System and method for measuring material level of stock bin based on MIMO array millimeter wave radar

Abstract

The invention discloses a stock bin material level measuring system and method based on an MIMO array millimeter wave radar, the stock bin material level measuring system based on the MIMO array millimeter wave radar comprises a group of millimeter wave radar sensing units, a communication module and an industrial control server which are interconnected, and the method adopts the MIMO array millimeter wave radar. The detection blind area limitation of a traditional narrow-beam radar is broken through, all-directional coverage measurement of the material surface is achieved, strong reflection sources such as bin wall corner reflection and a reinforcing cross beam are effectively restrained based on an interference map, weak signals are enhanced in combination with the dust environment self-adaptive compensation and directional focusing technology, effective material sites are accurately extracted through a statistical screening mechanism, and the detection accuracy is improved. And multi-path interference is eliminated, and finally, miniaturized and low-cost robust material level measurement is realized.

Description

Technical Field

[0001] This invention relates to millimeter-wave radar technology and industrial level sensing unit technology, specifically a silo level measurement system and method based on MIMO array millimeter-wave radar. Background Technology

[0002] Level measurement refers to measuring irregularly stacked materials to obtain real-time information such as material height and distribution characteristics. This information is used to estimate the volume or weight of the material stored in a container, playing a crucial role in production automation. As a highly efficient detection method, it is applied in various industrial environments, such as grain storage, feed storage, and the storage of materials like plastics, quartz sand, and rubber granules. Currently, level measurement products on the market are mainly divided into contact and non-contact types. Contact level gauges include contact level gauges, capacitive level gauges, and anti-rotation level gauges, while non-contact level gauges include radar level gauges, ultrasonic level gauges, and laser level gauges. Compared to contact level gauges, non-contact level gauges have advantages such as less susceptibility to environmental influences, higher accuracy, higher precision, and easier maintenance.

[0003] Millimeter-wave radar level gauges are non-contact level gauges, typically operating at frequencies as high as 60GHz or 77GHz. They possess strong penetration capabilities and are unaffected by environmental factors such as dust and steam. The interior of a silo is usually a small, enclosed environment, making it impossible to avoid the introduction of strong multipath clutter during detection. To mitigate the impact of multipath clutter, similar products on the market use narrow-beam antennas, reducing its physical influence. However, narrow-beam antennas present the following problems in practical applications:

[0004] 1) It can only measure the material surface condition in a small area, and the data representativeness is poor;

[0005] 2) Poor stability when continuously measuring the dynamic, random, rough material surface of the silo;

[0006] 3) It requires a lens antenna or horn antenna, which is not conducive to miniaturization and cost reduction. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of existing technologies by providing a silo level measurement system and method based on MIMO array millimeter-wave radar. This system is highly practical and easy to network, and this method can solve the problems of narrow-beam millimeter-wave radar level gauges, improving the robustness and accuracy of level measurement.

[0008] The technical solution to achieve the objective of this invention is:

[0009] A silo level measurement system based on MIMO array millimeter-wave radar includes an interconnected set of millimeter-wave radar sensing units, a communication module, and an industrial control server. Each millimeter-wave radar sensing unit is used to measure the level value. The communication module is used to establish a connection between the server and multiple millimeter-wave radar sensing units, transmit control commands and level measurement data, and the industrial control server is used to configure the operating parameters of the millimeter-wave radar sensing units, perform fusion analysis on multi-source measurement data, and implement privacy protection using a hardware-encrypted data security protocol.

[0010] A method for measuring silo level based on MIMO array millimeter-wave radar, the method employing the aforementioned silo level measurement system based on MIMO array millimeter-wave radar, comprising the following steps:

[0011] S1. Millimeter-wave radar is used to collect multiple frames of raw echo signals from the silo, the echo signals including material surface echo signals and interference noise signals;

[0012] S2. The spatial dimension data of the original echo signal is calibrated using a calibration sequence to ensure consistency across channels. A fast Fourier transform is then performed on the distance dimension to obtain the frequency domain echo signal (including material surface echo components and interference clutter components).

[0013] S3. Perform interference suppression processing on the frequency domain echo signal to suppress its clutter interference components;

[0014] S4. Perform loss compensation and directional focusing on the frequency domain echo signal (including material surface echo component and a small amount of interference clutter component) after interference suppression to enhance the weak signal;

[0015] S5. Peak detection is performed on the frequency domain echo signal (including material surface echo component and a small amount of interference clutter component) after interference suppression and enhancement to obtain the detection set. The effective material point is obtained by using the material point extraction method, and the true material level value is calculated. The process includes:

[0016] S501. Using the CFAR algorithm to analyze the range image Peak detection is performed to obtain a detection set that includes both valid and spurious sites. This includes the sampling point index and amplitude.

[0017] S502. Based on the material site extraction method, extract the valid material sites from the detection set, calculate the true material site distance based on the material site, and determine whether it is a valid material site by comparing it with the mean of the detection sequence. Valid material sites should meet the following requirements:

[0018] ,

[0019] In the formula, To detect the site sites in the sequence, To detect the total number of sequence sites, To detect the standard deviation of all sites in the sequence, Using the threshold factor, the sampling point index value of the effective material point is converted into a distance, and the material level measurement value can be obtained.

[0020] The millimeter-wave radar described in step S1 operates in a three-transmit, four-receive mode to collect multiple frames of echo data. Each frame of echo signal data has three dimensions: range dimension, Doppler dimension, and spatial dimension. The spatial dimension is composed of 12 virtual channels formed by the transmitting and receiving antennas.

[0021] The calibration process described in step S2 includes:

[0022] S201. Establish calibration sequence: Place the radar target at a preset distance, aligned with the center of the radar antenna in both the horizontal and vertical directions, collect echo data to create frequency calibration sequence and amplitude-phase calibration sequence. The frequency calibration sequence and amplitude-phase calibration sequence are the sampling point index value and complex signal value at the radar target in different virtual channels, respectively.

[0023] S202. Multiply the data from each channel of the echo signal by the compensation value of the frequency calibration sequence to perform frequency calibration:

[0024] ,

[0025] In the formula, Indicates the echo signal. For virtual channel numbers, where the frequency sequence is... The compensation value is expressed as:

[0026] ,

[0027] In the formula, Indicates the reference channel. and These represent the sampling rate and frequency modulation slope of the calibration data, respectively. and The sampling rate and frequency modulation slope of the data to be calibrated. Let be the sampling point vector of the data to be calibrated. The number of sampling points for the data to be calibrated. The number of sampling points for calibration data. This is the relative sampling magnification factor;

[0028] S203. Multiply the frequency-calibrated data by the amplitude-phase sequence compensation value to perform amplitude-phase calibration:

[0029] .

[0030] The interference suppression process for the echo signal described in step S3 includes:

[0031] S301. During the empty hopper calibration phase, an interference map is established. The interference map is a two-dimensional matrix of distance-amplitude, recording the strong reflection characteristics of the hopper wall edges, connections, hopper corner reflections, reinforced crossbeams, and close-range coupled wave characteristics.

[0032] S302. Apply attenuation suppression of more than 30dB to known strong reflection locations using an interference map.

[0033] The enhancement of weak signals mentioned in step S4 includes:

[0034] S401. Based on the feed properties and the internal environment of the warehouse, i.e., dust density, a loss compensation coefficient is set to compensate for the loss of the echo signal. The loss compensation coefficient is expressed as:

[0035] ,

[0036] In the formula, The distance between the radar and the material surface. Where is the particle radius of the material. The dielectric constant of the material. Dust density;

[0037] S402. Beam-weighted accumulation is used to cover the angular range of the material surface for directional focusing. The beam-weighting process is represented as follows:

[0038] ,

[0039] In the formula, Doppler index value, For the number of snapshots, For beam weights, The echo signal after loss compensation. Given the pointing angle, the weighted and accumulated distance image is represented as follows:

[0040] ,

[0041] In the formula, For frame index, For the number of frames captured, Accumulate the angle range for beam weighting.

[0042]

[0043] In the formula, To detect the site sites in the sequence, To detect the total number of sequence sites, To detect the standard deviation of all sites in the sequence, This is the threshold factor.

[0044] This technical solution employs a MIMO array millimeter-wave radar, overcoming the detection blind zone limitations of traditional narrow-beam radar to achieve omnidirectional coverage measurement of material surfaces. Based on the interference map, it effectively suppresses strong reflection sources such as silo wall corner reflections and reinforced beams. Combined with dust environment adaptive compensation and directional focusing technology, it enhances weak signals and uses a statistical screening mechanism to accurately extract effective material points, eliminating multipath interference. Ultimately, it achieves miniaturized, low-cost, and robust level measurement.

[0045] This system is highly practical and easy to network. This method can solve the problems of narrow-beam millimeter-wave radar level gauges and improve the robustness and accuracy of level measurement. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating the method used in the embodiment.

[0047] Figure 2 This is a schematic diagram of the system structure in the example embodiment. Detailed Implementation

[0048] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this is not intended to limit the scope of the invention.

[0049] Example:

[0050] Reference Figure 2 A silo level measurement system based on MIMO array millimeter-wave radar includes an interconnected set of millimeter-wave radar sensing units, a communication module, and an industrial control server. Each millimeter-wave radar sensing unit is used to measure the level value. The communication module is used to establish a connection between the server and multiple millimeter-wave radar sensing units, transmit control commands and level measurement data, and the industrial control server is used to configure the operating parameters of the millimeter-wave radar sensing units, perform fusion analysis on multi-source measurement data, and implement privacy protection using a hardware-encrypted data security protocol.

[0051] In this example, the millimeter-wave radar sensing unit operates at a frequency of 77-81 GHz, using a three-transmit, four-receive MIMO mode to transmit and receive radar signals, and processes the signals to obtain the level measurement value. The communication module enables bidirectional data interaction between the industrial control server and multiple millimeter-wave radar sensing units, transmitting control commands and level measurement results. It supports industrial-grade wired (RS-485 protocol) and wireless (LTE Cat.1 / 4G protocol) dual-mode communication, and the transmission mode can be flexibly configured according to the factory's on-site deployment conditions. The industrial control server acts as the system control hub, achieving collaborative management of multiple devices through real-time dynamic configuration of each millimeter-wave radar sensing unit. At the same time, it integrates measurement data from multiple sensing units to monitor the status of multiple compartments in real time. With the help of the engine (supporting the SM4 algorithm), an end-to-end security architecture is built, implementing double encryption and dynamic key management for the transmitted control commands and level data, meeting the industrial IoT privacy protection requirements specified in GB / T39276-2020 standard.

[0052] A method for measuring silo level based on MIMO array millimeter-wave radar, the method employing the aforementioned MIMO array millimeter-wave radar-based silo level measurement system, includes the following steps:

[0053] S1. Millimeter-wave radar is used to collect multiple frames of raw echo signals from the silo, the echo signals including material surface echo signals and interference noise signals;

[0054] S2. The spatial dimension data of the original echo signal is calibrated using a calibration sequence to ensure consistency across channels. A fast Fourier transform is then performed on the distance dimension to obtain the frequency domain echo signal (including material surface echo components and interference clutter components).

[0055] S3. Perform interference suppression processing on the frequency domain echo signal to suppress its clutter interference components;

[0056] S4. Perform loss compensation and directional focusing on the frequency domain echo signal (including material surface echo component and a small amount of interference clutter component) after interference suppression to enhance the weak signal;

[0057] S5. Peak detection is performed on the frequency domain echo signal (including material surface echo component and a small amount of interference clutter component) after interference suppression and enhancement to obtain the detection set. The effective material point is obtained by using the material point extraction method, and the true material level value is calculated. The process includes:

[0058] S501. Using the CFAR algorithm to analyze the range image Peak detection is performed to obtain a detection set that includes both valid and spurious sites. This includes the sampling point index and amplitude.

[0059] S502. Based on the material site extraction method, extract the valid material sites from the detection set, calculate the true material site distance based on the material site, and determine whether it is a valid material site by comparing it with the mean of the detection sequence. Valid material sites should meet the following requirements:

[0060] ,

[0061] In the formula, To detect the site sites in the sequence, To detect the total number of sequence sites, To detect the standard deviation of all sites in the sequence, Using the threshold factor, the sampling point index value of the effective material point is converted into a distance, and the material level measurement value can be obtained.

[0062] In this example, millimeter-wave radar is used to acquire multiple frames of raw echo signals from the silo. These raw echo signals include material surface echo signals and interference noise signals. Each frame of echo signal data has three dimensions: range, Doppler, and spatial. The spatial dimension is composed of 12 virtual channels formed by the transmitting and receiving antennas. A single frame of raw echo signal from the silo is represented as follows:

[0063] ,

[0064] In the formula, It is a sequence of sampling points. The distance between the target and the radar. At the speed of light, Sampling frequency, The relative speed between the target and the radar. The phase difference between each virtual channel. The relative half-wavelength distance between virtual channels. The angle of incidence, For noise interference, For clutter interference, the echo signal contains amplitude, phase, and frequency information;

[0065] The millimeter-wave radar described in step S1 operates in a three-transmit, four-receive mode, acquiring multiple frames of echo data. Each frame of echo signal data has three dimensions: range, Doppler, and spatial. The spatial dimension is composed of 12 virtual channels formed by the transmitting and receiving antennas. In this example, before the millimeter-wave radar sensing unit is installed in the silo, factory calibration is performed under a standard test environment. A 10cm diameter metal sphere target is placed at a preset position 5m away from the radar, ensuring that the center of the target is perfectly aligned with the center of the radar antenna array in both the horizontal and vertical directions. Ten frames of standard echo data are acquired, and frequency calibration sequences are created for each frame. With amplitude and phase calibration sequence ;

[0066] The calibration process described in step S2 includes:

[0067] S201. Establish calibration sequence: Place the radar target at a preset distance, aligned with the center of the radar antenna in both the horizontal and vertical directions, collect echo data to create frequency calibration sequence and amplitude-phase calibration sequence. The frequency calibration sequence and amplitude-phase calibration sequence are the sampling point index value and complex signal value at the radar target in different virtual channels, respectively.

[0068] S202. Multiply the data from each channel of the echo data by the compensation value of the frequency calibration sequence to perform frequency calibration:

[0069] ,

[0070] In the formula, Indicates the echo signal. For virtual channel numbers, where the frequency sequence is... The compensation value is expressed as:

[0071] ,

[0072] In the formula, Indicates the reference channel. and These represent the sampling rate and frequency modulation slope of the calibration data, respectively. and The sampling rate and frequency modulation slope of the data to be calibrated. Let be the sampling point vector of the data to be calibrated. The number of sampling points for the data to be calibrated. The number of sampling points for calibration data. This is the relative sampling magnification factor;

[0073] S203. Multiply the frequency-calibrated data by the amplitude-phase sequence compensation value to perform amplitude-phase calibration:

[0074] ,

[0075] The interference suppression process for the echo signal described in step S3 includes:

[0076] S301. During the empty hopper calibration phase, an interference map is established. The interference map is a two-dimensional matrix of distance-amplitude, recording the strong reflection characteristics of the hopper wall edges, connections, hopper corner reflections, reinforced crossbeams, and close-range coupled wave characteristics.

[0077] S302. Apply attenuation suppression of more than 30dB to known strong reflection locations using an interference map;

[0078] The enhancement of weak signals mentioned in step S4 includes:

[0079] S401. Based on the feed properties and the internal environment of the warehouse, i.e., dust density, a loss compensation coefficient is set to compensate for the loss of the echo signal. The loss compensation coefficient is expressed as:

[0080] ,

[0081] In the formula, The distance between the radar and the material surface. Where is the particle radius of the material. The dielectric constant of the material. Dust density;

[0082] S402. Beam-weighted accumulation is used to cover the angular range of the material surface for directional focusing. The beam-weighting process is represented as follows:

[0083] ,

[0084] In the formula, Doppler index value, For the number of snapshots, For beam weights, The echo signal after loss compensation. Given the pointing angle, the weighted and accumulated distance image is represented as follows:

[0085] ,

[0086] In the formula, For frame index, For the number of frames captured, Accumulate the angle range for beam weighting.

[0087] ,

[0088] In the formula, To detect the site sites in the sequence, To detect the total number of sequence sites, To detect the standard deviation of all sites in the sequence, This is the threshold factor.

Claims

1. A silo level measurement system based on MIMO array millimeter-wave radar, characterized in that, It includes an interconnected set of millimeter-wave radar sensing units, a communication module, and an industrial control server. Each millimeter-wave radar sensing unit is used to measure the level value. The communication module is used to establish a connection between the server and multiple millimeter-wave radar sensing units, transmit control commands and level measurement data, and the industrial control server is used to configure the operating parameters of the millimeter-wave radar sensing units, perform fusion analysis on multi-source measurement data, and use a hardware-encrypted data security protocol to achieve privacy protection.

2. A method for measuring silo level based on MIMO array millimeter-wave radar, wherein the method employs the silo level measurement system based on MIMO array millimeter-wave radar as described in claim 1, characterized in that, Including the following steps: S1. Millimeter-wave radar is used to collect multiple frames of raw echo signals from the silo, the raw echo signals including material surface echo signals and interference noise signals; S2. The spatial dimension data of the original echo signal is calibrated using a calibration sequence to ensure consistency across channels, and a fast Fourier transform is performed on the distance dimension. S3. Perform interference suppression processing on the frequency domain echo signal; S4. Perform loss compensation and directional focusing on the frequency domain echo signal after interference suppression to enhance the weak signal. The frequency domain echo signal after interference suppression contains material surface echo components and a small amount of interference noise components. S5. Peak detection is performed on the frequency domain echo signal after interference suppression and enhancement to obtain a detection set. The effective material level points are obtained using a material level extraction method, and the true material level value is calculated. The process includes: S501. Using the CFAR algorithm to analyze the range image Peak detection is performed to obtain a detection set that includes both valid and spurious material sites. This includes the sampling point index and amplitude. S502. Based on the material site extraction method, extract the valid material sites from the detection set, calculate the true material site distance based on the material site, and determine whether it is a valid material site by comparing it with the mean of the detection sequence. Valid material sites should meet the following requirements: ， In the formula, To detect the site sites in the sequence, To detect the total number of sequence sites, To detect the standard deviation of all sites in the sequence, Using the threshold factor, the sampling point index value of the effective material point is converted into a distance, and the material level measurement value can be obtained.

3. The method for measuring silo level based on MIMO array millimeter-wave radar according to claim 2, characterized in that, The millimeter-wave radar described in step S1 operates in a three-transmit, four-receive mode to collect multiple frames of echo data. Each frame of echo signal data has three dimensions: range dimension, Doppler dimension, and spatial dimension. The spatial dimension is composed of 12 virtual channels formed by the transmitting and receiving antennas.

4. The method for measuring silo level based on MIMO array millimeter-wave radar according to claim 2, characterized in that, The calibration process described in step S2 includes: S201. Establish calibration sequence: Place the radar target at a preset distance, aligned with the center of the radar antenna in both the horizontal and vertical directions, collect echo data to create frequency calibration sequence and amplitude-phase calibration sequence. The frequency calibration sequence and amplitude-phase calibration sequence are the sampling point index value and complex signal value at the radar target in different virtual channels, respectively. S202. Multiply the data from each channel of the echo data by the compensation value of the frequency calibration sequence to perform frequency calibration: ， In the formula, Indicates the echo signal. For virtual channel numbers, where the frequency sequence is... The compensation value is expressed as: ， In the formula, Indicates the reference channel. and These represent the sampling rate and frequency modulation slope of the calibration data, respectively. and The sampling rate and frequency modulation slope of the data to be calibrated. Let be the sampling point vector of the data to be calibrated. The number of sampling points for the data to be calibrated. The number of sampling points for calibration data. This is the relative sampling magnification factor; S203. Multiply the frequency-calibrated data by the amplitude-phase sequence compensation value to perform amplitude-phase calibration: 。 5. The method for measuring silo level based on MIMO array millimeter-wave radar according to claim 2, characterized in that, The interference suppression process for the echo signal described in step S3 includes: S301. During the empty hopper calibration phase, an interference map is established. The interference map is a two-dimensional matrix of distance-amplitude, recording the strong reflection characteristics of the hopper wall edges, connections, hopper corner reflections, reinforced crossbeams, and close-range coupled wave characteristics. S302. Apply attenuation suppression of more than 30dB to known strong reflection locations using an interference map.

6. The method for measuring silo level based on MIMO array millimeter-wave radar according to claim 2, characterized in that, The enhancement of weak signals mentioned in step S4 includes: S401. Based on the feed properties and the internal environment of the warehouse, i.e., dust density, a loss compensation coefficient is set to compensate for the loss of the echo signal. The loss compensation coefficient is expressed as: ， In the formula, The distance between the radar and the material surface. Where is the particle radius of the material. The dielectric constant of the material. Dust density; S402. Beam-weighted accumulation is used to cover the angular range of the material surface for directional focusing. The beam-weighting process is represented as follows: ， In the formula, Doppler index value, For the number of snapshots, For beam weights, The echo signal after loss compensation. Given the pointing angle, the weighted and accumulated distance image is represented as follows: ， In the formula, For frame index, For the number of frames captured, Accumulate the angle range for beam weighting; In the formula, To detect the site sites in the sequence, To detect the total number of sequence sites, To detect the standard deviation of all sites in the sequence, This is the threshold factor.

Images