An optimization method for improving the detection distance of ozone laser radar, its device and storage medium

Abstract

The present invention relates to a distance detection of ozone, and in particular to an optimization method for improving the detection distance of an ozone laser radar, comprising the following steps: selecting different convolution kernels according to different distance heights of the signal; denoising the noise in the echo signal by a sliding convolution filtering method to obtain a smooth signal; using the smoothed signal to calculate the extinction coefficient and the backscattering coefficient; and calculating the ozone concentration on the vertical contour line. The optimization method for improving the detection distance of an ozone laser radar provided by the present invention adopts a new convolution noise reduction algorithm, and its characteristic of changing the sliding window size retains rich high-altitude ozone information while smoothing the signal, and effectively reduces noise, thereby achieving an improvement in the signal-to-noise ratio. Through experimental verification, the detection distance is increased from 2.5km to 5.0km, greatly improving the ozone detection distance.

Description

Technical Field

[0001] The present invention relates to ozone distance detection, and in particular to an optimization method for improving the detection distance of an ozone laser radar, a device thereof, and a storage medium. Background Art

[0002] Ozone is one of the primary pollutants causing air quality problems, and it has a negative impact on human health and the ecological environment. In order to understand the characteristics of ozone generation and destruction and its transmission trend, obtain continuous and accurate ozone spatiotemporal distribution, and achieve precise prevention and control, laser radar is usually used to monitor the vertical height profile of atmospheric ozone concentration. However, due to the low signal-to-noise ratio of ozone echo signals in the atmosphere, it is difficult to extract effective signals. At present, the monitoring height of ozone can generally only cover the area below 3km near the ground, which cannot meet higher business application requirements.

[0003] At present, lidar usually uses the differential absorption (DIAL) method to monitor ozone. Its basic principle is to use the absorption characteristics of ozone to measure concentration. In addition to the effective signal, the echo signal measured by the lidar also contains certain noise and interference signals. These noise and interference signals cannot be effectively suppressed using the current common algorithms, which restricts the improvement of the maximum effective detection distance performance of the ozone lidar. Summary of the invention

[0004] In order to solve the above problem that the detection distance of ozone by laser radar cannot be improved, the present invention provides an optimization method for improving the detection distance of ozone laser radar. The specific technical solution is as follows:

[0005] An optimization method for improving the detection distance of ozone laser radar includes the following steps:

[0006] Select different convolution kernels according to the different distance heights of the signal;

[0007] The noise in the echo signal is denoised by the sliding convolution filtering method to obtain a smooth signal;

[0008] Use the smoothed signal to calculate the extinction coefficient and backscattering coefficient;

[0009] Calculate the ozone concentration along the vertical contour lines.

[0010] Preferably, when selecting different convolution kernels: a filter kernel with a small filter radius and a large center weight is selected in the near field; a filter kernel with a large filter radius and a flat center weight is selected in the far field.

[0011] Preferably, the method further comprises: first acquiring on and off echo signals, and calculating the absorption cross section of the corresponding wavelength.

[0012] Preferably, when the echo signal is subjected to denoising, the basic noise model is:

[0013] ;

[0014] In the formula, f(n) is the ideal echo signal without noise; s(n) is the noisy signal; e(n) is the noise;

[0015] σ is the noise intensity.

[0016] Furthermore, when the convolution kernel function is selected, the convolution kernel h will change with the distance to adapt to the signal strength at different heights. In the near field, the center weight of the convolution kernel is high; in the long field, the center weight of the convolution kernel is reduced. The convolution formula is as follows:

[0017] ;

[0018] In the formula, h is the convolution kernel, x is the original signal, and y is the processed signal.

[0019] Furthermore, when calculating the ozone concentration on the vertical profile, the calculated extinction coefficient and backscattering coefficient are substituted into the ozone concentration expression to obtain the ozone concentration at different heights on the vertical profile;

[0020] The ozone concentration expression is:

[0021] .

[0022] A device for an optimization method for improving the detection distance of an ozone laser radar, the device comprising: a processor, a memory and a program; the program is stored in the memory, and the processor calls the program stored in the memory to execute the steps of the above-mentioned optimization method for improving the detection distance of an ozone laser radar.

[0023] A computer-readable storage medium is configured to store a program, and the program is configured to execute the steps of the above-mentioned optimization method for improving the detection distance of ozone laser radar.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] An optimization method for improving the detection distance of ozone lidar provided by the present invention adopts a new convolution noise reduction algorithm. The characteristic of changing the sliding window size can smooth the signal while retaining rich high-altitude ozone information, and effectively reduce noise to achieve an improvement in the signal-to-noise ratio. Through experimental verification, the detection distance is increased from 2.5km to 5.0km, greatly improving the ozone detection distance.

[0026] Different convolution kernels are selected according to the different distances and heights of the signal: a kernel with a small filter radius and a large center weight is selected for the near field; a kernel with a large filter radius and a flat center weight is selected for the far field. This method can be used to distinguish and denoise the signal, effectively reduce noise, improve the signal-to-noise ratio, and increase the ozone detection distance of the lidar. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of this application;

[0028] Figure 2 is a detection graph of the existing algorithm in the embodiment;

[0029] Figure 3 It is a detection map using the algorithm of the present application. DETAILED DESCRIPTION

[0030] The present invention will be further described with reference to the accompanying drawings.

[0031] In order to increase the maximum effective detection distance of the ozone radar and improve the overall radar performance level by improving the signal-to-noise ratio of the lidar, the sliding convolution filtering algorithm can be used to further effectively suppress noise and interference signals and effectively improve the signal-to-noise ratio, thereby achieving an increase in the effective detection distance of the radar without changing the lidar hardware peripherals.

[0032] like Figure 1 As shown in the figure, this method is based on the laser radar ozone echo signal and adopts the convolution noise reduction method. By using different convolution kernels to process the noise of near-field and far-field signals, the laser radar signal-to-noise ratio is improved and the ozone detection distance is effectively increased. The specific technical scheme is as follows:

[0033] From the LiDAR equation, we know that

[0034] (1);

[0035] Then, the differential absorption method measures the distance Z to The ozone concentration between is expressed as:

[0036] (2);

[0037] Among them, B', E A ' and E M ' are collectively referred to as atmospheric interference terms (or atmospheric correction terms), where B' is the backscattering coefficient of the atmosphere (including atmospheric molecules and aerosols); E A ' and E M ' are the extinction coefficients of atmospheric aerosols and atmospheric molecules, respectively.

[0038] In the above formula, the echo signal P is doped with noise, which will submerge the effective information of the signal, resulting in a low signal-to-noise ratio and insufficient detection distance.

[0039] The optimized inversion process is as follows:

[0040] 1. Obtain the on and off echo signals and calculate the absorption cross section of the corresponding wavelength;

[0041] De-noising the signal: Traditional signal denoising methods will sacrifice more effective signals to reduce noise, which will greatly hinder the detection distance. In this regard, we creatively adopt a new signal smoothing method. The basic noise model is:

[0042] ;

[0043] Among them, f(n) is the ideal echo signal (noise-free); s(n) is the noisy signal; e(n) is the noise; σ is the noise intensity. Under ideal conditions, the noise and the ideal signal are independent of each other.

[0044] 2. Restoring f(n) by suppressing e(n) can achieve the purpose of noise reduction. Combining the signal characteristics of ozone, a sliding smoothing method is used here to set a reasonable convolution kernel function. The convolution kernel h will change with distance to adapt to the signal strength at different heights. In the near field, the center weight of the convolution kernel is high; in the long range, the center weight of the convolution kernel is reduced. The convolution formula is as follows:

[0045] ;

[0046] Among them, h is the convolution kernel, x is the original signal, and y is the processed signal.

[0047] 3. Use the smoothed signal to calculate the extinction coefficient and backscattering coefficient, and substitute each quantity into formula (2) to obtain the ozone concentration at different heights on the vertical profile.

[0048] Extinction coefficient: The ratio of a substance's attenuation or intensity reduction of incident light.

[0049] Wavelength absorption cross section: At a certain wavelength, the ratio of the number of photons that reach an area per unit time through an area unit to the number of photons per unit cross-sectional area at that wavelength.

[0050] Example: The data collected by ozone laser radar at the test site of Nanjing University was compared before and after algorithm optimization. The experimental results are as follows:

[0051] like Figure 2 As shown, the existing algorithm for ozone detection has an effective detection distance of 2.5 km.

[0052] like Figure 3As shown, the ozone detection algorithm of this application has an effective detection distance of 5 km.

[0053] Obviously, a new convolution noise reduction algorithm was adopted. Its characteristic of changing the sliding window size can smooth the signal while retaining rich high-altitude ozone information, and effectively reduce noise to achieve an improvement in the signal-to-noise ratio. Through experimental verification, the detection distance was increased from 2.5km to 5.0km. This method can greatly improve the ozone detection distance.

[0054] A device for an optimization method for improving the detection distance of an ozone laser radar, the device comprising: a processor, a memory and a program; the program is stored in the memory, and the processor calls the program stored in the memory to execute the steps of the above-mentioned optimization method for improving the detection distance of an ozone laser radar.

[0055] The memory and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines, such as through a bus connection. The memory stores computer-executable instructions for implementing the data access control method, including at least one software function module that can be stored in the memory in the form of software or firmware. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory.

[0056] The memory may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc. The memory is used to store programs, and the processor executes the programs after receiving the execution instruction.

[0057] The processor may be an integrated circuit chip having the ability to process signals. The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. The methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0058] A computer-readable storage medium is configured to store a program, and the program is configured to execute the steps of the above-mentioned optimization method for improving the detection distance of ozone laser radar.

[0059] The embodiments of the present invention are described with reference to flowcharts of methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate means for implementing the functions specified in the flowchart and / or.

[0060] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device that implements the functions specified in the flowchart.

[0061] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operation steps are executed on the computer or other programmable terminal device to produce computer-implemented processing, whereby the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in the flowchart.

[0062] The technical principle of the present invention is described above in conjunction with specific embodiments. These descriptions are only for explaining the principle of the present invention and cannot be interpreted as limiting the protection scope of the present invention in any way. Based on the explanations herein, those skilled in the art can associate other specific implementations of the present invention without paying creative labor, and these methods will fall within the protection scope of the claims of the present invention.

Claims

1. An optimization method for improving the detection distance of ozone laser radar, characterized in that: The following steps are involved: Select different convolution kernels according to the different distance heights of the signal; The noise in the echo signal is denoised by the sliding convolution filtering method to obtain a smooth signal; Use the smoothed signal to calculate the extinction coefficient and backscattering coefficient; Calculate ozone concentration on vertical contour lines; When selecting different convolution kernels: In the near field, a filter kernel with a small filter radius and a large center weight is selected; In the far field, a filter kernel with a large filter radius and a flat center weight is selected.

2. The optimization method for improving the detection distance of ozone laser radar according to claim 1, characterized in that: Also includes: First, obtain the on and off echo signals and calculate the absorption cross section of the corresponding wavelength.

3. The optimization method for improving the detection distance of ozone laser radar according to claim 2, characterized in that: When denoising the echo signal, the basic noise model is: ； In the formula, f(n) is an ideal echo signal without noise; s(n) is the noisy signal; e(n) is the noise; σ is the noise intensity.

4. The optimization method for improving the detection distance of ozone laser radar according to claim 3, characterized in that: When the convolution kernel function is selected, the convolution kernel h will change with the distance to adapt to the signal strength at different heights. In the near field, the center weight of the convolution kernel is high; in the far field, the center weight of the convolution kernel is reduced. The convolution formula is as follows: ； In the formula, h is the convolution kernel, x is the original signal, and y is the processed signal.

5. The optimization method for improving the detection distance of ozone laser radar according to claim 4, characterized in that: When calculating the ozone concentration on the vertical profile, the calculated extinction coefficient and backscattering coefficient are substituted into the ozone concentration expression to obtain the ozone concentration at different heights on the vertical profile; The ozone concentration expression is: ； Where B' is the atmospheric backscatter coefficient; E A ' is atmospheric aerosol; E M 'Extinction coefficient of atmospheric molecules; P is the echo signal P.

6. A device for optimizing the detection distance of ozone laser radar, characterized in that: The device comprises: Processor, memory and program; The program is stored in the memory, and the processor calls the program stored in the memory to execute the steps of an optimization method for improving the detection distance of ozone laser radar as described in claim 1.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium is configured to store a program, and the program is configured to execute the steps of an optimization method for improving the detection distance of ozone lidar as described in claim 1.

Images