the significance probability value can be obtained by mapping as follows:

[0061] where is the cumulative distribution function of the chi-square distribution with degrees of freedom , and represents the confidence level of the current observation vector belonging to the background distribution space .

[0062] If it satisfies:

[0063] where is the set confidence significance threshold (such as 0.05 or 0.01), it is determined that the current observation vector is statistically significantly deviated from the background structure, that is, it is determined that there may be artificial gamma source interference.

[0064] If it satisfies:

[0065] It indicates that the current gamma radiation data is not statistically significantly deviated from the statistical feature space under the condition of no artificial gamma source.

[0066] This step maps the observation count behavior to the abnormality judgment in the probability sense based on statistical inference theory, and constructs a direction-sensitive multi-dimensional deviation measure through Mahalanobis distance, which can effectively identify the complex pollution situation of abnormal overall energy spectrum structure without significant changes in single energy level. Compared with the traditional threshold comparison, this method has stronger collaborative discrimination ability and false alarm tolerance, and improves the response sensitivity to weak artificial source signals.

[0067] Step S105, if it is confirmed that there is artificial gamma source data in the gamma radiation data, a pollution event is classified and warned through a preset mode.

[0068] Specifically, after completing the identification of artificial gamma sources based on the statistical model, the pulse count data, amplitude normalized distribution, energy level distribution characteristics, and mutation trend indicators in the current sliding time window are taken as inputs to preliminarily classify the level and type of the pollution event. A plurality of pollution warning levels are preset, such as prompt level, warning level, alarm level, etc., corresponding to different response mechanisms. The judgment basis includes but is not limited to whether the high-energy pulse count ratio exceeds the set threshold, the duration of continuous windows with high deviation probability, whether a high-energy peak shift appears in the energy spectrum morphology, and whether the statistical slope appears a positive surge, etc.

[0069] The classification result corresponds to different preset response paths, for example, the prompt level automatically records logs and generates a background reminder, the warning level starts a live voice or light prompt, and the alarm level triggers a regional linkage response strategy.

[0070] In one possible implementation, step S105 further includes: if it is confirmed that there is artificial gamma source data in the gamma radiation data, obtaining the KL divergence between the amplitude normalized distribution in the current sliding window and the historical background distribution; constructing a joint deviation function based on the KL divergence, and supplementally verifying the judgment result through the joint deviation function; if the verification result corresponding to the supplementally verification meets a preset deviation condition, classifying and warning the pollution event through a preset mode, the preset mode including a location classification mode based on the spatial correlation between the pollution trigger source and the measurement point, an energy spectrum characteristic classification mode based on the proportion of the sudden increase component in different amplitude levels, and a time domain burst classification mode based on the change trend of the joint deviation function in the continuous sliding window.

[0071] Specifically, after confirming in step S104 that there is artificial gamma source interference in the current sliding window, the normalized energy spectrum distribution vector in the window is extracted , and the amplitude normalized distribution of the natural background gamma radiation spectrum is called, and the Kullback-Leibler divergence of the two is calculated as the structural deviation :

[0072] wherein, represents, and the Mahalanobis distance calculated in step S104 is called to construct a joint deviation function :

[0073] wherein and are adjustable weight parameters in the joint deviation function, satisfying , for weighting the balance between spectral deviation and statistical structure deviation. If the joint deviation exceeds a preset validation threshold , it is determined that the current pollution identification result has statistical significance and structural consistency, and the classification warning of the pollution event is started. The classification warning specifically includes but is not limited to the following three execution modes: position classification mode: based on the spatial coordinates (x, y) of the current measuring point in the device array, combined with the warning state and deviation intensity of the adjacent measuring points, a pollution source-measuring point spatial correlation map is constructed to judge whether the pollution presents a central diffusion trend or a single-point burst response, which is used to preliminarily locate the position type of the pollution source, such as a fixed point source, a flow source or a surface leakage; in the current normalized spectrum , the high-amplitude level corresponding to the pulse proportion is extracted, indicating the energy threshold index for dividing the high-amplitude level region, and the burst component proportion is calculated:

[0074] If exceeds the spectral burst threshold, it is determined that the pollution event has a high-energy enrichment feature, which can be classified as a high-energy spectrum event, such as industrial gamma source leakage or medical isotope near-field release; based on the joint deviation sequence in the continuous sliding window, the first-order difference is calculated, and it is judged whether there is a mutation:

[0075] wherein is the time-domain burst deviation judgment threshold. If the condition is met, the event is classified as a time-burst-type pollution, which is suitable for temporary radioactive source insertion, moving source approach, etc.

[0076] If the joint deviation does not exceed the preset validation threshold , indicates that the observed gamma radiation data in the current sliding time window, although it has been preliminarily determined to exist artificial gamma source interference in the previous stage, after being re-measured by the joint deviation function composed of the Mahalanobis distance and the KL divergence, the statistical deviation degree and the structural deviation degree do not reach the set lower limit of the reliable judgment, so the judgment result does not have sufficient statistical support and spectral deviation significance. In this case, the following treatment strategies, including but not limited to, will be triggered: postponing the triggering of the contamination event classification warning to avoid false alarms caused by one-time transient errors, short-term statistical disturbances or local probe abnormalities; the intermediate criterion of the observed vector , the deviation component , , the joint deviation amount in the current window is recorded as a cache record and written into the to-be-confirmed event log, marked as a "potential disturbance event", and the evolution behavior in the subsequent windows is continuously tracked; if the joint deviation amount presents a monotonic increasing trend or crosses the verification threshold , the event upgrading mechanism is triggered, the contamination event is re-confirmed, and the previous window is traced back as the warning starting point.

[0077] Please refer to Figure 2 , which shows a module schematic diagram of a radiation monitoring device provided by an embodiment of the present application. The device comprises an acquisition module 21 and a processing module 22, wherein, The acquisition module 21 is configured to acquire gamma radiation data and convert the gamma radiation data into current pulse signal data.

[0078] The processing module 22 is configured to amplify the current pulse signal data and extract the pulse waveform features in the current pulse signal after amplification; based on the pulse waveform features, the current pulse signal is classified into pulse count data of different amplitude levels; the pulse count data is input into a reference parameter comparison model, and whether there is artificial gamma source data in the gamma radiation data is judged based on the reference parameter comparison model; if it is confirmed that there is artificial gamma source data in the gamma radiation data, the contamination event is classified and warned by a preset mode.

[0079] In a possible implementation, the acquisition module 21 is configured to acquire gamma radiation data and convert the gamma radiation data into current pulse signal data, specifically including: absorbing the gamma radiation data in the irradiation area through a pre-set plastic scintillator to generate a scintillation light signal related to the gamma energy; transmitting the scintillation light signal to a photomultiplier tube, and under the condition of stable high voltage, the photomultiplier tube amplifies the photoelectrons in the scintillation light signal by multiple stages based on the multiplication electrode and outputs the current pulse signal data.

[0080] In a possible implementation, the processing module 22 is configured to amplify the current pulse signal data and extract the pulse waveform feature in the amplified current pulse signal: amplifying the current pulse signal data based on a preamplifier circuit; inputting the amplified current pulse signal data into a multi-stage comparator circuit, and performing grading processing on the amplified current pulse signal data according to a set amplitude threshold; extracting integral values of the graded current pulse signal data at different stages by using an integral window method; calculating a pulse discrimination parameter based on the integral values at different stages, and representing the pulse waveform feature based on the pulse discrimination parameter.

[0081] In a possible implementation, the processing module 22 is configured to grade the current pulse signal into pulse count data of different amplitude levels based on the pulse waveform feature, specifically including: inputting the pulse waveform feature into a preset multi-stage amplitude comparator, and outputting a response result corresponding to the current pulse signal under a preset condition based on the multi-stage amplitude comparator, the preset condition being that a plurality of amplitude thresholds corresponding to a plurality of γ energy ranges are set, and one γ energy range corresponds to one amplitude threshold; counting the number of pulses falling into each amplitude threshold based on the response result corresponding to the current pulse signal in each γ energy range; and obtaining the graded pulse count data corresponding to the number of pulses falling into each amplitude threshold.

[0082] In a possible implementation, before the processing module 22 inputs the pulse count data into the reference parameter comparison model and judges whether the artificial γ radiation source data exists in the γ radiation data based on the reference parameter comparison model, the reference parameter comparison model needs to be constructed, specifically including: obtaining a low-threshold integral count rate and a high-threshold integral count rate under different amplitude levels in an environment condition without artificial γ radiation source interference; calculating a ratio between the low-threshold integral count rate and the high-threshold integral count rate as an initial reference parameter; obtaining historical pulse count data in a continuous sliding time window, and constructing a time-dependent background fluctuation feature vector based on dynamic statistical indicators, the dynamic statistical indicators including a count rate mean statistical indicator, a variance statistical indicator, a skewness statistical indicator, and a kurtosis statistical indicator; constructing a statistical feature space in the environment condition without artificial γ radiation source interference based on the initial reference parameter and the background fluctuation feature vector by using a natural background γ radiation spectrum distribution; and constructing the reference parameter comparison model based on the statistical feature space.

[0083] In a possible implementation, the processing module 22 is configured to input the pulse count data into a reference parameter comparison model, and determine whether the artificial gamma radiation source data exists in the gamma radiation data based on the reference parameter comparison model, specifically including: obtaining the pulse count data at different amplitude levels in the current sliding time window, and constructing a multi-dimensional statistical vector; calculating the Mahalanobis distance of the multi-dimensional statistical vector relative to the initial reference parameter; mapping the Mahalanobis distance to a statistical significance probability value, and determining whether the statistical significance probability value is greater than a preset determination threshold; if the statistical significance probability value is greater than the preset determination threshold, it is determined that the artificial gamma radiation source data exists in the gamma radiation data.

[0084] In a possible implementation, the processing module 22 is configured to, if it is determined that the artificial gamma radiation source data exists in the gamma radiation data, perform classified early warning on the pollution event in a preset manner, specifically including: if it is determined that the artificial gamma radiation source data exists in the gamma radiation data, obtaining the KL divergence between the amplitude normalized distribution in the current sliding window and the historical background distribution; constructing a joint deviation function based on the KL divergence, and supplementally verifying the determination result by using the joint deviation function; if the verification result corresponding to the supplementally verification satisfies a preset deviation condition, performing classified early warning on the pollution event in a preset manner, and the preset manner includes a location classification manner based on the spatial correlation between the pollution trigger source and the measuring point, a spectral feature classification manner based on the proportion of the sudden increase component in different amplitude levels, and a time domain burst classification manner based on the change trend of the joint deviation function in the continuous sliding window.

[0085] It should be noted that the apparatus provided in the above embodiments is only used as an example for dividing the above functional modules, and in actual applications, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above described functions. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be described here.

[0086] The present application also provides an electronic device. Referring to Figure 3 , Figure 3 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. The electronic device can include at least one processor 301, at least one communication bus 302, a user interface 303, at least one network interface 304, and a memory 305.

[0087] The communication bus 302 is configured to realize the connection and communication between the components.

[0088] The user interface 303 can include a display screen (Display) and a camera (Camera), and the optional user interface 303 can further include a standard wired interface and a wireless interface.

[0089] The network interface 304 can optionally include a standard wired interface, a wireless interface (e.g., a WI-FI interface).

[0090] The processor 301 can include one or more processing cores. The processor 301 connects various parts within the server through various interfaces and lines, performs various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. The processor 301 can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 301 can be integrated with a combination of one or more of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU is mainly used to process an operating system, a user interface, and an application program, etc.; the GPU is used to render and draw the content to be displayed on the display screen; and the modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 301, but can be implemented by a separate chip.

[0091] The memory 305 can include a random access memory (RAM) and a read-only memory (ROM). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 can include a program storage area and a data storage area. The program storage area can store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 305 can optionally be at least one storage device located away from the above-mentioned processor 301. Referring to Figure 3 The memory 305, as a kind of computer storage medium, can include an operating system, a network communication module, a user interface module, and a radiation monitoring application program.

[0092] In Figure 3 In the electronic device shown, the user interface 303 is mainly used to provide an interface for the user to input, and obtain data input by the user; and the processor 301 can be used to call the radiation monitoring application stored in the memory 305, and when executed by one or more processors 301, causes the electronic device to perform the method described in one or more of the above embodiments. It should be noted that for the foregoing method embodiments, in order to simply describe, they are all described as a combination of a series of actions, but those skilled in the art should know that the present application is not limited to the order of the actions described, because according to the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0093] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors, causes the electronic device to perform the method described in one or more of the above embodiments.

[0094] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0095] In the several embodiments provided by the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the units is only a logical function division. There can be another division during actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical or other forms.

[0096] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0097] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0098] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable memory. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a memory and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned memory includes: a U disk, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0099] The above is only exemplary embodiments of the present application, and cannot limit the scope of the present application. That is, any equivalent changes and modifications made in accordance with the teachings of the present application are still within the scope of the present application.

[0100] The present application is intended to cover any variations, uses or adaptive changes of the present application, which follow the general principles of the present application and include common knowledge or conventional technical means in the technical field not disclosed in the present application.

Claims

1. A method of radiation monitoring, characterized by, The method comprises: acquiring gamma radiation data and converting the gamma radiation data into current pulse signal data; amplifying the current pulse signal data and extracting pulse waveform features in the current pulse signal after amplification; grading the current pulse signal into pulse count data of different amplitude levels based on the pulse waveform features; inputting the pulse count data into a reference parameter comparison model and determining whether artificial gamma radiation source data exists in the gamma radiation data based on the reference parameter comparison model; if the artificial gamma radiation source data is confirmed to exist in the gamma radiation data, classifying and warning a pollution event through a preset mode.

2. The method of claim 1, wherein, The acquisition of gamma radiation data and the conversion of the gamma radiation data into current pulse signal data specifically comprises: absorbing the gamma radiation data in the irradiation area by a pre-set plastic scintillator to generate scintillation light signals related to gamma energy; transmitting the scintillation light signals to a photomultiplier tube, which amplifies the photoelectrons in the scintillation light signals through multiple stages under stable high voltage conditions and outputs the current pulse signal data.

3. The method of claim 1, wherein, The amplification of the current pulse signal data and the extraction of pulse waveform features in the current pulse signal after amplification specifically comprise: amplifying the current pulse signal data based on a preamplifier circuit; inputting the amplified current pulse signal data into a multi-stage comparator circuit, which grades the amplified current pulse signal data according to a set amplitude threshold; extracting the integral values of the graded current pulse signal data at different stages through an integral window method; calculating pulse discrimination parameters based on the integral values at different stages and representing the pulse waveform features based on the pulse discrimination parameters.

4. The method of claim 1, wherein, The grading of the current pulse signal into pulse count data of different amplitude levels based on the pulse waveform features specifically comprises: inputting the pulse waveform features into a pre-set multi-stage amplitude comparator and outputting the corresponding response results of the current pulse signal under pre-set conditions based on the multi-stage amplitude comparator, wherein the pre-set conditions are that a plurality of amplitude thresholds corresponding to a plurality of gamma energy ranges are set, and one gamma energy range corresponds to one amplitude threshold; counting the number of pulses falling into each amplitude threshold based on the corresponding response results of the current pulse signal in each gamma energy range; obtaining the corresponding graded pulse count data based on the number of pulses falling into each amplitude threshold.

5. The method of claim 1, wherein, Before the inputting of the pulse count data into a reference parameter comparison model and the determination of whether artificial gamma radiation source data exists in the gamma radiation data based on the reference parameter comparison model, the reference parameter comparison model needs to be constructed, specifically comprising: acquiring low-threshold integral count rates and high-threshold integral count rates under different amplitude levels in an environment without artificial gamma radiation source interference; calculating the ratio between the low-threshold integral count rate and the high-threshold integral count rate as an initial reference parameter; The historical pulse count data in a continuous sliding time window is acquired, and a time-related background fluctuation feature vector is constructed based on dynamic statistical indicators, including a count rate mean statistical indicator, a variance statistical indicator, a slope statistical indicator, and a skewness statistical indicator. Based on the initial reference parameter and the background fluctuation feature vector, a statistical feature space under the condition of no artificial gamma source interference is constructed by a natural background gamma radiation spectrum distribution. The reference parameter comparison model is constructed based on the statistical feature space.

6. The method of claim 5, wherein, The pulse count data is input into the reference parameter comparison model, and whether there is artificial gamma source data in the gamma radiation data is judged based on the reference parameter comparison model, specifically including: The pulse count data under different amplitude levels in the current sliding time window is acquired, and a multi-dimensional statistical vector is constructed. The Mahalanobis distance of the multi-dimensional statistical vector relative to the initial reference parameter is calculated. The Mahalanobis distance is mapped to a statistical significance probability value, and it is judged whether the statistical significance probability value is greater than a preset judgment threshold. If the statistical significance probability value is greater than the preset judgment threshold, it is confirmed that there is artificial gamma source data in the gamma radiation data.

7. The method of claim 6, wherein, If it is confirmed that there is artificial gamma source data in the gamma radiation data, a pollution event is classified and warned through a preset manner, specifically including: If it is confirmed that there is artificial gamma source data in the gamma radiation data, the KL divergence between the amplitude normalized distribution in the current sliding window and the historical background distribution is acquired. A joint deviation function is constructed based on the KL divergence, and the judgment result is supplemented and verified through the joint deviation function. If the verification result corresponding to the supplementary verification meets a preset deviation condition, a pollution event is classified and warned through the preset manner, which includes a location classification manner based on the spatial correlation between the pollution trigger source and the measurement point, a spectrum feature classification manner based on the proportion of sudden increase components in different amplitude levels, and a time domain burst classification manner based on the change trend of the joint deviation function in the continuous sliding window.

8. A radiation monitoring device, characterized by The device includes an acquisition module and a processing module, wherein, The acquisition module is used to acquire gamma radiation data and convert the gamma radiation data into current pulse signal data. The processing module is used to amplify the current pulse signal data and extract the pulse waveform features in the amplified current pulse signal. The current pulse signal is classified into pulse count data of different amplitude levels based on the pulse waveform features. The pulse count data is input into the reference parameter comparison model, and whether there is artificial gamma source data in the gamma radiation data is judged based on the reference parameter comparison model. If it is confirmed that there is artificial gamma source data in the gamma radiation data, a pollution event is classified and warned through a preset manner.

9. An electronic device, comprising: The electronic device comprises a processor, a communication bus, a user interface, a network interface, and a memory for storing instructions, the user interface and the network interface being configured to communicate with other devices, and the processor being configured to execute the instructions stored in the memory to cause the electronic device to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method of any one of claims 1 to 7.

Images