CN-121998994-A - Slurry shield mud film maintenance test device and service life prediction method

CN121998994ACN 121998994 ACN121998994 ACN 121998994ACN-121998994-A

Abstract

The invention provides a slurry shield mud film maintenance test device and a service life prediction method, which relate to the technical field of mud film formation and comprise the steps of obtaining an original data set; the method comprises the steps of carrying out image preprocessing according to a surface image sequence to obtain preprocessed image data, carrying out crack form evolution feature extraction according to the preprocessed image data to obtain a crack form evolution feature set, carrying out prediction model construction processing according to the crack form evolution feature set, a slurry proportioning parameter, a stratum physical parameter and a film forming pressure value to construct and obtain a mud film life prediction model, and carrying out life calculation according to the mud film life prediction model to obtain a mud film residual life prediction value from a current state to a failure moment of a mud film. According to the invention, by combining automatic monitoring of a mud film observation mechanism and intelligent analysis based on fusion of image sequences and physical field parameters, a mud film life prediction model capable of quantitatively representing a crack evolution rule and predicting failure time is constructed.

Inventors

FENG KUN
WANG CHENLIN
LIANG XIAOMING
JIN ZHENYU

Assignees

西南交通大学

Dates

Publication Date: 20260508
Application Date: 20260410

Claims (10)

1. A lifetime prediction method, comprising: acquiring an original data set, wherein the original data set comprises a surface image sequence of a mud film during simulated stratum environment maintenance, a mud proportioning parameter, a stratum physical parameter and a film forming pressure value; Performing image preprocessing according to the surface image sequence, and performing partition contrast enhancement on a single frame image to inhibit specular highlight and performing self-adaptive segmentation based on image gray statistics to obtain preprocessed image data for eliminating non-uniform illumination interference; Extracting crack morphological evolution features according to the preprocessed image data, and obtaining a crack morphological evolution feature set representing crack development from punctiform initiation to linear penetration by detecting crack edge contours and carrying out morphological connection and background noise filtering on broken pixel segments; Performing prediction model construction according to the crack morphological evolution feature set, the slurry proportioning parameter, the stratum physical parameter and the film forming pressure value, and constructing to obtain a mud film life prediction model by fusing the crack space-time evolution feature and the physical field parameter and adopting a Bayes optimization strategy to fit a nonlinear mapping relation; and carrying out life calculation according to the life prediction model of the mud film and the acquired real-time maintenance data, and obtaining a mud film residual life prediction value from the current state to the failure moment of the mud film through model deduction.
2. The lifetime prediction method according to claim 1, wherein the image preprocessing is performed according to the sequence of surface images, wherein the preprocessing image data for eliminating non-uniform illumination interference is obtained by performing partition contrast enhancement on a single frame image to suppress specular highlights and performing adaptive segmentation based on image gray statistics, comprising: Carrying out wet surface specular reflection inhibition treatment according to the surface image sequence, dividing a single frame image into a plurality of subareas, carrying out independent self-adaptive histogram equalization, and compressing gray values of a highlight area while limiting the contrast stretching degree in each subarea to obtain the single frame image with enhanced contrast; carrying out crack foreground extraction processing according to the single frame image with enhanced contrast, and obtaining a preliminary crack foreground binary image by calculating the inter-class variance of the image to determine an optimal threshold value for dividing the gray level image into a foreground and a background; And carrying out background texture noise filtering treatment according to the preliminary crack foreground binary image, and analyzing the geometric characteristics of the communicated region to distinguish discrete noise formed by slurry particles from a continuous linear crack structure so as to obtain preprocessed image data for eliminating non-uniform illumination interference.
3. The lifetime prediction method according to claim 1, wherein the extracting of the morphological evolution features of the crack according to the preprocessed image data, by detecting the edge profile of the crack and performing morphological connection and background noise filtering on the broken pixel segment, a set of morphological evolution features of the crack characterizing the crack from punctiform initiation to linear through development is obtained, comprising: performing crack edge contour detection processing according to the preprocessed image data, and calculating an image gradient to identify a region with abrupt pixel gray value changes and position potential crack edges so as to obtain an edge intensity map containing edge position and direction information; Performing fracture crack segment connection treatment according to the edge intensity graph, and bridging edge pixel interruption caused by uneven illumination or mud film particle shielding by applying morphological expansion operation, so as to connect discrete edge points into a continuous linear structure, thereby obtaining a preliminary communicated crack skeleton graph; And carrying out background particle noise filtering and characteristic quantification treatment according to the preliminary connected crack skeleton diagram, and analyzing the length, area and length-width ratio geometric characteristics of the connected region to distinguish an elongated connected region formed by a real crack from an approximately round small-area noise formed by slurry particles so as to obtain a crack morphological evolution characteristic set for representing the crack development from punctiform initiation to linear penetration.
4. The life prediction method according to claim 1, wherein the predicting model construction process is performed according to the crack morphology evolution feature set, the slurry proportioning parameter, the stratum physical parameter and the film forming pressure value, and the mud film life prediction model is constructed by fusing the crack space-time evolution feature and the physical field parameter and fitting a nonlinear mapping relation by adopting a bayesian optimization strategy, and the method comprises the following steps: carrying out multi-source heterogeneous feature fusion treatment according to the crack morphological evolution feature set, the slurry proportioning parameter, the stratum physical parameter and the film forming pressure value, and obtaining a comprehensive feature vector for model training by associating and integrating the length and area space-time sequence of the crack with the specific gravity, viscosity and sand content of slurry and the porosity and permeability coefficient of stratum; model training and super-parameter optimization processing are carried out according to the comprehensive feature vectors, and an optimized prediction model foundation is obtained by inputting data into a gradient lifting frame and adopting a Gaussian process as an agent model to guide the optimal super-parameter combination of a search model; And carrying out prediction function construction processing according to the optimized prediction model base, and constructing and obtaining a mud film life prediction model by expressing the mapping relation between the crack development rate and the multi-physical-field condition as a weighted function form related to mud parameters, stratum parameters and film forming pressure.
5. The life prediction method according to claim 1, wherein the life calculation is performed according to the life prediction model of the sludge film and the acquired real-time maintenance data, and the predicted value of the residual life of the sludge film from the current state to the failure time is obtained by model deduction, comprising: Performing real-time feature extraction and vectorization processing according to the real-time maintenance data, and obtaining a real-time feature vector by performing pretreatment and crack morphological feature extraction processes on a mud film surface image obtained in real time and combining real-time mud proportioning, stratum parameters and film forming pressure; Performing model forward deduction processing according to the real-time feature vector, and calculating to obtain a prediction index sequence representing the development trend of the mud film crack under the current multi-physical-field condition by inputting the model forward deduction processing into the mud film life prediction model; and carrying out residual effective air retention time mapping treatment according to the prediction index sequence, and comparing and deducting the crack development trend output by the model with a preset failure criterion to obtain a predicted value of the residual life of the mud film from the current state to the failure moment.
6. The utility model provides a slurry shield mud membrane maintenance test device which characterized in that includes: the test box assembly (1), the test box assembly (1) comprises a test box body (11) and a test box cover plate (12), the test box cover plate (12) is in sealing connection with the test box body (11), and a permeable cylinder placing ring (13) is arranged in the test box body (11); The detachable permeable column assembly (2), the detachable permeable column assembly (2) is arranged in the test box assembly (1), the bottom of the detachable permeable column assembly (2) is fixedly connected with the permeable column placing ring (13), the detachable permeable column assembly (2) comprises an upper permeable column body (21), a detachable permeable column body (22) and a lower permeable column body (23), and the detachable permeable column body (22) is detachably and hermetically connected between the upper permeable column body (21) and the lower permeable column body (23); The environment control mechanism (3), the environment control mechanism (3) comprises a test environment total control (31), a pressure maintaining component (32), a humidity maintaining component (33) and a temperature maintaining component (34), the test environment total control (31) is fixedly arranged outside the test box body (11), and the test environment total control (31) is respectively connected with the pressure maintaining component (32), the humidity maintaining component (33) and the temperature maintaining component (34) in a point mode; The mud film observation mechanism (4), the mud film observation mechanism (4) comprises an image acquisition assembly (41) and an air pressure monitoring assembly (42), the image acquisition assembly (41) is arranged in the test box assembly (1), and the air pressure monitoring assembly (42) is communicated with the inside of the test box assembly (1); wherein, the test box component (1) and the detachable permeation column component (2) are made of transparent materials.
7. The slurry shield mud film maintenance test apparatus as set forth in claim 6, wherein said pressure maintaining unit (32) comprises an air compressor (321), a pressurizing pipe (322), an electronic air pressure gauge (323), an electric control air release valve (324) and a pressure stabilizing valve (325), said air compressor (321) is communicated with said test box unit (1) through said pressurizing pipe (322), said electronic air pressure gauge (323) is fixedly provided on said test box unit (11), said pressure stabilizing valve (325) is provided on said pressurizing pipe (322), said electric control air release valve (324) is fixedly provided on said test box cover plate (12), said humidity maintaining unit (331) comprises an electronic hygrometer (331), an electric control humidifier (332) and a water tank (333), said electronic hygrometer (331) is provided on said test box unit (11), said electric control humidifier (332) and said water tank (333) are provided inside said test box unit (11), said temperature maintaining unit (34) comprises an electronic thermometer (341), a temperature control assembly (342) and an electric control fan (342) are provided on said test box unit (342), the electric control air heater (343) and the ventilation fan (344) are arranged inside the test box body (11).
8. The slurry shield mud film maintenance test device according to claim 6, wherein the air pressure monitoring assembly (42) comprises an air pressure gauge box (421) and at least one air pressure gauge (422), and the air pressure gauge (422) is communicated with the detachable permeable column assembly (2).
9. The slurry shield mud film maintenance test device according to claim 8, wherein the mud film observation mechanism (4) further comprises a mud film air leakage outlet (423) and a mud film air leakage outlet pipe (424), the mud film air leakage outlet (423) is formed in the bottom of the detachable permeable cylinder (22) placing ring, and the mud film air leakage outlet pipe (424) is communicated with the mud film air leakage outlet (423) and the barometer (422).
10. The slurry shield mud film maintenance test device according to claim 8, wherein the barometer (422) is a graduated straight cylinder liquid barometer.

Description

Slurry shield mud film maintenance test device and service life prediction method Technical Field The invention relates to the technical field of slurry film forming, in particular to a slurry shield slurry film maintenance test device and a service life prediction method. Background In the shield tunnel construction, before the operation such as cutter changing is carried out by adopting the under-pressure cabin opening technology, a compact mud film is required to be formed on the face and high-pressure gas in the cabin is assisted, so that the face is effectively supported, the safety of the cabin opening and cutter changing operation is improved, and the stability of the face can be effectively controlled. The reliability core of the technology is that the mud film is easy to generate micro cracks and expand under the coupling action of a complex physical field, so that the air tightness of the mud film is extremely easy to lose, and the stability and the operation safety of a face are threatened, therefore, the long-term crack resistance and the air retention performance of the mud film under different mud formulas and stratum conditions are accurately evaluated through an indoor test before construction, the optimal proportioning and the guarantee of the construction safety and the economical efficiency are very important, at present, the method commonly adopted in the field is that after the mud film is formed by using a single permeable column device, pressurization or observation test is carried out in the same device immediately, the method can only simulate single-point working conditions, the test environment (pressure, temperature and humidity) is difficult to accurately control and keep constant, the real scene of the multi-long-term coupling action in the stratum cannot be effectively reproduced, in addition, the observation of the mud film cracking process is more dependent on manual visual observation or simple camera shooting record, the intelligent correlation between the observed crack morphology and the mud material parameters and the conditions, the service life condition cannot be carried out, the prediction scheme is low, the practical test is difficult to realize, the prediction of the cost is high, the intelligent prediction and the practical engineering system can be easily realized, and the real life is difficult to be predicted, and the practical experiment is difficult to be automatically predicted, and the practical and the life of the system can be predicted and well predicted. Based on the defects in the prior art, a slurry shield mud film maintenance test device and a service life prediction method are needed. Disclosure of Invention The invention aims to provide a slurry shield mud film maintenance test device and a service life prediction method so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: in a first aspect, the present application further provides a lifetime prediction method, including: acquiring an original data set, wherein the original data set comprises a surface image sequence of a mud film during simulated stratum environment maintenance, a mud proportioning parameter, a stratum physical parameter and a film forming pressure value; Performing image preprocessing according to the surface image sequence, and performing partition contrast enhancement on a single frame image to inhibit specular highlight and performing self-adaptive segmentation based on image gray statistics to obtain preprocessed image data for eliminating non-uniform illumination interference; Extracting crack morphological evolution features according to the preprocessed image data, and obtaining a crack morphological evolution feature set representing crack development from punctiform initiation to linear penetration by detecting crack edge contours and carrying out morphological connection and background noise filtering on broken pixel segments; Performing prediction model construction according to the crack morphological evolution feature set, the slurry proportioning parameter, the stratum physical parameter and the film forming pressure value, and constructing to obtain a mud film life prediction model by fusing the crack space-time evolution feature and the physical field parameter and adopting a Bayes optimization strategy to fit a nonlinear mapping relation; and carrying out life calculation according to the life prediction model of the mud film and the acquired real-time maintenance data, and obtaining a mud film residual life prediction value from the current state to the failure moment of the mud film through model deduction. In a second aspect, the present application provides a slurry shield mud film maintenance test apparatus, comprising: The test box assembly comprises a test box body and a test box cover plate, wherein the test box cover plate is in sealing connection with the test box body, and a p