Method and system for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurements

Abstract

The application provides a method and system for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement, wherein the method comprises: acquiring spatial distribution information of potassium feldspar single particles in a sample to be measured, quartz photoluminescence data and single particle infrared luminescence signals of potassium feldspar; scanning the sample to be measured and positioning the positions of multiple high radiation response particles according to the scanning result; calculating the particle distribution density of potassium feldspar according to the spatial distribution information, constructing a mineral spatial relationship graph through a graph convolution network based on the particle distribution density and the positions of the multiple high radiation response particles; using the mineral spatial relationship graph to aggregate the radiation measurement signals of adjacent particles, synchronously training the radiation measurement signals and the quartz photoluminescence data; fusing the training result and the single particle infrared luminescence signals, and establishing a heterogeneous adaptive standard growth curve according to the fusion result. The application improves the accuracy and reliability of the dating of incomplete sun-bleached samples.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radiation measurement, and in particular to a method and system for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement. BACKGROUND

[0002] In Quaternary sediment dating research, the potassium feldspar luminescence dating technique needs to solve the problems of incomplete bleaching sample signal weakening and mineral heterogeneity interference. In particular, in complex sedimentary environments such as fluvial facies and glacial facies, it is difficult for traditional methods to accurately establish a standard growth curve reflecting the true burial age, and there is an urgent need to develop a new dating method that can simultaneously process signal attenuation and mineral heterogeneity.

[0003] The current more advanced scheme adopts regional segmentation signal compensation technology. This method divides the sample into several measurement regions, obtains the signal intensity distribution of each region through high-resolution radiation scanning, then calculates the overall radiation signal using a regional weighted average algorithm, and combines a quartz standard growth curve for signal compensation correction.

[0004] This scheme uses fixed grid segmentation in regional division, which cannot adaptively identify the natural distribution characteristics of mineral particles, resulting in the introduction of artificial errors during signal compensation. The regional weighted algorithm is insufficient in extracting the signal characteristics of high radiation response regions, affecting the accuracy of the key interval of the growth curve. The signal compensation process lacks consideration of the spatial distribution of mineral heterogeneity, reducing the reliability of the dating results. SUMMARY

[0005] The present application provides a method and system for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement, to solve the problems of low precision and poor reliability in incomplete bleaching sample dating in the prior art.

[0006] In a first aspect, the present application provides a method for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement, comprising:

[0007] Obtaining the spatial distribution information of potassium feldspar single particles in the sample to be measured, quartz optically stimulated luminescence data, and single particle infrared luminescence signals of potassium feldspar;

[0008] Scanning the sample to be measured and positioning the positions of multiple high radiation response particles based on the scanning results;

[0009] Calculating the particle distribution density of the potassium feldspar based on the spatial distribution information, and constructing a mineral spatial relationship graph based on the particle distribution density and the positions of the multiple high radiation response particles through a graph convolution network;

[0010] Using the mineral spatial relationship graph to aggregate the radiation measurement signals of adjacent particles, and synchronously training the radiation measurement signals with the quartz optically stimulated luminescence data;

[0011] Fusing the training result with the single-particle infrared luminescence signal, and establishing a heterogeneous adaptive standard growth curve according to a fusion result.

[0012] Optionally, the particle distribution density of the potash feldspar is calculated according to the spatial distribution information, and a mineral spatial relationship graph is constructed by a graph convolution network based on the particle distribution density and the positions of the multiple high-radiation-response particles, including:

[0013] Particle size data of the particles in the potash feldspar and spacing data of the particles and adjacent particles are extracted from the spatial distribution information;

[0014] Particle distribution densities of the potash feldspar in different regions are calculated according to the spacing data and the particle size data;

[0015] Each particle is taken as a node, and node attributes include a radiation intensity value and a particle distribution density in a region to which the node belongs;

[0016] A mineral spatial relationship graph is generated based on the positions of the multiple high-radiation-response particles and the node attributes.

[0017] Optionally, the mineral spatial relationship graph is generated based on the positions of the multiple high-radiation-response particles and the node attributes, including:

[0018] Each node corresponding to each of the high-radiation-response particle positions is marked as a center node;

[0019] Spatial distances between each of the center nodes and all the remaining nodes are calculated;

[0020] Adjacent nodes having a spatial distance less than a preset distance threshold are searched with the center nodes as a reference;

[0021] Second radiation correlation coefficients of the center nodes and the corresponding adjacent nodes are calculated;

[0022] Key adjacent nodes having a second radiation correlation coefficient greater than a preset correlation threshold are screened from all the adjacent nodes, and a bidirectional connection relationship between the center nodes and the key adjacent nodes is established;

[0023] Node attributes of all the nodes and the bidirectional connection relationship are integrated to generate a mineral spatial relationship graph.

[0024] Optionally, the second radiation correlation coefficients of the center nodes and the corresponding adjacent nodes are calculated, including:

[0025] Radiation intensity values of the nodes at different time sampling points are spliced into radiation intensity time sequence data of the corresponding nodes;

[0026] segmenting the radiation intensity time series data according to a preset time window to obtain a plurality of groups of radiation intensity sub-sequences corresponding to the nodes;

[0027] calculating intensity change covariance of the center node and the adjacent nodes based on each group of radiation intensity sub-sequences corresponding to the center node and the adjacent nodes respectively, and determining a first radiation correlation coefficient of a time window based on a ratio of the intensity change covariance to a preset variance reference value;

[0028] calculating an average of the first radiation correlation coefficients of all time windows to obtain a second radiation correlation coefficient.

[0029] Optionally, the scanning the sample to be tested includes:

[0030] performing radiation intensity scanning on the sample to be tested to record a radiation intensity value of each potassium feldspar single particle;

[0031] selecting, from all particles, key particles with a radiation intensity value higher than a preset intensity value threshold according to a preset radiation intensity interval;

[0032] locating position coordinates of the key particles, and taking the position coordinates as the position of the high radiation response particle.

[0033] Optionally, the using the mineral spatial relationship diagram to aggregate radiation measurement signals of adjacent particles and synchronously training the radiation measurement signals with the quartz optically stimulated luminescence data includes:

[0034] sequentially taking each node in the mineral spatial relationship diagram as a target node;

[0035] extracting a set of adjacent nodes corresponding to the target node from the mineral spatial relationship diagram;

[0036] weighting and calculating a radiation intensity mean value of the set of adjacent nodes, and converting the radiation intensity mean value into an aggregated radiation measurement signal;

[0037] synchronously inputting the aggregated radiation measurement signal and the quartz optically stimulated luminescence data into a training model according to a predetermined time period, and updating weight parameters in the training model through iteration to make the aggregated radiation measurement signal and the quartz optically stimulated luminescence data reach a preset matching degree.

[0038] Optionally, the fusing the training result with the single-particle infrared stimulated luminescence signal includes:

[0039] extracting a weight parameter matrix after synchronous training from the training result;

[0040] Fuse the weight parameter matrix with the single-particle infrared stimulated luminescence signal to obtain a fusion result;

[0041] According to the fusion result, a corrected mapping relationship between radiation intensity and time is established;

[0042] Based on the corrected mapping relationship, a heterogeneous adaptive standard growth curve is generated.

[0043] In a second aspect, the present application provides a potassium feldspar optically stimulated luminescence dating standard growth curve establishment system based on radiation measurement, comprising:

[0044] An acquisition module is configured to acquire spatial distribution information of potassium feldspar single particles in a sample to be measured, quartz photoluminescence data, and single-particle infrared stimulated luminescence signals of potassium feldspar.

[0045] A scanning module is configured to scan the sample to be measured and locate positions of multiple high-radiation-response particles according to a scanning result.

[0046] A calculation module is configured to calculate a particle distribution density of the potassium feldspar according to the spatial distribution information, and construct a mineral spatial relationship graph through a graph convolution network based on the particle distribution density and the positions of the multiple high-radiation-response particles.

[0047] An aggregation module is configured to aggregate radiation measurement signals of adjacent particles by using the mineral spatial relationship graph, and synchronously train the radiation measurement signals and the quartz photoluminescence data.

[0048] A fusion module is configured to fuse a training result with the single-particle infrared stimulated luminescence signals, and establish a heterogeneous adaptive standard growth curve according to a fusion result.

[0049] In a third aspect, the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the potassium feldspar optically stimulated luminescence dating standard growth curve establishment method according to any one of the first aspect.

[0050] In a fourth aspect, the present application provides a computer storage medium storing computer program instructions, wherein the computer program instructions are executed by a processor to implement the potassium feldspar optically stimulated luminescence dating standard growth curve establishment method according to any one of the first aspect.

[0051] In the application, a method for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement is provided, the method comprising: acquiring spatial distribution information of potassium feldspar single particles in a sample to be measured, quartz photoluminescence data, and single particle infrared luminescence signals of potassium feldspar; scanning the sample to be measured, and positioning positions of multiple high radiation response particles according to a scanning result; calculating particle distribution density of the potassium feldspar according to the spatial distribution information, and constructing a mineral spatial relationship graph through a graph convolution network based on the particle distribution density and the positions of the multiple high radiation response particles; using the mineral spatial relationship graph to aggregate radiation measurement signals of adjacent particles, and synchronously training the radiation measurement signals and the quartz photoluminescence data; fusing a training result and the single particle infrared luminescence signals, and establishing a heterogeneous adaptive standard growth curve according to a fusion result.

[0052] The technical scheme provided by the application has the following beneficial effects:

[0053] The application realizes synchronous acquisition of multi-source data, provides comprehensive basic data support for subsequent analysis, accurately identifies key signal sources in the sample, provides positioning basis for signal enhancement, establishes a topological structure reflecting mineral spatial heterogeneity, provides spatial correlation basis for signal aggregation, realizes effective fusion of multi-source radiation signals, and improves signal quality. The application generates a high-precision growth curve that adapts to the heterogeneous characteristics of the sample.

[0054] Further, the application calculates the regional particle distribution density by extracting particle size and spacing data of potassium feldspar particles, takes each particle as a node containing radiation intensity and distribution density attributes, and constructs a mineral spatial relationship graph based on the positions of the high radiation response particles and the node attributes.

[0055] Moreover, the step establishes a topological network reflecting mineral spatial heterogeneity and radiation characteristics, provides an accurate spatial correlation model for subsequent signal aggregation by quantifying the relationship between particle distribution characteristics and radiation intensity, and effectively solves the signal deviation problem caused by ignoring mineral spatial distribution characteristics in traditional methods.

[0056] These aspects or other aspects of the application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can also be obtained by those skilled in the art without any creative effort.

[0058] Figure 1A flow chart of a method for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement is provided for the embodiments of the present application.

[0059] Figure 2 A structural schematic diagram of a system for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement is provided for the embodiments of the present application.

[0060] Figure 3 A structural schematic diagram of a computing device is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0061] In order to enable personnel in the technical field to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present application.

[0062] In some of the processes described in the specification and the claims of the present application and in the above-described accompanying drawings, a plurality of operations appear in a specific order, but it should be clearly understood that these operations can be executed or performed in parallel or in a sequence different from that in which they appear in the present text, and the serial numbers of the operations, such as 101, 102, etc., are merely used to distinguish different operations, and the serial numbers themselves do not represent any execution sequence. In addition, these processes can include more or fewer operations, and the operations can be executed or performed in sequence or in parallel. It should be noted that the descriptions of “first”, “second”, etc. in the present text are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do “first” and “second” represent different types.

[0063] In the potassium feldspar luminescence dating technology, the existing method adopts a signal processing method of fixed region division, which has obvious deficiencies: this method mechanically divides the sample into several measurement regions, and cannot accurately identify the natural distribution characteristics of mineral particles, resulting in the introduction of artificial errors in signal analysis. At the same time, this method is not sufficient for signal feature extraction in high radiation regions, and does not consider the spatial correlation between mineral particles, so that the accuracy of the growth curve established in the key time interval is reduced, affecting the reliability of the dating results.

[0064] To solve the above problems, the application provides a method for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement. The method first accurately obtains the spatial position and radiation characteristics of each mineral particle, then analyzes the distribution law and radiation correlation between the particles through an intelligent algorithm, and constructs a mineral spatial relationship diagram reflecting the real characteristics of the sample. On this basis, the quartz optically stimulated luminescence data and the potassium feldspar infrared signal are integrated, and finally a standard growth curve suitable for the characteristics of the sample is generated. This innovative scheme breaks through the limitations of traditional fixed area analysis, accurately captures the spatial distribution characteristics and radiation correlation of mineral particles, effectively solves the problems of signal analysis deviation and insufficient accuracy of key intervals, and improves the accuracy and reliability of dating results in complex sedimentary environments.

[0065] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0066] Figure 1 A flowchart of a method for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurement provided by the embodiments of the application is shown in Figure 1 The method comprises the following steps.

[0067] Step 101: Obtain the spatial distribution information of single particles of potassium feldspar in the sample to be measured, quartz optically stimulated luminescence data, and single particle infrared luminescence signal of potassium feldspar.

[0068] In step 101, the sample to be measured refers to a mixed sediment sample containing single particles of potassium feldspar, wherein the single particles of potassium feldspar are the main research object, but the sample may also contain other mineral particles. The spatial distribution information represents a data set containing the position coordinates and particle size of the potassium feldspar particles in the sample. The position coordinates are obtained by a microscopic imaging system, and the particle size is obtained by laser particle size analysis. The quartz optically stimulated luminescence data represent the measurement data of the change of the radiation signal intensity of the quartz mineral in the same layer as the potassium feldspar sample with time under light excitation. The single particle infrared luminescence signal represents the radiation signal intensity value of a single potassium feldspar particle under infrared light excitation.

[0069] In the embodiments of the present application, first, the sample is scanned using a microscopic imaging system to record the two-dimensional or three-dimensional position coordinates of each potassium feldspar particle. At the same time, a laser particle size analyzer is used to measure the particle size of each particle to form a spatial distribution information dataset. Under the same experimental conditions, a photoluminescence measurement instrument is used to obtain the photoluminescence data of the quartz mineral in the same sample. Finally, the infrared excitation device is used to measure the infrared luminescence signal intensity of the potassium feldspar particles one by one. These data will be input as the basis for subsequent analysis.

[0070] For example, after processing a river sediment sample, the position coordinates of 50 potassium feldspar particles are obtained by scanning using a microscopic imaging system, such as particle A coordinates 1.2mm 3.5mm and particle B coordinates 2.8mm 3.1mm. Laser particle size analysis shows that the particle size of particle A is 80μm and the particle size of particle B is 65μm. Simultaneous measurement of the same layer quartz sample obtains 10 time points of photoluminescence data, which are 1520, 1540, 1530 photons / second, etc. Using an infrared laser, the infrared luminescence signal of particle A is measured to be 1850 photons / second and the infrared luminescence signal of particle B is measured to be 2100 photons / second.

[0071] Step 102: Scanning the sample to be measured, and positioning the positions of a plurality of high radiation response particles according to the scanning results.

[0072] In step 102, the high radiation response particle refers to a particle that shows a radiation signal intensity higher than a preset threshold when detected by a radiation measurement instrument. A specific example is that under the same measurement conditions, the radiation count rate of some potassium feldspar particles is significantly higher than that of other particles (e.g., more than 2 standard deviations higher than the average value).

[0073] In the embodiments of the present application, a radiation scanner is used to measure each potassium feldspar particle point by point to record the radiation intensity value of each potassium feldspar particle. Statistical analysis is performed on all measurement data to determine the distribution characteristics of the radiation intensity. According to the preset intensity threshold standard, particles with radiation intensity significantly higher than the average level are screened out. Finally, the specific position coordinates of these high radiation particles in the sample are recorded.

[0074] For example, radiation scanning is performed on the aforementioned 50 potassium feldspar particles, and the average radiation intensity is measured to be 1550 photons / second with a standard deviation of 250. The threshold is set to be the average value plus twice the standard deviation, i.e., 2050 photons / second. It is found that the radiation intensity of particle B is 2100 photons / second, which exceeds the threshold, and it is determined to be the central node with position coordinates of 2.8mm 3.1mm.

[0075] Step 103: Calculating the particle distribution density of the potassium feldspar according to the spatial distribution information, and constructing a mineral spatial relationship graph through a graph convolution network based on the particle distribution density and the positions of the plurality of high radiation response particles.

[0076] In step 103, the particle distribution density represents the number of potassium feldspar particles per unit area or volume. The mineral spatial relationship diagram represents the spatial distribution relationship of potassium feldspar particles in a graph structure, where nodes represent particles and edges represent spatial associations. The mineral mainly refers to single potassium feldspar particles, but also includes other mineral particles in the sample that have spatial associations with potassium feldspar. These minerals together constitute the mineral composition of the sample and affect the measurement of the radiation signal.

[0077] In the embodiments of the present application, the spacing and particle size data of each particle are extracted from the spatial distribution information. The ratio of the number of particles in a certain range around each particle to the area of the region is calculated to obtain the particle distribution density. Each particle is converted into a graph node, and the node attributes include its radiation intensity and distribution density. The high-radiation particles are taken as the center nodes, and the connection relationship between the nodes is established according to the spatial distance and radiation correlation, and finally the mineral spatial relationship diagram is formed.

[0078] For example, taking particle B as the center, the particle distribution density within a 2mm radius range around it is calculated to be 8 particles / 0.12mm². The node attributes are established: particle B density 66.7 particles / mm², radiation intensity 2100 photons / sec. According to the 1.6mm spacing and the 0.54 radiation correlation coefficient, the connection relationship between particle B and particle A is established.

[0079] Step 104: using the mineral spatial relationship diagram, aggregating the radiation measurement signals of adjacent particles, and synchronously training the radiation measurement signals with the quartz photoluminescence data.

[0080] In step 104, the adjacency of adjacent particles is determined based on the spatial distance and radiation signal correlation between particles, and is specifically determined by the adjacency matrix set in the graph convolution network. For example, particles with a spacing less than 3 times the average particle size and a radiation signal correlation coefficient greater than 0.7 are defined as adjacent particles. The radiation measurement signal represents a composite signal obtained by integrating the radiation intensities of adjacent particles through a weighted average method.

[0081] In the embodiments of the present application, each node in the mineral spatial relationship diagram is traversed as a target node. The adjacent node set is found, and the weight coefficient is calculated according to the distance and radiation correlation between nodes. The average radiation intensity of adjacent nodes is calculated as the aggregation signal using the weighted average formula. After aligning the signal with the quartz photoluminescence data in time sequence, the training model is input, and the weight parameters are adjusted iteratively to match the two.

[0082] For example, take particle B as the center node, its adjacent nodes particle A weight 0.39, particle C weight 0.35. The calculated aggregate radiation signal is (0.39*1850+0.35*1750) / (0.39+0.35)=1802 photons / second. This signal is input into the model with the quartz data 1520 photons / second, and after 100 iterations, the matching error is reduced to an acceptable range.

[0083] Step 105: fuse the training results with the single-particle infrared luminescence signal, and establish a heterogeneous adaptive standard growth curve according to the fusion results.

[0084] In step 105, the heterogeneous adaptive standard growth curve represents a radiation intensity-time relationship curve that can automatically adapt to the heterogeneity characteristics of the sample.

[0085] In the embodiments of the present application, the weight parameter matrix is extracted from the trained model. The matrix is weighted and superimposed with the single-particle infrared luminescence signal. The corresponding relationship between radiation intensity and time is adjusted according to the superimposed results. The adjusted data points are used to fit and generate the final standard growth curve.

[0086] For example, the training weight [0.85, 0.12, 0.03] is extracted, and the superimposed calculation with the particle B infrared signal 1920 photons / second is 1906 photons / second. At the 8ka time point, the original signal 1802 photons / second is corrected to 1906 photons / second. The final generated standard growth curve is 1906 photons / second at 8ka and 2116 photons / second at 10ka.

[0087] This method accurately obtains the spatial distribution and radiation characteristics of potassium feldspar particles, constructs a spatial relationship graph reflecting the mineral heterogeneity, effectively integrates multi-source radiation signals, and establishes a growth curve that can accurately reflect the true characteristics of the sample, improving the accuracy of luminescence dating in complex sedimentary environments. The measurement deviation problem caused by ignoring the spatial correlation and signal heterogeneity in the traditional method is solved.

[0088] In order to solve the signal deviation problem caused by mineral heterogeneity in potassium feldspar luminescence dating, in some embodiments, step 103: based on the particle distribution density and the positions of the multiple high-radiation-response particles, a mineral spatial relationship graph is constructed through a graph convolution network according to the spatial distribution information, including:

[0089] Step 201: extract the particle size data of the particles in the potassium feldspar and the distance data between the particles and the adjacent particles from the spatial distribution information.

[0090] In step 201, the particle size data refers to the diameter size measurement of each potassium feldspar particle, which is obtained by laser particle size analysis. The spacing data refers to the straight-line distance between the center points of adjacent particles, which is calculated by particle coordinates. These two kinds of data together reflect the spatial distribution characteristics of the sample.

[0091] In the embodiments of the present application, the particle size measurement of each particle is first extracted from the obtained spatial distribution information, and then the center distance between each two adjacent particles is calculated according to the coordinate position of the particles. These basic data provide input for subsequent density calculation.

[0092] Step 202: Calculate the particle distribution density of potassium feldspar in different regions according to the spacing data and the particle size data.

[0093] In the embodiments of the present application, a circular region is drawn with each particle as the center, and the region radius is several times the particle size of the particle. The total number of particles in the region is counted, and then the particle distribution density value at the position is obtained by dividing the region area. The density values of all particles constitute a density distribution map. The specific implementation process is as follows: taking the position of each high radiation response particle as the center, taking 3 times the particle size of the particle as the radius to draw a circular region, counting the number of all particles in the region, and then dividing the region area to obtain the particle distribution density; a specific example is that in a river sediment sample, the particle size of the high radiation particle B is 65 μm, the center position 2.8 mm 3.1 mm is taken as the center of the circle, and the radius of 195 μm is taken as the radius to draw the region, a total of 8 particles are counted in the region, the area of the region is π×0.195²=0.119 mm², and then the particle distribution density of the region is 8 / 0.119=67.2 particles / mm², which will be used as a node attribute to construct a mineral spatial relationship graph.

[0094] Step 203: Each particle is taken as a node, and the node attribute includes the radiation intensity value and the particle distribution density in the region to which it belongs.

[0095] In step 203, the node is the basic unit in the graph structure, and each node represents a potassium feldspar particle. The node attribute includes the radiation intensity measurement value of the particle and the particle distribution density value in the region.

[0096] In the embodiments of the present application, each potassium feldspar particle is converted into a graph node, and the node attribute includes two key parameters: one is the radiation intensity value of the particle, which comes from the radiation scanning data; the other is the density value of the region where the particle is located, which comes from the calculation result of the previous step.

[0097] Step 204: Generate a mineral spatial relationship graph based on the positions of the plurality of high radiation response particles and the node attributes.

[0098] In the embodiments of the present application, first, the high-radiation particles are marked as central nodes, and then the adjacent nodes are screened and connected according to the spatial distance and radiation correlation. In the finally generated graph structure, the nodes contain attribute information, and the edges represent the spatial radiation correlation.

[0099] The following is a specific example:

[0100] In the analysis of the river sediment sample, for the identified high-radiation central particle B (coordinates 2.8mm 3.1mm, radiation intensity 2100 photons / sec), first, the distance data between it and the adjacent particle A (coordinates 1.2mm 3.5mm) is extracted, and a 1.6mm distance value is calculated by a two-dimensional distance formula sqrt(2.8-1.2)²+(3.1-3.5)², wherein the particle A has a particle size of 80μm, and the particle B has a particle size of 65μm. The analysis area is defined with the particle B as the center, and the distance of 3 times the particle size, i.e. 0.195mm, is taken as the radius. The area of this circular region is π×0.195²=0.119mm², and a total of 8 particles are contained in this region according to the statistics, and the particle distribution density is calculated as 8 / 0.119=67.2 particles / mm². The particle B is converted into a graph node, and the node attribute setting contains the radiation intensity value 2100 photons / sec and the density value 67.2 particles / mm². According to the 1.6mm distance between the particle B and the particle A and the 0.54 radiation correlation coefficient (the calculation formula is R=Cov(X,Y) / (σ_X σ_Y), wherein Cov represents the covariance, and σ represents the standard deviation), the connection relationship between the two is established. In constructing the complete mineral spatial relationship graph, the particle C (coordinates 4.5mm 2.9mm, radiation intensity 1750 photons / sec) is also included as a node, the distance between the particle C and the particle B is calculated as 1.7mm, and the radiation correlation coefficient between the two is calculated as 0.58 by the same method. The finally generated relationship graph contains three main nodes and two connection edges, which completely characterizes the spatial distribution characteristics and radiation correlation of the potassium feldspar particles in this region, and provides an accurate topological structure basis for subsequent signal aggregation.

[0101] In the embodiments of the present application, by constructing such a mineral spatial relationship graph, the spatial distribution characteristics and radiation correlation of the potassium feldspar particles can be accurately quantified, which provides an accurate topological structure basis for subsequent signal aggregation, and effectively solves the signal analysis deviation problem caused by ignoring spatial heterogeneity in the traditional method.

[0102] In order to further improve the construction accuracy of the mineral spatial relationship graph, in some embodiments, step 204: generating a mineral spatial relationship graph based on the positions of the plurality of high-radiation response particles and the node attributes, comprises:

[0103] Step 301: marking each node corresponding to the position of the high-radiation response particle as a central node.

[0104] In step 301, the center node refers to a key node with a dominant role in the mineral spatial relationship diagram, corresponding to the position of a high radiation response particle in the sample with radiation intensity higher than the average level. These nodes will serve as the core connection hub of the graph structure.

[0105] In the embodiment of the present application, first, from the identified high radiation response particles, particles with radiation intensity exceeding a certain threshold are selected, and the corresponding graph node is marked as a center node, providing a reference point for subsequent connection relationship construction.

[0106] Step 302: Calculate the spatial distance between each center node and all other nodes.

[0107] In step 302, the spatial distance refers to the actual physical distance between the center node and the particles represented by other nodes in the graph, which is calculated through particle coordinate data. This distance reflects the spatial proximity between particles.

[0108] In the embodiment of the present application, based on the position coordinate data of all particles, the two-dimensional plane distance between each center node and every other node is calculated, and a complete distance matrix is established, providing a basis for adjacent node search.

[0109] Step 303: Search for adjacent nodes with a spatial distance less than a preset distance threshold based on the center node.

[0110] In step 303, the adjacent node refers to a candidate connection node that is close enough to the center node in terms of spatial distance. The preset distance threshold is determined based on the multiple relationship of the average particle size.

[0111] In the embodiment of the present application, taking each center node as the center and the set distance threshold as the radius, all other nodes falling within the circular region are searched to form a preliminary set of adjacent node candidates.

[0112] Step 304: Calculate the second radiation correlation coefficient of the center node and the corresponding adjacent node.

[0113] In step 304, the second radiation correlation coefficient is an index that quantifies the degree of synchronous change of radiation signals between the center node and the adjacent node, obtained by statistical analysis of the radiation intensity time series data of both.

[0114] In the embodiment of the present application, the radiation intensity measurement sequence of the center node and the adjacent node in the same time period is selected, and the ratio of the covariance of the signal fluctuation of both to their respective standard deviations is calculated to obtain the correlation coefficient.

[0115] Step 305: From all adjacent nodes, filter out key adjacent nodes with a second radiation correlation coefficient greater than a preset correlation threshold, and establish a bidirectional connection relationship between the center node and the key adjacent nodes.

[0116] In step 305, the key adjacent node refers to a high-quality connection node that meets both the spatial distance requirement and has a high enough radiation correlation. The bidirectional connection relationship represents the mutual influence between nodes.

[0117] In the embodiments of the present application, for each adjacent node, when its radiation correlation with the center node exceeds the set threshold, a bidirectional connection edge is established, indicating that the two particles have a correlation in radiation characteristics.

[0118] Step 306: Integrate the node attributes of all nodes and the bidirectional connection relationship to generate a mineral spatial relationship graph.

[0119] In the embodiments of the present application, the attribute data of each node, including radiation intensity and particle density, and the connection relationship data between nodes, are integrated into a unified graph data structure, completing the construction of the mineral spatial relationship graph.

[0120] The following is a specific example:

[0121] Based on the analysis results of the river sediment samples, particle B has been determined as a high-radiation center node with coordinates 2.8mm 3.1mm, radiation intensity 2100 photons / sec, and node attributes including density value 67.2 particles / mm². First, particle B is labeled as the center node, and the distance between particle B and particle A with coordinates 1.2mm 3.5mm is sqrt(2.8-1.2)²+(3.1-3.5)²=1.6mm, and the distance between particle B and particle C with coordinates 4.5mm 2.9mm is 1.7mm. Set the distance threshold to 2mm, both particles A and C meet the adjacent condition. Then calculate the second radiation correlation coefficient R=Cov(X,Y) / (σ_X σ_Y) between the center node B and the adjacent node A, where X and Y represent the radiation intensity time series data of B and A respectively, Cov represents covariance, and σ represents standard deviation. By analyzing the radiation intensity sequence of 10 time points, the correlation coefficient of B and A is calculated as 0.54, and the correlation coefficient of B and C is calculated as 0.58. Set the correlation threshold to 0.55, only particle C meets the condition. Therefore, a bidirectional connection relationship between B and C is established in the mineral spatial relationship graph, and B and A do not establish a connection. The finally generated mineral spatial relationship graph contains three nodes: B with attributes 2100 photons / sec and 67.2 particles / mm², A with attributes 1850 photons / sec and 65.4 particles / mm², and C with attributes 1750 photons / sec and 62.8 particles / mm², and a connection edge B-C.

[0122] In the embodiments of the present application, through this refined graph structure construction method, the spatial correlation and radiation characteristic relationship between potassium feldspar particles can be more accurately captured, providing a more reliable topological structure basis for subsequent signal processing and growth curve establishment, effectively improving the accuracy of dating results.

[0123] To further improve the accuracy of the radiation correlation calculation, in some embodiments, step 304: the second radiation correlation coefficient of the center node and the corresponding adjacent node is calculated, comprising:

[0124] Step 401: splice the radiation intensity values of the nodes at different time sampling points into radiation intensity time sequence data of the corresponding nodes.

[0125] In step 401, the radiation intensity time sequence data refers to a data sequence formed by arranging the radiation intensity values collected at different time points of each node in chronological order. This data reflects the characteristics of the particle radiation intensity change over time.

[0126] In the embodiments of the present application, the radiation intensity values measured at multiple sampling time points of each node are connected in chronological order to form a complete time sequence data sequence, providing basic data for subsequent analysis.

[0127] Step 402: segment the radiation intensity time sequence data according to a preset time window to obtain multiple groups of radiation intensity subsequences of the corresponding nodes.

[0128] In step 402, the preset time window determines the number of data points included in each subsequence. The radiation intensity subsequence is a local data segment obtained by segmenting the complete time sequence data by a fixed time length.

[0129] In the embodiments of the present application, the complete radiation intensity time sequence data is segmented according to a set time window length, and each subsequence contains the same number of continuous time point data, ensuring the comparability of the data segments.

[0130] Step 403: based on each group of radiation intensity subsequences corresponding to the center node and the adjacent node respectively, calculate the intensity change covariance of the center node and the adjacent node, and based on the ratio of the intensity change covariance to a preset variance reference value, determine a first radiation correlation coefficient of a time window.

[0131] In step 403, the intensity change covariance is an index for measuring the synchronization of the radiation intensity fluctuations of two nodes. The preset variance reference value is a reference value for standardization processing, usually taking the geometric mean of the variances of the two nodes. The first radiation correlation coefficient is a quantitative index of the radiation intensity change synchronization of the center node and the adjacent node within a single time window, specifically reflecting the matching degree of the radiation signal fluctuations of the two nodes within the time period. Its calculation method is to divide the covariance value of the radiation intensity subsequences of the two nodes by the geometric mean of their variances. The coefficient takes a value between 0 and 1, and the larger the value, the more consistent the radiation change trend of the two nodes within the time window.

[0132] In the embodiment of the present application, for each group of time window subsequence data, the covariance value of the radiation intensity change of the center node and the adjacent node is calculated, and then divided by the geometric mean of the variance of the two nodes to obtain the first radiation correlation coefficient of the time window.

[0133] Step 404: Calculate the average of the first radiation correlation coefficients of all time windows to obtain the second radiation correlation coefficient.

[0134] In the embodiment of the present application, the first radiation correlation coefficients of all time windows are added and averaged to obtain the final second radiation correlation coefficient, which is used to determine whether to establish a connection relationship between nodes.

[0135] The following is a specific example:

[0136] Based on the radiation intensity time series data of particles B and C in the foregoing river sediment sample analysis, the radiation intensity values of particle B at 10 consecutive time points are 2100, 2095, 2112, 2088, 2120, 2090, 2105, 2085, 2118, and 2092 photons per second, and the corresponding time point data of particle C are 1750, 1745, 1760, 1738, 1772, 1735, 1755, 1730, 1765, and 1742 photons per second. The 10 time point data are divided into two time windows according to every 5 points as a group, the first window takes the first to fifth data points, the average value of the subsequence of particle B is 2103 photons per second, and the average value of the subsequence of particle C is 1753 photons per second. The covariance calculation formula is Cov=1 / nΣ(B_i-B_avg)(C_i-C_avg), where n=5, B_i and C_i represent the radiation intensity values of the two particles at the i-th time point, B_avg and C_avg represent the average value of the subsequence, and the covariance value is calculated as 142.6. The variance of the subsequence of particle B is 180.8, and the variance of the subsequence of particle C is 156.4. The variance reference value is √(180.8×156.4)=168.3, and the first radiation correlation coefficient is 142.6 / 168.3=0.847. The second window takes the sixth to tenth data points, and the covariance is calculated as 135.2, the variance reference value is 165.1, and the first radiation correlation coefficient is 0.819. The final second radiation correlation coefficient is (0.847+0.819) / 2=0.833.

[0137] In the embodiment of the present application, by using this method of periodical calculation and averaging, the radiation correlation between nodes can be more comprehensively and accurately evaluated, the misjudgment caused by single period data fluctuation can be avoided, and the accuracy and reliability of the mineral spatial relationship graph construction are improved.

[0138] To more accurately identify high radiation response particles in the sample, in some embodiments, step 102: scanning the sample to be tested, locating the position of the high radiation response particles according to the scanning results, includes:

[0139] Step 501: Perform radiation intensity scanning on the sample to be tested, and record the radiation intensity value of each potassium feldspar single particle.

[0140] In step 501, the radiation intensity scanning refers to the point-by-point detection of the sample using a radiation measuring instrument, and the radiation signal intensity value released by each potassium feldspar particle is recorded. These values reflect the radioactive characteristics of each particle.

[0141] In the embodiments of the present application, the surface of the sample is measured systematically using a radiation scanning device to ensure that all potassium feldspar particles are covered, and the radiation intensity readings at each detection point are recorded to form a complete radiation intensity distribution data set.

[0142] Step 502: According to the preset radiation intensity interval, filter out the key particles whose radiation intensity value is higher than the preset intensity value threshold from all particles.

[0143] In step 502, the preset radiation intensity interval is a filtering range set according to the overall radiation characteristics of the sample, which is used to distinguish between ordinary particles and high radiation particles. The preset intensity value threshold is usually the overall average radiation intensity plus a certain multiple of the standard deviation. The key particle refers to the central node.

[0144] In the embodiments of the present application, first, the average value and fluctuation range of the radiation intensity of all particles are calculated, then a reasonable intensity threshold is set, and finally the measured value of each particle is compared with the threshold to filter out special particles with radiation intensity higher than the threshold.

[0145] Step 503: Locate the position coordinates of the key particles, and use the position coordinates as the position of the high radiation response particles.

[0146] In step 503, the position coordinates of the key particles refer to the specific spatial positions of the filtered high radiation particles in the sample, which are obtained through the positioning system of the scanning device. These coordinates will serve as important reference points for subsequent analysis.

[0147] In the embodiments of the present application, for each high radiation particle filtered out, the coordinate recording function of the scanning device is called to accurately obtain its position information in the sample coordinate system, providing key node data for constructing a spatial relationship diagram.

[0148] The following is a specific example:

[0149] During the analysis of the river sediment sample, 50 potassium feldspar particles are systematically measured using a radiation scanner, and the radiation intensity value of each particle is recorded. By calculation, the average radiation intensity of all particles is 1550 photons / second, and the standard deviation is 250 photons / second. According to statistical principles, the screening threshold is set to the average value plus twice the standard deviation, i.e. 1550+2×250=2050 photons / second. In the measurement data, the radiation intensity of particle B reaches 2100 photons / second, which is significantly higher than the threshold, while the radiation intensity of particle A is 1850 photons / second and the radiation intensity of particle C is 1750 photons / second, both of which are lower than the threshold. Through the positioning system of the scanner, the position coordinates of particle B in the sample coordinate system are accurately obtained as 2.8mm 3.1mm, which are consistent with the coordinates recorded by the previous microscopic imaging system. At the same time, it is confirmed that the particle size of particle B is 65 microns, which is completely consistent with the previous embodiment. Finally, it is determined that particle B is a high radiation response particle, and its position coordinates will be used as a key node for the construction of the subsequent mineral spatial relationship diagram.

[0150] In the embodiments of the present application, through this systematic scanning and screening method, key particles with radiation characteristics in the sample can be reliably identified, providing accurate spatial positioning reference for subsequent analysis and ensuring the accuracy of the construction of the mineral spatial relationship diagram.

[0151] In order to more effectively integrate multi-source radiation signals, in some embodiments, step 104: using the mineral spatial relationship diagram, aggregating the radiation measurement signals of adjacent particles, and synchronously training the radiation measurement signals with the quartz optically stimulated luminescence data, includes:

[0152] Step 601: sequentially taking each node in the mineral spatial relationship diagram as a target node.

[0153] In step 601, the target node refers to the core potassium feldspar particle node in the current processing process, and each node in the mineral spatial relationship diagram will be processed in turn as the target node.

[0154] In the embodiments of the present application, each node in the mineral spatial relationship diagram is selected in turn as the core node of the current processing according to the node number order or the radiation intensity order, ensuring that all particle data can be fully utilized.

[0155] Step 602: extracting a set of adjacent nodes corresponding to the target node from the mineral spatial relationship diagram.

[0156] In step 602, the set of adjacent nodes refers to all adjacent nodes directly connected to the target node in the mineral spatial relationship diagram, and these nodes represent particles that have a correlation with the target node in terms of spatial position and radiation characteristics.

[0157] In the embodiments of the present application, according to the constructed mineral spatial relationship diagram, all adjacent nodes connected with the current target node are queried, and the complete attribute data of these nodes is obtained to prepare for signal aggregation.

[0158] Step 603: Weighted calculation of the average radiation intensity of the set of adjacent nodes, and conversion of the average radiation intensity into an aggregated radiation measurement signal.

[0159] In step 603, the weighted average radiation intensity is a composite signal value obtained by weighted average calculation of the radiation intensity of the adjacent nodes by considering the spatial distance and radiation correlation between the adjacent nodes and the target node.

[0160] In the embodiments of the present application, first, the weight coefficient of each adjacent node is determined, the distance and radiation correlation between nodes are comprehensively considered in the weight calculation, then the average radiation intensity of all adjacent nodes is calculated using the weighted average formula, and the aggregated radiation signal of the target node is obtained.

[0161] Step 604: According to a predetermined time period, the aggregated radiation measurement signal and the quartz photoluminescence data are synchronously input into a training model, and the weight parameters are iteratively updated in the training model to make the aggregated radiation measurement signal and the quartz photoluminescence data reach a preset matching degree.

[0162] In step 604, the training model refers to a machine learning model for fusing multi-source radiation signals, and a neural network model with adaptive weight adjustment function is specifically adopted, which can automatically optimize the weight distribution relationship between the aggregated radiation signal and the quartz photoluminescence data. Synchronous training refers to a process of aligning the aggregated radiation signal and the quartz photoluminescence data in time sequence, and iteratively adjusting the parameters of the model to make the two signals reach the best matching state. The preset matching degree refers to that the error rate between the aggregated radiation measurement signal and the quartz photoluminescence data is not more than 5%, and the matching degree is calculated according to the numerical difference percentage of the two signals at each time point, and the specific formula is |aggregated signal value-quartz signal value| / quartz signal value x 100%, when the value is less than or equal to 5%, it is determined that the matching requirement is met.

[0163] In the embodiments of the present application, the time-aligned aggregated radiation signal sequence and the quartz photoluminescence data sequence are input into the training model, the internal parameters are automatically adjusted through multiple iterations, the difference between the two signals is gradually reduced, and the preset matching standard is reached.

[0164] The following is a specific example:

[0165] In the analysis of the river sediment sample, based on the constructed mineral spatial relationship diagram, first select particle B as the target node, whose coordinates are 2.8mm 3.1mm, the radiation intensity is 2100 photons / second, and the density is 66.7 particles / mm². The adjacent node set directly connected with particle B is extracted from the relationship diagram, including particle A with coordinates 1.2mm 3.5mm and particle C with coordinates 4.5mm 2.9mm. According to the previous calculation, the distance between particles A and B is 1.6mm, the radiation correlation is 0.54, and the weight coefficient is 0.39; the distance between particles C and B is 1.7mm, the radiation correlation is 0.58, and the weight coefficient is 0.35. The weighted average formula is used to calculate the aggregated radiation signal, wherein the weight coefficient calculation considers the reciprocal square of the distance and the radiation correlation, and specifically, it is (0.39*1850+0.35*1750) / (0.39+0.35)=1802 photons / second. After aligning the aggregated signal with the synchronous collected quartz photoluminescence data 1520 photons / second according to the time sequence, the training model is input, and the model adjusts the weight parameters through the least square method iteration. After 100 iterations, the matching error of the two signals decreases from the initial 18.6% to 2.3%. The final training weight matrix is [0.85, 0.12, 0.03], which respectively corresponds to the contribution weights of particles B, A and C, wherein the weight calculation process uses the gradient descent algorithm optimization to ensure the best matching of the aggregated signal and the quartz data.

[0166] In the embodiments of the present application, through this systematic signal aggregation and synchronous training method, the radiation signal characteristics of potassium feldspar and quartz can be effectively integrated, the key information provided by the mineral spatial relationship diagram is fully utilized, and the accuracy and reliability of the final standard growth curve are improved.

[0167] In order to more accurately establish the standard growth curve reflecting the characteristics of the sample, in some embodiments, step 105: the training result is fused with the single-particle infrared luminescence signal, and a heterogeneous adaptive standard growth curve is established according to the fusion result, including:

[0168] Step 701: extracting the weight parameter matrix after synchronous training from the training result.

[0169] In step 701, the weight parameter matrix refers to the parameter set reflecting the importance of each node obtained through synchronous training, and each value in the matrix corresponds to the contribution weight of a node in the final signal.

[0170] In the embodiments of the present application, the internal parameters are extracted from the signal processing model after synchronous training, which record the relative contribution degree of different potassium feldspar particle nodes to the final radiation signal, forming a weight matrix.

[0171] Step 702: Fuse the weight parameter matrix with the single-particle infrared luminescence signal to obtain a fusion result.

[0172] In step 702, the fusion processing refers to a calculation process of combining the weight parameters with the single-particle infrared luminescence signal according to a specific rule, aiming to integrate radiation characteristic information from different sources.

[0173] In the embodiments of the present application, each parameter in the weight matrix is multiplied by the infrared luminescence signal of the corresponding node and then added to obtain a fusion signal value that comprehensively considers the spatial relationship and radiation characteristics.

[0174] Step 703: Establish a corrected mapping relationship between radiation intensity and time according to the fusion result.

[0175] In step 703, the radiation intensity is derived from the superposition processing result of the weight parameter matrix and the single-particle infrared luminescence signal, wherein the weight parameter matrix implicitly contains the radiation characteristic information extracted from the aggregate radiation measurement signal. The time corresponds to the time dimension information in the quartz optically stimulated luminescence data, which has been included when obtaining the quartz optically stimulated luminescence data, and has established a time correspondence relationship with the radiation measurement signal in the synchronous training process. The corrected mapping relationship refers to the new correspondence relationship between radiation intensity and time after fusion processing, which more accurately reflects the actual characteristics of the sample.

[0176] In the embodiments of the present application, according to the difference between the fusion signal value and the original radiation signal, an adjustment rule of radiation intensity with time is established to generate a more actual intensity-time correspondence table.

[0177] Step 704: Generate a heterogeneous adaptive standard growth curve based on the corrected mapping relationship.

[0178] In the embodiments of the present application, using the corrected radiation intensity-time correspondence data, a smooth standard growth curve is generated by curve fitting method, which fully considers the spatial heterogeneity and radiation characteristics of the sample. The specific implementation process is as follows: taking the radiation intensity sequence in the corrected mapping relationship as the vertical coordinate value and the corresponding time sequence as the horizontal coordinate value, an interpolation method is used to fit a continuous curve, wherein the data points in the high radiation response particle concentration interval are given a weight coefficient for intensive fitting; a specific example is as follows: for a potassium feldspar sample in a river sediment that is not completely bleached, 50 time point corrected radiation intensity values in the 0-10 ka time range are measured, and a 1.5 times weight is given to the 3-5 ka interval (high radiation particle concentration section) for curve fitting, and finally a luminescence growth curve reflecting the heterogeneous characteristics of the sample is generated.

[0179] The following is a specific example:

[0180] In the river sediment sample analysis, based on the weight parameter matrix [0.85, 0.12, 0.03] obtained through previous training, which respectively corresponds to the contribution weight of the central node particle B and its adjacent node particles A and C. Combined with the infrared luminescence signal intensity of each particle, wherein the particle B is 1920 photons / second, the particle A is 1850 photons / second, and the particle C is 1750 photons / second, the final signal value is calculated by using the weighted fusion formula, specifically 0.85*1920+0.12*1850+0.03*1750=1632+222+52.5=1906 photons / second. At the 8ka time point, the original aggregate radiation signal is 1802 photons / second, and the calculation correction coefficient is 1906 / 1802=1.058. The coefficient is applied to the signal correction of all time points, for example, the original signal of 2000 photons / second at the 10ka time point is corrected to 2000*1.058=2116 photons / second. The finally established standard growth curve corresponds to 1906 photons / second at the 8ka time point and 2116 photons / second at the 10ka time point.

[0181] In the embodiments of the present application, through the systematic signal fusion and curve generation method, the multi-source radiation characteristics of the sample can be fully integrated, a standard growth curve more accurately reflecting the actual deposition age can be established, and the problem of insufficient consideration of sample heterogeneity in the traditional method can be effectively solved, and the reliability of the dating result is improved.

[0182] Figure 2 A structure diagram of a system for establishing a standard growth curve for potassium feldspar luminescence dating based on radiation measurement is provided in the embodiments of the present application, as shown in Figure 2 The system comprises:

[0183] The acquisition module 21 is configured to acquire spatial distribution information of single particles of potassium feldspar in a sample to be measured, quartz optically stimulated luminescence data, and single particle infrared luminescence signals of potassium feldspar.

[0184] The scanning module 22 is configured to scan the sample to be measured and locate positions of a plurality of high radiation response particles according to a scanning result.

[0185] The calculation module 23 is configured to calculate a particle distribution density of the potassium feldspar according to the spatial distribution information, and construct a mineral spatial relationship graph by using a graph convolution network based on the particle distribution density and the positions of the plurality of high radiation response particles.

[0186] The aggregation module 24 is configured to aggregate radiation measurement signals of adjacent particles by using the mineral spatial relationship graph, and perform synchronous training on the radiation measurement signals and the quartz optically stimulated luminescence data.

[0187] The fusion module 25 is configured to fuse the training result with the single-particle infrared stimulated luminescence signal, and establish a heterogeneous adaptive standard growth curve according to the fusion result.

[0188] Figure 2 The radiation measurement-based K-feldspar optically stimulated luminescence dating standard growth curve establishment system can perform Figure 1 The radiation measurement-based K-feldspar optically stimulated luminescence dating standard growth curve establishment method has the implementation principle and technical effects as described above, and will not be described here in detail. The specific operation modes of each module and unit of the radiation measurement-based K-feldspar optically stimulated luminescence dating standard growth curve establishment system have been described in detail in the embodiments related to the method, and will not be described here in detail.

[0189] In one possible design, Figure 2 The radiation measurement-based K-feldspar optically stimulated luminescence dating standard growth curve establishment system can be implemented as a computing device, such as a computer. Figure 3 As shown in the figure, the computing device can include a storage component 31 and a processing component 32.

[0190] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0191] The processing component 32 is configured to execute the above Figure 1 The radiation measurement-based K-feldspar optically stimulated luminescence dating standard growth curve establishment method of the embodiments.

[0192] The processing component 32 can include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components, for executing the above method.

[0193] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0194] Of course, the computing device can also include other components, such as an input / output interface, a display component, a communication component, etc.

[0195] The input / output interface provides an interface between the processing component and peripheral interface modules, which can be output devices, input devices, etc.

[0196] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0197] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform, and the processing component, the storage component, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0198] The embodiment of the present application also provides a computer storage medium, which stores a computer program, and the computer program can implement the above-mentioned Figure 1 A method for establishing a potassium feldspar dosimetry standard growth curve based on radiation measurement.

[0199] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be described here.

[0200] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0201] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0202] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for establishing a potassium feldspar dosimetric standard growth curve based on radiation measurement, characterized in that, The method comprises the following steps: Obtaining the spatial distribution information of potassium feldspar single particles in the sample to be tested, the quartz optical photoluminescence data, and the single particle infrared photoluminescence signal of potassium feldspar; Scanning the sample to be tested, and positioning the positions of multiple high radiation response particles according to the scanning result; According to the spatial distribution information, calculating the particle distribution density of the potassium feldspar, and based on the particle distribution density and the positions of the multiple high radiation response particles, constructing a mineral spatial relationship graph through a graph convolution network; Using the mineral spatial relationship graph, aggregating the radiation measurement signals of adjacent particles, and synchronously training the radiation measurement signals and the quartz optical photoluminescence data; Fusing the training result and the single particle infrared photoluminescence signal, and establishing a heterogeneous adaptive standard growth curve according to the fusion result; The step of fusing the training result and the single particle infrared photoluminescence signal, and establishing a heterogeneous adaptive standard growth curve according to the fusion result comprises the following steps: Extracting a weight parameter matrix after synchronous training from the training result; Fusing the weight parameter matrix and the single particle infrared photoluminescence signal to obtain a fusion result; According to the fusion result, establishing a correction mapping relationship between radiation intensity and time; Based on the correction mapping relationship, generating a heterogeneous adaptive standard growth curve.

2. The method of claim 1, wherein, The step of calculating the particle distribution density of the potassium feldspar according to the spatial distribution information, and based on the particle distribution density and the positions of the multiple high radiation response particles, constructing a mineral spatial relationship graph through a graph convolution network comprises the following steps: Extracting particle size data of particles in the potassium feldspar and distance data between the particles and adjacent particles from the spatial distribution information; According to the distance data and the particle size data, calculating the particle distribution density of the potassium feldspar in different regions; Taking each particle as a node, and the node attribute comprising a radiation intensity value and a particle distribution density in a region to which the node belongs; Based on the positions of the multiple high radiation response particles and the node attribute, generating a mineral spatial relationship graph.

3. The method of claim 2, wherein, The step of generating a mineral spatial relationship graph based on the positions of the multiple high radiation response particles and the node attribute comprises the following steps: Marking each node corresponding to the position of the high radiation response particle as a center node; Calculating the spatial distance between each center node and all other nodes; Taking the center node as a reference, searching for adjacent nodes with a spatial distance less than a preset distance threshold value; Calculating a second radiation correlation coefficient of the center node and the corresponding adjacent node; From all adjacent nodes, screening key adjacent nodes with a second radiation correlation coefficient greater than a preset correlation threshold value, and establishing a bidirectional connection relationship between the center node and the key adjacent nodes; Integrating the node attribute of all nodes and the bidirectional connection relationship to generate a mineral spatial relationship graph.

4. The method of claim 3, wherein, The step of calculating the second radiation correlation coefficient of the center node and the corresponding adjacent node comprises the following steps: Splicing the radiation intensity values of the nodes at different time sampling points into radiation intensity time sequence data of the corresponding nodes; Dividing the radiation intensity time sequence data into multiple groups of radiation intensity sub-sequences according to a preset time window; The intensity change covariance of the center node and the adjacent node is calculated based on each group of radiation intensity subsequences corresponding to the center node and the adjacent node respectively, and a first radiation correlation coefficient of a time window is determined based on a ratio of the intensity change covariance to a preset variance reference value; An average of the first radiation correlation coefficients of all time windows is calculated to obtain a second radiation correlation coefficient.

5. The method of claim 1, wherein, The scanning of the sample to be tested includes: The radiation intensity of the sample to be tested is scanned, and the radiation intensity value of each potassium feldspar single particle is recorded; According to a preset radiation intensity interval, key particles with a radiation intensity value higher than a preset intensity value threshold are screened from all particles; The position coordinates of the key particles are located, and the position coordinates are taken as the positions of the high radiation response particles.

6. The method of claim 1, wherein, The mineral spatial relationship graph is used to aggregate the radiation measurement signals of adjacent particles, and the radiation measurement signals are synchronously trained with the quartz photoluminescence data, including: Each node in the mineral spatial relationship graph is sequentially taken as a target node; A set of adjacent nodes corresponding to the target node is extracted from the mineral spatial relationship graph; The radiation intensity mean value of the set of adjacent nodes is calculated by weighting, and the radiation intensity mean value is converted into an aggregated radiation measurement signal; According to a predetermined time period, the aggregated radiation measurement signal and the quartz photoluminescence data are synchronously input into a training model, and the weight parameters are iteratively updated in the training model to make the aggregated radiation measurement signal and the quartz photoluminescence data reach a preset matching degree.

7. A system for establishing a potassium feldspar luminescence dating standard growth curve based on radiation measurements, characterized by It includes: An acquisition module is configured to acquire spatial distribution information of potassium feldspar single particles in a sample to be tested, quartz photoluminescence data, and single-particle infrared luminescence signals of potassium feldspar; A scanning module is configured to scan the sample to be tested and locate positions of multiple high radiation response particles according to a scanning result; A calculation module is configured to calculate a particle distribution density of the potassium feldspar according to the spatial distribution information, and construct a mineral spatial relationship graph by a graph convolution network based on the particle distribution density and the positions of the multiple high radiation response particles; An aggregation module is configured to aggregate radiation measurement signals of adjacent particles by using the mineral spatial relationship graph, and synchronously train the radiation measurement signals with the quartz photoluminescence data; A fusion module is configured to fuse a training result with the single-particle infrared luminescence signals, and establish a heterogeneous adaptive standard growth curve according to a fusion result. The fusion of the training result and the single-particle infrared luminescence signals and the establishment of the heterogeneous adaptive standard growth curve according to the fusion result include: A weight parameter matrix synchronously trained from the training result is extracted; The weight parameter matrix is fused with the single-particle infrared luminescence signals to obtain a fusion result; A correction mapping relationship between radiation intensity and time is established according to the fusion result; Based on the correction mapping relationship, a heterogeneous adaptive standard growth curve is generated.

8. A computing device, comprising: The application relates to a kind of radiation measurement-based standard growth curve establishment methods for potassium feldspar luminescence dating, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be executed by the processing component, and the method is realized as claimed in any one of claims 1-6.

9. A computer storage medium, characterized in that The application relates to a kind of radiation measurement-based standard growth curve establishment methods for potassium feldspar luminescence dating, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be executed by the processing component, and the method is realized as claimed in any one of claims 1-6.

Images