Blasting Damage Control in Jointed Rock Tunnels: A Review with Numerical Validation of Water-Pressure Blasting

Abstract

Joints and other discontinuities in rock masses cause overbreak, underbreak, and instability during tunnel blasting. This paper reviews recent advances in damage control for jointed rock tunnels and validates key findings through numerical simulations. At the microscale, joints affect stress wave propagation, energy distribution, and crack growth patterns. We used ANSYS/LS-DYNA 19.0 to simulate 16 parametric cases and quantify the effects of joint geometry on blasting response. Results show that joint-to-borehole distance is the primary factor controlling damage distribution. A joint dip angle of 45° produces the most severe damage anisotropy, with cracks propagating preferentially along the joint plane. A three-dimensional tunnel model was then developed to assess water-pressure blasting. Compared with conventional methods, water-pressure blasting reduces damage depth by 20.4% and peak particle velocity by 57.6% in jointed rock. The paper also discusses parameter optimization methods, intelligent evaluation techniques, and dynamic control strategies. Engineering recommendations are provided for different geological conditions, including horizontally layered rock, inclined joints, and deep high-stress environments. This work offers both theoretical insights and practical guidance for precision blasting in jointed rock tunnels.

1. Introduction

Tunnel construction is an important part of national infrastructure development. As transportation networks extend into mountainous areas and deep underground, the geological conditions faced by tunnel projects are becoming increasingly complex [1]. Discontinuities such as joints and bedding planes are widespread in rock masses, fundamentally altering the mechanical properties of the rock [2] and its response to blasting [3]. The drill-and-blast method, due to its equipment flexibility and strong adaptability, remains the primary method for excavating hard rock tunnels [4]. However, the presence of joints makes the blasting process far more complex. On one hand, joints serve as weak planes that guide cracks to propagate preferentially along them [5], leading to an anisotropic distribution of damage. On the other hand, the reflection and transmission of stress waves at joints alter the energy propagation path, making blasting results difficult to predict and control. In tunnel engineering, the issues of overbreak and underbreak in jointed rock mass tunnels are much more severe than in intact rock [6], which not only increases construction costs but also threatens the long-term stability of the surrounding rock [7]. Therefore, understanding how joints influence rock blasting and using this knowledge to guide the selection of blasting techniques and the quantitative optimization of blasting schemes for tunnels in jointed rock masses is essential. The whole process is crucial for predicting tunnel excavation outcomes and preventing geological hazards and engineering accidents triggered by tunnel excavation.

To more clearly investigate the blasting mechanism in jointed rock masses, laboratory and field experiments have provided important insights into how joints affect blasting. Xiao et al. (2022) [8] used dynamic caustics in physical model tests to elucidate the localized damage effects of joints on rock blasting. They found that on the blast-facing side, joints promote damage, whereas on the back-blast side, they inhibit damage. Wang et al. (2025) [9] conducted large-scale field tests which showed that joint planes significantly affect the stress distribution and dynamic damage in the surrounding rock. The tunnel crown and side walls experienced the most severe damage, indicating that support design should focus on these areas. In addition, numerical simulation has become a mainstream approach for analyzing blasting in jointed rock masses. Zhang et al. (2023) [10], Zhou et al. (2022) [11], Huang et al. (2025) [12], and Song et al. (2022) [13] built numerical models using the finite element software ANSYS/LS-DYNA and the Particle Flow Code (PFC), systematically revealing the effects of joints on blasting stress waves. They calculated and validated the transmission and reflection coefficients of joints as well as energy ratios, and pointed out, for example, that the transmission coefficient increases as the joint angle grows. To make research results more credible, many researchers have adopted a combination of physical experiments and numerical simulations. Shen et al. (2024) [14] combined blasting experiments on red sandstone models with PFC2D discrete element simulations, revealing that the spacing between parallel joints controls the propagation and morphology of blast-induced cracks. Liang et al. (2024) [15] established a visual “wave–crack synchronization” experiment, supplemented by CDEM numerical simulations, to simultaneously capture stress wave evolution and blast-induced crack propagation in jointed rock masses. They revealed that diffraction at joint tips and stress superposition predominantly govern the initiation of wing cracks and the evolution of their length, and that high in situ stress significantly inhibits crack extension. Zuo et al. (2024) [16] performed blasting experiments on layered rock models coupled with numerical simulations, constructing a framework for blast crack propagation mechanisms under the control of bedding planes. They clarified the dominant effects of bedding planes on crack initiation positions, propagation paths, and coalescence processes.

Tunnel blasting control techniques developed with actual geological information are more suitable for practical engineering. A detailed geological survey and the acquisition of joint data is the first step in designing a blasting plan. Ba et al. (2020) [17] noted that digital compass contact measurements, close-range photogrammetry, and 3D laser scanning have become the three major techniques for geological surveying of structural planes in tunnels. Fan et al. (2021) [18] used Tunnel Seismic Prediction (TSP) technology to detect joints 100–150 m ahead of the tunnel face. Zheng et al. (2024) [19] developed an image recognition approach based on deep learning and neural networks, proposing a lithology identification neural network model that combines an attention mechanism with feature Brownian distance covariance. They applied this model to joint identification on rock faces and achieved a maximum accuracy of 97.6%. Under jointed rock mass conditions, researchers both in China and abroad have developed various blasting control technologies. Li et al. (2023) [20] built a three-dimensional numerical model combined with field tests to propose optimal borehole layout recommendations for tunnels in jointed rock masses under different initial in situ stress conditions. To improve the efficiency of blasting parameter optimization, Minh et al. (2021) [21] proposed an algorithmic model based on the Delphi language and AutoCAD platform to automatically generate tunnel contour blasting designs and study the influence of different factors on blasting outcomes. Given the complexity and high cost of traditional smooth blasting charging procedures, Wu et al. (2025) [22] proposed a new detonator-free cumulative spaced charging structure. Numerical simulations and other tests verified that this approach improves rock fragmentation and reduces surrounding rock damage under various rock conditions. In addition, He (2021) [23] reviewed the development and application of water pressure blasting technology in tunnel engineering, pointing out that hydro-blasting can effectively control overbreak and underbreak, resulting in a smoother tunnel profile. Directional blasting techniques such as shaped charge blasting can also achieve precise control of crack propagation paths through directed energy release. Combining these techniques yields Energy Concentrated Hydro-Blasting Technology (ECHBT), which integrates the advantages of both methods. Studies by Han et al. (2022) [24] and Xu et al. (2024) [25] show that ECHBT can enlarge the crushed zone of the rock mass, suppress far-field vibrations, and increase the success rate of main cracks propagating in the intended direction.

Based on the above analysis, there are three key gaps in the current research waiting to be filled. First, existing studies mostly remain at qualitative descriptions or single-parameter analyses, lacking quantitative relationships between joint parameters and blasting response that can directly guide engineering design. Second, new control technologies such as water pressure blasting and shaped charge blasting have mostly been studied under idealized two-dimensional conditions; their applicability and quantitative optimization strategies in three-dimensional jointed conditions are still unclear. Third, there is a lack of an intelligent blasting control framework that integrates geological exploration, parameter optimization, construction execution, and effect evaluation. The technologies at each stage are relatively independent, making it difficult to achieve dynamic optimization based on real-time feedback. This paper addresses these key issues through a systematic review and original numerical simulation studies. The review section systematically outlines the research progress on blasting in jointed rock masses from mechanistic understanding to technological application, constructing a complete framework of knowledge. More importantly, this study carried out two sets of numerical simulations using ANSYS/LS-DYNA: The first simulation, with 16 groups of parametric conditions, quantitatively analyzed the influence of joint geometric parameters on the dynamic response of blasting, establishing quantitative relationships between joint parameters and blasting response. The second simulation used a three-dimensional tunnel blasting model to quantitatively verify the damage control effectiveness of water pressure blasting technology in jointed rock masses. Theoretically, this work deepens the understanding of blasting mechanisms in jointed rock masses and provides data support for the development of more precise theoretical models. Practically, the validated effectiveness of the control techniques and the proposed parameter optimization strategies can be directly applied to engineering design, transforming blasting design from reliance on empirical judgment to scientific decision-making based on quantitative analysis. These advances are of great significance for improving the safety and cost-effectiveness of tunnel construction under complex geological conditions, and for promoting the transformation of tunnel blasting technology from experience-driven to data-driven, intelligent methods.

This paper is structured as a review article with numerical validation. The review section systematically summarizes the current understanding of joint effects on blasting and state-of-the-art control technologies. The numerical validation section presents original simulations that quantify the influence of joint parameters on blasting response and verify the effectiveness of water-pressure blasting. This combination provides both a comprehensive knowledge framework and quantitative data support for engineering applications.

2. Mechanisms of Joint Influence on Blasting Damage in Rock Masses

2.1. Basic Mechanisms of Rock Blasting

Rock blasting is a complex dynamic process. Most researchers believe this process primarily relies on the dynamic action of stress waves and the quasi-static action of explosive gases [26]. Upon detonation, the high-temperature, high-pressure detonation products generate a strong shock wave near the borehole wall, instantly exceeding the dynamic compressive strength of the rock and causing the rock around the borehole to undergo compressive crushing and pulverization, forming a crushed zone. During this process, the shock wave attenuates into a compressive stress wave that propagates outward. The radial propagation of the compressive wave induces circumferential tensile stress; when this exceeds the rock’s dynamic tensile strength, radial cracks rapidly initiate and extend, forming a fractured zone. Moreover, when the compressive stress wave encounters a free surface or a discontinuity such as a joint, it is reflected. The reflected tensile wave, if it exceeds the rock’s tensile strength, can induce spalling (layered fracturing) and drive further crack extension. As the distance from the blast increases, the energy of the compressive stress wave continues to dissipate, and its peak pressure decays below the rock’s elastic limit. The wave then transforms into a low-frequency, low-amplitude seismic wave that only causes elastic vibrations (recoverable deformation), forming a vibration zone. The rock breakage mechanism under blasting is illustrated in Figure 1 [27] (where R0 is the borehole radius). The propagation velocity, attenuation characteristics, and waveform of the stress wave have a decisive influence on the initial extent of rock breakage and the degree of damage [28,29].

After the stress wave, the explosive gases act in a quasi-static manner to complete crack coalescence and rock throw. The high-temperature, high-pressure explosive gases expand rapidly and force their way into the initial cracks. Through sustained gas pressure pressing against the crack walls, a splitting (wedging) effect is produced [30], causing further crack propagation and coalescence. The gas acts over a relatively longer duration; its quasi-static pressure is the main driving force for fully breaking the rock, forming the blasting crater, and throwing out rock fragments. The magnitude and duration of the gas pressure, as well as its seepage and distribution within the fracture network [31], directly influence the resultant fragment size, throw distance, and the cost-effectiveness of the blasting outcome. Stress waves and explosive gases work in tandem and complement each other during the blasting process, jointly accomplishing the rock breaking and excavation tasks [32]. Understanding the mechanisms of these two effects and their interrelationship is the basis for optimizing blast design, improving blasting efficiency, and controlling blasting hazards.

2.2. Microscopic Regulatory Role of Joints in Rock Blasting

Common discontinuities in natural rock masses, such as joints, bedding planes, and faults, have a pronounced micro-scale regulatory effect on the blasting-induced fracturing process, profoundly influencing stress wave propagation, energy attenuation, crack initiation and growth, the action of explosive gases, and the accumulation of damage. When a blasting stress wave propagates to a joint plane, its propagation is impeded and complex reflection and transmission phenomena occur [33]. This alters the original propagation path of the stress wave, reflecting some energy back toward the blast source while transmitting some into the rock on the other side of the joint. The change in wave path likewise causes an abrupt change in the displacement of the rock mass during blasting [34]. At the joint interface, stress waves also undergo energy dissipation and waveform scattering, which significantly accelerates the attenuation of wave energy and propagation velocity. Furthermore, joint geometric characteristics—such as infilling material [35], joint orientation [36], joint aperture, and surface roughness—directly regulate the reflection and transmission coefficients and the proportion of energy dissipated. These factors in turn alter the waveform of the stress wave transmitted through the joint, govern the effective fracturing energy that passes into the rock beyond the joint, and affect the subsequent stress distribution [10] and blasting fracture outcomes [11]. A dense network of joints will cause the stress wave energy to attenuate rapidly, limiting its effective breakage range.

To gain a better understanding of the micro-regulation by joints in the blasting fracture process, it is necessary to deeply investigate the coupled mechanisms of stress wave propagation, crack propagation, and damage evolution in jointed rock masses. In particular, the dynamic propagation of cracks and the cumulative evolution of damage together determine the final failure pattern and extent of the rock mass. Joints, as inherent discontinuities in the rock, can alter the stress field at crack tips, guiding existing cracks to propagate preferentially along the joint plane [37,38]. Stress concentration occurring at the joint tips leads to the formation of wing cracks and secondary cracks [39]. These effects ultimately and significantly change the expected fracture pattern of the rock mass under blasting [15]. Whether the influence of a joint on crack propagation is promotive or suppressive [16,40] mainly depends on the relative orientation between the joint and the principal stress field induced by blasting, the joint’s geometric characteristics, and the characteristics of the blast loading. When the orientation of the joint plane is unfavorable to the direction of the maximum tensile stress, the joint may act as an energy barrier, impeding stress waves from transmitting into certain regions and weakening the tensile stress field at crack tips, thereby suppressing crack propagation outward [12]. On the other hand, the explosive gases generated by the blast flow preferentially along the joint plane, altering the pressure field and enhancing the splitting (wedging) effect on the joint walls. This guides and promotes crack propagation.

While cracks propagate, internal damage in the rock mass is also accumulating [41]. Joints induce cracks to propagate preferentially along the structural planes, which exacerbates the damage in the material between the borehole and the joint [42], while the damage in the rock outside the joint is suppressed. In natural rock masses, joints typically occur in multiple sets of orientations [43], causing the blasting stress waves to undergo varying degrees of attenuation and scattering. Ultimately, by modulating the transmission coefficient and the fractal propagation of cracks [13], joints lead to damage zones that develop and accumulate anisotropically [44]. These anisotropic damage patterns are the direct micro-scale cause of engineering issues such as differences in surrounding rock stability, uneven excavation profiles, and overbreak/underbreak. Furthermore, blasting in deep rock is jointly controlled by joints and in situ stress, and the coupling of these factors significantly amplifies the non-uniformity of damage [45]. Yang et al. (2018) [46] and Ma et al. (2025) [47] have conducted studies via numerical simulation, theoretical analysis, and experiments. The results indicate that in situ stress has a guiding influence on blast-induced crack propagation, causing cracks to tend to develop in the direction of the maximum in situ stress while also reducing the overall damage extent in the jointed rock mass. As the in situ stress increases, both the proportion of fractures and the main crack length in the jointed rock mass show a declining trend [48,49]. In-depth research on the mechanical response and damage evolution of rock under dynamic loading [50] will help in establishing a dynamic damage constitutive model for jointed rock masses that more closely reflects reality [51]. These theoretical insights provide relevant scientific evidence for optimizing blasting parameters, improving energy utilization efficiency, and controlling damage to the surrounding rock. This achieves the dual goals of precise excavation profile formation and protection of the surrounding rock, offering important guidance for safe and efficient construction of deep tunnels.

2.3. Quantitative Influence of Joint Parameters on Blasting Dynamic Response

To further reveal the quantitative influence of joint geometric parameters on the dynamic response of rock blasting, the authors established a numerical blasting model of granite (4 m × 4 m) containing a single joint using ANSYS/LS-DYNA. The Arbitrary Lagrange–Euler (ALE) algorithm is used to deal with the dynamic coupling of multi-materials. The explosive and air elements were defined as ALE meshes, while the rock elements were defined as Lagrange meshes. In order to improve the computational efficiency and not affect the numerical simulation results, the two-dimensional model is adopted, that is, the thickness direction of the model is one unit width. The overall model grid is uniformly sized at 1 cm. Non-reflecting boundaries were applied to the four sides of the model. The borehole radius was 0.04 m, with the joint located on the right side of the borehole, as shown in Figure 2. A systematic study was conducted on four categories of joint parameters—joint-to-borehole distance (S = 0.3–1.3 m), joint trace length (L = 0.5–2 m), joint aperture (A = 0.01–0.04 m), and joint dip angle (δ = 30–75°)—and their quantitative effects on the rock mass’s blasting dynamic response. The joint parameter configurations and corresponding case IDs are listed in

Table 1. Joint-parameter configurations and corresponding case IDs.

Factor	1	2	3	4
I: Stand-off distance to the joint	S = 0.3 m	S = 0.6 m	S = 0.9 m	S = 1.3 m
II: Joint trace length	L = 0.5 m	L = 1 m	L = 1.5 m	L = 2 m
III: Joint aperture	A = 0.01 m	A = 0.02 m	A = 0.03 m	A = 0.04 m
IV: Joint inclination	δ = 30°	δ = 45°	δ = 60°	δ = 75°

.

The simulations are based on several key assumptions: (1) The rock material is modeled as a homogeneous, isotropic medium outside the joint zone. (2) The joint is modeled as a continuous weak plane filled with air. (3) The explosive detonation follows the JWL equation of state, assuming ideal detonation conditions. (4) The multi-scale mechanisms are coupled through the RHT damage model, which controls damage accumulation based on plastic strain and pressure history.

The air is defined by the keyword *MAT_NULL, its state equation is *EOS_LINEAR_POLYNOMIAL, and its density is 1.29 kg/m³. The specific parameters of explosives [52] are shown in

Table 2. Parameters for the explosive material and JWL EOS [52].

ρe (kg/m3)	VOD (m/s)	PCJ (GPa)	AJWL (GPa)	BJWL (GPa)	R1	R2	ω	E0 (GPa)
1320	6690	16	586	21.6	5.81	1.77	0.282	7.38

. The explosive was modeled using the *MAT_HIGH_EXPLOSIVE_BURN keyword with the Jones–Wilkins–Lee (JWL) [52] equation of state. The JWL equation of state describes the pressure-volume relationship of detonation products:

$$P=A\left(1-{\displaystyle \frac{\omega }{V{R}_1}}\right)e^{-R_{1}V}+B\left(1-{\displaystyle \frac{\omega }{V{R}_2}}\right)e^{-R_{2}V}+{\displaystyle \frac{\omega e}{V}}$$

(1)

where P is the pressure, V is the relative volume, e is the internal energy, and A, B, R₁, R₂, and ω are state equation parameters.

The rock was modeled using the Riedel–Hiermaier–Thoma (RHT) model, which is widely used for simulating the dynamic response of rock and concrete materials under blast loading. The RHT model can effectively describe the damage accumulation and strain rate effects in brittle materials. The key parameters for granite include: density ρ = 2660 kg/m³, Poisson’s ratio = 0.16 [52]. The complete RHT parameters are listed in

Table 3. Input parameters for RHT model for rock. (Note: * indicates an optional parameter, not a required parameter).

Parameter	Value	Parameter	Value
Mass density RO kg/m3	2660	Porosity exponent NP	3.0
Initial porosity ALPHA	0	Reference compressive strain-rate EOC	3 × 10−5
Crush pressure PEL (MPa)	125	Reference tensile strain rate EOT	3 × 10−6
Compaction pressure PCO (GPa)	6.0	Break compressive strain rate EC	3 × 1025
Hugoniot polynomial coefficient A1 (GPa)	25.7	Break tensile strain rate ET	3 × 1025
Hugoniot polynomial coefficient A2 (GPa)	37.84	Compressive strain rate dependence exponent BETAC	0.026
Hugoniot polynomial coefficient A3 (GPa)	21.29	Tensile strain rate dependence exponent BETAT	0.007
Parameter for polynomial EOS B0	1.22	Volumetric plastic strain fraction in tension PTF	0.001
Parameter for polynomial EOS B1	1.22	Compressive yield surface parameter GC *	0.53
Parameter for polynomial EOS T1 (GPa)	25.7	Tensile yield surface parameter GT *	0.7
Parameter for polynomial EOS T2	0.0	Erosion plastic strain EPSF	2.0
Elastic shear modulus SHEAR (GPa)	17	Shear modulus reduction factor XI	0.5
Compressive strength FC (MPa)	167.8	Damage parameter D1	0.04
Relative tensile strength FT *	0.04	Damage parameter D2	1.0
Relative shear strength FS *	0.21	Minimum damaged residual strain EPM	0.015
Failure surface Parameter A	2.44	Residual surface parameter AF	0.25
Failure surface Parameter N	0.76	Residual surface parameter AN	0.62
Lode angle dependence factor Q0	0.68	Gruneisen gamma GAMMA	0.0
Lode angle dependence factor B	0.05

.

To verify the reliability of the RHT model parameters, the numerical simulation results were compared with the laboratory-scale single-hole blasting experiments on Barre granite conducted by Dehghan Banadaki and Mohanty (2012) [52]. The granite sample had a diameter of 144 mm, a height of 150 mm, and a borehole diameter of 6.45 mm. A numerical model with identical dimensions and material parameters was established. Figure 3 shows the comparison between experimental and numerical results. The crushed zone and fracture zone in the simulation agree well with the experimental observations, confirming that the RHT model parameters used in this study are reasonable and can accurately simulate the damage and fracture behavior during rock blasting.

Figure 4, Figure 5, Figure 6 and Figure 7 clearly illustrate the differences in crack propagation patterns under various parameter settings. Combined with the specific data results shown in Figure 8, the following can be observed: the joint-to-borehole distance is the primary factor controlling the damage distribution. When the distance between the joint and the borehole increases from 0.3 m to 1.3 m, the proportion of damaged elements in the rock mass rises from 8.44% to 9.75%, and the length of the main crack on the right side of the borehole increases from 37.93 cm to 142.98 cm. This indicates that the shielding effect of the joint diminishes significantly as the distance increases. When the distance exceeds 0.6 m, damage in the rock mass on the right side of the joint tends to disappear. Increasing the joint trace length makes the wing cracks that extend from the joint tips more pronounced, while the length of the main crack on the right side of the borehole gradually decreases. A larger joint aperture intensifies the energy dissipation of stress waves at the joint, and the difference in peak particle velocity (PPV) between the incident side and the back-blast side monitoring points grows. At A = 0.04 m, the PPV difference reaches 35%. Meanwhile, explosive gases more easily penetrate along the joint plane, causing secondary splitting of the rock on both sides of the joint. Joint inclination has a significant guiding effect on crack propagation paths. When δ = 45°, cracks propagate preferentially along the joint plane and form through-going failure, and the PPV attenuation difference between the incident side and the back-blast side reaches its maximum. At δ = 75°, the main crack deviates under the influence of the joint and tends to propagate radially. These quantitative results not only verify the multi-scale regulatory effects of joint characteristics on blast energy shielding, crack propagation paths, and spatial damage distribution, but also provide key data support for optimizing blasting parameters based on joint characteristics in practical engineering. In particular, when determining critical parameters such as perimeter hole spacing and explosive charge distribution, the relative position of major joints to the boreholes should be fully considered.

It should be noted that the parametric study employed a single, continuous joint in a 2D model to isolate the effects of individual joint parameters. Natural rock masses typically contain multiple, intersecting, and non-persistent joint sets, forming complex Discrete Fracture Networks (DFNs). The quantitative relationships obtained from the single-joint model provide fundamental insights, but direct application to complex joint networks requires further verification. Future work should extend these findings to DFN-based models to capture the coupled effects of multiple joints.

3. Damage Control Techniques for Blasting in Jointed Rock Mass Tunnels

3.1. Blast Parameter Optimization Based on Joint Characteristics

3.1.1. Joint Information Acquisition Technologies

In the face of the adverse impacts of joints on blasting and engineering challenges such as overbreak, underbreak, and surrounding rock damage, a great number of advanced blasting technologies and intelligent methods have been introduced into tunnel engineering in recent years, aiming to achieve adaptive adjustment and parameter optimization of tunnel blasting under jointed conditions. Accurate geological exploration and acquisition of joint information are the foundation for optimizing tunnel blasting parameters. Digital compass contact surveys, close-range photogrammetry, and 3D laser scanning have become the three major technical methods supporting the geological investigation of structural planes in tunnels [17]. Tunnel Seismic Prediction (TSP) technology can detect joints up to 100–150 m ahead of the tunnel face [18]. Image recognition technology based on deep learning and neural networks has achieved a joint recognition accuracy of up to 97.6% on rock surfaces [19]. These technologies provide real-time geological data support for blast design.

3.1.2. Optimization Method of Blasting Parameters in Joint Rock-Mass Tunnel

After obtaining the actual joint distribution and lithology information, intelligent optimization of blasting parameters can be carried out to match the site-specific joint characteristics [53,54]. Smooth blasting, as a key controlled blasting technique for tunnel excavation, can create a smooth excavation surface along the tunnel design contour line by arranging perimeter boreholes reasonably and precisely controlling the charge in each hole, thereby minimizing disturbance to the surrounding rock [55]. Based on the earlier understanding of blasting mechanisms in jointed rock masses, the presence of discontinuities like joints and bedding planes makes it more likely to encounter poor smooth blasting profiles or severe overbreak in tunnels [56]. Therefore, when conducting smooth blasting in rock masses with joints, the influence of joint characteristics on blasting parameters must be fully considered. For deeply buried, high in situ stress tunnel environments, the effect of in situ stress must also be taken into account [57]. Li et al. (2023) [20], by establishing a 3D numerical model and conducting field tests, studied the variation in tunnel blasting results under the combined effect of in situ stress and joints, and on this basis proposed optimization recommendations for borehole layout in jointed rock tunnels under different initial in situ stress conditions. For example, under Class IV surrounding rock conditions, a perimeter hole spacing of 45 cm and a linear charge density of 0.375 kg/m yielded the best smooth blasting results. This finding provides a reference for how to adjust blasting parameters in deep, high-stress tunnels when joints are present. To improve the efficiency of blasting parameter optimization calculations, researchers have developed corresponding algorithmic tools. For example, Minh et al. (2021) [21] proposed an algorithmic model based on Delphi and AutoCAD, which can automatically generate tunnel contour blasting plans and study the influence of different factors on blasting outcomes. Using this model, a parametric analysis was conducted on factors such as the rock Protodyakonov coefficient, joint spacing, borehole diameter, and tunnel cross-sectional area. The results showed that the specific explosive consumption increased with higher rock Protodyakonov coefficient, larger joint spacing, and greater borehole diameter, but decreased with increasing tunnel cross-sectional area. This model provides a supporting tool for rapid optimization of smooth blasting parameters under specific geological conditions.

In addition to optimizing borehole layout parameters, optimization of the charging structure [58] and initiation network has also been a research focus in recent years for better control of tunnel smooth blasting results [59]. Traditional smooth blasting usually employs decoupled charge cartridges (alternating empty hole sections and charged sections) to control explosive energy, but this still requires detonators to link multiple charge segments and uses auxiliary materials like wooden spacers for fixing, making the process complex and costly. To address these issues, Wu et al. (2025) [22], based on the principle of emulsion explosive synergistic blasting, proposed a new cumulative deck charging structure that requires no detonators. Through numerical simulation, theoretical analysis, and field tests, they verified its effectiveness in improving rock fragmentation and reducing damage to the surrounding rock under different rock conditions. This provides a new design concept for charging during tunnel blasting under various joint conditions. In summary, with the aid of intelligent algorithms and simulation technology, one can dynamically optimize a series of parameters based on the characteristics of joints in the rock mass—including borehole diameter, hole spacing, charge amount, initiation sequence, delay timing, and special charging configurations—in order to dynamically adjust the blasting plan according to the actual situation. This shifts blast design from relying on experience to relying on data and intelligent computation, greatly improving the specificity and adaptability of the design.

3.2. Water Pressure Blasting Technology and Quantitative Verification

3.2.1. Principles of Water Pressure Blasting Technology

As a novel blasting method, water pressure blasting technology has demonstrated unique advantages in the construction of tunnels through jointed rock masses. It can better address challenges such as overbreak/underbreak, damage to surrounding rock, strong blast vibrations, and high tunnel dust levels. Water pressure blasting, also known as water medium blasting or water-sealed blasting, refers to a controlled blasting technique that uses water in the borehole as the medium for energy transmission and rock breakage [60]. The basic principle is to fill the borehole with a certain amount of water, and then detonate the explosive placed in the water or in water bags [61]. Compared to traditional air-deck (air-decoupled) charges [62], the incompressibility of water allows the explosive shock wave to be transmitted to the borehole wall more uniformly and persistently. When the shock wave passes through a water bag, virtually no energy is lost, so more energy is used for fracturing the rock, thereby improving the energy utilization of the explosive [63]. This produces smaller blast-induced vibrations, improves rock fragmentation, and to some extent reduces the specific charge and damage to the surrounding rock, enhancing the quality of the blasting excavation [64,65,66]. At the same time, water pressure blasting is an embodiment of green blasting: during the blast, the explosive’s instantaneous energy vaporizes the water in the collar water bag into steam under high temperature and pressure, which fully interacts with the dust, achieving dust suppression, cooling, and absorption of some harmful gases, thereby improving the working environment [67]. Water pressure blasting is especially suitable for surrounding rock that is heavily jointed and fractured. This is because the buffering and lubricating effects of water can reduce the impact of blast stress waves on joint planes, making the rock mass less prone to secondary fracturing or instability along the joints. He (2021) [23] studied the development and current applications of water pressure blasting technology in tunnel engineering, and pointed out that compared to ordinary blasting, water pressure blasting can effectively control overbreak/underbreak, produce a smoother tunnel profile, and reduce vibration disturbances to the surrounding rock.

3.2.2. Quantitative Verification of Water Pressure Blasting Damage Control Effect

To quantitatively assess the damage control effectiveness of water pressure blasting in jointed rock mass tunnels, the authors established a 3D tunnel smooth blasting model based on a certain tunnel project in Changchun and carried out numerical studies. For the 3D tunnel blasting model, an adaptive mesh distribution was adopted: coarser meshes (approximately 20 cm) were used in the far-field surrounding rock, while finer meshes were applied near the tunnel face, blast holes, and excavation contour to accurately capture the damage distribution. This local refinement strategy ensures numerical accuracy in the regions of interest while maintaining computational efficiency. The quantitative comparisons presented in this study were conducted under identical mesh configurations, ensuring that the observed trends reflect physical phenomena rather than numerical discretization effects. The surrounding rock volume excavated was 40 m × 40 m × 5 m. Monitoring points were marked on the tunnel face at an advance depth of 0.5 m, as shown in Figure 9. The borehole radius was 0.025 m, and the blast-hole layout on the tunnel face is shown in Figure 10. The tunnel charge configuration is shown in Figure 11. The tunnel excavation used a top-and-bottom bench method; in the simulation, a restart technique was employed: after completing blasting for the upper bench at the tunnel face, the lower bench blast excavation was initiated. The tunnel advance per round was 1.0 m, and the surrounding rock was Class V granite. This simulation comparatively analyzed three scenarios: (1) intact rock mass + conventional charge; (2) jointed rock mass with a 45° joint + conventional charge; and (3) jointed rock mass with a 45° joint + water bag charge.

Because the model is three-dimensional, the tunnel face at a 0.5 m advance was extracted to present the simulation results in more detail, as shown in Figure 12, Figure 13 and Figure 14. The damage evolution contour plots visually demonstrate the guiding effect of the joint and the superiority of water pressure blasting: the presence of the joint causes blast-induced damage to preferentially extend along the joint plane; under the same joint conditions (45° dip), using water-filled charges significantly reduced the area of the Excavation Damaged Zone (EDZ) around the tunnel periphery. Damage contour analysis (Figure 15) and quantitative data (Figure 16) indicate that after using water bag charges in the jointed rock mass, the damage depth in the surrounding rock near the joint decreased from 0.54 m to 0.43 m, a reduction of 20.4%. The average EDZ thickness was 0.18 m, about 17.37% less than that with conventional charging in the jointed rock; meanwhile, the peak particle velocity at each monitoring point was reduced by about 57.6% on average. This fully reflects the absorption and buffering effect of the water medium on stress waves, effectively mitigating the adverse impact of the joint on damage propagation. Compared to the intact rock mass, the presence of the joint caused the damage zone to extend about 54.3% further along the joint direction, whereas the water bag charge effectively suppressed this unfavorable extension. This numerical study provides a reliable theoretical basis and guidance for parameter optimization in the application of water pressure blasting technology in tunnels through jointed rock masses.

It should be noted that in practical applications, water pressure blasting parameters should be adjusted according to the characteristics of the joints. For rock masses with a high joint density or wide joint apertures, the amount of water in the borehole or the number of water bags can be increased appropriately to leverage a stronger water coupling effect, absorbing excess energy and suppressing the extent of damage. Conversely, for areas where joints are sparse and the surrounding rock is highly intact, the water amount should be controlled to avoid weakening the blast effect and causing underbreak. In addition, if a major joint plane dips toward one side, to prevent excessive damage to the surrounding rock on that side, one can consider using a longer water column in the corresponding boreholes for water stemming, in order to protect the rock outside the joint. These joint-specific parameter adjustments demonstrate the potential of water pressure blasting technology for dynamic control in blasting applications.

3.2.3. Other Emerging Blasting Control Technologies

Apart from water pressure blasting, The excavation compensation method is implemented for shallow buried large-span hard rock tunnel engineering, which can not only control the deformation of surrounding rock but also improve the efficiency of tunnel construction [68]. Directional blasting techniques such as shaped charge blasting achieve precise control of crack propagation paths by releasing energy in a targeted manner [69]. When applied in jointed rock masses, the orientation of the shaped charge groove should be adjusted according to the joint orientation. When it is necessary to utilize the joint plane to achieve pre-splitting, the shaped charge slot should face the main joint plane; when it is necessary to avoid excessive fracturing along a joint, the shaped charge slot should be offset from the joint plane as much as possible [70]. Energy-Concentrated Hydro-Blasting Technology (ECHBT) combines the dual advantages of water pressure and shaped charge blasting: it ensures directional energy release while using the water medium as a buffer to avoid excessive shock. Studies by Han et al. (2022) [24] and Xu et al. (2024) [25] have shown that ECHBT can enlarge the crushed zone in the rock mass, suppress far-field vibrations, and improve the success rate of main cracks propagating in the desired direction. This technology shows good application prospects in tunnels where joints are prevalent and strict profile control is required. In the future, the integration of these technologies with intelligent sensing and parameter optimization systems will form a comprehensive joint dynamic control blasting scheme from detection through execution.

3.3. Intelligent Evaluation and Dynamic Control

Intelligent evaluation of post-blast performance and dynamic adjustment of parameters are key steps to achieving precise control of blasting in jointed rock mass tunnels [71]. By obtaining feedback information through intelligent monitoring and evaluation technologies and combining it with the aforementioned parameter optimization methods, a closed-loop control system can be established to enable continuous optimization of the blasting plan.

Intelligent monitoring and evaluation technologies provide technical support for quantifying blasting outcomes. In terms of monitoring, aside from traditional measurements of vibration velocity (PPV) and stress–strain, Fiber Bragg Grating (FBG) sensors, which offer strong anti-interference capability and allow multi-point serial monitoring, have been widely used for real-time monitoring of damage and deformation in the surrounding rock [72]. In terms of evaluating blasting effects, digital image processing technology has become a mainstream approach. Considering rock fragment size (fragmentation) as a key evaluation index, He et al. (2024) [73] used an improved hybrid genetic Otsu algorithm combined with a marked watershed algorithm to achieve automatic identification of blast fragments in 0.96 s, with an accuracy up to 91.1%. A neural network model based on deep learning can integrate fragment size distribution and overbreak/underbreak volume [74,75] to output corresponding values and provide a composite score for tunnel blasting performance [76,77]. The half-hole ratio is another important indicator for assessing smooth blasting quality. Shi et al. (2024) [78] reconstructed the post-blast tunnel surface into a 3D profile using Structure from Motion (SFM) techniques, and then combined this with a RANSAC algorithm to compute the full cross-section half-hole ratio in a short time, with greatly improved accuracy. These intelligent evaluation techniques transform traditional qualitative judgments into quantitative analysis, providing precise evidence for parameter optimization.

The control mechanism of dynamic intelligent tunneling constitutes a closed-loop optimization system for blasting in jointed rock mass tunnels. The specific process includes the following aspects: first, use the aforementioned joint information acquisition techniques to ascertain the current geological conditions; then formulate optimized blasting parameters based on the joint characteristics; execute the blast and monitor the dynamic response in real time; obtain performance indicators through intelligent evaluation; and finally, use intelligent algorithms such as Bayesian optimization or genetic algorithms to analyze the causes of any deviations and adjust the parameters for the next cycle. The key is to establish a relational model between joint parameters, blasting parameters, and performance indicators. For instance, if the evaluation finds that overbreak volume exceeds the design value by 20%, the system automatically analyzes the matching relationship between joint dip angle and charge amount, and in the next cycle correspondingly reduces the charge in perimeter holes by 15–25% or adjusts the borehole layout parameters. For zones of dense joints, the system will proactively increase the number of water bags and optimize initiation delays to ensure damage is kept within the design range. Actual engineering projects have already verified the effectiveness of dynamic control. Zhou et al. (2024) [79], using real-time total station monitoring data and 3D laser scanning results combined with Python-based image processing under fractal theory, achieved precise quantification of overbreak and underbreak. Based on the analysis results, they dynamically adjusted the cut hole shape, detonator delay times, and the charging configuration of perimeter holes. After continuous optimization over multiple blasting cycles, the results showed that problems such as blasting vibrations and overbreak/underbreak were largely avoided, and the contour quality was significantly improved. This dynamic control mechanism is especially important in tunnel excavation with highly variable geological conditions—when encountering zones of dense joints or abrupt changes in rock type, the system can adjust parameters within 2–3 cycles, preventing loss of quality control due to geological changes.

Through the combination of intelligent evaluation and dynamic control, joints can be transformed from unfavorable factors in blasting into geological conditions that can be effectively utilized. By continuously learning from accumulated blasting data, the system gradually optimizes the parameter model and forms a knowledge base tailored to specific joint conditions. This “mechanism-guided + data-driven” control mode not only increases the success rate of individual blasts, but more importantly achieves stable control of blasting quality throughout the entire tunnel construction cycle, providing a systematic solution for the safe and efficient excavation of tunnels in jointed rock masses.

4. Engineering Applications and Recommendations for Technological Development

4.1. Engineering Application Recommendations Under Different Geological Conditions

The blasting of jointed rock mass tunnel needs to be differentiated according to the specific geological conditions. Based on the previous mechanism research and numerical verification, combined with engineering practice experience, the following targeted technical measures are suggested.

Blasting control for horizontally layered jointed hard rock tunnels: When a tunnel passes through stratified formations with bedding spacing ranging from centimeters to decimeters, the key to controlling the excavation profile and reducing vibrations [80] lies in a differentiated design. It is recommended to adjust the perimeter hole spacing at the tunnel crown and spandrels to 0.5–0.6 m and reduce it to 0.4–0.5 m at the haunches. Change the charging structure from continuous charge to decked (interval) charge, with the length of the decoupled (empty) sections controlled at 0.3–0.4 m. Soaking the stemming sections with water can effectively buffer the impact of stress waves on bedding planes. Adding empty (relief) holes between adjacent boreholes [81,82] will guide crack propagation. Based on the aforementioned mechanism analysis and numerical verification, these measures can effectively suppress large-scale peeling of the crown rock along bedding planes and are expected to reduce overbreak area by 50–60%. Support measures can also be added at the same time to control the large deformation of the surrounding rock [83].

Asymmetric blasting design and layered excavation method for inclined jointed rock: When joint dip angles are in the range of 30–60°, the damage distribution exhibits obvious anisotropy. Slabbing and block falls tend to occur at the tunnel shoulders and side walls, ultimately leading to a rough profile, increased overbreak, and a greater support workload. It is recommended to reduce the explosive charge in boreholes on the downhill side of the joint by 20–30% to decrease the damage on that side. Install 3–4 water bags in the boreholes adjacent to the main joint plane to utilize the buffering effect of water. During construction, the excavation advance should be determined according to the class of the surrounding rock. If the surrounding rock is Class IV, it is advised that the advance per round not exceed 1.5 m, as multiple rounds of advance can cause cumulative damage [84]. In addition, the tunnel can be excavated in two layers (upper and lower). If dividing the cross-section into two benches cannot meet vibration control requirements, it can be divided into upper, middle, and lower parts (or more) for blasting excavation.

Comprehensive control for deeply buried, high in situ stress jointed rock mass: Under high in situ stress, the coupled effect between joints and in situ stress is pronounced. In situ stress has a suppressive effect on the peak effective stress from blasting—the greater the in situ stress, the more evident the suppression. After blasting, the entire face tends to exhibit underbreak, which is unfavorable for blasting excavation. It is recommended to design the perimeter hole spacing as 0.4–0.5 m and appropriately increase the linear charge density. For boreholes aligned along the joint strike, slightly decrease the linear charge density and use decoupled charges. This comprehensive plan controls joint-guided damage while considering the inhibitory effect of high in situ stress on blasting. Attention should also be paid to the issue of instantaneous stress redistribution due to blasting excavation unloading [85], which will produce stress concentration phenomenon [86] and easily trigger rock drops [87] or rock bursts [88].

The above engineering recommendations are based on extensive practical validation and provide workable technical solutions for blasting in jointed rock mass tunnels under different geological conditions. The key is to identify the dominant controlling factors and implement targeted measures.

4.2. Recommendations for Development of the Technical System

The blasting technology of joint rock mass tunnel is developing towards intelligent, refined and green direction. In light of current advances and engineering needs, the following recommendations are offered:

• Intelligent planning and control proposed that deep-learning algorithms be used to enable autonomous optimization of blasting plans. Digital-twin frameworks, by constructing validated virtual models of jointed-tunnel blasting, may be used to forecast outcomes prior to construction and reduce on-site trial costs. An intelligent, data-driven decision system could learn from historical cases and adjust parameters in real time, thereby achieving adaptive control.
• Given the complexity of deep tunnels (dense joint–fracture networks, high in situ stress, elevated pore pressure and temperature), it is recommended that Discrete Fracture Networks (DFNs) be employed to characterize joint complexity. On this basis, multi-field coupling models—incorporating stress, seepage, temperature, and damage fields—should be developed to predict the long-term stability of surrounding rocks and to support life-cycle service assessment.
• Environmentally benign explosives, low-vibration blasting, and dust-reduction strategies are likely to become increasingly important. Particular attention should be given to non-explosive rock-breaking methods, such as CO₂ phase-transition fracturing, especially for urban tunnels and environmentally sensitive settings.

Taken together, these directions are expected to accelerate a shift from experience-driven to data-driven workflows, from rough operation to precision control, and from single-factor tuning to integrated technological applications. As these technologies mature, joints may be reframed from a limiting constraint to a managed—and even advantageous—geological condition, enabling safer, more efficient, and more environmentally responsible underground construction.

5. Conclusions

This paper provides a systematic review of the theoretical mechanisms, enabling technologies, and emerging trends in blasting-damage control for tunnels in jointed rock masses, thereby articulating a coherent framework from micro-mechanism to macro-level application. The principal conclusions are as follows:

• It is shown that joints regulate blasting damage through coupled, multi-scale mechanisms, for which quantitative relations are established. At the microscale, joints modify stress-wave trajectories, redistribute energy, and guide crack growth. Numerical evidence indicates that increasing the joint-to-borehole distance from 0.3 m to 1.3 m raises the proportion of damaged elements from 8.44% to 9.75%. A joint inclination of 45° is identified as the most adverse, promoting preferential crack propagation along the joint plane and leading to pronounced damage anisotropy. Joint aperture is found to be linearly related to PPV attenuation; each 0.01 m increase in aperture produces a 10–20% increase in the PPV-attenuation difference. Taken together, these results provide a defensible basis for precision blasting design informed by joint characteristics.
• Using three-dimensional tunnel-blasting simulations, the superiority of water-pressure blasting is quantitatively verified. Relative to conventional practice, damage depth in the jointed surrounding rock is reduced by 20.4%, and peak particle velocity (PPV) decreases by an average of 57.6%; overbreak is contained within 0.18–0.28 m. Mechanistically, the water medium acts as a buffer, transforming an impulsive shock into quasi-static pressure and thereby suppressing unfavorable damage transmission along joints. These findings support water-pressure blasting as an optimal control technique for jointed conditions.
• A technical scheme of “intelligent sensing—parameter optimization—dynamic control” is formulated. Deep-learning-based joint recognition achieves 97.6% accuracy. With the aid of intelligent algorithms and numerical simulation, blasting-parameter optimization is substantially improved in both accuracy and efficiency. The intelligent evaluation module enables rapid, quantitative assessment of key indices (e.g., fragment-size distribution, half-hole ratio). Through iterative updates across blasting cycles, dynamic control enhances contour quality and shifts practice from passively accommodating joints to actively leveraging them.
• Targeted parameter optimization should be implemented under varying geological scenarios. The deep integration of intelligence and automation is expected to enable autonomous plan optimization. Further refined research on multi-field coupling will elucidate damage evolution in deep and complex environments. Green and sustainable technologies—including non-explosive methods such as CO₂ phase-transition fracturing—are likely to gain prominence. Collectively, these directions will accelerate a transition from experience-driven to data-driven practice, from rough operation to precision control, and from single-factor tuning to integrated technological application.

Overall, the study provides a structured pathway for safe and efficient excavation in jointed rock masses and is of both theoretical and practical significance for advancing tunnel-blasting technology. Future work should focus on three key directions to achieve fully intelligent, precision blasting in jointed rock-mass tunnels: (1) Three-dimensional characterization of complex joint networks: Current studies predominantly use simplified 2D models or single-joint configurations. Future research should develop DFNs (Discrete Fracture Networks) based 3D models that capture the spatial distribution, connectivity, and mechanical properties of multiple joint sets. Integration of advanced geological survey technologies (3D laser scanning, borehole imaging, and seismic tomography) with numerical modeling will enable more realistic representations of jointed rock masses. (2) Prediction of damage evolution under multi-field coupling: Deep tunnel blasting involves the coupling of mechanical stress, temperature, groundwater seepage, and in situ stress fields. Future studies should develop multi-physics models that account for thermo-hydro-mechanical (THM) coupling effects on blasting-induced damage. Such models will be essential for predicting damage evolution in deep, high-stress, and water-rich geological environments. (3) AI-enabled, real-time decision-making: The integration of artificial intelligence with blasting design holds great promise for precision control. Machine learning algorithms can be trained on historical blasting data to predict optimal parameters. Real-time sensor feedback combined with digital twin technology will enable dynamic adjustment of blasting parameters during construction. This closed-loop intelligent system will transform tunnel blasting from experience-driven practice to data-driven, adaptive control.