S100: acquiring logging distribution data of a target location, wherein the logging distribution data comprises density distribution data, gamma distribution data and resistivity distribution data, wherein the density distribution data, the gamma distribution data and the resistivity distribution data are represented as 8-sector data; and amplifying the density distribution data, the gamma distribution data and the resistivity distribution data of the target location through an Akima interpolation method to acquire amplified logging distribution data, wherein the amplified logging distribution data is represented as 32-sector data, comprising amplified density distribution data, amplified gamma distribution data, and amplified resistivity distribution data;

wherein the step of amplifying the density distribution data, the gamma distribution data and the resistivity distribution data of the target location through the Akima interpolation method to acquire the amplified logging distribution data comprises:

for any two adjacent data points x_k and x_k+1, (k=0, 1, 2, . . . , n):

acquiring a constraint condition for a fitted curve:

wherein, y_k denotes an actual logging value at x_k; y_k+1 denotes an actual logging value at x_k+1; t_k denotes a slope of the fitted curve at x_k and x_k+1; t_k+1 denotes a slope at x_k+1 and x_k+2; S(x_k) denotes a value of a cubic polynomial interpolation equation at x_k; y_k′ denotes a slope of the actual logging value at x_k; and y_k+1′ denotes a slope of the actual logging value at x_k+1;

constructing the cubic polynomial interpolation equation:

wherein, S(x) denotes an interval function corresponding to an interval of [x_k×_k+1]; x_k and x_k+1 denote two adjacent data points; and c₀, c₁, c₂, and c₃ are constant parameters to be determined;

calculating an interpolation point for each interval based on the cubic polynomial interpolation equation;

calculating, based on the interpolation point, an interval interpolation equation through adjacent data points;

wherein, t_k and t_k+1 are expressed as follows:

wherein, m_k denotes the slope at the two data points x_k and x_k+1:

m_k+1 denotes a slope at two data points x_k+1 and x_k+2; m_k−1 denotes a slope at two data points x_k−1 and x_k; m_k−2 denotes a slope at two data points x_k−2 and x_k−1; m_k+2 denotes a slope at two data points x_k+2 and x_k+3; and thus, the amplified logging distribution data are acquired;

S200: standardizing the amplified logging distribution data to acquire standardized amplified logging distribution data, wherein the standardized amplified logging distribution data comprises standardized amplified density distribution data, standardized amplified gamma distribution data, and standardized amplified resistivity distribution data; and clustering the amplified logging distribution data to acquire clustered logging data;

S300: performing stratigraphic division based on the clustered logging data, acquiring stratum thicknesses d_str, and assigning stratum labels to the clustered logging data to acquire clustered logging data with stratigraphic information;

S400: performing dimensionality reduction on the clustered logging data with stratigraphic information to acquire first dimensionality-reduced logging data, and taking the first dimensionality-reduced logging data as a weight parameter of the standardized amplified logging distribution data to acquire a logging-while-drilling (LWD) feature dataset;

S500: acquiring, based on the LWD feature dataset, an LWD photoelectric data prediction curve through a trained missing curve prediction model;

wherein, the LWD photoelectric data prediction curve is acquired through the trained missing curve prediction model as follows:

forming the missing curve prediction model, comprising a t-channel image recognition network with 2t convolutional layers and 2t average pooling layers, wherein each channel comprises a first convolutional layer, a first average pooling layer, a second convolutional layer and a second average pooling layer that are sequentially connected; each convolutional layer has a different size; an f-th channel comprises a (4×f−1)×(4×f−1) first convolutional layer, a (4×f+4)×(4×f+4) second convolutional layer, and 2×2 pooling layers; and all the channels are connected together to one fully connected layer and one naive Bayesian decision maker;

denoting the first convolutional layer of a first channel, the first convolutional layer of a second channel, and the first convolutional layer of a third channel as a C1 layer, a C3 layer, and a C5 layer, respectively, wherein the C1 layer is configured to convolve an input image through 8 h_t×h_t convolution kernels; the C3 layer is configured to convolve an input image through 8 3h_t×3h_t convolution kernels; and the C5 layer is configured to convolve an input image through 8 5h_t×5 h_t convolution kernels, wherein a convolution result is calculated as follows:

wherein, con_{p, q}^l denotes the convolution result at a position (p, q); l denotes a current layer number; CON denotes a matrix covered by the convolution kernels; L and W denote a length and a width of the matrix covered by the convolution kernels; m1 and m2 denote serial numbers of lengths of the convolution kernels, m1 ranging from 1 to L, and m2 ranging from 1 to W; ker denotes a kernel function; and b denotes a corresponding bias term;

fitting the convolution result through a rectified linear unit (ReLU) function to acquire a fitted convolution result;

performing upsampling pooling on the fitted convolution result:

wherein, con_{m3, m4}^l denotes a two-dimensional matrix representation of a pooled convolution result; and (m3, m4) denotes a position of the convolution result;

converting the pooled convolution result into a tiled one-dimensional feature vector through a tiling layer;

integrating the tiled one-dimensional feature vector through the fully connected layer:

wherein, con_key^l denotes a one-dimensional matrix representation of a feature after the integration through the fully connected layer; l denotes the current layer number; key denotes an index value of a one-dimensional matrix; R denotes a length of a feature vector in an (l−1)-th layer; w denotes a weight matrix; b denotes the bias term; and r denotes a serial number of a data point of the feature vector;

introducing the naive Bayesian decision maker into the fully connected layer; and

taking the integrated tiled one-dimensional feature vector as the LWD photoelectric data prediction curve;

S600: calculating, based on the LWD photoelectric data prediction curve, an average value of photoelectric data of each stratum, and acquiring a lithology evaluation result of each stratum; and

S700: placing and drilling a well based on the average value of photoelectric data of each stratum and the lithology evaluation result of each stratum.