An ultra-long endurance TBM cutterhead cutter vibration signal acquisition system and method

Abstract

The application discloses a kind of super-long endurance TBM cutterhead cutter vibration signal acquisition system and method, it is related to the field of tunnel engineering equipment condition monitoring, including a plurality of wireless vibration signal acquisition nodes, the plurality of wireless vibration signal acquisition nodes are sequentially connected with control module, industrial router and host computer.Wireless vibration signal acquisition node integrates energy acquisition unit, through the mechanism of " edge sampling edge fills ", combined with three gears four states self-adaptive scheduling driven by energy prediction and LoRa+ local area network dual wireless communication, realize the shallow filling shallow release management of battery.The application solves the problem of power supply limitation, wiring difficulty and frequent maintenance of vibration monitoring system under the harsh working condition of TBM, realizes super-long endurance and maintenance-free operation.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering equipment condition monitoring technology, and in particular to a TBM cutterhead tool vibration signal acquisition system and method with ultra-long endurance. Background Technology

[0002] Full-face tunnel boring machines (TBMs) are core equipment for underground space development. Their cutterheads and cutterheads endure severe vibrations during tunneling, which are a major cause of damage to critical components. Therefore, effective vibration monitoring of the cutterheads and cutterheads is crucial for ensuring the safe and reliable tunneling of TBMs.

[0003] However, the harsh working environment of TBM cutterhead rotation, strong vibration, and high dust levels poses a significant challenge to traditional vibration monitoring technologies, resulting in very few successful field monitoring cases. Existing technologies mainly suffer from the following shortcomings: First, wiring is difficult and unreliable. Traditional wired monitoring solutions require laying cables on a rotating cutter head and leading out signals through slip rings. This is complex, costly, and the lines are easily worn or broken in continuous vibration and dusty environments, making it difficult to guarantee reliability.

[0004] Secondly, limited power supply leads to frequent maintenance. While battery-powered wireless sensors avoid wiring, battery capacity is limited, and their lifespan is further shortened under strong vibration conditions. This necessitates frequent shutdowns and dispatching personnel to hazardous areas to replace batteries, failing to meet the demands of long-term continuous tunneling and making it difficult for the monitoring system to achieve long-term maintenance-free operation.

[0005] Furthermore, existing energy management strategies are inefficient. To extend battery life, current technologies attempt to use energy harvesting (such as piezoelectric or electromagnetic energy harvesting) to assist battery power supply. However, common practices lack closed-loop coordinated control of the "energy harvesting-storage-consumption" link. Specifically, this manifests in the following ways: the energy harvesting link and the battery discharge link lack a low-loss automatic switching mechanism, resulting in energy loss; the fluctuating energy harvesting power is not dynamically matched with load power consumption, causing energy supply and demand mismatch; and the control of battery state of charge (SOC) often remains at the level of simple overcharge and over-discharge protection, without implementing shallow charge and shallow discharge strategies that can significantly extend battery life.

[0006] In summary, existing TBM cutterhead tool vibration monitoring technologies are limited by power supply, wiring, and maintenance issues, making it difficult to achieve long-term, continuous, reliable, and maintenance-free operation under harsh conditions. Therefore, there is an urgent need in this field for an ultra-long-endurance vibration signal acquisition solution that can construct a low-loss, predictable, and schedulable energy closed loop around "energy harvesting-storage-use". Summary of the Invention

[0007] To address the challenge of existing TBM cutterhead tool vibration monitoring technologies failing to achieve long-term, continuous, reliable, and maintenance-free operation under harsh conditions, this application proposes an ultra-long-endurance TBM cutterhead tool vibration signal acquisition system and method.

[0008] This application discloses a TBM cutter head tool vibration signal acquisition system with ultra-long battery life, including several wireless vibration signal acquisition nodes, wherein the several wireless vibration signal acquisition nodes are sequentially connected to a control module, an industrial router and a host computer. The wireless vibration signal acquisition nodes are fixed on the front and rear of the TBM cutter head and the inner surface of the hobbing cutter box. Each node independently acquires the three-dimensional vibration signal at its location. The control module is used to issue acquisition commands to each node, perform network-wide time synchronization, execute energy-power prediction and adaptive scheduling of sampling status, and receive and forward node data. The industrial router is powered via a wired connection and acts as a communication gateway to connect the control module and the host computer. The host computer runs monitoring software and receives and processes vibration data through the high-speed wireless LAN link of the industrial router.

[0009] Preferably, the wireless vibration signal acquisition node includes a protective shell, which includes a LoRa antenna, a reserved maintenance port, a local area network antenna, an aluminum alloy shell, an aluminum alloy top cover, and quick-release bolt holes. The LoRa antenna, the reserved maintenance port, and the local area network antenna are located on the side of the aluminum alloy shell, the aluminum alloy top cover is located on the top of the aluminum alloy shell, the quick-release bolt holes are located through the bottom of the aluminum alloy shell, and the aluminum alloy shell has a first cavity and a second cavity inside.

[0010] Preferably, a first circuit board and a lithium battery are disposed inside the first cavity, with the lithium battery positioned above the first circuit board. The first circuit board integrates an axial acceleration sensor, a microprocessor unit, and a wireless communication unit.

[0011] Preferably, an energy harvesting unit is provided inside the second cavity. The energy harvesting unit includes a coil, a permanent magnet, a guide rail, a bearing, a limiting spring, a rectification and filtering unit, a boost and voltage regulation unit, and a battery life management module. The rectification and filtering unit, the boost and voltage regulation unit, and the battery life management module are integrated on the second circuit board. The coil, permanent magnet, guide rail, bearing, and limiting spring are arranged coaxially and are located below the second circuit board.

[0012] Preferably, the control module is configured to execute the following power consumption prediction and frequency adaptation strategy: The average harvested power within the most recent N sampling windows Average operating power consumption SOC and cache usage are inputs. If the SOC is higher than 60%, increase the sampling rate or extend the sampling window; if If the SOC is below 40%, the system will be downgraded in the following order: reducing the sampling rate, shortening the sampling window, and enabling wake-up monitoring, until energy self-sufficiency is restored.

[0013] Preferably, the control module is further configured to perform shallow charge / discharge lifetime management: The SOC is corrected based on the charge / discharge curves reported by the wireless vibration signal acquisition nodes, limiting the SOC to slowly fluctuate within 30% to 80%. When the SOC is predicted to reach the upper limit of 80%, the boost duty cycle is reduced or trickle charging is switched. When the SOC is predicted to reach the lower limit of 30%, the next sampling window is postponed to ensure long-term small ΔSOC cycling of the battery, thereby significantly improving the battery's equivalent cycle life.

[0014] This application also discloses a method for acquiring vibration signals from a TBM tool turret with ultra-long battery life, implemented based on the aforementioned ultra-long battery life TBM tool vibration signal acquisition system, comprising the following steps: System deployment and initialization: Install wireless vibration signal acquisition nodes in the TBM cutter head and hobbing area, fix industrial routers and control modules, establish a wireless communication network and perform full network time synchronization; Energy harvesting link calibration, calibration of the output of electromagnetic induction energy harvesting unit, initialization of the range management system and setting of SOC shallow charge and discharge window; Energy prediction and adaptive scheduling: The control module periodically obtains the average energy extraction power and average load power of each node, predicts the net energy difference for the next cycle, and adaptively allocates sampling frequency and duty cycle to each node accordingly. Collaborative data acquisition and reporting: The control module issues a unified sampling window command, and each node synchronously performs vibration signal acquisition and vibration energy acquisition within the window to achieve "simultaneous acquisition and charging". After the sampling is completed, the data is reported and then enters a low-power sleep state. Lifecycle maintenance involves the control module performing network-wide energy balancing scheduling, migrating data acquisition tasks from low-energy nodes to high-energy nodes to balance the lifespan of all nodes in the network; and controlling nodes to enter a lifespan-friendly smooth shutdown or low-power self-sustaining mode when shutting down.

[0015] Preferably, during the system deployment and initialization, a LoRaMesh self-organizing network is established, including neighbor discovery, route calculation, and connectivity verification. The nodes and control modules transmit commands and status information via a LoRa link, and vibration data is transmitted via a high-speed wireless LAN link of an industrial router.

[0016] Preferably, the energy prediction and adaptive scheduling includes the following steps: Define the statistical period ; Node in each Internal measurement and calculation of average energy harvesting power and average load power The report is then sent to the control module. The control module calculates the power difference. Based on historical data from the most recent M statistical periods, the power difference for the next statistical period is predicted. ; The control module is based on the prediction Based on the current SOC of the node and the preset dual threshold criteria, the working state of the node is divided into continuous monitoring state, balanced monitoring state, intermittent monitoring state or wake-up monitoring state, and a corresponding sampling frequency, sampling window duration and sampling period are assigned to each state.

[0017] Preferably, in the collaborative acquisition, nodes simultaneously perform vibration signal acquisition and energy acquisition from environmental vibration within the same sampling window, and charge the battery to achieve simultaneous acquisition and charging; at the same time, power snapshot data is recorded and reported within the sampling window.

[0018] The beneficial effects of this invention are: 1. Synchronous linkage between sampling and charging, along with health window control, significantly extends node lifespan. Traditional wireless data acquisition nodes often employ a "sampling and then charging" or "passive energy harvesting" mode, resulting in frequent charge-discharge cycles and large SOC fluctuations. This invention establishes a closed loop of sampling-energy harvesting-SOC management during system startup, ensuring that nodes maintain a shallow charge-discharge (30%–80%) healthy range from the initial operation. This reduces the number of deep charge-discharge cycles by approximately 80%, and in long-term testing, increases cycle life by approximately 3.6 times (from approximately 500 cycles to over 1800 cycles).

[0019] 2. Window-level power snapshots enable energy budget visualization and support predictive scheduling closed-loop. This invention simultaneously records and reports power snapshots (energy harvesting power, load power, and SOC changes) within a sampling window, enabling the control module to quantify the energy budget of "simultaneous harvesting and charging" at a resolution of one second, achieving dynamic visibility of power. Experimental results show that this mechanism reduces energy prediction error from ±20% of traditional empirical estimates to ±3%, providing accurate feedforward data for subsequent scheduling.

[0020] 3. A predictive-driven three-level, four-state adaptive mechanism achieves a dual closed-loop energy feedforward and feedback system. Traditional systems often use fixed sampling frequencies and intervals, which can easily lead to excessive power consumption during low-energy periods or waste during high-energy periods. This invention automatically adjusts the sampling frequency and duty cycle by predicting the net energy difference over a future statistical period (5 minutes), achieving adaptive control of "high sampling when energy is sufficient and low sampling when energy is scarce." Experimental results show that, under the same conditions, the average power consumption is reduced by approximately 42%, while the effective monitoring duration is extended by approximately 2.3 times.

[0021] 4. LoRa+ LAN responsibility separation design improves communication reliability and energy efficiency. Traditional solutions often use a single-channel communication to carry commands and data, which is prone to system command delays or abnormal power consumption due to data channel congestion. This invention adopts a dual-link division of labor, with LoRa handling control and maintenance, and the industrial router's wireless LAN handling high-bandwidth data. This avoids channel contention and data loss. In actual testing, the overall network packet loss rate decreased from 3.8% to 0.2%, while the average communication power consumption decreased by approximately 30%.

[0022] 5. Achieving globally optimal energy consumption through network-wide lifetime-balanced scheduling based on the Max-Min principle. This invention periodically summarizes State of Charge (SOC) and power prediction, and the control module automatically migrates tasks from energy-deficient nodes to energy-rich nodes, maximizing the remaining lifetime of the "shortest-lived node" in the entire network. Experimental results show that, compared to a uniform allocation strategy, the lifetime-balanced scheduling of this invention can improve the overall maintenance-free operation time of the network by approximately 45%, significantly enhancing the long-term stability and reliability of the system.

[0023] 6. Flexible and reliable deployment capabilities ensure comprehensive coverage of TBM vibration signals. This invention supports the flexible and rapid deployment of multiple wireless nodes. Nodes can be installed in key locations such as the front and rear of the cutter head and inside the hobbing cutter box using welded bases and quick-release bolts. It eliminates the need for wired power supply and wired signal transmission, utilizing a wireless mesh self-organizing network to construct a full-space coverage monitoring network. Multiple monitoring points can be added without any damage to the TBM structure, greatly improving the flexibility of monitoring solutions under complex operating conditions. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the structure of the TBM cutter head tool vibration signal acquisition system with ultra-long battery life according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the wireless vibration signal acquisition node structure according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the energy harvesting unit structure according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating the energy-power prediction and three-level four-state scheduling process according to an embodiment of the present invention. Figure 5 This is a timing diagram of the unified sampling window in an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided with reference to the accompanying drawings and embodiments.

[0026] One embodiment of this application discloses a TBM tool head vibration signal acquisition system with ultra-long battery life, the structure of which is as follows: Figure 1 As shown, it includes several wireless vibration signal acquisition nodes, which are sequentially connected to a control module, an industrial router, and a host computer. In this embodiment, the number of wireless vibration signal acquisition nodes is 2-16.

[0027] Several wireless vibration signal acquisition nodes are fixed on the front and rear of the cutter head and the inner surface of the hobbing cutter box. Each node includes a sensor unit, a microprocessor unit (MCU, low-power microcontroller), an energy acquisition unit, a charging management and energy storage unit, a communication unit, and an IP67 sealed housing. The nodes simultaneously perform vibration sampling and vibration energy harvesting within a unified sampling window, and enter deep sleep mode outside the sampling window, retaining only the clock and wake-up channels to reduce average power consumption.

[0028] Each wireless vibration signal acquisition node measures approximately 200mm × 60mm × 60mm and weighs about 550g, encased in an IP67 aluminum alloy protective housing. The housing is made of 6061-T6 aluminum alloy and does not affect signal transmission. Before installing the node, a steel base with bolt holes, the same size as the bottom of the protective housing, is pre-welded onto the surface to be monitored. During installation, the bottom of the protective housing is placed against the base, and quick-release bolts are passed through the bolt holes on both the bottom of the housing and the base. The bolts are then tightened clockwise to complete the installation. For disassembly, the quick-release bolts are manually unscrewed counterclockwise to separate the node from the base. The assembly and disassembly process is simple and quick, without damaging the original structure of the cutter head.

[0029] The specific structure is as follows: Figure 2As shown, the wireless vibration signal acquisition node includes a protective housing, within which are housed a triaxial accelerometer 11, a microprocessor unit 12, a wireless communication unit 13, a lithium battery 14, and an energy acquisition unit. The energy acquisition unit includes a coil 151 (in this embodiment, a 5000-turn enameled copper wire coil), a permanent magnet 152 (in this embodiment, an N52-grade neodymium iron boron magnet), a guide rail 153, a bearing 154, a limit spring 155, a rectifier and filter unit 156, a boost and voltage regulator unit 157, and a battery life management module 158. The protective housing includes a LoRa antenna 161, a reserved maintenance port 162, a local area network antenna 163, an aluminum alloy housing 164, an aluminum alloy top cover 165 (connected to the aluminum alloy housing 164 by six bolts), and quick-release bolt holes 166 (for fixing to the device to be monitored via quick-release bolts and enabling rapid disassembly). The LoRa antenna 161, the reserved maintenance port 162, and the local area network antenna 163 are located on the side of the aluminum alloy housing 164, the aluminum alloy top cover 165 is located on the top of the aluminum alloy housing 164, and the quick-release bolt hole 166 is located through the bottom of the aluminum alloy housing 164.

[0030] The aluminum alloy housing 164 has a first cavity and a second cavity inside. The triaxial accelerometer 11, the microprocessor unit 12, and the wireless communication unit 13 are integrated on the first circuit board, which is placed in the first cavity. The lithium battery 14 is located directly above the first circuit board. The first circuit board and the lithium battery 14 are surrounded by cushioning foam for coverage and support.

[0031] like Figure 3 As shown, the rectifier filter unit 156, the boost regulator unit 157, and the battery management module 158 are integrated on the second circuit board and housed in the second cavity. The coil 151, permanent magnet 152, guide rail 153, bearing 154, and limiting spring 155 are arranged coaxially. The guide rail 153 is located on the innermost side, the bearing 154 is located on the outer side of the center of the guide rail 153, and the permanent magnet 152 is located on the outer side of the bearing 154 and is fixed relative to the bearing 154 (interference fit). The permanent magnet 152 and the bearing 154 can slide axially along the guide rail 153. The limiting spring 155 is located on the outer side of both ends of the guide rail and connected to both ends of the bearing 154. The coil 151 is located on the outermost side, maintaining a gap with the permanent magnet 152. The coil 151, permanent magnet 152, guide rail 153, bearing 154, and limiting spring 155 form a single unit, placed directly below the second circuit board and connected to it. The second circuit board and the coaxial core are surrounded by cushioning foam for support. Grooves are provided between the first and second cavities on both sides to accommodate wires connecting the first and second circuit boards.

[0032] The sensor unit uses a triaxial accelerometer with a range of ±16g. This sensor can capture the three-dimensional vibration signals of the cutterhead and roller cutter in the frequency band of 0.1Hz–10kHz, meeting the requirements of both low-frequency (structural resonance) and high-frequency (roller cutter rock breaking impact) characteristics in tunneling conditions.

[0033] The microprocessor unit uses an STM32 series low-power microcontroller (MCU) with an 80 MHz clock frequency, a built-in 12-bit ADC, and 1 MB of Flash memory. Its main functions include: analog signal acquisition and digital-to-analog conversion, digital filtering, data packing and buffering, control energy harvesting, and sleep / wake-up. The triaxial accelerometer is connected to the MCU via an SPI bus (SPI is a commonly used on-chip peripheral bus). The MCU has a built-in ring buffer and packing logic, which can store raw samples or perform downsampling / light compression as needed. The default sampling rate, sampling duration, and period are remotely controlled by the control module.

[0034] The energy harvesting unit is the core component for achieving ultra-long battery life, employing electromagnetic induction vibration energy extraction. It includes a mass-spring-damped tuning mechanism, an energy harvesting unit, a rectifier and filter unit 156, and a boost and regulator unit 157. The mass-spring-damped tuning mechanism comprises a permanent magnet 152, a guide rail 153, a bearing 154, and a limiting spring 155. The energy harvesting unit consists of a coil 151 and the permanent magnet 152. The natural frequency of the mass-spring-damped tuning mechanism matches the typical main oscillation frequency band of the TBM (0-500Hz). The energy harvesting unit consists of a 5000-turn enameled copper wire coil (DC resistance approximately 10 Ω) and an N52-grade neodymium iron boron magnet (surface magnetic field strength 1.2T). When the coil 151 and the permanent magnet 152 vibrate relative to each other, an AC voltage UAC is generated in the coil 151 according to the principle of electromagnetic induction, with an amplitude of 50–200 mV under typical operating conditions. The rectifier and filter unit 156 uses a low-dropout rectifier circuit to convert AC to DC, and simultaneously removes high-frequency noise through capacitor filtering. The boost regulator unit 157 uses a DC-DC boost chip, which can start with a minimum input voltage of 20 mV and output a stable voltage of 3.3 V–4.2 V.

[0035] The lithium battery 14 and the range management module 158 constitute a charging management and energy storage unit. The range management module 158 contains a range management system. The lithium battery 14 uses a single rechargeable lithium-ion battery (18650 battery) with a capacity of 5000 mAh. Under typical sampling frequency (1 kHz, 50% duty cycle), it can operate continuously for about 4 months in pure battery mode. With electromagnetic energy harvesting assistance, it can maintain a dynamic balance of SOC 30%–80% for a long time, achieving maintenance-free operation. The range management system includes overcharge, over-discharge, short circuit, and temperature protection functions, specifically: 1. Overcharge protection: When the cell voltage approaches the safe upper limit (4.2V), the charging switch is turned off, blocking the charging current from the energy harvesting unit or external power supply. 1. **Charging switch protection:** If the voltage drops below 4.15V, the charging switch resumes conduction. 2. **Over-discharge protection:** When the cell voltage approaches the set lower limit (3.3V), the discharge switch opens, cutting off the battery output while maintaining the input path of the energy harvesting unit. When the battery voltage recovers to above 3.35V, the battery output is restored. 3. **Short circuit protection:** Continuously monitors current changes in the circuit. When a short circuit occurs in the load, the instantaneous current rises sharply, triggering short circuit protection and quickly cutting off the battery input and output. After the short circuit is cleared, and after a delay and confirmation that the load has returned to normal, the battery input and output are restored. 4. **Temperature protection:** The battery management system has a built-in miniature thermistor that calculates the current temperature based on its resistance change. When the temperature exceeds the set upper limit (55 degrees Celsius) or falls below the set lower limit (-5 degrees Celsius), the battery input and output are cut off. After the temperature returns to the set range, the battery input and output are restored. The battery life management system tracks the State of Charge (SOC) and estimates the remaining battery capacity using voltage and current integration. It maintains operation within a shallow charge-discharge range of 30%–80% to significantly improve cycle life. The system incorporates an automatic switching circuit between the battery and the energy harvesting unit. When energy harvesting power is insufficient, the battery provides power; when harvesting power is sufficient, the system prioritizes power supply and charges the battery. Furthermore, the MCU estimates the State of Health (SOH) at sampling intervals for long-term node lifetime assessment.

[0036] The sampling-energy harvesting synchronization mechanism is implemented by the control module issuing a unified time base and a "sampling window" command: at the instant the window opens, the node synchronously starts triaxial sampling and energy harvesting charging. The power supply path adopts a parallel isolation structure (two power supplies are connected in parallel through "ideal diodes" in a redundant parallel method with no backflow), and the sampling power supply and charging path current are isolated.

[0037] Wireless communication unit 13 is a communication unit that operates by default in short packet, low duty cycle mode. Nodes aggregate and report data in a single "reporting slot" after the sampling window ends. To reduce energy consumption, nodes prioritize the shortest link strategy, either the nearest direct link or a single relay, and only enable multi-hop when link attenuation is severe. The wireless frame contains fields such as node ID, timestamp, sampling rate, number of axes, number of sample points, SOC, and estimated energy / power consumption per unit time. The communication unit includes a LoRa radio frequency unit and a local area network wireless unit, which are responsible for transmitting control commands and vibration data, respectively.

[0038] The LoRa RF unit operates at a frequency of 433 MHz, with a transmit power of 14 dBm and a receive sensitivity of up to [missing value]. 137dBm. Employing spread spectrum communication, the typical data transmission rate is 2.4 kbps, with a single packet length of 128 bytes. It is primarily used for command exchange, network maintenance, and status reporting between wireless vibration signal acquisition nodes and control modules. The LoRa link features long-range (typically greater than 500 m) and high anti-interference capabilities, maintaining stable communication even in complex environments such as underground tunnels and areas obstructed by metal structures.

[0039] The local area network (LAN) wireless unit conforms to the IEEE 802.11 standard, typically operates at a frequency of 2.4 GHz, and has a coverage range greater than 50 m in tunnel environments. For high-bandwidth vibration data transmission, nodes transmit collected triaxial vibration data in real-time to the control module via a high-speed LAN link from the industrial router, using an IP-based transmission protocol. The control module then forwards the data to the host computer via the industrial router. The host computer runs an IP-based receiving service, and the nodes, acting as data senders, establish connections with this service and push data. The data transmission employs data verification, sequence number marking, and acknowledgment retransmission or buffering mechanisms to ensure the integrity and real-time performance of the vibration data.

[0040] The control module, located in the personnel compartment behind the cutterhead, is responsible for network-wide scheduling, data aggregation, and energy management. Its function is to send and receive commands from the host computer and achieve energy balance by scheduling the energy consumption of the wireless vibration signal acquisition nodes, ensuring consistent lifespan across all nodes. The module formulates a data acquisition plan based on the host computer's monitoring requirements, including parameters such as sampling frequency, synchronization method, and data upload cycle, and then sends control commands to each node via the LoRa link. By adjusting the node's sampling interval, sampling frequency, transmission power, and working / sleep ratio, the energy consumption of each node in the network is balanced, preventing any single node from running out of power too quickly. The control module includes 32GB of storage, capable of caching vibration data for more than 30 consecutive days. The control module is connected to an industrial router via Gigabit Ethernet; this port integrates PoE power supply functionality, allowing direct power supply to the control module via a network cable.

[0041] The control module includes a communication unit (LoRa RF unit and LAN wireless unit) and a microprocessor unit (MCU) of the same standard as the wireless vibration signal acquisition node. It is responsible for time distribution, sampling plan distribution, energy-power consumption prediction and lifetime balancing scheduling, data reception, and buffering and forwarding. The control module is connected to an industrial router via a network cable and is powered by the router's 12V power supply. The control module receives aggregated data from the wireless vibration signal acquisition node through the LAN and forwards it to the IP-based receiving port of the host computer, ensuring orderly and reliable transmission.

[0042] The power consumption prediction and frequency adaptation strategy is executed in a closed loop by the control module: the control module uses the average power harvested within the most recent N sampling windows. Average operating power consumption SOC and cache usage are inputs. If the SOC has reached the set upper threshold (60%), then increase the sampling rate to the highest value (from the default 2000Hz to 4000Hz) and extend the sampling window duration (from the default 60-second sampling period of only 30 seconds to a continuous 60-second sampling period); if And if the SOC is below the lower threshold (40%), then follow the steps of "reducing the sampling rate, reducing the sampling window duration, and enabling wake-up monitoring (first reduce the sampling rate to 1000Hz, which will immediately reduce the instantaneous sampling power; after one sampling cycle, calculate if...) Still below Reduce the sampling window duration to 10 seconds (default 30 seconds), which will make Significantly reduced; after another sampling period, if Still below If wake-up monitoring is enabled, the node will stop sampling and maintain its energy harvesting state if it does not receive a specific sampling command. Reduce to 0, The energy self-sufficiency will be restored in the order of "basically remaining unchanged" until energy self-sufficiency is restored (i.e., the current energy self-sufficiency is maintained). Higher than before sampling stopped (And the SOC value has risen to above the lower threshold of 40%). Multiple wireless vibration signal acquisition nodes are allocated frequencies according to their current energy state to achieve network-wide lifetime balance.

[0043] Shallow charge / discharge lifespan management is implemented simultaneously with sampling and energy harvesting: the control module corrects the State of Charge (SOC) based on the charge / discharge curves reported by the wireless vibration signal acquisition node, limiting the SOC to slowly fluctuate within 30% to 80%. When the upper limit of 80% is predicted to be reached, the boost duty cycle is reduced or trickle charging is switched. When the lower limit of 30% is predicted to be reached, the next sampling window is postponed or the current window is shortened, ensuring long-term small ΔSOC cycling of the battery, thereby significantly improving the battery's equivalent cycle life.

[0044] The industrial router acts as a communication gateway, powered via a wired connection and communicating wirelessly with the control module and host computer. It uploads vibration data output from the control module to the host computer. Specifically, the industrial router, as a data transmission gateway, is connected to the control module via a network cable. The industrial router connects to the power network at the TBM operating site via a 12V power line, supplying power to itself and the connected control module, and communicating with the control module in real time. The router transmits the vibration data collected by the control module to the host computer via the local area network.

[0045] The industrial router is fixed near the 12V power supply box at the rear of the cutter head, with only a single 1.5m 12V DC power cable as the sole wiring harness on site. The industrial router acts as a communication gateway, providing local area network access and data forwarding functions between the control module and the host computer. The control module establishes an IP-based data transmission connection with the host computer through the local area network and continuously sends the collected vibration data to the host computer according to a preset data frame format. The host computer runs monitoring software, initiates the IP-based receiving service, and receives, verifies, parses, and archives the data from the control module.

[0046] The host computer monitoring software runs on the computer in the control room. It receives vibration data streams through the local area network, completes data storage, visualization, and analysis, thereby realizing remote real-time monitoring of the TBM tunneling process.

[0047] The TBM cutter head tool vibration signal acquisition system with ultra-long battery life in this embodiment adopts a hybrid architecture of fully wireless self-powered nodes + local wired gateways. Only one 12V power supply line needs to be configured for the industrial router. The remaining sensing nodes rely entirely on electromagnetic induction energy acquisition to achieve self-powering, thus achieving the effect of maintenance-free operation and ultra-long battery life.

[0048] Another embodiment of this application discloses a method for acquiring vibration signals from a TBM cutterhead with ultra-long battery life, implemented based on the aforementioned ultra-long battery life TBM cutterhead vibration signal acquisition system, comprising the following steps: S1. On-site deployment and installation.

[0049] S11. Equipment inventory and tag number planning.

[0050] S111. Confirm the equipment list. Number of vibration signal acquisition nodes N (N = 12~24), 1 control module, 1 industrial router, 1 host computer (monitoring software installed and IP-based receiving port configured).

[0051] S112. Establish node identity mapping. Each node has a factory-fixed device ID (immutable). Preferably, different projects can generate new node names (variable, facilitating location and maintenance). Naming scheme A: Two unsigned integers "01~99"; Naming scheme B: "Location + Serial Number", such as "Hoop Cutter 01~Hoop Cutter 32" "Cutter Head 01~Cutter Head 09". Print corrosion-resistant labels and affix them to the node shell.

[0052] S113. Plan installation sites and generate a sensor bitmap. Distribute sensor points evenly within the hobbing cutter boxes at the rear of the cutter head (at least 8 locations, 1 point per box), within several hobbing cutter boxes at the edge of the cutter head (at least 3 locations, 1 point per box), and at the center of the rear of the cutter head (≥1 location). Generate a "sensor bitmap" and archive it. The sensor bitmap includes both sensor locations and expected routing adjacency relationships.

[0053] S114. Compile a list of communication and power supply harnesses. Confirm that the only wired power source is the industrial router's 12V DC power cable (1.5m long, cross-sectional area ≥0.5mm², double insulation). The control module and the industrial router are connected by a single network cable. Preferably, a PoE power supply / communication hybrid is used, with all other links being wireless. The nodes and control module contain both LoRa and LAN communication units, while the host computer and industrial router only have a LAN unit. The LoRa unit is responsible for networking, commands, and status information, while the LAN handles high-bandwidth data.

[0054] S115. Create an "Installation and Initialization Work Sheet", which includes: Name-ID lookup table, sensor bitmap, router SSID / key, host computer IP-based receiving port address, maintenance log, and signature column.

[0055] S12, Node mechanical fixing and energy harvesting device assembly.

[0056] S121. Base Welding and Treatment. A steel base of the same size as the bottom surface of the node is welded to the surface of the equipment to be monitored, with M6 bolt holes provided on the base. The mounting surface is degreased and rusted with alcohol, and after drying, a 0.5mm nitrile rubber (NBR) pad is applied to increase friction and isolate vibration.

[0057] S122. The node is installed using M6 stainless steel quick-release bolts. When installing the node, align the bottom surface of the protective shell with the base, pass the quick-release bolts through the bolt holes on the bottom surface of the protective shell and the base, and tighten the quick-release bolts clockwise to complete the installation.

[0058] S123, Energy harvesting device orientation calibration: Align the relative motion axis of the coil-magnet energy harvesting assembly inside the node with the main vibration direction of the cutterhead (tunneling direction), with an angle error ≤ ±15°.

[0059] S124. Minimize antenna and metal obstruction. The hidden antenna face of the node shell faces the cavity to avoid direct obstruction by thick metal parts.

[0060] S13, Industrial Router Fixed and Power Supply Configuration.

[0061] S131. A dedicated bracket is installed next to the 12V power supply box at the rear of the cutter head; the industrial router is fixed at four points using M6 quick-release bolts.

[0062] S132. Connect to the 12V power supply box using a 1.5m DC power cord (≥0.5mm², double insulation).

[0063] S133. Industrial router initialization: Write the LAN name and password into the work order; set the transmit power to the highest level; enable the whitelist (only allow the control module and host computer to access).

[0064] S14. Placement and physical connection of the control module.

[0065] S142, Fixed installation: The control module and the router are arranged in the same bracket layer.

[0066] S142. Communication and Power Supply: The control module is connected to the router via a network cable, and the PoE interface is used to supply power and communicate with the control module via the network cable.

[0067] S143, Wireless Role Preset: The LoRa interface of the control module is used as the command broadcasting end; the local area network interface is used as the high-bandwidth vibration data uplink relay client (interconnected with the host computer through a router).

[0068] S2, Power On, Self-Test and Network Initialization.

[0069] S21, Power on.

[0070] S211. Power on in sequence: first the industrial router, then the control module, and finally each node, to avoid the wireless nodes running idle.

[0071] S212. Environmental Confirmation: There are no open flames or high-voltage operations in the vicinity. Personnel have been evacuated from the warning area for rotating parts.

[0072] S22, Node and control module power-on self-test.

[0073] S221, Node self-check.

[0074] Three-axis channel connectivity and bias: detection of X / Y / Z zero bias and noise.

[0075] MCU and Cache: Fast Read / Write Test.

[0076] RTC Preliminary Calibration: Drift < ±10 ppm compared to internal reference.

[0077] Energy extraction open circuit voltage test: UAC≥20 mV.

[0078] Battery voltage ≥ 3.3 V.

[0079] Once the node passes the test, it uploads status information to the control module: "READY / ID / Name / Battery Voltage / Power Harvesting Voltage" (LoRa).

[0080] S222, Control module self-test.

[0081] LoRa transmit / receive loopback successful.

[0082] LAN router association successful (RSSI≥) 65 dBm).

[0083] Fast read / write test of cache.

[0084] RTC calibration, deviation ≤ ±10 ppm.

[0085] S223, LoRa Mesh self-organizing network establishment.

[0086] S2231, Neighbor Discovery: The node sends "ID, Name, Signal Value, Battery Voltage / Power Capture Voltage" every 2 seconds at 433 MHz / 14 dBm.

[0087] S2232, Route Calculation: The control module aggregates the neighbor table and calculates routes based on signal thresholds ≥ The minimum number of hops at 110 dBm (usually 0 hops, indicating direct access from the node to the control module) is used to generate the uplink route; if necessary, one-hop or two-hop relays are specified for edge nodes.

[0088] S2233. Connectivity acceptance: The end-to-end packet loss rate is ≤1% (100 packets per window) and the round-trip delay is ≤100 ms to pass; nodes that fail will be automatically switched to the suboptimal relay and retested, with a maximum of 3 retries.

[0089] A soft positioning-routing coupling strategy is adopted: during the neighbor discovery phase, the signal value distribution is used to infer local spatial relationships, and a "short link, low hop count" topology is prioritized to reduce the transmission energy consumption of subsequent data reporting and reserve energy space for ultra-long endurance.

[0090] S224, time synchronization across the entire network.

[0091] The S2241 control module broadcasts a timestamp every 60 seconds (minimum unit 1ms).

[0092] S2242, the node performs fine-tuning (temperature drift compensation) on the RTC to converge the clock deviation of the entire network to ≤±1 ms.

[0093] S225. Network Locking and Baseline Preservation. The control module issues commands to freeze the current route and save the adjacency table, clock information, and node status information.

[0094] S3, Energy Harvesting Link Calibration and Threshold Setting (Initialization of the Ultra-Long Battery Life Core).

[0095] S31. Confirm the direction and amplitude of the energy harvesting device.

[0096] S311. Direction verification: Verify point by point that the energy harvesting axial arrow is consistent with the installation record, with an error ≤ ±15°.

[0097] S312. Amplitude Test: Under three typical working conditions—TBM stationary, TBM moving forward but not tunneling, and TBM tunneling at low speed—the open-circuit AC voltage U is measured. AC Record values ​​of ≥20 / 100 / 200mV and archive them.

[0098] S32, Rectifier-Filter-Boost Link Calibration.

[0099] S321, Rectification and Filtering: Confirm that the forward voltage drop of the rectifier bridge is ≤0.5 V@1 mA; the ripple factor is ≤10%.

[0100] S322, Boost Start-up: Confirm that the boost chip can start oscillating when the input is ≥20 mV; the output bus is preset to 4.2 V (the upper limit of single-cell lithium battery charging) and the steady-state deviation is ≤±1%.

[0101] S33, Initialization of the range management system and SOC estimation strategy.

[0102] S331, Enable Battery Life Management System: Enable overcharge (4.2V), over-discharge (3.3V), short circuit protection, and cell temperature protection. 10~60℃).

[0103] S332 and SOC estimation: A hybrid algorithm of "open-circuit voltage + coulomb integration" is adopted (refreshed every 1 second). The open-circuit voltage method is used for reference calibration after a static period of ≥10 min, and the coulomb integration is used for dynamic operating condition tracking; the initialization error is ≤±2%.

[0104] S333, Set shallow charge / discharge window: SOC lower limit 30%, upper limit 80%; this serves as the default health window. Cycling within this window can significantly extend lithium battery life (better than full charge / discharge).

[0105] S4. Data link initialization and monitoring.

[0106] S41, Host computer—Router access.

[0107] S411. The host computer connects to the industrial router via wireless LAN (SSID / key is on the work order) and obtains / sets a static IP address.

[0108] S412. The host computer starts the IP-based receiving service and sets the receiving port address.

[0109] S42. Control Module: The control module establishes an IP-based communication connection with the host computer. After completing the connection establishment and link connectivity confirmation, it enters the data push standby state.

[0110] S43, Access Authentication and Minimum Data Frame.

[0111] S431, the control module sends "HELLO_ (timestamp, version number, number of nodes)"; the host computer responds with "ACK".

[0112] S432, Minimum Test Frame: Send a 128-byte test data frame containing the example node ID and timestamp; after the host computer verifies the data, record "LINK_READY", and then enter the normal operation phase (detailed steps are in the operation section and will not be elaborated in this initialization section).

[0113] S5. Data Acquisition Task Generation and Power Consumption Prediction Scheduling (Energy First). The process is as follows: Figure 4 As shown.

[0114] S51, Energy-Power Consumption Inventory and Predictive Modeling.

[0115] S511, Set the network-wide statistical period as follows (5 min), in each It completes energy / power consumption sampling, uploading, and aggregation within the system.

[0116] S512, Node-side energy measurement and reporting.

[0117] Each node in Recorded at a resolution of 1 second: Energy extraction side voltage / current (after boost); Battery-side voltage / current; SOC is estimated using a hybrid method of "open-circuit voltage + coulomb integration" (refreshed every 1 second). Load operating time (the operating time of each sub-module such as sensing, processing, and local area network transmission).

[0118] This node calculates Internal average energy harvesting power and average load power And report "node energy information (node ​​ID, , , SOC).

[0119] S513, Control module side aggregation and prediction.

[0120] The control module is in each After completion, calculate the power difference between each node:

[0121] Linear regression is then performed based on the time series of the most recent M statistical periods (preferably M=6, covering approximately 30 minutes) to obtain the predicted power difference value for the next period. .

[0122] S514. Generate an "Energy Distribution Map": For all network nodes... , A two-dimensional table and a thermal visualization matrix are established based on SOC and spatial location to provide a basis for subsequent scheduling.

[0123] S52, Frequency-Duty Cycle-Mode Adaptive (Three Levels, Four States).

[0124] S521, with SOC and For dual threshold criteria (the upper and lower thresholds of SOC can be customized between 30% and 80%, and the optimal loop range of SOC is set to 40%-60% here, which does not conflict with the default healthy loop range of 30%-80%), four working states are defined: Continuous monitoring status: and ; Balance monitoring state: and ; Intermittent monitoring state: or ; Wake-up monitoring state: and .

[0125] S522, Parameter Distribution - Frequency and Duty Cycle Table, as shown in Table 1.

[0126] Table 1. Control module issues sampling parameters for this cycle based on status (independent for each node):

[0127] S523, Energy Priority and Task Granularity.

[0128] When multiple alternative nodes exist in the same spatial sub-region: Prioritize selecting... Larger Those with higher frequencies are responsible for high-frequency data acquisition; The lower-level device will automatically downgrade, reducing the number of times it sends signals over the local area network.

[0129] S6. Collaborative data acquisition and execution (data acquisition and energy harvesting are fully synchronized, with simultaneous data acquisition and charging). The process is as follows: Figure 5 As shown.

[0130] S61, Synchronous wake-up and window alignment.

[0131] S611, the control module broadcasts a sampling instruction, which includes the "start point of the next sampling window" and the "window length".

[0132] S612. After receiving the data, the node makes a fine adjustment based on the RTC and simultaneously enters the sampling window within an error of ≤±1 ms.

[0133] S62, Synchronous Sampling and Power Snapshot.

[0134] S621. Upon entering the window, simultaneous three-axis acceleration acquisition will be enabled, with the sampling frequency set in step 402. .

[0135] S622. The energy harvesting link remains in operation all time within the sampling window, while recording power snapshots: "energy harvesting side voltage / current, bus voltage, battery voltage / current, and SOC" are collected once per second.

[0136] S623. During sampling, a data frame is formed in the local buffer: frame header (0xAAAA), node ID, sampling point timestamp. Power snapshot array, triaxial data, frame tail (0xFFFF).

[0137] S63, Low-power transmission and retransmission strategy.

[0138] S631, Local Area Network (LAN) Data Push. After the sampling window begins, nodes sequentially report data frames to the control module via the LAN according to a predetermined transmission sequence list (to avoid energy consumption caused by retransmission due to concurrent conflicts).

[0139] S632, Retransmission upon Failure. If no confirmation is received within a timeout period, retransmission will be performed at most once; if it still fails, the data will be stored in the local Flash memory and merged and retransmitted in the next round (if energy is insufficient, storage will be prioritized to avoid repeated power consumption).

[0140] S64, Hibernation and Energy Replenishment after Window.

[0141] S641. After leaving the sampling window, the node shuts down high-power units (sensing, MCU high-speed clock, LAN transmission), leaving only RTC and LoRa in standby mode, with a static current ≤10 μA.

[0142] S642, the power acquisition link continues to work until SOC ≥ 80%, then switches to trickle maintenance; if SOC is still below 40%, the node actively reports "LOW_SOC" to the control module, providing a signal for the next round of sampling downgrading.

[0143] S7, Network-wide energy closed loop and lifespan balance (system-level optimization).

[0144] S71, Energy Telemetry Summary and Health Window Maintenance. The control module summarizes the network-wide SOC distribution every 10 minutes. The distribution and spatial heat map were used to check whether each node was still within the 30%–80% health window.

[0145] S72. Inter-node lifetime balancing scheduling (spatial equivalent replacement). Under the premise of ensuring unchanged spatial coverage, the load is allocated to nodes with more energy surplus, so as to maximize the remaining lifetime of the "shortest lifetime node" in the entire network (Max-Min).

[0146] S721. Construct sensor partitions. Based on the sensor bitmap and the empirical influence range, construct sub-region coverage sets; nodes within the same sub-region can be equivalently replaced.

[0147] S722, Task Migration. If both "low SOC and..." exist within the sub-region... Node A of “high SOC and” and “high SOC and If node B is a node that will migrate some of the window tasks from node A to node B in the next round.

[0148] S723, Link Energy Consumption Suppression. Within the range of alternatives, prioritize nodes with fewer route hops and better signal strength to undertake high-frequency data collection, shorten the reporting path, and reduce local area network energy consumption.

[0149] S8, Shutdown / Maintenance and Reset (Lifetime-Friendly Shutdown).

[0150] S81, Smooth shutdown of planned shutdown.

[0151] S811, the control module issues a "smooth shutdown" command, requiring all nodes to execute: Continue charging after the next sampling window to increase the SOC to 80%±1% (for better storage). Enter deep sleep mode (static current ≤ 3 μA), retaining only RTC and wake-up interrupt.

[0152] S812, the control module records the shutdown snapshot: the last SOC, temperature, cumulative cycle count, and last energy map, and archives it to the host computer.

[0153] S82, Self-protection and power saving during unplanned downtime. If the industrial router / host computer goes offline or the construction party stops the operation abruptly, the node automatically enters a local low-power self-sustaining strategy: only "wake-up monitoring" is retained, and power is continuously drawn to keep the SOC within a healthy window.

[0154] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A TBM tool head vibration signal acquisition system with ultra-long battery life, characterized in that, It includes several wireless vibration signal acquisition nodes, and the several wireless vibration signal acquisition nodes are sequentially connected to a control module, an industrial router and a host computer. The wireless vibration signal acquisition nodes are fixed on the front and rear of the TBM cutter head and the inner surface of the hobbing cutter box. Each node independently acquires the three-dimensional vibration signal at its location. The control module is used to issue acquisition commands to each node, perform network-wide time synchronization, execute energy-power prediction and adaptive scheduling of sampling status, and receive and forward node data. The industrial router is powered via a wired connection and serves as a communication gateway connecting the control module and the host computer. The host computer runs monitoring software and receives and processes vibration data through the high-speed wireless LAN link of the industrial router.

2. The TBM tool head vibration signal acquisition system with ultra-long battery life according to claim 1, characterized in that, The wireless vibration signal acquisition node includes a protective shell, which includes a LoRa antenna (161), a reserved maintenance port (162), a local area network antenna (163), an aluminum alloy shell (164), an aluminum alloy top cover (165), and quick-release bolt holes (166). The LoRa antenna (161), the reserved maintenance port (162), and the local area network antenna (163) are located on the side of the aluminum alloy shell (164), the aluminum alloy top cover (165) is located on the top of the aluminum alloy shell (164), and the quick-release bolt holes (166) are located through the bottom of the aluminum alloy shell (164). The aluminum alloy shell (164) has a first cavity and a second cavity inside.

3. The TBM tool head vibration signal acquisition system with ultra-long battery life according to claim 2, characterized in that, The first cavity is provided with a first circuit board and a lithium battery (14). The lithium battery (14) is located above the first circuit board. The first circuit board integrates an axial acceleration sensor (11), a microprocessor unit (12), and a wireless communication unit (13).

4. The TBM tool head vibration signal acquisition system with ultra-long endurance according to claim 2, characterized in that, An energy harvesting unit is provided in the second cavity. The energy harvesting unit includes a coil (151), a permanent magnet (152), a guide rail (153), a bearing (154), a limiting spring (155), a rectifier and filter unit (156), a boost and voltage regulator unit (157), and a battery life management module (158). The rectifier and filter unit (156), the boost and voltage regulator unit (157), and the battery life management module (158) are integrated on the second circuit board. The coil (151), the permanent magnet (152), the guide rail (153), the bearing (154), and the limiting spring (155) are arranged coaxially and are located below the second circuit board.

5. The TBM tool head vibration signal acquisition system with ultra-long battery life according to claim 1, characterized in that, The control module is configured to execute the following power prediction and frequency adaptation strategy: The average harvested power within the most recent N sampling windows Average operating power consumption SOC and cache usage are inputs. If the SOC is higher than 60%, increase the sampling rate or extend the sampling window; if If the SOC is below 40%, the system will be downgraded in the following order: reducing the sampling rate, shortening the sampling window, and enabling wake-up monitoring, until energy self-sufficiency is restored.

6. The TBM tool head vibration signal acquisition system with ultra-long endurance according to claim 5, characterized in that, The control module is also configured to perform shallow charge / discharge lifetime management: The SOC is corrected based on the charge / discharge curves reported by the wireless vibration signal acquisition nodes, limiting the SOC to slowly fluctuate within 30% to 80%. When the SOC is predicted to reach the upper limit of 80%, the boost duty cycle is reduced or trickle charging is switched. When the SOC is predicted to reach the lower limit of 30%, the next sampling window is postponed to ensure long-term small ΔSOC cycling of the battery, thereby significantly improving the battery's equivalent cycle life.

7. A method for acquiring vibration signals from a TBM tool turret with ultra-long battery life, characterized in that, The TBM cutter head tool vibration signal acquisition system based on any one of claims 1-6 includes the following steps: System deployment and initialization: Install wireless vibration signal acquisition nodes in the TBM cutter head and hobbing area, fix industrial routers and control modules, establish a wireless communication network and perform full network time synchronization; Energy harvesting link calibration, calibration of the output of electromagnetic induction energy harvesting unit, initialization of the range management system and setting of SOC shallow charge and discharge window; Energy prediction and adaptive scheduling: The control module periodically obtains the average energy extraction power and average load power of each node, predicts the net energy difference for the next cycle, and adaptively allocates sampling frequency and duty cycle to each node accordingly. Collaborative data acquisition and reporting: The control module issues a unified sampling window command, and each node synchronously performs vibration signal acquisition and vibration energy acquisition within the window to achieve "simultaneous acquisition and charging". After the sampling is completed, the data is reported and then enters a low-power sleep state. Lifecycle maintenance involves the control module performing network-wide energy balancing scheduling, migrating data acquisition tasks from low-energy nodes to high-energy nodes to balance the lifespan of all nodes in the network; and controlling nodes to enter a lifespan-friendly smooth shutdown or low-power self-sustaining mode when shutting down.

8. The method for acquiring vibration signals of a TBM tool turret with ultra-long battery life according to claim 7, characterized in that, In the system deployment and initialization, a LoRaMesh self-organizing network is established, including neighbor discovery, route calculation and connectivity verification; The nodes and control modules transmit commands and status information via a LoRa link, and vibration data is transmitted via a high-speed wireless LAN link of an industrial router.

9. The method for acquiring vibration signals of a TBM tool turret with ultra-long battery life according to claim 8, characterized in that, The energy prediction and adaptive scheduling includes the following steps: Define the statistical period ; Node in each Internal measurement and calculation of average energy harvesting power and average load power The report is then sent to the control module. The control module calculates the power difference. Based on historical data from the most recent M statistical periods, the power difference for the next statistical period is predicted. ; The control module is based on the prediction Based on the current SOC of the node and the preset dual threshold criteria, the working state of the node is divided into continuous monitoring state, balanced monitoring state, intermittent monitoring state or wake-up monitoring state, and a corresponding sampling frequency, sampling window duration and sampling period are assigned to each state.

10. The method for acquiring vibration signals of a TBM tool turret with ultra-long battery life according to claim 9, characterized in that, In the collaborative acquisition, nodes simultaneously perform vibration signal acquisition and energy acquisition from environmental vibration within the same sampling window, and charge the battery to achieve simultaneous acquisition and charging; at the same time, power snapshot data is recorded and reported within the sampling window.

Images