A self-cleaning system for a focused ultrasonic transducer

Abstract

The application relates to the technical field of ultrasonic wind measurement equipment, and discloses a self-cleaning system for a transmitting ultrasonic transducer, which comprises a protective cover module, a mechanical driving module and an intelligent control module; the mechanical driving module comprises a direct-current motor and a wiper linkage structure, and a time, data acquisition and data processing module is designed in the intelligent control module; the system records continuous operation time through a timing module, calculates fluctuation variance of instantaneous wind speed data by using a data processing module, and extracts mean value drop difference and fluctuation standard deviation of analog-digital conversion sampling values; when the operation time reaches a preset period or the above data features are greater than corresponding set thresholds or limits, the intelligent control module generates a cleaning instruction to control the direct-current motor to drive the wiper linkage to perform physical wiping on the surface of the transducer. According to the application, quantitative on-demand active cleaning is realized through multi-source data features, dust interference on ultrasonic signals is eliminated, and the measurement accuracy is ensured.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic wind measurement equipment technology, specifically a self-cleaning system for a through-beam ultrasonic transducer. Background Technology

[0002] Through-beam ultrasonic anemometers measure wind speed based on the ultrasonic time-of-flight method and are widely used in environments such as underground ventilation systems in coal mines. The measurement accuracy of this type of sensor depends on the stable transmission of the ultrasonic signal between transducers. In actual working conditions, the airflow in roadways carries dust, and dust particles easily adhere to the surface of the ultrasonic transducers. Dust accumulation absorbs and attenuates ultrasonic energy, leading to a decrease in the signal sampling value at the receiving end and abnormal fluctuations in instantaneous wind speed data, thus increasing wind speed measurement errors or causing monitoring failure.

[0003] Existing dust prevention and maintenance methods have structural and logical limitations. Due to the physical characteristics of ultrasonic transmission, adding a solid protective lens to the outside of the transducer would directly block the sound wave propagation path, forcing the transducer to remain exposed. Traditional maintenance mainly relies on manual periodic wiping from elevated positions. Considering the high installation location of underground sensors, manual operation poses safety hazards and is difficult to handle random dust accumulation. If a single, fixed-cycle automatic mechanical cleaning device is used, the fixed action interval cannot match the actual degree of dust accumulation. When the dust concentration in the environment increases sharply, the cleaning action often lags behind the dust accumulation process, causing inaccurate measurement results during this period; while in relatively clean environments, continuous timed mechanical actions will cause ineffective wear and tear on the motor and cleaning components. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a self-cleaning system for through-beam ultrasonic transducers. This system solves the problems of existing ultrasonic wind speed sensors being prone to signal attenuation and measurement distortion due to dust accumulation in dusty environments, the difficulty of traditional manual maintenance, and the inability of single, fixed-cycle mechanical cleaning methods to achieve on-demand cleaning based on actual dust accumulation.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a self-cleaning system for a through-beam ultrasonic transducer, comprising a mechanical drive module and an intelligent control module.

[0006] The mechanical drive module includes a DC motor and a wiper linkage structure. The DC motor drives the wiper linkage structure to clean the ultrasonic transducer of the through-beam ultrasonic anemometer. The intelligent control module is connected to the DC motor and internally includes a timing module, a data acquisition module, and a data processing module.

[0007] This system constructs a parallel monitoring and triggering mechanism based on multi-source signal characteristics. Dust accumulation on the surface of the ultrasonic transducer physically interferes with the ultrasonic signal, causing data fluctuations in both instantaneous wind speed data and underlying sampled data. The system extracts these data characteristics to execute cleaning decisions.

[0008] The timing module records the continuous running time and generates a cleaning instruction when the continuous running time reaches the preset cleaning interval.

[0009] The data acquisition module acquires instantaneous wind speed data and analog-to-digital conversion sampled values ​​from the ultrasonic transducer receiver. The data processing module calculates the variance of the instantaneous wind speed data. This calculation involves performing a sliding window filter on the instantaneous wind speed data and extracting a continuous sequence of instantaneous wind speed samples within a set time window. The variance is then calculated by determining the dispersion of each instantaneous wind speed sample point within the sequence relative to the mean instantaneous wind speed within the set time window. The data processing module generates a cleaning command when the variance exceeds a calibrated dynamic variance threshold. This calibrated dynamic variance threshold is obtained offline based on baseline normal turbulent flow characteristics under dust-free conditions.

[0010] Dust adherence absorbs and attenuates ultrasonic energy. The data processing module extracts the state characteristics of the analog-to-digital conversion (ADC) sampled values ​​within a set time window. When the data processing module determines that the mean decrease of the ADC sampled values ​​relative to the calibration reference value within the set time window exceeds a set mean decrease threshold, or that the standard deviation of the ADC sampled values ​​within the set time window exceeds a set standard deviation threshold, a cleaning instruction is generated. The calibration reference value is the reference value calibrated in a dust-free, clean air environment; the set mean decrease threshold is calculated based on the maximum energy attenuation rate allowed by the sensor's measurement accuracy; and the set standard deviation threshold is based on the judgment limit for signal jitter caused by high-frequency environmental noise.

[0011] The intelligent control module receives the cleaning command and sends a drive command to the DC motor. After the current cleaning action is completed, the intelligent control module resets its internal timing module and simultaneously resets the data cache within all set time windows in the data processing module to its initial state.

[0012] The wiper linkage structure includes a brush head connecting arm and a brush. A DC motor drives the brush head connecting arm and the brush to reciprocate and perform the wiping and cleaning action. A spring is installed at the connection between the brush head connecting arm and the brush, which pulls the brush inward to fit it against the functional surface of the ultrasonic transducer.

[0013] To prevent obstruction of the ultrasonic transmission path, a mechanical drive module is located at the probe of the through-beam ultrasonic anemometer, comprising a motor housing and couplings. A DC motor is mounted inside the motor housing, with its output shaft extending out and sequentially connected to the external brush head connecting arm via couplings. A positioning block is located at the upper part of the motor housing, its installation position radially flush with the outer diameter of the ultrasonic transducer. The positioning block physically obstructs the brush's range of motion in the axial direction. During non-cleaning periods, the brush remains at the limit position where the positioning block is located.

[0014] The system also includes a transducer bracket and an L-shaped fixing plate. The transducer bracket has pre-drilled mounting holes on its plate surface that match the outer diameter of the ultrasonic transducer. The main housing of the ultrasonic transducer passes through these mounting holes and is secured by pre-drilled sealing rings and retaining rings. The lower end of the transducer bracket is fitted with the L-shaped fixing plate. The vertical surface of the L-shaped fixing plate is mechanically connected to the transducer bracket via pre-drilled bolt holes and fastening bolts. The horizontal surface of the L-shaped fixing plate has elongated holes for assembly and fixation with the motor housing.

[0015] The system also includes a protective cover module that covers the outside of the through-beam ultrasonic anemometer. The main structure of the protective cover module is a stainless steel protective cover. The protective cover is composed of a split-type assembled shell and movable connecting parts to form an openable structure. The relative opening and closing operation between the shells is realized through the movable connecting parts, which can both physically cover and isolate dust and facilitate the maintenance of internal equipment.

[0016] The intelligent control module is integrated inside the through-beam ultrasonic anemometer. The through-beam ultrasonic anemometer sends drive commands directly to the external DC motor via a pre-set aerial cable, and simultaneously uses the aerial cable to provide the electrical energy required for the DC motor to operate.

[0017] This invention provides a self-cleaning system for through-beam ultrasonic transducers. It has the following beneficial effects: 1. This invention generates cleaning instructions by extracting the fluctuation variance of instantaneous wind speed data and the mean descent difference and fluctuation standard deviation of analog-to-digital conversion sampled values, and comparing them in parallel with the set dynamic variance threshold and mean descent threshold. This data processing logic quantifies the attenuation and distortion of ultrasonic signals caused by dust obstruction, transforms the system's operation mode into on-demand proactive cleaning, reduces the ineffective wear of mechanical parts caused by fixed-cycle cleaning, and avoids wind speed measurement errors caused by untimely cleaning.

[0018] 2. The present invention provides a positioning block in the upper part of the motor box that is radially flush with the outer diameter of the ultrasonic transducer. The positioning block physically blocks the range of motion of the brush in the axial direction. The mechanical limiting structure ensures that the brush can stay stably in the limited position during non-cleaning periods, preventing the brush from staying on the functional surface of the ultrasonic transducer and eliminating the physical obstruction of the linear transmission path of the sound wave by the cleaning mechanism itself.

[0019] 3. This invention integrates the intelligent control module into the wind speed and direction sensor, directly sending drive commands and providing operating power to the external DC motor via a connecting cable, eliminating reliance on a separate power supply. Combined with the inward tension of the brush by the mechanical spring and the assembly adjustment allowance provided by the elongated holes on the L-shaped fixing plate, the wiring and alignment installation process of the wind measuring device in complex tunnel environments is simplified. Attached Figure Description

[0020] Figure 1 This is a three-dimensional structural diagram of the present invention; Figure 2 This is a schematic diagram of the system architecture of the present invention; Figure 3 This is a flowchart of the method of the present invention; Figure 4 This is a flowchart illustrating the working logic of a self-cleaning system for a through-beam ultrasonic transducer provided in an embodiment of the present invention. Figure 5 The graph shows the variation of wind speed measurement error over time under the three cleaning systems of this invention.

[0021] The components include: 1. Brush; 2. Brush head connecting arm; 3. Spring; 4. Coupling; 5. Motor box; 6. DC motor; 7. Ultrasonic transducer; 8. Transducer bracket; 9. Positioning block; 10. L-shaped fixing plate; 11. Fixing ring; 12. Fastening bolt. Detailed Implementation

[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Please see the appendix Figure 1 and attached Figure 2 This invention provides a self-cleaning system for a through-beam ultrasonic transducer, which may include: a protective cover module, a mechanical drive module, and an intelligent control module.

[0024] The protective cover module is an openable stainless steel protective cover that covers the outside of the through-beam anemometer. The protective cover module isolates and protects against dust carried in the airflow of the tunnel.

[0025] The mechanical drive module is located at the sensor probe. The mechanical drive module includes a wiper linkage structure and a DC motor 6. This DC motor 6 is a ZH3.6 / 18 intrinsically safe DC motor for mining, having obtained safety certification, to meet the safety and explosion-proof requirements of underground coal mines. The wiper linkage structure includes a brush head connecting arm 2 and a brush 1. The operation of the DC motor 6 drives the brush head connecting arm 2 and the brush 1 to reciprocate, performing the wiping and cleaning action.

[0026] The intelligent control module is integrated into the mine wind speed and direction sensor (e.g., the GFC25XD mine wind speed and direction sensor). The mine wind speed and direction sensor supplies power to the DC motor 6 in the mechanical drive module and transmits control commands via a connecting cable. The through-beam ultrasonic transducer self-cleaning system is linked to and shares power with the mine wind speed and direction sensor, eliminating the need for additional independent power supply equipment.

[0027] Please see the appendix Figure 3 and attached Figure 4 This invention provides a self-cleaning method for a through-beam ultrasonic transducer, comprising the following steps: S10, The system powers on and performs initialization operations; S20, the timing module runs and records the system running time, and generates a cleaning instruction when the recorded time reaches the preset cleaning interval. S30, the data acquisition module acquires the instantaneous wind speed data from the wind speed sensor and the analog-to-digital conversion sampling value from the receiver of the ultrasonic transducer 7 in real time; S40, the data processing module performs sliding window filtering and variance calculation on the instantaneous wind speed data, and generates a cleaning command when the fluctuation variance of the instantaneous wind speed data is greater than the set dynamic threshold. S50, the data processing module calculates the analog-to-digital conversion sampled values ​​and generates a cleaning instruction when the mean of the analog-to-digital conversion sampled values ​​decreases by more than a set threshold or the standard deviation of the fluctuation exceeds a set threshold. S60, the intelligent control module receives the cleaning command generated in any one of steps S20, S40 or S50 and sends a drive command to the DC motor 6. S70, DC motor 6 receives drive command and starts, and drives brush 1 to perform physical cleaning action on the surface of ultrasonic transducer 7 through coupling 4. S80, after the cleaning action is completed, the intelligent control module will reset the timing module, reset the monitoring data cache and return to step S20.

[0028] The self-cleaning system for the through-beam ultrasonic transducer is equipped with a protective cover module. In this embodiment, the protective cover module completely covers the outside of the through-beam ultrasonic anemometer. The main structure of the protective cover module is a stainless steel protective cover.

[0029] The stainless steel protective cover is designed to be openable. The technical purpose of this design is to maintain overall external protection while also accommodating the daily maintenance needs of the internal equipment. In actual engineering assembly, this openable structure is specifically manifested as a combination of a modular housing and movable connecting components, which enable relative opening and closing operations between the housings. Regarding the specific structural form of the movable connecting components, such as the use of metal hinges or latch-type locking devices, those skilled in the art can perform conventional mechanical design and configuration based on the actual installation space and operational convenience requirements of the downhole ventilation device. The specific component selection and assembly methods are well-known technologies in the field and will not be elaborated upon here.

[0030] The protective cover module forms an enveloping isolation layer around the through-beam ultrasonic anemometer. In the actual operating conditions of underground coal mine ventilation systems, when the airflow passes through the area where the anemometer is located, the main shell structure of the stainless steel protective cover can physically block large dust particles carried in the airflow. Based on this enveloping spatial layout, not only is the spatial distance for dust to enter the core working area of ​​the sensor increased, but the direct adhesion of large amounts of contaminating particles to the sensor surface is also reduced. The physical protective structure reduces the probability of contamination of the core sensing components, and combined with the mechanically driven cleaning action within the system, it provides a stable working environment for the through-beam ultrasonic anemometer.

[0031] In this embodiment, the mechanical drive module is integrated with the existing sensor hardware via an external support component. Specifically, the transducer bracket 8 has a pre-drilled mounting hole on its plate surface that matches the outer diameter of the ultrasonic transducer 7. The main body housing of the ultrasonic transducer 7 passes through this mounting hole and is directly secured by the original sealing ring and retaining ring 11. Based on this assembly structure, the system reuses the original fixing components of the ultrasonic transducer 7, thereby reducing the overall manufacturing cost.

[0032] Based on the support foundation established by the transducer bracket 8, an L-shaped fixing plate 10 is fitted at the lower end of the transducer bracket 8. The vertical surface of the L-shaped fixing plate 10 is mechanically connected to the transducer bracket 8 through bolt holes and fastening bolts 12. In addition, an elongated hole is provided on the horizontal surface of the L-shaped fixing plate 10, which is used for assembly and fixation with the housing of the motor box 5. By providing the elongated hole, this embodiment provides a certain installation adjustment margin during the assembly process, which facilitates the alignment of the mechanical components.

[0033] The system's power components are integrated inside the motor housing 5. A DC motor 6 is installed inside the motor housing 5. The output shaft of the DC motor 6 extends out of the motor housing 5 and is sequentially connected to the external brush head connecting arm 2 via coupling 4. Based on this power transmission path, the rotational output of the DC motor 6 is converted into the mechanical oscillation of the brush head connecting arm 2. In the structural design of the actuator, a spring 3 is provided at the connection point between the brush head connecting arm 2 and the brush 1. The spring 3 is used to pull the brush 1 inward. During mechanical operation, the tension applied by the spring 3 causes the brush 1 to tightly grip and adhere to the functional surface of the ultrasonic transducer 7, thereby ensuring the actual scraping and cleaning effect.

[0034] Existing wind sensors using protective housings with transparent lenses severely affect ultrasonic wave propagation, leading to a sharp increase in the measurement error of wind speed calculated using the ultrasonic time-of-flight method. Furthermore, since wind sensors in coal mines are generally installed at high locations, traditional cleaning methods rely on manual climbing, which is unsafe in the underground environment, increasing the difficulty and challenges of cleaning. Therefore, this embodiment abandons the lens obstruction design and adopts a direct-contact cleaning solution. Considering the linear transmission characteristics of through-beam wind speed sensors, this embodiment incorporates a corresponding limiting structure design to avoid mechanical interference. Specifically, a positioning block 9 is provided on the upper part of the motor housing 5. The positioning block 9 is radially aligned with the outer diameter of the ultrasonic transducer 7. The purpose of this size and position correlation is to ensure that the positioning block 9 and the brush 1 resting there do not obstruct the ultrasonic transducer 7's ultrasonic wave transmission and reception surfaces, eliminating interference from external physical components on the sound wave transmission path.

[0035] Furthermore, the positioning block 9 simultaneously provides axial mechanical restraint for the brush 1 in its spatial structure. With the spring 3 continuously applying inward tension, if the swing trajectory of the brush 1 deviates from the functional surface area of ​​the ultrasonic transducer 7, it may slide backward and fall into the gap of the motor housing 5. The positioning block 9 physically blocks the range of motion of the brush 1 in the axial direction, preventing the actuator from jamming. During non-cleaning periods monitored by the wind speed sensor, the system controls the brush 1 to swing and remain stationary at the restraint position of the positioning block 9. The specific forming and fixing method of the positioning block 9 can be performed by those skilled in the art using conventional sheet metal processing techniques; it is well-known in the field and will not be elaborated upon here.

[0036] In this embodiment, the intelligent control module of the system is integrated into the mine wind speed and direction sensor. Based on this hardware electrical architecture, the mine wind speed and direction sensor directly sends drive commands to the external DC motor 6 via a connecting cable, and simultaneously uses the same cable to provide the electrical energy required for the operation of the DC motor 6. This setup eliminates the dependence of the external self-cleaning actuator on an independent power supply, and realizes low-level hardware linkage control between the wind measurement mode and the self-cleaning mode on the physical circuit.

[0037] To address the uncertainty of cleaning needs arising from the complex dust environment underground, this embodiment expands upon the single timed cleaning mode. The intelligent control module constructs a multi-mode adaptive intelligent triggering mechanism based on real-time monitoring of multi-source data features. This mechanism executes autonomous prediction and physical cleaning actions through the following parallel logical steps: S101, the system has an independent timing module for executing time-cycle-based predictive triggers. The timing module records the continuous running time of the current monitoring cycle. During actual equipment deployment, technicians preset fixed cleaning intervals in the system based on historical dust concentration data from the current wind tunnel. When the time value recorded by the timing module reaches this preset interval, the intelligent control module generates a cleaning command.

[0038] S102, the system data processing module executes anomaly detection triggering based on the fluctuation characteristics of wind speed data. In engineering applications, when dust adheres to the surface of the ultrasonic transducer 7, the transmission of the ultrasonic signal becomes unstable due to physical interference, which manifests as abnormal fluctuations in instantaneous wind speed data in macroscopic measurement results. Based on the above physical causal relationship, the system continuously acquires instantaneous wind speed sampling data during operation and performs sliding filtering on the sequence data within a set time window. In order to accurately quantify the degree of wind speed fluctuation without limiting specific mathematical formulas, the data processing module extracts a continuous instantaneous wind speed sampling sequence based on the set time window, calculates the dispersion of each instantaneous wind speed sampling point in the sequence from the mean instantaneous wind speed within the window, and thus obtains the fluctuation variance of the instantaneous wind speed data.

[0039] The system compares the variance of the instantaneous wind speed data obtained through calculation with a set dynamic variance threshold in real time. This dynamic variance threshold is obtained through offline calibration based on the baseline normal turbulent flow characteristics of the current roadway ventilation section under dust-free conditions. When the variance of the instantaneous wind speed data is found to be greater than the aforementioned dynamic variance threshold, it indicates that the signal transmission is likely to be distorted due to dust obstruction, and the intelligent control module then generates a cleaning command.

[0040] S103, the system data processing module executes a state determination trigger based on the underlying analog-to-digital conversion sampling values. At the lowest physical signal level, under ideal conditions of no obstruction and no dust, the analog-to-digital conversion sampling values ​​received by the ultrasonic transducer 7 should remain within a relatively stable range. However, dust accumulation on the surface of the ultrasonic transducer 7 directly absorbs and attenuates ultrasonic energy, leading to an overall decrease in the average value or irregular abrupt changes in the distribution of the analog-to-digital conversion sampling values ​​acquired by the ultrasonic transducer 7. The system synchronously monitors the characteristics of the analog-to-digital conversion sampling values ​​within a set time window, and its logical determination criteria are expressed as follows: ; In the above judgment criteria, This indicates a cleaning instruction. This represents the difference in the mean of the analog-to-digital conversion sampled values ​​within the current time window relative to the baseline value calibrated by the device in a dust-free, clean air environment. This indicates the set threshold for the decrease in mean value, which is calculated and set by the researchers based on the maximum energy decay rate allowed by the sensor's measurement accuracy. It represents the standard deviation of the fluctuation of the analog-to-digital conversion sampled values ​​within the current time window. This represents the set standard deviation threshold for fluctuation, reflecting the system's judgment limit for signal jitter caused by high-frequency environmental noise. This represents the logical OR operator. Based on this criterion, once the mean decrease exceeds the set mean decrease threshold, or the standard deviation of fluctuation exceeds the set standard deviation of fluctuation threshold, the system generates a cleaning command to initiate a self-cleaning process, thereby restoring the ultrasonic signal transmission strength. To ensure the completeness of the above statistical calculations, the set time window must contain at least two valid sampling points.

[0041] S104, execute collaborative decision-making and mechanical reset actions. The above three judgment logics are monitored in parallel in the system, forming a multi-dimensional decision-making closed loop. When the intelligent control module captures any cleaning command generated in steps S101, S102, or S103, the intelligent control module sends a drive command to the DC motor 6 through the control circuit. After receiving the command, the DC motor 6 starts and drives the brush 1 to complete reciprocating physical cleaning within a specified angle through the mechanical transmission structure. The left and right swing angle of the brush 1 is preset and controlled in the wind speed sensor. After the current cleaning action is completed, the intelligent control module clears its internal timing module, simultaneously resets the data cache in all sliding time windows to the initial state, and controls the brush 1 to return to the limit position and stop, waiting to enter the next cycle and feature monitoring loop.

[0042] The multi-mode intelligent triggering closed loop constructed in this embodiment realizes the leap from timed passive cleaning to on-demand active cleaning. This not only effectively avoids measurement failures or errors caused by untimely cleaning, but also reduces meaningless mechanical movements, significantly extends the service life of motors and brushes, and improves the overall reliability of mining wind measurement equipment.

[0043] Specific application examples: To more intuitively illustrate the working process of the multi-mode adaptive intelligent triggering mechanism in this invention, a specific application example is given, using the main return airway with high dust concentration in a coal mine as an application scenario: In this scenario, the through-beam ultrasonic anemometer is installed at the top of the tunnel (approximately 3.5 meters from the floor). When the system is powered on and initialized (step S10), the technician sets the following parameters: Timed cycle: The preset cleaning interval is 24 hours.

[0044] Wind speed variance threshold: The dynamic variance threshold is set to 0.5 (m / s). 2 .

[0045] AD sampling threshold: Mean decline threshold Set to 15% of the baseline value; standard deviation of fluctuation threshold The settings are based on the ambient background noise.

[0046] The system operation process is as follows: Phase 1 (Normal Operation): During the first 12 hours of sensor operation, the variance of wind speed data remains between 0.1 and 0.2 (m / s). 2 The AD mean decrease rate was 2%, which did not reach the set dynamic variance threshold and mean decrease threshold, so the brush remained at the positioning block.

[0047] Phase Two (Sudden High-Concentration Dust Interference): During the 14th hour of operation, a coal-cutting operation occurred upstream in the roadway, sending a large amount of dust through the airflow to the sensors. While the physical shield blocked large coal particles, fine dust quickly adhered to the transducer surface, causing distortion of the ultrasonic signal. The system detected a sharp fluctuation in instantaneous wind speed data, with the calculated variance of the fluctuation jumping to 0.8 (m / s). 2 At this point, the wind speed data fluctuation characteristic trigger condition was met (0.8 > 0.5), and the intelligent control module immediately issued a cleaning command, driving the motor to perform three cycles of cleaning. After cleaning, the variance returned to 0.15 (m / s). 2 At the same time, the system will reset the timing module and the data cache.

[0048] Phase Three (Slow Dust Accumulation Interference): During the subsequent 20 hours of operation (20 hours after the last cleaning, but before reaching the 24-hour timer cycle), airborne dust particles slowly and evenly adhered to the transducer surface. The wind speed signal did not fluctuate drastically (variance not exceeded limits), but the system's underlying monitoring showed a slow decrease of 18% in the mean value of the AD sampling values. This was determined according to the established underlying signal logic judgment criteria. At this time, the difference in mean decrease (18%) has exceeded the set threshold for mean decline. (15%). Therefore, the condition for triggering the AD sampling value status determination (S103) is met, and the system generates a cleaning command. It then actively performed the cleaning action again, restored the energy transmission intensity of the ultrasound, and reset the timer to zero again.

[0049] Phase Four (Timed Cull-Off Trigger): During the subsequent 24 hours, due to the downhole shutdown and maintenance, the airflow is clean, and none of the above-mentioned anomaly detections are triggered. When the continuous running time recorded by the timing module reaches the preset 24 hours, the timing cycle prediction trigger condition is met, and the system performs routine preventative cleaning, completing closed-loop monitoring for multi-dimensional decision-making.

[0050] As can be seen from this application example, the present invention effectively avoids mechanical wear caused by sweeping even when there is no dirt, and achieves a second-level response in the event of sudden pollution, realizing true on-demand cleaning.

[0051] Experimental verification and comparative analysis with accompanying figures: To verify the system's effectiveness under complex working conditions, the applicant constructed a coal mine roadway simulation test chamber with dust injection capabilities and conducted a 72-hour continuous comparative experiment. Please refer to the attached document. Figure 5 Experimental Results and Data Analysis: First, regarding the appendix Figure 5 The coordinate axes are interpreted as follows: the horizontal axis (X-axis) represents the running time (hours), indicating the continuous running time of the coal mine underground simulation test chamber. The test cycle ranges from 0 to 72 hours, reflecting the working duration of the sensor in a continuously dusty environment. The vertical axis (Y-axis) represents the measurement error (%), indicating the percentage deviation between the wind speed data measured by the ultrasonic anemometer and the actual standard wind speed. A larger value indicates more severe dust accumulation on the sensor surface, leading to greater distortion of the ultrasonic signal; a smaller and more stable value indicates a cleaner sensor surface and more accurate measurements.

[0052] Based on this, the specific performance of the three experimental groups is as follows: Control group A: A traditional through-beam ultrasonic anemometer (without a self-cleaning device) was used.

[0053] exist Figure 5In the graph, the curve marked with a solid black line and a hollow square represents control group A. Graphically, this curve exhibits a parabolic trend that increases continuously over time. Physically, this indicates that without any cleaning measures, dust accumulation on the transducer surface will continuously build up with increasing operating time, leading to accelerated attenuation of ultrasonic energy. Experimental data clearly reflects this: control group A experienced a sharp increase in error 12 hours after the start of the experiment, with the measurement error exhibiting a non-linear and rapid amplification. Around 40 hours later, the error exceeded 15%, rendering it completely unusable and causing the sensor to fail entirely.

[0054] Control group B: A wind speed sensor with a traditional timed cleaning mechanism was used (only set to clean once every 24 hours).

[0055] exist Figure 5 In the graph, the curve marked with a black dashed line and a hollow circle represents control group B. Graphically, this curve exhibits a distinct sawtooth pattern. The error gradually increases from 0 to 24 hours, then drops to a low point at the designated cleaning node (24 hours); it then rises again, dropping once more at 48 hours. This physically represents the limitations of current conventional timed cleaning logic: while fixed cleaning allows for a brief recovery of the error, traditional timed cleaning cannot respond promptly during the gap between cleaning nodes, especially when facing sudden high concentrations of dust. Experimental data shows that the maximum error still spikes to over 8% during the blind zone, still generating significant amounts of dirty data.

[0056] Experimental Group C: The self-cleaning system of the through-beam ultrasonic transducer of the present invention (with multi-mode adaptive triggering enabled) is used.

[0057] exist Figure 5 In the diagram, the black dotted line connected to the curve marked with a gray filled triangle represents experimental group C. Graphically, this curve remains closely aligned with the bottom X-axis, exhibiting extremely small sawtooth fluctuations, representing the sudden cleaning commands issued in real-time by the system. Its physical meaning is a direct manifestation of the technical solution of this invention: it illustrates that the underlying judgment criteria based on wind speed variance and AD sampling values, as described in the above embodiments, were incorporated. Afterwards, once the control module detects in the background sliding window calculation that dust accumulation is causing the error to rise (about to exceed the set threshold), the system will immediately clean as needed, instantly reducing the error back to the minimum. Experimental data shows that, thanks to real-time perception and intervention of multi-source data, the maximum peak value was strictly limited throughout the entire 72-hour test cycle, and the wind speed measurement error was consistently controlled within an accuracy range of 2.5%.

[0058] Comprehensive comparison of mechanical losses: In the aforementioned 72-hour experiment, control group B mechanically and consistently performed three cleaning cycles regardless of the actual contamination levels. In contrast, experimental group C (the present invention), despite experiencing several random impacts from high-concentration dust, was precisely identified by the system's intelligent judgment logic and only underwent four necessary cleaning cycles. The system does not rigidly adhere to fixed cycles, ensuring extremely high data accuracy throughout its entire lifecycle while avoiding ineffective wear on the transducer's anti-corrosion coating and brushes caused by frequent, indiscriminate scraping.

[0059] In summary, the experimental data and the attached curves fully demonstrate that the present invention has achieved substantial technical breakthroughs in improving the accuracy of wind measurement in harsh dust environments, reducing mechanical wear, and ensuring the long-term reliable operation of equipment.

Claims

1. A self-cleaning system for a through-beam ultrasonic transducer, characterized in that, Includes mechanical drive modules and intelligent control modules; The mechanical drive module includes a DC motor (6) and a wiper linkage structure, wherein the DC motor (6) drives the wiper linkage structure to perform cleaning on the ultrasonic transducer (7) of the through-beam ultrasonic anemometer. The intelligent control module is connected to the DC motor (6) and includes a timing module, a data acquisition module, and a data processing module. The timing module is used to record the continuous running time and generate a cleaning instruction when the preset cleaning interval period is reached; The data acquisition module is used to acquire instantaneous wind speed data and analog-to-digital conversion sampling values ​​from the receiving end of the ultrasonic transducer (7); The data processing module is used to calculate the fluctuation variance of the instantaneous wind speed data, and generate a cleaning instruction when the fluctuation variance is greater than the calibrated dynamic variance threshold, or when the mean decrease difference of the analog-to-digital conversion sampled value relative to the calibrated benchmark value is greater than the set mean decrease threshold within the set time window, or when the fluctuation standard deviation of the analog-to-digital conversion sampled value is greater than the set fluctuation standard deviation threshold within the set time window. The intelligent control module is used to receive the cleaning command and send the drive command to the DC motor (6).

2. The self-cleaning system for a through-beam ultrasonic transducer according to claim 1, characterized in that, The wiper linkage structure includes a brush head connecting arm (2) and a brush (1). The DC motor (6) drives the brush head connecting arm (2) and the brush (1) to swing back and forth to perform the cleaning. A spring (3) is provided at the connection between the brush head connecting arm (2) and the brush (1). The spring (3) pulls the brush (1) inward to adhere it to the surface of the ultrasonic transducer (7).

3. The self-cleaning system for a through-beam ultrasonic transducer according to claim 2, characterized in that, The mechanical drive module is located at the probe of the through-beam ultrasonic anemometer. The mechanical drive module also includes a motor box (5) and a coupling (4). The DC motor (6) is installed inside the motor box (5). The output shaft of the DC motor (6) passes through the motor box (5) and is connected to the external brush head connecting arm (2) through the coupling (4).

4. The self-cleaning system for a through-beam ultrasonic transducer according to claim 3, characterized in that, The upper part of the motor box (5) is provided with a positioning block (9). The installation position of the positioning block (9) is radially flush with the outer diameter of the ultrasonic transducer (7). The positioning block (9) physically blocks the range of motion of the brush (1) in the axial direction. During non-cleaning periods, the brush (1) stays at the limit position where the positioning block (9) is located.

5. A self-cleaning system for a through-beam ultrasonic transducer according to claim 4, characterized in that, It also includes a transducer bracket (8) and an L-shaped fixing plate (10). The plate surface area of ​​the transducer bracket (8) is pre-opened with an assembly hole that matches the outer diameter of the ultrasonic transducer (7). The main body shell of the ultrasonic transducer (7) passes through the assembly hole and is pressed and fixed at the end face by a pre-set sealing ring and a fixing ring (11).

6. A self-cleaning system for a through-beam ultrasonic transducer according to claim 5, characterized in that, The lower end of the transducer bracket (8) is fitted with the L-shaped fixing plate (10). The vertical surface of the L-shaped fixing plate (10) is mechanically connected to the transducer bracket (8) through preset bolt holes and fastening bolts (12). The horizontal surface of the L-shaped fixing plate (10) is provided with elongated holes, which are used to assemble and fix with the housing of the motor box (5).

7. The self-cleaning system for a through-beam ultrasonic transducer according to claim 1, characterized in that, It also includes a protective cover module that covers the outside of the through-beam ultrasonic anemometer. The main structure of the protective cover module is a stainless steel protective cover. The protective cover is composed of a split-type assembled shell and movable connecting parts to form an openable structure. The relative opening and closing operation between the shells is realized through the movable connecting parts.

8. A self-cleaning system for a through-beam ultrasonic transducer according to claim 1, characterized in that, The intelligent control module is integrated inside the through-beam ultrasonic wind speed sensor. The through-beam ultrasonic wind speed sensor sends the drive command directly to the external DC motor (6) through a preset aviation cable, and simultaneously uses the aviation cable to provide the DC motor (6) with the power required for operation.

9. A self-cleaning system for a through-beam ultrasonic transducer according to claim 1, characterized in that, The data processing module is used to perform sliding window filtering calculation on the instantaneous wind speed data, extract a continuous instantaneous wind speed sampling sequence based on a set time window, and calculate the degree of dispersion between each instantaneous wind speed sampling point in the instantaneous wind speed sampling sequence and the mean instantaneous wind speed in the set time window to obtain the fluctuation variance. The calibrated dynamic variance threshold is obtained offline based on the baseline normal turbulence characteristics under dust-free conditions. The calibration reference value is the reference value calibrated in a dust-free, clean air environment; The set mean decline threshold is calculated and set based on the maximum energy decay rate allowed by the sensor measurement accuracy; The set standard deviation threshold for fluctuation is based on the judgment limit for signal jitter caused by high-frequency environmental noise.

10. A self-cleaning system for a through-beam ultrasonic transducer according to claim 1, characterized in that, The intelligent control module is also used to reset the internal timing module after the current cleaning action is completed, and simultaneously reset the data cache in all set time windows in the data processing module to the initial state.

Images