Low-power-consumption vibration monitoring system and method based on multi-stage triggering and frequency conversion sampling

Abstract

The invention discloses a low-power-consumption vibration monitoring system and method based on multi-stage triggering and frequency conversion sampling. The low-power-consumption vibration monitoring system comprises a hardware assembly, a logic state and key control parameters. The hardware assembly comprises an MEMS sensor, a logic control unit, an annular buffer area and a main controller, the logic states comprise a standby state, an early warning state and a trigger state, and the key control parameters comprise an early warning threshold value Th1, a trigger threshold value Th2, sampling frequency and a time window. The scheme provides a monitoring method which not only can maintain microampere (muA) level ultra-low standby power consumption, but also can completely record the whole process (including high-frequency characteristics before impact) of the impact event.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent sensing and signal processing, in particular to a low-power vibration monitoring system and method based on multi-stage triggering and variable frequency sampling. BACKGROUND

[0002] As a core means of equipment health assessment and unexpected accident capture, vibration monitoring has expanded from simple pedometers to complex fields such as logistics asset tracking, precision manufacturing and semiconductors, infrastructure monitoring, etc. with the development of Industry 4.0 and Internet of Things, and needs to meet the vibration data acquisition requirements in different scenarios. The existing vibration monitoring technology faces the core contradiction that high-fidelity data capture and ultra-low power / low data volume are difficult to balance. Specifically, the "sleep-wake" mechanism leads to the loss of key vibration characteristics, full-time high-frequency sampling causes energy consumption and bandwidth bottlenecks, and the traditional single threshold triggering method is prone to false triggering or missing reports in complex industrial noise environments, which cannot meet the industry's demand for low power consumption and complete capture of the whole process of sudden events.

[0003] Therefore, a solution is needed. SUMMARY

[0004] (I) Technical problems solved In view of the deficiencies in the prior art, the present application provides a low-power vibration monitoring system and method based on multi-stage triggering and variable frequency sampling to solve the problems raised in the background art.

[0005] (II) Technical solutions

[0006] To achieve the above purpose, the present application is realized by the following technical solutions: A low-power vibration monitoring system and method based on multi-stage triggering and variable frequency sampling, comprising hardware components, logic states and key control parameters. The hardware components include a MEMS sensor, a logic control unit, a ring buffer and a main controller, the logic states include a standby state, a pre-warning state and a trigger state, the key control parameters include a pre-warning threshold Th1, a trigger threshold Th2, a sampling frequency and a time window; the ring buffer is a FIFO, the main controller is an MCU, the standby state is state 0, the pre-warning state is state 1, the trigger state is state 2, the time window is Tw, the logic control unit is a core control subject, the sampling frequency is divided into two kinds of a first sampling frequency F1 and a second sampling frequency F2, according to the pre-warning threshold Th1 and the trigger threshold Th2 and the time window, the sampling frequency is dynamically switched between F1 (low frequency) and F2 (high frequency), and then the system is driven to flow among the standby state, the pre-warning state and the trigger state, only when the trigger condition of “Th2+Tw” is met, an interrupt is sent to wake up the sleeping MCU, and the whole process does not need the MCU to intervene in the basic control, so that data integrity is ensured and ultra-low power consumption is realized.

[0007] Preferably, S1: in the standby mode, the MEMS sensor is controlled to collect vibration data at the first sampling frequency F1, and the data is cyclically written into the ring buffer (FIFO); S2: the amplitude of the vibration data is monitored in real time, and when the amplitude exceeds the pre-warning threshold Th1, the sampling frequency of the MEMS sensor is switched to the second sampling frequency F2, wherein F2>F1; S3: after being switched to the second sampling frequency F2, the high-frequency vibration data is continuously written into the ring buffer (FIFO); S4: it is judged whether the amplitude of the vibration data exceeds the trigger threshold Th2 within a preset time window, wherein Th2>Th1; S5: if the trigger threshold Th2 is exceeded, a trigger signal is generated to wake up the main controller (MCU), and the main controller reads the historical vibration data containing the time before the trigger from the ring buffer (FIFO); S6: if the trigger threshold Th2 is not exceeded within the time window, the sampling frequency is restored to the first sampling frequency F1, and the main controller (MCU) is not woken up Preferably, the size of the ring buffer (FIFO) is configured to be at least capable of storing the second sampling frequency F2 data for T_hist duration, so as to ensure that the data read by the main controller (MCU) contains complete rising edges of vibration events.

[0008] Preferably, in step S5, the data sequence read by the main controller (MCU) contains two parts: the first part is the background vibration data collected at the first sampling frequency F1, and the second part is the event vibration data collected at the second sampling frequency F2. The main controller (MCU) normalizes the time axis of the two parts of data when processing according to the time stamp or data index.

[0009] Preferably, the MEMS sensor is a MEMS accelerometer with programmable logic or a state machine, and the frequency switching logic in steps S2 and S6 is completely executed inside the MEMS sensor without the intervention of the main controller (MCU).

[0010] Preferably, the MEMS sensor is internally integrated with a FIFO memory and a logic control unit; and the main controller (MCU) is only woken up from a sleep state after receiving a trigger signal sent by the MEMS sensor.

[0011] Preferably, the logic control unit is configured to perform the variable frequency sampling and threshold judgment logic as claimed in any one of claims 2 to 5.

[0012] (Three) beneficial effects The present application provides a low-power consumption vibration monitoring system and method based on multi-level triggering and variable frequency sampling. The following beneficial effects are provided: 1. Ultra-low power consumption: standby low-frequency sampling + pre-warning / alarm stage does not wake up the MCU, only true triggering wakes up; the frequency switching logic is executed inside the sensor without the intervention of the MCU, greatly reducing the system energy consumption.

[0013] 2. Data integrity: pre-warning stage increases the sampling rate and retains the vibration "pre-oscillation" data, FIFO stores low-frequency background + high-frequency event data, fully covers the whole process of vibration event, and avoids the loss of vibration characteristics.

[0014] 3. Strong anti-interference: multi-level threshold logic of "Th1 pre-warning + Th2+Tw triggering" is adopted to reduce false alarms and missed reports in complex industrial environments.

[0015] 4. Data efficiency: variable frequency sampling replaces full-time high-frequency sampling, reduces the amount of invalid data, and relieves the transmission bandwidth bottleneck. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is the system diagram of the present application; Figure 2 is the system state machine transition diagram of the present application, showing the jump conditions of the standby, pre-warning, and triggering states; Figure 3The sampling frequency and acceleration waveform timing diagram of the application. The horizontal axis is time, and the vertical axis is the acceleration amplitude. The frequency mutation at the initial stage of the waveform rising and the effect that the MCU wakes up with a lag but the data is not lost are clearly shown. Figure 4 The FIFO data structure diagram of the application. The storage structure of mixed low-frequency historical data and high-frequency real-time data is shown.

[0017] The figure includes hardware components; MEMS sensors; logic control units; ring buffers; main controllers; logic states; standby states; early warning states; trigger states; key control parameters; early warning thresholds Th1; trigger thresholds Th2; sampling frequencies; time windows. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the application.

[0019] Please refer to Figures 1-4 The embodiments of the application provide a technical solution: including hardware components, logic states and key control parameters.

[0020] The hardware components include MEMS sensors, logic control units, ring buffers and main controllers, the logic states include standby states, early warning states and trigger states, and the key control parameters include early warning thresholds Th1, trigger thresholds Th2, sampling frequencies and time windows; the ring buffer is FIFO, the main controller is MCU, the standby state is state 0, the early warning state is state 1, the trigger state is state 2, the time window is Tw, the logic control unit is the core control subject, the sampling frequency is divided into two kinds of first sampling frequency F1 and second sampling frequency F2, and the sampling frequency is dynamically switched between F1 (low frequency) and F2 (high frequency) according to the early warning threshold Th1, the trigger threshold Th2 and the time window, so as to drive the system to flow in the standby state, the early warning state and the trigger state. Only when the trigger condition of "Th2+Tw" is met, the MCU in sleep mode is sent an interrupt to wake up. The whole process does not need the intervention of the MCU in basic control, which guarantees data integrity and realizes ultra-low power consumption.

[0021] In detail, S1: in the standby mode, the MEMS sensor is controlled to collect vibration data at the first sampling frequency F1, and the data is cyclically written into the ring buffer (FIFO); S2: Real-time monitoring of the amplitude of the vibration data, when the amplitude exceeds the pre-warning threshold Th1, the sampling frequency of the MEMS sensor is switched to the second sampling frequency F2, where F2>F1; S3: After switching to the second sampling frequency F2, continue to write high-frequency vibration data into the ring buffer (FIFO); S4: Determine whether the amplitude of the vibration data exceeds the trigger threshold Th2 within a preset time window, where Th2>Th1; S5: If the trigger threshold Th2 is exceeded, a trigger signal is generated to wake up the main controller (MCU), and the main controller reads the historical vibration data containing the trigger time from the ring buffer (FIFO); S6: If the trigger threshold Th2 is not exceeded within the time window, the sampling frequency is restored to the first sampling frequency F1, and the main controller (MCU) is not awakened.

[0022] The size of the ring buffer (FIFO) is configured to be at least capable of storing T_hist duration of second sampling frequency F2 data, ensuring that the data read by the main controller (MCU) contains complete vibration event rising edge.

[0023] In step S5, the data sequence read by the main controller (MCU) contains two parts: the first part is the background vibration data collected at the first sampling frequency F1, and the second part is the event vibration data collected at the second sampling frequency F2. The main controller (MCU) normalizes the time axis reconstruction of the two parts of data according to the time stamp or data index when processing.

[0024] The MEMS sensor is a MEMS accelerometer with programmable logic or state machine, and the frequency switching logic in steps S2 and S6 is completely executed inside the MEMS sensor without the need for intervention of the main controller (MCU).

[0025] The MEMS sensor is internally integrated with a FIFO memory and a logic control unit; the main controller (MCU) is only awakened from the sleep state after receiving the trigger signal sent by the MEMS sensor.

[0026] The logic control unit is configured to perform the variable frequency sampling and threshold judgment logic as claimed in any one of claims 2 to 5.

[0027] Data comparison Power consumption comparison: Traditional scheme (high frequency all the time): power consumption P_high; Traditional scheme (low frequency + wake up): power consumption P_low (but loses the first T_loss data); This solution features standby power consumption P_opt (only briefly increasing frequency during micro-vibrations) and complete waveform capture. (Note: Actual measurement data shows that P_opt is much smaller than P_high and slightly larger than P_low, achieving the optimal balance of energy efficiency.)

[0028] Waveform completeness: The comparison chart shows that this solution can restore the initial micro-crack vibration (high frequency, low amplitude) at the moment of impact, while the traditional low-frequency wake-up solution completely loses this information.

[0029] Solution Analysis: 1. Extremely low power consumption, suitable for long-term monitoring scenarios: In standby mode, it samples at a low frequency F1, and the core power consumption comes only from the MEMS sensor; in the early warning stage, it only switches to a high frequency F2 through the internal logic of the sensor, without waking up the MCU; the MCU interrupt is only triggered when a real impact meets the "Th2+Tw" condition, avoiding the energy waste of full-time high-frequency sampling, reducing the loss of invalid MCU wake-up, and greatly extending the device's battery life.

[0030] 2. Complete and intact data, capturing the entire event chain: After the frequency is increased during the early warning stage, the FIFO continuously stores high-frequency data, preserving the details of the "pre-shaking" of the vibration event; the data read by the MCU after triggering includes low-frequency background, high-frequency rising edge and peak data, completely covering the entire event process, solving the pain point of the traditional "sleep-wake" mechanism losing the oscillation characteristics.

[0031] 3. Strong anti-interference capability, reducing false alarm and missed alarm rates: It adopts a multi-level threshold logic of "Th1 warning + Th2 + Tw trigger" instead of a single threshold. Th1 filters potential events, and Th2 + time window verifies valid events, which can accurately distinguish between noise disturbances and real vibration events in the industrial environment, effectively reducing false triggers and missed alarms.

[0032] 4. Highly efficient data transmission, alleviating bandwidth pressure: By dynamically adjusting the data volume through frequency conversion sampling, high-frequency data is generated only when necessary, avoiding the massive amounts of invalid data caused by constant high-frequency sampling. This reduces data transmission and storage pressure, alleviates bandwidth bottlenecks, and adapts to the data transmission needs of various scenarios such as logistics tracking and infrastructure monitoring.

[0033] Working principle: 1. Standby phase: The sensor samples at a low frequency F1 and writes the data to the FIFO. The system power consumption is extremely low during this period.

[0034] 2. Warning Phase (Frequency Variable): When the detected change in acceleration exceeds the small "warning threshold" Th1 (indicating that an event may have occurred, but it is uncertain whether it is a strong impact), the system immediately increases the sampling rate to F2.

[0035] Innovation: The main controller MCU is not woken up at this time; the sampling rate is adjusted only inside the sensor or in the low-power front end.

[0036] High-frequency data begins to rapidly fill the FIFO, thus preserving the high-frequency details of the "pre-impact" in the FIFO.

[0037] 3. Trigger / Retreat Phase: Case A (True Impact): If the acceleration continues to rise and exceeds the "trigger threshold" Th2 within the timeout window Tw, it is considered a valid impact.

[0038] Lock the FIFO and stop overwriting (or mark the current pointer).

[0039] Send an interrupt to wake up the main controller MCU.

[0040] The MCU reads the contents of the FIFO (at this time, the FIFO contains: the historical low-frequency background + the high-frequency rising edge after frequency conversion + the high-frequency peak after triggering).

[0041] Case B (False Alarm / Micro-shock): If the time does not exceed Th2 within the Tw time period, it is determined to be an environmental disturbance.

[0042] The sensor automatically reduces the sampling rate back to F1.

[0043] It does not wake up the MCU, thus saving a lot of power consumption.

[0044] Technical effects of implementing this solution: This solution employs an innovative design of "multi-level triggering + variable frequency sampling," achieving extremely low power consumption through "low-frequency standby + front-end variable frequency + on-demand MCU wake-up." It captures the entire event chain data by retaining vibration "pre-warning" data and storing it in a complete FIFO, thus preventing the loss of vibration characteristics. Multi-level threshold and time window verification logic enhances anti-interference capabilities and reduces false alarms and missed alarms in complex environments. Simultaneously, variable frequency sampling reduces invalid data volume and alleviates transmission and storage bandwidth pressure. This solution precisely addresses the core contradiction of traditional technologies—the difficulty of balancing high-fidelity data with low power consumption / low data volume—and is fully adaptable to the industrial vibration monitoring needs of various scenarios, including logistics asset tracking, precision manufacturing, and infrastructure monitoring.

[0045] The hardware components of this invention include: a MEMS sensor; a logic control unit; a ring buffer; a main controller; logic states; a standby state; an early warning state; a trigger state; key control parameters; an early warning threshold Th1; a trigger threshold Th2; a sampling frequency; and a time window. These components are all general-purpose standard parts or components known to those skilled in the art. Their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods. The problem addressed by this invention is the core contradiction faced by existing vibration monitoring technologies: the difficulty in simultaneously achieving high-fidelity data capture and ultra-low power consumption / low data volume. Specifically, the "sleep-wake" mechanism leads to the loss of key vibration start-up characteristics, and full-time high-frequency sampling causes energy consumption and bandwidth bottlenecks. Furthermore, traditional single-threshold triggering methods are prone to false triggering or missed alarms in complex industrial noise environments, failing to meet the industry's demand for low power consumption and complete capture of the entire process of sudden events. This invention provides a monitoring method that can maintain ultra-low standby power consumption at the microampere (μA) level while completely recording the entire process of an impact event (including high-frequency characteristics before the impact).

[0046] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0047] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A low-power vibration monitoring system based on multi-level triggering and frequency conversion sampling, characterized in that: This includes hardware components, logic states, and key control parameters; The hardware components include MEMS sensors, a logic control unit, a circular buffer, and a main controller. The logic states include standby, warning, and trigger states. The key control parameters include a warning threshold Th1, a trigger threshold Th2, a sampling frequency, and a time window. The circular buffer is a FIFO, the main controller is an MCU, the standby state is state 0, the warning state is state 1, the trigger state is state 2, and the time window is Tw. The logic control unit serves as the core control entity. The sampling frequency is divided into two types: a first sampling frequency F1 and a second sampling frequency F2. Based on the warning threshold Th1, the trigger threshold Th2, and the time window, the sampling frequency is dynamically switched between F1 (low frequency) and F2 (high frequency), thereby driving the system to flow between the standby, warning, and trigger states. Only when the trigger condition "Th2+Tw" is met is an interrupt sent to wake up the dormant MCU. The entire process requires no MCU intervention in basic control, ensuring data integrity while achieving ultra-low power consumption.

2. A monitoring method for a low-power vibration monitoring system based on multi-level triggering and frequency conversion sampling as described in claim 1, characterized in that, The monitoring method includes the following steps: S1: In standby mode, control the MEMS sensor to collect vibration data at the first sampling frequency F1 and write the data cyclically into the circular buffer (FIFO). S2: Monitor the amplitude of vibration data in real time. When the amplitude exceeds the warning threshold Th1, switch the sampling frequency of the MEMS sensor to the second sampling frequency F2, where F2>F1. S3: After switching to the second sampling frequency F2, continue to write high-frequency vibration data into the circular buffer (FIFO). S4: Determine whether the amplitude of the vibration data exceeds the trigger threshold Th2 within the preset time window, where Th2>Th1; S5: If the trigger threshold Th2 is exceeded, a trigger signal is generated to wake up the main controller (MCU), and the main controller reads historical vibration data containing the data prior to the trigger moment from the circular buffer (FIFO); S6: If the trigger threshold Th2 is not exceeded within the time window, the sampling frequency is restored to the first sampling frequency F1, and the main controller (MCU) is not woken up.

3. The low-power vibration monitoring method based on multi-level triggering and frequency conversion sampling according to claim 2, characterized in that: The size of the circular buffer (FIFO) is configured to store at least T_hist duration of second sampling frequency F2 data, ensuring that the data read by the main controller (MCU) includes the complete rising edge of the vibration event.

4. The low-power vibration monitoring method based on multi-level triggering and frequency conversion sampling according to claim 3, characterized in that: In step S5, the data sequence read by the main controller (MCU) includes two parts: the first part is the background vibration data collected at the first sampling frequency F1, and the second part is the event vibration data collected at the second sampling frequency F2. When processing, the main controller (MCU) reconstructs the time axis of the two parts of data according to the timestamp or data index.

5. The low-power vibration monitoring method based on multi-level triggering and frequency conversion sampling according to claim 4, characterized in that: The MEMS sensor is a MEMS accelerometer with programmable logic or a state machine. The frequency switching logic in steps S2 and S6 is executed entirely within the MEMS sensor without the intervention of the main controller (MCU).

6. The low-power vibration monitoring system based on multi-level triggering and frequency conversion sampling according to claim 1, characterized in that: The MEMS sensor integrates a FIFO memory and a logic control unit; the main controller (MCU) only wakes up from the sleep state after receiving a trigger signal from the MEMS sensor.

7. The low-power vibration monitoring method based on multi-level triggering and frequency conversion sampling according to claim 5, characterized in that: The logic control unit is configured to perform the frequency conversion sampling and threshold determination logic as described in any one of claims 2 to 5.

Images