Radiation contamination simulation investigation equipment control method based on ultraviolet fluorescence signals

Abstract

The invention relates to a radiation contamination simulation detection equipment control method based on an ultraviolet fluorescence signal, which is applied to radiation contamination simulation detection equipment, and the equipment comprises a detection probe. The method comprises the steps that a fluorescence intensity original signal and an ambient light interference signal of a target area are collected, the distance between a detection probe and a target surface is obtained, a fluorescence simulation agent is sprayed on the target area and irradiated by an ultraviolet LED lamp, and the simulation agent is used for emitting fluorescence with the wavelength larger than 450 nm under excitation of ultraviolet light; based on the distance and the ambient light interference signal, performing coupling compensation on the fluorescence intensity original signal; and calculating a radiation simulation value according to the calibration fluorescence intensity signal, and outputting information such as the radiation simulation value, the safety level and the component type. According to the method, through non-radiation simulation agent replacement, signal calibration and multi-dimensional analysis technical means, the safety of radiation detection training is improved, the health threat of a real radioactive source to trainees is eliminated, and the accuracy and the information amount of a detection result are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of radiation detection simulation, and in particular relates to a control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals. Background Art

[0002] After a nuclear accident, radiation contamination typically occurs in the form of point, line, and surface sources. Examples include point-source contamination on personnel and vehicle surfaces, line-source contamination on roads, and surface-source contamination over large areas. Traditional radiation detection training often relies on live equipment and real radioactive sources (such as gamma radiation sources). However, real radioactive sources can cause external radiation damage to trainees and pose risks in the management of radioactive materials. Furthermore, live equipment training faces challenges such as a disconnect between operational procedures and simulation scenarios, and insufficient detection sensitivity. Consequently, a safe and efficient simulated radiation detection method is urgently needed. Summary of the Invention

[0003] Based on this, it is necessary to provide a control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals to address the above technical problems, so as to improve the safety and efficiency of detection and enhance the authenticity and reliability of simulation training.

[0004] In a first aspect, the present application provides a control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals, which is applied to radiation contamination simulation inspection equipment. The radiation contamination simulation inspection equipment includes a detection probe, including:

[0005] The original fluorescence intensity signal and ambient light interference signal of the target area are collected, and the distance between the detection probe and the target surface is obtained. The target area is sprayed with a fluorescent simulant and illuminated by an ultraviolet LED lamp. The simulant is used to emit fluorescence with a wavelength greater than 450nm under ultraviolet light excitation;

[0006] Based on the distance and ambient light interference signals, the original fluorescence intensity signal is coupled and compensated to obtain a calibrated fluorescence intensity signal;

[0007] The radiation simulation value is calculated based on the calibrated fluorescence intensity signal, and multi-dimensional radiation characteristic information is output. The multi-dimensional radiation characteristic information includes the radiation simulation value, safety level, component type and contamination trend prediction results.

[0008] In one embodiment, coupling compensation is performed on the original fluorescence intensity signal based on the distance and the ambient light interference signal to obtain a calibrated fluorescence intensity signal, including:

[0009] According to the ambient light interference signal and distance, the ambient light reference component is calculated using the following formula:

[0010]

[0011] Among them, Ibase is the ambient light reference component, I ambient is the ambient light interference signal, d is the distance, k1 is the ambient light ratio, and k2 is the ambient light attenuation coefficient;

[0012] Calculate the preliminary calibration signal based on the original fluorescence intensity signal and the ambient light reference component;

[0013] Based on the preliminary calibration signal and distance, the calibration fluorescence intensity signal is calculated using the following formula:

[0014]

[0015] Among them, I c To calibrate the fluorescence intensity signal, I temp is the preliminary calibration signal, Δd is the difference between the distance d and the preset standard detection distance, k3 is the fluorescence attenuation compensation coefficient, and k4 is the distance sensitivity correction parameter.

[0016] In one embodiment, the radiation simulation value is calculated by the following steps, including:

[0017] According to the calibration experimental data of fluorescent simulants with different concentrations, the corresponding calibration fluorescence intensity signal samples and equivalent radiation dose reference values ​​are collected;

[0018] Based on the calibration experimental data, calibration fluorescence intensity signal samples and equivalent radiation dose reference values, the least squares fitting process is performed to obtain the quadratic mapping model parameters;

[0019] The calibrated fluorescence intensity signal is input into the quadratic mapping model parameters to obtain the radiation simulation value.

[0020] In one embodiment, the component type is calculated by the following steps, including:

[0021] Extracting spectral characteristic parameters according to the fluorescence spectrum corresponding to the calibrated fluorescence intensity signal, the spectral characteristic parameters including characteristic peak wavelength, half-height width and characteristic peak area;

[0022] The spectral characteristic parameters are input into a lightweight convolutional neural network and similarity matching is performed with the simulants in a preset simulant spectral database to obtain the component type. The preset simulant spectral database stores standard spectral characteristic parameters of various simulants.

[0023] The ambient light interference intensity is calculated according to the ambient light interference signal, and the confidence of the component type is weightedly adjusted based on the ambient light interference intensity and the distance between the detection probe and the target area to obtain an updated confidence.

[0024] In one embodiment, the contamination trend prediction result is calculated by the following steps, including:

[0025] Collect radiation simulation values, distance changes between the detection probe and the target area, and ambient temperature and humidity data over N time periods to construct a time series input data set. The ambient temperature and humidity data includes temperature, humidity, and wind speed.

[0026] A two-dimensional diffusion equation is established based on Fick's second law. The environmental parameters in the time series input data set are input into the two-dimensional diffusion equation as dynamic terms to generate a spatiotemporal evolution model of contamination concentration.

[0027] The output sequence of the spatiotemporal evolution model of contamination concentration is input into the LSTM neural network for training, and the radiation simulation value prediction sequence and diffusion impact range heat map of the preset future time period are output to obtain the contamination trend prediction results.

[0028] In one embodiment, the security level is calculated by the following steps, including:

[0029] Collect radiation simulation value samples under different ambient light intensities and different detection distances, and divide the basic threshold intervals based on the K-means clustering algorithm to obtain the basic level threshold model. The basic threshold intervals include safe, warning, dangerous and serious;

[0030] According to the intensity of ambient light interference and the distance between the detection probe and the target area, the basic level threshold model is dynamically modified to obtain the dynamic level threshold;

[0031] The radiation simulation value is compared with the dynamic level threshold to obtain the safety level.

[0032] In one embodiment, collecting the original fluorescence intensity signal and the ambient light interference signal of the target area and obtaining the distance between the detection probe and the target surface includes:

[0033] The ranging module of the detection probe emits laser pulses and receives echo signals, and calculates the distance through the flight time;

[0034] The UV sensing module based on the detection probe obtains the fluorescence signal in the 450-600nm band under the excitation of the UV LED and obtains the original fluorescence intensity signal. The UV sensing module includes a photomultiplier tube, a photodiode and a charge-coupled device;

[0035] The ambient light signal in the 400-700nm band is collected by the photosensitive sensing module of the detection probe, and the ambient light signal is band-pass filtered to obtain the ambient light interference signal.

[0036] In a second aspect, the present application also provides a radiation contamination simulation detection equipment control system based on ultraviolet fluorescence signals, which is applied to radiation contamination simulation inspection equipment. The radiation contamination simulation inspection equipment includes a detection probe, including:

[0037] A data acquisition module is used to collect the original fluorescence intensity signal and ambient light interference signal of the target area, and obtain the distance between the detection probe and the target surface. The target area is sprayed with a fluorescent simulant and illuminated by an ultraviolet LED lamp. The simulant is used to emit fluorescence with a wavelength greater than 450nm under ultraviolet light excitation;

[0038] A signal calibration module is used to perform coupling compensation on the original fluorescence intensity signal based on the distance and ambient light interference signal to obtain a calibrated fluorescence intensity signal;

[0039] The radiation analysis module is used to calculate the radiation simulation value based on the calibrated fluorescence intensity signal and output multi-dimensional radiation characteristic information, which includes the radiation simulation value, safety level, component type and contamination trend prediction results.

[0040] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in the first aspect when executing the computer program.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps in the first aspect when the computer program is processed.

[0042] The above-mentioned control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals, by spraying a simulant that can emit fluorescence with a wavelength greater than 450nm and using ultraviolet LEDs for excitation, can replace the real radiation source with a non-radiation fluorescence signal, eliminating the risk of ionizing radiation damage, and effectively solving the problem of radioactive material hazards in traditional actual installation training. Secondly, when collecting the original fluorescence intensity signal, the ambient light interference signal and the detection distance are obtained simultaneously. Combined with distance compensation and environmental correction, the interference of distance changes and ambient light noise on the detection signal is further eliminated, significantly improving the detection sensitivity and data reliability under complex lighting conditions. Finally, the radiation simulation value is calculated based on the calibrated fluorescence intensity signal, and multi-dimensional radiation characteristic information such as radiation simulation value, safety level, component type and contamination trend prediction results is output, which not only improves the practicality of the detection results, but also improves the dynamic evaluation capability of radiation scene simulation.

[0043] Compared to traditional radiation detection training methods, this method, through the use of non-radioactive simulants, signal calibration, and multi-dimensional analysis techniques, not only improves the safety of radiation detection training and eliminates the health threat posed to trainees by real radioactive sources, but also significantly enhances the accuracy and information content of detection results. The output of multi-dimensional radiation signature information makes the simulations during training more realistic and complex than those at actual nuclear accident sites, helping trainees gain a more comprehensive and in-depth understanding and mastery of radiation detection skills, enabling them to perform their work more calmly and efficiently in actual nuclear accident response. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 A flow chart of a method for controlling a radiation contamination simulation detection device based on ultraviolet fluorescence signals provided by an exemplary embodiment of the present invention;

[0046] Figure 2 A schematic structural diagram of a control system for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals provided as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0048] In one embodiment, Figure 1 As shown, a method for controlling radiation contamination simulation detection equipment based on ultraviolet fluorescence signals is provided. This embodiment uses the method as an example to illustrate the application of the method to radiation contamination simulation inspection equipment. The radiation contamination simulation inspection equipment includes a detection probe. In this embodiment, the method includes the following steps:

[0049] S101: Collect the original fluorescence intensity signal and the ambient light interference signal of the target area, and obtain the distance between the detection probe and the target surface, wherein the target area is sprayed with a fluorescent simulant and irradiated by an ultraviolet LED lamp, and the simulant is used to emit fluorescence with a wavelength greater than 450nm under ultraviolet light excitation.

[0050] Specifically, the simulant sprayed on the target area can be a water-soluble simulant containing environmentally friendly organic phosphors (mainly composed of organic compounds of carbon, hydrogen, oxygen, and nitrogen). When irradiated by an ultraviolet LED lamp with a wavelength of 420nm and an emission power exceeding 60mW, it can be stimulated to produce fluorescence with a wavelength exceeding 450nm. Schematically, the water-soluble simulant can be modified from deionized water, a fluorescent agent, a fluorine-containing copolymer, a low-carbon polyol, and a variety of additives. It is a green, environmentally friendly, non-toxic, non-corrosive, and non-radiative aqueous simulant liquid. The fluorescence reflection spectrum can then be collected by the ultraviolet sensor detection circuit inside the detection probe to generate the original fluorescence intensity signal. In addition, natural light, incandescent lamps, and other light sources in the environment may generate interference signals. The detection probe can synchronously collect ambient light interference signals through the sensor to distinguish effective fluorescence from background noise. In addition, since the fluorescence intensity decays with increasing distance, based on the detection probe, the distance between the probe and the target surface can also be measured in real time by infrared ranging or ultrasonic sensors, providing the necessary parameters for subsequent signal calibration.

[0051] S102: Based on the distance and the ambient light interference signal, coupling compensation is performed on the original fluorescence intensity signal to obtain a calibrated fluorescence intensity signal.

[0052] Specifically, based on the distance between the detection probe and the target surface, a physical model (such as the inverse square law) can be used to compensate for the attenuation of the original fluorescence intensity signal to ensure consistent detection accuracy at different distances. In addition, the ambient light interference signal can be preliminarily filtered out by a filter for light with wavelengths other than 450nm and above, and further deducted from the background noise by a digital filtering algorithm such as Kalman filtering. Schematically, the signal amplitude can be adjusted by the distance parameter first, and then the ambient light interference baseline can be subtracted. Finally, a calibrated fluorescence intensity signal can be obtained, thereby ensuring detection sensitivity and anti-interference capability.

[0053] S103: Calculate the radiation simulation value according to the calibrated fluorescence intensity signal and output multi-dimensional radiation characteristic information, where the multi-dimensional radiation characteristic information includes the radiation simulation value, safety level, component type and contamination trend prediction result.

[0054] Specifically, the calibrated fluorescence intensity signal is linearly related to the concentration of the simulant. Therefore, it can be converted into a radiation simulation value by presetting a calibration curve. And a multi-level threshold can be preset to automatically match the safety level according to the radiation simulation value. For example, when the radiation simulation value exceeds the alarm threshold, an audible and visual alarm is triggered, such as a buzzer ringing continuously, to attract the attention of the training personnel. Furthermore, the fluorescence emission spectrum characteristics of the simulant can be matched and identified through a spectral library, and the component type can be output in combination with its water solubility, non-toxicity and other characteristics. Schematically, it is also possible to store historical detection data and use a time series analysis algorithm to predict the radiation contamination diffusion or attenuation trend, and output the contamination trend prediction result. In addition, the above-mentioned multi-dimensional radiation characteristic information can be displayed in Chinese on the LCD screen, and supports serial communication to be uploaded to the software control system.

[0055] The aforementioned control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals collects the original fluorescence intensity signal and ambient light interference signal of the target area, and determines the distance between the detection probe and the target surface. This method comprehensively gathers key data that influences radiation detection, providing a data foundation for subsequent radiation analysis and processing. Furthermore, by spraying a fluorescent simulant on the target area and irradiating it with an ultraviolet LED lamp, the simulant's characteristic of emitting fluorescence at a specific wavelength under ultraviolet light excitation creates a safe and controllable simulated radiation environment. This avoids the risk of ionizing radiation damage associated with traditional real-world radiation sources, simplifies the radioactive material management process, and reduces the cost and difficulty of control. Secondly, the original fluorescence intensity signal is coupled and compensated based on distance and ambient light interference signals, significantly improving signal quality and accuracy. The introduction of the distance parameter effectively corrects for signal intensity deviations caused by variations in the distance between the detection probe and the target surface, ensuring that the signal more accurately reflects the radiation simulation conditions in the target area. The processing of the ambient light interference signal accurately eliminates ambient light noise, further ensuring the purity and reliability of the fluorescence intensity signal. Finally, the radiation simulation value is calculated based on the calibrated fluorescence intensity signal, and information such as the radiation simulation value, safety level, component type and contamination trend prediction results are output, which greatly expands the application value and practicality of the detection results, making the simulated training scene highly close to the complexity of the real nuclear accident scene, providing trainees with a more comprehensive and in-depth learning opportunity, and helping them to master radiation detection skills.

[0056] In one embodiment, collecting the original fluorescence intensity signal and the ambient light interference signal of the target area and obtaining the distance between the detection probe and the target surface includes:

[0057] The ranging module of the detection probe emits laser pulses and receives echo signals, and calculates the distance through the flight time;

[0058] The UV sensing module based on the detection probe obtains the fluorescence signal in the 450-600nm band under the excitation of the UV LED and obtains the original fluorescence intensity signal. The UV sensing module includes a photomultiplier tube, a photodiode and a charge-coupled device;

[0059] The ambient light signal in the 400-700nm band is collected by the photosensitive sensing module of the detection probe, and the ambient light signal is band-pass filtered to obtain the ambient light interference signal.

[0060] Specifically, the ranging module can emit a laser pulse with a wavelength of 650nm at a fixed frequency. After being reflected by the target surface, the echo signal is received by the PIN photodiode. By measuring the time difference between the emission and reception of the laser pulse, the distance between the detection probe and the target surface can be calculated by combining the speed of light and the time of flight. Furthermore, the ultraviolet sensor module of the detection probe can be constructed by using a photomultiplier tube, a photodiode and a charge-coupled device. Among them, the photomultiplier tube can convert weak fluorescence signals into electrical signals and amplify them, and the photodiode can also convert fluorescence signals into electrical signals. The charge-coupled device is an image sensor that can output fluorescence signals as digital signals. By adopting an integrated design of photomultiplier tubes, photodiodes and charge-coupled devices, a three-layer signal acquisition system can be constructed. In principle, a miniaturized photomultiplier tube can be selected, and the photocathode response band is 300-650nm. When the fluorescence signal intensity is less than 10nW / cm 2 When the photomultiplier tube works first, it converts the weak light signal into electrical pulses, which are output after amplification and shaping, thus meeting the detection requirements of low-concentration simulants. Silicon photodiodes can be used to cover the 400-1100nm band. When the fluorescence signal intensity is 10nW / cm 2 -1μW / cm 2 Within the range, it works in reverse bias mode. The photocurrent is converted into a voltage signal after the transimpedance amplifier and sampled by digital-to-analog conversion, with a small linear error. In addition, a linear array charge-coupled device can be integrated, with a response band of 350-1000nm. When the fluorescence signal intensity is greater than 1μW / cm 2 When the light is emitted, the charge-coupled device can start spectral scanning to obtain the emission spectrum in the 450-600nm band. By comparing it with the built-in spectral library, the composition identification and concentration calibration of the fluorescent substance can be realized.

[0061] Specifically, the light sensor module can use an integrated photodiode covering the 400-700nm band to collect ambient light signals such as sunlight and incandescent lamps in real time. To distinguish ambient light from fluorescent signals, the collected ambient light signal can be bandpass filtered to remove unnecessary wavelengths while retaining portions that may overlap with the fluorescent signal, thereby obtaining the ambient light interference signal.

[0062] In one embodiment, coupling compensation is performed on the original fluorescence intensity signal based on the distance and the ambient light interference signal to obtain a calibrated fluorescence intensity signal, including:

[0063] According to the ambient light interference signal and distance, the ambient light reference component is calculated using the following formula:

[0064]

[0065] Among them, I base is the ambient light reference component, I ambient is the ambient light interference signal, d is the distance, k1 is the ambient light ratio, and k2 is the ambient light attenuation coefficient;

[0066] Calculate the preliminary calibration signal based on the original fluorescence intensity signal and the ambient light reference component;

[0067] Based on the preliminary calibration signal and distance, the calibration fluorescence intensity signal is calculated using the following formula:

[0068]

[0069] Among them, I c To calibrate the fluorescence intensity signal, I temp is the preliminary calibration signal, Δd is the difference between the distance d and the preset standard detection distance, k3 is the fluorescence attenuation compensation coefficient, and k4 is the distance sensitivity correction parameter.

[0070] Specifically, in actual detection scenarios, ambient light in the 450-600nm band, such as sunlight and incandescent lamps, can produce co-frequency interference with the fluorescence signal. Therefore, a computational model can be constructed by combining the ambient light spectral characteristics and distance attenuation to isolate this interference component. Furthermore, the ambient light reference component can be calculated based on the ambient light interference signal and distance using the above formula. In a darkroom environment, a standard light source can be used to simulate the spectral distribution of sunlight and illuminate the target area. A monochromator and a power meter can be used to simultaneously measure the total ambient light intensity of 400-700nm and the intensity of the 450-600nm sub-band (the true interference component). Linear fitting of multiple data sets can be performed to determine the proportional relationship between the two, namely the ambient light contribution coefficient k1. The ambient light attenuation coefficient k2 can be calculated by measuring the attenuation of the 450-600nm sub-band intensity with distance within a range of 0-20cm while maintaining the ambient light intensity constant, and fitting the attenuation curve using the least squares method. Finally, the ambient light reference component can be calculated. This component reflects the interference intensity on fluorescence detection at the current distance and ambient light level. The ambient light reference component is then subtracted from the raw fluorescence intensity signal to obtain a preliminary calibration signal.

[0071] Furthermore, the attenuation of the fluorescence signal with distance is affected by factors such as the diffuse reflectance characteristics of the target surface and the nonlinearity of the probe's receiving angle. Therefore, the above formula can be used to calibrate the fluorescence intensity signal. This formula comprehensively considers the attenuation of the fluorescence signal with distance and the effect of distance on signal detection sensitivity, thereby achieving precise calibration of the fluorescence intensity signal.

[0072] In one embodiment, the radiation simulation value is calculated by the following steps, including:

[0073] According to the calibration experimental data of fluorescent simulants with different concentrations, the corresponding calibration fluorescence intensity signal samples and equivalent radiation dose reference values ​​are collected;

[0074] Based on the calibration experimental data, calibration fluorescence intensity signal samples and equivalent radiation dose reference values, the least squares fitting process is performed to obtain the quadratic mapping model parameters;

[0075] The calibrated fluorescence intensity signal is input into the quadratic mapping model parameters to obtain the radiation simulation value.

[0076] Specifically, gradient samples can be prepared based on a water-soluble fluorescent simulant at volume concentrations of 0.1%, 0.5%, 1%, 5%, and 10%. Under fixed distance and stable ultraviolet excitation conditions, 100 sets of calibration fluorescence intensity signals and corresponding equivalent radiation dose reference values ​​are collected for each concentration sample. Subsequently, based on the above data, the least squares method can be used for fitting. This method can further determine the model parameters by minimizing the difference between the observed values ​​and the model predicted values, thereby obtaining the quadratic mapping model parameters that can best fit the experimental data, so that the model can accurately convert the calibration fluorescence intensity signal into the corresponding radiation dose value. Finally, the collected calibration fluorescence intensity signal is input into the above-mentioned quadratic mapping model to obtain a radiation simulation value. This radiation simulation value reflects the radiation intensity level of the target area under specific conditions, providing key data for subsequent safety assessments and decision-making.

[0077] In one embodiment, the component type is calculated by the following steps, including:

[0078] Extracting spectral characteristic parameters according to the fluorescence spectrum corresponding to the calibrated fluorescence intensity signal, the spectral characteristic parameters including characteristic peak wavelength, half-height width and characteristic peak area;

[0079] The spectral characteristic parameters are input into a lightweight convolutional neural network and similarity matching is performed with the simulants in a preset simulant spectral database to obtain the component type. The preset simulant spectral database stores standard spectral characteristic parameters of various simulants.

[0080] The ambient light interference intensity is calculated according to the ambient light interference signal, and the confidence of the component type is weightedly adjusted based on the ambient light interference intensity and the distance between the detection probe and the target area to obtain an updated confidence.

[0081] Specifically, in the fluorescence spectrum, the characteristic peak wavelength refers to the wavelength position at which the substance produces the strongest fluorescence radiation. Different substances will form characteristic peaks at specific wavelengths due to differences in molecular structure. Schematically, the significant peak positions in the spectrum can be identified by peak detection algorithms such as the derivative method and the local maximum search method, and the corresponding wavelength values ​​can be recorded to obtain the characteristic peak wavelength. The half-height width refers to the spectral line width of the characteristic peak at half the peak value, which reflects the broadening degree of the spectral peak and is related to factors such as the purity of the substance and intermolecular interactions. The characteristic peak area refers to the integrated area under the characteristic peak curve, which is positively correlated with the concentration of the substance. Numerical integration methods such as the trapezoidal method can be used to integrate the spectral data within the characteristic peak range to obtain the characteristic peak area. The above parameters can effectively characterize the properties of fluorescent substances.

[0082] Specifically, this embodiment can use a lightweight convolutional neural network MobileNet to process the extracted spectral feature parameters. The preset simulated agent spectrum database pre-stores the standard spectral feature parameters of a variety of simulated agents, including the characteristic peak wavelength, half-height width, characteristic peak area, and corresponding component type labels of each simulated agent. Based on the lightweight convolutional neural network MobileNet, its input layer can receive the vectorized representation of the spectral feature parameters, and after processing by the convolution layer, pooling layer and fully connected layer, output the similarity score with the simulated agent in the database. Schematically, the matching degree between the input feature and the standard feature in the database can be calculated by cosine similarity or Euclidean distance, and the similarity threshold can be set according to the actual detection requirements. Subsequently, the top k simulated agents with the highest similarity can be taken as candidate component types, where k≥1, and finally the most likely component type is determined by the softmax function or voting mechanism.

[0083] Furthermore, the impact of ambient light interference and detection distance on the detection results can be comprehensively considered to further improve the reliability of component type identification. For example, the intensity of the ambient light interference signal can be calculated using a root mean square or peak detection method. The larger the value, the more significant the interference of ambient light on the fluorescence signal. By establishing an initial confidence level, the confidence level of the component type is weighted and adjusted based on the ambient light interference intensity and the distance between the detection probe and the target area, resulting in an updated confidence level, which further quantifies the reliability of the component results.

[0084] In one embodiment, the contamination trend prediction result is calculated by the following steps, including:

[0085] Collect radiation simulation values, distance changes between the detection probe and the target area, and ambient temperature and humidity data over N time periods to construct a time series input data set. The ambient temperature and humidity data includes temperature, humidity, and wind speed.

[0086] A two-dimensional diffusion equation is established based on Fick's second law. The environmental parameters in the time series input data set are input into the two-dimensional diffusion equation as dynamic terms to generate a spatiotemporal evolution model of contamination concentration.

[0087] The output sequence of the spatiotemporal evolution model of contamination concentration is input into the LSTM neural network for training, and the radiation simulation value prediction sequence and diffusion impact range heat map of the preset future time period are output to obtain the contamination trend prediction results.

[0088] Specifically, this time-series input dataset comprehensively reflects the radiation status and environmental conditions of the target area at different time points, providing a foundation for subsequent model development and prediction. Fick's second law, which describes the temporal and spatial variation of a substance's concentration in a medium, is suitable for modeling the diffusion of radiation contamination. Therefore, a two-dimensional diffusion equation can be established based on Fick's second law. Since environmental parameters such as temperature, humidity, and wind speed have a significant impact on the diffusion of radiation contamination—for example, wind speed affects the speed and direction of contaminant propagation, while temperature and humidity affect the deposition and diffusion of contaminants—the diffusion equation can be numerically solved using the finite difference method. The environmental parameters in the time-series input dataset are used as dynamic boundary conditions to generate a concentration field sequence consisting of N time steps, with each time step corresponding to a gridded concentration distribution matrix for the target area. Subsequently, a long-short-term memory network can be used to nonlinearly model the spatiotemporal evolution of contamination concentrations to achieve trend prediction.

[0089] Schematically, the concentration field sequence output by the two-dimensional diffusion equation can be expanded into a one-dimensional vector and concatenated with the radiation simulation value sequence and the distance change sequence to form an input sequence containing spatiotemporal characteristics. This sequence is input into a long-short-term memory network, which outputs a radiation simulation value prediction sequence and a diffusion influence range heat map. A continuous distribution heat map can be generated by interpolating the concentration field matrix at the prediction time, using a color gradient to represent the concentration level. The influence range is defined by a boundary threshold to create a diffusion influence range heat map. This diffusion influence range heat map can be used to visually display future contamination trends, providing a quantitative basis for emergency response and personnel protection.

[0090] In one embodiment, the security level is calculated by the following steps, including:

[0091] Collect radiation simulation value samples under different ambient light intensities and different detection distances, and divide the basic threshold intervals based on the K-means clustering algorithm to obtain the basic level threshold model. The basic threshold intervals include safe, warning, dangerous and serious;

[0092] According to the intensity of ambient light interference and the distance between the detection probe and the target area, the basic level threshold model is dynamically modified to obtain the dynamic level threshold;

[0093] The radiation simulation value is compared with the dynamic level threshold to obtain the safety level.

[0094] Specifically, the radiation simulation value samples under different ambient light intensities and detection distances cover a wide range of possible detection conditions, providing a comprehensive data foundation for subsequent analysis. Each set of samples includes ambient light intensity, detection distance, and the corresponding radiation simulation value. Subsequently, based on the K-means clustering algorithm, the radiation simulation value is used as a clustering feature, and the Euclidean distance is used to measure the similarity between samples. The number of clusters is preset to four. By calculating the similarity between data points, the data is divided into different clusters, thereby determining the four basic level threshold ranges: safe, warning, dangerous, and severe. Furthermore, because the intensity of ambient light interference affects the accuracy of the radiation simulation value, and changes in detection distance affect the signal strength and distribution, the basic level threshold model can be dynamically modified based on the ambient light interference intensity and the distance between the detection probe and the target area to obtain a more accurate dynamic level threshold that is more adapted to the current detection conditions. The current safety level can then be determined based on which dynamic level threshold range the radiation simulation value falls within. This safety level intuitively reflects the safety status of the current radiation environment, providing clear guidance to operators so that they can take appropriate protective measures in a timely manner.

[0095] like Figure 2 As shown, based on the same inventive concept, the embodiment of the present application also provides a control system 200 for radiation contamination simulation detection equipment based on ultraviolet fluorescent signals for realizing the above-mentioned control method for radiation contamination simulation detection equipment based on ultraviolet fluorescent signals. The implementation scheme for solving the problem provided by this system is similar to the implementation scheme described in the above-mentioned method. Therefore, the specific limitations of one or more embodiments of the control system for radiation contamination simulation detection equipment based on ultraviolet fluorescent signals provided below can be found in the above-mentioned limitations on the control method for radiation contamination simulation detection equipment based on ultraviolet fluorescent signals, and will not be repeated here. And the system is applied to radiation contamination simulation inspection equipment, wherein the radiation contamination simulation inspection equipment includes a detection probe. Schematically, the system includes:

[0096] Data acquisition module 201 is used to collect the original fluorescence intensity signal and ambient light interference signal of the target area, and obtain the distance between the detection probe and the target surface, wherein the target area is sprayed with a fluorescent simulant and illuminated by an ultraviolet LED lamp, and the simulant is used to emit fluorescence with a wavelength greater than 450nm under ultraviolet light excitation;

[0097] A signal calibration module 202 is configured to perform coupling compensation on the original fluorescence intensity signal based on the distance and ambient light interference signal to obtain a calibrated fluorescence intensity signal;

[0098] The radiation analysis module 203 is used to calculate the radiation simulation value according to the calibrated fluorescence intensity signal and output multi-dimensional radiation characteristic information, which includes the radiation simulation value, safety level, component type and contamination trend prediction result.

[0099] In the above system, the data acquisition module 201 collects the target area's original fluorescence intensity signal, ambient light interference signals, and the distance between the detection probe and the target surface, providing a comprehensive data foundation for subsequent detection and analysis, ensuring the system can perceive the target area's real-time status information. The signal calibration module 202 couples and compensates the original fluorescence intensity signal based on the distance and ambient light interference signals, effectively eliminating the interference and attenuation effects of environmental and distance factors on the signal. This ensures that the calibrated fluorescence intensity signal more accurately reflects the actual situation in the target area, providing reliable data support for subsequent calculations. The radiation analysis module 203 calculates the radiation simulation value based on the calibrated fluorescence intensity signal and outputs multidimensional radiation signature information including the radiation simulation value, safety level, component type, and contamination trend prediction results. This not only enables a quantitative assessment of the radiation situation in the target area, but also provides a basis for safety protection decision-making through safety level determination, clarifies material properties through component type identification, and uses contamination trend prediction results to plan countermeasures in advance. Compared to single radiation detection methods, this system can comprehensively analyze the target area from multiple dimensions and levels, greatly improving the accuracy, reliability, and practicality of simulated radiation detection.

[0100] In an exemplary embodiment, the present invention further provides a computer device comprising a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals described herein. A multi-core processor is preferred to improve the system's parallel processing capabilities. The memory provides sufficient temporary storage space to support program execution and data processing. The memory capacity should be large enough to accommodate large amounts of supply information and computing tasks.

[0101] In an exemplary embodiment, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the control method of the radiation contamination simulation detection equipment based on ultraviolet fluorescence signals of the present application.

[0102] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A control method for radiation contamination simulation detection equipment based on ultraviolet fluorescence signals, characterized in that: Applied to a radiation contamination simulation inspection device, the radiation contamination simulation inspection device includes a detection probe, and the method includes: Collecting the original fluorescence intensity signal and the ambient light interference signal of the target area, and obtaining the distance between the detection probe and the target surface, wherein the target area is sprayed with a fluorescent simulant and illuminated by an ultraviolet LED lamp, and the simulant is used to emit fluorescence with a wavelength greater than 450nm under ultraviolet light excitation; Based on the distance and the ambient light interference signal, coupling compensation is performed on the original fluorescence intensity signal to obtain a calibrated fluorescence intensity signal; A radiation simulation value is calculated according to the calibrated fluorescence intensity signal, and multi-dimensional radiation characteristic information is output. The multi-dimensional radiation characteristic information includes the radiation simulation value, safety level, component type and contamination trend prediction result.

2. The method according to claim 1, characterized in that The performing coupling compensation on the original fluorescence intensity signal based on the distance and the ambient light interference signal to obtain a calibrated fluorescence intensity signal includes: According to the ambient light interference signal and the distance, the ambient light reference component is calculated using the following formula: Among them, I base is the ambient light reference component, I ambient is the ambient light interference signal, d is the distance, k1 is the ambient light ratio, and k2 is the ambient light attenuation coefficient; Calculating a preliminary calibration signal based on the fluorescence intensity raw signal and the ambient light reference component; According to the preliminary calibration signal and the distance, the calibration fluorescence intensity signal is calculated by the following formula: Among them, I c is the calibration fluorescence intensity signal, I temp is the preliminary calibration signal, Δd is the difference between the distance d and the preset standard detection distance, k3 is the fluorescence attenuation compensation coefficient, and k4 is the distance sensitivity correction parameter.

3. The method according to claim 1, characterized in that The radiation simulation value is calculated by the following steps, including: According to the calibration experimental data of the fluorescent simulant at different concentrations, corresponding calibration fluorescence intensity signal samples and equivalent radiation dose reference values ​​are collected; performing a least squares fitting process based on the calibration experimental data, the calibration fluorescence intensity signal sample and the equivalent radiation dose reference value to obtain quadratic mapping model parameters; The calibrated fluorescence intensity signal is input into the secondary mapping model parameters to obtain the radiation simulation value.

4. The method according to claim 1, wherein The component type is calculated by the following steps, including: Extracting spectral characteristic parameters according to the fluorescence spectrum corresponding to the calibrated fluorescence intensity signal, wherein the spectral characteristic parameters include characteristic peak wavelength, half-height width and characteristic peak area; Inputting the spectral characteristic parameters into a lightweight convolutional neural network and performing similarity matching with the simulants in a preset simulant spectral database to obtain the component type, wherein the preset simulant spectral database stores standard spectral characteristic parameters of multiple simulants; The ambient light interference intensity is calculated according to the ambient light interference signal, and the confidence of the component type is weightedly adjusted based on the ambient light interference intensity and the distance between the detection probe and the target area to obtain an updated confidence.

5. The method according to claim 1, wherein The contamination trend prediction result is calculated by the following steps, including: Collecting the radiation simulation value, the distance change between the detection probe and the target area, and the ambient temperature and humidity data within N time periods to construct a time series input data set, where the ambient temperature and humidity data includes temperature, humidity, and wind speed; Establishing a two-dimensional diffusion equation based on Fick's second law, and inputting the environmental parameters in the time series input data set as dynamic terms into the two-dimensional diffusion equation to generate a spatiotemporal evolution model of contamination concentration; The output sequence of the spatiotemporal evolution model of contamination concentration is input into the LSTM neural network for training, and the radiation simulation value prediction sequence and diffusion impact range heat map of the preset future time period are output to obtain the contamination trend prediction result.

6. The method according to claim 4, characterized in that The security level is calculated by the following steps, including: Collect radiation simulation value samples under different ambient light intensities and different detection distances, and divide the basic threshold intervals based on the K-means clustering algorithm to obtain a basic level threshold model, wherein the basic threshold intervals include safe, warning, dangerous and serious; Dynamically modifying the basic level threshold model according to the ambient light interference intensity and the distance between the detection probe and the target area to obtain a dynamic level threshold; The radiation simulation value is compared with the dynamic level threshold to obtain the safety level.

7. The method according to claim 1, characterized in that The collecting of the original fluorescence intensity signal and the ambient light interference signal of the target area and obtaining the distance between the detection probe and the target surface includes: The distance is obtained by calculating the time of flight based on the laser pulse emitted by the distance measuring module of the detection probe and the received echo signal; Based on the ultraviolet sensing module of the detection probe, a fluorescence signal in the 450-600nm band under the excitation of the ultraviolet LED is obtained to obtain an original fluorescence intensity signal. The ultraviolet sensing module includes a photomultiplier tube, a photodiode and a charge-coupled device; The ambient light signal in the 400-700 nm band is collected by the photosensitive sensor module of the detection probe, and the ambient light signal is band-pass filtered to obtain the ambient light interference signal.

8. A radiation contamination simulation detection equipment control system based on ultraviolet fluorescence signals, characterized in that: Applicable to radiation contamination simulation inspection equipment, the radiation contamination simulation inspection equipment includes a detection probe, and the system includes: a data acquisition module for collecting the original fluorescence intensity signal and the ambient light interference signal of the target area, and obtaining the distance between the detection probe and the target surface, wherein the target area is sprayed with a fluorescent simulant and illuminated by an ultraviolet LED lamp, and the simulant is used to emit fluorescence with a wavelength greater than 450nm under ultraviolet light excitation; a signal calibration module, configured to perform coupling compensation on the original fluorescence intensity signal based on the distance and the ambient light interference signal to obtain a calibrated fluorescence intensity signal; The radiation analysis module is used to calculate the radiation simulation value based on the calibrated fluorescence intensity signal and output multi-dimensional radiation characteristic information, wherein the multi-dimensional radiation characteristic information includes the radiation simulation value, safety level, component type and contamination trend prediction result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images