A training method, application method and related system for waveform data processing model

Abstract

The present application discloses a training method, application method and related system for a waveform data processing model, which relates to the fields of environmental radioactivity detection and particle physics technology. The training method includes obtaining raw waveform data and its corresponding real particle label data through a TPC detector; performing downsampling processing and data compression processing on the raw waveform data in sequence; inputting the pre-processed raw waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel in which it is located, determining the network loss based on the predicted probability values ​​of each category of the output particles and the real labels of the particles; updating the network parameters of the waveform data processing model through back propagation based on the network loss, and obtaining a trained waveform data processing model after multiple iterations. Using the trained waveform data processing model for particle identification not only saves data storage and reduces the amount of calculation, but also improves the accuracy of the multi-dimensional information particle identification results.

Description

Technical Field

[0001] The present application relates to the fields of environmental radioactivity detection and particle physics technology, and in particular to a training method, an application method and a related system for a waveform data processing model. Background Art

[0002] Ionizing radiation, including alpha and beta radiation, is radiation that can ionize atoms or molecules. While the human body cannot typically perceive ionizing radiation directly, exposure to radiation exceeding safe doses or sustained internal exposure from ingested radiation sources can cause serious harm. Therefore, monitoring alpha, beta, and other radiation levels in various environments is crucial. Alpha and beta radiation levels in daily life are very low, posing challenges to the discrimination capabilities of measuring instruments and achieving low-background detection.

[0003] To address this issue, a new measurement method based on the low-background measurement of α and β radiation using a TPC (Time Projection Chamber) detector has been developed. TPC uses waveform data generated by the drift of electron clouds generated by ionizing radiation in an electric field to determine particle trajectory and energy loss. In many ionizing radiation detection applications, the radioactivity of the sample being measured is relatively low, and the similarity and overlap between background particles and particles of interest make the development of TPC data processing methods crucial.

[0004] Currently, the mainstream TPC data processing method is a multi-dimensional information particle identification method based on track detection. This method primarily extracts features from waveform signals, measures physical information such as particle track, energy, and energy loss, and integrates multiple pieces of information to identify particles. However, in this multi-dimensional information particle identification method, the acquisition of multi-dimensional information typically requires large amounts of data acquisition and storage, as well as the processing of large amounts of data, which poses challenges to the hardware's data storage and processing capabilities. Furthermore, the analysis and feature selection of multi-dimensional information are crucial to the effectiveness and performance of the algorithm. In high-dimensional data spaces, feature selection and the curse of dimensionality can affect the accuracy and efficiency of the algorithm, making it susceptible to empirical factors and subject to significant errors, resulting in inaccurate and unreliable multi-dimensional information particle identification results.

[0005] In summary, how to improve the accuracy of multi-dimensional information particle identification results while saving data storage and reducing computational complexity, so as to solve the problems of large data volume, strong influence of subjective factors and low accuracy in traditional multi-dimensional information particle identification methods, has become a technical problem that needs to be solved urgently in this field. Summary of the Invention

[0006] The purpose of this application is to provide a training method, application method and related system of a waveform data processing model, which can improve the accuracy of multi-dimensional information particle identification results while saving data storage and reducing calculation amount.

[0007] To achieve the above objectives, this application provides the following solutions:

[0008] In a first aspect, the present application provides a method for training a waveform data processing model, the method comprising:

[0009] Obtain raw waveform data and its corresponding particle real label data; the raw waveform data is multi-channel particle waveform data collected by using a TPC detector, the TPC detector includes multiple channels, and the multiple channels synchronously collect particle waveform data to form the multi-channel particle waveform data; the particle real label data includes a real label for characterizing the actual category of the particle.

[0010] The original waveform data is sequentially subjected to downsampling processing and data compression processing to obtain pre-processed original waveform data.

[0011] The preprocessed raw waveform data is input into each channel of the waveform data processing model according to the spatial dimension of the channel, and the network loss is determined based on the predicted probability values ​​of each category of particles and the true labels of the particles output by the waveform data processing model; the waveform data processing model is a model with a multi-channel network structure established based on an artificial neural network, and the network structure of each channel includes multiple convolution layers and multiple pooling layers. The waveform data processing model is constructed by first passing through all convolution layers and then passing through all pooling layers.

[0012] According to the network loss, the network parameters of the waveform data processing model are updated through back propagation to perform multiple iterations on the waveform data processing model to obtain a trained waveform data processing model.

[0013] Optionally, downsampling processing and data compression processing are sequentially performed on the original waveform data to obtain pre-processed original waveform data, specifically including:

[0014] The original waveform data is downsampled by adopting an interval resampling method to obtain the downsampled original waveform data.

[0015] The downsampled raw waveform data is compressed and saved in a sparse tensor format according to the order of the channel numbers of the TPC detector to obtain the preprocessed raw waveform data.

[0016] Optionally, the particle categories include alpha particles, beta particles and muons.

[0017] In a second aspect, the present application provides an application method of a waveform data processing model, the application method of the waveform data processing model comprising:

[0018] Target waveform data is acquired, where the target waveform data is multi-channel particle waveform data of the particles to be measured collected using a TPC detector.

[0019] Downsampling processing and data compression processing are sequentially performed on the target waveform data to obtain preprocessed target waveform data.

[0020] The preprocessed target waveform data is input into each channel of the waveform data processing model according to the spatial dimension of the channel in which it is located to obtain the predicted probability value of each category of the particle to be measured; the waveform data processing model is a model trained according to the training method of the waveform data processing model described in the first aspect.

[0021] Optionally, after the step of inputting the preprocessed target waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel to obtain the predicted probability value of each category of the particle to be detected, the application method of the waveform data processing model further includes:

[0022] Set the confidence threshold.

[0023] The predicted probability value corresponding to the highest category among the predicted probability values ​​of each category of the particles to be tested is compared with the confidence threshold. The case where the predicted probability value is less than the confidence threshold is determined as a background case, and the case where the predicted probability value is greater than or equal to the confidence threshold is determined as a particle case. All background cases are eliminated and all particle cases are retained.

[0024] In a third aspect, the present application provides a training system for a waveform data processing model, the training system for the waveform data processing model comprising:

[0025] The first data acquisition module is used to obtain raw waveform data and its corresponding particle real label data; the raw waveform data is multi-channel particle waveform data collected by using a TPC detector, and the TPC detector includes multiple channels, and the multiple channels synchronously collect particle waveform data to form the multi-channel particle waveform data; the particle real label data includes a real label for characterizing the actual category of the particle.

[0026] The first preprocessing module is used to perform downsampling processing and data compression processing on the original waveform data in sequence to obtain preprocessed original waveform data.

[0027] A network loss determination module is used to input the preprocessed raw waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel, and determine the network loss based on the predicted probability values ​​of each category of particles and the true labels of the particles output by the waveform data processing model; the waveform data processing model is a model with a multi-channel network structure established based on an artificial neural network, and the network structure of each channel includes multiple convolution layers and multiple pooling layers. The waveform data processing model is constructed by first passing through all convolution layers and then passing through all pooling layers.

[0028] A training module is used to update the network parameters of the waveform data processing model through back propagation according to the network loss, so as to iterate the waveform data processing model multiple times to obtain a trained waveform data processing model.

[0029] Optionally, the first preprocessing module includes:

[0030] The downsampling unit is used to downsample the original waveform data by adopting an interval resampling method to obtain the downsampled original waveform data.

[0031] A data compression unit is used to compress and save the downsampled original waveform data into a sparse tensor format in the order of the channel numbers of the TPC detector to obtain the preprocessed original waveform data.

[0032] Optionally, the particle categories include alpha particles, beta particles and muons.

[0033] In a fourth aspect, the present application provides an application system of a waveform data processing model, the application system of the waveform data processing model comprising:

[0034] The second data acquisition module is used to acquire target waveform data, where the target waveform data is multi-channel particle waveform data of the particles to be measured collected by using a TPC detector.

[0035] The second preprocessing module is used to perform downsampling processing and data compression processing on the target waveform data in sequence to obtain preprocessed target waveform data.

[0036] A target particle category prediction module is used to input the preprocessed target waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel to obtain the predicted probability value of each category of the particle to be measured; the waveform data processing model is a model trained according to the waveform data processing model training method described in the first aspect.

[0037] Optionally, the application system of the waveform data processing model further includes:

[0038] A confidence threshold setting module, used to set a confidence threshold;

[0039] The screening module is used to compare the predicted probability value corresponding to the highest category among the predicted probability values ​​of each category of the particles to be tested with the confidence threshold, determine the case where the predicted probability value is less than the confidence threshold as a background case, and determine the case where the predicted probability value is greater than or equal to the confidence threshold as a particle case, and eliminate all background cases and retain all particle cases.

[0040] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0041] The present application provides a training method, application method, and related system for a waveform data processing model. By combining TPC detection technology, downsampling and data compression technology, and neural network technology, and applying them to specific waveform processing scenarios, on the one hand, multi-channel particle waveform data is collected by a TPC detector and pre-processed by downsampling and data compression, thereby obtaining waveform data with high compression ratio and high resolution, effectively saving data storage, reducing the amount of calculation, lowering the difficulty of operation, and improving the efficiency of waveform data processing and particle type prediction. On the other hand, in the context of multi-channel particle waveform data from a TPC detector, based on machine learning algorithms such as artificial neural networks, the multi-channel particle waveform data is input into each channel of the waveform data processing model according to the spatial dimension of the channel, and the corresponding particle identification results are output. By training the waveform data processing model, the trained waveform data processing model is endowed with high-precision particle identification capabilities, achieving high-precision particle identification, and solving the problems of large data volume and low accuracy in traditional multi-dimensional information particle identification methods. The model can be widely applied to high-sensitivity detection scenarios of α and β radionuclides in radiation environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. 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 creative work.

[0043] Figure 1 A flowchart of a method for training a waveform data processing model provided in Example 1 of the present application.

[0044] Figure 2 This is a schematic diagram of the structure of the TPC detector provided in Example 1 of the present application.

[0045] Figure 3This is a schematic diagram of the waveform data processing method provided in Example 1 of the present application.

[0046] Figure 4 A waveform diagram of the original waveform data provided in Example 1 of the present application.

[0047] Figure 5 This is a waveform diagram of the original waveform data provided in Example 1 of the present application after data compression processing.

[0048] Figure 6 This is a waveform diagram of the original waveform data provided in Example 1 of the present application after downsampling.

[0049] Figure 7 This is a structural diagram of the waveform data processing model provided in Example 1 of the present application.

[0050] Figure 8 This is a schematic diagram of the structure of the convolution layer and pooling layer of each channel of the waveform data processing model provided in Example 1 of the present application.

[0051] Figure 9 This is a schematic diagram of the structure of the fully connected output part of the waveform data processing model provided in Example 1 of the present application.

[0052] Figure 10 This is a starting point distribution diagram of the β sample cases provided in Example 1 of the present application.

[0053] Figure 11 This is a distribution diagram of the starting points of the remaining cases of the β sample after screening provided in Example 1 of the present application.

[0054] Figure 12 This is a starting point distribution diagram of the background sample case provided in Example 1 of this application.

[0055] Figure 13 This is a distribution diagram of the starting points of the remaining cases of the background sample after screening provided in Example 1 of the present application.

[0056] Figure 14 A flowchart of the application method of the waveform data processing model provided in Example 2 of the present application.

[0057] Figure 15 This is a structural block diagram of the training system for the waveform data processing model provided in Example 3 of the present application.

[0058] Figure 16 This is a structural block diagram of the application system of the waveform data processing model provided in Example 4 of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0060] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0061] Example 1

[0062] like Figure 1 As shown, this embodiment proposes a training method for a waveform data processing model, which includes the following steps.

[0063] Step S1: Obtain raw waveform data and its corresponding true particle label data. The raw waveform data is multi-channel particle waveform data acquired using a TPC detector. The TPC detector includes multiple channels that synchronously acquire particle waveform data to form the multi-channel particle waveform data. The true particle label data includes true labels that characterize the actual categories of the particles.

[0064] Step S2: performing downsampling processing and data compression processing on the original waveform data in sequence to obtain pre-processed original waveform data.

[0065] Step S3: Input the preprocessed raw waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel in which it is located, and determine the network loss based on the predicted probability values ​​of each category of particles output by the waveform data processing model and the true label of the particles. The network loss refers to the network loss value of the loss function. The waveform data processing model is a model with a network structure with multiple channels established based on an artificial neural network. The network structure of each channel corresponds to the input of the preprocessed raw waveform data of one spatial dimension. The network structure of each channel includes multiple convolutional layers and multiple pooling layers, and is constructed by first passing through all convolutional layers and then passing through all pooling layers. The convolution kernel sizes of different convolutional layers in the same channel are different.

[0066] Step S4: updating the network parameters of the waveform data processing model through back propagation according to the network loss, so as to iterate the waveform data processing model multiple times to obtain a trained waveform data processing model.

[0067] In this embodiment, step S2 sequentially performs downsampling processing and data compression processing on the original waveform data to obtain pre-processed original waveform data, which specifically includes the following steps.

[0068] Step S21 : downsampling the original waveform data by adopting interval resampling to obtain downsampled original waveform data.

[0069] Step S22: compress and save the downsampled original waveform data into a sparse tensor format according to the order of the channel numbers of the TPC detector to obtain the preprocessed original waveform data.

[0070] In this embodiment, the particle types mainly include α particles, β particles, and muons, etc. In other words, the particle types can be α particles, β particles, or muons, etc.

[0071] like Figure 2 As shown, this embodiment adopts a TPC detector, which consists of two parts: a drift region and an end cap readout detector. The end cap readout detector mainly includes a tank body and two end faces. The two end faces are arranged opposite to each other. The two end faces are the cathode plane and the anode plane respectively, and the remaining faces are the sides of the tank body, as shown in FIG. Figure 2 As shown, a readout bar is provided on the anode plane for reading out electronic signals. In this TPC detector, the drift region is filled with a working gas such as argon or isobutane and is subjected to a uniform drift electric field. When charged particles pass through the drift region, they ionize the working gas molecules, generating electron-ion pairs. Under the influence of the uniform drift electric field within the drift region, electrons drift toward the anode region, generating avalanche amplification. Ultimately, the electronic signal is read out by the readout bar of the end cap readout detector. After pre-amplification and waveform digitization, multi-channel raw waveform data is obtained.

[0072] In this embodiment, the data from a single channel of a TPC detector is a one-dimensional array consisting of a series of numbers. Because a TPC detector has multiple channels, the data obtained by the TPC detector is a multidimensional array, with the number of dimensions equal to the number of channels. Multi-channel particle waveform data from a TPC detector refers to the particle waveform data from multiple channels obtained through preamplification and waveform digitization in the TPC detector.

[0073] The waveform data processing process based on the TPC detector in this embodiment mainly includes a data preprocessing algorithm, a neural network algorithm, an algorithm training method, and an algorithm deployment.

[0074] The data preprocessing algorithm involves resampling the multi-channel particle waveform data (i.e., the raw waveform data) collected by the TPC detector through interval resampling. Using the peak position of the waveform as the interval center, appropriate intervals are selected for interval resampling while preserving essential waveform information. This reduces the waveform sampling rate and compresses the data according to the TPC detector's channel number, saving it in a sparse tensor format. The first column of the sparse tensor contains the channel number of the triggered channel, and the second column contains the corresponding waveform data. The sparse tensor also stores the data's dimensionality and data type.

[0075] In this embodiment, interval resampling includes interval sampling and interval resampling. Interval sampling is performed first, and then interval resampling is performed on the basis of interval sampling. Since the characteristic of the sampled waveform is a pulse signal, the peak position of the pulse is used as the center of the interval, and a suitable interval is selected as the sampling interval. This interval needs to include the complete rising and falling edges of the pulse. Then, interval resampling is performed within the above-mentioned sampling interval. Specifically, sampling is performed once every several points within the sampling interval, and all the sampled points are combined into a new waveform signal. The interval between the sampling points needs to ensure that the original waveform pulse signal is not distorted. At this point, the entire interval resampling is completed.

[0076] The neural network algorithm includes dividing the above-mentioned preprocessed raw waveform data into different dimensions, and feeding the preprocessed raw waveform data of different dimensions as input into a multi-channel waveform data processing model. For the preprocessed raw waveform data, the waveform data of each dimension is used as the input of a channel. The waveform data of each dimension is first processed by multiple convolution layers and then by a pooling layer. The convolution layer uses a set of learnable filters (convolution kernels) to perform convolution operations on the input to capture features of different scales. The output of the convolution layer is downsampled by the pooling layer to reduce the dimension of the feature map. After the data output by the pooling layer of each channel is reduced in dimensionality, it is spliced ​​into a one-dimensional tensor form, and the spliced ​​data is input into a neural network module composed of multiple fully connected layers. In each fully connected layer, the input is weighted and summed, and nonlinear transformation is performed through an activation function, such as the ReLU function, to further extract features and perform classification, and map the input to the output space. The last layer activation function uses the sigmoid function to compress the output to the range of [0, 1]. The network finally normalizes the output particle identification results through the Softmax layer.

[0077] In this embodiment, the pre-processed waveform data is divided into dimensions (such as X dimension and Y dimension) and input into a multi-channel artificial neural network in parallel. The meaning of the dimension division is the same as the channel division when the TPC detector is collected. The number of dimensions is equal to the number of channels of the neural network and the number of types of detector channels, that is, the number of dimensions, the number of channels of the neural network and the number of types of detector channels are all equal. Each channel corresponds to waveform data of one dimension. The waveform data of each dimension is first processed by multiple convolution layers and then by pooling layers. After the data output by the pooling layer of each channel is reduced in dimensionality, it is spliced ​​into a one-dimensional tensor form. The spliced ​​data is input into a neural network composed of multiple fully connected layers and output. The network normalizes the output particle identification results through the Softmax layer. The network parameters are updated by comparing with the original true labels and through back propagation and iterative training.

[0078] During preprocessing, the multidimensional array has been converted into a multidimensional sparse tensor. Therefore, the waveform data processing model receives input in a multidimensional sparse tensor format. The output can be viewed as the predicted probability of each particle event being classified as a specific particle type, i.e., the confidence value. Particle types can include alpha particles, beta particles, and muons. Therefore, the waveform data processing model outputs the predicted probability of the current particle event being an alpha particle, the predicted probability of the current particle event being a beta particle, and the predicted probability of the current particle event being a muon.

[0079] In this embodiment, the algorithm training method involves dividing the collected particle waveform data and background waveform data into training and test sets according to a certain ratio. In this embodiment, the sample ratio of the training set to the test set is 8:2. The training set is used for parameter updating and training of the neural network, while the test set is used to evaluate the network's performance on unseen data. Furthermore, during network training, a dynamic learning rate algorithm and regularization techniques such as dropout layers are employed to further improve network accuracy and prevent overfitting. In this embodiment, the algorithm deployment involves loading the particle event data and background event data into a trained waveform data processing model and outputting the results. The output is a predicted probability value, which can be considered the predicted probability of each event being classified as a particle of different types, i.e., the confidence level. The output is strictly filtered by setting a confidence threshold. Only particles with consistent labels and predicted probabilities greater than or equal to the confidence threshold are considered correctly classified, thereby achieving a high background rejection rate and a high particle retention rate.

[0080] Label consistency refers to the consistency between the predicted label of the particle output by the waveform data processing model and its actual label. In practical applications, a trained waveform data processing model is used to classify and screen particles. The particle waveform data can be input into the trained waveform data processing model, and the trained waveform data processing model can predict the probabilities of different particle categories corresponding to the particle waveform data. The predicted probability value A corresponding to the category Q with the largest predicted probability value is selected and compared with the confidence threshold B. If the predicted probability value A is greater than or equal to the confidence threshold B, then the category corresponding to the particle waveform data is determined to be Q. Q can be one of the categories of alpha particles, beta particles, or muons. If the predicted probability value A is less than the confidence threshold B, then the category corresponding to the particle waveform data is determined to be background. Here, background refers to all other signals or noise other than the target signal during the measurement process. This case is a background case.

[0081] Figure 3 The schematic diagram of the waveform data processing method based on the TPC detector is shown in FIG. Figure 3 As shown, the specific process of waveform data processing mainly includes the following contents.

[0082] 1) Based on the raw waveform data collected by the TPC detector, the raw waveform data is multi-channel particle waveform data, and then the collected raw waveform data is subjected to a preprocessing process of downsampling and data compression. Figure 4 It is a waveform diagram of the original waveform data; Figure 5 It is a waveform diagram of the original waveform data after data compression processing; Figure 6 The figure below is a waveform diagram of the original waveform data after downsampling. During preprocessing, the original waveform data is first downsampled by selecting a fixed interval where the pulse peak is located, and then downsampling is performed using interval sampling to obtain the downsampled waveform data. The downsampled waveform data is then combined into waveform data in a sparse tensor format according to the channel number sequence. Only the positions and values ​​of the non-zero elements need to be stored to obtain the preprocessed original waveform data. The channel number sequence is determined by the TPC detector. The channel number sequence of the TPC detector can be channel number 1 data, channel number 2 data, channel number 3 data, and so on.

[0083] 2) The pre-processed raw waveform data is input into the pre-built waveform data processing model in different dimensions for training. In this embodiment, the model is first trained through multiple convolution layers and then through pooling layers. Figure 7 and Figure 8As shown, the preprocessed raw waveform data obtained in 1), i.e., the waveform data in sparse tensor format, is input into the waveform data processing model dimensionally, with each dimension corresponding to a neural network input channel in the model. In each input channel, the waveform data is first converted to a normal tensor format, and then normalized. The normalized waveform data first passes through multiple connected convolutional layers, and then through a pooling layer to output the data for each channel.

[0084] 3) The data of each channel output by the pooling layer is processed by dimensionality reduction and splicing, and the final particle type identification result is output through the fully connected layer. Figure 7 and Figure 9 As shown, the result of each channel in 2) is still a two-dimensional tensor. In order to facilitate the dimensionality reduction and feature extraction of the fully connected layer, the data output by each channel of the neural network is reduced in dimensionality and then combined and spliced ​​into a one-dimensional tensor form. The spliced ​​data is input into a neural network composed of multiple fully connected layers to output the identification result representing the particle type. The last layer activation function uses the sigmoid function. The network outputs the identification result through the Softmax layer. The identification result is a one-dimensional tensor with a length of M, where M is the number of categories of particles to be identified. Each value of the vector represents the predicted probability value of the input case corresponding to each particle, that is, the confidence value. By comparing with the real label representing the actual category of the particle, back propagation and iterative training are performed to update the network parameters of the waveform data processing model and obtain a trained waveform data processing model, thereby improving the performance of the waveform data processing model and improving the classification accuracy of the particle category by the waveform data processing model.

[0085] 4) The target waveform data is loaded into the trained waveform data processing model, and the output result of the network Softmax layer is used as the predicted probability value. The target waveform data is screened by setting the probability-related confidence threshold.

[0086] The target waveform data is different from the original waveform data. The original waveform data is used to train the waveform data processing model to obtain a trained waveform data processing model. The target waveform data is the waveform data to be classified and screened, which is actually processed by the trained waveform data processing model after the trained waveform data processing model is obtained. The trained waveform data processing model is used to perform particle classification and case screening on the target waveform data. During data screening, cases whose output predicted probability values ​​are greater than or equal to the confidence threshold are retained, while cases whose output predicted probability values ​​are less than the confidence threshold are screened out. By setting a confidence threshold, this embodiment increases the screening rate of background cases while retaining higher-confidence cases.

[0087] This embodiment takes 512 sampling point data of 240 channels acquired by the TPC detector as an example to illustrate the specific implementation process of the technical solution of this embodiment. The specific process includes the following contents.

[0088] (1) Downsampling and data compression are performed on the acquired multi-channel sampling waveforms.

[0089] The TPC detector used in this embodiment has a total of 240 data readout channels, with X-dimension and Y-dimension each occupying 120 channels, and each channel has a total of 512 sampling points.

[0090] The peak value of each channel waveform is roughly located near the 150th sampling point, such as Figures 4 to 6 As shown, using the preprocessing method based on downsampling and data compression of this embodiment, the interval between the 50th and 250th sampling points is selected, and the channel waveforms are retained to 100 sampling points by sampling at intervals of one point. The above operation is performed on the waveforms of all channels. The downsampled data is then combined into a sparse tensor format according to the channel number sequence.

[0091] (2) The processed waveform data is input into the neural network in different dimensions, first through multiple convolutional layers, and then through pooling layers for processing and training.

[0092] In this embodiment, the waveform data in the sparse tensor format obtained in (1) is input into a dual-channel neural network in batches, with the X dimension and Y dimension corresponding to a neural network input channel respectively. In each input channel, the waveform data is first converted to a normal tensor format, and then the waveforms of the 120 channels of the data are normalized. The processed waveform data first passes through three connected convolutional layer networks, and then passes through the pooling layer to output the channel results. Finally, each channel outputs a tensor with a shape of 15×4.

[0093] (3) After reducing the dimensionality of the data output from each channel and splicing them together, the identification results are output through the fully connected layer.

[0094] In this embodiment, the data output by each channel of the neural network in (2) is flattened and reduced in dimension, and each channel obtains a one-dimensional tensor of length 60. Then, each one-dimensional tensor is sequentially combined and spliced ​​into a one-dimensional tensor of length 120. The spliced ​​one-dimensional tensor is input into a neural network module consisting of three fully connected layers to output the identification result. The output of the network is a one-dimensional tensor of length 2, each value representing the predicted probability value of the input case corresponding to the corresponding particle. By comparing with the label, back propagation is performed and iterative training is performed.

[0095] (4) Load the data onto the trained network, regard the output of the network Softmax layer as the predicted probability value, and filter the data by setting the confidence threshold related to the probability.

[0096] In this embodiment, cases with predicted probabilities greater than or equal to the confidence threshold are retained, while cases with predicted probabilities less than the confidence threshold are filtered out. After setting a reasonable confidence threshold, the changes in the hit position distribution of cases before and after screening are analyzed, ultimately allowing case particles to be accurately screened.

[0097] This embodiment sets a reasonable confidence threshold, identifying events with output confidence values ​​greater than or equal to the confidence threshold as desired event particles and retaining them. Events with output confidence values ​​less than the confidence threshold are identified as background data and are screened out. This embodiment conducted experiments on β-sample events using the aforementioned confidence threshold screening method. The experimental results showed that the number of β-sample events before screening was 87,340, while the number of remaining β-sample events after screening was 48,242, resulting in a retention rate of 55.0%. The number of background sample events before screening was 19,161, while the number of remaining background sample events after screening was 302, resulting in a screening rate of 98.4%. Figures 10 to 13 It is the distribution diagram of the starting points obtained by different sample data before and after the background cases are filtered out using the above algorithm, where Figure 10 is the distribution diagram of the starting points of the β sample cases, Figure 11 is the distribution diagram of the starting points of the remaining cases of the β sample after screening, Figure 12 is the distribution map of the starting points of background sample cases, Figure 13 This is the distribution diagram of the starting points of the remaining cases of the background sample after screening. Figures 10 to 13 It can be seen intuitively that the starting point distribution of the β sample events is more concentrated, indicating that the vast majority of background events outside the incident window are filtered out, while the particles considered to be real events within the incident window are largely retained. This demonstrates the effectiveness of the technical solution of this embodiment in screening event particles. The starting point distribution of background events typically forms two prominent circular patterns, with fewer events in the outer circle and more in the inner circle. This is due to the stepped mechanical structure of the incident window. This demonstrates the accuracy of the starting point distribution of the technical solution of this embodiment.

[0098] This embodiment establishes, trains, and applies a waveform data processing model based on a TPC detector. This significantly compresses the data volume while retaining relatively complete waveform information, effectively reducing the amount and complexity of data calculations and the consumption of computer memory and other hardware during algorithm processing. Furthermore, this embodiment directly processes the raw waveform data output by the TPC detector, rather than extracting multidimensional physical features. This allows for maximum utilization of the waveform and time information present in the waveform, avoiding the subjectivity and large errors associated with feature selection in traditional algorithms and improving the accuracy of particle waveform data classification and screening. Compared to traditional algorithms, this method offers a simpler approach, greater program portability, and higher background resolution.

[0099] Example 2

[0100] like Figure 14 As shown, this embodiment provides an application method of a waveform data processing model. The application method is actually a method for using the waveform data processing model trained in Example 1. The application method of the waveform data processing model includes the following steps.

[0101] 101. Obtain target waveform data, where the target waveform data is multi-channel particle waveform data of a particle to be measured collected by using a TPC detector.

[0102] 102. Perform downsampling processing and data compression processing on the target waveform data in sequence to obtain pre-processed target waveform data.

[0103] 103. Input the preprocessed target waveform data into each channel of a waveform data processing model according to the spatial dimension of the channel to obtain a predicted probability value for each category of the particle to be detected. The waveform data processing model is a model trained according to the waveform data processing model training method described in Example 1.

[0104] In this embodiment, after step 103, the application method of the waveform data processing model may further include the following steps.

[0105] 104. Set the confidence threshold.

[0106] 105. Compare the predicted probability value corresponding to the highest category among the predicted probability values ​​for each category of the particles to be tested with the confidence threshold to screen out background events and particle events. Specifically, cases with predicted probabilities less than the confidence threshold are considered background events, while cases with predicted probabilities greater than or equal to the confidence threshold are considered particle events. All background events are eliminated, and all particle events are retained.

[0107] Example 3

[0108] like Figure 15 As shown, this embodiment provides a training system for a waveform data processing model, which is actually the system corresponding to the training method of Example 1. The training system for the waveform data processing model includes the following functional modules.

[0109] The first data acquisition module M1 is used to obtain original waveform data and its corresponding particle real label data; the original waveform data is multi-channel particle waveform data collected by using a TPC detector, and the TPC detector includes multiple channels, and the multiple channels synchronously collect particle waveform data to form the multi-channel particle waveform data; the particle real label data includes a real label for characterizing the actual category of the particle.

[0110] The first preprocessing module M2 is used to perform downsampling processing and data compression processing on the original waveform data in sequence to obtain preprocessed original waveform data.

[0111] The network loss determination module M3 is used to input the preprocessed original waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel, and determine the network loss based on the predicted probability values ​​of each category of particles output by the waveform data processing model and the true label of the particles; the waveform data processing model is a model with a multi-channel network structure established based on an artificial neural network, and the network structure of each channel includes multiple convolution layers and multiple pooling layers. The waveform data processing model is built by first passing through all convolution layers and then passing through all pooling layers.

[0112] The training module M4 is used to update the network parameters of the waveform data processing model through back propagation according to the network loss, so as to iterate the waveform data processing model multiple times to obtain a trained waveform data processing model.

[0113] In this embodiment, the first pre-processing module M2 includes the following functional units.

[0114] The downsampling unit is used to downsample the original waveform data by adopting an interval resampling method to obtain the downsampled original waveform data.

[0115] A data compression unit is used to compress and save the downsampled original waveform data into a sparse tensor format in the order of the channel numbers of the TPC detector to obtain the preprocessed original waveform data.

[0116] Example 4

[0117] like Figure 16 As shown, this embodiment provides an application system of a waveform data processing model, which is actually a system corresponding to the application method of embodiment 2. The application system of the waveform data processing model includes the following functional modules.

[0118] The second data acquisition module N1 is used to acquire target waveform data, where the target waveform data is multi-channel particle waveform data of the particles to be measured acquired by using a TPC detector.

[0119] The second preprocessing module N2 is used to perform downsampling processing and data compression processing on the target waveform data in sequence to obtain preprocessed target waveform data.

[0120] The target particle category prediction module N3 is used to input the preprocessed target waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel, so as to obtain the predicted probability value of each category of the particle to be measured; the waveform data processing model is a model trained according to the training method of the waveform data processing model described in Example 1.

[0121] The confidence threshold setting module N4 is used to set the confidence threshold.

[0122] The screening module N5 is used to compare the predicted probability value corresponding to the highest category among the predicted probability values ​​of each category of the particles to be tested with the confidence threshold, determine the case where the predicted probability value is less than the confidence threshold as a background case, and determine the case where the predicted probability value is greater than or equal to the confidence threshold as a particle case, and eliminate all background cases and retain all particle cases.

[0123] This embodiment pre-processes the multi-channel signal waveform data of the TPC detector obtained through pre-amplification and waveform digitization to obtain waveform data with a high compression ratio while maintaining a relatively high resolution. The waveform data of each channel is simultaneously input into a pre-built and trained multi-channel artificial neural network model according to the spatial distribution to obtain the particle identification result. In the context of utilizing the original waveform data of the TPC detector, high-precision particle identification is achieved based on machine learning algorithms such as artificial neural networks. Under the premise of achieving a reasonable retention rate for the particles of interest, most background events are screened out. In addition, making full use of the time and space information of the interaction between the incident particles and the TPC detector hidden in the signal waveform of the TPC detector can effectively increase the particle identification capability, which is particularly suitable for achieving high-sensitivity detection of α and β radioactive nuclides in a radiation environment, and has broad application prospects in the fields of environmental radioactivity detection, radiopharmaceutical research, etc.

[0124] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A training method for a waveform data processing model, characterized in that: The training method of the waveform data processing model includes: Obtaining raw waveform data and its corresponding real particle label data; the raw waveform data is multi-channel particle waveform data acquired using a TPC detector, the TPC detector including multiple channels, the multiple channels synchronously acquiring particle waveform data to form the multi-channel particle waveform data; the real particle label data includes a real label for characterizing the actual category of the particle; The original waveform data is sequentially downsampled and compressed to obtain preprocessed original waveform data; wherein, an interval resampling method is adopted, with the peak position of the waveform as the interval center, a suitable interval is selected, and the original waveform data is interval resampled while retaining basic waveform information to obtain downsampled original waveform data; according to the order of the channel numbers of the TPC detector, the downsampled original waveform data is compressed and saved in a sparse tensor format, the first column of the sparse tensor is the channel number of the triggered channel, and the second column is the corresponding waveform data, and the sparse tensor saves the dimension size and data type of the data to obtain the preprocessed original waveform data; The preprocessed raw waveform data is input into each channel of the waveform data processing model according to the spatial dimension of the channel, and the network loss is determined based on the predicted probability value of each category of particles and the true label of the particles output by the waveform data processing model; the waveform data processing model is a model with a multi-channel network structure established based on an artificial neural network, and the network structure of each channel includes multiple convolution layers and multiple pooling layers. The waveform data processing model is constructed by first passing through all convolution layers and then passing through all pooling layers. According to the network loss, the network parameters of the waveform data processing model are updated through back propagation to perform multiple iterations on the waveform data processing model to obtain a trained waveform data processing model.

2. The training method of the waveform data processing model according to claim 1, characterized in that: The particle classes include alpha particles, beta particles, and muons.

3. An application method of a waveform data processing model, characterized in that: The application method of the waveform data processing model includes: Acquire target waveform data, where the target waveform data is multi-channel particle waveform data of the particles to be measured collected using a TPC detector; performing downsampling processing and data compression processing on the target waveform data in sequence to obtain preprocessed target waveform data; The preprocessed target waveform data is input into each channel of the waveform data processing model according to the spatial dimension of the channel to obtain the predicted probability value of each category of the particle to be measured; the waveform data processing model is a model trained according to the training method of the waveform data processing model according to claim 1.

4. The application method of the waveform data processing model according to claim 3, characterized in that: After the step of inputting the preprocessed target waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel to obtain the predicted probability value of each category of the particles to be detected, the application method of the waveform data processing model further includes: Set confidence threshold; The predicted probability value corresponding to the highest category among the predicted probability values ​​of each category of the particles to be tested is compared with the confidence threshold. The case where the predicted probability value is less than the confidence threshold is determined as a background case, and the case where the predicted probability value is greater than or equal to the confidence threshold is determined as a particle case. All background cases are eliminated and all particle cases are retained.

5. A training system for a waveform data processing model, characterized in that: The training system of the waveform data processing model includes: A first data acquisition module is configured to acquire raw waveform data and corresponding particle real label data; the raw waveform data is multi-channel particle waveform data acquired using a TPC detector, the TPC detector comprising multiple channels, the multiple channels synchronously acquiring particle waveform data to form the multi-channel particle waveform data; the particle real label data comprises a real label for characterizing the actual category of the particle; A first preprocessing module is used to sequentially perform downsampling processing and data compression processing on the original waveform data to obtain preprocessed original waveform data; the first preprocessing module includes a downsampling unit and a data compression unit, the downsampling unit is used to adopt an interval resampling method, with the peak position of the waveform as the interval center, select a suitable interval, and perform interval resampling on the original waveform data while retaining basic waveform information to obtain downsampled original waveform data; the data compression unit is used to compress the downsampled original waveform data into a sparse tensor format in the order of the channel numbers of the TPC detector, the first column of the sparse tensor is the channel number of the triggered channel, and the second column is the corresponding waveform data, and the sparse tensor stores the dimension size and data type of the data to obtain the preprocessed original waveform data; a network loss determination module, configured to input the preprocessed raw waveform data into each channel of a waveform data processing model according to the spatial dimension of the channel, and determine the network loss based on the predicted probability values ​​of each particle category and the true labels of the particles output by the waveform data processing model; the waveform data processing model is a model with a multi-channel network structure established based on an artificial neural network, wherein the network structure of each channel includes multiple convolutional layers and multiple pooling layers, and the waveform data processing model is constructed by first passing through all convolutional layers and then passing through all pooling layers; A training module is used to update the network parameters of the waveform data processing model through back propagation according to the network loss, so as to iterate the waveform data processing model multiple times to obtain a trained waveform data processing model.

6. The waveform data processing model training system according to claim 5, characterized in that: The particle classes include alpha particles, beta particles, and muons.

7. An application system of a waveform data processing model, characterized in that: The application system of the waveform data processing model includes: A second data acquisition module is used to acquire target waveform data, wherein the target waveform data is multi-channel particle waveform data of the particles to be measured acquired by using a TPC detector; A second preprocessing module is used to perform downsampling processing and data compression processing on the target waveform data in sequence to obtain preprocessed target waveform data; A target particle category prediction module is used to input the preprocessed target waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel in which it is located, so as to obtain the predicted probability value of each category of the particle to be measured; the waveform data processing model is a model trained according to the waveform data processing model training method according to claim 1.

8. The application system of the waveform data processing model according to claim 7, characterized in that: The application system of the waveform data processing model also includes: A confidence threshold setting module, used to set a confidence threshold; The screening module is used to compare the predicted probability value corresponding to the highest category among the predicted probability values ​​of each category of the particles to be tested with the confidence threshold, determine the case where the predicted probability value is less than the confidence threshold as a background case, and determine the case where the predicted probability value is greater than or equal to the confidence threshold as a particle case, and eliminate all background cases and retain all particle cases.

Images