A method for simulating and predicting indoor radioactive distribution

Abstract

The application discloses a kind of simulation prediction methods of indoor radioactivity distribution, comprising: constructing indoor building model, the radioactivity absorption simulation of different radionuclide sources is carried out, and the photon flux of corresponding position in the room of indoor building model is obtained;Calculate the ideal absorption dose rate of room irradiation under the specific activity condition of different radionuclide sources;Photon flux is used as input data, and target absorption dose rate is used as output data;Training data set is constructed, random forest regression model is trained, and trained random forest regression model is output;The photon flux of radionuclide in different positions in the room collected by photon detector is input into trained random forest regression model, and the distribution of indoor room irradiation absorption dose rate is obtained.The application can be used for radioactivity level analysis in multi-target scene, and solve the problem that previous building material radioactivity level analysis can only be calculated by empirical formula, so that it is not accurate enough.

Description

Technical Field

[0001] This invention relates to the field of indoor radioactivity research, and specifically to a method for simulating and predicting indoor radioactivity distribution. Background Technology

[0002] Radioactive hazards from building materials mainly include two types: external radiation and internal radiation. External radiation primarily originates from the radioactive nuclides uranium and thorium they contain and their decay products. These nuclides attack human cells by emitting gamma rays, causing tissue damage and thus generating radioactive hazards. The main nuclides involved are 226Ra, 232Th, and 40K. Internal radiation is produced by the decay product 222Rn of the nuclide 226Ra in the building's air. Radon is an inert gas, colorless and odorless. Because radon gas itself decays and produces radioactive decay products, these decay products can be inhaled by the lungs when a person breathes them. These decay products further decay into alpha particles that damage human tissues, ultimately leading to lung cancer.

[0003] In current technologies, when studying indoor radioactivity, the air absorbed dose rate at the center of a standard room model can be obtained using empirical formulas to preliminarily assess the radioactivity level of building materials. However, this assessment can only be achieved through complex calculations. Currently, empirical formulas such as the internal and external radiation indices are mainly used to evaluate the radioactivity level of building materials, without simulating and quantifying actual radiation scenarios. In reality, the distribution of radioactivity indoors is also affected by wall thickness, density, door and window size, and building structure. Therefore, existing evaluation methods primarily provide a rapid assessment of building material radioactivity; their accuracy and the distribution of radioactivity in practical applications require further in-depth research. Summary of the Invention

[0004] To address the aforementioned shortcomings of existing technologies, this invention provides a method for simulating and predicting indoor radioactivity distribution. By simulating radioactivity at different locations within a building and using machine learning techniques, it is possible to accurately predict the absorbed dose rate based on the photon flux at different locations within the building.

[0005] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0006] A method for simulating and predicting indoor radioactivity distribution is provided, comprising the following steps:

[0007] S1: Construct an indoor building model, simulate the radioactive absorption of different radionuclide sources on the indoor building model, and obtain the photon flux at the corresponding location in the room of the indoor building model;

[0008] S2: Based on the specific activity of the radionuclide source used in the radioactive absorption simulation process in step S1, substitute the specific activity of each radionuclide source into the standard room irradiation model, and calculate the ideal absorbed dose rate of room irradiation under different radionuclide source specific activity conditions.

[0009] S3: Using the photon flux obtained in step S1 as input data and the target absorbed dose rate obtained in step S2 as output data, construct a training dataset, input it into the random forest regression model, train the random forest regression model, and output the trained random forest regression model.

[0010] S4: The photon flux of radionuclides at different locations in the room is collected by a photon detector and input into a trained random forest regression model. The output is the radiation absorbed dose rate at different locations in the room, thus obtaining the distribution of the radiation absorbed dose rate in the room.

[0011] Further, step S1 includes:

[0012] S11: Based on the set indoor room dimensions, wall thickness, and density, construct an indoor building model in OpenMC software and determine the radionuclide source terms;

[0013] S12: Calculate the photon source intensity S for each radionuclide source based on the specific activity of the radionuclide source term. u ;

[0014]

[0015] Where u is the number of the radionuclide source, C u ρ is the specific activity of the radionuclide source, ρ is the density of the wall, and E is the specific activity of the radionuclide source. i For the emission energy of gamma photons, Y i , where i is the branching ratio corresponding to the gamma photon emission energy, i is the gamma ray number during the emission of the radioactive nuclide source, and I is the number of gamma rays;

[0016] S13: Based on the calculated photon source intensity S for each radionuclide source u The photon source intensity S of each radionuclide source u Calculate the gamma strength S of the walls assigned to different areas of the interior building model. v S v =S u ×V v Where v is the wall number of the interior architectural model, V v For the volume of the wall;

[0017] S14: Divide the interior of the room in the interior building model into a regular grid of n1×n2×n3, where n1, n2, and n3 are the number of regular grids in the length, width, and height directions, respectively. The dimensions of each grid are: length Δx, width Δy, and height Δz.

[0018]

[0019] Among them, L x ,L y ,L z These are the length, width, and height dimensions of the room.

[0020] S15: Use the Tally module in OpenMC software to record the wall's gamma intensity S v Under certain conditions, the flux distribution of gamma rays inside the room is determined, and the number of photons N in each regular grid within the room is obtained. p p is the number of the regular grid;

[0021] S16: Based on the number of photons N p Calculate the photon flux φ at the center of each regular grid. p ;

[0022]

[0023] Where A is the volume of the regular grid, and Δt is the time interval for recording the gamma-ray flux;

[0024] S17: Return to step S12 and reset the specific activity of the radionuclide source to C. u ′, and satisfy C u ′≠C u Then, perform steps S12-S16 to calculate the specific activity of the radionuclide source as C. u The photon flux at the center of each regular grid point at time ′;

[0025] S18: Specific activity until m times the radionuclide source is reset. Then, the photon flux at the center of each regular grid point after each reset of the specific activity of the radionuclide source was obtained, thus obtaining the photon flux data of the corresponding location in the room simulated by the radionuclide source for each time. Let p be the photon flux at the center of the regular grid after the specific activity of the radionuclide source is reset for the mth time.

[0026] S19: Repeat steps S12-S18 to obtain the photon flux data of the corresponding location in the room for each radionuclide source. The specific activity of the U-th radionuclide source is reset for the m-th time at the center of the regular grid p. The resulting photon flux, U, represents the number of different types of radioactive nuclide sources.

[0027] Furthermore, the method for calculating the room's absorbed radiation dose rate in step S2 is as follows:

[0028]

[0029] Among them, D m λ represents the ideal absorbed dose rate for all radionuclide sources within the room. u denoted as the ideal absorption coefficient of the u-th radionuclide source.

[0030] Further, step S3 includes:

[0031] S31: Retrieve the photon flux corresponding to the specific activity of each radionuclide source after each reset as input data. The ideal absorbed dose rate D is calculated in step S2. m As output data y m =D m This forms a training dataset (x) to predict the absorbed dose rate at the center of a regular grid p within a room. m ,y m );

[0032] S32: Create a training dataset (X,Y), where X = (x1,x2,…,x…) m Y = (y1, y2, ..., y) m X is the input training dataset, and y is the output training dataset;

[0033] S33: Construct a random forest regression model;

[0034]

[0035] Where y is the output of the random forest regression model, g is the decision tree number, and f g (x) is the prediction function of the g-th decision tree, where x is the input variable and n is the input variable. trees Let be the number of decision trees, j be the training sample number of the node in the g-th decision tree, and y be the number of the decision trees. j Let n be the training target value of the j-th node in the g-th decision tree. L Let be the number of training samples in the leaf nodes of the g-th decision tree;

[0036] S34: Randomly select M training data groups with replacement from the training dataset (X,Y) as training samples to generate the training set S for the prediction function of the g-th decision tree. g ={(x1,y1),(x2,y2),…,(x M ,y M )},(xM ,y M () represents the Mth training data set extracted;

[0037] S35: Transfer the training set S g Input the prediction function f of the g-th decision tree g In (x), for the prediction function f g (x) is trained, and during the training process, when splitting each node of the g-th decision tree, the data is obtained from the input data. Several subsets of data are randomly selected as input subsets x′. M ,

[0038] S36: Using input data Selected input subset x′ M Train the split nodes and calculate the impurity of each node after training.

[0039]

[0040] Where t is the node number, d is the training sample number of the input node, and I(g) t Let y be the impurity of the t-th node, |D| be the number of training sample nodes under node t, and y be the impurity of the t-th node. d The target absorbed dose rate for the training samples of the node. This represents the average absorbed dose rate output after node training.

[0041] S37: Calculate the total impurity I(g) after splitting the g-th decision tree;

[0042]

[0043] Where |D1| is the number of left nodes in the decision tree split, and |D2| is the number of right nodes in the decision tree split. For the impurity of the left node, The impurity of the right node;

[0044] S38: Set the minimum threshold I0 of the total impurity after splitting the decision tree. If I(g)≤I0, stop splitting the decision tree and take the split node as the leaf node of the g-th decision tree. If I(g)>I0, return to step S35 and split the g-th decision tree again until I(g)≤I0 is satisfied, and then output the leaf node of the g-th decision tree.

[0045] S39: Repeat steps S34-S38 to train each decision tree in the random forest regression model in turn, obtain the leaf nodes of each decision tree, and output the trained random forest regression model.

[0046] The beneficial effects of this invention are as follows: This invention can be used for radioactivity level analysis in multi-target scenarios, solving the problem that previous analyses of building material radioactivity levels could only be calculated using empirical formulas, resulting in insufficient accuracy. By simulating the source terms, geometric parameters, and detectors of indoor building models, and training photon flux and absorbed dose rate using machine learning, the indoor radioactivity distribution under different structural parameters such as wall thickness, density, and materials can be predicted and simulated, providing more refined data support for monitoring indoor radioactivity levels in buildings.

[0047] This invention reduces the complex impact of wall self-shielding and the energy attenuation effect of gamma rays propagating in the air on simulation results by using a black-box approach. It can simulate the distribution of indoor radioactivity levels in multiple application scenarios, and can provide certain data support, especially in laboratory designs with strict requirements for background radiation dose. Attached Figure Description

[0048] Figure 1 This is a flowchart of a method for simulating and predicting indoor radioactivity distribution. Detailed Implementation

[0049] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0050] like Figure 1 As shown, a method for simulating and predicting indoor radioactivity distribution includes the following steps:

[0051] S1: Construct an indoor building model, and simulate the radioactive absorption of different radionuclide sources on the indoor building model to obtain the photon flux at corresponding locations within the room of the indoor building model. Step S1 specifically includes:

[0052] S11: Based on the set indoor room dimensions, wall thickness, and density, construct the indoor building model in OpenMC software and determine the radionuclide source terms. In this embodiment, the indoor room structure is constructed using standard dimensions: 4m long, 5m wide, and 2.8m high, with a wall thickness of 0.2m and a density of 2350 kg / m³. 3The determination of radionuclide source terms is based on naturally occurring uranium-series, thorium-series, and potassium-40 radionuclides in building materials. Key parameters include the active concentration units (Bq / kg) of the three nuclides 226Ra, 232Th, and 40K; the primary gamma-ray energies and branching ratios of the nuclides; potassium-40 (K-40): primary energy 1460.8 keV, emission probability 0.1066; uranium-series (Ra-226): includes multiple gamma rays released from the decay of 214Pb and 214Bi; thorium-series (Th-232): includes multiple gamma rays released from the decay of 228Ac and 212Pb.

[0053] S12: Calculate the photon source intensity S for each radionuclide source based on the specific activity of the radionuclide source term. u ;

[0054]

[0055] Where u is the number of the radionuclide source, C u ρ is the specific activity of the radionuclide source, ρ is the density of the wall, and E is the specific activity of the radionuclide source. i For the emission energy of gamma photons, Y i , where i is the branching ratio corresponding to the gamma photon emission energy, i is the gamma ray number during the emission of the radioactive nuclide source, and I is the number of gamma rays;

[0056] S13: Based on the calculated photon source intensity S for each radionuclide source u The photon source intensity S of each radionuclide source u Calculate the gamma strength S of the walls assigned to different areas of the interior building model. v S v =S u ×V v Where v is the wall number of the interior architectural model, V v For the volume of the wall;

[0057] S14: Divide the interior of the room in the interior building model into a regular grid of n1×n2×n3, where n1, n2, and n3 are the number of regular grids in the length, width, and height directions, respectively. The dimensions of each grid are: length Δx, width Δy, and height Δz.

[0058]

[0059] Among them, L x ,L y ,L z These are the length, width, and height dimensions of the room.

[0060] S15: Use the Tally module in OpenMC software to record the wall's gamma intensity S vUnder certain conditions, the flux distribution of gamma rays inside the room is determined, and the number of photons N in each regular grid within the room is obtained. p p is the number of the regular grid;

[0061] S16: Based on the number of photons N p Calculate the photon flux φ at the center of each regular grid. p ;

[0062]

[0063] Where A is the volume of the regular grid, and Δt is the time interval for recording the gamma-ray flux;

[0064] S17: Return to step S12 and reset the specific activity of the radionuclide source to C. u ′, and satisfy C u ′≠C u Then, perform steps S12-S16 to calculate the specific activity of the radionuclide source as C. u The photon flux at the center of each regular grid point at time ′;

[0065] S18: Specific activity until m times the radionuclide source is reset. Then, the photon flux at the center of each regular grid point after each reset of the specific activity of the radionuclide source was obtained, thus obtaining the photon flux data of the corresponding location in the room simulated by the radionuclide source for each time. Let p be the photon flux at the center of the regular grid after the specific activity of the radionuclide source is reset for the mth time.

[0066] S19: Repeat steps S12-S18 to obtain the photon flux data of the corresponding location in the room for each radionuclide source. The specific activity of the U-th radionuclide source is reset for the m-th time at the center of the regular grid p. The resulting photon flux, U, represents the number of different types of radioactive nuclide sources.

[0067] This embodiment can use the matplotlib module in OpenMC software to plot the heat map of gamma photon flux, as well as the photon flux distribution at different slices along the X, Y, and Z axes.

[0068] S2: Based on the specific activity of the radionuclide sources used in the radioactive absorption simulation in step S1, substitute the specific activity of each radionuclide source into the standard room irradiation model to calculate the ideal absorbed dose rate of room irradiation under different radionuclide source specific activities. The method for calculating the room irradiation absorbed dose rate in step S2 is as follows:

[0069]

[0070] Among them, D m λ represents the ideal absorbed dose rate for all radionuclide sources within the room. u denoted as the ideal absorption coefficient of the u-th radionuclide source.

[0071] In this embodiment, the concrete room measures 4m long, 5m wide, and 2.8m high, with walls having a thickness of 20cm and a density of 2350kg / m³. The formula for calculating the absorbed dose rate of the entire room is:

[0072] D = 0.92C Ra +1.1C Th +0.08C K ;

[0073] Where D is the absorbed dose rate of the entire room, in nGy / h; C Ra C Th C K These are the specific activities of 226Ra, 232Th, and 40K, respectively, in Bq / kg.

[0074] S3: Using the photon flux obtained in step S1 as input data and the target absorbed dose rate obtained in step S2 as output data, construct a training dataset, input it into the random forest regression model, train the random forest regression model, and output the trained random forest regression model. Step S3 specifically includes:

[0075] S31: Retrieve the photon flux corresponding to the specific activity of each radionuclide source after each reset as input data. The ideal absorbed dose rate D is calculated in step S2. m As output data y m =D m This forms a training dataset (x) to predict the absorbed dose rate at the center of a regular grid p within a room. m ,y m );

[0076] S32: Create a training dataset (X,Y), where X = (x1,x2,…,x…) m Y = (y1, y2, ..., y) m X is the input training dataset, and y is the output training dataset;

[0077] S33: Construct a random forest regression model;

[0078]

[0079] Where y is the output of the random forest regression model, g is the decision tree number, and fg (x) is the prediction function of the g-th decision tree, where x is the input variable and n is the input variable. trees Let be the number of decision trees, j be the training sample number of the node in the g-th decision tree, and y be the number of the decision trees. j Let n be the training target value of the j-th node in the g-th decision tree. L Let be the number of training samples in the leaf nodes of the g-th decision tree;

[0080] S34: Randomly select M training data groups with replacement from the training dataset (X,Y) as training samples to generate the training set S for the prediction function of the g-th decision tree. g ={(x1,y1),(x2,y2),…,(x M ,y M )},(x M ,y M () represents the Mth training data set extracted;

[0081] S35: Transfer the training set S g Input the prediction function f of the g-th decision tree g In (x), for the prediction function f g (x) is trained, and during the training process, when splitting each node of the g-th decision tree, the data is obtained from the input data. Several subsets of data are randomly selected as input subsets x′. M ,

[0082] S36: Using input data Selected input subset x′ M Train the split nodes and calculate the impurity of each node after training.

[0083]

[0084] Where t is the node number, d is the training sample number of the input node, and I(g) t Let y be the impurity of the t-th node, |D| be the number of training sample nodes under node t, and y be the impurity of the t-th node. d The target absorbed dose rate for the training samples of the node. This represents the average absorbed dose rate output after node training.

[0085] S37: Calculate the total impurity I(g) after splitting the g-th decision tree;

[0086]

[0087] Where |D1| is the number of left nodes in the decision tree split, and |D2| is the number of right nodes in the decision tree split. For the impurity of the left node, The impurity of the right node;

[0088] Each decision tree selects a training subset by random sampling, and the feature selection when splitting a node is a subset randomly selected from the training subset. This randomness ensures the diversity among different decision trees and reduces the risk of overfitting.

[0089] S38: Set the minimum threshold I0 of the total impurity after splitting the decision tree. If I(g)≤I0, stop splitting the decision tree and take the split node as the leaf node of the g-th decision tree. If I(g)>I0, return to step S35 and split the g-th decision tree again until I(g)≤I0 is satisfied, and then output the leaf node of the g-th decision tree.

[0090] S39: Repeat steps S34-S38 to train each decision tree in the random forest regression model in turn, obtain the leaf nodes of each decision tree, and output the trained random forest regression model.

[0091] S4: The photon flux of radionuclides at different locations in the room is collected by a photon detector and input into a trained random forest regression model. The output is the radiation absorbed dose rate at different locations in the room, thus obtaining the distribution of the radiation absorbed dose rate in the room.

[0092] This invention can be used for radioactivity level analysis in multi-target scenarios, solving the problem that previous analyses of building material radioactivity levels could only be calculated using empirical formulas, resulting in insufficient accuracy. By simulating the source terms, geometric parameters, and detectors of indoor building models, and training photon flux and absorbed dose rate using machine learning, the indoor radioactivity distribution under different structural parameters such as wall thickness, density, and materials can be predicted and simulated, providing more refined data support for monitoring indoor radioactivity levels in buildings.

[0093] This invention reduces the complex impact of wall self-shielding and the energy attenuation effect of gamma rays propagating in the air on simulation results by using a black-box approach. It can simulate the distribution of indoor radioactivity levels in multiple application scenarios, and can provide certain data support, especially in laboratory designs with strict requirements for background radiation dose.

Claims

1. A method for simulating and predicting indoor radioactivity distribution, characterized in that, Includes the following steps: S1: Construct an indoor building model, simulate the radioactive absorption of different radionuclide sources on the indoor building model, and obtain the photon flux at the corresponding location in the room of the indoor building model; S2: Based on the specific activity of the radionuclide source used in the radioactive absorption simulation process in step S1, substitute the specific activity of each radionuclide source into the standard room irradiation model, and calculate the ideal absorbed dose rate of room irradiation under different radionuclide source specific activity conditions. S3: Using the photon flux obtained in step S1 as input data and the target absorbed dose rate obtained in step S2 as output data, construct a training dataset, input it into the random forest regression model, train the random forest regression model, and output the trained random forest regression model. S4: The photon flux of radionuclides at different locations in the room is collected by a photon detector and input into a trained random forest regression model. The output is the radiation absorbed dose rate at different locations in the room, thus obtaining the distribution of the radiation absorbed dose rate in the room. Step S1 includes: S11: Based on the set indoor room dimensions, wall thickness, and density, construct an indoor building model in OpenMC software and determine the radionuclide source terms; S12: Calculate the photon source intensity of each radionuclide source based on the specific activity of the radionuclide source term. S u ; ； in, u This is the numbering of the radioactive nuclide source. Specific activity of the radionuclide source. For wall density, E i To emit energy for gamma photons Y i The branching ratio corresponding to the gamma photon emission energy. i This refers to the numbering of gamma rays emitted during the radioactive emission process of a radionuclide source. I The number of gamma rays; S13: Based on the calculated photon source intensity of each radionuclide source S u The photon source intensity of each radionuclide source S u Calculate the gamma strength of the walls assigned to different areas of the interior building model. ; ,in, v Number the walls of the interior architectural model. For the volume of the wall; S14: Divide the interior of the room in the interior architectural model into... Regular mesh, These represent the number of regular grid cells distributed along the length, width, and height directions, respectively, and the length of each grid cell is... ,Width ,high The dimensions are; ； in, These are the length, width, and height dimensions of the room. S15: Record the wall's gamma intensity using the Tally module in OpenMC software. The flux distribution of gamma rays inside the room under certain conditions was determined, and the number of photons in each regular grid within the room was obtained. N p , p The number of the regular grid; S16: Based on the number of photons N p Calculate the photon flux at the center of each regular grid. ; ； in, A The volume of the regular mesh. The time interval for recording gamma ray flux; S17: Return to step S12 and reset the specific activity of the radionuclide source to... And satisfy Then, perform steps S12-S16 to calculate the specific activity of the radionuclide source. The photon flux at the center of each regular grid point; S18: Until after m The specific activity of the radionuclide source was reset. Then, the photon flux at the center of each regular grid point after each reset of the specific activity of the radionuclide source was obtained, thus obtaining the photon flux data of the corresponding location in the room simulated by the radionuclide source for each time. , For the first m After resetting the specific activity of the radionuclide source, a regularized grid is created. p Photon flux at the center location; S19: Repeat steps S12-S18 to obtain the photon flux data of the corresponding location in the room for each radionuclide source. , For the first U Radionuclide sources in a regular grid p The location of the center is the first m The specific activity of the radionuclide source was reset. The resulting photon flux U This refers to the number of different types of radionuclide sources.

2. The method for simulating and predicting indoor radioactivity distribution according to claim 1, characterized in that, The method for calculating the room's absorbed radiation dose rate in step S2 is as follows: ； in, This represents the ideal absorbed dose rate for all radionuclide sources within the room. For the first u The ideal absorption coefficient of a radioactive nuclide source.

3. The method for simulating and predicting indoor radioactivity distribution according to claim 2, characterized in that, Step S3 includes: S31: Retrieve the photon flux corresponding to the specific activity of each radionuclide source after each reset as... Input data And calculate the ideal absorbed dose rate in step S2. As output data To form a regular grid within the predicted room p Training data set of absorbed dose rate at the center location ; S32: Create a training dataset , , , For the input training dataset, To output the training dataset; S33: Construct a random forest regression model; ， ； in, y This is the output of the random forest regression model. g The numbering of the decision tree, For the first g The prediction function of a decision tree. x For input variables, For the number of decision trees, j For the first g The training sample number of the node in the decision tree. y j For the first g The first decision tree j The training target value for each node. n L For the first g The number of training samples in the leaf nodes of a decision tree; S34: From the training dataset Random draws with replacement M The training data set is used as the training sample to generate the first training data set. g Training set of prediction functions for decision trees , For the first one drawn M One training data set; S35: Transfer the training set Enter the first g Prediction function of a decision tree In the middle, for the prediction function Training was conducted, and during the training process, the first g When a decision tree splits at each node, it starts from the input data. Randomly select several subsets of data as input subsets , ; S36: Using input data Selected input subset Train the split nodes and calculate the impurity of each node after training. ， ； in, t For the node number, d This refers to the training sample number of the input node. For the first t The impurity of each node, For nodes t The number of training sample nodes, The target absorbed dose rate for the training samples of the node. This represents the average absorbed dose rate output after node training. S37: Calculate the... g Total impurity after splitting a decision tree ; ； in, The number of left nodes in a decision tree split. The number of right nodes in a decision tree split. For the impurity of the left node, The impurity of the right node; S38: Set the minimum threshold for total impurity after the decision tree splits. ,like If the decision tree stops splitting nodes, the currently split node becomes the first node. g The leaf nodes of the decision tree; if Then return to step S35 and re-process the first... g The decision tree splits its nodes until the condition is met. When, output the first g Leaf nodes of a decision tree; S39: Repeat steps S34-S38 to train each decision tree in the random forest regression model in turn, obtain the leaf nodes of each decision tree, and output the trained random forest regression model.

Images