A thin-spray material for fracturing and swelling surrounding rock and a multi-scale evaluation optimization method for implementation effect

Abstract

The application discloses a kind of thin spray material for broken and expanded surrounding rock and multi-scale evaluation optimization method of implementation effect, by mass percentage, thin spray material includes 50%-60% composite cement powder, 20%-25% organic glue powder, 18%-22% coal-based solid waste powder, 3%-8% fine-grained sandstone, 0.5%-1.5% nanocellulose crystal and 6.0% additive;And based on the proportion of thin spray material, a kind of multi-scale evaluation optimization method of thin spray material spraying implementation effect in well is developed.The application realizes the effective bonding of coal roadway surrounding rock by the specific proportion design of thin spray material, constructs the cross-scale three-dimensional interpenetrating network structure of "rigid skeleton-flexible interface-nano enhancement" on the micro scale;And through the weight absolute value weighted calculation of multi-parameter adjustment conflict, combined with "weight-measured" two-dimensional decision matrix, the dual discrimination of data fluctuation and performance failure is realized, relying on the closed-loop iteration logic of "monitoring-computing-adjustment-re-spray", the proportion debugging cycle is reduced, and the service life of spray layer is improved.

Description

Technical Field

[0001] This invention belongs to the field of coal mine surrounding rock spraying and sealing technology, specifically involving a thin spraying material for fractured and swollen surrounding rock and a multi-scale evaluation and optimization method for its implementation effect. Background Technology

[0002] In recent years, China's social and economic development has been rapid, with coal making a significant contribution.

[0003] Because deep roadways are situated in a high-stress environment, the surrounding rock is highly susceptible to fragmentation and deformation under the combined disturbance of mining stress. Excavated roadways, due to long-term exposure to ventilation, are prone to weathering and spalling, resulting in a significant reduction in strength. Furthermore, in water-rich rock strata, fragmentation and fissures create channels for groundwater seepage, softening the rock mass and drastically reducing its strength. This often leads to disasters such as roof collapse, sidewall indentation, and severe floor heave, seriously hindering safe production operations in underground coal mines. Therefore, it is urgent to adopt active support measures to seal the surrounding rock and coordinate deformation with the surrounding rock to optimize the stress state when the roof and sidewalls are under pressure, ensuring the safe and efficient mining of deep coal resources.

[0004] Currently, thick shotcrete is widely used in underground coal mines to seal the surface of fractured and swollen surrounding rock, forming a continuous shell structure on the rock surface to integrate the fractured rock mass into a unified load-bearing body. Due to its convenient construction steps, low raw material costs, and high final strength, it is widely used in most mines in my country. However, traditional thick shotcrete, after solidification, exhibits a rigid and brittle structure with extremely low ductility. When the roadway is affected by ground pressure or mining-induced stress, the shotcrete layer cannot absorb energy through plastic deformation. The bonding surface between the shotcrete layer and the surrounding rock is prone to peeling due to stress concentration. Once the surrounding rock undergoes fracture and swelling deformation, the thick shotcrete layer, lacking flexible buffering, will accelerate overall instability, leading to sudden collapse and detachment without warning, triggering a chain reaction of roof falls, resulting in support damage and casualties. Furthermore, traditional dry-spraying processes require on-site mixing of cement and aggregates. During spraying, the high-pressure airflow carries a large amount of particulate matter, resulting in a rebound rate of 25%-40% for thick-layer sprayed concrete. The thicker the sprayed layer, the greater the rebound. Dust concentration at the work site exceeds the standard by more than 10 times. These waste materials containing accelerators accumulate in the roadway, increasing cleaning costs and continuously releasing alkaline dust, worsening the underground ventilation environment. Inhalable dust during operation remains difficult to control, and long-term operation can easily lead to silicosis. Simultaneously, micro-cracks in the sprayed layer allow groundwater to seep in, causing softening of the surrounding rock. The concrete itself cracks under wet-dry cycles, losing its sealing function, and further weathering and spalling occur after the surrounding rock is exposed. Actual measurements show that the effective lifespan of the sprayed concrete layer in water-rich roadways is less than 50% of the design value. Acidic groundwater and hydrogen sulfide gas in coal mines can dissolve the calcium components in cement stone, forming expansive products such as gypsum. The high porosity of thick-layer sprayed concrete allows corrosive media to penetrate, causing anchor (cable) corrosion and concrete pulverization, ultimately leading to the loss of support function.

[0005] The reasons why traditional thick shotcrete is prone to failure can be systematically attributed to the following: its brittle structure fractures suddenly without warning when pressure is applied to the roadway, leading to instantaneous loss of support function; the high concentration of dust generated by the dry-spraying process deteriorates the working environment, and the accumulation of rebound material exacerbates pollution; simultaneously, water seepage through microcracks accelerates weathering, and its weak resistance to acid and alkali corrosion leads to rapid pulverization and failure under the erosion of acidic groundwater and hydrogen sulfide. These three defects work synergistically to significantly shorten the effective lifespan of the support system. To overcome the above defects of thick shotcrete materials, some mines are currently trying to use modified cement-based materials for thin-layer shotcreting operations. Compared with traditional concrete, thin-layer shotcrete materials show significant advantages in sealing the surrounding rock in underground coal mines. The elongation of thin-layer shotcrete materials modified with organic components can reach 150%-200%, far exceeding the less than 5% of concrete, and its affinity with coal and rock materials is greatly improved compared with inorganic materials. When pressure is applied to the roadway, the shotcrete layer tightly bonds to the surrounding rock and absorbs energy through flexible deformation, producing only network microcracks rather than brittle collapse, significantly reducing the risk of sudden roof collapse. By adding functional additives such as reactive fillers and preservatives, a dense cementitious substance can be formed, effectively blocking the infiltration of groundwater and acidic media. Field tests show that in water-rich or acidic environments, the lifespan of a thin-layer sprayed coating can reach 8-10 years, more than 60% longer than the 2-3 years of concrete. Furthermore, by adopting a wet spraying process, the dust concentration is reduced to 20 mg / m³. 3 The following (90% reduction compared to dry shotcrete), rebound rate <10%, significantly improves the working environment and reduces the risk of silicosis. The spray layer thickness is only 5-8mm (compared to 50-150mm for thick shotcrete), reducing material usage to 1 / 20-1 / 25. Two people per shift can spray 10-15 meters, increasing construction efficiency by 3 times compared to concrete. Although the unit price is higher than concrete, the reduced usage, labor, and transportation costs result in overall cost savings of over 30%. For example, Chinese patent CN112062523B discloses a thin-film spray material for reinforcement, beautification, support, and sealing of coal mine roadways, along with its preparation and application methods. It aims to improve the overall toughness and deformation coordination of the material by introducing a portion of redispersible polymer powder and inorganic cement-based composite. However, its main component is still cement, resulting in complex processes, high costs, insufficient adaptability to surrounding rock deformation, and stringent construction precision requirements. The aging risk and deformation resistance of the thin-film spray layer in actual mine environments still need to be verified on-site.

[0006] To improve the existing thin-layer sprayed concrete (SJC) support and sealing system in coal mines and address the shortcomings of current SJCC materials such as poor toughness, easy detachment, and difficulty in solidification, a new type of SJCC sealing material is urgently needed. This material should solidify quickly, tightly bond to and seal fractured and swollen surrounding rock, possess a certain degree of toughness, coordinate with the deformation of the surrounding rock, and adapt to the autonomous stress regulation of the surrounding rock. However, the effectiveness of SJCC operations lacks high-precision real-time monitoring methods and dynamic visualization presentation techniques. Existing technologies can only make empirical inferences based on indirect data such as pressure curves, resulting in limited reliability of quantitative indicators. Furthermore, an evaluation system for the effectiveness of SJCC support has not yet been established. Existing standards often use conventional pressure index evaluation models, ignoring the unique complexity of the crack network characteristic of SJCC technology. This makes it impossible to accurately assess the dynamic process of cracking and instability of the sprayed layer, leading to a disconnect between evaluation results and actual modification effects. Therefore, a corresponding evaluation method is needed for quantitative analysis and dynamic visualization of the sealing support effect after SJCC operations, along with methods for improving and optimizing material proportions. Summary of the Invention

[0007] One of the objectives of this invention is to provide a thin-spray material for sealing shattered surrounding rock, which has strong adhesion, good wall adhesion, controllable setting time, good sealing effect, corrosion resistance, and the ability to coordinate deformation with the surrounding rock, so as to achieve effective sealing and reinforcement of shattered surrounding rock.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A thin-film spraying material for sealing shattered surrounding rock, comprising, by mass percentage: 50%-60% composite cement powder, 20%-25% organic adhesive powder, 18%-22% coal-based solid waste powder, 3%-8% fine-grained sand and gravel, 0.5%-1.5% nano-cellulose crystals, and additives totaling no more than 6.0% of the powder mass of the thin-film spraying material. The additives include 0.5%-1.0% carbon nanotubes, 0.5%-0.8% water-reducing agent, 0.5%-1.0% expanding agent, 0.3%-0.6% early-strength agent, 0.5%-0.8% thickener, 0.8%-1.0% interfacial coupling agent, and 1.0%-1.2% film-forming aid. The proportion of additives should be less than 6.0% of the total mass of the pure powder of the thin-film spraying material. The dosage of each admixture is not fixed and should be flexibly configured according to the specific geological conditions of the surrounding rock. In addition, the water-cement ratio of the mixture should be dynamically adjusted between 0.8:1 and 1.5:1 to meet the needs of different working conditions.

[0010] Preferably, the composite cement powder is prepared by blending silicate cement and sulfoaluminate cement at a mass ratio of 1.5:1. This can meet the requirements of rapid sealing support in downhole operations, ensuring rapid hydration and no expansion conflict between components.

[0011] Preferably, the organic adhesive powder is a redispersible ethylene-vinyl acetate copolymer adhesive powder with a minimum film-forming temperature of ≤5℃ and an alkali resistance of 100% film integrity after 28 days of immersion in saturated Ca(OH)2.

[0012] Preferably, the coal-based solid waste powder is obtained from coal gangue solid waste discarded during underground coal mining and fly ash captured after calcination in power plants. The coal gangue powder, after high-temperature calcination and grinding activation, and the decarbonized fly ash powder are mixed in a 1:1 mass ratio to form a mixed powder. The specific surface area of ​​the coal gangue powder after grinding and activation is ≥480±20 m². 2 / kg, activity index ≥75% at 7d, ≥95% at 28d, ≤5% residue on 45μm sieve, median D50 diameter 18±1μm, D90 particle size ≤38μm. Fly ash powder specific surface area ≥350 m². 2 / kg, activity index ≥75% for 28d, SO3 content ≤1.8%, glass microsphere content ≥70%, 45μm sieve residue ≤12%, D50 median diameter 12-15μm, D90 particle size ≤45μm.

[0013] Preferably, the fine-grained sand and gravel should be ground into fine sand particles with a particle size of <0.15mm, a mud content of ≤1.0%, and a moisture content of ≤0.5%. The nanocellulose crystals have a diameter of <15nm and an aspect ratio of >40.

[0014] Preferably, the carbon nanotubes in the additive are multi-walled carbon nanotubes with a purity ≥98%, a carboxyl content of 1.8%-2.5%, an oxygen content ≤1.5%, and a length of 10-30 μm.

[0015] Preferably, the water-reducing agent is a mixed powder prepared by compounding polycarboxylate ether high-efficiency water-reducing agent and modified lignin sulfonate at a mass ratio of 3:1, and PVP dispersant powder is added at a mass ratio of 0.3% to 0.5% of the two powders.

[0016] Preferably, the expanding agent is an ettringite-type expanding agent.

[0017] Preferably, the early-strength agent is a mixed powder prepared by compounding lithium carbonate and sodium gluconate retarder at a mass ratio of 25:1, with a purity >98% and a particle size distribution of 5-8 μm after D50 sieving. The thickener is a mixed powder prepared by compounding hydroxypropyl methylcellulose powder and pregelatinized starch at a mass ratio of 10:1, with a water dissolution time <30s.

[0018] Preferably, the interface coupling agent is a silane interface coupling agent, and the silane type is KH-792 diamino type with a purity ≥98%.

[0019] Preferably, the film-forming aid is a mixed liquid prepared by compounding dodecyl alcohol ester and propylene glycol phenyl ether at a mass ratio of 3:1, and its hydrolytic stability is <5%.

[0020] Preferably, the water-cement ratio of the thin-spray material is 1.2:1 to 1.5:1.

[0021] The second objective of this invention is to provide the preparation process and application method of the above-mentioned thin-spray material for sealing shattered surrounding rock, which has a simple preparation process and is convenient for on-site construction in coal mines.

[0022] The on-site configuration and usage methods are as follows:

[0023] (1) Inorganic powder premixing: According to the set ratio, first put the composite cement powder, coal-based solid waste powder, fine sand and gravel and nanocellulose crystals into the forced mixer. Turn on the mixer and stir at a low speed of 100-150 rpm for 5 minutes to avoid dust flying and material separation. Then add dry powder additives in sequence: multi-walled carbon nanotubes, polycarboxylate ether high-efficiency water-reducing agent mixed powder, ettringite-based expansion agent, lithium carbonate early strength agent, hydroxypropyl methylcellulose powder thickener, increase the speed to 200 rpm and stir for 5 minutes to make the powder fully and evenly dispersed to obtain a dry and uniform basic powder mixture, ensuring that there are no obvious lumps and the color is consistent (mainly gray-black), which is conducive to subsequent hydration reaction and adhesion improvement.

[0024] (2) Preparation of inorganic slurry: Slowly add 60%-70% of the total water to the mixed powder in step (1), and stir at 200-300 rpm for 5-10 minutes to form an inorganic slurry. During the water addition process, the silane interface coupling agent should be poured in slowly and evenly to avoid excessive local concentration. Adjust the mixer to 200-250 rpm and stir at a medium speed for 3-4 minutes to promote the uniform combination of water and powder. This step produces a uniform wet slurry with a certain fluidity (moderate viscosity) and no segregation or sedimentation. At this point, the slurry has a preliminary sealing effect, but the adhesion needs to be enhanced;

[0025] (3) Toughening with organic adhesive powder: After adding the remaining 30%-40% of the water, add the organic adhesive powder material. Add it in small batches, evenly, to prevent clumping. Adjust the mixer speed to 400-450 rpm and stir at high speed for 10 minutes to quickly disperse the adhesive powder and activate its bonding properties until the slurry becomes uniformly viscous. Then, evenly sprinkle in the film-forming aid and continue stirring at high speed at 400-450 rpm for another 10 minutes. This step yields a highly adhesive, fluid, and pumpable composite slurry.

[0026] (4) Coal mine underground spraying: Remove loose rocks, accumulated water, oil stains and metal fragments in the roadway, and use a high-pressure water gun (pressure ≥5MPa) to wash the surface of the surrounding rock and residual coal dust on the ground to ensure that there is no floating dust accumulation at the working face. Turn on the roadway ventilation system (wind speed ≥0.5m / s), detect the gas concentration (≤0.5%) and oxygen content (≥20%), set up work warning fences, mark the danger area and isolate it.

[0027] After the slurry is pumped into the hopper of the rotary wet spraying machine, it is stirred at a low speed (100-150 rpm) to maintain uniformity and prevent sedimentation until spraying. The wet spraying machine pumps the slurry obtained in step (3) to the nozzle to mix with compressed air (spraying pressure 0.5-0.8MPa), and the air flow rate is controlled at 8-10m³ / h. 3 / min.

[0028] The nozzle is 0.8-1.2 m away from the surrounding rock surface. The surrounding rock surface is divided into several strips (strip width ≤ 1.5 m). Within each strip, an independent Z-shaped path is executed, spraying horizontally from the bottom left starting point to the right end point. During the vertical upward movement of 50 cm, it is simultaneously sprayed horizontally to the left end point. This cycle is repeated until the top, forming a continuous "Z" shaped zigzag line. It is ensured that the next trajectory covers 50% of the width of the previous trajectory and that there are no missed areas.

[0029] During spraying, the operator should stand with feet apart, rotating at a constant speed of 15° / s around the waist, ensuring the nozzle remains perpendicular to the rock surface. A low horizontal movement of 0.3-0.5 m / s is used to ensure a coating thickness of 3-5 mm, while a vertical movement of 0.8-1.0 m / s reduces sagging after spraying. After spraying, a dense, elastic film forms on the surrounding rock surface, which, when tightly bonded, seals cracks and resists fragmentation and deformation. Immediately after spraying, the pipelines and spraying equipment should be flushed with clean water to prevent blockages.

[0030] Preferably, a forced mixer with hydraulic continuously variable transmission is used for large-scale underground engineering operations, supporting segmented speed control in both dry and wet stages. Simultaneously, a rotor-type wet spraying machine is selected for spraying, featuring dual-piston pumps and corrosion-resistant pipelines.

[0031] It should be noted that, given the heterogeneous nature of the surrounding rock geological conditions in the well, conducting pre-spraying tests before large-scale spraying, and dynamically verifying the weight of various monitoring parameters calculated in real time, as well as adjusting the material ratio and construction process, are key preliminary measures to ensure that the thin-spray support system is accurately adapted to different chemical conditions and to avoid failure risks from the source.

[0032] The third objective of the present invention is to provide a multi-scale evaluation method for the implementation effect of pre-spraying test of underground thin spraying material and the corresponding thin spraying material ratio. Core evaluation indicators including the hanging wall sag degree after spraying, the bonding force with surrounding rock, the solidification speed, the sealing effect, and the coordinated deformation capacity are obtained through on-site monitoring, so as to solve the problem of "difficult characterization" of the spraying implementation effect. In view of the fact that on-site monitoring can only obtain the final state result of the sprayed layer surface, and it is difficult to perspective the transient evolution mechanism of internal microcrack initiation, force chain fracture and interface debonding, the present invention further couples FLAC3D-PFC3D cross-scale numerical simulation, constructs the mapping relationship of "macro stress field - micro particle motion", and dynamically restores the whole life cycle process of the sprayed layer from the outside to the inside, from local damage to macro instability.

[0033] To achieve the above objectives, the present invention realizes continuous tracking of the health state of the sprayed layer and dynamic adjustment of material ratio by collecting six types of parameters in real time and combining with attention weight indicator early warning.

[0034] The specific implementation technical scheme is as follows:

[0035] S1. Obtain on-site working conditions according to the geological characteristics of the surrounding rock, and the on-site working conditions fall into 9 typical working conditions;

[0036] S2. According to the on-site working conditions, select the preset thin spraying material corresponding to the on-site working conditions, that is, the thin spraying material corresponding to the above typical working conditions, to carry out a preliminary spraying test on the surrounding rock;

[0037] S3. Collect various parameters of the sprayed layer after spraying: after the sprayed layer is initially set, collect six types of monitoring parameters synchronously, including sag amount, bonding strength, solidification time, permeability coefficient, gas leakage rate, and coordinated deformation coefficient; the specific steps are as follows:

[0038] (1) A laser range finder arranged on site combined with a scale is used to measure the sag degree after spraying. Immediately after spraying, select 3-5 test points (1m apart) on the vertical rock surface, and measure the sag distance between the bottom of the sprayed coating and the preset horizontal reference line. The test is completed within 1-2 minutes after spraying to capture the initial rheological behavior of the material. The sag degree shall be ≤5mm to be qualified;

[0039] (2) A portable bonding force tester is used to carry out on-site pull-out test. 24 hours after the spraying is cured, a pull-out head (50mm in diameter) is installed on the sprayed layer, a tensile force is applied at a constant rate (0.5MPa / s) until failure, and the peak force (MPa) is recorded. The test points cover different lithology areas (such as sandstone, mudstone, coal rock), and every 10m 2 has at least 3 measurement points. The bonding strength is qualified when the bonding to sandstone / mudstone is ≥0.8MPa and the bonding to coal rock is ≥0.7MPa;

[0040] (3) Use a Vicat needle penetration meter in conjunction with an infrared temperature recorder to test the setting time. After spraying, insert a standard needle (1 mm in diameter) into the sprayed surface and measure the needle penetration depth every 30 seconds until initial setting (needle penetration < 1.5 mm) and final setting (needle penetration < 0.5 mm). Record the initial setting time, final setting time, and ambient temperature and humidity. Repeat the test 3 times and take the average value. An initial setting time ≥ 10 min and a final setting time ≤ 20 min are considered acceptable.

[0041] (4) Water pressure permeation test and gas sealing test. After 24 hours of curing, a sealing cavity (100mm in diameter) was installed in the sprayed area, and a water pressure of 0.3MPa was applied and maintained for 30 minutes. The water pressure permeation flow rate (cm / s) after spraying was measured. Simultaneously, a micro-gas chamber was pre-embedded between the thin sprayed layer and the surrounding rock interface and filled with helium. After 24 hours, a gas sample was extracted through the reserved gas extraction hole, and the helium concentration decay was analyzed using a portable intrinsically safe helium analyzer to calculate the sealing efficiency. To ensure effective sealing of harmful gases such as methane and resistance to corrosion and water pressure after spraying, the permeability coefficient was ≤1×10⁻⁶. -6 The gas flow rate is cm / s and the sealing efficiency is ≥95% (helium leakage rate <5%) to be considered qualified;

[0042] (5) Implement loading tests and use displacement sensors to evaluate the coordinated deformation capability. After spraying, three fiber optic displacement sensors are evenly pasted on the surface of the sprayed layer along the diagonal of the loading plate. At the same time, drill holes 5 cm deep in the surrounding rock at the corresponding positions and embed the same type of sensors to avoid interference from the sprayed layer. The cyclic load adopts a trapezoidal wave with a loading frequency of 0.5 Hz and is applied in stages at a rate of 0.1 MPa / s, with each stage held for 30 s. The strain data of the sprayed layer and the surrounding rock are collected simultaneously. The ratio of the two is the coordinated deformation coefficient (%). A coordination coefficient ≥ 100% indicates that the sprayed layer can completely coordinate deformation. Taking into account the risk of sprayed layer debonding failure and coordination mismatch between the coating and the surrounding rock deformation, a coordinated deformation coefficient ≥ 90% for thin-film sprayed materials is considered qualified.

[0043] S4. Data Fusion and Bias Reduction, and Attention Weight Allocation. The Dynamic Time Warping (DTW) algorithm is used to align timestamps from multiple data sources, eliminating phase deviations caused by acquisition response delays. An attention mechanism weight allocation model is employed to eliminate misjudgments that may be caused by single parameter anomalies. Based on the sample distribution under different parameter values, the information entropy E is obtained by calculating the amount of uncertain information carried by the parameter using the entropy weight method. i The higher the value of E, the more dispersed the distribution of the parameter sampling data (the greater the dispersion), and the worse the performance stability; the information entropy E i The attention weight w is obtained after exponential normalization. i Essentially, this parameter represents its contribution weight in the failure assessment under the current operating condition, guiding the priority of subsequent material proportion adjustments. A higher weight indicates greater dispersion and uncertainty in the parameter, corresponding to a larger adjustment range for the components.

[0044] Information entropy E i The calculation depends on the specific gravity p ij Its standard mathematical definition is as follows:

[0045]

[0046] In the formula x ij is the original monitoring value of the i-th parameter in the j-th sample, and n is the total number of samples.

[0047] Calculate the information entropy E of the i-th parameter. i：

[0048]

[0049] In the formula, ln is the natural logarithm, and E is the natural logarithm when the data are completely identical (minimum uncertainty). i E approaches 0; when the data distribution is uniform (where uncertainty is greatest), E i Approaching 1.

[0050]

[0051] Information entropy E of the parameter i The higher the value, the more discriminative / predictive information this parameter carries, and the greater the attention weight w assigned to it. i (The more discrete the data, the greater the entropy value and the richer the failure discrimination information contained.) Six monitoring parameters were included: sag, bond strength, setting time, permeability coefficient, gas leakage rate, and compatibility deformation coefficient. Weights were assigned to these six parameters, and exponential normalization was used to adjust the weights of all parameters to w. i ∈(0,1), and the sum of all weights ∑w i =1. By fusing historical accident data with the weight distribution of real-time monitoring through a Bayesian network, the contribution of each parameter in the failure determination is dynamically adjusted, and the weight value of each monitoring parameter after spraying is calculated under the current surrounding rock conditions.

[0052] A 3D dynamic display of the spraying effect is provided. The PFC3D-FLAC3D coupled model is used to dynamically simulate the spraying construction, the bonding between the sprayed layer and the surrounding rock, the stress-induced deformation of the roadway rock strata after curing, and the dynamic process of sprayed layer cracking and failure. The main steps are as follows:

[0053] (1) The far-field region was established using FLAC3D to create the finite element rock strata and tunnel excavation area. PFC3D 6.0 or above was selected, and the FLAC3D module was configured in the Tools option. The far-field region used zone elements in FLAC3D to simulate the continuous medium of the rock strata, and hexahedral meshes were used. Zone face skins were used to group the faces, automatically dividing the six faces of the model into "east, west, top, bottom, north, south". The constitutive model of the rock strata was the mohr-coulumb model, and the parameters of rock density, elastic modulus, cohesion, internal friction angle, and friction coefficient were set according to the mine geological exploration data. Interface contact surfaces were set at the contact surfaces of different rock strata to avoid force transmission distortion that could lead to mesh failure and error. The gravity acceleration was set to 9.8, and the pressure of the overlying rock strata was calculated as: overlying rock strata unit weight × gravity acceleration × vertical depth of the rock strata. The calculated overlying strata pressure is applied to the "top" surface of the zone region. The lateral pressure coefficient is set to a value between 1.2 and 1.5, depending on the specific circumstances, and the lateral stress is distributed along the model's lateral direction according to a gradient. The velocity fix command is used to constrain the model's displacement boundary conditions. Static equilibrium calculations are then initiated until the local unbalanced force ratio reaches 1e. -3 Equilibrium state; the stress in the excavated area is relieved step by step using the relax excavate command, simulating the dynamic process of stress redistribution after excavation. The model is then restarted for static equilibrium calculation until the local unbalanced force ratio reaches 1e. -3 Equilibrium state;

[0054] Establish a data interaction wall between the FLAC3D far-field region and the PFC3D near-field region. After the tunnel in the FLAC3D far-field region is excavated, the wall-zone command is used to establish a connection between the zone region and the ball region, and the stress, displacement, and acceleration data calculated by FLAC3D are alternately mapped to the discrete particle region of the near-field PFC3D.

[0055] (2) Testing the basic physical and mechanical parameters of the thin-slurry material. The following parameters were measured: sag (flow distance and critical yield stress were measured to calibrate the initial viscosity and damping coefficient of the slurry); tensile strength (tensile strength, elongation at break, and elastic modulus were measured to calibrate pb_ten and pb_emod of the parallel bond model); shear strength (cohesion, internal friction angle, and shear modulus were measured to calibrate pb_coh, pb_fa, and kratio of the parallel bond model); and bonded rock sample strength (interfacial bond strength was measured to calibrate tensile_strength and coh of the Flat-Joint model). The tests include shear viscosity (measured to calibrate fluid constitutive parameters), load-bearing strength (measured peak and residual load-bearing capacity to calibrate particle contact stiffness kn and ks), and nanoindentation testing after core sampling of surrounding rock (measured nanohardness, indentation modulus, and creep compliance to calibrate micro-contact stiffness, correct constitutive parameters of hydration products (CSH, AFt), and reflect the micro-mechanical response of organic-inorganic hybrid structures). These tests provide accurate parameters for selecting contact models between particles inside thin-sprayed materials and contact models between thin-sprayed materials and rough surrounding rock.

[0056] (3) The near-field region uses PFC3D to establish a discrete element particle spraying area. The sprayed slurry is simplified into a discrete particle flow. A particle generation area is created at a distance of 1.2m from the surface of the surrounding rock of the tunnel (the width of the particle generation area is set to Generate_space). After the particles are generated, they are uniformly grouped into ball_generate before spraying. The angle of slurry spraying is set (defined as a function of time mech.time.total), and the nozzle height is the height of the surrounding rock minus the product of the spraying time and box_size. The sprayed slurry particles are given a spraying velocity in the x-direction (defined as a function of the spraying angle) and a spraying velocity in the y-direction (defined as a function of the spraying angle), and the nozzle's movement trajectory in the z-direction (defined as a function of time). Initial density and damping are also assigned to the particles. The specific implementation code is as follows:

[0057] [Generate_space=0.5]; Size of the particle generation region.

[0058] def Spray_slurry; the fish function defines the spray point.

[0059] loop foreach bp ball.groupmap("pentu")

[0060] if ball.pos.x(bp)-ball.radius(bp)<1.2 then

[0061] Spray_slurry=false

[0062] exit

[0063] endif

[0064] endloop

[0065] Spray_slurry=true

[0066] end

[0067] def Reciprocating_spraying; the fish function implements cyclic reciprocating spraying

[0068] whilestepping

[0069] if Spray_slurry then

[0070] jiaodu=math.pi*0.18*math.sin(2*math.pi*0.5*mech.time.total)

[0071] z_pos=8-box_size*mech.time.total

[0072] command

[0073] ball group "ball_generate"

[0074] ball generate group "pentu" number 100 radius 0.02 0.05 box [1.2-box_size] ...

[0075] 1.2 [-box_size*0.5] [box_size*0.5] @z_pos [z_pos+box_size]

[0076] ball attribute vel-x [50*math.cos(jiaodu)] vel-y [50*math.sin(jiaodu)] range ...

[0077] group "pentu"

[0078] ball attribute density 5e3 damp 0.03 range group "pentu"

[0079] endcommand

[0080] endif

[0081] end

[0082] The thin-spray material created in this invention contains cement hydration gel products, organic adhesive powder, nanocellulose, and carbon nanotube components. To simultaneously simulate bond strength, tensile toughness, and fracture behavior, the Parallel Bond Model is used for the internal (ball-ball) structure of the thin-spray material, and the interparticle contact parameters are calibrated using data obtained from the aforementioned experiments. Considering the strength increase after curing, a fish function is written to realize the increase in coating strength over time.

[0083] The `strength_evolution` function is defined to update the strength over time.

[0084] loop while time <total_time

[0085] local t_factor=1-exp(-0.1*mech.time.total); time factor

[0086] loop foreach contact cmat cid

[0087] contact.prop(cid, 'pb_coh')=pb_coh*t_factor

[0088] contact.prop(cid, 'pb_ten')=pb_ten*t_factor

[0089] endloop

[0090] endloop

[0091] Considering the sliding friction between the rough surrounding rock and the sprayed particles, the Flat-Joint model built into PFC3D is used for the ball-fact between the thin-layer sprayed material and the rock contact surface. This model can simulate the mechanical interlocking behavior between the peak-valley structure of the surrounding rock surface and the thin-layer sprayed material through the polygonal contact surface. The surface roughness parameters of the surrounding rock and the adhesion force between the thin-layer sprayed material and the surrounding rock, as measured by the above experiments, are used to calibrate the interparticle contact parameters.

[0092] (4) Write the fish function to apply cyclic dynamic pressure load to the boundary of the FLAC3D model to simulate the process of cracking of the sprue and debonding failure from the surrounding rock under the action of dynamic pressure disturbance. Use trapezoidal wave to simulate the periodic pressure of the rock strata (pressure increase-pressure stabilization-pressure decrease stage) to avoid numerical oscillation caused by rectangular wave.

[0093] fish define apply_cyclic_load; defines the cyclic dynamic load function.

[0094] Parameter settings (need to be adjusted according to operating conditions):

[0095] local period=10.0; Load period (seconds)

[0096] local peak_pressure=5e6; Peak pressure (Pa)

[0097] local ramp_ratio=0.3; trapezoidal wave boost / buck duration ratio

[0098] Calculate the current load value (trapezoidal wave):

[0099] local t_cycle=math.mod(global.time, period)

[0100] local t_ramp=period*ramp_ratio

[0101] local pressure=0.0

[0102] if t_cycle<=t_ramp

[0103] pressure = peak_pressure * (t_cycle / t_ramp); Pressure boosting stage

[0104] else if t_cycle<=period-t_ramp

[0105] pressure = peak_pressure; stabilizing pressure range

[0106] else

[0107] pressure = peak_pressure * (1.0 - (t_cycle - (period - t_ramp)) / t_ramp); pressure drop section

[0108] command

[0109] zone face apply stress-normal [pressure] range group 'top'

[0110] endif

[0111] endcommand

[0112] end

[0113] The model parameters were corrected and solved until convergence equilibrium was reached. Based on a comparison between the observed cracking evolution of the sprayed layer in the field and the dynamic evolution process of the coupled calculation, the contact model parameters were continuously adjusted for correction until the local unbalanced force ratio reached 1e. -3 The equilibrium state.

[0114] (5) Visualization of post-processing of coupled calculation results. By writing the fish function, the evolution law of energy dissipation of surrounding rock under dynamic pressure disturbance is displayed intuitively, and tensile cracks and shear cracks are displayed in groups, so as to clearly show the force chain breakage and crack dynamic evolution process, and evaluate the spraying effect.

[0115] fish def energy_now; Extraction of energy dissipated from surrounding rock

[0116] While stepping

[0117] loop foreach local z_p zone.list

[0118] if zone.model(z_p) is 'null'

[0119] stress_two=zone.stress.int(z_p)

[0120] stress_one=zone.stress.min(z_p)

[0121] stress_three=zone.stress.max(z_p) zone.extra(z_p,2)=math.abs(zone.vol(z_p)*1 / (2*zone.prop(z_p,'young'))*(stress_one^2+stress_two^2+stress_three^2-2*z one.prop(z_p,'poisson')*(stress_one*stress_two+stress_one*stress_three+stress_two*stress_three))-zone.extra(z_p,1))

[0122] zone.extra(z_p,3) = math.abs(zone.work.elastic.total(z_p))

[0123] zone.extra(z_p,4)=math.abs(zone.work.plastic.total(z_p));zone.extra(z_p,5)=math.abs(zone.work.elastic.total(z_p))+math.abs(zone.work.plastic.total(z_p))

[0124] endif

[0125] endloop

[0126] end

[0127] import itasca as it; Call the python library to extract tensile cracks and shear cracks respectively

[0128] import numpy as np

[0129] it.command("python-reset-state false")

[0130] def export_crack_info():

[0131] cracks = it.dfn.fracture.list()

[0132] data_tensor = []

[0133] data_shear = []

[0134] n = 0

[0135] m = 0

[0136] for crack in cracks:

[0137] if it.dfn.name(it.dfn.fracture.Fracture.dfn(crack)) == 'crack_shear_tension':

[0138] pos_x = it.dfn.fracture.Fracture.pos_x(crack)

[0139] pos_y = it.dfn.fracture.Fracture.pos_y(crack)

[0140] pos_z = it.dfn.fracture.Fracture.pos_z(crack)

[0141] data_tensor.append([n,pos_x,pos_y,pos_z])

[0142] n += 1

[0143] else:

[0144] pos_x = it.dfn.fracture.Fracture.pos_x(crack)

[0145] pos_y = it.dfn.fracture.Fracture.pos_y(crack)

[0146] pos_z = it.dfn.fracture.Fracture.pos_z(crack)

[0147] data_shear.append([m,pos_x,pos_y,pos_z])

[0148] m += 1

[0149] It should be noted that the FLAC3D-PFC3D cross-scale numerical simulation used in this invention aims to construct a visual mechanism analysis tool. Its core function is to intuitively reveal the entire dynamic evolution process of the thin spray layer under dynamic pressure disturbance, from "force chain network reconstruction - microcrack initiation and propagation - macroscopic cracking and instability," thus assisting in explaining the intrinsic mechanism of spray layer failure and the rationality of its control logic. Given the significant heterogeneity and randomness of the in-situ geological environment, the particle contact model parameters involved in the numerical simulation only need to be calibrated by referring to the macroscopic mechanical trends obtained from laboratory tests, and do not require precise quantitative matching with field measurement data. Therefore, the specific numerical indicators output by the numerical simulation, such as stress cloud maps, dissipated energy cloud maps, crack density, force chain fracture rate, and particle kinetic energy, are only used as auxiliary references for evaluating the spraying effect and not as the basis for determining whether the spray layer is qualified. The final determination of the spraying effect is still based on the parameter weight deviation and the measured physical parameter thresholds.

[0150] (8) Evaluation and Targeted Control Measures for Thin-Layer Spraying Support under Multiple Failure Factor Coupling: For the complex working conditions of thin-layer spraying support in downhole fractured surrounding rock where lithology and fracture degree are intertwined, an attention weight system for six monitoring parameters (sag w1, bond strength w2, setting time w3, water pressure permeability coefficient w4, gas leakage rate w5, and coordinated deformation coefficient w6) is dynamically constructed based on the entropy weight method. Combined with FLAC3D-PFC3D cross-scale dynamic simulation, the deterioration mechanism of the sprayed layer is revealed, thereby enabling precise control of the material ratio. Given that this invention needs to establish a universal evaluation system for the variable downhole environment, the following will present the commonly used material composition ratios under nine typical working conditions obtained from downhole pre-spraying tests in multiple mining areas, and the reference attention weight configuration scheme after dynamic calculation and normalization:

[0151] Based on the "Engineering Rock Mass Classification Standard" (GB / T 50218-2014), the "Classification Method of Surrounding Rock for Coal Mine Development and Preparation Roadways" (MT / T 1239-2025), and the geological characteristics of surrounding rock in coal mine roadways, the degree of rock fragmentation is quantified into three levels: slightly fragmented, moderately fragmented, and extremely loosely fragmented. The specific definition standards are as follows: Slightly fragmented is defined as rock mass that is intact, with undeveloped joints and fissures, and a fissure density ≤ 2 fissures / m³. 2 Rock mass integrity coefficient K v ≥0.75, Point Load Strength Index I s(50) ≥2.0MPa, with only a small amount of weathered scum on the surface of the surrounding rock and no obvious seepage or dripping; moderately fractured is defined as relatively fractured rock mass with well-developed joints and fissures, and a fissure density between 2 and 5 fissures / m³. 2 Between, the rock mass integrity coefficient K v Between 0.55 and 0.75, the point load strength index I s(50) Between 1.0 and 2.0 MPa, the surrounding rock surface contains locally loose fragments, accompanied by slight water seepage or damp patches; extremely loose and fractured is defined as extremely fractured rock mass, with highly developed joints and fissures forming a network, and a fissure density >5 fissures / m³. 2 Rock mass integrity coefficient K v <0.55, point load strength index I s(50) <1.0MPa, the surrounding rock is fragmented or loose, accompanied by obvious water seepage, water inrush (flow rate >5L / min) or coal and rock pulverization.

[0152] ① For conditions involving slightly fractured mudstone, a flexible formulation is typically used: 52.7% composite cement + 25% organic binder powder + 18.5% coal-based solid waste + 3% fine-grained sand and gravel + 0.8% nanocellulose + 5.2% additives (including 0.7% water-reducing agent, 0.8% carbon nanotubes, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.8% interface coupling agent, and 1.0% film-forming aid). Typical weightings are w6=0.25, w2=0.22, w1=0.18, w3=0.15, w4=0.10, and w5=0.10.

[0153] ② For moderately fractured mudstone, a typical formulation is 52% composite cement + 24% organic binder powder + 19% coal-based solid waste + 3.2% fine-grained sand and gravel + 1.0% nanocellulose + 5.5% additives (including 0.8% water-reducing agent, 0.9% carbon nanotubes, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interfacial coupling agent, and 1.0% film-forming aid), compounded with 0.8% KH-792 silane coupling agent. Typical weightings are w2=0.30, w6=0.25, w4=0.18, w1=0.12, w5=0.10, w3=0.05;

[0154] ③ For extremely loose mudstone conditions, a high-thixotropic formulation is typically used, consisting of 50.5% composite cement + 24% organic binder powder + 20% coal-based solid waste + 3.7% fine-grained sand and gravel + 1.5% nanocellulose + a total of 5.5% admixtures (including 0.8% water-reducing agent, 0.9% carbon nanotubes, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interfacial coupling agent, and 1.0% film-forming aid). Typical weightings are w4=0.22, w5=0.20, w2=0.18, w1=0.15, w6=0.15, and w3=0.10.

[0155] ④ For sandstone with slight fracturing, a rigid formula is typically used: 59.7% composite cement + 20% organic binder powder + 20% coal-based solid waste + 5.0% fine-grained sand and gravel + 0.5% nanocellulose + 4.8% additives (including 0.7% water-reducing agent, 0.7% carbon nanotubes, 0.7% expanding agent, 0.4% early-strength agent, 0.5% thickener, 0.8% interface coupling agent, and 1.0% film-forming aid). Typical weightings are w2=0.35, w6=0.25, w3=0.15, w1=0.10, w4=0.08, and w5=0.07.

[0156] ⑤ For moderately fractured sandstone, a toughening formula typically uses 58.7% composite cement + 20% organic binder powder + 20% coal-based solid waste + 5.0% fine-grained sand and gravel + 0.7% nanocellulose + 5.4% admixtures (including 0.8% water-reducing agent, 0.8% carbon nanotubes, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 1.0% film-forming aid), compounded with 0.5% nano-SiO2. Typical weightings are w6=0.30, w2=0.25, w1=0.15, w3=0.12, w4=0.10, w5=0.08.

[0157] ⑥ For extremely loose sandstone conditions, a puncture-resistant formula is typically used, consisting of 55.7% composite cement, 20% organic adhesive powder, 20% coal-based solid waste, 8.0% fine-grained sand and gravel, 0.5% nanocellulose, and a total of 5.4% admixtures (including 0.8% water-reducing agent, 0.8% carbon nanotubes, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 1.0% film-forming aid). Typical weightings are w1=0.30, w2=0.25, w6=0.20, w4=0.10, w5=0.08, and w3=0.07.

[0158] ⑦ For slightly fractured coal and rock conditions, a high airtight formula is typically used, consisting of 50.6% composite cement + 24% organic binder powder + 20% coal-based solid waste + 3.1% fine-grained sand and gravel + 0.8% nanocellulose + a total of 5.3% additives (including 0.7% water-reducing agent, 0.8% carbon nanotubes, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 1.0% film-forming aid). Typical weightings are w5=0.35, w2=0.25, w6=0.20, w3=0.10, w4=0.05, and w1=0.05.

[0159] ⑧ For coal and rock with moderate crushing conditions, a high-permeability formula is typically used, consisting of 50.4% composite cement, 23% organic binder powder, 20% coal-based solid waste, 3.3% fine-grained sand and gravel, 1.2% nanocellulose, and a total of 4.9% admixtures (including 0.7% water-reducing agent, 0.8% carbon nanotubes, 0.7% expanding agent, 0.4% early-strength agent, 0.5% thickener, 0.8% interface coupling agent, and 1.0% film-forming aid). Typical weightings are w2=0.35, w5=0.25, w6=0.15, w1=0.10, w4=0.08, and w3=0.07.

[0160] ⑨ For extremely loose coal and rock conditions, a paste-like formulation is typically used, consisting of 50.7% composite cement, 25% organic binder powder, 20% coal-based solid waste, 3.0% fine-grained sand and gravel, 1.3% nanocellulose, and a total of 5.0% admixtures (including 0.7% water-reducing agent, 0.8% carbon nanotubes, 0.7% expanding agent, 0.4% early-strength agent, 0.5% thickener, 0.8% interfacial coupling agent, and 1.1% film-forming aid). Typical weightings are w5=0.30, w4=0.20, w2=0.20, w1=0.10, w6=0.10, and w3=0.10.

[0161] It should be noted that the multi-scale evaluation method for the implementation effect of downhole thin-layer spraying material disclosed in this invention has the dual functions of pre-spraying test adaptation verification and formal spraying post-effect monitoring. It can be used to verify the suitability of material proportions during the pre-spraying stage, providing a basis for subsequent large-scale construction, and also to assess the health status of the sprayed layer throughout its entire lifecycle after formal spraying. The above weight values ​​represent the objective weight distribution obtained after calculation of each monitoring parameter under corresponding typical working conditions and empirical proportions, and are only used as a benchmark reference for achieving targeted control. In actual engineering applications, the material proportions should be dynamically interpolated and adjusted according to the weight deviations based on different surrounding rock geological conditions.

[0162] S5. To address the issue that complex geological conditions in coal mines make it difficult to directly match the nine established typical working conditions, the target area is initially classified into one of the nine most closely related typical working conditions based on the on-site lithology and degree of fracturing. A small-scale pre-spraying test is then conducted using the benchmark material ratio corresponding to that working condition. After the sprayed layer has initially set, six monitoring parameters are simultaneously collected: sagging, bond strength, setting time, permeability coefficient, gas leakage rate, and compatibility deformation coefficient. The real-time attention weight vector w is then calculated. i After setting the expression = (w1, w2, w3, w4, w5, w6), calculate each weight w one by one. i With the selected typical operating condition benchmark weight w i0 The deviation (i.e., w) i -w i0 / w i0The top three weights with deviations exceeding 15% were selected as core control targets. Based on the physical meaning of these three weights, and using the benchmark ratio of the selected typical working conditions as a reference, the dosage of the material composition was adjusted by linear weighted interpolation. After adjustment, the coating was applied again and monitored. If the weight deviation still exceeded 15%, the remaining weights with deviations below 15% were further extracted as secondary control targets. The material ratio was then fine-tuned, and a second re-spraying test was conducted. This verification process was repeated until all weight deviations were controlled within 15%, at which point large-scale downhole spraying operations were carried out. This ensures accurate adaptation of thin-spray support parameters under complex working conditions, avoids blind trial and error without benchmarks, and ensures that the support system is always in optimal service condition. The specific calculation process is as follows:

[0163] A quantitative mapping relationship was established between six monitoring parameters (sagging w1, bond strength w2, setting time w3, permeability coefficient w4, helium leakage rate w5, and compatibility deformation coefficient w6) and the content of each component of the thin-film sprayed material. The change in the weight of each parameter only points to the adjustment direction of a specific material component, avoiding the problem of being unable to determine addition or subtraction when multiple parameters conflict.

[0164] Table 1 Mapping Table of Monitoring Parameter Weights and Material Group Allocation Ratios

[0165]

[0166] Note: Adjustment amplitude coefficient k i The percentage adjustment for each 1% weight deviation of the component (as a percentage of the component's baseline value). The amplitude coefficient k described in this document... i The unit is % / 1%, and its value needs to be divided by 100 to convert it into a decimal slope when calculating.

[0167] The weight deviation reflects the uncertainty of the parameter under the current operating conditions. A high weight indicates that the parameter has a large dispersion and the corresponding performance is unstable, requiring an increase in the dosage of the corresponding material component to improve the performance; a low weight indicates that the parameter has tended to be stable, and the material input in this direction can be appropriately reduced (but not below the minimum limit).

[0168] Table 2. Rules for determining the direction of weight deviation and the direction of material adjustment

[0169]

[0170] The following are the specific calculation formulas for the weight of each parameter. All formulas are based on the baseline mix ratio of the current round (initially the baseline mix ratio for typical working conditions, and the mix ratio adjusted in the previous round for subsequent rounds) and are calculated incrementally.

[0171] Define the weight of the i-th parameter in round t as w. i (t), the corresponding typical operating condition baseline weight is w.i0 Relative weight deviation δ i The formula for calculating (t) is:

[0172] δ i (t) = (w i (t)-w i0 ) / w i0 ×100%

[0173] When |δ i When (t)|≤15%, the parameter weight is considered to have converged, and no adjustment is made; when δ i When (t)|>15%, this parameter is marked as the core regulatory target and enters the adjustment calculation process.

[0174] For the parameter w marked as the target point i The adjustment amount Δx for the material component x to which it is directed is calculated using the following formula:

[0175] Δx = x(t)×(k i / 100)×(δ i (t)-sign(δ i (t)×15%)

[0176] In the formula:

[0177] x(t) — The mass percentage of this component in the current proportion of the formula in round t (i.e., the base value for calculation in this round);

[0178] k i —The adjustment factor corresponding to this parameter (i.e., the percentage adjustment of the component amount corresponding to each 1% excess weight deviation).

[0179] δ i (t)-sign(δ i (t))×15%——Net deviation exceeding the 15% threshold.

[0180] The adjusted dosage of this component is: x(t+1) = x(t) + Δx

[0181] When multiple parameter weights exceed the limit simultaneously and point to the same material component (e.g., w2, w4, w5, and w6 all point to organic adhesive powder), the following rules are applied for superposition calculation to avoid directional conflicts:

[0182] If all parameters exceeding the standard for this component are adjusted in the same direction (either all increasing or all decreasing), then algebraic summation can be performed directly:

[0183] Δx 总 =Σ Δx j (j represents the index of all parameters that exceed the standard for this component)

[0184] If some parameters require an increase in a component while others require a decrease (this is extremely rare in actual operating conditions because the physical meanings of each parameter differ), then priority is assigned according to the absolute value of each parameter's weight (parameters with larger absolute weights have greater say in adjustments):

[0185] Δx 总 =Σ(w j ×Δx j ) / Σ|w j | (Weighted average)

[0186] This weighted average formula ensures that parameters with larger weights (i.e., performance directions with greater uncertainty and greater need for improvement under current operating conditions) play a dominant role in adjusting the proportions.

[0187] When more than three parameter weights exceed the limit simultaneously, a step-by-step adjustment strategy is adopted to avoid uncontrollable interaction effects caused by adjusting too many components at once. All exceeding parameters are adjusted according to |δ... i Sort the data in descending order, and select the top three as the core targets for this round. The sorting is as follows:

[0188] ① First priority: |δ i The largest parameter (with the largest weight deviation and the highest uncertainty);

[0189] ② Second priority: |δ i The second largest parameter;

[0190] ③ Third priority: |δ i The third largest parameter.

[0191] If two parameters |δ i If the measured values ​​are equal, the parameter indicating the non-compliant value is selected first; if all measured values ​​are compliant or all are non-compliant, the parameter indicating the organic adhesive powder is selected first. Then, a step-by-step adjustment strategy is adopted to adjust the ratio:

[0192] ① Step 1: Adjust the ratio only for the first three targets, and leave the other parameters that exceed the standard unchanged for the time being.

[0193] ② Step Two: Conduct a second round of spraying verification based on the adjusted mixing ratio.

[0194] ③ Step 3: In the second round, if the original first three targets have converged, no further adjustments will be made; if the original fourth and fifth targets still exceed the standard, they will be promoted to new core targets.

[0195] ④ Step 4: Adjust a maximum of three targets per round to ensure that the cause and effect of each adjustment are traceable and explainable.

[0196] ⑤ Step 5: If |δ| increases after two consecutive rounds of adjustment of a certain parameter, it indicates that the adjustment in that direction is ineffective. Then, the adjustment direction will be automatically reversed and the adjustment amount will be doubled. At the same time, the material composition corresponding to the parameter may have a nonlinear response relationship. It is recommended to combine it with FLAC3D-PFC3D simulation to assist in the analysis.

[0197] Regardless of the final result of the superposition, the final dosage of any component must not exceed the range specified in the patent claims. If the superposition exceeds the upper limit, the upper limit value shall be used; if it is lower than the lower limit, the lower limit value shall be used. This constraint is a hard constraint and takes precedence over all the above calculations. Simultaneously, the sum of the mass percentages of all components must equal 100%. When a component is adjusted, the remaining components are scaled proportionally according to their proportion in the current formulation to ensure total conservation.

[0198] Let the adjusted new doping amount of the target component be x(t+1), and the current doping amounts of the remaining n-1 components be a1, a2, ..., a n-1 The scaling factor λ is calculated as follows:

[0199] λ=(100%-x(t+1)) / (a1+a2+...+a n-1 )

[0200] The adjusted new dosages of the remaining components are: a j (t+1)=a j ×λ

[0201] This scaling only adjusts the relative proportions of non-target components and does not change the determined dosage of target components.

[0202] In practical engineering, two types of conflict situations may occur: (A) the weight deviation is within ±15% but the measured parameter value is unqualified; (B) the weight deviation exceeds ±15% but the measured parameter value is qualified. The following is a systematic conflict identification and handling logic.

[0203] In each round of calculation, in addition to the weighted deviation, the pass rate of each measured parameter is calculated independently. The measured deviation rate η is defined as follows. i :

[0204] η i =(Measured value) i -Qualification threshold i ) / Qualified threshold i ×100%

[0205] Among them, for parameters that are "the larger the better" (bond strength, coefficient of conformity to deformation), η i <0 indicates non-compliance; for parameters that are "the smaller the better" (sag, permeability coefficient, gas leakage rate), η i>0 indicates non-compliance; for "range constraint" type parameters (setting time), calculate the deviations for initial setting and final setting respectively.

[0206] ① When a certain parameter weight w i (t) and the baseline w i0 If the deviation is ≤15%, no adjustment is required according to the weighting standard. However, if the measured physical value of this parameter does not reach the qualified threshold specified in the patent, the measured value has a veto power—even if the weighting deviation is compliant, the component indicated by this parameter must still be adjusted. The adjustment amount is forcibly calculated according to the following formula:

[0207] Δx 强制 = x(t)×k i ×max(|η i |,10%)

[0208] Where max(|η i |, 10%) ensures that even if the measured deviation is small, a minimum adjustment of at least 10% is still applied to drive improvement, avoiding ineffectiveness due to too small an adjustment range.

[0209] ② When the weight deviation of a certain parameter is >15%, the ratio should be adjusted according to the weight standard. However, if the measured physical value of the parameter already meets the qualified threshold, a high weight only indicates that the monitoring data of the parameter has a large dispersion (i.e., large fluctuations in multiple measurement results), and does not necessarily mean that the current average value is unqualified. In this case, two situations need to be distinguished:

[0210] (a) Data fluctuation type: The measured mean is acceptable, but the standard deviation is large. In this case, the high weight is due to data instability rather than mean deviation. Priority should be given to improving the consistency of construction techniques (such as mixing time and spraying pressure) rather than adjusting material proportions. The material adjustment amount should be halved: Δx = 0.5 × Δx 原公式 .

[0211] (b) Occasional compliance type: The measured mean is barely qualified and close to the threshold. In this case, the high weighting indicates a potential risk of instability. The original formula should be fully adjusted to prevent problems before they occur.

[0212] Automatic discrimination formula: Calculate the coefficient of variation (CV) i =Standard deviation / Mean.

[0213] If CV i >15% → Classified as "Data Fluctuation Type" → Δx = 0.5 × Δx 原公式 (Prioritize improving construction techniques); if CV i ≤15% → classified as "occasional achievement" → Δx=Δx 原公式 (Full adjustment of material ratio).

[0214] By combining the two dimensions of weight deviation and measured value compliance, a 2×2 decision matrix is ​​formed to ensure that there is a unique adjustment path under any working condition.

[0215] Table 3. Two-dimensional decision matrix with weights and actual measurements

[0216]

[0217] S6. After each round of adjustment, using the adjusted ratio as the new baseline, spray, collect, and calculate again to enter the next iteration (t→t+1), until all |δ i | ≤ 15% and all measured physical properties are qualified. The key rule is: weighted benchmark w i0 The weighting remains unchanged (always the original baseline weight of the selected typical working condition), but the weighting baseline x(t) is updated round by round. That is, the weighting deviation in round t+1 is still relative to the original w. i0 The calculation is used to determine convergence, while the incremental calculation of the material proportions is based on the actual proportion x(t) after the previous adjustment (used to guide the new adjustment). This design ensures that the weight deviation always has a fixed convergence target (w). i0 The ratio, starting from the latest actual value, gradually approaches the optimal solution in each round. The iteration passes only if all termination conditions are met:

[0218] ①Condition 1 (Weight Convergence): The weight deviations of all six parameters satisfy |δ i ≤15%;

[0219] ②Condition Two (Measured Qualified): All six measured physical parameters meet the qualified thresholds specified in the patent (sag ≤ 5mm, bonded sandstone / mudstone ≥ 0.8MPa, bonded coal / rock ≥ 0.7MPa, initial setting ≥ 10min, final setting ≤ 20min, permeability coefficient ≤ 1×10⁻⁶). -6 cm / s, sealing efficiency ≥95%, and coefficient of deformation ≥90%);

[0220] ③Condition 3 (Variation Control): Coefficient of Variation (CV) of all parameters i ≤15% (data stability meets standards);

[0221] ④ Condition 4 (Maximum Iteration Protection): If the mixture still fails to converge after five consecutive iterations, the current mix proportion at the end of the fifth iteration will be used as the final construction mix proportion. At the same time, a failure risk assessment report will be generated, pushed to the ground monitoring terminal, and an early warning signal will be triggered, switching to the manual expert intervention mode.

[0222] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0223] 1. This invention is not a simple physical mixture of various cementitious materials, but rather, through a specific ratio design, it constructs a cross-scale three-dimensional interpenetrating network structure of "rigid skeleton - flexible interface - nano-reinforcement" at the microscale. The thin-spray material created by this invention introduces organic adhesive powder into inorganic cement-based materials for modification and toughening. After hydration, it generates ettringite (AFt) and CSH gel to provide a high-strength skeleton structure. The flexible coating formed on the surface of the high-strength skeleton after the organic adhesive powder is redispersed forms an organic-inorganic hybrid interpenetrating structure of "organic coating inorganic core," achieving effective bonding of coal and rock, possessing both the rigidity of the inorganic phase and the ductility of the organic phase. At the same time, the flexible coating of organic adhesive powder forms a hydrophobic barrier on the inner wall of the pores, blocking the penetration path of corrosive media. The volcanic ash effect of coal-based solid waste significantly consumes Ca(OH)2 in the system, reducing alkalinity and significantly weakening the risk of sulfate crystallization expansion damage and acid corrosion. Furthermore, the dense three-dimensional network constructed from nanocellulose and carbon nanotubes further enhances the chemical stability of the microstructure, extending the material's service life in humid, high-sulfur mining environments. The compounded coal solid waste powder fills micropores, reducing porosity; the active SiO2 / Al2O3 within it participates in secondary hydration reactions, generating more CASH gel, resulting in a denser structure. The addition of nanocellulose accelerates the hydration reaction rate with its hydrophilic groups, and the hydrogen-bonded network enhances the thixotropy of the slurry, reducing spray rebound. Simultaneously, its embedding in the gel network forms "fiber bridges," inhibiting microcrack propagation. Carbon nanotubes interspersed within the hydration products form a conductive network, improving the material's antistatic properties and enhancing tensile strength. This thin-film spray material, acting as a stress transfer medium, can disperse locally concentrated stress to deeper, stable rock masses during crushing and deformation in tunnels. Through flexible deformation, it absorbs some energy, slowing the failure process and preventing stress abrupt changes that could lead to rock mass buckling. It is this specific combination and ratio matching of "cement compounding to regulate the hydration process, polymer emulsion to construct a flexible interface, nanomaterials to induce interface enhancement, and industrial solid waste to fill micropores" that produces a significant synergistic effect among the components, thereby breaking through the bottleneck of traditional single cementitious materials with high brittleness and weak interfacial bonding, and realizing the improvement of the material's macroscopic mechanical properties and interfacial bonding properties.

[0224] Compared to traditional thick-film shotcrete, thin-film shotcrete uses only 20%-30% of the material, and the use of coal-based solid waste to replace cement reduces raw material costs by 30%. Furthermore, traditional concrete spraying requires high-pressure equipment, resulting in a rebound rate of 25%-40% and generating significant dust, easily leading to material waste and harming worker health. This thin-film shotcrete achieves low-pressure spraying with a rebound rate of <5% and no dust through dynamic adjustment of the water-cement ratio and flow control with a thickener. Traditional thick-film shotcrete suffers from cracking (high rigidity), poor sealing (porosity >15%), and low bond strength (0.5-1.0 MPa). This thin-film shotcrete incorporates a flexible membrane for adaptive swelling (extensibility >15%), nano-filling for density (porosity <5%), and interfacial coupling agent reinforcement, effectively sealing the swelling surrounding rock and coordinating deformation with the surrounding rock under pressure, regulating rock mass stress and preventing buckling.

[0225] 2. Currently, the traditional hammering method is often used on-site to judge whether the sprayed layer has debonded and cracked, relying on the crispness of the sound. This method is susceptible to operator experience differences, resulting in a misjudgment rate as high as 40%. Furthermore, each test requires interrupting construction and only captures a localized, instantaneous state. This invention significantly improves monitoring accuracy by collecting six types of parameters in real time through multi-source on-site monitoring and eliminating data bias through a data fusion algorithm. It employs FLAC3D-PFC3D coupled numerical modeling. FLAC3D accurately simulates the macroscopic rock mass stress field and deformation patterns, while PFC3D meticulously depicts the microscopic mechanical behavior and interface effects of the thin-layer sprayed material at the particle-level. Through a data interaction interface, a closed-loop analysis of "macroscopic stress field → microscopic particle movement → macroscopic deformation feedback" is achieved, a phenomenon that traditional single models cannot capture.

[0226] This invention overcomes the limitations of traditional thin-layer shotcrete support, which relies on trial and error based on human experience, by using a dynamic early warning and targeted control mechanism based on attention weights. It quantifies the range of mix proportion adjustments using a linear interpolation formula based on weight deviations, calculates multi-parameter adjustment conflicts by weighting absolute values ​​of weights, and achieves dual identification of data fluctuations and performance failures by combining a two-dimensional decision matrix of "weight-measurement". Relying on the closed-loop iterative logic of "monitoring-calculation-adjustment-re-spraying", it shortens the mix proportion debugging cycle under atypical working conditions by more than 50% and increases the overall service life of the shotcrete layer by more than 60%. While eliminating material waste caused by blind adjustments, it minimizes the misadjustment rate of construction quality and reduces maintenance costs by more than 70% compared to traditional thick shotcrete.

[0227] Most importantly, compared with existing technologies such as CN112062523B, this application represents a fundamental leap from "passive material improvement" to "active intelligent control." CN112062523B is limited to the physical compounding of adhesive powder and cement, and does not address the dynamic evaluation and feedback mechanisms during construction, thus failing to solve the compatibility problem in heterogeneous surrounding rock environments underground. This invention integrates "attention weight control," "FLAC3D-PFC3D cross-scale evaluation," and "weight deviation linear interpolation adjustment" to form a closed-loop material optimization system. Specifically, the attention weight mechanism quantifies the uncertainty of each monitoring parameter using the entropy weight method, identifying the core failure risk sources under the current working conditions; the cross-scale evaluation method reveals the mechanical evolution mechanism behind the weight deviation, providing a predictive basis for the adjustment direction; and the linear interpolation adjustment formula transforms abstract weight data into specific material dosage correction values, realizing a shift from "qualitative experience" to "quantitative calculation." The synergistic effect of these three elements is as follows: cross-scale simulation provides physical meaning for weight allocation, weight early warning indicates the priority for interpolation adjustment, and interpolation calculation implements the control instructions. This combined approach effectively overcomes the shortcomings of traditional methods, such as blind adjustment and delayed feedback, and achieves dynamic optimization and adaptation of the thin-spray support system throughout its entire life cycle. Attached Figure Description

[0228] Figure 1 This is a graph showing the viscosity of the thin-sprayed material prepared in Example 1 of the present invention as a function of shear rate, as measured by a rheometer.

[0229] Figure 2 This is a schematic diagram of the main components of the thin-film material prepared in Example 1 of the present invention;

[0230] Figure 3 This is an XRD analysis diagram of the phase composition and content of the thin-spray material prepared in Example 1 of the present invention before and after hydration.

[0231] Figure 4 These are scanning electron microscope images of the bonding interfaces between the thin-sprayed material prepared in Example 1 of this invention and surrounding rocks of different lithologies.

[0232] Figure 5 These are comparison images of the fracture surface of ordinary concrete material after it is broken and the thin-sprayed material prepared by this invention after it is bent.

[0233] Figure 6 This refers to the test of the sag and setting time of the thin-sprayed material prepared in Example 1 of the present invention;

[0234] Figure 7 This is a tensile strength test of the thin-film sprayed material prepared in Example 1 of the present invention;

[0235] Figure 8This is a test of the load-bearing capacity of the thin-sprayed material prepared in Example 1 of the present invention after coupling with a rock sample;

[0236] Figure 9 This is a test of the bonding strength of the thin-sprayed material prepared in Example 1 of the present invention after bonding with the rock sample matrix;

[0237] Figure 10 This is a schematic diagram of the multi-source data acquisition and processing process after downhole spraying in Embodiment 2 of the present invention;

[0238] Figure 11 This is a schematic diagram of the process of dynamically extrapolating the surrounding rock fragmentation deformation and the cracking and debonding failure of the sprayed layer in FLAC-PFC coupled numerical simulation in Embodiment 2 of the present invention.

[0239] Figure 12 This is a comparison diagram of the cracking failure of the original thick shotcrete and the treatment effect of the thin shotcrete closed section in Embodiment 2 of the present invention. Detailed Implementation

[0240] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0241] Unless otherwise specified, the raw materials used in the following implementation examples are all commercially available and readily available products.

[0242] The raw materials used in the following implementation examples all meet the following requirements: The composite cement powder is prepared by blending PO42.5R with high early strength characteristics (R type) and CSA42.5 sulfoaluminate cement with high active mineral content at a mass ratio of 1.5:1. The organic polymer powder is a redispersible ethylene-vinyl acetate copolymer powder with a minimum film-forming temperature ≤5℃ and 100% film integrity after 28 days of alkali resistance immersion in saturated Ca(OH)2.

[0243] The coal-based solid waste powder is obtained from coal gangue solid waste mined underground and fly ash captured after calcination in power plants. The coal gangue powder, which has been activated by high-temperature calcination and grinding, and the fly ash powder that has undergone decarbonization treatment are mixed into a powder at a mass ratio of 1:1.

[0244] The specific surface area of ​​coal gangue powder after grinding and activation is ≥480±20 m². 2 / kg, activity index ≥75% at 7d, ≥95% at 28d, ≤5% residue on 45μm sieve, median diameter of D50 is 18±1μm, and particle size of D90 is ≤38μm.

[0245] fly ash powder with a specific surface area ≥350 m² 2 / kg, activity index ≥75% for 28d, SO3 content ≤1.8%, glass microsphere content ≥70%, 45μm sieve residue ≤12%, D50 median diameter 12-15μm, D90 particle size ≤45μm.

[0246] The fine-grained sand and gravel should be ground into fine sand particles with a particle size of <0.15mm, a mud content of ≤1.0%, and a moisture content of ≤0.5%. The nanocellulose crystals should have a diameter of <15nm and an aspect ratio of >40.

[0247] The carbon nanotubes in the additive are multi-walled carbon nanotubes with a purity ≥98%, a carboxyl content of 1.8%-2.5%, an oxygen content ≤1.5%, and a length of 10-30 μm.

[0248] The water-reducing agent is a mixed powder prepared by compounding polycarboxylate ether high-efficiency water-reducing agent and modified lignin sulfonate at a mass ratio of 3:1, and PVP dispersant powder is added at a mass ratio of 0.3% to 0.5% of the two powders.

[0249] The expanding agent is selected as an ettringite-type expanding agent, with an SO3 content of 28%-32%, an Al2O3 content of 7%-9%, and a 45μm sieve residue fineness of <5%.

[0250] The early strength agent is made by compounding lithium carbonate and sodium gluconate retarder at a mass ratio of 25:1, with a purity of >98% and a particle size distribution of 5-8μm after D50 sieving.

[0251] The thickener is a mixed powder prepared by compounding hydroxypropyl methylcellulose powder and pregelatinized starch at a mass ratio of 10:1, with a bulk density of 0.35-0.45 g / cm³. 3 Dissolving time in water < 30s.

[0252] The interface coupling agent is a silane interface coupling agent, with the silane type being KH-792 diamino and a purity ≥98%. The film-forming aid is a mixed liquid prepared by compounding dodecyl alcohol ester and propylene glycol phenyl ether at a mass ratio of 3:1, with a hydrolytic stability <5%. The water-cement ratio of the thin-film sprayed material is 1.2:1-1.5:1. The water used should meet the requirements of the "Standard for Water Used in Concrete" (JGJ63-2006).

[0253] Example 1

[0254] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 50% composite cement powder, 20% organic adhesive powder, 22.5% coal-based solid waste powder, 7% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.8% carbon nanotubes, 0.8% water-reducing agent, 0.5% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.7% interfacial coupling agent, and 1.0% film-forming aid, prepared at a water-cement ratio of 1.3:1.

[0255] The specific preparation steps are as follows: Only small batches of thin-spray material samples were prepared for performance testing during the experiment. At least five samples were prepared for each group of experiments, and the average value was taken after removing the maximum and minimum values. A HOBART N50 vertical planetary mixer was used for mixing. Step 1: After adding composite cement powder, coal-based solid waste powder, fine-grained sand and gravel, and nanocellulose crystals, the mixture was stirred at 300 rpm for 5 minutes. Then, carbon nanotubes, water-reducing agent mixed powder, expanding agent, early-strength agent, and thickener dry powder were added sequentially. The speed was increased to 500 rpm and stirred for 10 minutes until the powder was homogeneous, grayish-black, without lumps, and had no grainy feel when rubbed. Step 2: Subsequently, 60% of the total water volume was slowly poured in with clean water, and silane interface coupling agent was added dropwise simultaneously. The mixture was stirred at 400 rpm for 10 minutes to form a smooth, paste-like, non-bleeding inorganic matrix material. Step 3: The remaining 40% of the water volume was added, and the organic adhesive powder was added in three batches, with a 2-minute interval between each batch. Stir at 700 rpm for 10 minutes until the slurry is glossy and viscous with a string length of ≥5 cm. This yields a novel organic-inorganic thin-spray sealing material.

[0256] Example 2

[0257] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 52.7% composite cement powder, 25% organic adhesive powder, 18.5% coal-based solid waste powder, 3% fine-grained sand and gravel, 0.8% nanocellulose crystals, 0.8% carbon nanotubes, 0.7% water-reducing agent, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.8% interface coupling agent, and 0.9% film-forming aid, prepared at a water-cement ratio of 1.3:1.

[0258] The preparation method is the same as in Example 1.

[0259] Example 3

[0260] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 52% composite cement powder, 24% organic adhesive powder, 19% coal-based solid waste powder, 3.2% fine-grained sand and gravel, 1.0% nanocellulose crystals, 0.9% carbon nanotubes, 0.8% water-reducing agent, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 0.9% film-forming aid, prepared at a water-cement ratio of 1.3:1.

[0261] The preparation method is the same as in Example 1.

[0262] Example 4

[0263] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 50.5% composite cement powder, 24% organic adhesive powder, 20% coal-based solid waste powder, 3.7% fine-grained sand and gravel, 1.5% nanocellulose crystals, 0.9% carbon nanotubes, 0.8% water-reducing agent, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 0.9% film-forming aid, prepared at a water-cement ratio of 1.4:1.

[0264] The preparation method is the same as in Example 1.

[0265] Example 5

[0266] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 59.7% composite cement powder, 20% organic adhesive powder, 20% coal-based solid waste powder, 5.0% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.7% carbon nanotubes, 0.7% water-reducing agent, 0.7% expanding agent, 0.4% early-strength agent, 0.5% thickener, 0.8% interface coupling agent, and 1.0% film-forming aid, prepared at a water-cement ratio of 1.2:1.

[0267] The preparation method is the same as in Example 1.

[0268] Example 6

[0269] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 58.7% composite cement powder, 20% organic adhesive powder, 20% coal-based solid waste powder, 5.0% fine-grained sand and gravel, 0.7% nanocellulose crystals, 0.8% carbon nanotubes, 0.8% water-reducing agent, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, 1.0% film-forming aid, and 0.5% nano-SiO2, prepared at a water-cement ratio of 1.2:1.

[0270] The preparation method is the same as in Example 1.

[0271] Example 7

[0272] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 55.7% composite cement powder, 20% organic adhesive powder, 20% coal-based solid waste powder, 8.0% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.8% carbon nanotubes, 0.8% water-reducing agent, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 1.0% film-forming aid, prepared at a water-cement ratio of 1.2:1.

[0273] The preparation method is the same as in Example 1.

[0274] Example 8

[0275] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 50.6% composite cement powder, 24% organic adhesive powder, 20% coal-based solid waste powder, 3.1% fine-grained sand and gravel, 0.8% nanocellulose crystals, 0.8% carbon nanotubes, 0.7% water-reducing agent, 0.8% expanding agent, 0.5% early-strength agent, 0.6% thickener, 0.9% interface coupling agent, and 1.0% film-forming aid, prepared at a water-cement ratio of 1.4:1.

[0276] The preparation method is the same as in Example 1.

[0277] Example 9

[0278] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 50.4% composite cement powder, 23% organic adhesive powder, 20% coal-based solid waste powder, 3.3% fine-grained sand and gravel, 1.2% nanocellulose crystals, 0.8% carbon nanotubes, 0.7% water-reducing agent, 0.7% expanding agent, 0.4% early-strength agent, 0.5% thickener, 0.8% interfacial coupling agent, and 1.0% film-forming aid, prepared at a water-cement ratio of 1.4:1.

[0279] The preparation method is the same as in Example 1.

[0280] Comparative Example 1

[0281] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 50.7% composite cement powder, 22.5% coal-based solid waste powder, 7% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.8% carbon nanotubes, 0.8% water-reducing agent, 0.5% expanding agent, 0.5% early-strength agent, and 0.6% thickener, prepared at a water-cement ratio of 1.3:1.

[0282] The preparation method is the same as in Example 1.

[0283] Comparative Example 2

[0284] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 60.7% composite cement powder, 18% coal-based solid waste powder, 7% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.7% carbon nanotubes, 0.7% water-reducing agent, 0.5% expanding agent, 0.4% early-strength agent, and 0.5% thickener, prepared at a water-cement ratio of 1.2:1.

[0285] The preparation method is the same as in Example 1.

[0286] Comparative Example 3

[0287] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 50% composite cement powder, 25.8% coal-based solid waste powder, 7% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.8% carbon nanotubes, 0.8% water-reducing agent, 0.5% expanding agent, 0.5% early-strength agent, and 0.6% thickener, prepared at a water-cement ratio of 1.4:1.

[0288] The preparation method is the same as in Example 1.

[0289] Comparative Example 4

[0290] A thin-film spray material for sealing shattered surrounding rock is made of the following components by weight percentage: 55% composite cement powder, 20% coal-based solid waste powder, 10.7% fine-grained sand and gravel, 0.5% nanocellulose crystals, 0.7% carbon nanotubes, 0.7% water-reducing agent, 0.5% expansion agent, 0.4% early-strength agent, and 0.5% thickener, prepared at a water-cement ratio of 1.2:1.

[0291] The preparation method is the same as in Example 1.

[0292] Comparative Example 5

[0293] A thin-film spraying material for sealing shattered surrounding rock is made of the following components by weight percentage: 52% composite cement powder, 20% coal-based solid waste powder, 7% fine-grained sand and gravel, 2.2% nanocellulose crystals, 0.7% carbon nanotubes, 0.7% water-reducing agent, 0.5% expanding agent, 0.4% early-strength agent, and 0.5% thickener, prepared at a water-cement ratio of 1.4:1.

[0294] The preparation method is the same as in Example 1.

[0295] Indoor tests were conducted to assess the basic physical and mechanical properties of the thin-film sprayed material prepared above: slurry viscosity test, setting time test, microstructure observation of adhesion after spraying with different rock types, comparison of end morphology after bending with ordinary concrete material, wall sag test, tensile strength test, bearing capacity test after application to rock samples, and adhesion strength test with rock sample matrix. The comparison results after the tests are shown in Table 4.

[0296] Table 4 Indoor test data

[0297]

[0298]

[0299] Depend on Figure 1The viscosity-shear rate curve in the slurry rheological test results shows that this thin-film spraying material is a typical shear-thinning fluid. Its viscosity-shear rate variation characteristic is suitable for the process requirements of different stages of thin-film spraying construction: the viscosity of the slurry remains at 10 at low shear rates (<0.01 1 / s). 6 The extremely high viscosity (Pa·s) ensures that the thin-film spray material maintains internal structural stability during the static waiting and low-speed conveying stages, effectively preventing solid-liquid separation and particle sedimentation, and guaranteeing uniform and stable material performance. When the thin-film spraying operation stops, the internal shear force disappears, the shear rate decreases rapidly, and the viscosity quickly returns to a high viscosity state. On the one hand, this prevents the material from flowing and sagging under its own weight after spraying, ensuring uniform spray layer thickness and stable formation, thus improving the forming quality of the thin-film spray layer. On the other hand, the high viscosity enhances the early structural strength of the material, accelerates film formation and sealing, and allows for earlier implementation of its functions of surrounding rock sealing, water blocking, and dust suppression. At high shear rates (>100 1 / s), the viscosity of this thin-film spray material decreases from 10... 6 The Pa·s drops sharply to less than 5 Pa·s. During the spraying operation, under the high shear action of the high-pressure pump of the spray gun, the material flow resistance is significantly reduced. The low viscosity under high shear not only avoids particle deposition that could cause blockage of the delivery pipe and spray gun nozzle, thus improving the stability of construction delivery, but also improves the atomization and spraying effect of the material. This makes it easier for the thin-spray material to spread evenly on the surface of the surrounding rock, improving the uniform coverage of the thin-spray layer and ultimately ensuring the construction quality of the thin-spray sealing.

[0300] Depend on Figure 2As shown in the schematic diagram of the organic-inorganic hybrid modified thin-film spray material, the core components of this thin-film spray material achieve performance gains through hydration reactions and microstructure regulation. The working principle is based on the chemical properties and hydration process of the components themselves: the CSH gel generated by the hydration of silicate cement fills the micropores of the inorganic skeleton and improves the density of the matrix; the hydration rate of sulfoaluminate cement is higher, and the generated ettringite whiskers are interspersed in the CSH gel network. The two together form a continuous rigid load-bearing skeleton. At the same time, the formation of ettringite can compensate for the shrinkage of cement hydration, avoid early shrinkage cracking of the spray layer, and ensure the initial load-bearing strength. Redispersible ethylene-vinyl acetate copolymer powder is uniformly dispersed in the inorganic slurry during the wet mixing stage. During curing, after water evaporation, the powder particles fuse and form a continuous, flexible polymer film on the surface of the inorganic skeleton, the inner walls of the pores, and at particle contact points. This film dissipates external stress by utilizing the 150%-200% elongation of the ethylene-vinyl acetate chain segments, preventing brittle fracture of the rigid skeleton. Simultaneously, it seals the capillary pores of the inorganic skeleton, improving impermeability and corrosion resistance. The surface of nanocellulose crystals is rich in hydroxyl groups, which can form hydrogen bonds with cement hydration products. On one hand, it adsorbs onto the surface of cement particles, slowing down the early hydration rate and increasing the thixotropy of the slurry to inhibit spray sagging, while simultaneously promoting the hydration reaction rate. On the other hand, it intersects in the pores of the CSH gel and ettringite skeleton, forming a nanoscale fiber bridging structure. When microcracks initiate, it can transfer stress across the crack ends, inhibiting crack propagation. The active SiO2 and Al2O3 in the coal-based solid waste powder participate in the secondary hydration reaction to generate additional CSH gel, which further densifies the skeleton structure; multi-walled carbon nanotubes are interspersed in the hydration products to form a reinforcing network, which improves the tensile strength and antistatic properties of the material and is suitable for the support requirements of fractured and swollen surrounding rock.

[0301] Depend on Figure 3XRD analysis of the phase composition and content of the thin-sprayed material before and after hydration showed that the main hydration products during the hydration process were divided into two main parts: a crystalline phase mainly composed of ettringite, a small amount of calcium hydroxide, and unhydrated cement / fly ash / silica particles, and a cementitious phase mainly composed of CSH gel and EVA polymer film. Before hydration, the material mainly consisted of crystalline phases such as quartz (44.6%), metakaolin (20.7%), alite (11.8%), belite (9.6%), and mullite (6.7%). Among them, alite and belite, as the main clinker minerals, provided the initial cementing activity, while the significant presence of metakaolin reflected the characteristic phases of coal-based solid waste after high-temperature calcination. After hydration, the original alite and belite characteristic peaks basically disappeared, indicating that they were largely consumed and transformed into gel products during the hydration reaction. At the same time, significant characteristic peaks of ettringite (14.1%) and newly formed hydrated calcium silicate (CSH, 13.4%) appeared in the spectrum. In addition, there were a small amount of inert crystalline phases that did not participate in the reaction (such as unreacted quartz and residual Portlandite). This phase transformation process directly demonstrates that the thin-sprayed material undergoes a full pozzolanic reaction and cement hydration reaction during hydration. By generating a dense amorphous gel phase (CSH) and interlocking needle-like ettringite crystals, the internal pores are effectively filled, thereby constructing a robust rigid skeleton structure and significantly improving the material's mechanical strength and sealing density.

[0302] Depend on Figure 4Microscopic electron microscopy images of the bonding interfaces between the thin-layer sprayed material and surrounding rocks of different lithologies show that at the interface between the sprayed layer and the coal / rock, the material closely conforms to the rough and uneven contours of the coal / rock. The material is fully embedded in the pores and gaps in the coal / rock surface, forming a tightly interlocking microstructure. Hydration products such as calcium silicate gel and ettringite generated during hydration uniformly fill the interfacial gaps. After forming a film with the organic adhesive powder in the system, a continuous overlap is formed at the interface. The nanoscale reinforcing components further densify the interfacial structure, resulting in a stable bonding interface. At the interface between the sprayed layer and the mudstone, the material is tightly bonded. The thin-layer sprayed material adapts well to the surface microstructure characteristics of the mudstone, achieving a continuous and uniform adhesion between the material and the mudstone surface. In this system, the active powders exert a volcanic ash effect, continuously promoting the hydration reaction to generate more hydration gel products. Organic adhesive powder forms a flexible bonding film at the interface, which, combined with the bridging effect of nanocellulose and multi-walled carbon nanotubes, further strengthens the interfacial bonding, allowing the sprayed layer and mudstone to bond as a whole, resulting in a dense interfacial bond. At the sprayed layer-sandstone interface, the sprayed material fully fills the interparticle gaps on the sandstone surface, forming a perfect interlocking and interlocking with the rough structure of the sandstone surface, resulting in a tight and continuous bond. In the composite system, hydration products continuously accumulate and fill microscopic voids at the interface, organic components form good bridging and bonding at the interface, and nano-reinforcing components further enhance the overall strength of the interface, resulting in strong interlocking force and high bonding stability.

[0303] The thin-spray material prepared by this invention synergistically imparts excellent adhesion to three types of interfaces: the hydration gel generated by the hydration of inorganic cement and ettringite provide sufficient bonding strength and densely fill the interfacial gaps; the active mineral powder continuously promotes hydration and further optimizes the interfacial pore structure; the organic adhesive powder forms a continuous adhesive film at the interface, improving the interface's adaptability and bonding toughness; and nanocellulose and multi-walled carbon nanotubes play a nano-bridging role, further strengthening the overall interfacial bonding force, ultimately achieving excellent adhesion and filling effects for three different lithologies.

[0304] Depend on Figure 5Comparative images of the fracture surface of ordinary concrete and the thin-film sprayed material prepared by this invention after bending show that ordinary concrete is a brittle material, which fractures directly under stress, with a clean and direct fracture surface, exhibiting typical brittle failure characteristics. In contrast, this material can be bent into a large-curvature arc without fracturing, indicating a significant improvement in toughness and deformation capacity. It can withstand greater deformation under external forces without failure, demonstrating a much stronger bending resistance than ordinary concrete. Ordinary concrete completely separates into two parts after fracture, losing its structural integrity. This thin-film sprayed material maintains its overall structural integrity after large deformation bending, without fracturing or separation, indicating a significantly improved crack resistance. It is less prone to brittle fracture under stress and structural deformation, maintaining structural integrity and improving structural safety and reliability. This property makes the material more suitable for engineering scenarios that require certain deformation, dynamic loads, or surrounding rock deformation (such as shotcreting of surrounding rock in shattered tunnels and slope support). It can adapt to the deformation of the base / surrounding rock without cracking or breaking, making up for the defects of ordinary concrete, which is brittle and prone to fracture and cracking. Combined with the design of fiber reinforcement and hybrid modification, it also shows that the material has achieved significant optimization in toughening and crack resistance.

[0305] Depend on Figure 6 The sag test of the thin-sprayed material showed that it did not sag significantly on walls with different roughness (all less than 5mm), indicating that the material has good rheological resistance and excellent wall adhesion. The average initial setting time of the tested materials was 15 minutes, and the average final setting time was 60 minutes.

[0306] Depend on Figure 7 The tensile strength test results of the thin-film sprayed material show that its tensile strength continuously increases with curing time: Early strength development is rapid, reaching nearly 2 MPa after about one day of curing, meeting the engineering requirements for early stress and rapid load-bearing; the strength develops steadily and follows a logarithmic growth law, maintaining strength growth with increasing curing time, and showing no signs of stagnation even after 14 days of curing, indicating good long-term performance development; the final strength reaches more than 3.5 times the initial strength, demonstrating excellent strength development potential. As curing progresses, the final tensile strength of this material can reach over 7 MPa. For geotechnical / cement-based cementitious materials, this tensile strength is already at a high level. Typically, the tensile strength of these traditional materials is only about 1 / 10 of their compressive strength, and the tensile strength of conventional ordinary concrete is only 1-5 MPa. The final tensile performance of this material is superior to most traditional similar materials.

[0307] A self-made test fixture was used to test the compressive bearing capacity of a thin-film sprayed material applied to rock samples. A 40mm × 40mm × 40mm rock sample was sandwiched between two 40mm × 40mm × 80mm blocks. A 5mm layer of the thin-film sprayed material was applied to the bottom of all three samples. After curing, a uniaxial press was used to apply pressure to the middle block, and the force-displacement curves were measured. The bearing capacity of the thin-film sprayed material was obtained through equivalent calculation.

[0308]

[0309] In the formula: F max —The maximum load applied to the intermediate block during the experiment;

[0310] b — the width of the middle active block.

[0311] The LLC value represents the ability of a thin-film sprayed material to support loose blocks. The larger the LLC value, the better the load-bearing performance of the thin-film sprayed material, and the better the load-bearing effect in actual engineering. Conversely, the smaller the LLC value, the worse the load-bearing effect.

[0312] Depend on Figure 8 The bearing capacity test results after coupling the thin-layer sprayed material with the rock sample show that, at 3 days of curing, the coupled sample already reached a peak bearing capacity of over 100 N / cm. This indicates that the thin-layer sprayed material cures quickly and can form an effective bearing structure with the rock mass in a short time, providing rapid support resistance to the surrounding rock. This is suitable for the support needs of tunnels and other rock mass engineering projects, which require timely control of early deformation and maintenance of surrounding rock stability after excavation. The peak bearing capacity at 7 days exceeds 135 N / cm, indicating that the material's strength continues to increase with age, and the overall bearing capacity after coupling with the rock sample is further improved, enabling it to continuously adapt to the gradually developing loads of the surrounding rock. After reaching the peak bearing capacity, the curve does not experience a precipitous drop in stress, but rather decreases slowly with increasing displacement. Even at the stage of large deformation, it can still maintain a high residual bearing capacity and will not suddenly become unstable. At the same time, the ultimate deformation that the 7-day sample can withstand is much greater than that of the 3-day sample, indicating that the coupled structure can maintain its bearing capacity under large surrounding rock deformation, which is consistent with the support characteristics of large deformation in soft / fractured surrounding rock, and can absorb the deformation energy of the surrounding rock to maintain the stability of the support structure.

[0313] Cylindrical rock specimens, each 50 mm high and 50 mm in diameter, were used. The rock specimens consisted of sandstone, mudstone, and coal. One end of each rock specimen was bonded to a 50 mm diameter circular iron sheet using epoxy resin. A 5 mm × 5 mm × 30 mm iron rod was then attached to the circular iron sheet. A thin-film spray material, mixed according to a fixed ratio, was then evenly applied between the two rock specimens, with a coating thickness of 2 mm. The tensile strength was tested after a 3-day curing period, and the bond strength between the spray material and the rock was calculated using the following formula:

[0314]

[0315] Where: F——Tensile force on the specimen (unit: N);

[0316] D—Diameter of the rock specimen (unit: mm) 2 ).

[0317] Depend on Figure 9 The bond strength test results between the thin-film sprayed material and rock show that the peak tensile bond strength between the material and mudstone reaches 2.33 MPa, far exceeding the 0.8 MPa specified in the "Technical Specification for Shotcrete in Coal Mines" (MT / T 1037-2017). The bond strength between the material and sandstone reaches 1.83 MPa. In the tensile test, the final failure mode was the fracture of the sandstone specimen itself; failure did not occur at the bonding interface between the material and sandstone. This indicates that the bond strength between the material and sandstone is higher than the tensile strength of the sandstone itself, the bonding interface is sufficiently stable, and the bonding effect is excellent. The bond strength between the thin-sprayed material and the coal body was measured to be 0.71 MPa, which is lower than the minimum requirement of 0.8 MPa. This is because the coal itself is a porous soft rock with extremely low strength. In the test, the coal matrix was broken first, and the measured value is actually the tensile strength of the coal body rather than the strength of the bonding interface. This result precisely shows that the thin-sprayed material has excellent wetting and bonding performance to the coal body. The strength of the bonding interface is already higher than the tensile strength of the low-strength coal body itself, and the bonding reliability is sufficient. It still belongs to the category of excellent bonding effect.

[0318] The proportions in the above implementation examples are only based on laboratory tests. After the proportions of the main components are adjusted within a given variable range, the various physical and mechanical properties still meet the usage requirements after laboratory tests.

[0319] Engineering Application Case 1

[0320] A thin-film spraying material for sealing fractured and swollen surrounding rock and a multi-scale evaluation and optimization method for spraying effect are presented. During actual downhole construction, the mix proportion needs to be adjusted according to specific surrounding rock conditions. A flowchart illustrating the mix proportion calculation and adjustment process is shown below. Figure 10 As shown, taking a section of a mine in Yan'an City, Shaanxi Province, China, with muddy rock under dynamic pressure disturbance as an example, the law of rock fragmentation and deformation of the surrounding rock in the tunnel is revealed through field data acquisition and fusion analysis and FLAC-PFC coupled numerical simulation, as well as the dynamic process of cracking, debonding and failure of the shotcrete layer, and targeted design of thin shotcrete material ratio is given.

[0321] Based on preliminary on-site assessment, the condition is classified as moderately fractured mudstone. Initial benchmark mix design (typical condition for moderately fractured mudstone):

[0322] Composite cement 52% + organic binder powder 24% + coal-based solid waste 19% + fine-grained sand and gravel 3.2% + nanocellulose 1.0% + total admixtures 5.5% (water-reducing agent 0.8%, carbon nanotubes 0.9%, expanding agent 0.8%, early-strength agent 0.5%, thickener 0.6%, interface coupling agent 0.9%, film-forming aid 1.0%), benchmark weight: w 10 =0.12, w 20 =0.30, w 30 =0.05, w 40 =0.18, w 50 =0.10,w 60 =0.25.

[0323] Select a representative tunnel section 5-10m long (with a sprayed surface area of ​​approximately 50-100m²). 2 Preliminary spraying tests were conducted, and six types of monitoring parameters were collected and attention weights were calculated. In the first round (t=1), the measured weights were w1(1)=0.18, w2(1)=0.38, w3(1)=0.04, w4(1)=0.15, w5(1)=0.09, and w6(1)=0.16.

[0324] Deviation calculation:

[0325] δ1=(0.18-0.12) / 0.12=+50%→Exceeds the standard (in the direction of sag)

[0326] δ2=(0.38-0.30) / 0.30=+27%→Exceeds the standard (bonding direction)

[0327] δ3=(0.04-0.05) / 0.05=-20%→Exceeds the standard (in the direction of solidification time)

[0328] δ4 = (0.15 - 0.18) / 0.18 = -17% → Exceeds the standard (permeability direction)

[0329] δ5 = (0.09 - 0.10) / 0.10 = -10% → Compliant

[0330] δ6=(0.16-0.25) / 0.25=-36%→Exceeds the standard (coordinates the deformation direction)

[0331] Target ranking (in descending order of |δ|): w1(+50%)>w6(-36%)>w2(+27%)>w3(-20%)>w4(-17%)

[0332] Select the first three target points: w1, w6, and w2.

[0333] Adjustment calculations for the first three targets:

[0334] (1) w1 target point (drooping degree), pointing to thickener and nanocellulose:

[0335] Thickener Δx = +0.00168%, Nanocellulose Δx = +0.00175%

[0336] (2) w6 target point (coordinated deformation coefficient), pointing to organic adhesive powder, nanocellulose, and carbon nanotubes:

[0337] Organic gum powder Δx = −0.06048%, nanocellulose Δx = −0.00126%, carbon nanotubes Δx = −0.000756%

[0338] (3) w2 target point (bonding strength), pointing to organic adhesive powder and interfacial coupling agent:

[0339] Organic adhesive powder Δx = +0.0288%, interfacial coupling agent Δx = +0.00063%

[0340] Multi-target stacking treatment of the same component:

[0341] Organic rubber powder: -0.06048% from w6 and +0.0288% from w2 → weighted average = (0.16 × (-0.060480) + 0.38 × (+0.028000)) / (0.16 + 0.38) = +0.00178370% → new value of organic rubber powder = 24.0000% + 0.001784% = 24.001784%;

[0342] Nanocellulose: −0.001260% from w6 and +0.001750% from w1 → Weighted average = (0.18×(+0.001750)+0.16×(−0.001260)) / (0.18+0.16)=+0.00033353% → New value of nanocellulose = 1.0000%+0.000334%=1.000334%;

[0343] Carbon nanotubes: only w6 → Δ = −0.000756% → new value = 0.899244%

[0344] Thickener: w1 → Δ = +0.001680% → New value = 0.6017%

[0345] Interface coupling agent: w2 → Δ = +0.000648% → new value = 0.9006%

[0346] The total adjusted new values ​​of the five components amount to 27.403671%, the total original values ​​of the unadjusted components amount to 78.1000%, and the remaining percentage is 72.596329%.

[0347] Scaling factor λ = 72.596329% / 78.1000% = 0.929530

[0348] The remaining components after adjustment:

[0349] Composite cement = 52% × 0.929530 = 48.335560%

[0350] Coal-based solid waste = 19% × 0.929530 = 17.661070%

[0351] Fine-grained sand and gravel = 3.2% × 0.929530 = 2.974496%

[0352] Water-reducing agent = 0.8% × 0.929530 = 0.743624%; Expanding agent = 0.8% × 0.929530 = 0.743624%; Early strength agent = 0.5% × 0.929530 = 0.464765%; Film-forming aid = 1.0% × 0.929530 = 0.929530%.

[0353] Using the adjusted ratio from the first round as the new baseline x(1), spray again, collect six types of parameters, calculate the weight wᵢ(2), and recalculate the deviation δᵢ(2). At this point, although w3 and w4 were not targets in the first round, they are included in the adjustment if they still exceed the standard in the second round. Repeat the above process until the iteration termination condition is met. The calculation process for each round is completely consistent with the first round, except that the baseline ratio is updated round by round.

[0354] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A thin-film sprayed material for sealing shattered surrounding rock, characterized in that, By mass percentage, it includes: 50%-60% composite cement powder, 20%-25% organic adhesive powder, 18%-22% coal-based solid waste powder, 3%-8% fine sand and gravel, 0.5%-1.5% nanocellulose crystals and additives, wherein the additives account for no more than 6.0% of the total mass of the thin-spray material powder; The additives include 0.5%-1.0% carbon nanotubes, 0.5%-0.8% water-reducing agent, 0.5%-1.0% expanding agent, 0.3%-0.6% early strength agent, 0.5%-0.8% thickener, 0.8%-1.0% interfacial coupling agent, and 1.0%-1.2% film-forming aid; The water-cement ratio of the thin-film sprayed material is 0.8:1-1.5:

1.

2. The thin-film sprayed material for sealing shattered surrounding rock according to claim 1, characterized in that, The composite cement powder is prepared by mixing silicate cement and sulfoaluminate cement in a mass ratio of 1.5:

1. The organic adhesive powder is a redispersible ethylene-vinyl acetate copolymer adhesive powder; The water-reducing agent is prepared by compounding polycarboxylic acid ether and modified lignin sulfonate at a mass ratio of 3:1, and PVP dispersant powder is added at a mass ratio of 0.3% to 0.5% of the powder of both. The expanding agent is selected as a calcium vanadium-type expanding agent; The early strength agent is prepared by compounding lithium carbonate and sodium gluconate at a mass ratio of 25:1; The thickener is prepared by compounding hydroxypropyl methylcellulose powder and pregelatinized starch at a mass ratio of 10:

1. The film-forming aid is prepared by compounding dodecyl alcohol ester and propylene glycol phenyl ether at a mass ratio of 3:

1.

3. A multi-scale evaluation and optimization method for the implementation effect of downhole thin-film spraying material, characterized in that, The specific implementation plan is as follows: S1. Obtain on-site working conditions based on the geological characteristics of the surrounding rock; S2. Based on the site conditions, select the preset material ratio corresponding to the site conditions and conduct a preliminary spraying test on the surrounding rock; S3. Collect various parameters of the sprayed layer after spraying: sag, adhesion strength, curing time, permeability coefficient, gas leakage rate, and compatibility deformation coefficient at multiple test points; S4. Data Processing: Timestamps are aligned using a dynamic time warping algorithm to eliminate phase deviations. The information entropy E of each parameter is calculated based on the entropy weight method. i The formula is: in, x ij It is the original monitoring value of the i-th parameter in the j-th sample, and n is the total number of samples; The attention weights w of each parameter are obtained by exponentially normalizing the information entropy Eᵢ. i : ； S5. Calculate the deviation δ between the attention weights of each parameter and the baseline weights corresponding to the field conditions. i The formula is: δ i (t) = (w i (t)-w i0 ) / w i0 ×100% Among them, w i0 The baseline weights are the values ​​corresponding to the on-site working conditions. When |δ i When |≤15%, it is determined that the corresponding parameter weights have converged, and no adjustment of the thin-spray material composition ratio is performed; When |δ i When |>15%, the corresponding parameter is marked as the core control target point, and the distribution ratio of the thin-spray material group is adjusted by linear weighted interpolation. S6. Apply the adjusted thin-spray material to the surrounding rock in a pilot spraying test, repeating S3 until the iteration ends.

4. The multi-scale evaluation method according to claim 3, characterized in that, In step S5, the adjustment amount Δx of the thin-spray material composition is calculated according to the following formula: Δx = x(t)×(k i / 100)×(δ i (t)-sign(δ i (t))×15%) Where x(t) is the mass percentage of the corresponding component in the current round, and k is the adjustment range coefficient corresponding to the corresponding parameter; When multiple parameter deviations simultaneously satisfy |δ i When the adjustment amounts are all greater than 15% and point to the same material component and have the same adjustment direction, the adjustment amounts are directly summed algebraically. When multiple parameter deviations simultaneously satisfy |δ i When the percentage is greater than 15% and points to the same material component but the adjustment direction is opposite, the following weighted average formula shall be used for calculation: Δx 总 =Σ(w j ×Δx j ) / Σ|w j | Where w j Let Δx be the attention weight for the j-th parameter. j The adjustment amount for each parameter is individually pointed to for the corresponding component; The adjusted proportion of the thin-film spray material shall not exceed the range specified in claim 1; if it does, the boundary value shall be taken. The sum of the adjusted proportions of the thin-film sprayed material equals 100%. After the corresponding components are adjusted, the remaining components are scaled proportionally according to their proportion in the current proportion.

5. The multi-scale evaluation and optimization method according to claim 4, characterized in that, In step S5, the mapping relationship between the weights of each parameter and the composition of the thin-spray material is as follows: The sag w1 is adjusted to target the thickener and nanocellulose crystals, with adjustment coefficients k1=0.8% / 1% and k1'=0.5% / 1%, respectively. The bond strength w2 is adjusted for both the organic polymer powder and the interfacial coupling agent, with adjustment coefficients of k2=1.0% / 1% and k2'=0.6% / 1%, respectively. The setting time w3 is adjusted to target the early-strength agent and the water-reducing agent, with adjustment coefficients of k3=0.5% / 1% and k3'=0.3% / 1%, respectively. The permeability coefficient w4 was adjusted to target organic polymer powder and film-forming aids, with adjustment coefficients of k4=0.7% / 1% and k4'=0.5% / 1%, respectively. The gas leakage rate w5 is adjusted to target the organic polymer powder and the interfacial coupling agent, with adjustment coefficients of k5=0.8% / 1% and k5'=0.5% / 1%, respectively. The coordination deformation coefficient w6 points to organic polymer powder, nanocellulose crystals and carbon nanotubes, with adjustment amplitude coefficients of k6=1.2% / 1%, k6'=0.6% / 1% and k6''=0.4% / 1%, respectively.

6. The multi-scale evaluation and optimization method according to claim 5, characterized in that, The direction of adjustment of the solidification time parameter w3 is opposite to the direction of the sign of Δx calculated by the formula: When δ3(t) > +15%, reduce the content of the early strength agent; When δ3(t) < -15%, increase the content of the early strength agent; The adjustment direction of the other five monitoring parameters is the same as the direction of the sign of Δx calculated by the formula.

7. The multi-scale evaluation method according to claim 5, characterized in that, The acceptable threshold values ​​for each parameter in step S3 are: sag ≤ 5 mm; bond strength: bonded sandstone / mudstone ≥ 0.8 MPa, bonded coal / rock ≥ 0.7 MPa; initial setting time ≥ 10 min, final setting time ≤ 20 min; permeability coefficient ≤ 1 × 10⁻⁶. -6 cm / s; Gas leakage rate < 5%; Coordinated deformation coefficient ≥ 90%; When the weight deviation |δ i If the percentage is ≤15% but the measured parameter value does not meet the acceptable threshold, a forced adjustment shall be made according to the following formula: Δx 强制 = x(t)×k i ×max(|η i |,10%) Where η i η represents the measured deviation rate. i =(Measured value) i -Qualification threshold i ) / Qualified threshold i ×100%; When the weight deviation |δ i The percentage is over 15%, but the measured parameter values ​​meet the acceptable threshold and the coefficient of variation (CV) is... i When the value is greater than 15%, it is determined to be a data fluctuation type, and the adjustment amount is half of Δx; When the weight deviation |δ i The percentage is over 15%, but the measured parameter values ​​meet the acceptable threshold and the coefficient of variation (CV) is... i If the percentage is ≤15%, it is determined to be an occasional achievement, and the adjustment amount is Δx.

8. The multi-scale evaluation method according to claim 5, characterized in that, When more than three parameter weight deviations simultaneously satisfy |δ i When the value exceeds 15%, a step-by-step adjustment strategy is adopted: All parameters exceeding the standard are set according to |δ i |Sorted in descending order, the first three parameters are selected as the core targets for this round of ratio adjustment, and the rest satisfy |δ i The parameter value of 15% will not be adjusted for the time being; Based on the adjusted ratio, a pilot spraying test was conducted. If the original fourth and fifth target points still satisfy |δ i If the percentage is greater than 15%, then it will be adjusted to a new core target. A maximum of three target points can be adjusted per round; If |δ| increases after two consecutive adjustments of a certain parameter, the adjustment direction will be automatically reversed and the adjustment amount will be doubled.

9. The multi-scale evaluation and optimization method according to claim 4, characterized in that, The iteration termination condition in step S6 is: Condition 1: The weighted deviations of all six parameters satisfy |δ i ≤15%; Condition 2: All six parameters meet the acceptable threshold. Condition 3: Coefficient of Variation (CV) of the six parameters i ≤15%; If the mix proportions fail to converge after five consecutive iterations, the current mix proportion at the end of the fifth iteration will be used as the final construction mix proportion. At the same time, a failure risk assessment report will be generated, pushed to the ground monitoring terminal, and an early warning signal will be triggered.

10. The multi-scale evaluation and optimization method according to claim 3, characterized in that, Step S3 is followed by a FLAC3D-PFC3D multi-scale numerical simulation step: (1) A far-field continuous medium model of rock strata was established using FLAC3D, and a near-field discrete particle spraying area model was established using PFC3D. The alternating mapping of stress, displacement and acceleration data between FLAC3D and PFC3D was realized through the data interaction interface. (2) The physical and mechanical parameters of the thin-sprayed material were calibrated through indoor tests, including sag, tensile strength, shear strength, bond strength, rheological viscosity and bearing capacity, as well as the nano-indentation modulus of the surrounding rock; (3) In PFC3D, the Parallel Bond Model is used to simulate the contact between particles inside the thin-sprayed material, and the Flat-Joint Model is used to simulate the mechanical interlocking behavior between the thin-sprayed material and the surrounding rock. (4) Apply cyclic dynamic pressure load to the boundary of the FLAC3D model to simulate the dynamic process of cracking and debonding failure of the sprayed layer under dynamic pressure disturbance. (5) By writing the fish function, the evolution law of energy dissipation of the surrounding rock and the propagation process of tensile cracks and shear cracks are dynamically displayed, and the spraying effect is evaluated.

Images