Self-adaptive data noise reduction method based on environmental characteristics and applied to chemical industry park

Abstract

The invention relates to the technical field of environmental monitoring, and discloses a self-adaptive data noise reduction method based on environmental characteristics and applied to a chemical industry park. The method comprises the following steps: S1, detecting environmental characteristics of the static pressure water level gauge, and starting a cleaning mode of the static pressure water level gauge based on the environmental characteristics; s2, acquiring water level information based on a static pressure water level gauge, and acquiring turbidity information based on a turbidity sensor; s3, the turbidity information is corrected based on historical tide data, and the historical tide data comprise historical water level information and historical turbidity information. The method has the advantages that the method adapts to complex tidal modes and environmental changes, the accuracy and reliability of turbidity monitoring are remarkably improved, and meanwhile certain anomaly detection and self-adaptive capacity is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of environmental monitoring, and in particular to an adaptive data denoising method based on environmental characteristics applied to a chemical park. Background Art

[0002] In chemical parks, real-time monitoring of water quality and gas is a core requirement for ensuring environmental safety and compliance with factory operations. Among them, water quality monitoring, as a key link, directly affects the management effect of emission control, pollution control and ecological environment protection in the park. In Binhai Chemical Park, turbidity is one of the important indicators of water quality, and its monitoring is of great significance for reflecting the concentration of suspended particulate matter in water bodies and potential pollution risks.

[0003] Changes in the concentration of suspended particulate matter in water bodies may be caused by a variety of factors, including the combined effects of natural environmental changes and human activities. For example, in coastal chemical parks, if there are accidents such as tank leaks, pipeline ruptures, or rainwater erosion of emissions, the concentration of suspended particulate matter in the water body may increase significantly. These accidents are usually accompanied by the influx of large amounts of sediment, organic matter, or chemical particles, which rapidly deteriorate the water quality. In addition, potential pollutants (such as untreated wastewater discharged from chemical production processes) may directly affect turbidity levels. These wastewaters may contain a large amount of undecomposed particulate matter, toxic heavy metal ions, or oily substances, which will not only lead to increased turbidity, but may also pose a serious threat to the water body's ecosystem and downstream water safety.

[0004] If turbidity changes are not monitored and effectively controlled in a timely manner, a series of environmental and safety problems may arise. For example, high turbidity waters will significantly reduce the survival capacity of aquatic plants and animals and destroy the local ecological balance; suspended particles may absorb pollutants and become a carrier for the spread of pollutants, aggravating environmental pollution in downstream areas; more seriously, these pollutions may endanger the safety of residents' drinking water and even cause a public health crisis. In addition, abnormal turbidity may also mask the presence of other more dangerous pollutants in the water (such as chemical toxins or highly toxic gases), resulting in pollution incidents not being identified and handled in a timely manner.

[0005] The special geographical environment of the Binhai Chemical Park further exacerbates the complexity of monitoring. For example, the flow of water caused by tidal phenomena can significantly increase the concentration of suspended particulate matter, causing abnormally high readings of turbidity sensors. Such abnormal readings may not only cause false alarms and waste valuable emergency resources, but may also cover up more serious pollution problems. In addition, if water level measurement equipment is used to detect water levels in real time (such as static pressure water level sensors), suspended matter and dissolved matter in tidal movements may also form dirt on the detection part of the water level measurement equipment, affecting the measurement accuracy of the equipment and resulting in inaccurate water level data. Summary of the invention

[0006] In order to adapt to complex tidal patterns and environmental changes, significantly improve the accuracy and reliability of turbidity monitoring, and at the same time have certain anomaly detection and adaptive capabilities, the present application provides an adaptive data denoising method based on environmental characteristics applied to chemical parks.

[0007] In the first aspect, the present application provides an adaptive data denoising method based on environmental characteristics applied to a chemical park, which adopts the following technical solution:

[0008] An adaptive data denoising method based on environmental characteristics applied to a chemical park comprises the following steps:

[0009] S1. Detect the environmental characteristics of the static pressure water level gauge and start the cleaning mode of the static pressure water level gauge based on the environmental characteristics;

[0010] S2. Obtain the current water level fluctuation range information based on the static pressure water level gauge and obtain turbidity information based on the turbidity sensor;

[0011] S3. Correcting the turbidity information based on historical tidal data, wherein the historical tidal data includes historical water level information and historical turbidity information.

[0012] Optionally, the static pressure water level gauge body includes an outer shell, a piezoelectric ceramic and a sensor arranged inside the outer shell, the outer shell is columnar, and a connecting port is opened on the bottom surface of the outer shell, and the connecting port is used for external fluid to enter so as to contact the sensor inside the outer shell; the piezoelectric ceramic is installed inside the static pressure water level gauge body, and the top of the static pressure water level gauge is connected to an external host through an armored steel cable, and the piezoelectric ceramic is used to output ultrasonic vibrations to the static pressure water level gauge body.

[0013] Optionally, the static pressure water level gauge body further includes an ultrasonic probe and a convex ring located in the communicating port, the ultrasonic probe is located in the outer shell, and the ultrasonic probe is used to detect an ultrasonic detection signal of the convex ring.

[0014] Optionally, the cross section of the outer shell is circular and gradually becomes thicker from both ends to the middle.

[0015] Optionally, S1 includes the following sub-steps:

[0016] S11. Control the piezoelectric ceramic to emit an ultrasonic measurement signal;

[0017] S12. Acquire an ultrasonic detection signal based on an ultrasonic probe, and perform dimension expansion on the ultrasonic detection signal; wherein the waveform of the ultrasonic detection signal is generated by the physical properties of the ultrasonic emission waveform and the interface, and carries relevant effective information related to the measured medium and the interface;

[0018] S13. Perform noise reduction processing on the ultrasonic detection signal;

[0019] S14. Perform non-local mean filtering;

[0020] S15. Input the ultrasonic detection signal after non-local mean filtering into the pre-trained neural network model to obtain the dirt thickness; wherein the pre-trained neural network model is trained with the echo time and echo amplitude of the pre-calibrated ultrasonic reflection echo waveform as input and the calibrated dirt thickness as output, wherein the outer diameter, inner diameter and average wall thickness of the pipe are determined, the composition of the seawater is determined, and the waveform of the ultrasonic reflection echo includes the echo peak of the outer wall of the pipe, the secondary echo of the outer wall of the pipe, the echo of the inner wall of the pipe, the secondary echo of the inner wall of the pipe and the tertiary echo of the inner wall of the pipe;

[0021] S16. Determine whether the dirt thickness is greater than a preset threshold. If so, activate the piezoelectric ceramic to emit ultrasonic vibration for cleaning.

[0022] Optionally, the S12 includes the following steps:

[0023] S121. Assume that the ultrasonic detection signal is ,in, , , , ;

[0024] S122. One-dimensional ultrasonic detection signal Do dimensional expansion and add additional noise to the original A random signal with a constant power spectral density and a virtual observation signal To assist in the analysis, , i=2, 3, ..., M; For Random signals of the same distribution; Construct the virtual observation matrix P, , .

[0025] The S13 comprises the following steps:

[0026] S131. Decompose the virtual observation matrix P to obtain a square matrix ;

[0027] S132. Square Matrix Perform a two-dimensional wave atom transformation and obtain the coefficient matrix ;

[0028] S133. Coefficient Matrix The coefficients in are threshold filtered to obtain the coefficient matrix after noise reduction. ;

[0029] S134. The coefficient matrix after noise reduction Perform a two-dimensional wave atomic inverse transform to obtain the filtered ;

[0030] S135. The result matrix after denoising The one-dimensional noise reduction result of the signal is obtained by accumulating and averaging in the longitudinal direction. .

[0031] Optionally, S3 includes the following steps:

[0032] S31. Determine the corresponding water level range in the historical tidal data according to the current water level fluctuation range;

[0033] S32. Extracting historical turbidity information corresponding to the water level range from the historical tidal database;

[0034] S33. Combined with historical tidal data, analyze the typical impact pattern of tidal fluctuations on turbidity;

[0035] S34. Select a correction coefficient based on a typical impact mode, and calculate the corrected turbidity based on the correction coefficient and the real-time turbidity.

[0036] In the second aspect, the present application provides an adaptive data denoising method based on environmental characteristics applied to a chemical park, which adopts the following technical solution:

[0037] An adaptive data denoising device based on environmental characteristics applied to a chemical park, comprising:

[0038] A cleaning module, used to detect the environmental characteristics of the static pressure water level gauge and start the cleaning mode of the static pressure water level gauge based on the environmental characteristics;

[0039] A data acquisition module, used to acquire water level information based on a static pressure water level gauge and to acquire turbidity information based on a turbidity sensor;

[0040] The turbidity correction module is used to correct the turbidity information based on historical tidal data, wherein the historical tidal data includes historical water level information and historical turbidity information.

[0041] In a third aspect, the present application provides a computer device, which adopts the following technical solution:

[0042] A computer device comprises a processor, wherein the processor runs a program of any one of the above-mentioned adaptive data denoising methods based on environmental characteristics applied to a chemical park.

[0043] In a fourth aspect, the present application provides a storage medium, which adopts the following technical solution:

[0044] A storage medium stores a program of any one of the above-mentioned adaptive data denoising methods based on environmental characteristics applied to a chemical park.

[0045] In summary, the present application includes at least one of the following beneficial technical effects:

[0046] 1. By cleaning and compensating the environmental characteristics (such as scale and temperature) of the static pressure water level gauge, the measurement error is effectively reduced to ensure the high accuracy of the water level data. At the same time, the historical tidal data and real-time water level fluctuation range analysis are introduced to dynamically adjust the real-time turbidity value by calculating the correction coefficient, which significantly reduces the abnormal turbidity deviation caused by tidal fluctuations.

[0047] 2. The method combines the average water level and its changing speed, and can make adaptive adjustments to different typical modes such as high tide, low tide and non-tidal factors, thereby improving the system's adaptability to complex tidal environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flowchart of an adaptive data denoising method based on environmental characteristics applied to a chemical park in a certain embodiment of the present application.

[0049] Figure 2 It is a schematic longitudinal section diagram of the main body of the static pressure water level gauge in a certain embodiment of the present application.

[0050] Figure 3 It is a schematic cross-sectional view of the main body of a static pressure water level gauge in a certain embodiment of the present application.

[0051] Figure 4 This is a module connection diagram of an adaptive data denoising system based on environmental characteristics applied to a chemical park in a certain embodiment of the present application.

[0052] Reference numerals:

[0053] 11. Outer shell; 12. Convex ring; 13. Sensor; 102. Piezoelectric ceramics; 103. Connecting port; 5. Armored steel cable. DETAILED DESCRIPTION

[0054] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0055] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0056] The present application embodiment discloses an adaptive data denoising method based on environmental characteristics applied to a chemical park, referring to Figure 1 , including S1-S3.

[0057] S1. Detect the environmental characteristics of the static pressure water level gauge and start the cleaning mode of the static pressure water level gauge based on the environmental characteristics.

[0058] Due to the influence of tides, the water level will change frequently. Only by accurately obtaining the range of water level changes in the early stage can it be matched with historical tidal data in the later stage. However, it is found in actual operation that the static pressure water level placed in seawater is easily affected by dirt or scale. Scale deposition will cover the pressure surface of the sensor, increasing the resistance to pressure transmission. And the blockage of the pressure surface may cause the pressure to be unable to be fully transmitted to the sensor chip, resulting in a low measurement value. In addition, the scale itself has weight, which will increase the pressure on the sensor surface. If the scale deposition is heavy, the pressure value measured by the sensor will be high, resulting in an overestimation of the water level. The deposition of scale may cause changes in the density near the liquid, especially the adhesion of scale affects the fluidity, which may cause uneven local liquid density, thereby affecting the accuracy of the calculation. Scale deposition may also cause physical damage to the sensor, such as corroding the pressure diaphragm or blocking the exhaust hole, so that the measurement error accumulates over time, or affects the zero drift of the sensor (the change of the initial reference point of the measurement), resulting in long-term unstable measurement data.

[0059] Therefore, the embodiment of the present application provides a static pressure water level gauge capable of self-cleaning, referring to Figure 2 and Figure 3The main body of the static pressure water level gauge includes an outer shell 11 and a sensor 13 arranged inside the outer shell 11. Specifically, the outer shell 11 is a cylinder with an axis. In different embodiments, the cross section of the outer shell 11 can be of different shapes. As an example, the cross section of the outer shell 11 is circular. When the outer shell 11 is placed underwater, the tidal water flow will repeatedly and continuously impact the outer shell 11. The impact of the water flow has two sides. On the one hand, the impurities contained in the tidal water will continuously deposit on the contact surface between the outer shell 11 and the water under the flow of the water flow. On the other hand, the impact of the water flow will accelerate the detachment of impurities on the surface of the outer shell 11, and the two reach a dynamic balance.

[0060] The top surface of the outer shell 11 is connected to the external host through the armored steel cable 5. The bottom surface of the outer shell 11 is provided with a communication port 103, which is used for external fluid to enter and contact the sensor 13 inside the outer shell 11. In different embodiments, the communication port 103 can have different shapes, such as circular, square or other shapes. In order to make the shape of the communication port 103 not affect the water flow flowing into the inner part of the outer shell 11, or to make the inflowing water flow not produce differences when the outer shell 11 is at different rotation angles, therefore, preferably, the communication port 103 is selected to be circular, and the channel of the outer shell 11 concave by the communication port 103 is cylindrical. The pressure input port of the sensor 13 is arranged on the side of the channel. In different embodiments, the sensor 13 can adopt different models, such as an isolated diffused silicon sensitive element or a ceramic capacitor pressure sensitive sensor, to convert static pressure into an electrical signal. Here, the input position for obtaining water flow pressure information is called a pressure input port.

[0061] The piezoelectric ceramic 102 is installed inside the main body of the static pressure water level gauge. In different embodiments, the piezoelectric ceramic 102 can start vibration according to a preset time interval or according to the degree of dirt accumulation through the intelligent control module, and vibrate at a specific frequency and amplitude to remove dirt from the surface of the sensor 13, keep the surface of the sensor 13 clean, and ensure that the measurement accuracy is not affected by dirt accumulation. At present, in some existing technical routes, the degree of dirt accumulation can be measured by laser detection or ultrasonic detection or other methods. As an example, in this solution, a convex ring 12 surrounding the outer shell 11 is designed inside the connecting port. The convex ring 12 can not only reduce the impact of water flow fluctuations on the internal area of ​​the connecting port, but also provide echoes for ultrasonic waves.

[0062] For piezoelectric ceramics 102, when an electric field is applied in the polarization direction of the crystal, the lattices in these crystals will be deformed due to strong internal stress under the action of the electric field force. Under the action of the alternating electric field, the internal stress and deformation of the crystal will change periodically, thereby generating mechanical vibration. Specifically, if pressure is applied to the piezoelectric ceramic 102, it will generate a potential difference, which is called the positive piezoelectric effect. Conversely, if voltage is applied, mechanical stress is generated, which is called the inverse piezoelectric effect. If the pressure is a high-frequency vibration, high-frequency current is generated. When a high-frequency electrical signal is applied to the piezoelectric ceramic 102, a high-frequency sound signal is generated, that is, mechanical vibration, which is the so-called ultrasonic vibration.

[0063] Since the power-on position of the piezoelectric ceramic 102 is determined, the characteristic direction of the piezoelectric ceramic 102 is determined. Here, the characteristic direction of the piezoelectric ceramic 102 is the vibration direction of the piezoelectric ceramic 102 at the corresponding power-on position. In a certain embodiment, the vibration direction of the piezoelectric ceramic 102 is designed to form a certain angle with the axis of the outer shell 11, and the ultrasonic wave generated by the piezoelectric ceramic 102 can be used to clean the side of the outer shell 11. In another embodiment, the cross-section of the outer shell 11 is circular and gradually becomes thicker from both ends to the middle. In this way, the vibration direction of the piezoelectric ceramic 102 forms a certain angle with the outer surface of the outer shell 11, which can also play a cleaning role.

[0064] Specifically on the scale of natural days, the composition and content of wastewater discharged by chemical plants every day are relatively fixed, and the composition of seawater is also relatively fixed. However, due to changes in the speed of water flow, such as the influence of tides, the thickness of dirt and other substances formed will not be uniform. If the piezoelectric ceramic is started regularly every day to clean the static pressure water level gauge, on the one hand, there will be a problem of over-cleaning, because the formation of the dirt layer is actually relatively slow. On the other hand, starting ultrasonic vibration more frequently, such as cleaning once a day, may cause adverse effects on internal components, such as loose solder joints, etc., reducing service life.

[0065] It should be noted that the causes and components of scale or dirt are usually complex, and the generation characteristics of dirt will be a nonlinear function composed of multiple time-related parameters. If only theoretical numerical quantitative analysis is performed, it will not only be difficult, but also difficult to estimate the accuracy. Therefore, this application provides an embodiment. Since the piezoelectric ceramic itself will emit an ultrasonic signal, the appropriate frequency can be adjusted to detect the ultrasonic detection signal, thereby measuring the thickness of the dirt. This is because the ultrasonic detection signal carries a lot of information related to the characteristics of the reflecting surface. However, since the ultrasonic detection signal is usually mixed with a lot of noise and interference, specific steps are required to reduce the noise.

[0066] Specifically, S1 includes the following steps S11-S15.

[0067] S11. Control the piezoelectric ceramic to emit an ultrasonic measurement signal.

[0068] In this step, an ultrasonic signal is generated by the vibration of piezoelectric ceramics, and the physical properties of ultrasonic waves such as propagation and reflection in elastic media are used to obtain an echo signal containing dirt information on the sensor surface.

[0069] Piezoelectric ceramics are core devices that use the piezoelectric effect to achieve mechanical vibration and electrical signal conversion. When an electric field is applied in the polarization direction of the crystal, the lattice in the piezoelectric ceramic will deform under the action of the electric field force, and generate high-frequency mechanical vibrations with the periodic changes of the electric field. This mechanical vibration further forms an ultrasonic signal in the surrounding medium. Ultrasonic waves are mechanical vibration waves with a frequency higher than 20 kHz. They have a wide frequency range. High-frequency bands (such as 10MHz) are usually selected in underwater detection to obtain higher resolution and sensitivity. The propagation speed and reflection characteristics of ultrasonic signals are closely related to the physical properties of the propagation medium (such as density and acoustic impedance). Therefore, the reflected echo contains rich information about the medium.

[0070] The specific process of controlling the piezoelectric ceramic to emit an ultrasonic signal includes the following stages. First, an AC voltage of a specific frequency is generated by the control module to excite the piezoelectric ceramic, causing it to vibrate and emit an ultrasonic signal. In this embodiment, in order to optimize the penetration ability and resolution of the signal, an operating frequency of 10 MHz and a pulse width of 100 ns are selected. The emission of high-frequency ultrasonic waves can effectively improve the detection ability of thin layers of dirt while avoiding excessive signal attenuation. Subsequently, the ultrasonic wave propagates through an elastic medium (such as water) and is reflected when it encounters a medium interface (such as a dirt layer and a sensor surface). The arrival time (i.e., transit time) of the echo signal is proportional to the distance from the interface, while the amplitude of the reflected signal reflects the difference in acoustic impedance between the media. Based on these characteristics, it can be used to further calculate the thickness of the dirt.

[0071] Ultrasonic sensors (i.e., ultrasonic probes) are key components for transmitting and receiving ultrasonic signals. Their working principle is based on the inverse piezoelectric effect of piezoelectric ceramics. Driven by high-frequency AC signals, piezoelectric ceramics generate mechanical vibrations, convert electrical energy into mechanical wave energy and transmit it to the medium. When the ultrasonic wave is reflected by the interface, the mechanical vibration received by the ultrasonic probe is converted back into an electrical signal for system analysis. In this step, the design of the probe must ensure that its frequency matches the resonant frequency of the piezoelectric ceramic to maximize signal energy and reduce losses.

[0072] In order to further improve the temporal and spatial resolution of ultrasonic signals, the transmitted pulse signal adopts a narrow pulse form. This design helps to reduce the overlap of echo signals, allowing the system to more accurately distinguish multiple interface reflection echoes. In addition, the width and frequency of the pulse signal also need to comprehensively consider the attenuation characteristics and detection depth requirements in the water flow environment to ensure that the signal can effectively cover the target area.

[0073] S12. Acquire an ultrasonic detection signal based on an ultrasonic probe, and perform dimension expansion on the ultrasonic detection signal; wherein the waveform of the ultrasonic detection signal is generated from the ultrasonic emission waveform and the physical properties of the interface, and carries a large amount of effective information related to the measured medium and interface.

[0074] When controlling the piezoelectric ceramic to emit an ultrasonic measurement signal, the system needs to generate a specific AC voltage to excite the piezoelectric ceramic to emit an ultrasonic signal, and after it receives the reflected echo, it converts the mechanical vibration information into an electrical signal for processing.

[0075] In the specific implementation, the operating frequency of the piezoelectric ceramic must be consistent with its resonant frequency to ensure that the generated ultrasonic energy is maximized and the sensitivity of the measurement is improved. As an example, when 10MHz is selected as the operating frequency, the control circuit generates an excitation voltage with a frequency of 10MHz to drive the piezoelectric ceramic. Under this AC voltage drive, the piezoelectric ceramic is converted into ultrasonic pulses in an efficient manner.

[0076] In order to obtain sufficient time resolution and detection accuracy, the duration of the transmitted pulse is set to 100ns, which ensures that the pulse width can match the requirements of the measurement environment and improves the signal penetration and echo reception sensitivity.

[0077] Specifically, S12 includes the following steps S121 and S122.

[0078] S121. Assume that the ultrasonic detection signal is ,in, , , , .

[0079] It is the result of the ultrasonic signal emitted by piezoelectric ceramics being reflected at the medium interface and then received by the ultrasonic probe. It is effective ultrasonic echo information, reflecting the physical properties of the medium interface (such as dirt layer, sensor surface), including the thickness and acoustic impedance of the interface. k, α and β are parameters. is a noise signal, which may come from:

[0080] Dielectric scattering: When ultrasound encounters a non-smooth surface, random scattering signals are generated.

[0081] Environmental disturbances: such as mechanical vibrations or electromagnetic noise.

[0082] System noise: signal acquisition and transmission noise from the electronic equipment itself.

[0083] Mathematically, the signal can be modeled as: . It is the effective component of the signal and can usually be decomposed into the superposition of multiple echoes: Among them, P is the number of echoes, which represents the multiple reflections of ultrasound at different medium interfaces or the same interface. is the i-th echo signal, whose amplitude and phase reflect the characteristics of the medium interface. is the echo delay, that is, the time difference from the emission of the ultrasonic wave to its return, which reflects the position or thickness of the medium interface. There are obvious similarities.

[0084] is the noise component, which can be white noise or another type of random signal, and its characteristics are usually described by a statistical distribution (such as a Gaussian distribution).

[0085] S122. One-dimensional ultrasonic detection signal Do dimensional expansion and add additional noise to the original A random signal with a constant power spectral density and a virtual observation signal To assist in the analysis, , i=2, 3, ..., M; For Random signals of the same distribution; Construct the virtual observation matrix P, , M is the number of extended observation signals, .

[0086] One-dimensional signal Medium, Noise There may be some overlap with the frequency domain characteristics of the effective signal g(t), making it difficult to separate by direct filtering. In this case, by adding random noise and constructing a matrix form, the spatial distribution characteristics of the two-dimensional signal can be used to enhance the ability to distinguish between the signal and the noise.

[0087] Constructed virtual observation signal for: , i=2, 3,…,M. is the original ultrasonic detection signal, For Random signals of the same distribution, M is the number of expanded observation signals. In this way, the generated virtual observation signals can retain the characteristics of the original signals while diluting the correlation of the noise on the two-dimensional plane.

[0088] The M virtual observation signals Organized into a matrix P, P is an M×N matrix, N is the number of sampling points. It is randomly distributed in the matrix, while the effective signal g(t) has a high similarity between the matrix columns. By constructing the matrix P, the original one-dimensional signal is converted into a two-dimensional representation, which provides a basis for the subsequent use of algorithms such as wave atom transformation to extract effective signals.

[0089] Ultrasonic detection signal The waveform is generated by the ultrasonic emission waveform and the physical characteristics of the medium interface, and contains relevant information about the medium and the medium interface. It needs to be retained as much as possible in the process of noise reduction. , which usually comes from the scattering of the ultrasonic detection signal by the medium, or from interference from the external environment.

[0090] In this step, the dimension M of the dimensional expansion needs to be selected in combination with the computational complexity and the noise reduction effect. To make the selection, and p should be a relatively small integer so that M is larger than the period length of the fluctuation signal. As an example, P=2 is taken here.

[0091] S13. Perform noise reduction processing on the ultrasonic detection signal.

[0092] The virtual observation matrix P is a two-dimensional matrix constructed from the one-dimensional signal after dimension expansion. Each row represents a virtual observation signal , which consists of the original signal and random noise The key characteristics of the matrix are as follows:

[0093] 1. The original signal g(t) remains consistent in each row, thus forming a regular structure with high correlation between the columns of the matrix. This regularity manifests itself as a highly regular texture characteristic in two-dimensional space, similar to an oscillating texture in the vertical direction.

[0094] 2. Added random noise It is an independent signal with the same distribution as the original noise, with low correlation. The randomness of noise is manifested as disordered distribution in a two-dimensional matrix, making it difficult to form a continuous structure or significant texture.

[0095] 3. There may be significant noise peaks in the one-dimensional signal. These peaks are diluted through row and column distribution in the two-dimensional matrix, reducing their interference with the overall signal processing.

[0096] By considering the elements of the matrix P as the grayscale values ​​of pixels, the signal can be represented as a two-dimensional image. In this representation: the regularity of the effective signal makes the image present an oscillating texture in the vertical direction; random noise appears as randomly distributed isolated points or regions in the image, and its significance is weak. The advantages of this two-dimensional representation are: the characteristics of the signal are more intuitive and clear, which is convenient for processing using spatial filtering technology (such as wave atom transformation); the noise appears more dispersed in two-dimensional space, making it easier to separate.

[0097] Wave atom transform is a sparse representation method suitable for multidimensional signals, which can decompose the signal into basis function representations with local characteristics and directional selectivity. By applying wave atom transform on a two-dimensional matrix, the separation effect of noise and signal is significantly improved:

[0098] Wavelet atomic transformation can decompose a two-dimensional matrix into a sparse coefficient matrix. The main information of the signal is concentrated in a few large-amplitude coefficients, while the coefficients corresponding to the noise are more evenly distributed and have smaller amplitudes. Through threshold filtering (such as hard threshold or soft threshold filtering), large-amplitude signal coefficients can be retained while removing low-amplitude noise coefficients.

[0099] The high correlation of the signal in the two-dimensional matrix makes it appear as a concentrated energy distribution in the transformed coefficient matrix, which is easy to separate from the noise. The wave atom transform can also handle local changes in the signal, avoiding the information loss in the traditional Fourier transform.

[0100] Perform an inverse wave atom transformation on the filtered coefficient matrix to obtain a two-dimensional matrix after noise reduction. The signal is further restored to a one-dimensional form by averaging the rows and columns of the matrix. .

[0101] In ultrasonic detection, the matrix P contains the superposition results of multiple virtual observation signals. The two-dimensional wave atom transformation significantly improves the signal quality by extracting the regular texture (corresponding to the effective signal) in the matrix and removing the randomly distributed isolated points (corresponding to the noise).

[0102] Specifically, in a certain embodiment, S13 includes the following steps S131-S135.

[0103] S131. Decompose the virtual observation matrix P to obtain a square matrix .

[0104] In this step, several square sub-matrices are extracted from the virtual observation matrix P , which lays the foundation for the subsequent two-dimensional wave atomic transformation and noise reduction processing. This operation decomposes the complete matrix P into overlapping small matrices through frame technology to ensure that the signal characteristics are fully preserved during the noise reduction process and avoid data loss.

[0105] The reason for decomposing the virtual observation matrix P is that the original matrix P is usually large, and directly transforming and processing it will result in high computational complexity and large memory usage. By decomposing the matrix into smaller square sub-matrices , which can reduce the computational burden and more effectively analyze signal characteristics in a local area.

[0106] Specifically, frame segmentation and overlapping processing are required here.

[0107] Framing: Divide the columns of matrix P into several groups, each group of columns constitutes a square sub-matrix .

[0108] Overlap: A certain overlap degree η (usually 50%) is set between the columns of adjacent square matrices to ensure the continuity of information between adjacent frames and avoid the loss of boundary information caused by segmentation.

[0109] Square Matrix The formula is ,in, represents the kth column of the matrix P. M is the square side length of the matrix, , indicating the position of the starting column of the matrix.

[0110] The number of square matrices c can be calculated by the following formula: . Where N is the number of columns of matrix P.

[0111] For example, set the side length M and overlap η of the square matrix, such as M=64, η=50%. Calculate the starting column k of each square matrix: . Extract the corresponding columns from the matrix P according to the size of k and M to generate a square matrix Repeat the above steps for each i until all columns of the matrix P have been processed.

[0112] S132. Square Matrix Perform a two-dimensional wave atom transformation and obtain the coefficient matrix .

[0113] Two-dimensional wave atomic transform is a sparse representation technology, which is particularly suitable for processing local and directional features contained in two-dimensional matrix data. It can decompose the matrix signal into a sparse coefficient matrix, where: the energy of the effective signal g(t) is concentrated in a small number of high-amplitude transform coefficients; the noise h(t) is distributed in the entire domain of the transform space, showing as low-amplitude coefficients.

[0114] By decomposing, the matrix It can be expressed as: .in, is the signal in the wave atomic basis function The projection coefficient on Represents a basis function with localization and direction selectivity. The two-dimensional wave atomic transform can capture the regular texture in the signal (such as the oscillation characteristics in the vertical direction), while the noise appears as disordered high-frequency components.

[0115] Specifically, the square matrix Projected into the wave atom basis space, we get a sparse coefficient matrix : .in It is the transformed sparse coefficient matrix, which contains the distribution information of effective signals and noise.

[0116] First, the matrix Perform a Fast Fourier Transform (FFT) to convert the signal into the frequency domain: .

[0117] Then Apply the directivity and multiscale analysis of wave atom basis functions to decompose the signal into sparse coefficients: ,in, is the wave atom basis function, j and m represent the scale and direction index respectively.

[0118] The non-zero coefficients in are concentrated in a small number of directions and scales, representing the main information of the signal, while the coefficients of the noise are more dispersed and have lower amplitudes.

[0119] S133.Yes The coefficients in are threshold filtered to obtain the coefficient matrix after noise reduction. .

[0120] By setting a threshold λ, coefficients with amplitudes lower than the threshold (mainly noise coefficients) can be eliminated, and coefficients with larger amplitudes (mainly signal coefficients) can be retained.

[0121] The threshold filter formula is: ,in, Representation coefficient matrix ; λ is the threshold used to distinguish between signal and noise.

[0122] For the value of λ, the threshold can be dynamically adjusted according to the statistical characteristics of the coefficient matrix. For example, based on the noise model of Gaussian distribution, λ is determined using the mean or variance:

[0123] , where σ is the noise standard deviation and N is the number of matrix elements.

[0124] Here, hard threshold filtering is used for filtering. Coefficients below the threshold are directly set to zero:

[0125] .

[0126] In this step, the effective signal can be concentrated on a small number of high-amplitude coefficients, and the key information of the signal can be retained through filtering, thereby ensuring the integrity of the signal after noise reduction.

[0127] S134. The coefficient matrix after noise reduction Perform a two-dimensional wave atomic inverse transform to obtain the filtered .

[0128] right Perform a two-dimensional wave atomic inverse transform to obtain the filtered .in, is the sparse coefficient matrix after threshold filtering; represents the inverse wave atom transform operation; is the reconstructed denoising matrix. In the reconstruction process, the wave atom basis function A linear combination of is used to generate the signal: .in, yes The non-zero coefficients in ; is the wave atom basis function. Through this process, the matrix signal after noise reduction The key characteristics of the valid signal are retained while random noise is eliminated.

[0129] During the reconstruction process, according to The non-zero coefficients in select the corresponding wave atom basis functions for weighted combination. Each basis function The directionality and scale characteristics of the signal ensure that the local characteristics of the signal can be restored. Since the threshold filter has eliminated the low-amplitude noise coefficient, the noise component is significantly suppressed during reconstruction, and the resulting matrix signal The signal-to-noise ratio is greatly improved. The reconstructed result is a matrix that is consistent with the original matrix. Two-dimensional matrices of uniform size , but the noise is removed and the signal is enhanced.

[0130] S135. The result matrix after denoising The one-dimensional noise reduction result of the signal is obtained by accumulating and averaging in the longitudinal direction. .

[0131] In this step, the denoising square matrix processed in the previous step is Reassemble into a complete noise reduction matrix This step solves the data redundancy and boundary discontinuity problems introduced by framing and overlapping processing by weighted averaging, thereby generating a continuous and complete matrix.

[0132] By weighted averaging, the values ​​of multiple square matrices can be effectively fused in the overlapping area, solving the signal redundancy problem and smoothing the boundary transition. The weighted formula is:

[0133]

[0134] in, is the pixel value at position (k, l) in the reconstruction matrix; S(k, l) is the set of square matrices covering position (k, l); is a square matrix The weight at position (k,l).

[0135] In this embodiment, the weight Gaussian weighting is used. The weight is calculated using the Gaussian distribution function. The center weight of the overlapping area is larger, and the edge weight gradually decreases. The formula is:

[0136] , where σ controls the width of the Gaussian distribution.

[0137] S14. Perform non-local mean filtering.

[0138] In the above steps, the original one-dimensional signal f(t) is converted into a two-dimensional matrix P through dimension expansion, and the denoised complete matrix P' is generated through denoising. In order to be applied to actual scenarios, the denoised matrix signal needs to be restored to the same one-dimensional form f'(t) as the original signal.

[0139] Signal restoration is achieved by weighted superposition of the row data of the matrix P′. The specific operation includes the following steps:

[0140] Each row Represents a signal copy generated during the virtual observation signal expansion process. By weighted accumulation of these rows, the main features of the original signal can be restored. The formula is as follows:

[0141] , where M is the number of rows in the matrix; is the weight coefficient of the i-th row. Here, we assume that each row has the same weight, then The restored one-dimensional signal f′(t) is the final form of the denoised signal.

[0142] S15. Input a pre-trained neural network model to obtain the dirt thickness; wherein the pre-trained neural network model is trained with the echo time and echo amplitude of the pre-calibrated ultrasonic reflection echo waveform as input and the calibrated dirt thickness as output, wherein the outer diameter, inner diameter, and average wall thickness of the pipe are determined, the composition of seawater is determined, and the waveform of the ultrasonic reflection echo includes the echo peak of the outer wall of the pipe, the secondary echo of the outer wall of the pipe, the echo of the inner wall of the pipe, the secondary echo of the inner wall of the pipe, and the tertiary echo of the inner wall of the pipe.

[0143] The training of neural network models is already a fairly mature technology, and will not be described in detail here. It should be noted that in this embodiment, in order to provide rich sample data, a sample set is provided in a simulation manner. Specifically, the steps for obtaining the sample set are as follows:

[0144] S151. Simulate a one-dimensional signal f′(t) in a real scene as an input signal; wherein, , A is the signal amplitude, corresponding to the characteristics of different thicknesses; f is the signal frequency, simulating the physical environment characteristics; is the phase, corresponding to the echo time; n(t) is the noise component, which conforms to the Gaussian distribution.

[0145] S152. Generate the corresponding target output according to the signal amplitude and phase settings; wherein the thickness , k1 and k2 are coefficients. Environmental parameters .

[0146] S153. Use a random number generator and a mathematical model to simulate real signals, and introduce a certain range of random noise to simulate environmental interference.

[0147] S16. Determine whether the dirt thickness is greater than a preset threshold. If so, activate the piezoelectric ceramic to emit ultrasonic vibration for cleaning.

[0148] When the fouling thickness predicted by the neural network exceeds a set threshold, the ultrasonic cleaning operation of the hydrostatic water level gauge is automatically triggered.

[0149] S2. Obtain current water level fluctuation range information based on the static pressure water level gauge, and obtain turbidity information based on the turbidity sensor.

[0150] The static pressure information of the water body is collected in real time through the static pressure water level meter, and the water level value is calculated according to the formula. The turbidity data of the water body is obtained through the turbidity sensor to reflect the changes in the concentration of suspended particles. The collected data of the static pressure water level meter and the turbidity sensor are recorded synchronously, and the two types of data are synchronized through the timestamp.

[0151] S3. Correcting the turbidity information based on historical tidal data, wherein the historical tidal data includes historical water level information and historical turbidity information.

[0152] Specifically, in a certain embodiment, S3 includes the following steps S31-S34.

[0153] S31. Determine the corresponding water level range in the historical tidal data based on the current water level fluctuation range.

[0154] The current water level fluctuation range is ,in, The lowest water level monitored in real time. is the highest water level monitored in real time. This range can be calculated through a short time window of real-time water level information: ,in, is the water level at the current moment, and w is the length of the time window (such as 10 minutes).

[0155] The water level fluctuation range of historical tidal data is: , the matching condition is , that is, find all historical tidal records that overlap with the current water level fluctuation range. For example, if the current water level is 2.45m-2.50m, the matching range is the historical water level interval [2.4m, 2.6m].

[0156] S32. Extract historical turbidity information within the corresponding water level range from the historical tide database.

[0157] The historical tidal database stores the correlation information between water level and turbidity, which usually includes: the sampling time of tidal data, the historical water level value at the corresponding time, and the turbidity value at the corresponding time.

[0158] The data structure can be represented as: .

[0159] S33. Combined with historical tidal data, analyze the typical impact pattern of tidal fluctuations on turbidity.

[0160] The current average water level is calculated based on the water level fluctuation range, and the current typical impact mode is determined by the average water level and the speed of change of the average water level.

[0161] Based on the average water level and its changing speed, combined with historical tidal data, the current typical impact mode is determined:

[0162] In the high tide impact mode, turbidity usually increases because the water flow rolls up more suspended matter. In the low tide impact mode, turbidity usually decreases due to the sedimentation of particles. In the non-tidal impact mode, the change in turbidity is related to non-tidal factors, such as industrial emissions, rainfall, etc. Different typical impact modes correspond to different correction factors K, which are calibrated by historical data and stored in the database.

[0163] S34. Select a correction coefficient based on a typical impact mode, and calculate the corrected turbidity based on the correction coefficient and the real-time turbidity.

[0164] According to the typical impact mode (high tide, low tide, non-tidal) determined by S33, the corresponding correction coefficient database is searched. Combined with the current average water level and the speed of change, the most matching correction coefficient K is selected.

[0165] Using the formula Calculate the corrected turbidity.

[0166] Reference Figure 4 The present application also discloses an adaptive data noise reduction device based on environmental characteristics for use in a chemical park, comprising:

[0167] A cleaning module, used to detect the environmental characteristics of the static pressure water level gauge and start the cleaning mode of the static pressure water level gauge based on the environmental characteristics;

[0168] A data acquisition module, used to acquire water level information based on a static pressure water level gauge and to acquire turbidity information based on a turbidity sensor;

[0169] The turbidity correction module is used to correct the turbidity information based on historical tidal data, wherein the historical tidal data includes historical water level information and historical turbidity information.

[0170] An embodiment of the present application further discloses a computer device, including a processor, in which runs a program of any one of the above-mentioned adaptive data denoising methods based on environmental characteristics applied to a chemical park.

[0171] The embodiment of the present application also discloses a storage medium storing a program of any one of the above-mentioned adaptive data denoising methods based on environmental characteristics applied to a chemical park.

[0172] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An adaptive data denoising method based on environmental characteristics applied to a chemical park, characterized in that: The following steps are involved: S1. Detect the environmental characteristics of the static pressure water level gauge and start the cleaning mode of the static pressure water level gauge based on the environmental characteristics; S2. Obtain the current water level fluctuation range information based on the static pressure water level gauge and obtain turbidity information based on the turbidity sensor; S3. Correcting the turbidity information based on historical tidal data, wherein the historical tidal data includes historical water level information and historical turbidity information.

2. The method for adaptive data denoising based on environmental characteristics applied to a chemical park according to claim 1, characterized in that: The main body of the static pressure water level gauge includes an outer shell, a piezoelectric ceramic and a sensor arranged inside the outer shell. The outer shell is columnar, and a connecting port is opened on the bottom surface of the outer shell. The connecting port is used for external fluid to enter and contact the sensor inside the outer shell. The piezoelectric ceramic is installed inside the main body of the static pressure water level gauge. The top of the static pressure water level gauge is connected to an external host through an armored steel cable. The piezoelectric ceramic is used to output ultrasonic vibrations to the main body of the static pressure water level gauge.

3. The method for adaptive data denoising based on environmental characteristics applied to a chemical park according to claim 2, characterized in that: The static pressure water level gauge body also includes an ultrasonic probe and a convex ring located in the communication port. The ultrasonic probe is located in the outer shell and is used to detect the ultrasonic detection signal of the convex ring.

4. The method for adaptive data denoising based on environmental characteristics applied to a chemical park according to claim 3, characterized in that: The cross section of the outer shell is circular and gradually becomes thicker from both ends to the middle.

5. The method for adaptive data denoising based on environmental characteristics applied to a chemical park according to claim 4, characterized in that: The S1 comprises the following sub-steps: S11. Control the piezoelectric ceramic to emit an ultrasonic measurement signal; S12. Acquire an ultrasonic detection signal based on an ultrasonic probe, and perform dimension expansion on the ultrasonic detection signal; wherein the waveform of the ultrasonic detection signal is generated by the physical properties of the ultrasonic emission waveform and the interface, and carries relevant effective information related to the measured medium and the interface; S13. Perform noise reduction processing on the ultrasonic detection signal; S14. Perform non-local mean filtering; S15. Input the ultrasonic detection signal after non-local mean filtering into the pre-trained neural network model to obtain the dirt thickness; wherein the pre-trained neural network model is trained with the echo time and echo amplitude of the pre-calibrated ultrasonic reflection echo waveform as input and the calibrated dirt thickness as output, wherein the outer diameter, inner diameter and average wall thickness of the pipe are determined, the composition of the seawater is determined, and the waveform of the ultrasonic reflection echo includes the echo peak of the outer wall of the pipe, the secondary echo of the outer wall of the pipe, the echo of the inner wall of the pipe, the secondary echo of the inner wall of the pipe and the tertiary echo of the inner wall of the pipe; S16. Determine whether the dirt thickness is greater than a preset threshold. If so, activate the piezoelectric ceramic to emit ultrasonic vibration for cleaning.

6. The method for adaptive data denoising based on environmental characteristics applied to a chemical park according to claim 5, characterized in that: The S12 comprises the following steps: S121. Assume that the ultrasonic detection signal is ,in, , , , ; It is the result of the ultrasonic signal emitted by piezoelectric ceramics being reflected at the medium interface and then received by the ultrasonic probe. It is effective ultrasonic echo information, which is used to reflect the physical characteristics of the medium interface; is the noise signal; P is the number of echoes, representing multiple reflections of ultrasound at different medium interfaces or the same interface; is the i-th echo signal, whose amplitude and phase reflect the characteristics of the medium interface; is the time delay of the echo, that is, the time difference from the emission to the return of the ultrasonic wave, which reflects the position or thickness of the medium interface; k, α and β are parameters; S122. One-dimensional ultrasonic detection signal Do dimensional expansion and add additional noise to the original A random signal with a constant power spectral density and a virtual observation signal To assist in the analysis, , i=2, 3, ..., M; For Random signals of the same distribution; Construct the virtual observation matrix P, , M is the extended dimension; ； The S13 comprises the following steps: S131. Decompose the virtual observation matrix P to obtain a square matrix , The formula is ,in, represents the kth column of the matrix P, M is the square side length of the matrix, , represents the position of the starting column of the matrix; S132. Square Matrix Perform a two-dimensional wave atom transformation and obtain the coefficient matrix ; S133. Coefficient Matrix The coefficients in are threshold filtered to obtain the coefficient matrix after noise reduction ; S134. The coefficient matrix after noise reduction Perform a two-dimensional wave atomic inverse transform to obtain the filtered ; S135. The result matrix after denoising The one-dimensional noise reduction result of the signal is obtained by accumulating and averaging in the longitudinal direction. .

7. The method for adaptive data denoising based on environmental characteristics applied to a chemical park according to claim 5, characterized in that: The S3 comprises the following steps: S31. Determine the corresponding water level range in the historical tidal data according to the current water level fluctuation range; S32. Extracting historical turbidity information corresponding to the water level range from the historical tidal database; S33. Combined with historical tidal data, analyze the typical impact pattern of tidal fluctuations on turbidity; S34. Select a correction coefficient based on a typical impact mode, and calculate the corrected turbidity based on the correction coefficient and the real-time turbidity.

8. An adaptive data denoising device based on environmental characteristics applied to a chemical park, characterized in that: include: A cleaning module, used to detect the environmental characteristics of the static pressure water level gauge and start the cleaning mode of the static pressure water level gauge based on the environmental characteristics; A data acquisition module, used to acquire water level information based on a static pressure water level gauge and to acquire turbidity information based on a turbidity sensor; The turbidity correction module is used to correct the turbidity information based on historical tidal data, wherein the historical tidal data includes historical water level information and historical turbidity information.

9. A computer device, characterized in that: It comprises a processor, in which runs a program of the adaptive data denoising method based on environmental characteristics applied to a chemical park as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: A program for an adaptive data denoising method based on environmental characteristics applied to a chemical park as described in any one of claims 1 to 7 is stored.

Images