Intelligent geological condition sensing system of cantilever tunneling machine based on dual-band sound wave analysis

Abstract

The invention provides an intelligent geological condition sensing system of a cantilever tunneling machine based on dual-band sound wave analysis, which can solve the problem that the geological condition encountered by the cantilever tunneling machine during construction is unclear, and realizes safe, efficient, intelligent and refined construction of the cantilever tunneling machine. The system is composed of a dual-band high-sensitivity sound pickup, a data wireless transmission antenna and an integrated industrial personal computer, and can collect sound wave data of a cutting head, a tunnel face and surrounding rock in the construction process of the cantilever tunneling machine in real time, and the sound wave data of the cutting head, the tunnel face and the surrounding rock can be obtained through processing methods such as wavelet denoising and Fourier transform on collected audio signals. The system can effectively remove noise signals, analyzes and identifies surrounding rock types in real time in combination with a machine learning algorithm, and can provide a scientific basis for selection of construction parameters of the cantilever tunneling machine, so that parameter setting in the construction process is optimized, the construction safety is improved, reliable support is provided for intelligent construction of tunnel engineering, and the construction efficiency is improved. And special requirements of construction of the cantilever heading machine are met.

Description

Technical Field

[0001] The present invention relates to tunnel geological identification and prediction technology, and in particular to an intelligent geological condition perception system for a cantilever tunnel boring machine based on dual-band acoustic wave analysis. Background Art

[0002] Cantilever TBMs (TBMs) have been widely used in tunnel construction due to their advantages, including minimal construction disturbance, high flexibility, low construction costs, and excellent construction quality. This is particularly true in areas with restricted blasting (such as airports, residential areas, and wildlife reserves) and in loose surrounding rock environments. These areas typically have stricter requirements for ground subsidence, so the key to reducing subsidence lies in accurately predicting the geological conditions ahead of the tunnel face, enabling the rational selection of excavation parameters and implementing refined construction. However, traditional TBM construction methods rely heavily on operator experience to select excavation parameters, lacking effective geological information. This empirically-based parameter selection can lead to inappropriate excavation plans, resulting in construction safety issues such as excessive surface subsidence. Therefore, to improve construction safety, enhance the intelligence level of TBM methods, and further promote refined construction management, there is an urgent need to develop an intelligent method that can accurately identify geological conditions to guide TBM operations. Currently, experts and scholars have recognized the challenges posed by unclear geological conditions in tunnel construction and have proposed various geological identification and prediction methods to achieve precise construction in complex geological environments: Chinese patent application number 202410780441.1 discloses a full-time domain induced polarization detection system and method for tunnel advanced geological prediction. The system receives observation data and performs inversion by controlling the control instructions of a constant current high-power transmitting module and a multi-channel parallel acquisition receiving module. During the inversion process, the minimum gradient support constraint is used instead of the traditional smooth constraint, and the data morphology is introduced as prior information into the inversion objective function to update the model parameters. This method can realize the acquisition of multi-source information such as apparent resistivity and polarizability, and perform inversion imaging of multiple parameters such as zero-frequency resistivity, relaxation time, polarizability, and frequency correlation coefficient, significantly improving detection efficiency and anti-electromagnetic interference capabilities. At the same time, this technology is particularly suitable for detecting the source of sudden water disasters ahead of the tunnel face during tunnel construction, meeting the urgent needs for tunnel safety and early warning capabilities. However, the system is complex and inconvenient to install, which affects the progress of tunnel construction.

[0003] Chinese patent application number 202111333726.3 discloses a method and system for tunnel advance geological prediction based on in-drilling perception of geochemical characteristics. The method first conducts advance drilling on several preset boreholes, and collects rock samples from each borehole in stages during the drilling process. By performing geochemical tests on the collected rock samples, the geochemical characteristics of different boreholes are obtained. Based on these staged rock sample positions and geochemical characteristic data, data fitting and spatial interpolation methods are used to construct a three-dimensional grid model in front of the tunnel face. By performing local singularity analysis on the three-dimensional grid model, abnormal areas are identified, thereby achieving accurate prediction of tunnel advance geology.

[0004] Chinese patent application number 201910335107.4 discloses a nondestructive testing method for tunnel linings based on audio analysis. This method uses a mobile terminal with a recording function to collect audio signals from percussion on the tunnel lining at a set sampling frequency. The collected percussion signals are cropped to retain the valid signal portion. A calibration threshold is calculated based on the parameters of the valid signal, and a void index is further calculated based on the characteristics of the valid signal. The calculated void index is compared with the calibration threshold, and the quality of the tunnel lining is determined by comparing the results. This method can only detect surface defects in existing tunnel linings and is not suitable for cantilever tunnel boring machine construction.

[0005] Chinese patent application number 202410322033.1 discloses an adaptive tunnel void identification method and device based on acoustic vibration detection technology. This method identifies tunnel lining voids through the following steps: establishing an acoustic-solid coupling finite element model of the tunnel and obtaining the acoustic response frequency band of the void; establishing a physical model of the concrete structure with the void, impacting the model with a hammer, and collecting acoustic signals through a microphone; constructing a variational modal decomposition model and using a particle swarm algorithm to optimize the number of decomposition levels and penalty factors. Based on the optimization results, the acoustic signal is modally decomposed to obtain multiple sub-signals; using wavelet packet decomposition and sensitive energy band analysis, the energy distribution and energy entropy of each sub-signal are calculated to identify void defects in the concrete. However, this device is only applicable to tunnels in the operational phase.

[0006] Chinese patent application number 202211628988.7 discloses a method and system for monitoring rock fracture and instability in tunnel excavation based on acoustic emission. This method deploys N data acquisition columns within a monitoring area and divides the area into N-1 monitoring sub-areas. When an acoustic emission source occurs, the monitoring sub-area where the acoustic emission source is located is determined by comparing and analyzing the acoustic emission ring counts, acoustic emission energy, and elastic wave arrival time obtained by each acquisition sensor. Within the identified monitoring sub-area, the acoustic emission source is further located by analyzing the time difference between the acoustic emission elastic waves arriving at the acquisition sensors on both sides of the area. Finally, based on the monitoring curves of acoustic emission ring counts and energy versus time, an early warning of surrounding rock fracture and instability is issued.

[0007] Currently, tunnel geological identification and prediction technologies primarily include physical and electromagnetic methods. However, both methods have limitations and are difficult to adapt to the environmentally sensitive construction methods of cantilever tunnel boring machines (TBMs). Traditional physical methods typically rely on explosives as a seismic source, which is not suitable for TBM construction environments. The use of explosives can cause significant disturbance to the construction site and surrounding environment. Furthermore, physical methods are highly sensitive to external factors such as noise, climate change, and construction activities within the geological environment, which can lead to increased measurement errors. Electromagnetic methods are subject to electromagnetic interference from electrical equipment and machinery on the construction site, which can affect the accuracy of measurement results and reduce the reliability of predictions. Furthermore, electromagnetic equipment is often complex and difficult to operate, and actual application on the construction site requires additional technical support and operational experience. Currently, there is no geological prediction method specifically designed for TBM construction. This technological gap results in a low level of intelligent operation for TBM construction, making it difficult to effectively improve construction efficiency and posing safety risks during refined construction. Therefore, there is an urgent need to develop a geological identification method suitable for TBM construction environments to enhance intelligent operation, ensure construction safety, and improve efficiency. Summary of the Invention

[0008] In response to the problems existing in the above-mentioned prior art, the present invention provides a cantilever tunnel boring machine geological condition intelligent perception system based on dual-band acoustic wave analysis, which can solve the problem of unclear geological conditions encountered by the cantilever tunnel boring machine during construction and realize its safe, efficient, intelligent and refined construction. The system consists of a dual-band high-sensitivity microphone, a data wireless transmission antenna and an integrated industrial control computer. It can collect real-time acoustic wave data of the cutting head, tunnel face and surrounding rock during the construction of the cantilever tunnel boring machine. By performing wavelet denoising, Fourier transform and other processing methods on the collected audio signals, the system can effectively remove noise signals and combine machine learning algorithms to analyze and identify surrounding rock types in real time. It can provide a scientific basis for the selection of construction parameters of the cantilever tunnel boring machine, thereby optimizing the parameter setting during the construction process, improving construction safety, and providing reliable support for the intelligent construction of tunnel engineering, meeting the special needs of cantilever tunnel boring machine construction.

[0009] The present invention is achieved through the following technical solutions: a cantilever tunnel boring machine geological condition intelligent perception system based on dual-band acoustic wave analysis, including a dual-band high-sensitivity sound pickup device, a data wireless transmission system and an acoustic wave signal processing system. The dual-band high-sensitivity pickup device includes a capacitive pickup and a dynamic pickup, which is responsible for collecting sound wave data from the cutting head, tunnel face and surrounding rock during the construction of the cantilever tunneling machine; The wireless data transmission system includes a storage unit and an antenna module located inside the pickup device, which are responsible for signal storage and wireless transmission. When the pickup captures an acoustic signal, the internal storage unit immediately records it. The storage unit uses an efficient data recording mechanism to ensure the integrity and accuracy of the acoustic signal. The recorded data files are wirelessly transmitted in real time to the terminal data processing platform via the built-in antenna module. After receiving the data, the terminal processing platform performs data processing and analysis to achieve real-time monitoring and evaluation of the surrounding rock type and structural surface of the tunnel face. The acoustic signal processing system uses dual-band high-sensitivity pickup devices placed in front and behind the boom tunnel boring machine to collect audio signals generated by the boom tunnel boring machine's cutting head when breaking rock. After noise reduction processing, the system extracts the frequency spectrum of the tunnel face during rock breaking to construct time-frequency domain characteristic parameters. Based on an established time-frequency domain characteristic library of rock mass types, the system identifies the current tunnel face rock type and infers its physical and mechanical properties, providing operators with detailed construction guidance.

[0010] As a further preferred technical solution, the dual-band high-sensitivity sound pickup device is installed in front of the cab and at the rear hydraulic control system of the boom tunnel boring machine.

[0011] As a further preferred technical solution, the capacitive pickup and the dynamic coil pickup are installed on a shock-absorbing bracket.

[0012] As a further preferred technical solution, the housing and shock-absorbing bracket of the pickup are made of high-density polyethylene (HDPE).

[0013] As a further preferred technical solution, the antenna module of the data wireless transmission system adopts a built-in layout.

[0014] As a further preferred technical solution, Fourier transform is performed on the signal data after background noise removal and noise reduction, and the signal data is converted from the time domain to the frequency domain, thereby obtaining the time-frequency domain information of the sound wave.

[0015] As a further preferred technical solution, the method for using the system includes the following steps: (1) Calibrate the microphone based on the specific conditions of the construction site to ensure the resolution and accuracy of data collection; (2) The device performs self-tests to confirm that all sensors are working properly, check whether the wireless connection is stable, and whether the terminal platform can synchronize information to ensure the normal operation of the entire device system; (3) Collecting audio data of ambient noise in the tunnel when the boom tunnel boring machine is not started; (4) Start the cantilever tunnel boring machine, but do not cut, to collect the noise generated by the cantilever tunnel boring machine; (5) After the boom tunnel boring machine starts cutting, the intelligent terminal system performs preliminary processing based on the real-time collected sound wave data, including filtering, noise reduction and other steps to filter out invalid noise signals and extract effective signals reflecting the geological conditions; (6) Identify the rock type of the current tunnel face through machine learning algorithms and guide the boom tunneling machine operator to optimize the tunneling parameters; (7) After completing the excavation operation of a section, check the status of the pickup to ensure that the pickup can maintain the best working condition during the next section excavation.

[0016] The advantages of the present invention are: (1) The present invention provides an intelligent sensing method for geological conditions of a cantilever tunnel boring machine based on dual-band acoustic wave analysis. The method collects acoustic wave data of the cutting head, tunnel face and surrounding rock during the construction of the cantilever tunnel boring machine in real time through a dual-band high-sensitivity microphone, and analyzes and identifies the surrounding rock type in real time.

[0017] (2) The dual-band high-precision pickup consists of a dynamic pickup responsible for collecting low-frequency signals and a capacitive pickup responsible for collecting high-frequency signals. The mechanical design reduces the transmission of the cantilever tunnel boring machine body vibration to the pickup, and the digital signal processing method reduces the impact of environmental noise on the pickup, ensuring the quality of the audio data collected by the pickup.

[0018] (3) This device can analyze the sound wave information generated when cutting rocks through the main and auxiliary pickups installed in the front and rear hydraulic control systems of the boom tunnel boring machine cab, and realize the identification of the current rock type on the tunnel face.

[0019] (4) The present invention collects the working noise data of the boom tunnel boring machine after starting it, and uses a main and auxiliary dual microphone to reduce the mechanical noise of the boom tunnel boring machine itself and the environmental noise, thereby enhancing the effective signal and obtaining high-quality sound wave data; (5) The present invention conducts indoor and field tests, uses high-precision audio acquisition equipment to collect audio signals generated by a boom tunnel boring machine cutting different rock types, uses signal decomposition technology to extract frequency domain features, and constructs a rock type time-frequency domain feature library. Subsequently, in the project, Fourier transform is performed on the noise-reduced signal data to obtain time-frequency domain information, and the time-frequency domain features are input into the feature library. The rock type identification results are output to guide operators to optimize tunneling parameters.

[0020] (6) The detection system is integrated and installed on the cantilever tunnel boring machine. No additional operation is required after the detection system is started. The intelligent terminal system automatically identifies the rock type of the tunnel face. The degree of automation and intelligence is higher than that of existing patents.

[0021] (7) The operation process of the detection device is simple, convenient and fast. After the sound pickup device is started, the intelligent terminal system runs a machine learning algorithm (support vector machine) to process and analyze the collected sound wave data, analyze and identify the surrounding rock type in real time, and provide a scientific basis for the selection of construction parameters of the cantilever tunnel boring machine.

[0022] (8) The detection device does not require additional drilling, will not cause potential damage to the tunnel structure, and does not require the installation of complex equipment, saving time for equipment installation and debugging, and does not interfere with the normal construction organization of the tunnel, ensuring that the construction progress is carried out in an orderly manner as planned.

[0023] (9) The present invention does not require the use of artificial seismic sources as transmitting waves, thus eliminating the risks caused by blasting and avoiding disturbance to the surrounding rock. However, the electromagnetic method faces electromagnetic interference generated by electrical equipment and machinery at the construction site, which affects the accuracy of the measurement results. Therefore, the present invention is suitable for tunnels built in environmentally sensitive areas.

[0024] (10) The physical method is highly sensitive to external factors such as noise in the geological environment, climate change, and construction activities, which can easily lead to increased errors in the measurement results. The present invention adopts a dual-band high-sensitivity microphone and uses a variety of noise reduction methods to effectively improve the signal quality and ensure the accuracy of the measurement results.

[0025] (11) This device has a built-in battery and wireless transmission module, and its outer shell is made of high-density polyethylene (HDPE), which can withstand harsh underground construction environments and facilitate detection work.

[0026] (12) Different types of microphones have significant differences in frequency response characteristics, which are mainly reflected in multiple dimensions such as sensitivity, amplification, and restoration of sounds of different frequencies. In the actual operation of the boom tunneling machine, the frequency of cracks generated inside the rock is usually in the high-frequency range above 100 kHz, while the sound frequency generated by the friction between the pick and the rock is relatively low. In order to achieve high-precision conversion of the original sound signal and retain the original characteristics and details of the sound to the greatest extent, this device uses different types of microphones to collect high-frequency and low-frequency sound wave signals respectively. The dynamic microphone is used to collect low-frequency signals, and its good response characteristics in the low-frequency band are used to ensure accurate capture of low-frequency signals. At the same time, a capacitive microphone is used to collect high-frequency signals. With its high sensitivity to high-frequency signals and excellent restoration ability, it can effectively collect high-frequency sounds. This device integrates these two different types of microphones into one, and constructs an efficient pickup system that can comprehensively and accurately collect sound signals of different frequencies.

[0027] (13) During the actual installation process, in order to ensure the stability of the sound intensity and facilitate subsequent accurate analysis, the sound pickup device was installed in front of the cab and at the rear hydraulic control system of the boom tunnel boring machine. The boom tunnel boring machine will inevitably generate strong vibrations during the tunneling operation. This vibration will be transmitted to the inside of the microphone through mechanical conduction, which may cause the microphone to generate mechanical noise, interfere with the collection of the original sound signal, and may also change the proportion of the microphone to pick up sounds of different frequencies, thereby adversely affecting the subsequent signal analysis work. In order to effectively solve this problem, this device installs the microphone on a specially designed shock-absorbing bracket. Through the special structure and material properties of the shock-absorbing bracket, it can effectively isolate the vibration from being transmitted to the microphone, thereby significantly reducing the noise caused by vibration and providing high-quality data support for subsequent analysis and application.

[0028] (14) The antenna module is designed with a built-in layout. This design not only optimizes the overall structure of the sensor and maintains its smooth and simple appearance, but also improves the antenna's protection capabilities in harsh construction environments, ensuring the stability and reliability of wireless communication. The built-in antenna design also reduces the impact of external factors such as mechanical collisions and dust intrusion on antenna performance, thereby ensuring the continuity of data transmission and the immediate receipt of operating instructions. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a structural schematic diagram of the sound pickup device of the present invention.

[0030] Figure 2 This is a schematic diagram of the dual-pickup noise reduction principle of the present invention.

[0031] Figure 3 It is a three-dimensional stereogram of the time domain and frequency domain of the sound wave of the present invention.

[0032] Figure 4 It is a flow chart of the detection principle of the present invention. DETAILED DESCRIPTION

[0033] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0034] In this preferred embodiment, a cantilever tunnel boring machine geological condition intelligent perception system based on dual-band acoustic wave analysis is provided, which includes three key subsystems: a dual-band high-sensitivity pickup device, a data wireless transmission system, and an acoustic signal processing system. The pickup is composed of an integrated dual-band pickup, and its main function is to accurately capture the amplitude and frequency information of the surrounding rock acoustic wave signal. The data wireless transmission system is responsible for transmitting the audio data collected by the high-sensitivity pickup to the acoustic signal processing system in a high-speed, real-time transmission mode without loss, thereby ensuring the timeliness and reliability of data transmission. The acoustic signal processing system is equipped with a powerful data storage and processing analysis module. By utilizing its powerful data processing and analysis capabilities, it uses an efficient noise reduction algorithm to process the collected signals and extract spectral features with diagnostic value, so as to realize the rock type identification and structural surface distribution of the shallow surrounding rock of the tunnel face and its rear, thereby providing a scientific basis for the safety monitoring of tunnel construction and the assessment of rock stability. The detailed functions and structures of each system are as follows: The dual-band high-precision pickup device, such as Figure 1 As shown: It includes a capacitive pickup 5 (collecting 10Hz-20kHz low-frequency signals) and a dynamic pickup 6 (collecting 100kHz-500kHz high-frequency signals), which are responsible for collecting acoustic wave data from the cutting head, tunnel face, and surrounding rock during the construction of the cantilever tunneling machine; The wireless data transmission system includes a storage unit 3 and an antenna module 2, which are responsible for signal storage and wireless transmission. When the microphone captures an acoustic signal, it is converted into an electrical signal using the principle of electromagnetic induction. The electrical signal is transmitted to the storage unit 3 through an internal circuit, and the storage unit 3 immediately records it. The storage unit 3 uses an efficient data recording mechanism to ensure the integrity and accuracy of the acoustic signal. The recorded data file is wirelessly transmitted in real time to the terminal data processing platform via the built-in antenna module 2. After receiving the data, the terminal processing platform performs data processing and analysis to achieve real-time monitoring and evaluation of the surrounding rock type and structural surface of the tunnel face. The acoustic signal processing system uses dual-band high-sensitivity pickup devices placed in front and behind the boom tunnel boring machine to collect audio signals generated by the boom tunnel boring machine's cutting head when breaking rock. After noise reduction processing, the system extracts the frequency spectrum of the tunnel face during rock breaking to construct time-frequency domain characteristic parameters. Based on an established time-frequency domain characteristic library of rock mass types, the system identifies the current tunnel face rock type and infers its physical and mechanical properties, providing operators with detailed construction guidance.

[0035] Furthermore, in actual tunnel excavation operations, the cantilever tunnel boring machine itself will also generate noise. Therefore, after the cantilever tunnel boring machine is started, the cutting action is temporarily not performed to collect the noise data generated during its operation. When the cutting operation begins and the mixed noise signal is collected, the machine operation noise can be separated and removed from the mixed signal based on the inherent noise data of the machine collected in the early stage by using the signal processing algorithm, thereby achieving noise reduction processing of the cutting signal. At the same time, the environmental noise in the tunnel is very complex. This strong noise causes significant interference to the acoustic detection data, seriously affecting the quality of the signal, resulting in a significant reduction in the signal-to-noise ratio in the data, and thus weakening the ability to identify effective signals. Therefore, a main and auxiliary dual microphone is used to perform environmental noise reduction processing: the main microphone arranged in front of the cantilever tunnel boring machine and close to the face is responsible for collecting the audio signal generated by the interaction between the cutting teeth and the face; and the auxiliary microphone arranged behind the cantilever tunnel boring machine mainly focuses on collecting environmental noise signals. The schematic diagram of the dual microphone noise reduction arrangement is shown as follows. Figure 2 As shown in the figure, the audio processing system analyzes the signals collected simultaneously by the primary and secondary microphones, comparing multi-dimensional data such as the time and frequency domain characteristics and phase information of the signals collected by the two microphones. Ambient noise typically appears simultaneously in the signals collected by the primary and secondary microphones, with highly similar characteristics. However, the useful sound signal captured by the primary microphone often appears at a lower amplitude in the signal collected by the secondary microphone. This characteristic can be exploited to effectively identify noise.

[0036] On this basis, signal processing techniques such as equalization and gain adjustment are used to further enhance the effective signal strength generated by the boom tunnel boring machine cutting the rock mass, while also suppressing noise components and improving signal quality. After removing background noise and mechanical vibration noise data, the signal is further processed using bandpass filtering technology, effectively removing random noise components from the signal while retaining the useful frequency range. This ensures the accuracy and reliability of the remaining signal, providing high-quality data support for subsequent rock mass identification and analysis.

[0037] Systematically conduct indoor and field tests to construct a time-frequency domain feature library for rock types. Using high-precision audio acquisition equipment, we collect audio signals generated by boom tunnel boring machines cutting various rock types in both indoor simulation environments and actual engineering sites. Because raw audio signals are typically complex mixed signals, signal decomposition technology is used to process and analyze the collected audio signals, decomposing them into different frequency components from which frequency domain features containing rock properties are extracted. The recognition accuracy for granite, limestone, and sandstone reached 92%, a 15% improvement over traditional acoustic emission methods. We collated, summarized, and classified a large amount of frequency domain feature data corresponding to different rock types to construct a time-frequency domain feature library for rock types that comprehensively and accurately reflects the corresponding relationship between rock types and their time-frequency domain features.

[0038] Perform Fourier Transform (FT) on the signal data after background noise removal and noise reduction, converting it from the time domain to the frequency domain, thereby obtaining the time-frequency domain information of the sound wave, such as Figure 3 . Fourier transform can reveal the frequency components of the signal and its dynamic characteristics over time, which is crucial for the identification of rock mass types. By analyzing the frequency domain characteristics, the characteristic frequencies and response patterns of different rock masses can be captured. The extracted time-frequency domain features are input into the established time-frequency domain feature library of rock mass types. This feature library is based on a large amount of rock mass sample data. By comparing and analyzing the time-frequency domain characteristics of different rock masses, a relationship model between the rock mass type and its corresponding time-frequency domain characteristics is constructed. Finally, by using support vector machines (SVM), according to the input time-frequency domain features, it is mapped to a high-dimensional space through a kernel function, and the optimal classification hyperplane is found. The identification results of the rock mass type of the tunnel face are output to achieve accurate classification of different rock mass types, thereby guiding the boom tunneling machine operator to optimize the tunneling parameters, improve tunneling efficiency, and realize the intelligent tunneling of the boom tunneling machine.

[0039] The capacitive pickup 5 and the dynamic pickup 6 are mounted on a shock-absorbing bracket 1. The capacitive pickup 5, the dynamic pickup 6, the wireless module 2, the storage unit 3 and the power supply module 7 are integrated on the shock-absorbing bracket 1. The connection circuits of each unit are arranged inside the shock-absorbing bracket. A dust cover 4 is provided on the front side of the wireless module 2, the storage unit 3 and the power supply module 7. The capacitive pickup 5 and the dynamic pickup 6 are mounted side by side on the outside of the dust cover 4.

[0040] The housings of the capacitive pickup 5 and dynamic pickup 6, as well as the shock-absorbing bracket 1, are all made of high-density polyethylene (HDPE), with a thickness of 5-10mm and an impact strength of ≥10kJ / m². These materials offer excellent durability and lightweight properties, making them resistant to multiple adverse factors such as moisture, dust, and mechanical impact in harsh underground construction environments. This creates a stable and optimal working environment for the pickups, effectively ensuring their continued operation.

[0041] The antenna module of the wireless data transmission system adopts a built-in layout. The shock-absorbing bracket 1 uses a thickened hollow template of 50×50×2, with four solid legs of 5×5×10. The antenna module is integrated into the shock-absorbing bracket template, and the wireless transmission efficiency is optimized through an impedance matching circuit.

[0042] like Figure 4 The method for using the system includes the following steps: (1) Calibrate the microphone based on the specific conditions of the construction site to ensure the resolution and accuracy of data collection; (2) The device performs self-tests to confirm that all sensors are working properly, check whether the wireless connection is stable, and whether the terminal platform can synchronize information to ensure the normal operation of the entire device system; (3) Collecting audio data of ambient noise in the tunnel when the boom tunnel boring machine is not started; (4) Start the cantilever tunnel boring machine, but do not cut, to collect the noise generated by the cantilever tunnel boring machine; (5) After the boom tunnel boring machine starts cutting, the intelligent terminal system performs preliminary processing based on the real-time collected sound wave data, including filtering, noise reduction and other steps to filter out invalid noise signals and extract effective signals reflecting the geological conditions; (6) Identify the rock type of the current tunnel face through machine learning algorithms and guide the boom tunneling machine operator to optimize the tunneling parameters; (7) After completing the excavation operation of a section, check the status of the pickup to ensure that the pickup can maintain the best working condition during the next section excavation.

[0043] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent geological condition perception system for a boom roadheader based on dual-band acoustic wave analysis, characterized in that: It includes a dual-band high-sensitivity pickup device, a wireless data transmission system, and an acoustic signal processing system. The dual-band high-sensitivity pickup device includes a capacitive pickup and a dynamic pickup, which is responsible for collecting sound wave data from the cutting head, tunnel face and surrounding rock during the construction of the cantilever tunneling machine; The wireless data transmission system includes a storage unit and an antenna module located inside the pickup device, which are responsible for signal storage and wireless transmission. When the pickup captures an acoustic signal, the internal storage unit immediately records it. The storage unit uses an efficient data recording mechanism to ensure the integrity and accuracy of the acoustic signal. The recorded data files are wirelessly transmitted in real time to the terminal data processing platform via the built-in antenna module. After receiving the data, the terminal processing platform performs data processing and analysis to achieve real-time monitoring and evaluation of the surrounding rock type and structural surface of the tunnel face. The acoustic signal processing system uses dual-band high-sensitivity pickup devices placed in front and behind the boom tunnel boring machine to collect audio signals generated by the boom tunnel boring machine's cutting head when breaking rock. After noise reduction processing, the system extracts the frequency spectrum of the tunnel face during rock breaking to construct time-frequency domain characteristic parameters. Based on an established time-frequency domain characteristic library of rock mass types, the system identifies the current tunnel face rock type and infers its physical and mechanical properties, providing operators with detailed construction guidance.

2. The intelligent geological condition perception system for a cantilever roadheader based on dual-band acoustic wave analysis according to claim 1 is characterized in that: The dual-band high-sensitivity sound pickup device is installed in front of the cab and at the rear hydraulic control system of the boom tunneling machine.

3. The intelligent geological condition perception system for a cantilever roadheader based on dual-band acoustic wave analysis according to claim 2 is characterized in that: The capacitive pickup and the dynamic coil pickup are mounted on a shock-absorbing bracket.

4. The intelligent geological condition perception system for a cantilever roadheader based on dual-band acoustic wave analysis according to claim 3 is characterized in that: The pickup housing and shock mounts are made of high-density polyethylene (HDPE).

5. The intelligent geological condition perception system for a cantilever roadheader based on dual-band acoustic wave analysis according to claim 1 is characterized in that: The antenna module of the data wireless transmission system adopts a built-in layout.

6. The intelligent geological condition perception system for a boom roadheader based on dual-band acoustic wave analysis according to claim 1 is characterized in that: The signal data after background noise removal and noise reduction is subjected to Fourier transform, which converts it from the time domain to the frequency domain to obtain the time-frequency domain information of the sound wave.

7. The intelligent geological condition perception system for a boom roadheader based on dual-band acoustic wave analysis according to claim 1 is characterized in that: The method of using the system includes the following steps: (1) Calibrate the microphone based on the specific conditions of the construction site to ensure the resolution and accuracy of data collection; (2) The device performs self-tests to confirm that all sensors are working properly, check whether the wireless connection is stable, and whether the terminal platform can synchronize information to ensure the normal operation of the entire device system; (3) Collecting audio data of ambient noise in the tunnel when the boom tunnel boring machine is not started; (4) Start the cantilever tunnel boring machine, but do not cut, to collect the noise generated by the cantilever tunnel boring machine; (5) After the boom tunnel boring machine starts cutting, the intelligent terminal system performs preliminary processing based on the real-time collected sound wave data, including filtering, noise reduction and other steps to filter out invalid noise signals and extract effective signals reflecting the geological conditions; (6) Identify the rock type of the current tunnel face through machine learning algorithms and guide the boom tunneling machine operator to optimize the tunneling parameters; (7) After completing the excavation operation of a section, check the status of the pickup to ensure that the pickup can maintain the best working condition during the next section excavation.

Images