Vibration Safety Threshold and Control Technology for Blasting to Prevent Seawater Intrusion in Coastal Tunnel Sections Near Faults

Abstract

Coastal underground engineering projects are prone to seawater intrusion during blasting operations, posing significant risks to the safety of construction personnel and the structural integrity of the projects. To ensure the safety of blasting operations in areas at risk of seawater intrusion, this study focuses on a section of a coastal tunnel that is at risk of such intrusion. Using fracture mechanics theory and silo theory analysis methods, the minimum safe distance between the workface and the fault to prevent seawater intrusion is determined. Numerical simulations are employed to analyze the dynamic response of the surrounding rock and the attenuation of vibrations as blasting excavation progresses near the fault-controlled zone. This study also explores the impact of dynamic excavation on fault stability. By employing a regression analysis, this study establishes quantitative relationships between the amount of explosive used and the peak particle velocity (PPV) at different distances, as well as between the range of rock damage and PPV at various distances. This analysis allows for the determination of a safe PPV threshold to prevent seawater intrusion in the fault-controlled area. The accuracy of the computational model is validated using field-measured data. Finally, an optimized blasting design and strategy based on electronic detonator initiation are proposed for the control area, ensuring construction safety. This study provides theoretical and technical references for achieving safe and efficient blasting excavation in coastal underground engineering projects.

1. Introduction

During tunnel excavation using drilling and blasting methods, the instantaneous nature of explosive detonation impacts the surrounding rock mass of the tunnel, causing vibrational effects. This can result in surrounding rock damage, deterioration of the surrounding rock properties, and even safety risks such as rock bursts and spalling [1,2,3]. Near fault fracture zones, phenomena like rock sliding and stress concentration are also likely to occur. If the fault fracture zone is near a marine area, the damage caused by blasting to the surrounding rock and its connection can directly induce seawater intrusion. The occurrence of these adverse effects and risks not only directly affects construction safety but also has detrimental impacts on the stability and functionality of the underground project. Therefore, in the blasting construction of coastal underground projects, it is essential to strictly control the blasting damage and vibration of the surrounding rock to ensure construction safety and project quality.

Seawater intrusion is a phenomenon that occurs when the underground water balance changes due to geological changes or other external factors in coastal areas. Fault fracture zones are one of the common geological conditions in coastal underground projects. Therefore, when tunneling crosses fault fracture zones with abundant water, it is crucial to mitigate the effects of blasting vibration to ensure the stability of the surrounding rock and fault fracture zones, thereby preventing seawater intrusion. Additionally, in response to changes in the stability of the surrounding rock during blasting, it is effective to pre-define seawater intrusion risk control sections, determine the minimum safe distance, and promptly adjust the blasting design. These measures are crucial for ensuring construction safety. Extensive studies [4,5,6,7] have been conducted by scholars domestically and internationally on this topic, utilizing methods such as theoretical analysis, model testing, field measurement data analysis, and numerical simulation.

To determine the factors, extent, and patterns of fault impacts on tunnels primarily regarding the surrounding rock’s stability, existing studies have explored the effects of fault fracture changes, water seepage on surrounding rock stability, and the importance of grouting reinforcement for surrounding rock protection [8,9,10]. Luo et al. [11], through an orthogonal experiment using numerical simulation, analyzed the dynamic response characteristics of the surrounding rock during tunneling processes that cross the fault zone under multiple coupled factors. They pointed out that the sensitivity of the tunnel blast-induced dynamic response to faults is ranked as the fault width being greater than the surrounding rock grade, which is greater than the fault dip angle.

Research to determine the minimum safe distance for seawater intrusion risk segments involves both theoretical analysis and numerical simulation methods. A theoretical analysis uses static theoretical models, such as fracture mechanics, material mechanics, damage mechanics, the silo theory, the catastrophe theory, and the limit equilibrium theory, as well as dynamic theoretical models addressing excavation disturbance and fluid flow changes [12,13,14,15,16]. Numerical simulation methods mainly include continuous medium mechanics and numerical analysis methods, such as finite element and finite difference, and discontinuous medium numerical analysis methods, such as discrete element and smooth particle hydrodynamics [17,18]. Huang et al. [19] and Song et al. [20] analyzed the evolution process of water and mud inrush during tunnel construction through field tests and laboratory model tests. Zarei et al. [21] evaluated the critical instability states of different rock types by establishing mechanical models and derived predictive formulas for the critical safe distance of rock burst prevention plates. Li et al. [22] categorized disaster-causing structures leading to water inrush into two types: water-blocking rock body failure and filling structure instability. The difference is that the former involves complete or fractured rock bodies where failure occurs due to tensile bending, compressive shear failure, or internal fracture expansion, leading to the loss of water-blocking capability, while the latter involves rock bodies composed of filling media, with failure occurring due to local seepage or overall sliding instability.

For controlled blasting in coastal tunnels, reducing the blast vibration intensity is crucial. Currently, millisecond delay blasting technology using electronic detonators is widely applied in engineering for controlled blasting. Electronic detonators can precisely control the delay time, increase the number of detonation stages, and control the charge per stage while ensuring drilling progress, making the blasting process more precise and safer. Extensive research [23,24,25] shows that millisecond delay blasting with electronic detonators can reduce vibration effects by 30% to 60% compared to blasting using nonel detonators.

Currently, research on seawater intrusion mainly focuses on conditions without external forces, with limited studies on seawater intrusion under external forces such as blasting excavation. Due to the lack of analysis on the impacts of blasting, using research findings from conditions without external forces to guide blasting excavation and construction design in seawater intrusion control sections leads to inapplicability. Therefore, it is necessary to analyze seawater intrusion induced by blasting excavation. Accurately and efficiently estimating the zones at risk of seawater intrusion, clarifying the impact of blasting excavation on the dynamic response of the surrounding rock (fault), and predicting the range of rock damage and peak vibration induced by blasting are crucial for preventing seawater intrusion disasters caused by blasting excavation.

This paper focuses on a coastal tunnel excavation project with seawater intrusion risk. Using a theoretical analysis, numerical simulation, and field experiments, this study investigates the risk zone range for preventing seawater intrusion in the section near faults based on blasting design and determines the minimum safe distance between the workface and the fault. The aim of this study is to guide grouting operations. Additionally, this paper analyzes the quantitative relationship between charging parameters, the damage range of surrounding rock, and the peak vibration velocity, deriving a formula to calculate the damage range of surrounding rock due to blasting excavation near faults. It proposes a vibration-reducing blasting design for the control section and redefines the risk control zone range for seawater intrusion. This research will provide theoretical and technical references for the design and safe construction of underground tunnel excavations in coastal areas.

2. Engineering Background

This paper is based on a coastal tunnel project, which consists of two transportation tunnels with a total length of approximately 3 km. Due to the coastal environment, there is a significant risk of seawater intrusion during construction, necessitating strict control over blasting vibrations and damage to the surrounding rock. Additionally, excavation predominantly involves granite rock, classified as Grade I and Grade II, making rock-breaking particularly challenging. Therefore, this project is a representative coastal tunnel project and will serve as a typical case in this field. The research results have practical significance.

2.1. Geologic Situation

The project site is located in a hilly area, with the east, west, and north sides facing the sea. Based on field surveys and exploration, the lithology distribution in this area is as follows: the eastern part mainly consists of monzogranite, the central part mainly consists of gabbro, and the western part mainly consists of gneissic granite. The surface layer is primarily composed of Quaternary residual slope deposits. According to the “Specifications for Design of Highway Tunnel JTG3370.1-2018” [26], the surrounding rock in this area is generally stable, with the rock mass primarily consisting of Grade I and II surrounding rock. However, there are local structural fracture zones and dense joint zones where the rock mass quality is slightly poorer, classified as Grade V surrounding rock.

The groundwater conditions in the reservoir area are characterized by weathered faults and structural fractures, which serve as the primary storage areas for bedrock fissure water. Within the range from the surface to 20–30 m underground, reticulated fissure phreatic water is mainly distributed, with atmospheric precipitation being the primary recharge source. According to the water-bearing conditions of the rock and soil mass and the aquifer media, the groundwater types in the reservoir area include reticulated fissure phreatic water and vein fissure water. Reticulated fissure phreatic water is primarily found in the Quaternary residual slope deposits and the upper weathered crust, with water abundance determined by the actual topography. Vein fissure water is mainly distributed in bedrock fracture zones with a generally weak water abundance. The reservoir area has good natural shielding conditions, making it difficult for outer sea waves to penetrate, with wind waves being the main influence. The lowest tidal water level in the nearby sea area over the years was −3.8 m.

2.2. Fault and Jointed Zones

The geological exploration of the project area has revealed 14 fault zones and 7 densely jointed zones (WJ1 to WJ7), as shown in Figure 1a. The faults are oriented as follows: NE-trending faults F1, F2, F3, and F16; NW-trending faults F7 and F18; NWW-trending fault F8; nearly SW-trending faults F9, F12, F15, and F17; and NNE-trending faults F10 (F11), F13, and F14. Within the construction area, the abnormal zones mainly include the nearly E-W-trending fault F15, NNE-trending faults F10 (F11) and F14, and NE-trending faults F3 and F16. These fault zones intersect with the transportation tunnel. Among the densely jointed zones, WJ1, WJ2, and WJ3 have a general NE trend, which is close to the trend in the main regional structures. WJ4 and WJ7 have a NW trend, which aligns with the gradient direction of the southeast side of the excavation area where the terrain drops sharply. WJ5 and WJ6 have a NWW trend.

Among these, fault F10, which has a weaker mechanical property and is closest to the sea, has the highest potential risk for seawater intrusion. Therefore, this study selects the F10 fault as the focus area for controlled blasting design to prevent seawater intrusion. The geological structure of this section is shown in Figure 1b.

3. Control Zone for Seawater Intrusion Risk Based on Conventional Blasting Techniques in Transportation Tunnels

3.1. Conventional Blasting

The cross section of the transportation tunnel features a straight-wall three-center arch structure. The tunnel is 10 m wide and 9 m high with an arch height of 3.3 m. The section structure design is illustrated in Figure 2. The surrounding rock is primarily hard granite with a rock firmness coefficient f > 15. The excavation area is 83 m², classifying it as a large cross-section tunnel.

The conventional blasting scheme, depicted in Figure 3, involves using No. 2 rock emulsion explosives, with electronic detonators initiating blasts in segmented delays. Each blasting cycle advances by 2.8 m, and the unit explosive consumption is 1.08 kg/m³. This conforms to the safety regulations for blasting and meets the requirements for rock excavation and blasting construction.

3.2. Determining the Critical Section for Seawater Intrusion during Conventional Blasting

Under the disturbance caused by blasting excavation, the surrounding rock of the transportation tunnel will undergo significant damage. As the tunnel advances towards fault F10, the excavation workface moves closer to the fault, increasing the potential damage to the surrounding rock near the fault. When the damage extends beyond a certain range and connects with the fault fracture zone, it can induce seawater intrusion. To avoid seawater intrusion, it is necessary to pre-design grouting reinforcement areas. Thus, it is essential to determine the minimum distance from the excavation section to the fault, which is the safe threshold distance between the workface and the fault to prevent seawater intrusion. This distance will guide the grouting work. Considering the site conditions, the disturbance from blasting construction and the unloading of surrounding rock after blasting construction are taken into account. Using the calculation formulas for the damage range of the surrounding rock based on the fracture mechanics theory and silo theory, the minimum safe distance between the workface and the fault when conventional blasting induces seawater intrusion can be determined. This safe threshold distance ensures that the excavation process does not compromise the integrity of the surrounding rock and helps in planning the necessary grouting reinforcement to prevent seawater intrusion.

3.2.1. Minimum Safe Distance under Unloading Disturbance of Surrounding Rock

Current research on water inflow from fault zones mainly employs structural mechanics and damage mechanics theories to establish mechanical models. Typically, fault fracture zones and dense joint zones are treated as semi-infinite bodies. However, in reality, the length and width of fault fracture zones are finite, and the shapes and scales of different fault zones within the same project vary. Therefore, the size of the faults should also be considered. This study adopts a calculation model based on the silo theory, limit equilibrium theory, and catastrophe theory [12,27,28], as shown in Equation (1). This method simplifies the fault into a three-phase material consisting of solid, liquid, and gas. The scenario where the fault is orthogonal to the tunnel excavation direction (as shown in Figure 4) is simplified into a rock plug model. A silo model is established to calculate the crustal stress and analyze the minimum safe distance from the excavation face to the fault zone when shear failure occurs in the surrounding rock. The expression is as follows:

$$D_f={\displaystyle \frac{\left(\gamma R_h\left(\sin \theta -\mu l\cos \theta /2R_h\right)/\mu \right)\left(1-\exp \left(-\mu Kz/R_h\right)\right)+\gamma l\cos \theta +{\gamma }_{w}h}{4\left(\sum {\gamma }_i{H}_i\tan \psi +c\right)}}{D}_0,$$

(1)

where γ is the volumetric weight of the rock mass in the fault fracture zone, kN/m³. R_h is the hydraulic radius of the cross-section, R_h = 0.5bl/(b + l); b is the width of the fault fracture zone; and l is the length of the fault fracture zone, m. θ is the dip angle of the fault fracture zone. μ is the friction coefficient. K is the lateral pressure coefficient. z is the burial depth of the transportation tunnel, m. γ_w is the volumetric weight of the water in the fault fracture zone, kN/m³. γ_i is the volumetric weight of the overlying soil layer in the fault fracture zone, kN/m³. H_i is the thickness of the overlying rock layer, m. ψ is the saturated internal friction angle of the surrounding rock. c is the saturated cohesion of the surrounding rock, kPa. D₀ is the equivalent diameter of the transportation tunnel, m.

This model allows for the assessment of the critical safe distance by considering the mechanical properties of the surrounding rock and the stress conditions, thereby guiding the design of grouting reinforcement areas to prevent seawater intrusion during tunnel excavation near faults. Based on the engineering geological and hydrological conditions, the parameter values are as follows: γ = 26 kN/m³, R_h = 2 m, μ = 0.1, l = 4 m, K = 1.2, z = 160 m, γ_w = 9.8 kN/m³, γ_i = 28 kN/m³, θ = 90°, H_i = 160 m, ψ = 30°, c = 123 kPa, and D₀ = 10.7 m. Using these values, the calculated minimum safe distance from the excavation face to the F10 fault under unloading disturbance is D_f = 2.05 m.

3.2.2. Minimum Safe Distance under Blasting Dynamic Load Disturbance

During tunnel blasting construction, full-face blasting is performed, involving group hole delay initiation. The disturbance effect on the surrounding rock accumulates with each blasting cycle. Therefore, a coupled model of the fractured rock mass, blasting stress, and seepage based on the fracture mechanics theory is used. This model analyzes the dynamic stress intensity factors of the crack’s compressive shear failure under a blasting dynamic load, determining the minimum safe distance near the fault under the disturbance of cyclic blasting [29], as shown in the following formula:

$$D_{bdl}=r_a\cdot \exp \left[\left(H_{rp}^{1-a}-H_{rap}^{1-a}\right)/\left(1-a\right)m_b^{2a-1}\right],$$

(2)

where r_a is the equivalent excavation radius, m. H_rap is the stress within the plastic zone distributed on the surrounding rock, MPa. α is the attenuation coefficient, generally taken as 3. m_b is the empirical parameter of the rock mass. H_rp is the uniform load acting on the tunnel periphery after blasting excavation, MPa.

Based on the site construction parameters, the equivalent excavation radius r_a is determined to be 5.4 m. The plastic zone stress on the surrounding rock is approximately equal to the initial pressure of the surrounding rock, H_rap, which is 10.0 MPa. According to the design parameters of the blasting scheme for non-controlled sections of the tunnel, the uniform load acting on the tunnel periphery after blasting excavation, H_rp, is calculated to be 6.2 MPa. The surrounding rock structure in the tunnel excavation area is intact with slightly weathered surface structures. Combining the quantified Geological Strength Index (GSI) chart, the GSI value for this area is determined to be 90, with a corresponding m_b value of 0.5. Finally, the minimum safe distance from the excavation face to the F10 fault under blasting excavation is calculated to be D_bdl = 2.55 m.

3.2.3. Control Zone Range for Seawater Intrusion Risk Using Conventional Blasting

Based on the previously determined minimum safe distances under two conditions, the minimum safe distance D for conventional blasting near the F10 fault is the maximum value of the two conditions, which is 2.55 m.

$$D=\max \left(D_f,D_{bdl}\right),$$

(3)

Therefore, the seawater intrusion risk control zone is delineated, as shown in Figure 5. Since the control zone should include the minimum safe distances at both ends of the fault as well as the fault width, the grouting reinforcement zone for the actual surrounding rock is approximately 9.1 m. Otherwise, in the fault-controlled zone, the appropriate control of blasting schemes and standards should be applied in both the fault-controlled zone and the reinforcement zone. This will prevent excessive damage to the surrounding rock, thereby mitigating potential seawater intrusion risks and ensuring the safe and orderly progress of the project.

4. Vibration Safety Threshold for Blasting Excavation Near Fault F10

Due to the complexity of detonation wave propagation and the dynamic response of the surrounding rock, existing instruments can only monitor limited dynamic response data such as the vibration velocity and displacement in real time within the blast zone. They cannot achieve the dynamic monitoring and analysis of full-field blast stress and damage. Therefore, numerical simulation is an effective method to address these issues. By establishing a three-dimensional computational model under blasting loads, the entire process of tunnel blasting construction can be accurately simulated to study the characteristics and changing patterns of the dynamic response of the surrounding rock.

When blasting operations approach the reinforcement zone, it is necessary to dynamically monitor the damage range of the surrounding rock caused by blasting excavation and propose control measures to ensure that the damage remains within a safe range. In practical engineering, the blasthole acoustic testing method is generally used to detect the damage range of the surrounding rock [30,31]. However, this method is not suitable for predicting surrounding rock damage before blasting, and its cumbersome operational procedures are not conducive to regular engineering use.

Thus, it is necessary to find indicators that can characterize the blasting damage range of the surrounding rock, which can be easily applied in practical engineering to guide the dynamic adjustment of the blasting design for the seawater intrusion control zone. There is a correlation between surrounding rock damage and rock strain energy. Rock strain is proportional to the peak blasting vibration velocity, indicating a close relationship between the peak blasting vibration velocity and rock blasting damage. On construction sites, the waveform of the blasting vibration velocity is easier to obtain. Therefore, by studying the quantitative relationship between the vibration velocity and damage range, it becomes easier to characterize the damage to the surrounding rock caused by blasting vibration. This approach can guide the dynamic adjustment of the on-site blasting plan to ensure construction safety.

4.1. Modeling

4.1.1. Establishing the Computational Model

To ensure safety during blasting excavation near fault F10, it is crucial to determine the safe vibration velocity threshold. This section outlines the process used to determine this threshold, specifically focusing on fault F10. Based on the contour map, a three-dimensional topographic map was drawn. Combined with the stratigraphic and lithological data of the fault-controlled zone, a three-dimensional computational model was established using the stratum structure method. As shown in Figure 6, the modeling area includes four strata and five lithologies: heavily weathered gabbro, moderately weathered gabbro, slightly weathered gabbro, slightly weathered monzonitic granite, and the fault itself. The tunnel crosses slightly weathered gabbro and the fault. Considering boundary effects, the physical model dimensions were set at 240 m in length, 280 m in width, and 260 m in height. The tunnel is located 160 m below the highest point of the surface. The F10 fault is 4 m wide, orthogonal to the tunnel, and spans the entire computational model with a dip and strike of 90°. Using the HYPERMESH v2019 software, the model was discretized into a mesh. The maximum mesh element size was set to 6.8 m, while the finite element mesh element size near the tunnel was refined to 1.5 m. The model consisted of approximately 1.87 million hexahedral mesh elements. In the model, the exposed surface boundary was set as a free boundary condition, and the inner boundary of the rock mass was set as a non-reflective boundary. Calculations were performed using the LS-DYNA v11.0 software.

4.1.2. Material Parameters

The RHT material model [32] was chosen as the constitutive model for the rock. The RHT model takes into account pressure hardening, strain hardening, and damage effects, and it can describe the changes in the material’s state from the elastic phase to the damage and failure phases. This allows for the simulation of the dynamic response and damage fracture process of the rock under blasting loads. The damage accumulation process for the material is as follows:

$$D_r={\displaystyle \sum \Delta {\epsilon }_{\mathrm{pl}}/{\epsilon }_{\mathrm{p}}^{\mathrm{failure}}},$$

(4)

$${\epsilon }_{\mathrm{p}}^{\mathrm{failure}}=D_1{(P^{*}-P_{\mathrm{spall}}^{*})}^{D_2}\ge {\epsilon }_{\mathrm{f}}^{\min },$$

(5)

where ${\epsilon }_{\mathrm{p}}^{\mathrm{failure}}$ is the plastic strain at failure; ${\epsilon }_{\mathrm{f}}^{\min }$ is the minimum failure strain; D₁ and D₂ are the damage constants; P* is the normalized pressure of uniaxial compression strength; and $P_{\mathrm{spall}}^{\ast }$ is the normalized spallation strength. D_r is the rock damage range, which can be viewed using history variable #4 in the post-processing step.

The mechanical parameters of the rocks are shown in

Table 1. Mechanical parameters of rocks.

Rock Formation	Elastic Modulus (GPa)	Poisson’s Ratio	Density (Kg/m3)	Internal Friction Angle (°)	Cohesive Forces (MPa)
Strongly weathered gabbro	5.76	0.25	2620	30.4	3.37
Moderately weathered gabbro	6.04	0.25	2810	33.8	3.74
Slightly weathered gabbro	59.92	0.21	2860	53.2	10.23
Slightly weathered diorite	26.89	0.18	2630	50.2	9.18
Fault 10 (F10)	0.90	0.30	2530	20.29	0.71

.

4.1.3. Equivalent Blast Load (EBL) Method

Tunnel excavation and rock breaking result from the application of blasting loads to the rock mass through a series of blastholes. In numerical simulations, the large number of blastholes and their small sizes complicate the modeling process and significantly extend the computation time. To enhance computational efficiency, this study adopts the equivalent blast load (EBL) method [33,34,35]. In this approach, the load is applied to the equivalent elastic zone boundary, which corresponds to the tunnel’s contour surface. By utilizing the Chapman–Jouguet theory of condensed explosive detonation waves and the principle of waveform superposition [36], the attenuation pattern of the blasting load as it propagates through the rock mass is calculated. This allows for the creation of a full-process blasting load curve that is applied to the tunnel excavation contour surface, thus simplifying the simulation and improving computational efficiency. This method avoids the establishment of blastholes and has strong applicability for the numerical simulation of tunnel blasting.

(a)

Rock breaking mechanism under blasting loads

In the actual construction process of tunnel blasting excavation, strip charges and blastholes are usually in a decoupled charging construction. In this scenario, the peak load is related to the coupling coefficient between the explosive and the blasthole. According to explosion dynamics, the pressure on the blasthole wall—i.e., the load pressure transmitted from the blasthole explosion to the rock mass—can be calculated using the following equation [1,37,38]:

$$P_1={\displaystyle \frac{{\rho }_0{D}^2}{2(\gamma +1)}}{a}^{2\gamma }{l}_e{n}_p,$$

(6)

where P₁ is the peak equivalent blast loads (PEBLs) of a strip charge; ρ₀ is the density of explosives, kg/m³; D is the detonation velocity of explosives, m/s; γ is the adiabatic index of explosive detonation products, which can generally be approximated as 3; a is the radial decoupling coefficient between the explosive and the blasthole, a = d_c/d_b, with d_c being the charge diameter and d_b being the blasthole diameter; and l_e is the axial loading coefficient. n_p is the pressure amplification coefficient when the explosive blast products expand and collide with the blasthole wall, typically taken as 10.

After detonation, a stress wave is generated and propagates through the rock mass, as shown in Figure 7. The explosive detonation causes varying degrees of damage in the rock mass. Based on the propagation characteristics of stress waves in the near-field blasting zone, the area is divided into an impact wave zone and a stress wave zone. The degree of damage further categorizes these zones into a crushed zone, a fractured zone, and an elastic zone. In the elastic zone, the generated seismic waves may cause damage to the auxiliary structures of the rock mass.

According to the Mises criterion [33,34], if the stress intensity at any point within the rock exceeds the rock’s failure strength under uniaxial stress conditions, the rock will fail. The generation of the crushed and fractured zones is related to the mechanical properties of the rock. The formulas for calculating the radius of the crushed zone and the fractured zone can be derived as follows:

$$\left\{\begin{array}{l}{r}_c={\left(P_1{\sigma }_i/{\sigma }_{cd}{\sigma }_r\right)}^{1/\alpha }{r}_b\\ r_f={\left(P_1{\sigma }_i/{\sigma }_{td}{\sigma }_r\right)}^{1/\beta }{r}_b\\ {\sigma }_i={\sigma }_r{\left[{\left(1+b\right)}^2+\left(1+b^2\right)-2\mu {\left(1-b\right)}^2\left(1-\mu \right)\right]}^{1/2}/\sqrt{2}\end{array}\right.,$$

(7)

where σ_i is the stress intensity at any point within the rock; σ_cd and σ_td are the dynamic uniaxial compressive strength and tensile strength of the rock, respectively. σ_r is the radial stress within the rock; b is the lateral stress coefficient; α is the attenuation coefficient of the explosive shock wave in the rock for the crushed zone; β is the attenuation coefficient of the explosive stress wave in the rock for the fractured zone; r_b is the blasthole radius; and μ is the rock’s Poisson’s ratio.

The attenuation of blast-induced stress waves within the rock mass primarily occurs in the elastic zone. As these stress waves propagate through the rock, their intensity diminishes following a power law exponential decay, which is described by the following relationship [39]:

$$P_R=P_1/{\overline{r}}^n,$$

(8)

where P_R is the peak stress value at a distance R from the blast source; $\overline{r}$ is the relative distance, defined as $\overline{r}=R/d_c$; and n is the stress attenuation index, which, for rock types with high integrity, is defined as n = 2 ± μ/(1 − μ). The positive sign in the formula corresponds to the shock wave propagation region, while the negative sign corresponds to the stress wave propagation region. Given that the surrounding rock of the tunnel in this study is primarily granite, which is well consolidated and exhibits high integrity, this formula is appropriate for the conditions analyzed in this research.

(b)

Calculation of PEBLs on excavation contour for different blastholes

In analyzing the load attenuation laws of multi-hole blasting in a tunnel, the effect of temporary cavities on the attenuation process of a single-hole blasting load in the rock mass can be neglected. However, the formation of these cavities significantly influences the redistribution of the blast load. Thus, the analysis should focus on the geometric relationship between the cavities, the excavation contour surface, and the distribution of blastholes. Based on the function of the blastholes, the tunnel’s cross-section blastholes can be classified into three categories: cut holes, easer holes, and contour holes.

In the context of tunnel blasting, cut holes play a crucial role by creating the initial free surface necessary for subsequent blasting operations. During the blasting process, cut holes are subjected to strong rock confinement under single free surface conditions. To ensure that the blasting energy effectively breaks the rock within the range of the minimum resistance line, the wedge cut method is typically employed. This method is particularly effective in generating a new free surface. As illustrated in Figure 8a, the elastic boundary of the cut hole zone is used to define the boundary of the wedge cut zone. Within this zone, the cut holes generate transverse cracks that eventually intersect, leading to the ejection of the central rock mass and the formation of a temporary cavity. This results in the creation of an effective elastic boundary for the propagation of stress waves during the blasting process. For wedge-shaped cut hole blasting, the boundary of the cut zone can be approximated as the effective elastic boundary of the stress wave propagation generated by cut hole blasting. Therefore, there are four assumptions for calculation:

a)

The mutual influence between adjacent cut holes is neglected.

b)

The blasting process is approximated by treating the cylindrical explosive charge as if it is detonating within a semi-infinite medium.

c)

Cylindrical explosives undergo continuous charging.

d)

There is no consideration of the load attenuation in the minimum resistance line area.

As shown in Figure 8a, to solve the PEBL expression for the energy transfer from the cut holes to the excavation contour surface, the first step is to calculate the blast load on the equivalent elastic zone boundary of the cut holes. Then, according to Formula (8), the stress wave load on the equivalent elastic zone boundary is translated to the excavation contour surface. Finally, the PEBL for the cut holes on the excavation contour surface is obtained. Similarly, the PEBLs for the easer holes and contour holes on the excavation contour surface can be determined. The schematic diagrams of their equivalent boundaries are shown in Figure 8b,c. The calculation method for the PEBL of easer holes is consistent with that of cut holes, with the primary difference being the load influence coefficient during group hole detonation. However, contour holes serve a special function; they are characterized by dense spacing and small explosive charges and have the purpose of controlling the shaping of the excavation contour surface. Typically, a disturbed zone of 30 to 50 cm of rock is formed in this area under the influence of the blast load, creating a fractured zone. This zone provides a space and pathways for the expansion of gasses generated by the contour hole blasts, causing the rock layer to gradually detach and ultimately form the excavation contour surface, leaving behind half-hole marks on the contour. In engineering practice, the proportion of half-hole marks left by contour holes is also an important indicator for evaluating the effectiveness of smooth blasting. Therefore, it can be approximated that the equivalent elastic zone boundary of the contour hole blast load coincides with the excavation contour surface. The expressions for the PEBLs of these three types of blastholes are summarized in

Table 2. The parameters of the emulsion explosive.

Blasthole	PEBL *
	Equivalent Elastic Zone Boundary	Excavation Contour Surface
Cut hole	$P_R=k_c{P}_1{\left(r_b/r_c\right)}^{2+\mu /\left(1-\mu \right)}{\left(r_c/r_f\right)}^{2-\mu /\left(1-\mu \right)}$	$P_{cut}=P_R{\left(r_e/r_q\right)}^{2-\mu /\left(1-\mu \right)}$
Easer hole	$P_R=k_e{P}_1{\left(r_b/r_c\right)}^{2+\mu /\left(1-\mu \right)}{\left(r_c/r_f\right)}^{2-\mu /\left(1-\mu \right)}$	$P_{ear}=P_R{\left(r_e/r_q\right)}^{2-\mu /\left(1-\mu \right)}$
Contour hole	—	$P_{con}=\left(n_p{L}_b/L_p\right)P_1$

* where k_c and k_e represent the load influence coefficients for the group detonation of cut holes and easer holes, respectively. r_e is the distance from the center of the blast zone of the blastholes with the same delay time to the equivalent elastic zone boundary; r_q is the minimum distance from the center of the blast zone of the blastholes with the same delay time to the excavation contour surface. n_p is the number of contour holes detonated in the same delay time. L_b is the perimeter of the blasthole, and L_p is the length of the line connecting the centers of the contour holes. The meanings of the other symbols remain consistent with their previous definitions.

.

Regarding the selection of the load influence coefficients for the group detonation of cut holes and easer holes, researchers have suggested that it is related to the number of blastholes and the arrangement method [40,41]. According to the stress wave superposition calculation method for the simultaneous detonation of multiple blastholes, the coefficient is determined by calculating the equivalent surface as the cylindrical surface formed along the blasthole axis in the rock mass at the equivalent elastic zone boundary. This coefficient is given by the ratio of the surface area of the cylindrical surface formed by multiple blastholes to the surface area of the cylindrical surface at the equivalent elastic boundary, as shown in the following equation:

$$\left\{\begin{array}{l}{k}_{cut}=n_{cut}{S}_b/S_{cut}=\pi r_b{h}_b{n}_{cut}/\left[\left(\pi r_c+l_1+l_2\right)H\right]\\ k_{ear}=n_{ear}{S}_b/S_{ear}=\pi r_b{h}_b{n}_{ear}/\left[\left(\pi r_c+l_3+l_4\right)H\right]\end{array}\right.,$$

(9)

where n_cut and n_ear represent the number of cut holes and number of easer holes detonated in the same delay time, respectively; H is the excavation advance per round; and l₁ to l₄ correspond to the specific lengths shown in Figure 8. k_cut and k_ear are coefficients whose values depend on the number of cut holes and the arrangement of the holes. These coefficients align with the results from existing studies.

(c)

Blasting load time history curve

In numerical simulations, the blasting load curve applied to the excavation contour surface is a function of load versus time. The most commonly used form is currently the “triangular” shape. To obtain the time history curve of the blasting load, the key is to determine the time at which the PEBL occurs and the time at which the load decreases to zero. With these values, the functional expression of the blasting load time history curve can be derived.

$$P\left(t\right)=\left\{\begin{array}{l}0 \left(t=0\right)\\ PBEL \left(t=t_r\right)\\ 0 \left(t=t_s\right)\end{array}\right.,$$

(10)

Then, the loading and unloading times for a triangular blasting load can be calculated using the following empirical formula [42]:

$$\left\{\begin{array}{l}{t}_r=12\sqrt{r^{2-\mu }}{Q}^{0.05}/K\\ t_s=84\sqrt[3]{r^{2-\mu }}{Q}^{0.2}/K\end{array}\right.,$$

(11)

where K is the bulk modulus of the rock mass; Q is the amount of explosive charge in the blasthole.

In summary, the EBL curve for the entire blasting design can be derived. The expression is as follows:

$$P\left(t\right)=P_{cut}\left(t\right)+P_{ear}\left(t+\Delta t_1\right)+\cdots P_{ear}\left(t+\Delta t_i\right)+P_{con}\left(t+\Delta t_{i+1}\right),$$

(12)

where Δt is the delay time of the detonator.

Based on the above discussion, transportation tunnel blasting uses a No. 2 rock emulsion explosive with a density of 1100 kg/m³, a detonation velocity of 3500 m/s, and an explosive diameter of 32 mm. Using Formulas (1) and (2), the time history curves of the EBL for each segment of the blastholes under the conventional blasting scheme can be calculated, as shown in Figure 9.

4.2. Model Feasibility Verification

To validate the accuracy of the model and parameters, a verification experiment was conducted. In front of the excavation workface in the model, three blast vibration monitoring points were arranged at positions identical to those used in the actual site blasting process. Monitoring point 1 (1#) is located 86 m from the excavation workface. The distance between each of the three monitoring points is 20 m. The layout of the monitoring points is illustrated in Figure 10a. As shown in Figure 10b, the TC-4850 vibration meter (Manufactured by Chengdu Zhongke Measurement and Control Co., Ltd., Chengdu, Sichuan, China.) was used for on-site blast vibration monitoring. Three-directional vibration velocity sensors were utilized, which are capable of accurately recording blast vibration signals.

Figure 11 illustrates the comparison between the measured and simulated vibrations at monitoring point 1.

Table 3. Statistics of simulated PPV and measured PPV at each measurement point.

Monitoring Point	Distance from Explosive Source (m)	Direction	Simulated PPV (cm/s)	Measured PPV (cm/s)	Relative Error (%)
1#	86	X	1.642	1.743	5.74
		Y	1.611	1.903	14.80
		Z	1.274	1.412	9.77
		Resultant	2.466	2.208	11.68
2#	106	X	1.642	1.45	13.24
		Y	1.441	1.298	11.02
		Z	0.982	0.930	5.59
		Resultant	1.892	1.617	17.01
3#	126	X	0.813	0.822	1.09
		Y	0.946	0.792	19.44
		Z	0.868	0.766	13.32
		Resultant	1.269	1.089	16.53

presents the PPVs from both the numerical simulations and field measurements at all three monitoring points. The results from the numerical simulations are generally slightly higher than the actual field measurements. For monitoring point 1, the maximum relative error between the simulated and measured peak vibration velocity occurs in the vertical direction with an error of 14.80%. For monitoring point 2, the maximum relative error is in the radial direction, amounting to 13.24%. For monitoring point 3, the maximum relative error appears in the vertical direction with an error of 19.44%. The relative errors at monitoring points 1 and 2 are both less than 14.80%. At monitoring point 3, the relative error exceeds 15%. It is evident that the accuracy of the blast vibration calculations improves with increasing proximity to the blast source. Overall, the comparison between the simulated and measured data shows that while there are some discrepancies, the numerical simulations provide a reasonable approximation of the actual blast vibrations. The relative errors observed, particularly those below 20%, affirm that the simulation model can reliably predict blast vibrations for engineering applications [43]. This validation underscores the model’s utility in practical scenarios, especially in tunnel excavation projects.

4.3. Simulation Plans

Based on the blasting physical model near fault F10, the excavation workface was set at the junction between the control zone near the fault and the surrounding rock reinforcement control zone. This setup is illustrated in Figure 12. Using conventional blasting schemes for tunnel construction as a foundation, six different blasting models (Models I to VI) were designed with varying amounts of explosive charges. The single-hole charge amount for each section of the blasthole was increased by 0.2 kg increments. All other conditions remained constant.

To assess the impact of varying explosive charges on the tunnel excavation process, the charge amounts for each section of blastholes in Models I to VI were substituted into the PEBL calculation formula. The PEBLs for the six models were computed based on the specific charge amounts. The results are presented in

Table 4. Blast parameters for Models I to VI.

No.	Blasthole Type	Model I		Model II		Model III		Model IV		Model V		Model VI
		Qm * (kg)	P(MPa)	Qm * (kg)	P(MPa)	Qm * (kg)	P(MPa)	Qm * (kg)	P(MPa)	Qm * (kg)	P(MPa)	Qm * (kg)	P(MPa)
1	Cut holes	70.2	42.6	72.8	43.7	75.4	45.3	78.0	46.8	80.6	47.9	83.2	48.8
2	Easer holes (Row 1)	54.6	20.8	57.2	21.8	59.8	22.6	62.4	23.2	65.0	23.8	67.6	24.3
3	Easer holes (Row 2)	21.0	18.5	22.0	19.3	23.0	20.2	24.0	20.7	25.0	21.1	26.0	21.6
4	Easer holes (Row 3)	4.2	2.1	4.4	2.2	4.6	2.3	4.8	2.4	5.0	2.5	5.2	2.6
5	Easer holes (Row 4)	21.0	17.9	22.0	18.8	23.0	19.6	24.0	20.1	25.0	20.5	26.0	20.9
6	Floor holes (Row 1)	46.2	18.8	48.4	19.7	50.6	20.6	52.8	21.1	55.1	21.6	57.2	22.0
7	Easer holes (Row 5)	18.9	16.4	19.8	17.1	20.7	17.9	21.6	18.3	22.5	18.8	23.4	19.3
8	Contour holes	28.8	15.1	33.6	16.6	38.4	17.6	43.2	18.6	48.0	19.6	52.8	20.5
9	Floor holes (Row 2)	4.2	1.9	6.2	2.0	8.2	2.1	10.2	2.2	12.2	2.3	14.2	2.4

* Q_m is the maximum explosive charges per delay.

. Using Formula (12), the action times for each section of blasting load for the six models were determined. This provided the full-time history curves of the EBLs for each model. The calculated EBLs were then applied to the respective excavation sections of Models I to VI to simulate the blasting conditions under different explosive charges. This simulation aimed to mimic the actual tunnel construction scenarios and evaluate the structural responses.

4.4. Analysis of Blast Vibration Attenuation Patterns in Tunnels

In the simulation model of tunnel excavation, five measurement points were strategically placed along the central line at the bottom of the transportation tunnel. These points are designated as P1 through P5. The five points are positioned at intervals of 10 m from each other. P1 is located closest to the workface at a distance of 10 m. The arrangement for P2 to P5 was designed to monitor the propagation of the blasting effects on the tunnel structure at different distances from the workface. The specific layout of these measurement points is illustrated in Figure 13.

To study the quantitative relationship between the explosive charge and peak particle velocity (PPV), this paper combines the Sadovsky formula with the processing of the calculated data [44]. The Sadovsky formula is widely used to describe the attenuation pattern of blast-induced vibrations. Numerous studies have shown that the particle velocity caused by explosive energy is closely related to factors such as the charge weight, the distance from the measurement point, the properties of the rock and soil, site conditions, and the blasting method. Typically, the prediction and analysis of particle velocity are carried out using the Sadovsky formula, which is generally expressed as follows:

$$PPV=k{\left(Q^{1/3}/R\right)}^{\alpha }=k{q}^{\alpha },$$

(13)

where Q is the explosive charge per delay; R is the distance from the measurement point to the blast source; k is a constant related to site conditions, rock and soil properties, and the blasting method; and α is the velocity influence coefficient related to the blasting site and charge structure.

The simulation results show the PPV at various measuring points under different explosive charge conditions, as illustrated in Figure 14. It can be observed that when the explosive charge is the same, the PPV at the measurement points decreases as the distance from the blast center increases. Conversely, when the distance from the blast center is the same, the PPV at the measurement points is positively correlated with the explosive charge.

The PPV curves for each measurement point are shown in Figure 15. As the explosive charge increases, the vertical and composite PPVs at the P1 measurement point show the largest increase, rising by 2.2 cm/s and 2.8 cm/s, respectively. The P4 measurement point’s vertical PPV is more sensitive to changes in explosive charge. From Model I to Model VI, the single-hole charge increased by approximately 22.9%, resulting in the vertical and composite PPVs at this measurement point increasing by 17.02% and 20.91%, respectively. Since the greater the distance from the blast center, the smaller the PPV, the base value of the PPV at more distant points is relatively small. Therefore, when the explosive charge increases, the increase in the PPV is also smaller.

Using the least squares method, a regression analysis was performed on the PPV data from various measurement points under different blasting designs. The attenuation patterns of the PPV at each measurement point were obtained, as shown in Figure 16. By fitting the data using linear regression and validating it through the Sadovsky formula, fitting equations for the vertical and composite directions were derived. The coefficient of determination R² for these directions were found to be 89.3% and 96.6%, respectively. This indicates a significant correlation between the simulated PPV, the explosive charge, and the distance from the blast center. The corresponding expressions are as follows:

$$\left\{\begin{array}{l}PP{V}_Z=90.92{\left(Q^{1/3}/R\right)}^{1.13}\\ PP{V}_{\mathrm{Re}}=57.97{\left(Q^{1/3}/R\right)}^{0.67}\end{array}\right.,$$

(14)

4.5. Relationship between PPV and Surrounding Rock Damage

4.5.1. Determination of Surrounding Rock Damage

Damage to surrounding rock occurs when the blasting action causes cracks to form or existing cracks to propagate and expand, resulting in a reduction in the rock mass’s elastic modulus and deterioration of its mechanical properties. Under the influence of blasting loads, cracks in the tunnel’s surrounding rock can be visible. However, changes in the internal structure of the surrounding rock that have not yet resulted in visible cracks, or damage within the unexcavated surrounding rock, cannot be directly observed. In practical engineering, the range of damage to the surrounding rock caused by blasting is generally determined using the blasthole acoustic wave testing method. The principle behind this method is that blasting alters the structure of the surrounding rock, leading to the deterioration of the rock mass’s mechanical properties. As a result, the speed and time of acoustic wave propagation through the altered rock structure change, allowing for the determination of the damage range in the surrounding rock. To assess the extent of damage to the surrounding rock, damage mechanics is employed. The relationship between the elastic modulus of the surrounding rock and the acoustic wave is used to introduce a damage variable D_v (where 0 < D_v < 1). The greater the value of D, the more severe the damage. The calculation formula for the damage variable is as follows [45,46]:

$$D_v=1-E_e/E_0=1-{\left(v_1-v_0\right)}^2=1-{\left(1-\eta \right)}^2,$$

(15)

where E_e is the elastic modulus of the rock mass before blasting, Pa. E₀ is the elastic modulus of the rock mass after blasting, Pa. v₁ is the velocity of the acoustic wave measured in the rock mass before blasting, m/s. v₀ is the velocity of the acoustic wave measured in the rock mass after blasting, m/s. η is the rate of change in the longitudinal wave velocity before and after blasting.

According to the “Technical specification for excavation construction of rock-foundation of hydraulic structures” (SL47-2020) [47], a 10% reduction in the longitudinal wave velocity before and after blasting is considered the minimum critical value for determining rock mass damage. By substituting the critical damage threshold η = 10% into the formula, the damage variable D_v is calculated to be 0.19. This means that when the damage variable D_v > 0.19, the rock mass is considered to have sustained damage.

4.5.2. An Analysis of the Impact of Explosive Charge on Damage to the Surrounding Rock

Using the LS-PREPOST 4.6 software, the results of the excavation models in the fault-controlled zone with different explosive charges were analyzed. The surrounding rock damage cloud diagrams for Models I to VI were extracted using the History Variable #4 keyword. The damage range in the rock mass was determined based on the damage variable D_v > 0.19. The surrounding rock damage ranges for the excavation models with different explosive charges (Models I to VI) are shown in Figure 17.

In the damage cloud diagrams in Figure 15, it is evident that the maximum damage to the surrounding rock generally occurs at the arch and the floor of the tunnel. For maximum section charges of 45 kg and 47 kg, the extent and severity of surrounding rock damage are relatively small. The maximum damage ranges for each model are 2.4 m, 2.6 m, 3.1 m, 3.8 m, 4.5 m, and 5.4 m, respectively. The relationship between the surrounding rock damage range and the maximum explosive charges per delay is shown in Figure 18. The coefficient of determination R² for the linear fit is 98.8%, and the fitted relationship is given by the following equation:

$$D_s=0.036Q_m+0.191,$$

(16)

where D_s is the predicted value of the surrounding rock damage, m. Q_m is the maximum explosive charges per delay, kg.

This indicates that the growth rate of the damage range increases with the explosive charge. Furthermore, the surrounding rock’s damage range is positively correlated with the maximum section charge. Therefore, by adjusting the maximum section charge in the blasting plan, the damage range of the surrounding rock caused by excavation blasting can be controlled.

4.5.3. Regression Analysis

On blasting sites, monitoring data for vibration intensity are relatively easy to obtain. Therefore, the blast-induced vibration intensity is commonly used to assess the impact of blasting on different types of buildings, structures, facilities, and other protective objects. However, the degree of damage to the surrounding rock is often difficult to monitor on-site. Numerous researchers [48,49,50,51] have conducted studies on the damage to surrounding rock caused by blasting through extensive field experiments, monitoring, and numerical simulations. These studies have established that at a constant blast radius, the damage range in the rock mass is exponentially related to the PPV. This allows the PPV to be used to assess the extent of damage to the surrounding rock after blasting, facilitating its application on-site. The relationship is expressed as follows:

$$v=C{e}^{w{D}_s},$$

(17)

where C and ω are regression coefficients.

Also, to analyze this relationship using linear regression, we first take the natural logarithm of both sides of the equation. After transformation, we let y = lnv, a = w, x = D_s, and b = lnC, resulting in the equation y = ax + b. The linear regression fit results for the relationship between surrounding rock’s damage range and the PPV and the fitting equations for each measurement point are shown in Figure 19. The coefficients of determination R² for these fitting equations are quite high, respectively, namely 98.1%, 94.7%, 98.5%, 97.0%, and 95.1%, indicating the strong reliability of the relationship. Notably, measurement point P3 has the highest coefficient of determination, and its regression equation is

$$v=2.25e^{0.37D_s}$$

(18)

Compared to other measurement points, the PPV at P3 more effectively characterizes the surrounding rock’s damage range caused by the blast. Therefore, the PPV at P3, located 30 m from the blast center, is chosen as the control basis. Using the safety threshold for the PPV at this point and Equation (16), the surrounding rock’s damage range during blasting near fault F10 can be predicted.

4.6. Determination of PPV Safety Threshold in Seawater Intrusion Risk Control Zones

During the design phase, numerical simulation methods can be used to study the relationship between the extent of damage to the surrounding rock caused by blasting excavation and the PPV at the workface. However, in actual construction, the PPV generated by blasting is typically obtained through on-site monitoring. Due to the hazards posed by flyrock during blasting, monitoring the PPV at the excavation face or nearby workfaces is highly challenging and dangerous. In the previous section, it was concluded that the relationship between the surrounding rock’s damage range and the PPV measured 30 m from the workface has the highest fitting reliability. Therefore, considering practical construction factors, it is reasonable to use the damage and PPV fitting formula for the point 30 m from the workface as a basis for assessment. This allows for the calculation of the PPV 30 m behind the workface when the surrounding rock’s damage range reaches the fault, which serves as the PPV safety threshold for blasting in fault-controlled zones.

It has previously been established that the minimum safe distance to prevent seawater intrusion during conventional blasting near fault F10 is 2.55 m. If the surrounding rock’s damage range exceeds 2.55 m, it will directly intersect with the fault. Thus, it is essential to ensure that the blast-induced surrounding rock’s damage range is less than 2.55 m. Substituting this value into Equation (18) yields a PPV safety threshold of v = 5.50 cm/s at a distance of 30 m behind the workface. This value will serve as the PPV safety threshold for preventing seawater intrusion in fault-controlled zones, guiding subsequent controlled blasting designs.

5. Controlled Blasting Techniques in the Fault-Controlled Zone

The primary objective of controlled blasting in fault-adjacent sections is to manage the extent of damage to the surrounding rock caused by blasting loads, thereby preventing seawater intrusion and ensuring construction safety. By establishing reasonable blasting vibration safety threshold criteria, the impact of blasting vibrations on the stability of the surrounding rock can be accurately assessed. These criteria can also be used to guide the design and optimization of blasting plans in controlled sections.

5.1. The Design Process for Controlled Basting

Previously, we defined the risk control area for seawater intrusion, established the blasting vibration safety threshold, and discussed the conventional blasting plans for the transportation tunnel. Based on this information, the blasting plan for the fault-adjacent excavation control sections will be dynamically adjusted. Using the minimum safe distance calculation method, the seawater intrusion risk control zone will be redefined under different optimized blasting plans. Three-dimensional computational models will be created for each optimized plan. By comparing the PPV at 30 m behind the workface with the established safety threshold, continuous optimization will be carried out. The goal is to determine an efficient excavation blasting design that meets the safety threshold requirements for vibration velocity. Ultimately, this approach will enable dynamic blasting design for seawater intrusion control zones. The design process is illustrated in Figure 20.

5.2. Optimized Blasting Plan Using Electronic Detonator

To address the issue of rock mass damage exceeding control thresholds when using conventional blasting techniques in fault-adjacent sections, and to prevent the risk of seawater intrusion caused by the connection of blast-induced cracks with fault zones, a vibration reduction blasting design using electronic detonators for hole-by-hole detonation is adopted. This approach aims to maintain high-efficiency tunnel excavation while mitigating the risk of seawater intrusion in fault-controlled sections by controlling blasting vibrations and minimizing damage to the surrounding rock.

Electronic detonators have been widely used for reducing ground vibration in tunnel blasting. However, the delay time of electronic detonators needs to be set manually, and the reasonable time parameters can give full play to its advantages of high precision and safety. Moreover, the core parameters of the initiation circuit for tunnel blasting using electronic detonators are the delay interval between blastholes (Δt_p) and the delay interval between delay groups (Δt_k), from which the delay time series of all EDs can be obtained.

To determine the delay parameters for electronic detonators, we proposed a calculation method based on the single-hole blasting field test and waveform superposition theory. This method was successfully applied in tunnel blasting engineering, yielding excellent results. The calculation process is illustrated in Figure 21. First, the measured vibration waveform of single-hole blasting was fitted to a continuous waveform in the time domain by Fast Fourier Transform. The effective section of the fitted waveform was identified and intercepted as the basic waveform for superposition by the time-energy density method. Second, the multi-hole superposition waveform with different delay intervals was obtained according to the waveform superposition theory. By analyzing the variation law of the particle peak velocity of the superposition waveform, the optimal delay intervals could be determined. For detailed calculations, please refer to references [25,52,53], as this paper will not elaborate on them.

Using the calculation method, the delay interval between blastholes (Δt_p) was determined to be 4 ms, and the delay interval between delay groups (Δt_k) was determined to be 30~40 ms. The sequential hole detonation plan is shown in Figure 22. In the blasting plan, the locations of the holes and the charge are the same as those in conventional blasting. The cut holes utilized a hole-by-hole detonation method with a delay interval of 4 ms between blastholes. The delay interval between the cut holes and easer holes was set to 40 ms to ensure that the explosive energy of the cut holes was fully utilized. For the other groups, the delay interval between delay groups was set to 30 ms.

5.3. Experimental Results and Analysis

The measured blast vibration waveform at 30 m from the workface is shown in Figure 23. In the figure, it can be observed that the maximum resultant velocity is 5.35 cm/s, which is slightly below the safety threshold of 5.5 cm/s. The peak occurs at 202.6 ms, corresponding to the easer holes rather than the cut holes. Therefore, in subsequent construction, the delay time of the detonators in this group can be adjusted based on the timing of the PPV occurrence to ensure hole-by-hole detonation. This adjustment will enable more precise control over the PPV, ensuring that it meets the safety requirements for blasting.

6. Conclusions

This study focused on the section of a coastal tunnel project where there is a risk of seawater intrusion utilizing theoretical analysis, numerical simulation, and field experiments to investigate the safe vibration threshold and control techniques. The main conclusions are as follows.

Based on conventional blasting techniques and using a calculation method derived from the fracture mechanics theory and silo theory, the minimum safe distance between the working face and the fault to prevent seawater intrusion was calculated for two scenarios: blasting construction and surrounding rock unloading. After a comprehensive analysis, the minimum safe distance for preventing seawater intrusion was determined to be 2.55 m. This distance was used to define the risk control zone for seawater intrusion when using conventional blasting techniques.

A three-dimensional geostructural blasting calculation model was established using numerical simulation. By comparing the numerical simulation data with field-measured blasting vibration data, it was found that within a distance of less than 100 m from the blast center, the relative error of the PPV in all directions was less than 14.8%. This validated the reliability of the calculation model and material parameters. Additionally, through multiple linear regression analyses, quantitative relationships were established between the amount of explosive used and PPV at different distances as well as between the extent of rock damage and PPV at various distances. After verification using empirical formulas, the relationship between the peak particle velocity 30 m behind the working face and the rock damage was selected as the predictive formula, and the safe PPV threshold for preventing seawater intrusion in the fault control zone was determined to be 5.50 cm/s.

By adjusting the explosive charge per stage and the delay time, an optimized blasting scheme using electronic detonators was proposed. The measured vibration data show that the PPV was below the safe threshold, meeting the PPV safety requirements for the seawater intrusion risk control zone. This method is suitable for blasting excavation in the fault control zone.

These results can serve as a reference for safe and efficient blasting excavation in coastal tunnel projects, particularly in regions prone to seawater intrusion.