Sound-based particle water content detection method and device

Abstract

The invention provides a sound-based particle water content detection method and device, the detection method is applied to a detection system, the detection system comprises a horizontally arranged plate, and the detection method comprises the following steps: controlling particles to impact the surface of the plate in a free falling body manner according to a preset height so as to generate an impact sound signal; collecting impact sound signals; performing noise reduction processing on the collected sound signal to obtain a target sound signal; extracting a target spectrum feature of the target sound signal; and comparing the target frequency spectrum characteristic with a pre-established corresponding relation between the frequency spectrum characteristic and the water content to obtain the particle water content corresponding to the target frequency spectrum characteristic, thereby realizing non-contact measurement, avoiding the problem of sensor pollution, having high precision and real-time online monitoring capability, and being low in equipment cost, strong in environmental adaptability and suitable for popularization and application. The efficiency and the reliability of particle water content detection are remarkably improved, and the method is suitable for various industrial scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of sludge particle moisture detection, and in particular to a sound-based particle moisture content detection method and device. Background Art

[0002] In industrial fields such as sewage treatment and solid waste treatment, the detection of sludge particle moisture content is of great significance for process optimization and cost control.

[0003] Traditional methods for measuring moisture content primarily include drying and weighing methods and near-infrared spectroscopy. The drying and weighing method calculates moisture content by measuring the difference in mass between the sludge before and after drying. This method requires a long drying process and is difficult to meet the needs of real-time monitoring. While near-infrared spectroscopy allows for rapid detection, the equipment is expensive and has strict requirements for the uniform distribution of sludge particles. In practical applications, it is susceptible to interference from uneven particle distribution, resulting in reduced measurement accuracy. Furthermore, contact-based detection methods suffer from the problem of sensor contamination by sludge adhesion, requiring frequent maintenance, increasing both cost and operational complexity. Summary of the Invention

[0004] The present invention provides a method and device for detecting the water content of particles based on sound, so as to solve the technical problem in the prior art that the water content of particles cannot be detected with high precision and without contact.

[0005] In one aspect, the present invention provides a method for detecting the moisture content of particles based on sound, the method being applied to a detection system comprising a horizontally arranged plate, the method comprising: Controlling the particles to impact the surface of the plate in a free-fall manner at a preset height to generate an impact sound signal; collecting the impact sound signal; Perform noise reduction processing on the collected sound signal to obtain the target sound signal; extracting target frequency spectrum features of the target sound signal; The target spectrum feature is compared with a pre-established correspondence between spectrum features and water content to obtain the water content of the particles corresponding to the target spectrum feature.

[0006] According to a sound-based particle moisture content detection method provided by the present invention, the detection system includes an acoustic sensor array and a sound card, which are arranged on the side or bottom surface of the board. The collection of the impact sound signal includes: controlling the acoustic sensor array to collect the impact sound signal and convert the collected impact sound signal into an analog electrical signal; The sound card is controlled to convert the analog electrical signal into a digital signal.

[0007] According to a sound-based particle moisture content detection method provided by the present invention, the acoustic sensor array includes a piezoelectric sensor and a microphone; The step of controlling the acoustic sensor array to collect the impact sound signal and converting the collected impact sound signal into an analog electrical signal includes: controlling the piezoelectric sensor and the microphone to independently collect the impact sound signal, and converting the collected impact sound signal into an analog electrical signal; The controlling the sound card to convert the analog electrical signal into a digital signal comprises: The sound card is controlled to synchronously receive the analog electrical signals output by the piezoelectric sensor and the microphone, and convert the analog electrical signals into digital signals.

[0008] According to a sound-based particle moisture content detection method provided by the present invention, the detection system further includes: pellet dosing device and soundproof box; wherein the particle delivery device and the plate are located in the soundproof box; The controlling particles to hit the surface of the plate in a free-fall manner at a preset height includes: The particles are controlled by the particle delivery device to hit the surface of the plate in a free-fall manner at a preset height.

[0009] According to a sound-based particle moisture content detection method provided by the present invention, the correspondence between the spectral characteristics and the moisture content is established in the following manner: Prepare particle samples with different water contents; allowing each particle sample to freely fall onto the surface of the plate at a preset height, and collecting an impact sound signal; Extracting the spectrum characteristics of the impact sound signal; Through machine learning algorithms, a correspondence model between spectrum characteristics and water content is established.

[0010] According to a sound-based particle moisture content detection method provided by the present invention, the target spectrum characteristics include: Main frequency, spectral center of gravity and high frequency attenuation slope; Wherein, the main frequency is the frequency value with the largest amplitude in the spectrum of the target sound signal; The spectrum center of gravity is the energy distribution center of the target sound signal spectrum; The high-frequency attenuation slope is the attenuation slope of the target sound signal spectrum in the high-frequency band.

[0011] According to a sound-based particle moisture content detection method provided by the present invention, the target spectrum characteristics further include: spatiotemporal characteristics; The spatiotemporal features are extracted in the following way: The temporal and spatial characteristics of the target sound signal are extracted using a time series analysis method and a spatial analysis method; wherein the temporal and spatial characteristics include the propagation speed, reflection intensity and propagation path of the sound wave.

[0012] According to a sound-based particle moisture content detection method provided by the present invention, the number of the acoustic sensor arrays is multiple and they are respectively arranged at different positions of the board, and the collecting of the impact sound signal includes: In the process of collecting the impact sound signal, a sound source localization model is constructed using a plurality of acoustic sensor arrays; Based on the sound source localization model, the actual spatial coordinates of the particle impact point are inferred using the sound wave arrival time difference algorithm; Using the sound source localization model to identify whether there are multiple particles that collide synchronously or bounce back multiple times; If it exists, the effective impact sound signal generated by the single particle impact is extracted.

[0013] According to the present invention, a method for detecting the moisture content of particles based on sound is provided, which utilizes the sound source localization model to identify whether there are multiple particles that collide synchronously or bounce back multiple times, including: For each impact sound signal recorded by the acoustic sensor array, extract its arrival timestamp; Calculate the time difference between different acoustic sensor arrays in the same impact event; If the time differences recorded by all acoustic sensor arrays conform to the expected TDOA distribution of a single impact point, it is judged to be a single particle impact; If there is an expected TDOA distribution beyond a single impact point, it is judged to be a multi-particle synchronous impact or a multi-particle rebound impact.

[0014] On the other hand, the present invention also provides a device for detecting the moisture content of particles based on sound, wherein the device is applied to a detection system, wherein the detection system includes a horizontally arranged plate, and the device includes: a delivery module, configured to control the particles to impact the surface of the plate in a free-fall manner at a preset height to generate an impact sound signal; A signal collection module, configured to collect the impact sound signal; A noise reduction module is used to perform noise reduction processing on the collected sound signals to obtain the target sound signals; A feature extraction module, configured to extract target spectrum features of the target sound signal; The water content determination module is used to compare the target spectrum characteristics with the pre-established correspondence between spectrum characteristics and water content to obtain the particle water content corresponding to the target spectrum characteristics.

[0015] The sound-based particle moisture content detection method and device provided by the present invention controls the free fall of particles to hit a plate and collects the impact sound signal. After noise reduction processing and spectrum feature extraction, it is compared with a pre-established correspondence to determine the moisture content. This realizes non-contact measurement and avoids the problem of sensor contamination. At the same time, it has high-precision, real-time online monitoring capabilities, low equipment cost, strong environmental adaptability, significantly improving the efficiency and reliability of particle moisture content detection, and is suitable for a variety of industrial scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 1 is a flow chart of a method for detecting moisture content of particles based on sound provided in an embodiment of the present invention; Figure 2 1 is a schematic structural diagram of a device for detecting moisture content of particles based on sound provided in an embodiment of the present invention; Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0019] Figure 1 FIG2 is a flow chart of a method for detecting moisture content of particles based on sound according to an embodiment of the present invention. The detection method can be applied to a detection system, which can include a horizontally arranged plate.

[0020] See also Figure 1 The sound-based particle moisture content detection method may include the following steps 101 to 105.

[0021] Step 101: Control particles to hit the surface of a board in a free-fall manner at a preset height to generate impact sound signals.

[0022] In this step, particles generally refer to sludge particles, but other types of particles are also possible and are not listed here. The preset height can be set according to actual needs. For example, it can generally be set to any value between 10 and 50 cm, preferably 30 cm. The plate can be steel or other materials, such as ceramic or aluminum alloy.

[0023] Step 102: Collect impact sound signals.

[0024] In this step, the detection system may include an acoustic sensor array and a sound card, which are arranged on the side or bottom surface of the board.

[0025] Collecting impact sound signals may include: controlling the acoustic sensor array to collect impact sound signals and converting the collected impact sound signals into analog electrical signals; Controls the sound card to convert analog electrical signals into digital signals.

[0026] This step achieves efficient acquisition and digital processing of impact sound signals by introducing a combination of an acoustic sensor array and a sound card. The acoustic sensor array accurately collects impact sound signals and converts them into analog electrical signals, while the sound card converts the analog signals into digital signals, providing a high-quality data foundation for subsequent signal processing and analysis. This step not only improves the stability and accuracy of signal acquisition, but also enhances anti-interference capabilities, ensuring the reliability of the detection process. The sound card can be a 24-bit ADC sound card with a sampling rate of 44.1kHz.

[0027] Step 103: Perform noise reduction processing on the collected sound signal to obtain a target sound signal.

[0028] In this step, the collected sound signal is subjected to noise reduction processing, which can include wavelet denoising and adaptive filtering. Wavelet denoising uses wavelet transforms to process the collected sound signal. Wavelet transforms can decompose the signal into sub-signals with different frequency components, effectively separating high-frequency noise from low-frequency valid signals. For example, selecting the db4 wavelet basis for soft threshold processing can remove high-frequency noise from the signal while retaining the low-frequency valid signal components. Adaptive filtering uses an adaptive filtering module (such as a filter based on the minimum mean square error (LMS) algorithm) to suppress environmental noise. Adaptive filters can dynamically adjust filtering parameters based on the characteristics of the environmental noise, thereby better adapting to different detection environments.

[0029] The noise reduction processing of the collected sound signals may also include: Acceleration compensation of particles; The compensation formula is A norm = A raw / (2 gh ), where h is the height of the particle, that is, the distance from the particle's free fall to the impact plate. A raw is the original acceleration signal, that is, the amplitude of the sound signal collected directly from the acoustic sensor array, g is the acceleration due to gravity, A norm is the compensated acceleration signal, i.e., the amplitude of the compensated sound signal. This compensation can eliminate signal amplitude differences caused by varying particle placement heights, further improving signal consistency and comparability, thereby enhancing the reliability and accuracy of the detection method.

[0030] Step 104: extract target frequency spectrum features of the target sound signal.

[0031] In this step, principal component analysis (PCA) or random forest can be used to screen key features, that is, to screen target spectrum features.

[0032] Target spectrum characteristics may include: main frequency ( f peak ), spectral center of gravity ( S centroid ) and the high frequency attenuation slope ( α high The dominant frequency reflects the primary vibration characteristic of the signal. The center of gravity of the spectrum reflects the concentration of energy distribution. The high-frequency attenuation slope reflects the ability of water to absorb high-frequency sound waves.

[0033] Among them, the main frequency is the frequency value with the largest amplitude in the spectrum of the target sound signal, which can be expressed as formula (1): (1); in, is a time domain signal x ( t ) corresponds to the amplitude in the frequency domain, Indicates taking the maximum value.

[0034] The center of gravity of the spectrum is the energy distribution center of the target sound signal spectrum, which can be expressed as formula (2): (2); in, represents the i-th frequency point, The position of the center of gravity of the spectrum is closely related to the moisture content of the particles. When the moisture content increases, the center of gravity shifts to low frequencies.

[0035] The high-frequency attenuation slope is the attenuation slope of the target sound signal spectrum in the high-frequency band (e.g., 5-20kHz), which is related to the viscous damping of water and can be expressed as formula (3): (3); in, f 1 and f 2 represents two frequency points, f 2 is greater than f 1. Usually choose the linear region of the high-frequency attenuation segment; Indicates from f 1 to f 2 decibel value of signal attenuation.

[0036] Step 105 : Compare the target spectrum feature with the pre-established correspondence between the spectrum feature and the water content to obtain the particle water content corresponding to the target spectrum feature.

[0037] In this step, the target spectrum feature can be used as input and the particle moisture content can be used as output.

[0038] In this embodiment, by controlling the free fall of particles to hit the plate and collecting the impact sound signal, after noise reduction processing and spectral feature extraction, the water content is determined by comparing it with the pre-established correspondence, thereby realizing non-contact measurement and avoiding the problem of sensor contamination. At the same time, it has high-precision, real-time online monitoring capabilities, low equipment cost, strong environmental adaptability, and significantly improves the efficiency and reliability of particle moisture content detection. It is suitable for a variety of industrial scenarios.

[0039] In one embodiment of the present specification, the acoustic sensor array includes a piezoelectric sensor and a microphone; Controlling the acoustic sensor array to collect the impact sound signal and converting the collected impact sound signal into an analog electrical signal, including: controlling the piezoelectric sensor and the microphone to independently collect impact sound signals, and converting the collected impact sound signals into analog electrical signals; Control the sound card to convert analog electrical signals into digital signals, including: The sound card is controlled to synchronously receive the analog electrical signals output by the piezoelectric sensor and the microphone and convert them into digital signals.

[0040] In this embodiment, by introducing an acoustic sensor array (including piezoelectric sensors and microphones) and a sound card, multi-source acquisition and synchronous digital processing of impact sound signals are achieved. This not only improves the stability and accuracy of signal acquisition, but also enhances signal reliability through multi-sensor data fusion, further enhancing the accuracy and anti-interference capabilities of moisture content detection, providing stronger technical support for high-precision detection in complex industrial environments. The piezoelectric sensor has a frequency response range of 20Hz-20kHz and a sensitivity of 50mV / g.

[0041] In one embodiment of this specification, the detection system further includes: pellet dosing device and soundproof box; Wherein, the particle delivery device and the plate are located in a soundproof box; Control the particles to hit the surface of the board in a free-fall manner according to a preset height, including: The particles are controlled by the particle delivery device to hit the surface of the plate in a free fall manner at a preset height.

[0042] In this example, by adding a particle delivery device and a soundproof enclosure to the detection system, particles can freely fall onto the plate at a precisely controlled height, and signal acquisition is performed in a low-noise environment. This effectively eliminates the impact of inconsistent delivery heights and background noise on test results, further improving detection accuracy and repeatability while enhancing the system's environmental adaptability, ensuring high-precision and high-reliability online detection of particle moisture content. The background noise level of the soundproof enclosure is <30dB.

[0043] There are many types of particle delivery devices, such as a funnel-type delivery device and a mechanical delivery device.

[0044] The funnel-type delivery device consists of a funnel and a height-adjustable bracket. A controllable valve is located at the bottom of the funnel to control the release of particles. The mechanical delivery device consists of a robotic arm, a particle clamping device, and a height-adjustable bracket. The robotic arm is driven by a motor, enabling precise control of the particle delivery position and height.

[0045] In one embodiment of this specification, the correspondence between the spectral characteristics and the water content is established in the following manner: Step 1: Prepare particle samples with different water contents; In this step, different moisture contents can be prepared as needed, for example, 30% to 95% moisture content. The actual moisture content (i.e., moisture content) can be calibrated by drying. The particle size is generally 3-5 mm. The moisture content can be expressed by the following formula (4): (4); in, Indicates the moisture content, represents the wet weight of the particles, Indicates the dry weight of the particles.

[0046] Step 2: allowing each particle sample to freely fall from a preset height and impact the surface of the board, and collecting the impact sound signal; In this step, the sampling rate of the impact sound signal is generally greater than 44.1kHz, and the duration covers the entire impact process (approximately 0.1-0.5 seconds). If the plate is a steel plate, the surface roughness Ra is less than 0.8μm.

[0047] Step 3: extracting the frequency spectrum characteristics of the impact sound signal; In this step, the target spectrum characteristics may include: main frequency (100-2000 Hz), spectrum center of gravity (200-1500 Hz), and high-frequency attenuation slope (10-30 dB / oct).

[0048] Step 4: Use machine learning algorithms to establish a corresponding relationship model between spectrum characteristics and water content; The corresponding relationship model can be expressed by formula (5): = (5); Where f() represents a function that predicts the water content based on the input spectral features.

[0049] In this embodiment, by establishing a correspondence model between spectral features and moisture content, a scientific and reliable quantitative basis is provided for the detection of particle moisture content. This model, based on the spectral features of the impact sound signals of particle samples with different moisture contents, is trained and optimized using a machine learning algorithm. It accurately maps the extracted spectral features to moisture content values. This not only improves the accuracy and efficiency of moisture content detection, but also enhances the adaptability and generalization capabilities of the detection system, enabling it to handle a variety of particle materials and complex practical application scenarios, providing strong support for real-time monitoring and optimization of industrial processes. It is understood that as the number of feature types of the target spectral features increases, corresponding feature types will also be added when establishing the correspondence.

[0050] In this embodiment, after establishing the correspondence model, the correspondence model can also be verified. For example, the mean absolute error (MAE) of the correspondence model can be verified. ) and the coefficient of determination ( ).

[0051] The mean absolute error is shown in the following formula (6): (6); in, is the sample size, is the actual water content of the i-th sample, is the predicted moisture content of the i-th sample.

[0052] The coefficient of determination is shown in the following formula (7): (7); in, It represents the average value of the actual moisture content of all samples.

[0053] In one embodiment of this specification, the target spectrum characteristics further include: spatiotemporal characteristics; Spatiotemporal features are extracted in the following ways: The temporal and spatial characteristics of the target sound signal are extracted using time series analysis methods and spatial analysis methods; the temporal and spatial characteristics include the propagation speed, reflection intensity and propagation path of the sound wave.

[0054] In this embodiment, the feature extraction dimensions of the sound signal are further enriched by introducing spatiotemporal features (such as the propagation speed, reflection intensity, and propagation path of the sound wave). These spatiotemporal features, combined with spectral features, can more comprehensively describe the sound signal generated by particle impact, thereby improving the accuracy and reliability of water content prediction.

[0055] In one embodiment of the present specification, there are multiple acoustic sensor arrays, which are respectively arranged at different positions of the board to collect impact sound signals, including: In the process of collecting impact sound signals, multiple acoustic sensor arrays are used to build a sound source localization model; Based on the sound source localization model, the time difference of arrival of sound waves is used to infer the actual spatial coordinates of the particle impact point; Use the sound source localization model to identify whether there are multiple particles hitting simultaneously or bouncing back multiple times; If it exists, the effective impact sound signal generated by the single particle impact is extracted.

[0056] In this embodiment, by deploying multiple acoustic sensor arrays and constructing a sound source localization model, precise location of particle impact points and identification of multiple particle impacts are achieved. Using a time-difference-of-arrival algorithm, a single-particle impact can be effectively distinguished from simultaneous or rebounding impacts of multiple particles, extracting the effective signal from a single particle. This significantly improves the detection system's anti-interference capability and signal processing accuracy, ensuring accurate measurement of particle moisture content even under complex conditions (such as simultaneous impacts of multiple particles), further enhancing the practicality and reliability of the detection method.

[0057] In one embodiment of the present specification, a sound source localization model is used to identify whether there are multiple particles that collide synchronously or bounce back multiple times, including: For each impact sound signal recorded by the acoustic sensor array, extract its arrival timestamp; Calculate the time difference between different acoustic sensor arrays in the same impact event; If the time differences recorded by all acoustic sensor arrays conform to the expected TDOA distribution of a single impact point, it is judged to be a single particle impact; If there is an expected TDOA distribution beyond a single impact point, it is judged to be a multi-particle synchronous impact or a multi-particle rebound impact.

[0058] In this embodiment, by analyzing the arrival timestamps and time differences of the impact sound signals recorded by the acoustic sensor array, it is possible to accurately determine whether the impact event is a single-particle impact or a multi-particle impact. This not only improves the accuracy of impact event identification, but also enhances the robustness of the detection system, enabling it to operate stably in complex industrial environments and effectively avoiding the influence of multi-particle interference on the water content detection results.

[0059] The core principle of this invention is based on the following sound wave generation mechanism. The kinetic energy of sludge particles hitting the plate is converted into sound energy, and the characteristics of the sound waves are affected by the viscoelasticity, density, and internal moisture content of the particles. High-moisture sludge (e.g., >80%): There is a lot of internal moisture, strong viscous damping, low sound wave amplitude, the main frequency is biased towards low frequencies (100-500Hz), and high-frequency attenuation is rapid (e.g., attenuation slope >20dB / oct). Low-moisture sludge (e.g., <60%): The particles are hard, the impact energy release is concentrated, the sound wave amplitude is high, and the spectrum exhibits broadband characteristics (e.g., main frequency >1kHz, high-frequency retention >50%).

[0060] Table 1 below shows an example of the measurement of the moisture content of the pellets.

[0061] Table 1

[0062] From Table 1 above, we can see that the main frequency of the high water content sample (S3) is significantly lower than that of the low water content sample (S1), and the center of gravity of the spectrum shifts to the left, which is consistent with theoretical expectations. The model prediction mean absolute error MAE = 1.8%, , verify the feasibility.

[0063] Based on the same general inventive concept, the present invention also protects a device for detecting the moisture content of particles based on sound, such as Figure 2 As shown, Figure 2 Schematic diagram of the structure of a device for detecting moisture content in particles based on sound, provided by an embodiment of the present invention. The device for detecting moisture content in particles based on sound, provided by the present invention, is described below. The device for detecting moisture content in particles based on sound described below can be used in conjunction with the method for detecting moisture content in particles based on sound described above.

[0064] The device for detecting the moisture content of particles based on sound includes a delivery module 201 , a signal collection module 202 , a noise reduction module 203 , a feature extraction module 204 and a moisture content determination module 205 .

[0065] The delivery module 201 is used to control the particles to hit the surface of the plate in a free-fall manner according to a preset height to generate an impact sound signal; The signal collecting module 202 is used to collect the impact sound signal; The noise reduction module 203 is used to perform noise reduction processing on the collected sound signal to obtain the target sound signal; The feature extraction module 204 is used to extract the target spectrum features of the target sound signal; The water content determination module 205 is used to compare the target spectrum feature with a pre-established correspondence between spectrum features and water content to obtain the particle water content corresponding to the target spectrum feature.

[0066] Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention.

[0067] like Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call logic instructions in the memory 330 to execute the sound-based particle moisture content detection method.

[0068] Furthermore, the logic instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0069] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the sound-based particle moisture content detection method provided by the above methods.

[0070] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the sound-based particle moisture content detection method provided by the above methods.

[0071] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0072] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for detecting the moisture content of particles based on sound, characterized in that: The detection method is applied to a detection system, wherein the detection system includes a horizontally arranged board, and the detection method includes: Controlling the particles to impact the surface of the plate in a free-fall manner at a preset height to generate an impact sound signal; collecting the impact sound signal; Perform noise reduction processing on the collected sound signal to obtain the target sound signal; extracting target frequency spectrum features of the target sound signal; The target spectrum feature is compared with a pre-established correspondence between spectrum features and water content to obtain the water content of the particles corresponding to the target spectrum feature.

2. The method for detecting the moisture content of particles based on sound according to claim 1, characterized in that: The detection system includes an acoustic sensor array and a sound card, which are arranged on the side or bottom surface of the board. The collection of the impact sound signal includes: controlling the acoustic sensor array to collect the impact sound signal and convert the collected impact sound signal into an analog electrical signal; The sound card is controlled to convert the analog electrical signal into a digital signal.

3. The method for detecting the moisture content of particles based on sound according to claim 2, characterized in that: The acoustic sensor array includes a piezoelectric sensor and a microphone; The step of controlling the acoustic sensor array to collect the impact sound signal and converting the collected impact sound signal into an analog electrical signal includes: controlling the piezoelectric sensor and the microphone to independently collect the impact sound signal, and converting the collected impact sound signal into an analog electrical signal; The controlling the sound card to convert the analog electrical signal into a digital signal comprises: The sound card is controlled to synchronously receive the analog electrical signals output by the piezoelectric sensor and the microphone, and convert the analog electrical signals into digital signals.

4. The method for detecting the moisture content of particles based on sound according to claim 1, characterized in that: The detection system also includes: pellet delivery device and soundproof box; wherein the particle delivery device and the plate are located in the soundproof box; The controlling particles to hit the surface of the plate in a free-fall manner at a preset height includes: The particles are controlled by the particle delivery device to hit the surface of the plate in a free-fall manner at a preset height.

5. The method for detecting the moisture content of particles based on sound according to claim 1, characterized in that: The corresponding relationship between the spectrum characteristics and the water content is established in the following way: Prepare particle samples with different water contents; allowing each particle sample to freely fall onto the surface of the plate at a preset height, and collecting an impact sound signal; Extracting the spectrum characteristics of the impact sound signal; Through machine learning algorithms, a correspondence model between spectrum characteristics and water content is established.

6. The method for detecting the moisture content of particles based on sound according to claim 1, characterized in that: The target spectrum characteristics include: Main frequency, spectral center of gravity and high frequency attenuation slope; The main frequency is the frequency value with the largest amplitude in the spectrum of the target sound signal; The spectrum center of gravity is the energy distribution center of the target sound signal spectrum; The high-frequency attenuation slope is the attenuation slope of the target sound signal spectrum in the high-frequency band.

7. The method for detecting the moisture content of particles based on sound according to claim 6, characterized in that: The target spectrum characteristics also include: spatiotemporal characteristics; The spatiotemporal features are extracted in the following way: The temporal and spatial characteristics of the target sound signal are extracted using a time series analysis method and a spatial analysis method; wherein the temporal and spatial characteristics include the propagation speed, reflection intensity and propagation path of the sound wave.

8. The method for detecting the moisture content of particles based on sound according to claim 2, characterized in that: There are multiple acoustic sensor arrays, which are respectively arranged at different positions of the board. The collecting of the impact sound signal includes: In the process of collecting the impact sound signal, a sound source localization model is constructed using a plurality of acoustic sensor arrays; Based on the sound source localization model, the actual spatial coordinates of the particle impact point are inferred using the sound wave arrival time difference algorithm; Using the sound source localization model to identify whether there are multiple particles that collide synchronously or bounce back multiple times; If it exists, the effective impact sound signal generated by the single particle impact is extracted.

9. The method for detecting the moisture content of particles based on sound according to claim 8, characterized in that: The sound source localization model is used to identify whether there are multiple particles that collide synchronously or bounce back multiple times, including: For each impact sound signal recorded by the acoustic sensor array, extract its arrival timestamp; Calculate the time difference between different acoustic sensor arrays in the same impact event; If the time differences recorded by all acoustic sensor arrays conform to the expected TDOA distribution of a single impact point, it is judged to be a single particle impact; If there is an expected TDOA distribution beyond a single impact point, it is judged to be a multi-particle synchronous impact or a multi-particle rebound impact.

10. A device for detecting the moisture content of particles based on sound, characterized in that: The detection device is applied to a detection system, wherein the detection system includes a horizontally arranged board, and the detection device includes: a delivery module, configured to control the particles to impact the surface of the plate in a free-fall manner at a preset height to generate an impact sound signal; A signal collection module, configured to collect the impact sound signal; A noise reduction module is used to perform noise reduction processing on the collected sound signals to obtain the target sound signals; A feature extraction module, configured to extract target spectrum features of the target sound signal; The water content determination module is used to compare the target spectrum characteristics with the pre-established correspondence between spectrum characteristics and water content to obtain the particle water content corresponding to the target spectrum characteristics.

Images