SVM (Support Vector Machine)-based tunnel earthquake advanced prediction intelligent seismic source

Abstract

The invention requests to protect an SVM-based tunnel earthquake advanced prediction intelligent seismic source, and the seismic source comprises a hammerhead adjustment structure which can replace hammerheads with different contact areas or add a counterweight to adjust and convert parameters of the hammerheads; in order to adapt to the design of a hammer head, a seismic source hammer rod part is specially designed, including an internal protective connecting hole design and a thread structure, and a hammer rod is specially designed; based on hammerhead structure selection of the SVM, geological information related to the optimal hammerhead combination is converted into a data set, the SVM is trained, and the most suitable hammerhead combination is predicted according to field geological environment information for construction. According to the technical scheme, the problems that in the prior art, a pneumatic seismic source device is large in energy transmission efficiency fluctuation, high in operation complexity, poor in structural stability and insufficient in flexibility under different geological conditions can be effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of seismic exploration technology, and in particular to an intelligent seismic source for tunnel seismic advanced prediction based on SVM. BACKGROUND

[0002] In modern society, environmental awareness is gradually developing, and in the tunnel exploration industry, all factors are gradually considered to comprehensively select a suitable seismic source as an impact generating device. Explosive sources, as a technology that has been used for a long time and has obvious effects, have been widely used and can still be used in tunnel advanced prediction today. However, when using explosive sources, the following difficulties are faced: the safety of construction personnel, the destructive impact on the surrounding ecological environment, the delay of the progress of the construction task, and the increase of the operation cost. The geological radar method is susceptible to mechanical interference in the hole and has a short detection distance; the direct current method only has good survey results for water-bearing fault anomaly detection.

[0003] The pneumatic seismic source has better adaptability than the explosive source, the geological radar method, and the direct current method, and can clearly identify various abnormal geologies in tunnel geological exploration. At the same time, the pneumatic seismic source can adjust the contact area and mass of the hammer head according to different types of geology, and can obtain more accurate and clear geological data. The pneumatic seismic source is more green and safe than the explosive source in tunnel geological advanced prediction, and has better protection for the environment, which can ensure the safety of construction personnel. In terms of construction progress, the pneumatic seismic source has higher efficiency than the drilling technology and the explosive source, and the construction time is compressed from 3-4 hours to less than an hour, and the tunnel engineering can be implemented simultaneously during exploration. In terms of cost, the pneumatic seismic source has lower cost, reduces the number of construction personnel to 2, and reduces the labor cost and transportation and storage cost.

[0004] Considering various factors, the use of the pneumatic seismic source in tunnel advanced prediction exploration is becoming more and more widespread.

[0005] SVM is a supervised learning algorithm that can be used to solve multi-classification problems, and its core idea is to "find the optimal classification / regression boundary and maximize the model generalization ability", which performs particularly well in small sample and high-dimensional data scenarios. Its core advantages are:

[0006] Strong generalization ability in small samples: based on the "structural risk minimization" theory (rather than the "empirical risk minimization" of traditional algorithms), it can avoid overfitting in scenarios with small sample sizes (such as limited experimental data for pneumatic seismic source debugging) and ensure the prediction accuracy of new data.

[0007] Excellent performance in high-dimensional data: even if the feature dimension (such as the number of data attributes) is much larger than the sample size, SVM can efficiently handle it through kernel functions.

[0008] Strong anti-interference: only rely on support vector, not sensitive to noise data, high model robustness.

[0009] Good interpretability: the physical meaning of the optimal hyperplane is clear (maximize the interval), and the support vector can be explained "which samples play a key role in classification / regression", which is convenient for analyzing the model decision logic.

[0010] In the mode selection of adjusting the seismic source, the SVM is used for the most suitable structure prediction, and the "precise nonlinear modeling ability, small sample generalization ability and anti-interference" of the SVM are used to solve the core pain points of the adjustable pneumatic seismic source, such as "complex parameter coupling, few experimental samples, many working condition interferences and difficult multi-target balance", realize the upgrade of "from empirical adjustment to data-driven precise adjustment", and the value is particularly significant in the scene of high seismic source performance requirement in geological exploration.

[0011] Disadvantages of prior art

[0012] 1) Fixed structure limitation: the contact area and mass of the traditional hammer head are fixed, and it is difficult to meet the needs of different geological environments such as soft soil and hard rock, resulting in large fluctuation of energy transmission efficiency (10%~60%).

[0013] 2) Complex operation: in the case of limited geological conditions, it is very cumbersome to know which type of hammer head combination has better geological data.

[0014] 3) Abnormal signals cannot be automatically identified: when the pneumatic seismic source fails to excite, the generated signal is an abnormal signal, which cannot be used as an effective signal for interpretation, and the traditional pneumatic seismic source cannot automatically identify and remind, and must be identified and analyzed by professional personnel.

[0015] 4) Lack of flexibility: the structure of the existing pneumatic seismic source cannot be flexibly transported in the tunnel environment, and it is difficult to adapt to the complex environment in the tunnel.

[0016] 5) Large environmental damage: explosive seismic source causes irreparable damage to the internal environment of the tunnel and causes damage to the environment. Summary of the invention

[0017] The present application aims to solve the problems of the prior art. An intelligent seismic source for tunnel seismic advanced prediction based on SVM is proposed. The technical scheme of the present application is as follows:

[0018] The intelligent seismic source for tunnel seismic advanced prediction based on SVM comprises:

[0019] A hammer head adjusting structure, the hammer head adjusting structure comprises a hammer head replacing device and a replaceable hammer head, the replaceable hammer head is used to replace according to geological conditions to adapt to different environments;

[0020] A mass adjustment structure is provided, which has a specialized hammer rod design. The hammer rod is equipped with a structure that allows for the addition of counterweights to change the overall mass of the adaptive pneumatic vibration source device and improve its adaptability.

[0021] The connection protection structure is equipped with a protective device to prevent damage to the internal connection hole and the threads of the hammer replacement device, ensuring the structural stability of the adaptive adjustable pneumatic vibration source device under high impact force. The specific structure of the hammer replacement device has a mechanical driving relationship with the connection protection structure. The counterweight added in the mass adjustment structure is related to the position of the connection protection structure to ensure that the connection stability between the hammer replacement device and the adaptive adjustable pneumatic vibration source device is not damaged while adjusting the overall mass.

[0022] Furthermore, it also includes a tracked power structure, which is arranged around the adaptive adjustable pneumatic vibration source device to drive the adaptive adjustable pneumatic vibration source device to move within the tunnel.

[0023] Furthermore, it also includes a battery power supply system, which is installed on the adaptive adjustable pneumatic vibration source device and electrically connected to the hammer replacement device and the mass adjustment structure to provide the power required for operation.

[0024] Furthermore, the adaptive adjustable pneumatic vibration source device has a modular design, and the battery power supply system, as one of the modular components, is detachably connected to the other components of the adaptive adjustable pneumatic vibration source device.

[0025] Furthermore, it also includes a support vector machine (SVM), which is used to select the optimal combination of hammerhead structures based on geological information.

[0026] Furthermore, the hammer rod is provided with an extension structure, which surrounds the hammer head to increase the additional space for counterweight.

[0027] Furthermore, a protective device is provided at the connection between the hammer rod and the hammer head replacement device. The protective device includes a threaded protective component surrounding the hammer rod to protect the threads from damage under high impact force.

[0028] Furthermore, the counterweight installation structure is positioned in relation to the hammer replacement device, ensuring that the connection stability between the hammer replacement device and the hammer rod is not affected during the overall mass adjustment of the adaptive pneumatic vibration source device.

[0029] Furthermore, it also includes a closed air pump, which is installed inside the adaptive adjustable pneumatic vibration source device and is used to independently generate pneumatic pressure.

[0030] Furthermore, the adaptive adjustable aerodynamic source device also includes a time-domain pulse peak energy threshold screening mechanism, which is set in the signal processing module of the adaptive adjustable aerodynamic source device to identify and exclude low-quality and invalid signals in real time.

[0031] The advantages and beneficial effects of this invention are as follows:

[0032] The invention comprises: an adaptive adjustable pneumatic vibration source consisting of two parts that can be independently adjusted, with replaceable hammers of different contact areas and a counterweight structure with variable mass.

[0033] Specialized pneumatic vibration source hammer head structure: To accommodate the addition of counterweight structures, the hammer head length is extended, allowing for sufficient counterweight placement. Additionally, connecting holes are added inside the vibration source hammer head to facilitate connection to the structural components.

[0034] Optimize the energy supply structure: abandon the traditional battery structure and choose direct energy supply inside the construction site to reduce the bottleneck of battery energy.

[0035] The pneumatic vibration source features a lightweight and modular design: the battery, pneumatic vibration source, and encapsulated air pump are integrated into a modular design, allowing for disassembly and installation. This facilitates long-distance transportation and use by construction personnel.

[0036] Integrated air pump enclosure: The air pump is encapsulated in a cuboid structure, along with the air pump control device and other equipment. This not only protects the air pump structure but also lowers the barrier to entry for use, reduces manual operation, and achieves a high degree of automation.

[0037] Combined with efficient area detection algorithms: The collected data is detected in real time using algorithms to effectively identify abnormal signals.

[0038] Selection of hammer combination modes based on SVM: The limited hammer combination modes (contact area and mass of different models) are selected by supervising learning algorithm "SVM". SVM is trained by a database composed of collected optimal hammer combination mode data. The limited geological conditions on site are transformed into specified data, and SVM is used to predict the optimal hammer combination. Attached Figure Description

[0039] Figure 1 This invention provides a preferred embodiment of a three-view diagram of an aerodynamic vibration source;

[0040] Figure 2 This is a schematic diagram of the hammer head structure of the hammer impact source according to a preferred embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram of the hammer replacement device;

[0042] Figure 4 This is a schematic diagram of the internal connection holes of the hammer rod;

[0043] Figure 5 This is a diagram of the hammer head after it has been fitted with a hammer;

[0044] Figure 6 This is a schematic diagram of the hammer rod after the addition of a counterweight;

[0045] Figure 7 These are three views of the integrated air pump structure;

[0046] Figure 8 This is a perspective view of an integrated air pump vibration source. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0048] The technical solution of the present invention to solve the above-mentioned technical problems is:

[0049] 1. Reduce the amplitude of energy transfer efficiency fluctuations: Reduce the amplitude of energy transfer efficiency fluctuations of the air gun source in different geological conditions, making it more stable, reducing the difficulty of future data analysis, and improving data accuracy.

[0050] 2. Integrated Air Pump Packaging: The air pump and control unit are integrated into a single package, reducing manual operation and increasing the proportion of automated operation. The air pump is encapsulated in a cuboid structure, protecting it from impact damage.

[0051] 3. Reduce operational complexity: Simplify complex operations, lower the professional requirements for construction personnel, improve applicability, and reduce financial expenditure. Increase the proportion of automation within the seismic source.

[0052] 4. Environmental and Safety Design: Compressed air is used as the power source to avoid chemical pollution. The pneumatic vibration source will not cause destructive damage to the tunnel interior and provides better protection for the safety of construction personnel.

[0053] 5. Selection of Hammer Combination Patterns Based on SVM: The supervised learning algorithm "SVM" is used to select a limited number of hammer combination patterns (contact area of ​​different models and whether to add counterweights). The SVM is trained on a database of collected optimal hammer combination pattern data. The limited geological conditions on site are transformed into specified data, and the SVM performs optimal hammer combination prediction.

[0054] 6. Real-time on-site screening: Use "time-domain pulse peak + energy threshold" to quickly eliminate signals without effective excitation (such as no peak value or too low energy) to avoid the accumulation of invalid data.

[0055] 7. Tracked structure: Choosing to use a tracked structure instead of a wheeled structure can adapt to more complex and harsh environments in tunnels, while also being more reliable and less likely to suffer structural damage.

[0056] 8. Battery power supply: Using batteries as the power source to power the motor and air pump not only eliminates the need for on-site environmental restrictions and reduces time wasted due to communication between departments, but also gives the seismic source vehicle greater flexibility and better adaptability to complex environments.

[0057] 9. Thread protection design: To prevent damage to the internal thread structure of the hammer head due to the force, a physical resistance design is provided at the connection contact surface.

[0058] 10. Specialized extended hammer head structure design: In order to change the overall mass of the hammer impact source, the hammer head is specially designed with an extended hammer head structure, which can accommodate a larger counterweight.

[0059] Working principle:

[0060] The first step is to collect specific geological information within the tunnel in advance and feed it into an SVM (Support Vector Machine) to determine the optimal hammerhead structure combination. The SVM database is built from data obtained from previous experiments. After training the SVM using this database, it can predict the hammerhead structure combination most likely suitable for the geological conditions. Then, the hammerheads are configured according to the SVM predictions, and this structure combination can obtain better geological signals in this tunnel situation.

[0061] The second step is on-site operation. Upon arrival at the site, the seismic source trolley moves through the tunnel using its self-powered track system to the pre-positioned hammering location. Once in position, the trolley activates its internal air pump. After sufficient air pressure is reached, the seismic source is triggered for hammering. Based on geological exploration requirements, after the initial seismic source activation, invalid or low-quality signals are eliminated using real-time "time-domain pulse peak + energy threshold" analysis. Supplementary hammering is then performed at the corresponding location to supplement signals that do not meet geological requirements.

[0062] This pneumatic vibration source vehicle is divided into three parts: the upper part is the pneumatic vibration source, the middle part is the enclosed air pump structure, and the lower part is the tracked power part.

[0063] The enclosed air pump structure provides air pressure to the impact source through a flexible air tube, and the battery at the rear of the air pump structure powers the entire source vehicle. The enclosed structure of the source vehicle enhances structural independence, improves efficiency, and reduces preparation work for operation.

[0064] The seismic source trolley uses a tracked power structure as its power source, providing high flexibility and adapting to various environments inside the tunnel; during construction, its own weight acts as a reaction force to buffer the seismic source.

[0065] The pneumatic source hammer structure of the seismic source trolley can use hammers of different specifications to obtain more effective geological data. The impact source generates impact on the tunnel sidewall through air pressure. The mass and contact area of ​​the hammer rod can be adjusted to correspond to different geological conditions to meet the working requirements of the device for signal collection. This application provides an adaptive adjustable pneumatic vibration source device, comprising: a hammer head adjustment structure, which includes a hammer head replacement device and a replaceable hammer head, the replaceable hammer head being used to replace the hammer head according to geological conditions to adapt to different environments; a mass adjustment structure, which has a specialized hammer rod design, the hammer rod having a structure for adding counterweights to change the overall mass of the adaptive adjustable pneumatic vibration source device and improve adaptability; and a connection protection structure, which has a protective device to prevent damage to the internal connection holes and the threads of the hammer head replacement device, ensuring the structural stability of the adaptive adjustable pneumatic vibration source device under high impact forces. The specific structure of the hammer head replacement device and the connection protection structure have a mechanical driving relationship, and the counterweight addition in the mass adjustment structure is related to the positional relationship of the connection protection structure, to ensure that the connection stability between the hammer head replacement device and the adaptive adjustable pneumatic vibration source device is not compromised while adjusting the overall mass.

[0066] The adaptive pneumatic vibration source device, employing the technical solution of this embodiment, includes a hammer replacement device and a specialized hammer rod design. By replacing different hammers, it adapts to various geological conditions, and by adding counterweights to the hammer rod, the overall mass is adjusted. This design leverages the flexibility of the hammer replacement device and the adaptability of the mass adjustment structure, solving the problem of large fluctuations in energy transfer efficiency under different geological environments. The connection protection structure prevents damage to the threaded structure under high impact forces, ensuring the structural stability of the device when adjusting the hammers and counterweights, reducing operational complexity, and improving the reliability and service life of the device. By monitoring the status of the hammer replacement device and the counterweight addition, the operating parameters of the pneumatic vibration source device can be adjusted in real time, enabling it to achieve optimal operating conditions under various complex geological conditions. This improves the accuracy and efficiency of tunnel advance prediction and solves the problems of complex operation and unstable energy transfer efficiency of traditional pneumatic vibration source devices under different geological conditions.

[0067] Furthermore, in this embodiment, a tracked power structure is also included, which is arranged around the adaptive adjustable pneumatic vibration source device to drive the adaptive adjustable pneumatic vibration source device to move within the tunnel.

[0068] In this embodiment, the adaptive pneumatic seismic source device integrates a replaceable hammerhead structure, an adjustable counterweight system, and an integrated air pump and battery power supply component. Driven by a tracked power structure, it achieves efficient and flexible movement in complex tunnel environments. Utilizing an SVM algorithm, the device quickly selects the most suitable hammerhead combination from six preset modes (large, medium, and small contact area hammers, and whether or not counterweights are added) based on on-site geological information such as rock strength, water inflow status, weathering degree, and fracture level. This intelligent selection mechanism overcomes the performance fluctuation problem of traditional seismic sources under different geological conditions, significantly improving the stability of energy transfer efficiency. The integrated air pump structure not only protects the air pump from external environmental damage but also reduces manual operation, enhancing the system's automation and reliability. The real-time on-site screening function quickly eliminates invalid signals using the "time-domain pulse peak + energy threshold" standard, ensuring data quality and laying the foundation for subsequent accurate geological analysis. Furthermore, the lightweight and modular design of the pneumatic seismic source device facilitates transportation and installation by construction personnel, reducing operational difficulty and costs, and improving overall work efficiency and safety. The tracked power structure design enables the device to adapt to various terrains within tunnels, ensuring the smooth progress of exploration work. In summary, the adaptive adjustable pneumatic seismic source device of this embodiment, through the combination of intelligent algorithms and structural innovation, achieves high-precision and high-efficiency earthquake prediction under complex geological conditions.

[0069] Furthermore, in this embodiment, a battery power supply system is also included. The battery power supply system is mounted on the adaptive adjustable pneumatic vibration source device and electrically connected to the hammer replacement device and the mass adjustment structure to provide the power required for operation.

[0070] In this embodiment, the adaptive pneumatic vibration source device integrates a battery power system. This system is closely connected to the hammer replacement device and the mass adjustment structure, and is responsible for supplying the power required for the entire device to operate. This design eliminates the limitation of external energy sources, enhances the portability and independent operation capability of the vibration source device, and eliminates the need for additional wiring or carrying bulky traditional power supplies in tunnel environments, reducing operational complexity and dependence on the site. The introduction of the battery power system not only simplifies on-site preparation and reduces safety risks, but also improves work efficiency. Especially in complex working conditions requiring rapid movement and response, the immediacy and flexibility of battery power are particularly prominent, ensuring the continuity and accuracy of geological exploration work.

[0071] Furthermore, the modular design allows for easy maintenance and replacement of the battery power system, further improving equipment reliability and lifespan. Combined with the intelligent selection mechanism of the SVM (Search Engine Monitoring System), the battery power system provides a stable and timely energy source for the pneumatic seismic source, enabling the equipment to quickly adjust according to the optimal hammer combination pattern predicted by the SVM, achieving efficient and environmentally friendly geological data acquisition. Overall, the battery power system reduces manpower requirements while also enhancing the application efficiency and technological advancement of pneumatic seismic sources in the field of tunnel advance prediction.

[0072] Furthermore, in this embodiment, a modular design is adopted, with the battery power supply system being one of the modular components, which is detachably connected to other components of the adaptive adjustment aerodynamic vibration source device.

[0073] In this embodiment, the pneumatic vibration source device features a modular design, with the battery power supply system detachably connected to the main body of the pneumatic vibration source as an independent module, enhancing the device's portability and ease of maintenance. This modular design not only simplifies transportation but also allows construction personnel to quickly replace and repair components on-site, ensuring operational continuity and efficiency. Combined with an integrated air pump structure, the device can operate autonomously without external power, reducing dependence on external conditions and enhancing adaptability and flexibility in complex environments such as tunnels. Furthermore, the introduction of the SVM algorithm enables real-time, intelligent hammer combination mode selection based on on-site geological conditions, significantly improving the accuracy and efficiency of geological data acquisition. The on-site real-time anomaly signal filtering mechanism further ensures data quality, avoiding interference from invalid data and making subsequent data analysis more accurate and reliable. In summary, the pneumatic vibration source device provided in this embodiment not only optimizes energy transfer efficiency and reduces operational complexity but also improves its applicability and safety in tunnel advance prediction through intelligent algorithms and modular design, providing a more efficient and accurate geological exploration method for tunnel engineering.

[0074] Furthermore, in this embodiment, a support vector machine (SVM) is also included, which is used to select the optimal combination of hammerhead structures based on geological information.

[0075] In this embodiment, a Support Vector Machine (SVM) is integrated into the pneumatic seismic source system to select the optimal hammerhead structure combination based on geological information collected on-site, such as rock strength, water inrush status, weathering degree, and fracture level. The SVM, through a pre-trained database containing past experimental data and the correspondence between geological features and hammerhead combinations, can quickly and accurately predict the hammerhead contact area suitable for the current geological conditions and whether additional counterweight is needed. This design greatly simplifies the on-site operation process, avoids the uncertainty of manual experience-based adjustments, and ensures that the pneumatic seismic source operates optimally in different geological environments, thereby improving the accuracy and efficiency of geological exploration. The SVM's prediction results directly guide the hammerhead replacement device on the seismic source trolley for rapid replacement, automating the hammerhead and counterweight adjustment process and reducing the complexity of manual operation. Combined with a real-time on-site signal screening algorithm, invalid or low-quality excitation signals can be instantly eliminated, ensuring high-quality data acquisition and further improving the accuracy and reliability of geological analysis. The integrated design of the air pump structure and battery power supply not only enhances the system's portability but also improves its adaptability and safety in tunnel environments. Overall, this invention enables precise adjustment and optimization of aerodynamic sources in tunnel advance prediction exploration through intelligent selection of SVM and efficient real-time on-site detection, significantly improving the efficiency and data quality of exploration work.

[0076] Furthermore, in this embodiment, the hammer rod is provided with an extension structure, which surrounds the hammer head to increase the additional space for the counterweight.

[0077] In this embodiment, the extended structure of the hammer rod surrounds the hammer head, providing additional space for mounting a counterweight. This design allows for adjustment of the hammer head mass according to specific geological conditions, enhancing the adaptability of the pneumatic source to different geological environments. By precisely controlling the hammer head mass and contact area, energy transfer efficiency can be optimized, improving the accuracy and clarity of geological signal acquisition. The extended structure not only expands the functionality of the hammer head but also ensures stable installation of the counterweight, reducing operational complexity and enabling on-site personnel to configure the pneumatic source more quickly and effectively to adapt to varying tunnel geological conditions. Furthermore, the flexible addition of the counterweight refines the adjustment range of the pneumatic source, allowing for the acquisition of high-quality geological data even in transition zones between soft soil and hard rock, thereby comprehensively improving the reliability and efficiency of advanced prediction.

[0078] Furthermore, in this embodiment, a protective device is provided at the connection between the hammer rod and the hammer head replacement device. The protective device includes a threaded protection component surrounding the hammer rod, which is used to protect the threads from damage under high impact force.

[0079] In this embodiment, a protective device is provided at the connection between the hammer rod and the hammer head replacement device. This device specifically includes a threaded protective component surrounding the hammer rod. Through the design of a physical baffle, the threaded protective component can effectively resist the reaction force under high impact, preventing damage to the internal connecting hole of the hammer rod and the threaded structure of the hammer head replacement device, thus ensuring the reliability and durability of hammer head replacement and counterweight installation. This design avoids difficulties in hammer head replacement due to thread damage in complex geological environments, thereby maintaining the efficient operation of the pneumatic vibration source, reducing maintenance costs, and improving equipment lifespan and operational safety. In practical applications, the design flexibility of the threaded protective component means that it can adapt to hammer heads of various sizes and types, providing a solid physical basis for the precise adjustment of the pneumatic vibration source under different geological conditions, thereby ensuring the accuracy and stability of geological data acquisition.

[0080] Furthermore, in this embodiment, the counterweight installation structure is positioned in relation to the hammer replacement device, ensuring that the connection stability between the hammer replacement device and the hammer rod is not affected during the overall mass adjustment of the adaptive pneumatic vibration source device.

[0081] In this embodiment, the counterweight addition structure of the adaptively adjustable pneumatic seismic source is positionally correlated with the hammer replacement device, ensuring the connection stability between the hammer replacement device and the hammer rod when adjusting the overall mass. This design allows the pneumatic seismic source to dynamically adjust the hammer contact area and hammer rod mass according to changes in on-site geological conditions, thereby optimizing the energy transfer efficiency of seismic waves. By integrating the air pump, control device, and battery into a single package, not only is the air pump protected from external damage, but the operation process is also simplified, improving the degree of automation. Combined with the SVM algorithm, the optimal hammer combination is predicted from the database based on characteristics such as rock strength, water inrush state, weathering degree, and fracture level, realizing the transformation from experience-based adjustment to data-driven adjustment. In addition, the use of battery power and tracked movement increases the flexibility and environmental adaptability of the device, especially in confined spaces such as tunnels, effectively improving work efficiency. On-site screening mechanisms using time-domain pulse peaks and energy thresholds can immediately eliminate invalid or low-quality signals, ensuring the accuracy and reliability of subsequent data analysis, ultimately achieving high-quality acquisition and analysis of geological exploration signals. In other embodiments, the durability and ease of maintenance of the aerodynamic source can be improved by further optimizing the design of the internal connection holes of the hammer rod and the physical baffle structure of the hammer head, thereby reducing the risk of signal distortion caused by internal wear or damage.

[0082] Furthermore, in this embodiment, a closed air pump is also included. The closed air pump is disposed inside the adaptive adjustable pneumatic vibration source device and is used to independently generate pneumatic pressure.

[0083] In this embodiment, the enclosed air pump 2 is housed within the adaptively adjustable pneumatic vibration source device to independently generate pneumatic pressure. This design allows the pneumatic vibration source to be self-sufficient, eliminating reliance on an external air source and enhancing the system's portability and ease of operation. The enclosed air pump 2 is connected to the pneumatic vibration source via a flexible air tube, enabling it to move freely within the tunnel while continuously providing stable and sufficient air pressure. This ensures consistent energy release with each impact, thereby improving the quality and consistency of geological data acquisition. Furthermore, the integrated encapsulation of the enclosed air pump 2 not only protects the pump from the complex environment within the tunnel but also reduces the time and manpower required for on-site deployment and operation, making the entire system more compact and reliable. This allows for higher efficiency in tunnel advance prediction, reduces fluctuations in energy transfer efficiency, and optimizes the geological data acquisition process.

[0084] The innovative design of the adaptive pneumatic vibration source device, combined with a closed air pump 2, demonstrates its high flexibility and adaptability in the face of different geological environments. By selecting the optimal hammer combination mode through SVM and supplementing it with real-time on-site screening using "time-domain pulse peak + energy threshold," it ensures that each hammer strike generates an effective signal that meets the requirements of geological exploration, greatly improving data availability and prediction accuracy. This solution, which comprehensively considers energy efficiency, operational simplification, environmental protection, and personnel safety, marks a significant step forward for tunnel advance prediction technology towards intelligence and efficiency.

[0085] Furthermore, in this embodiment, a time-domain pulse peak energy threshold screening mechanism is also included. The screening mechanism is set in the signal processing module of the adaptive adjustment aerodynamic vibration source device to identify and exclude low-quality and invalid signals in real time.

[0086] In this embodiment, the time-domain pulse peak energy threshold screening mechanism is integrated into the signal processing module of the adaptively adjusted aerodynamic source. This mechanism monitors the energy peak value and duration of the excitation signal to instantly identify and eliminate low-quality signals generated due to excitation failure or insufficient energy. Specifically, after the aerodynamic source excites vibration signals from the tunnel sidewall, the signal processing module quickly analyzes the time-domain characteristics of these signals. If the pulse peak value is below a set threshold or the energy distribution does not meet a preset standard, it is determined to be an abnormal signal and immediately blocked, not included in subsequent data analysis. This screening method not only reduces the accumulation of invalid data but also effectively improves data quality, ensuring the accuracy of subsequent SVM predictions of geological conditions. Through this mechanism, on-site operators can focus on high-quality signal analysis, greatly improving work efficiency and reducing the complexity of data analysis.

[0087] Furthermore, in this embodiment, a database is also included, which stores historical data and model training results to assist on-site operators in quickly determining the optimal combination of hammerhead structures.

[0088] In this embodiment, the pneumatic seismic source trolley integrates an adaptive hammerhead adjustment device, an integrated air pump structure, and a tracked mobile system, achieving efficient, flexible, and safe operation in tunnel geological exploration. The adaptively adjustable pneumatic seismic source optimizes for different geological conditions by replacing hammerheads with different contact areas and using a variable-mass counterweight structure, improving the stability of energy transfer efficiency. The integrated air pump structure not only encapsulates the air pump and control device but also incorporates a battery power system, reducing external dependence and enhancing the equipment's independence and adaptability. The tracked design ensures the reliable movement of the seismic source trolley in complex tunnel environments. Simultaneously, the hammerhead replacement device and specialized extended hammerhead structure design allow for rapid hammerhead replacement and counterweight addition without damaging the internal threads, meeting diverse geological exploration needs. Combined with SVM's intelligent mode selection, based on characteristics such as rock strength, water inflow status, weathering degree, and fracture level, the most suitable hammerhead combination mode is predicted, further improving the quality and efficiency of geological data acquisition. Real-time signal screening on-site quickly identifies and eliminates invalid or low-quality signals, ensuring the accuracy and reliability of subsequent analysis. In addition, the database's decision support function stores historical data and algorithm training results, providing on-site operators with immediate references, simplifying operating procedures, and improving decision-making efficiency.

[0089] In the process of using the adaptive adjustable pneumatic seismic source device of this application for tunnel earthquake early prediction, the on-site operator first collects geological information within the tunnel, including key parameters such as rock strength, water inflow status, weathering degree, and fracture grade. Subsequently, this data is input into a pre-set Support Vector Machine (SVM) database. Through intelligent algorithm analysis, the best hammer structure combination matching the current geological conditions is selected from six preset modes with different contact areas and whether or not counterweights are added. Based on the SVM prediction results, the operator quickly adjusts the hammer contact area and overall mass of the pneumatic seismic source using a hammer replacement device and mass adjustment structure. After adjustment, the pneumatic seismic source uses high-pressure gas generated inside an integrated encapsulated air pump, which is transmitted to the hammer through a flexible air tube to perform the hammering action. After each hammering, the signal processing module immediately activates a time-domain pulse peak energy threshold screening mechanism to analyze the excitation signal in real time, eliminating invalid or low-energy signals to ensure data quality. Furthermore, the battery power system onboard the device provides stable power support throughout the process, eliminating the need for external power supply and enhancing independent operation capabilities within the tunnel. Driven by a tracked power structure, the adaptively adjustable pneumatic vibration source device moves flexibly inside the tunnel to the predetermined hammering position, preparing for the next stage of exploration. The entire process is efficient and automated, significantly reducing operational complexity and improving the accuracy and efficiency of geological exploration.

[0090] When the pneumatic seismic source trolley operates in a tunnel environment, construction personnel first input geological information from the tunnel, including rock hardness, groundwater conditions, rock weathering degree, and fracture development, into a pre-set Support Vector Machine (SVM) system. Based on its trained model, the SVM system automatically selects the most suitable hammer structure for the current geological conditions from six pre-stored combinations of hammer contact area size (large, medium, small) and whether or not counterweights are added. Subsequently, based on the SVM's predictions, the hammer is replaced using a hammer replacement device, and a corresponding weight is added using the hammer rod extension structure to adjust the overall mass of the pneumatic seismic source trolley, thereby optimizing the seismic wave excitation effect. During hammer replacement and counterweight addition, physical baffles in the connecting protective structure effectively prevent potential damage to the internal threads and connecting holes of the hammer rod from high impact forces. Once ready, the trolley activates its battery power system, and the integrated encapsulated air pump structure begins inflation. When the air pressure reaches the predetermined value, the pneumatic seismic source is triggered to strike the hammer. The signal processing module on the trolley then activates, using a screening mechanism of "time-domain pulse peak + energy threshold" to analyze the signal generated by each hammer strike in real time, automatically identifying and eliminating invalid or low-quality signals to ensure the accuracy and reliability of the collected data. Throughout the operation, the tracked structure provides stable power support for the trolley, enabling it to move freely within the tunnel and adapt to complex terrain. Through the above steps, the adaptively adjustable pneumatic vibration source trolley can intelligently select the optimal hammer combination according to different geological conditions, achieving efficient and accurate geological data acquisition and providing strong technical support for tunnel advance prediction.

[0091] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0092] Three views of the overall structure of the aerodynamic source Figures 1-8 As shown:

[0093] Adaptive hammerhead adjustment device: The hammerhead's contact area with the tunnel is changed via a hammerhead replacement device, allowing for the selection and installation of appropriate structures. A counterweight can also be added behind the hammerhead to adapt to different geological types. The hammerhead contact surface diameters are 6cm, 12cm, and 18cm; the counterweight is 10kg. By changing the hammerhead's contact area and hammer shaft mass, different amplitudes and bandwidths can be obtained to cope with different geological types and obtain clearer images.

[0094] Anti-thread damage structure: A physical baffle structure is used to prevent damage to the threads of the internal connecting hole of the hammer rod and the hammer head replacement device due to reaction force, which would prevent the counterweight and hammer head from being changed.

[0095] Battery power: Choosing to use batteries as a power source to power the motor and air pump not only eliminates the limitations of the construction site environment and reduces the time wasted due to communication between departments, but also gives the seismic source vehicle greater flexibility and better adapts to complex environments.

[0096] Integrated air pump packaging: The air pump structure, control device, battery and other devices are packaged in the same structure. This integrated design reduces operational complexity, lowers the probability of equipment damage, and avoids the loss of structural parts.

[0097] Tracks: Choosing to use tracks as the mobility structure allows for better adaptation to tunnel environments.

[0098] SVM (Support Vector Machine)

[0099] The support vector machine dataset has four features: rock strength (MPa), water inrush state (level), weathering degree (level), and fracture level (level); and six labels: three areas (large, medium, and small) and whether or not weight is added, forming six labels. The relationship between features and labels is as follows:

[0100]

[0101] Table 1

[0102] During training, Support Vector Machines (SVMs) employ a one-to-one strategy, using the RBF (Gaussian kernel) as the kernel function, with regularization parameters C=100 and gamma=1. The kernel function maps data from a low-dimensional feature space to a high-dimensional feature space. A larger c penalizes misclassification, resulting in high accuracy on the training set but weak generalization ability. A smaller c penalizes misclassification less, allowing errors to be treated as noise points, leading to stronger generalization ability. Gamma controls the influence of sample points on the decision boundary; a smaller gamma value reduces the risk of overfitting, while a larger value increases the risk.

[0103] Using SVM to train the database, with the training set and test set divided into a 7:3 ratio, the final selection of the corresponding hammer combination reached 94.5%.

[0104] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0105] The above embodiments should be understood as illustrative only and not as limiting the scope of protection of the present invention. After reading the description of the present invention, those skilled in the art can make various alterations or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. An SVM-based tunnel earthquake early warning intelligent seismic source, characterized in that, Comprise: Hammer head adjusting structure, the hammer head adjusting structure includes hammer head replacement device and replaceable hammer head, the replaceable hammer head is used to replace according to geological conditions to adapt to different environment; Mass adjusting structure, the mass adjusting structure has specialized hammer rod design, the hammer rod is provided with the structure of additional assembly weight, to change the overall mass of the adaptive adjusting pneumatic vibrator device, improve adaptability; Connection protection structure, the connection protection structure is provided with the protection device that prevents the internal connection hole and the thread of the hammer head replacement device are damaged, ensure that the adaptive adjusting pneumatic vibrator device is stable under high impact force, wherein the specific structure of the hammer head replacement device and the existence mechanical drive relationship of the connection protection structure, the position relationship of the additional assembly weight in the mass adjusting structure is associated with the connection protection structure, to ensure that the overall mass is adjusted while not damaging the connection stability of the hammer head replacement device and the adaptive adjusting pneumatic vibrator device.

2. The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1, characterized in that, Also include track power structure, the track power structure is arranged around the adaptive adjusting pneumatic vibrator device, for driving the adaptive adjusting pneumatic vibrator device moves in the tunnel. 3.The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1 or 2, characterized in that, Also include battery power supply system, the battery power supply system is arranged on the adaptive adjusting pneumatic vibrator device, is electrically connected with the hammer head replacement device and the mass adjusting structure, for providing the power required for operation.

4. The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 3, characterized in that, The adaptive adjusting pneumatic vibrator device has a modular design, the battery power supply system is one of the modular components, and is detachably connected with other components of the adaptive adjusting pneumatic vibrator device.

5. The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1, characterized in that, Also include support vector machine (SVM), the support vector machine is used to select the optimal hammer head structure combination according to geological information.

6. The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1 or 5, characterized in that, The hammer rod is provided with an extension structure, and the extension structure is arranged around the hammer head, for increasing the additional space of the weight.

7. The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 6, characterized in that, The connection between the hammer rod and the hammer head replacement device is provided with a protection device, and the protection device includes a thread protection component around the hammer rod, for protecting the thread from being damaged under high impact force.

8. The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1 or 7, characterized in that, The additional assembly weight structure is associated with the position of the hammer head replacement device, to ensure that the connection stability of the hammer head replacement device and the hammer rod is not affected during the overall mass adjustment of the adaptive adjusting pneumatic vibrator device. 9.The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1, characterized in that, Also include closed air pump, the closed air pump is arranged inside the adaptive adjusting pneumatic vibrator device, for independently generating pneumatic pressure. 10.The SVM-based tunnel seismic advanced prediction intelligent seismic source according to claim 1, characterized in that, The adaptive adjusting pneumatic vibrator device also includes a time-domain pulse peak energy threshold screening mechanism, the screening mechanism is arranged in the signal processing module of the adaptive adjusting pneumatic vibrator device, for real-time identification and exclusion of low-quality and invalid signals. SVM-based hammer head structure selection: SVM is a supervised learning algorithm that can be used to solve multi-classification problems. It can be trained by collecting data sets, divided into training sets and test sets. After training the training set, the "optimal classification" of internal differentiation of different options is obtained. Then the test set is used to verify its accuracy, which has strong robustness and generalization ability. In this seismic source, the field geological conditions are converted into specified data types and input into the SVM. The hammer head combination most suitable for the field geological conditions is predicted by SVM, improving efficiency.

Images