A method for measuring interval liquid level slope using millimeter wave radar

Abstract

The invention discloses a method for measuring the liquid level gradient of an interval by using a millimeter wave radar, which belongs to the technical field of liquid level measurement, and specifically comprises: collecting a liquid level image of a liquid level at a measuring point by using an image sensor, performing grayscale processing, calculating the grayscale sum, measuring the light intensity of the environment by a photoelectric sensor, and obtaining the corresponding grayscale sum standard from a preset light reflection comparison table; when the grayscale sum is less than the standard, no millimeter wave radar ranging is performed; if the standard is met, the millimeter wave radar is used to transmit electromagnetic waves and receive echo signals, and the turbidity of the liquid level is identified by calculating the energy of the echo signals; if the turbidity is lower than the preset turbidity standard, the millimeter wave radar transmits electromagnetic waves again, calculates the liquid level height according to the time delay of the echo signal, and then obtains the liquid level gradient between different measuring points; otherwise, continues monitoring until the conditions are met; the invention improves the accuracy, reliability and real-time performance of liquid level monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid level measurement, and in particular to a method for measuring the slope of an interval liquid level by using a millimeter wave radar. Background Art

[0002] In the field of industrial and environmental monitoring, liquid level and its changes are crucial monitoring parameters. These parameters are not only related to production efficiency, product quality and resource management, but also directly affect environmental protection and safe operation. Traditional liquid level measurement methods include float type, pressure type, ultrasonic type, etc. These methods have their own unique advantages and limitations in different application scenarios.

[0003] In recent years, with the development of millimeter-wave radar technology, its application in liquid level measurement has gradually attracted attention. Millimeter-wave radar can achieve millimeter-level measurement accuracy, meeting the needs of most industrial and environmental monitoring. Its high-precision measurement capability makes it excel in applications where liquid level height needs to be strictly controlled. Millimeter-wave radar can work stably in high temperature, high pressure, and corrosive environments, and is suitable for liquid level monitoring in open environments. Millimeter-wave radar measures liquid level height by emitting electromagnetic waves and receiving echo signals, without direct contact with the liquid, avoiding some common problems in traditional methods. This non-contact measurement method reduces equipment wear and maintenance requirements while improving measurement safety.

[0004] Although millimeter-wave radar has many advantages in liquid level measurement, it still faces some challenges in practical application. When measuring liquid level, millimeter-wave radar is easily affected by floating objects and turbidity on the liquid surface, which will affect the reflection intensity of electromagnetic waves and the measurement accuracy. Summary of the invention

[0005] The purpose of the present invention is to provide a method for measuring the slope of liquid level in an interval by using millimeter wave radar to solve the following technical problems:

[0006] When measuring liquid level, millimeter wave radar is easily affected by floating objects and turbidity on the liquid surface, which will affect the reflection intensity of electromagnetic waves and the measurement accuracy.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for measuring the slope of a liquid level in an interval using a millimeter wave radar comprises the following steps:

[0009] A plurality of measuring points are set, and a liquid surface image of the liquid surface at each measuring point is collected by an image sensor, the liquid surface image is grayed to obtain a gray image, and the grayscale sum of the gray image is calculated;

[0010] The light intensity of the environment where the measuring point is located is collected by a photoelectric sensor, and the corresponding grayscale sum standard is obtained from a preset light reflection comparison table according to the light intensity. When the grayscale sum is less than the corresponding grayscale sum standard, the millimeter wave radar ranging of the liquid surface is not performed;

[0011] When the grayscale sum is greater than or equal to the corresponding grayscale sum standard, the millimeter wave radar transmits electromagnetic waves to the liquid surface, receives the echo signal, and identifies the turbidity of the liquid surface by calculating the energy of the liquid surface echo signal;

[0012] If the turbidity level is lower than the turbidity standard, the millimeter-wave radar will transmit electromagnetic waves to the liquid surface again to calculate the liquid level according to the time delay of the echo signal and the millimeter-wave radar transmission height, and then obtain the liquid level slope according to the liquid level height difference between different measuring points; if the turbidity level is greater than or equal to the turbidity standard, monitoring will continue and the monitoring time will be recorded.

[0013] As a further solution of the present invention: the calculation process of the grayscale sum is:

[0014] Obtain a grayscale histogram of the grayscale image, and obtain the grayscale steps g in the grayscale image from the grayscale histogram i The number m, i∈[1,m], and the number of pixels n corresponding to each grayscale level i , calculate the grayscale sum G according to the value of each grayscale level and the corresponding number of pixels, .

[0015] As a further solution of the present invention: during operation, the exposure parameter settings of the image sensor remain fixed.

[0016] As a further solution of the present invention: the process of establishing the illumination reflection comparison table is as follows:

[0017] Under different light intensities, the liquid surface images of pure water samples were collected using an image sensor with the same exposure parameters, and the corresponding grayscale sum was calculated based on the liquid surface image. The mean of the grayscale sum calculated through multiple collections was set as the grayscale sum standard under the corresponding light intensity.

[0018] As a further solution of the present invention: the initial height range of the image sensor relative to the liquid surface at the measuring point is A±a, a is the error value, and the setting distance range between the container where the pure water sample is located and the image sensor is also A±a. At the farthest setting distance, the size of the container is still larger than the imaging angle of view of the image sensor.

[0019] As a further solution of the present invention: the process of the millimeter wave radar judging the turbidity of the liquid surface is:

[0020] The electromagnetic wave power emitted by the millimeter wave radar is P t , the received echo signal power is P r , then the calculation formula for the turbidity of the liquid surface is:

[0021] ;

[0022] In the formula, G o is the antenna gain, λ is the radar wavelength, σ is the radar cross section of the target, which is used to indicate the turbidity of the liquid surface, R is the estimated distance between the radar and the liquid surface, and L is the system loss.

[0023] As a further solution of the present invention: the duration during which the turbidity level of the liquid surface at each measuring point is greater than or equal to the turbidity standard is counted, the durations are sorted, and a prompt is given to repair the channel at the measuring point with the longest duration.

[0024] As a further solution of the present invention: it also includes historical data of image sensors, photoelectric sensors and millimeter wave radar measurements at statistical measurement points, wherein the historical data includes the grayscale sum of liquid surfaces with different turbidity levels under different light intensities, and deep learning model training is performed on the historical data to obtain the corresponding relationship between any grayscale sum and turbidity degree under different light intensities, and the difference between the turbidity prediction value output by the model and the actual turbidity value is evaluated through a loss function, and the model parameters are adjusted through an optimization algorithm to minimize the value of the loss function until convergence.

[0025] Beneficial effects of the present invention:

[0026] The present invention combines the data of image sensors, photoelectric sensors and millimeter-wave radars to more accurately judge the liquid level conditions in complex environments, reducing the errors that may be caused by a single sensor. First, the grayscale sum of the liquid surface image is calculated, and the degree of reflection of the liquid surface is expressed based on the grayscale sum, thereby avoiding the adverse effects of floating objects and turbidity on the measurement on the liquid surface. Then, the turbidity degree of the liquid surface is further accurately judged by analyzing the millimeter-wave radar echo energy. At the same time, the detection of liquid surface fluctuation is also realized, thereby more accurately avoiding the influence of undesirable liquid surface conditions on the measurement. When the liquid level condition reaches the standard, the liquid level height is calculated based on the time delay of the millimeter-wave radar echo signal. The present invention significantly improves the accuracy, reliability and real-time performance of liquid level monitoring, and further improves the anti-interference performance of millimeter-wave radar measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention will be further described below in conjunction with the accompanying drawings.

[0028] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] See also Figure 1 As shown, the present invention is a method for measuring the slope of the liquid level in an interval by using a millimeter wave radar, comprising the following steps:

[0031] 1. Collect the liquid surface image and calculate the total grayscale:

[0032] Several measuring points are set, and the image sensor is used to collect the liquid surface image of each measuring point. The image sensor can be installed in a suitable position to ensure that the liquid surface can be clearly captured.

[0033] The collected liquid surface image is grayed out. This process converts the color image into a grayscale image for subsequent analysis.

[0034] Calculate the grayscale sum of the grayscale image. The grayscale sum refers to the sum of the grayscale values ​​of all pixels in the grayscale image, which is used to represent the reflectivity of the liquid surface. The higher the grayscale sum, the stronger the reflectivity of the liquid surface, which may mean that the liquid surface is clearer.

[0035] 2. Collect light intensity and obtain the grayscale sum standard:

[0036] The light intensity of the environment where the measurement point is located is collected through the photoelectric sensor. The photoelectric sensor can sense the light intensity in the environment in real time.

[0037] According to the collected light intensity, the corresponding grayscale sum standard is obtained from the preset light reflection comparison table. This comparison table is established in advance through experiments or empirical data, and is used to determine a reasonable grayscale sum standard under different light conditions.

[0038] If the calculated grayscale sum is less than the corresponding grayscale sum standard, it means that the current liquid surface reflectivity is low, which may be due to turbidity of the liquid surface or other interference factors. At this time, millimeter-wave radar ranging is not performed on the liquid surface to avoid misjudgment.

[0039] 3. Determine the turbidity of the liquid surface:

[0040] When the grayscale sum is greater than or equal to the corresponding grayscale sum standard, it means that the liquid surface is relatively clear at this time, and electromagnetic waves are emitted to the liquid surface through the millimeter wave radar, and echo signals are received.

[0041] The turbidity of the liquid surface is identified by calculating the energy of the liquid surface echo signal. The energy of the echo signal reflects the reflection characteristics of the liquid surface. The higher the energy, the stronger the liquid surface reflection ability, which usually means that the liquid surface is clearer. On the contrary, it means that there may be more impurities or foam on the liquid surface, or the liquid surface is more volatile, resulting in unstable reflection signals. Therefore, the millimeter wave radar itself can be used to further accurately identify the turbidity and fluctuation of the liquid surface.

[0042] 4. Calculate the liquid level and slope:

[0043] If the turbidity of the liquid surface is lower than the preset turbidity standard, the liquid surface is considered suitable for further measurement.

[0044] The millimeter-wave radar transmits electromagnetic waves to the liquid surface again, and calculates the height of the liquid surface based on the time delay of the echo signal and the emission height of the millimeter-wave radar. The time delay refers to the time interval between the emission of the electromagnetic wave and the reception of the echo signal. Combined with the emission height of the radar, the height of the liquid surface can be accurately calculated.

[0045] The liquid surface slope is obtained based on the liquid surface height difference between different measuring points. The liquid surface slope refers to the height difference of the liquid surface between two adjacent measuring points, which reflects the degree of inclination of the liquid surface.

[0046] 5. Continuous monitoring and feedback:

[0047] If the turbidity of the liquid surface is greater than or equal to the turbidity standard, the system will continue to monitor and record the monitoring time until the conditions are met before measuring again. This ensures the accuracy and reliability of the measurement results.

[0048] Through the above steps, the present invention realizes accurate measurement of liquid level slope, and through multi-sensor fusion technology, effectively improves the anti-interference ability and measurement accuracy of the system. This method is particularly suitable for liquid level monitoring in complex environments, such as those with fluctuations, foam or suspended matter.

[0049] In a preferred embodiment of the present invention, the grayscale sum is calculated as follows:

[0050] When there are many floating objects or turbidity in the water, the reflectivity of the water will gradually decrease, and the reflectivity is directly reflected in the decrease of the gray value of the liquid surface image. Therefore, the present invention represents the reflectivity by the gray value sum. First, the gray histogram of the gray image is obtained, and the gray level g in the gray image is obtained from the gray histogram. i The number m, i∈[1,m], and the number of pixels n corresponding to each grayscale level i , calculate the grayscale sum G according to the value of each grayscale level and the corresponding number of pixels, The effect of floating objects on the liquid surface and water turbidity on the reflectivity of the water is reflected by calculating the sum of the products of all grayscale levels and the number of pixels.

[0051] In another preferred embodiment of the present invention, during operation, the exposure parameter settings of the image sensor remain fixed, thereby preventing the parameter changes from having an objective impact on the grayscale values ​​of pixels in the image.

[0052] In a preferred case of this embodiment, the process of establishing the illumination and reflection comparison table is as follows:

[0053] First, a series of liquid surface images of pure water samples are collected using an image sensor with the same exposure parameter settings under various light intensity environments. Here, "pure water" refers to an ideal water body without any impurities, suspended matter or foam, to ensure that the collected images can truly reflect the impact of light intensity on the grayscale of the liquid surface without being disturbed by changes in the water quality itself. Each time the image is collected, the current ambient light intensity value is accurately recorded. Then, for each collected liquid surface image, the system uses an image processing algorithm to calculate its grayscale sum. This step may involve converting a color image into a grayscale image, and then accumulating and summing the grayscale values ​​of all pixels to obtain the grayscale sum of the image. In order to improve the accuracy and reliability of the data, the above process needs to be repeated multiple times under each light intensity to collect enough data samples. After completing the data collection under all light intensities, for each specific light intensity, the average of the grayscale sums obtained from multiple collections is calculated. This average is regarded as the standard grayscale sum of the pure liquid surface image under this light intensity. Finally, all the light intensities and their corresponding standard grayscale sums are organized into a detailed comparison table. This table constitutes a light reflection comparison table, which provides a reliable reference for judging the light intensity of the current environment based on the grayscale sum of the liquid surface image collected in real time. The light reflection comparison table established in this way not only helps to improve the accuracy of environmental adaptability judgment before millimeter-wave radar ranging, but also lays a solid foundation for the intelligent operation of the entire system.

[0054] In another preferred case of this embodiment, the initial height range of the image sensor relative to the liquid surface at the measuring point is A±a, a is the error value, and the setting distance range between the container where the pure water sample is located and the image sensor is also A±a, so as to ensure consistency. At the same time, because in the case of field measurement, the measured liquid surface is relatively wide and occupies the entire sensor's viewing angle, so during the sample collection process, even at the farthest setting distance, the size of the container is still larger than the imaging viewing angle of the image sensor.

[0055] In another preferred embodiment of the present invention, the process of the millimeter wave radar determining the turbidity of the liquid surface is:

[0056] The millimeter wave radar emits electromagnetic waves to the liquid surface with a transmission power of P t , which is the energy actively emitted by the radar system to detect the target object. The radar receives the echo signal reflected from the liquid surface and measures its power as P r , this echo signal contains the energy information reflected from the liquid surface, and the radar equation is used to calculate the radar cross section σ of the target. The radar equation is a basic physical formula that describes the relationship between radar system parameters and target characteristics. The specific formula is as follows:

[0057] ;

[0058] In the formula, G o is the antenna gain, λ is the radar wavelength, R is the estimated distance between the radar and the liquid surface, and L is the system loss.

[0059] The radar cross section σ reflects the reflection characteristics of the liquid surface. For a clear liquid surface, the scattering cross section is small; while for a turbid liquid surface containing impurities, foam or suspended matter, the scattering cross section is large. Therefore, the calculated σ value can effectively evaluate the turbidity of the liquid surface. If the calculated radar cross section σ is lower than the preset turbidity standard, it means that the liquid surface is relatively clear and suitable for further millimeter wave radar ranging. If the radar cross section σ is higher than the turbidity standard, it means that there are more impurities, foam or fluctuations on the liquid surface, which is not suitable for immediate ranging, and monitoring will continue until conditions improve.

[0060] In another preferred embodiment of the present invention, the duration of the turbidity of the liquid surface at each measuring point being greater than or equal to the turbidity standard is counted, and for each measuring point exceeding the turbidity standard, the system records its duration. The duration of the excess of the standard for all measuring points is collected and stored in a database or data structure. The system sorts according to these durations to determine which measuring points have a high turbidity level for a long time. Based on the sorting results, the system automatically generates a report or prompt indicating which measuring points have channels with more serious turbidity problems. In particular, for those measuring points with the longest excess time, the system will give clear prompts and recommend that these channels be inspected and repaired first.

[0061] In another preferred embodiment of the present invention, not only the liquid level state is monitored in real time, but also a deep learning model is trained on historical data, so that the turbidity of the liquid surface under different lighting conditions can be more accurately predicted. The training process is:

[0062] 1. Data Collection and Storage:

[0063] The image sensor, photoelectric sensor and millimeter wave radar measurement data of each measurement point are continuously collected. These data include key parameters such as the total grayscale of the liquid surface, the energy of the echo signal, and the light intensity.

[0064] All collected data are stored in a central database for subsequent analysis and processing.

[0065] 2. Historical data collation:

[0066] Organize and preprocess the stored historical data to ensure data quality and consistency. This includes removing outliers, filling in missing data, and standardizing data formats from different sources.

[0067] Special attention is paid to the sum of grayscale values ​​of liquid surfaces with different turbidity levels under different light intensities, and these data are used as key input features for training deep learning models.

[0068] 3. Selection and training of deep learning models:

[0069] Select a suitable deep learning model architecture, such as a convolutional neural network (CNN) or a recurrent neural network (RNN), to establish a mapping relationship from the total grayscale and light intensity to the turbidity of the liquid surface.

[0070] The selected model is trained using historical data. During the training process, the model learns how to predict the turbidity of the liquid surface based on the input grayscale sum and light intensity.

[0071] 4. Loss function and optimization:

[0072] Define a suitable loss function to measure the difference between the predicted value of the turbidity output by the model and the actual value. Commonly used loss functions include mean square error (MSE) or cross entropy loss.

[0073] The gradient of the loss function with respect to the model parameters is calculated through the back-propagation algorithm, and the model parameters are adjusted using an optimization algorithm (such as stochastic gradient descent, Adam, etc.) to minimize the value of the loss function.

[0074] The training process is repeated until the value of the loss function no longer decreases significantly or reaches the preset training rounds, at which point the model is considered to have converged.

[0075] 5. Model evaluation and validation:

[0076] Use a portion of the reserved historical data as a validation set to evaluate the trained model. Calculate the performance indicators of the model on the validation set, such as accuracy, recall, F1 score, etc.

[0077] If the model performance does not meet expectations, you can consider adjusting the model architecture, adding regularization terms, changing the optimization algorithm or parameters, etc. to further improve the model performance.

[0078] 6. Real-time prediction and application:

[0079] Once the model is trained and validated, it can be deployed in the actual system for real-time prediction. Whenever new measurement data is available, the system uses the trained model to predict the turbidity of the liquid surface.

[0080] Based on the prediction results, the system can automatically adjust the monitoring strategy, issue warnings, or prompt managers to take corresponding actions.

[0081] 7. Continuous learning and updating:

[0082] In order to cope with environmental changes and drift in data distribution, the system also supports online learning and model update functions. As new data continues to accumulate, the system can regularly retrain and optimize the model to ensure that its predictive performance always remains at a high level.

[0083] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for measuring the slope of a liquid level in an interval using a millimeter wave radar, characterized in that: The following steps are involved: A plurality of measuring points are set, and a liquid surface image of the liquid surface at each measuring point is collected by an image sensor, the liquid surface image is grayed to obtain a gray image, and the grayscale sum of the gray image is calculated; The light intensity of the environment where the measuring point is located is collected by a photoelectric sensor, and the corresponding grayscale sum standard is obtained from a preset light reflection comparison table according to the light intensity. When the grayscale sum is less than the corresponding grayscale sum standard, the millimeter wave radar ranging of the liquid surface is not performed; When the grayscale sum is greater than or equal to the corresponding grayscale sum standard, the millimeter wave radar transmits electromagnetic waves to the liquid surface, receives the echo signal, and identifies the turbidity of the liquid surface by calculating the energy of the liquid surface echo signal; If the turbidity level is lower than the turbidity standard, the millimeter-wave radar will transmit electromagnetic waves to the liquid surface again to calculate the liquid level according to the time delay of the echo signal and the millimeter-wave radar transmission height, and then obtain the liquid level slope according to the liquid level height difference between different measuring points; if the turbidity level is greater than or equal to the turbidity standard, monitoring will continue and the monitoring time will be recorded.

2. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 1, characterized in that: The calculation process of the grayscale sum is: Obtain a grayscale histogram of the grayscale image, and obtain the grayscale steps g in the grayscale image from the grayscale histogram i The number m, i∈[1,m], and the number of pixels n corresponding to each grayscale level i , calculate the grayscale sum G according to the value of each grayscale level and the corresponding number of pixels, .

3. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 1, characterized in that: During operation, the exposure parameter settings of the image sensor remain fixed.

4. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 3, characterized in that: The process of establishing the illumination reflection comparison table is as follows: Under different light intensities, the liquid surface images of pure water samples were collected using an image sensor with the same exposure parameters, and the corresponding grayscale sum was calculated based on the liquid surface image. The average of the grayscale sums calculated from multiple collections was set as the grayscale sum standard under the corresponding light intensity.

5. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 4, characterized in that: The initial height range of the image sensor relative to the liquid surface at the measuring point is A±a, where a is the error value. The setting distance range between the container where the pure water sample is located and the image sensor is also A±a. At the farthest setting distance, the size of the container is still larger than the imaging angle of view of the image sensor.

6. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 1, characterized in that: The process of the millimeter wave radar judging the turbidity of the liquid surface is as follows: The electromagnetic wave power emitted by the millimeter wave radar is P t , the received echo signal power is P r , then the calculation formula for the turbidity of the liquid surface is: ； In the formula, G o is the antenna gain, λ is the radar wavelength, σ is the radar cross section of the target, which is used to indicate the turbidity of the liquid surface, R is the estimated distance between the radar and the liquid surface, and L is the system loss.

7. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 1, characterized in that: The duration during which the turbidity of the liquid surface at each measuring point is greater than or equal to the turbidity standard is counted, the durations are sorted, and a prompt is given to repair the channel at the measuring point with the longest duration.

8. The method for measuring the liquid level slope of an interval using millimeter wave radar according to claim 1, characterized in that: It also includes historical data of image sensors, photoelectric sensors and millimeter-wave radar measurements at statistical measurement points, wherein the historical data includes the grayscale sum of liquid surfaces with different turbidity levels under different light intensities, and a deep learning model is trained on the historical data to obtain the corresponding relationship between any grayscale sum and the turbidity level under different light intensities. The difference between the turbidity prediction value output by the model and the actual turbidity value is evaluated through a loss function, and the model parameters are adjusted through an optimization algorithm to minimize the value of the loss function until convergence.

Images