Self-powered vibration sensor based on multi-mode energy capture

Abstract

The invention provides a self-powered vibration sensor based on multi-mode energy capture, and the sensor comprises a multi-mode vibration energy capture unit which comprises a piezoelectric ceramic piece and an electromagnetic induction coil-permanent magnet assembly, and the piezoelectric ceramic piece and the electromagnetic induction coil-permanent magnet assembly are installed on the surface of equipment in parallel, the piezoelectric ceramic is used for capturing 100-200Hz high-frequency vibration energy; the electromagnetic induction coil-permanent magnet assembly is used for capturing low-frequency vibration energy of 5-50 Hz; the energy acquisition power supply management circuit comprises a double-path rectification filter circuit, a first bidirectional DC-DC converter and a super capacitor / lithium battery hybrid energy storage unit; the first bidirectional DC-DC converter is connected with the two-way rectification filter circuit and the super capacitor / lithium battery hybrid energy storage unit at the same time, and is used for controlling charging and discharging of the super capacitor / lithium battery hybrid energy storage unit based on a preset dynamic impedance matching network and an energy priority scheduling strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of vibration monitoring, and in particular, relates to a self-powered vibration sensor based on multi-modal energy capture. BACKGROUND

[0002] In a source network load integrated smart park, the stable operation of various devices such as transformers, energy storage batteries, fans, and pumps is crucial. The vibration signals generated by these devices during operation can effectively reflect their health status, so monitoring the vibration signals is a key means to ensure the safe and efficient operation of the devices.

[0003] Traditional vibration sensors mostly rely on external power supply, such as mains or batteries. In a distributed source network load device scenario, using mains power supply requires large-scale wiring, which not only has high cost, but also has great difficulty in wiring in some complex environments, and also lacks flexibility; while battery power supply faces the problem of regular replacement, which not only increases maintenance cost, but also pollutes the environment with waste batteries.

[0004] Some existing self-powered vibration sensor technologies, such as self-powered vibration sensors based on piezoelectric ceramics, can convert part of the vibration energy into electrical energy, but have many limitations. Such sensors can usually only adapt to vibrations in a specific frequency range, and in the low-frequency vibration (such as transformer core vibration frequency 50-100Hz) or complex vibration environment commonly seen in source network load devices, the energy capture efficiency is low, making it difficult to meet the power consumption requirements of stable sensor operation.

[0005] In order to solve the above problems, people have been seeking an ideal technical solution. SUMMARY

[0006] The purpose of the present application is to overcome the shortcomings of the prior art, and to provide a self-powered vibration sensor based on multi-modal energy capture.

[0007] In order to achieve the above purpose, the technical scheme adopted by the present application is: a self-powered vibration sensor based on multi-modal energy capture, comprising: A multi-modal vibration energy capture unit, comprising a piezoelectric ceramic sheet and an electromagnetic induction coil-permanent magnet assembly installed in parallel on the surface of a device, the piezoelectric ceramic is used to capture high-frequency vibration energy of 100-200Hz; the electromagnetic induction coil-permanent magnet assembly is used to capture low-frequency vibration energy of 5-50Hz; The energy harvesting power management circuit includes a dual-channel rectification filter circuit, a first bidirectional DC-DC converter, and a super capacitor / lithium battery hybrid energy storage unit. The dual-channel rectification filter circuit rectifies and filters voltage signals output by a piezoelectric ceramic and voltage signals output by an electromagnetic induction coil-permanent magnet assembly. The first bidirectional DC-DC converter is connected to the dual-channel rectification filter circuit and the super capacitor / lithium battery hybrid energy storage unit, and is used to control charging and discharging of the super capacitor / lithium battery hybrid energy storage unit based on a preset dynamic impedance matching network and an energy priority scheduling strategy. A vibration sensor is connected to the super capacitor / lithium battery hybrid energy storage unit and is used to collect vibration information. A LoRa wireless transmission module is used for data transmission. A microcontroller is used to collect voltage states of the super capacitor, and control working states of the vibration sensor and the LoRa wireless transmission module according to the voltage states.

[0008] In a possible embodiment, the dynamic impedance matching network includes an impedance detection unit and a switch capacitor array. The impedance detection unit is connected to the dual-channel rectification filter circuit, and is used to sample output voltages and output currents of the dual-channel rectification filter circuit in real time, and calculate source impedance according to the collected output voltages and output currents. The switch capacitor array includes eight channels of different capacitance values and corresponding switches, and is used to dynamically control the number of opened switches to update capacitor combinations, so that a matching error between a load impedance and the source impedance meets a preset error condition.

[0009] In a possible embodiment, the switch circuit array dynamically controls the number of opened switches, including: A vibration frequency is obtained by analyzing the vibration information through fast Fourier transform, and a source impedance range is determined according to the vibration frequency. A matching capacitance value is calculated according to a formula C=1 / (2πfz), where C is the capacitance, f is the vibration frequency, and Z is the source impedance. The matching capacitance value is controlled according to the calculated matching capacitance value to control switch actions of the corresponding capacitance circuit, so as to realize the matching capacitance value.

[0010] In a possible embodiment, the switch capacitor array is arranged at a distance of less than 10 mm from an output end of the dual-channel rectification filter circuit, and a width of a switch wiring of the switch capacitor array is greater than or equal to 20 mil.

[0011] In a possible embodiment, the energy priority scheduling strategy includes: Vibration stage: After the piezoelectric ceramic and the electromagnetic induction coil-permanent magnet assembly are independently rectified, the first bidirectional DC-DC converter is directly used to charge the super capacitor, and the voltage of the super capacitor is detected, when the voltage of the super capacitor is greater than 4.5V, the super capacitor charges the lithium battery in a trickle mode through the second DC-DC conversion circuit; At the same time, the super capacitor is used as the power supply to supply power to the vibration sensor, the LoRa wireless transmission module and the microprocessor by default; when the voltage of the super capacitor is less than 3.3V, the lithium battery is switched to the power supply, and the lithium battery supplies power to the vibration sensor, the LoRa wireless transmission module and the microprocessor; wherein, during the discharging process of the lithium battery, the lithium battery is still reversely charged through the double rectification filter circuit, the first bidirectional DC-DC converter, the super capacitor and the second DC-DC conversion circuit; Vibration-free sleep stage: the first bidirectional DC-DC converter enters low-power standby, only the lithium battery supplies power to the wake-up module of the first bidirectional DC-DC converter, and the super capacitor is in a power-off protection state to avoid leakage current loss.

[0012] The microprocessor collects the supply voltage of the power supply in real time, when the supply voltage is greater than or equal to 3.3V, the vibration sensor and the LoRa wireless transmission module are woken up, the vibration sensor collects data at a frequency of 10Hz, and the LoRa wireless transmission module transmits data once every 60s; when the supply voltage is less than 2.5V, the LoRa wireless transmission module is turned off, and the vibration sensor enters an intermittent wake-up mode, wakes up once every 500ms to collect vibration information, and the low-power monitoring function of the multi-modal vibration energy capture unit and the microprocessor is retained.

[0013] In a possible embodiment, when the supply voltage of the power supply is greater than 4V, the microprocessor increases the data collection frequency of the vibration sensor to 100Hz to obtain accurate device vibration information; When the voltage of the power supply is greater than or equal to 3.3V and less than 4V, the microprocessor reduces the data collection frequency of the vibration sensor to 10Hz to balance the monitoring accuracy and energy consumption; When the voltage of the power supply is greater than or equal to 2.5V and less than 3.3V, the LoRa transmission interval is extended from 60s to 300s.

[0014] The present application has outstanding substantial characteristics and significant progress compared with the prior art, in particular: 1. The wideband energy capture capability is significantly improved: the piezoelectric-electromagnetic multi-modal energy collection technology is fused, and the average output power is greater than or equal to 50muW in the vibration frequency range of 5-200Hz, which is increased by 300% compared with the single piezoelectric scheme.

[0015] 2. Low power consumption and long-term operation are achieved: Through adaptive energy management and low power consumption design, the average power consumption of the system is significantly reduced; with the help of supercapacitors, it can work continuously for 365 days in a moderate vibration environment without external power supply, which greatly reduces the operation and maintenance costs.

[0016] 3. Significantly effective precise monitoring and collaborative optimization: By extracting feature parameters through fast Fourier transform, the amount of data transmitted per transmission and the power consumption of transmission are significantly reduced.

[0017] In summary, this invention, through innovative multimodal energy harvesting technology, adaptive energy management strategy, and low-power design, enables vibration sensors to operate under long-term self-powered conditions in complex vibration environments, improves the reliability and accuracy of equipment vibration monitoring, reduces operation and maintenance costs, and provides strong support for the stable operation of power generation, grid, load and storage systems and smart parks. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of the present invention.

[0019] Figure 2 This is a schematic diagram of energy routing logic.

[0020] Figure 3 This is the circuit diagram of the first bidirectional DC-DC converter of the LT8302.

[0021] Figure 4 This is a circuit architecture diagram for energy management.

[0022] Figure 5 This is a schematic diagram of the sensor signal processing and wireless transmission circuit.

[0023] Figure 6 This is a system workflow diagram.

[0024] In the figure, 1. Piezoelectric ceramic sheet; 2. Electromagnetic induction coil; 3. Permanent magnet; 4. Elastic support; 5. Dual-channel rectifier and filter circuit; 6. Supercapacitor; 7. MCU microcontroller; 8. MEMS accelerometer; 9. LoRa wireless transmission module; 10. Equipment surface. Detailed Implementation

[0025] The existing technical solutions for self-generating vibration sensors mainly consist of the following parts: Energy Harvesting Module: This module uses a single piezoelectric ceramic sheet as the energy harvesting element. Utilizing the piezoelectric effect of the ceramic, it converts the mechanical energy generated by equipment vibration into electrical energy. The piezoelectric ceramic sheet is typically directly attached to the equipment surface. When the equipment vibrates, the piezoelectric ceramic sheet deforms, thereby generating an electrical signal.

[0026] Energy management circuit: This circuit generally includes a rectifier bridge, a filter capacitor, and a voltage regulator. The rectifier bridge converts the alternating current generated by the piezoelectric ceramic into direct current, the filter capacitor is used to smooth the voltage fluctuations, and the voltage regulator converts the unstable direct current voltage into a stable direct current voltage to power the subsequent sensors and other circuits.

[0027] Sensor module: A MEMS vibration acceleration sensor is used to collect vibration data, and a microcontroller (MCU) is used for data processing and control. The MEMS sensor can measure the vibration acceleration of the device, and the MCU analyzes and processes the collected data.

[0028] Transmission module: The processed data is uploaded to the park management platform through wireless communication technologies such as Bluetooth or Wi-Fi, so that the operator can real-time understand the vibration state of the device.

[0029] However, this existing technical solution has obvious deficiencies: 1. Low energy harvesting efficiency: The single piezoelectric ceramic energy harvesting method cannot cover a wide range of vibration frequencies (5-200Hz) of the source network load storage device, especially in low frequency (such as below 20Hz) vibration scenarios, the output power is extremely low, usually less than 10μW, which is difficult to support the continuous and stable work of the sensor (generally power consumption is about 50μW).

[0030] 2. Poor power supply stability: The existing energy management circuit lacks adaptive adjustment mechanism and cannot adjust the power supply strategy in real time according to the energy harvesting and load power consumption. When the vibration energy fluctuates, it cannot guarantee the stable power supply of the sensor and the transmission module, resulting in the interruption of data collection and transmission, affecting the continuity and accuracy of device state monitoring.

[0031] 3. High system power consumption: The traditional sensor and transmission module are not optimized for self-power generation scenarios, and the sleep current is large, such as the Bluetooth module sleep current is about 10μA. In the case of limited self-power generation energy, high-power system components will quickly consume energy and shorten the working time of the sensor.

[0032] 4. The traditional "or gate diode" merging scheme has the limitation of one-way energy flow. When the piezoelectric / electromagnetic output voltage is inconsistent, it will generate 15%-30% of circulating loss, and cannot use the energy storage unit to reverse power supply in weak vibration, resulting in high power supply interruption rate of the system in intermittent vibration scenarios.

[0033] 5. The power consumption of Bluetooth or Wi-Fi transmission module is high, which will exacerbate the consumption of energy in the case of self-power generation, further reducing the stability of the system.

[0034] The present application aims to solve the problems of low energy capture efficiency, unstable power supply and high system power consumption of the existing self-powered vibration sensor in the application scene of source network load storage smart park. Through innovative multi-modal energy collection technology, adaptive energy management strategy and low power consumption design, long-term self-powered operation of the vibration sensor in a complex vibration environment is realized, the reliability and accuracy of equipment vibration monitoring are improved, the operation and maintenance cost is reduced, and strong support is provided for the stable operation of the source network load storage system and the smart park.

[0035] The technical solutions of the present application will be further described in detail through specific embodiments.

[0036] As Figures 1-6 The structure diagram, circuit diagram and working principle diagram of the self-powered vibration sensor based on multi-modal energy capture.

[0037] Embodiment 1 The present embodiment provides a self-powered vibration sensor based on multi-modal energy capture, as shown in Figure 1 The self-powered vibration sensor based on multi-modal energy capture comprises: A multi-modal vibration energy capture unit, comprising a piezoelectric ceramic sheet 1 and an electromagnetic induction coil 2-permanent magnet 33 assembly, wherein the piezoelectric ceramic sheet 1 is directly installed on the surface of the equipment 10 and closely adheres to the equipment; and the electromagnetic induction coil 2-permanent magnet 33 assembly is suspendedly installed on the surface of the equipment 10 through an elastic support 4.

[0038] Specifically, the piezoelectric ceramic sheet 1 is made of lead zirconate titanate (PZT-5H) material, with a thickness of 0.5 mm and silver electrode plated on the surface, for capturing high-frequency vibration energy of 100-200 Hz; the electromagnetic induction coil 2 is wound with enameled wire, with 2000 turns and a wire diameter of 0.1 mm; the permanent magnet 33 is made of neodymium iron boron N35, with an air gap spacing of 2 mm; and the electromagnetic induction coil 2 and the permanent magnet 33 form an assembly for capturing low-frequency vibration energy of 5-50 Hz.

[0039] It can be understood that, since the piezoelectric ceramic sheet 1 and the electromagnetic induction coil 2-permanent magnet 33 assembly are both arranged on the surface of the equipment 10, a wide frequency vibration range of 5-200 Hz can be covered.

[0040] In addition, in actual use, the vibration coupling coefficient of the multi-modal vibration energy capture unit can be optimized by using ANSYS simulation software, for example, by adjusting the elastic modulus of the elastic support 4, the relative position of the piezoelectric ceramic sheet 1 and the electromagnetic induction coil 2-permanent magnet 33 assembly and other parameters, to ensure that the energy capture efficiency can reach more than 85% under different frequency vibrations.

[0041] The energy harvesting power management circuit comprises a dual-channel rectification filter circuit 5, a first bidirectional DC-DC converter and a super capacitor 6 / lithium battery hybrid energy storage unit.

[0042] The dual-channel rectification filter circuit 5 is configured to rectify and filter voltage signals output by the piezoelectric ceramic 1 and the electromagnetic induction coil 2-permanent magnet 33 assembly respectively.

[0043] Specifically, the dual-channel rectification filter circuit 5 comprises a bridge rectification circuit composed of Schottky diodes SBD1040 and a rectification circuit composed of an LTC4357 synchronous rectification chip. The alternating current output by the piezoelectric ceramic 1 is converted into direct current by the bridge rectification circuit, and the alternating current output by the electromagnetic induction coil 2-permanent magnet 33 assembly is rectified by the synchronous rectification chip.

[0044] The first bidirectional DC-DC converter is connected to the dual-channel rectification filter circuit 5 and the super capacitor 6 / lithium battery hybrid energy storage unit, and is configured to control the charging and discharging of the super capacitor 6 / lithium battery hybrid energy storage unit based on a preset dynamic impedance matching network and an energy priority scheduling strategy.

[0045] Specifically, the first bidirectional DC-DC converter adopts an LT8302 first bidirectional DC-DC converter with a switching frequency of 500 kHz and supports ±3A current bidirectional flow. For details, refer to Figure 2 , which comprises an LT8302 isolated flyback converter and a current transformer FC-SCT4.2-1:50.

[0046] The super capacitor 6 is a 1F / 5.5V super capacitor 6, and the lithium battery is a 100mAh lithium polymer battery (3.7V).

[0047] Further, as shown in Figure 4 , the energy harvesting power management circuit of the present embodiment comprises a dynamic power path management, battery detection protection and control circuit in addition to the hybrid energy storage unit composed of the lithium battery backup battery and the super capacitor 6 energy storage element. The dynamic power path management, battery detection protection and control circuit jointly control the switching of the preset dynamic impedance matching network and the execution of the energy priority scheduling strategy to control the charging and discharging of the super capacitor 6 / lithium battery hybrid energy storage unit.

[0048] Specifically, the energy priority scheduling strategy comprises: Vibration stage: after the piezoelectric ceramic and the electromagnetic induction coil 2-permanent magnet 33 assembly are independently rectified, the first bidirectional DC-DC converter directly charges the super capacitor 6, and detects the voltage of the super capacitor 6 at the same time, when the voltage of the super capacitor 6 is greater than 4.5V, the super capacitor charges the lithium battery in a trickle mode to prevent overcharging; at the same time, the super capacitor 6 is used as the power supply by default to supply power to the vibration sensor, the LoRa wireless transmission module 9 and the microprocessor 7; when the voltage of the super capacitor 6 is less than 3.3V, the lithium battery is switched to the power supply, and the lithium battery supplies power to the vibration sensor, the LoRa wireless transmission module 9 and the microprocessor 7; wherein, during the discharging process of the lithium battery, the lithium battery is still reversely charged by the double rectification filter circuit 5, the first bidirectional DC-DC converter, the super capacitor 6 and the second DC-DC conversion circuit, that is, the lithium battery is in a floating charging mode. Specifically, the second DC-DC conversion circuit can adopt an existing DC-DC conversion circuit, of course, in some embodiments, the LT8302 bidirectional DC-DC converter can also be used.

[0049] No-vibration sleep stage: the first bidirectional DC-DC converter enters low-power standby, only the lithium battery supplies power to the wake-up module of the first bidirectional DC-DC converter, and the super capacitor 6 is in a power-off protection state to avoid leakage current loss.

[0050] In one embodiment, the dynamic impedance matching network includes an impedance detection unit and a switched capacitor array.

[0051] The impedance detection unit is connected with the double rectification filter circuit 5, and is used for real-time sampling of output voltage and output current of the double rectification filter circuit 5, and calculating source impedance according to the collected output voltage and output current; The switched capacitor array includes 8 paths of different capacitance capacitors and corresponding switches, and is used for dynamically controlling the number of opened switches to update the capacitor combination, so that the matching error of the load impedance and the source impedance meets the preset error condition. Specifically, the switch is a MOSFET switch, and the 8 paths of different capacitance capacitors are connected to or disconnected from the circuit through the respective MOSFET switches; specifically, the 8 paths of different capacitance capacitors are 10μF, 22μF, 47μF, 100μF×2, 220μF, 470μF and 1000μF respectively; the MOSFET switch is a SI2302 MOSFET switch, and the equivalent capacitance dynamically adjusts in the range of 10μF-2000μF.

[0052] The switched capacitor array updates the capacitor combination at a timing (every 10ms) to make the matching error of the load impedance and the source impedance less than 3%. Wherein, the capacitor combination can be obtained by using an improved genetic algorithm.

[0053] In a possible embodiment, the switch circuit array updates the capacitor combination, including the following steps: The vibration information is analyzed by fast Fourier transform to obtain a vibration frequency, and a source impedance range is determined according to the vibration frequency, for example, 50 Hz corresponds to a source impedance of 200 Ω, and 20 Hz corresponds to 50 Ω. The matching capacitor value is calculated according to the formula C = 1 / (2πfz), wherein C is the capacitor, f is the vibration frequency, and Z is the source impedance. The MOSFET switch of the corresponding capacitor value circuit is controlled according to the calculated matching capacitor value, and the matching capacitor value is realized, such as 20 Hz / 50 Ω matching 82 μF capacitor.

[0054] It can be understood that the energy loss after matching is reduced by 40%, and the output power is increased from 5 μW to 15 μW under the condition of 20 Hz / 0.5 g vibration.

[0055] In a possible embodiment, the switch capacitor array is less than 10 mm away from the output end of the two-way rectification filter circuit 5, so as to reduce the line inductance (less than 1 nH); and the MOSFET switch of the switch capacitor array has a width of more than 20 mil, so as to reduce the on-resistance (less than 100 mΩ). In addition, a complete ground plane can be laid on the bottom layer of the PCB to suppress the switching noise interference.

[0056] It can be understood that the embodiment adopts a three-level architecture of “two-way independent rectification, two-way DC-DC, and super capacitor 6 / lithium battery hybrid energy storage”, breaks through the one-way conduction limitation of the diode “or gate” through dynamic impedance matching and energy priority scheduling, and simultaneously realizes the improvement of energy harvesting efficiency.

[0057] Further, the effect of the embodiment is tested: Normal operation (vibration frequency 50 Hz / vibration amplitude 0.8 g) scene: piezoelectric output 30 μW power, electromagnetic output 20 μW power, super capacitor 6 charging 0.2 mWh per hour; Night light load (vibration frequency 20 Hz / vibration amplitude 0.4 g) scene: piezoelectric output 5 μW power, electromagnetic output 12 μW power, lithium battery power supply (discharge current 5 μA), and the system maintains operation.

[0058] Under the test condition of vibration amplitude 0.5 g and frequency 20 Hz, the output power of the sensor of the application is 15 μW, and the output power of the single piezoelectric scheme is only 5 μW; when the frequency is 100 Hz, the output power of the application is 80 μW, and the output power of the single piezoelectric scheme is 20 μW.

[0059] In a mixed scene, the average output power is greater than or equal to 40 μW, which meets the system power consumption (25 μW) requirement.

[0060] In addition, the embodiment also comprises a vibration sensor connected with the super capacitor 6 / lithium battery hybrid energy storage unit, for collecting vibration information. Preferably, the vibration sensor adopts a three-axis MEMS acceleration sensor 8ADXL362, supports a ±16g range, and has a standby mode current <500nA; the three-axis MEMS acceleration sensor 8ADXL362 collects vibration information based on a periodicity (wake-up interval 50ms) and a high-resolution mode (0.25mg / LSB), and has a single collection power consumption <5μJ, effectively reducing the power consumption of the sensor.

[0061] In use, the three-axis MEMS acceleration sensor 8ADXL362 is fixed on the support of the energy storage battery pack through a magnetic base, the vibration frequency of this position is mainly 50-150Hz, and the amplitude is about 0.4g. At the same time, the piezoelectric ceramic sheet 1 is closely attached to the surface of the support of the energy storage battery pack, ensuring that high-frequency vibration energy can be effectively captured; the electromagnetic induction coil 2-permanent magnet 33 assembly is suspendedly installed through the elastic support 4, with a distance of 5mm from the surface of the support of the energy storage battery pack, for capturing low-frequency vibration energy.

[0062] Specifically, a signal processing circuit matched with the vibration sensor is also provided, as shown in FIG. 5. Figure 5 The vibration information collected by the vibration sensor is sent into the microprocessor 7 for subsequent processing after being processed by the signal conditioning circuit and A / D conversion, and is transmitted to the upper computer through the wireless transmission module.

[0063] The microcontroller collects the voltage state of the super capacitor 6 and controls the working state of the vibration sensor and the LoRa wireless transmission module 9. Specifically, the microprocessor 7 collects the voltage in real time, wakes up the three-axis MEMS acceleration sensor 8ADXL362 and the LoRa wireless transmission module 9 when the voltage ≥3.3V, and closes the LoRa wireless transmission module 9 when the voltage <2.5V, only leaving the low-power monitoring function of the energy harvesting unit and the MCU; In the high-voltage mode, the three-axis MEMS acceleration sensor 8ADXL362 collects data at a frequency of 10Hz, and the LoRa module transmits data once every 60s; In the low-voltage mode, the sensor enters an intermittent wake-up mode and wakes up once every 500ms to collect data, thereby realizing adaptive switching of the power supply state and optimizing the energy utilization efficiency.

[0064] In addition, when the first bidirectional DC-DC converter detects that the voltage of the super capacitor 6 is >4V, the microprocessor 7 increases the data collection frequency according to the adjustment strategy to obtain accurate device vibration information. When the first bidirectional DC-DC converter detects that the lithium battery voltage < 3.0V, the microprocessor 7 reduces the data acquisition frequency to 10Hz according to the interval adjustment strategy to balance the monitoring accuracy and energy consumption; When the first bidirectional DC-DC converter detects that the lithium battery voltage < 3.0V, the microprocessor 7 prolongs the LoRa transmission interval according to the interval adjustment strategy to reduce transmission power consumption, for example, prolongs the LoRa transmission interval from 60s to 300s, thereby reducing power consumption.

[0065] The LoRa wireless transmission module 9 is used for data transmission. Specifically, the LoRa wireless transmission module 9 adopts SX1278.

[0066] The embodiment further provides an implementation method: the microcontroller wakes up the three-axis MEMS acceleration sensor 8 ADXL362 every 100ms (10Hz) to collect three-axis acceleration data once, and enters sleep after 50ms, and the single collection power consumption is 4μJ; the collection frequency is increased to 20Hz every day during the working period (8:00-20:00, device high load running period), and is reduced to 5Hz during the remaining period.

[0067] When the edge server detects that the 100Hz main frequency amplitude exceeds the threshold value (0.5g) for 3 times in succession and is accompanied by an increase in the 20Hz low-frequency component, it is determined that the transformer core is loose, and an early warning information is pushed through the park management platform, and the response time is <10s.

[0068] Further, a 5-200Hz vibration environment (amplitude 0.2-0.6g) is simulated in the laboratory for continuous operation test, the self-powered vibration sensor continuously works for 7 days, the voltage of the super capacitor 6 is always maintained at 2.8-3.8V, and the data upload success rate is 99.5%.

[0069] The transformer support bolt is artificially loosened (a 60Hz abnormal vibration component is introduced) to perform fault test, and the system accurately identifies within 2 transmission periods (2 minutes), which is 7.5 times higher than the existing single piezoelectric scheme (response time 15 minutes).

[0070] In addition, the embodiment is also integrated with a source network load storage system. Specifically, the microprocessor 7 of the self-powered vibration sensor performs fast Fourier transform on the original vibration data collected by the vibration sensor, extracts characteristic parameters, and transmits them to the park edge server through LoRa, and the vibration parameters include main frequency, vibration amplitude, energy spectrum density, and vibration frequency. The edge server combines the operation parameters of the source network load storage equipment, such as the temperature of the energy storage battery pack and the power grid load data, uses a machine learning algorithm (such as a random forest algorithm) to identify abnormal vibration modes of the equipment, such as bearing wear and support loosening, and outputs an early warning signal to the park management platform.

[0071] In addition, the energy parameters such as the voltage of the super capacitor 6 and the state of charge of the lithium battery can be uploaded through LoRa for the park management platform to adjust the equipment operation mode (such as the switching frequency of the energy storage converter) accordingly, further improving the vibration energy output.

[0072] Specifically, one workflow of the embodiment is as follows: When vibration occurs, the multi-modal vibration energy capturing unit captures vibration energy and charges the hybrid energy storage unit. When the hybrid energy storage unit is charged, the vibration sensor is powered on to start collecting vibration information and sends the collected vibration information and the operation information of the super capacitor 6 and the lithium battery to the source network and storage system.

[0073] It is judged whether the number of sent data reaches a preset number, if yes, it enters a low-power-consumption operation and judges whether the accumulated sleep time reaches a preset time such as 50 ms, if yes, it exits the sleep mode and collects data again.

[0074] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application but not to limit it; although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that: the specific embodiments of the present application can be modified or some technical features can be replaced by equivalent ones; without departing from the spirit of the technical solutions of the present application, they should be covered in the technical solution range of the present application.

Claims

1. A self-powered vibration sensor based on multimodal energy harvesting, characterized in that, include: A multimodal vibration energy harvesting unit includes a piezoelectric ceramic sheet and an electromagnetic induction coil-permanent magnet assembly mounted in parallel on the surface of the device. The piezoelectric ceramic sheet is used to harvest high-frequency vibration energy of 100-200Hz, and the electromagnetic induction coil-permanent magnet assembly is used to harvest low-frequency vibration energy of 5-50Hz. The energy harvesting power management circuit includes a dual-channel rectifier and filter circuit, a first bidirectional DC-DC converter, and a supercapacitor / lithium battery hybrid energy storage unit. The dual-channel rectifier and filter circuit rectifies and filters the voltage signal output from the piezoelectric ceramic and the voltage signal output from the electromagnetic induction coil-permanent magnet assembly, respectively. The first bidirectional DC-DC converter is connected to both the dual-channel rectifier and filter circuit and the supercapacitor / lithium battery hybrid energy storage unit, and is used to control the charging and discharging of the supercapacitor / lithium battery hybrid energy storage unit based on a preset dynamic impedance matching network and energy priority scheduling strategy. A vibration sensor, connected to the supercapacitor / lithium battery hybrid energy storage unit, is used to collect vibration information; LoRa wireless transmission module, used for data transmission; The microcontroller is used to collect the voltage status of the supercapacitor and control the operation of the vibration sensor and LoRa wireless transmission module based on the voltage status.

2. The self-powered vibration sensor based on multimodal energy harvesting according to claim 1, characterized in that, The dynamic impedance matching network includes an impedance detection unit and a switched capacitor array; The impedance detection unit is connected to the dual-channel rectifier and filter circuit, and is used to sample the output voltage and output current of the dual-channel rectifier and filter circuit in real time, and calculate the source impedance based on the sampled output voltage and output current. The switched capacitor array includes eight capacitors with different capacitance values ​​and corresponding switches, which are used to dynamically control the number of switches turned on to update the capacitor combination so that the load impedance and source impedance matching error meet the preset error conditions.

3. A self-powered vibration sensor based on multimodal energy harvesting according to claim 2, characterized in that, The switching circuit array dynamically controls the number of switches that are turned on, including: The vibration frequency is obtained by analyzing the vibration information through Fast Fourier Transform, and the source impedance range is determined based on the vibration frequency. The matching capacitance value is calculated using the formula C=1 / (2πfz), where C is the capacitance, f is the vibration frequency, and Z is the source impedance. The switching action of the corresponding capacitance circuit is controlled based on the calculated matching capacitance value to achieve the matching capacitance value.

4. A self-powered vibration sensor based on multimodal energy harvesting according to claim 2 or 3, characterized in that, The distance between the switched capacitor array and the output terminal of the dual-channel rectifier filter circuit is <10mm; and the width of the switching trace of the switched capacitor array is ≥20mil.

5. A self-powered vibration sensor based on multimodal energy harvesting according to claim 1, 2, or 3, characterized in that, Energy priority scheduling strategies include: Vibration stage: After the piezoelectric ceramic and the electromagnetic induction coil-permanent magnet assembly are rectified independently, they are directly charged to the supercapacitor through the first bidirectional DC-DC converter. At the same time, the supercapacitor voltage is detected. When the supercapacitor voltage is >4.5V, the supercapacitor charges the lithium battery in trickle mode through the second DC-DC conversion circuit. Meanwhile, by default, a supercapacitor is used as the power source to power the vibration sensor, LoRa wireless transmission module, and microprocessor. When the supercapacitor voltage is <3.3V, the lithium battery is switched to power the vibration sensor, LoRa wireless transmission module, and microprocessor. During the discharge process of the lithium battery, it is still charged in reverse through a dual-circuit rectifier and filter circuit, a first bidirectional DC-DC converter, a supercapacitor, and a second DC-DC conversion circuit. Vibration-free sleep phase: The first bidirectional DC-DC converter enters low-power standby mode, with only the lithium battery powering the wake-up module of the first bidirectional DC-DC converter, and the supercapacitor is in a power-off protection state to avoid leakage current loss.

6. A self-powered vibration sensor based on multimodal energy harvesting according to claim 1, 2, or 3, characterized in that, The vibration sensor uses a triaxial MEMS accelerometer ADXL362, which acquires vibration information based on a periodic, high-resolution mode.

7. A self-powered vibration sensor based on multimodal energy harvesting according to claim 6, characterized in that, The microprocessor performs a fast Fourier transform on the vibration information collected by the vibration sensor, extracts feature parameters, and transmits them to the park edge server via a LoRa wireless transmission module. The vibration parameters include the dominant frequency, vibration amplitude, energy spectral density, and vibration frequency.

8. A self-powered vibration sensor based on multimodal energy harvesting according to claim 5, characterized in that, The microprocessor collects the power supply voltage in real time. When the power supply voltage is ≥3.3V, it wakes up the vibration sensor and LoRa wireless transmission module. The vibration sensor collects data at a frequency of 10Hz, and the LoRa wireless transmission module transmits data once every 60s. When the power supply voltage is <2.5V, the LoRa wireless transmission module is turned off, and the vibration sensor enters an intermittent wake-up mode, waking up once every 500ms to collect vibration information, while retaining the low-power monitoring function of the multimodal vibration energy capture unit and the microprocessor.

9. A self-powered vibration sensor based on multimodal energy harvesting according to claim 8, characterized in that, When the power supply voltage is greater than 4V, the microprocessor increases the data acquisition frequency of the vibration sensor to 100Hz in order to obtain accurate equipment vibration information. When 3.3V ≤ power supply voltage < 4V, the microprocessor reduces the data acquisition frequency of the vibration sensor to 10Hz to balance monitoring accuracy and energy consumption. When 2.5V ≤ power supply voltage < 3.3V, the LoRa transmission interval is extended from 60s to 300s.

Images