阅读说明：本技术 一种光声信号降噪系统及其降噪方法 (Photoacoustic signal noise reduction system and noise reduction method thereof ) 是由 万青 蔡善忠 时悦 宋鹏才 刘欢 于 2020-07-08 设计创作，主要内容包括：本发明提供了一种光声信号降噪系统及其降噪方法,包括输入信号处理单元、参考信号产生单元、相敏检测单元、低通滤波单元、卡尔曼滤波单元和输出信号处理单元；输入信号处理单元对输入的光声信号进行前置放大后发送至相敏检测单元中；参考信号产生单元产生两路有相位差的参考信号；相敏检测单元将上述前置放大后的输出的光声信号分别与两路参考信号进行复合,产生两路检测输出信号,该两路检测输出信号发送至低通滤波单元中消除高频部分；随后再输出至卡尔曼滤波单元；卡尔曼滤波单元接收到低通滤波器输出值后,进一步进行卡尔曼滤波降噪,将滤波降噪结果输出至输出信号处理单元中。(The invention provides a photoacoustic signal noise reduction system and a noise reduction method thereof, wherein the photoacoustic signal noise reduction system comprises an input signal processing unit, a reference signal generating unit, a phase-sensitive detection unit, a low-pass filtering unit, a Kalman filtering unit and an output signal processing unit; the input signal processing unit pre-amplifies the input photoacoustic signal and sends the amplified photoacoustic signal to the phase-sensitive detection unit; the reference signal generating unit generates two paths of reference signals with phase difference; the phase-sensitive detection unit respectively combines the output photoacoustic signals after pre-amplification with two reference signals to generate two detection output signals, and the two detection output signals are sent to a low-pass filtering unit to eliminate a high-frequency part; then outputting the data to a Kalman filtering unit; and after receiving the output value of the low-pass filter, the Kalman filtering unit further performs Kalman filtering noise reduction, and outputs a filtering noise reduction result to the output signal processing unit.)

1. A photoacoustic signal noise reduction system, characterized by: the phase-sensitive detection device comprises an input signal processing unit, a reference signal generating unit, a phase-sensitive detection unit, a low-pass filtering unit, a Kalman filtering unit and an output signal processing unit; the output end of the input signal processing unit and the output end of the reference signal generating unit are respectively in signal connection with the input end of the phase-sensitive detection unit; the output end of the phase-sensitive detection unit is in signal connection with the input end of the low-pass filtering unit, and the output end of the low-pass filtering unit is in signal connection with the input end of the Kalman filtering unit; the output end of the Kalman filtering unit is in signal connection with the input end of the output signal processing unit; wherein:

the input signal processing unit is used for pre-amplifying the input photoacoustic signal and then sending the amplified photoacoustic signal to the phase-sensitive detection unit;

the reference signal generating unit is used for generating a reference signal and directly sending the reference signal to the phase-sensitive detection unit, and phase-shifting the reference signal and then sending the reference signal to the phase-sensitive detection unit;

the phase-sensitive detection unit is used for respectively compounding the output photoacoustic signals subjected to pre-amplification with reference signals and phase-shifted reference signals to generate two paths of detection output signals, and the two paths of detection output signals are sent to the low-pass filtering unit;

the low-pass filtering unit respectively filters two paths of detection output signals generated by the phase-sensitive detection unit and outputs the two paths of detection output signals to the Kalman filtering unit;

after receiving the output value of the low-pass filter, the Kalman filtering unit further performs Kalman filtering noise reduction, and outputs a filtering noise reduction result to the output signal processing unit;

and the output signal processing unit carries out addition operation according to the results output by the two Kalman filtering units to obtain the amplitude of the photoacoustic signal to be detected, and inversely calculates the gas concentration in the photoacoustic cell.

2. A photoacoustic signal noise reduction system as set forth in claim 1, wherein: the phase-sensitive detection unit is used for carrying out proportional operation on the pre-amplified input photoacoustic signal and the reference signal and the phase-shifted reference signal respectively to obtain two paths of detection output signals; the reference signal and the phase-shifted reference signal are orthogonal sinusoidal signals.

3. A photoacoustic signal noise reduction system as set forth in claim 2, wherein: the phase-sensitive detection unit comprises a first digital multiplier and a second digital multiplier, and the pre-amplified input photoacoustic signal and the reference signal are input into the first digital multiplier; the pre-amplified input photoacoustic signal and the phase-shifted reference signal are input into a second digital multiplier, and the outputs of the two paths of digital multipliers are led into a low-pass filtering unit.

4. A photoacoustic signal noise reduction system as set forth in claim 2, wherein: the low-pass filtering unit is an FIR filter or an IIR filter.

5. A noise reduction method of a photoacoustic signal noise reduction system is characterized in that: the method comprises the following steps:

s1: the photoacoustic signal excited in the photoacoustic cell is pre-amplified by the input signal processing unit to obtain a pre-amplified output photoacoustic signal S_i(t)，C isThe amplitude of the signal is such that,

s2: pre-amplified output photoacoustic signal S_i(t) inputting into the phase-sensitive detection unit, and combining with the reference signal generated by the reference signal generation unit and the phase-shifted reference signal to obtain two detection output signals, which are respectively:

where A (t) is time-varying noise superimposed in amplitude after extracting a signal of known frequency; s_r(t) is a known frequency signal, sin ω t and cos ω t in the formula; nT is the width of the fourier time window; n is the number of fundamental wave cycles; a (t) is much less than n_i(t) and A (t)>C；

S3: the low-pass filter unit filters and outputs two paths of detection output signals generated by the phase-sensitive detection unit to eliminate high-frequency components except cut-off frequency, and the output Y of each path of photoacoustic signal processed by the low-pass filter₁Or Y₂Comprises the following steps:

s4: the output Y of the photoacoustic signal processed by the low-pass filter is input into a Kalman filtering unit for Kalman filtering noise reduction, and the superimposed noise is further reduced by continuously predicting and correcting the output value by the Kalman filtering unitSound: let the effective value after the low-pass filter processing be X' (t),

let the effective value of the output obtained by measurement be z (t), the measurement error be v (t), and z (t) ═ x (t) + v (t); w (t) and V (t) are both time-varying Gaussian noise portions, the variances of W (t) and V (t) beingCovarianceAccording to the output X (t) of the low-pass filter at the time t, the estimated value X (t +1) of the output of the low-pass filter at the time t +1 is solved as follows:

and (3) estimating value at the next moment: x (t +1) | (t) ═ X (t);

the covariance matrix is predicted as: p (t +1) | (t) ═ P (t) + w (t);

the state is updated as follows:

x (t +1) ═ X (t +1) | (t) + K (t +1) [ Z (t +1) -X (t +1) | (t) ]; k (t +1) is the Kalman gain;

the covariance matrix is updated as: p (t +1) ═ 1-K (t +1) ] P (t +1) | (t);

the effective value after filtering and noise reduction by the Kalman filtering unit is as follows:

s5: the output signal processing unit calculates the amplitude of the acousto-optic signal by addition according to the result of Kalman filtering noise reductionAnd further calculating the concentration of the gas in the acousto-optic cell; the relationship between the photoacoustic signal and the gas concentration is: u. of_λ＝aσ_λI_λC_cell；

Wherein u is_λIs the intensity of the photoacoustic signal excited in the photoacoustic cell, a is the proportionality coefficient, σ_λIs the absorption coefficient of the gas at the wavelength λ; i is_λIs the intensity of the incident light at wavelength λ; c_cellIs the cell constant of the photoacoustic cell; the light for excitation is a monochromatic laser or infrared light, and σ is determined according to the change of gas concentration_λWill also follow the changes.

Technical Field

The invention relates to the technical field of photoacoustic signal detection and processing, in particular to a photoacoustic signal noise reduction system and a photoacoustic signal noise reduction method.

Background

With the gradual advance of the power industry into the era of high-voltage and large power grids, the reliability requirement on the fault diagnosis technology of power equipment is higher and higher, and therefore the precision requirement on the concentration detection of the dissolved gas in the oil is higher and higher. The concentration of dissolved gas in transformer oil is detected by a photoacoustic spectroscopy technology, and the analysis of different gas concentration ratios is the main method for judging potential operation faults of the transformer at present. The principle of detecting gas concentration by photoacoustic spectroscopy is shown in fig. 1, and the detection device mainly comprises an excitation light source, a photoacoustic cell, a signal detection part and the like. The gas in the photoacoustic cell is irradiated by a monochromatic light with adjustable frequency, the gas absorbs the light energy to transit from a ground state to an excited state and is immediately deactuated in a mode of releasing heat energy, the released heat energy generates periodic heating to the surrounding medium according to the modulation frequency of the light, so that the medium generates periodic pressure fluctuation, the sound pressure is detected by a microphone and converted into a photoacoustic signal, the more the gas molecules, the stronger the photoacoustic pressure wave generated, and the stronger the converted photoacoustic signal, so that the gas concentration can be quantitatively analyzed. However, such signals are inevitably mixed with strong environmental noise and circuit noise, the extraction and noise reduction of the signals by using the lock-in amplifier is the main method for the noise reduction processing of the photoacoustic signals at present, when the gas concentration is low or the online monitoring system is in a very harsh natural environment, a strong magnetic field generated by a large voltage and a large current of the transformer has large interference on the signals, so that the very weak photoacoustic signals are mixed with the strong noise, and the traditional lock-in amplifier is difficult to filter the strong noise efficiently, so that the signal processing method is difficult to meet the requirement on the detection accuracy of the low-concentration gas.

Disclosure of Invention

In view of this, the present invention provides a photoacoustic signal noise reduction system and a noise reduction method thereof, which can combine a kalman filter and a lock-in amplifier, enhance noise reduction capability, and do not increase a time constant.

In one aspect, the invention provides a photoacoustic signal noise reduction system, which comprises an input signal processing unit, a reference signal generating unit, a phase-sensitive detecting unit, a low-pass filtering unit, a kalman filtering unit and an output signal processing unit; the output end of the input signal processing unit and the output end of the reference signal generating unit are respectively in signal connection with the input end of the phase-sensitive detection unit; the output end of the phase-sensitive detection unit is in signal connection with the input end of the low-pass filtering unit, and the output end of the low-pass filtering unit is in signal connection with the input end of the Kalman filtering unit; the output end of the Kalman filtering unit is in signal connection with the input end of the output signal processing unit; wherein:

the input signal processing unit is used for pre-amplifying the input photoacoustic signal and then sending the amplified photoacoustic signal to the phase-sensitive detection unit;

the reference signal generating unit is used for generating a reference signal and directly sending the reference signal to the phase-sensitive detection unit, and phase-shifting the reference signal and then sending the reference signal to the phase-sensitive detection unit;

the phase-sensitive detection unit is used for respectively compounding the output photoacoustic signals subjected to pre-amplification with reference signals and phase-shifted reference signals to generate two paths of detection output signals, and the two paths of detection output signals are sent to the low-pass filtering unit;

the low-pass filtering unit respectively filters two paths of detection output signals generated by the phase-sensitive detection unit and outputs the two paths of detection output signals to the Kalman filtering unit;

after receiving the output value of the low-pass filter, the Kalman filtering unit further performs Kalman filtering noise reduction, and outputs a filtering noise reduction result to the output signal processing unit;

and the output signal processing unit carries out addition operation according to the results output by the two Kalman filtering units to obtain the amplitude of the photoacoustic signal to be detected, and inversely calculates the gas concentration in the photoacoustic cell.

On the basis of the above technical solution, preferably, the phase-sensitive detection unit performs proportional operation on the pre-amplified input photoacoustic signal and the reference signal and the phase-shifted reference signal respectively to obtain two paths of detection output signals; the reference signal and the phase-shifted reference signal are orthogonal sinusoidal signals.

Further preferably, the phase-sensitive detection unit includes a first digital multiplier and a second digital multiplier, and the pre-amplified input photoacoustic signal and the reference signal are input into the first digital multiplier; the pre-amplified input photoacoustic signal and the phase-shifted reference signal are input into a second digital multiplier, and the outputs of the two paths of digital multipliers are led into a low-pass filtering unit.

Further preferably, the low-pass filtering unit is an FIR filter or an IIR filter.

In another aspect, the present invention further provides a noise reduction method for a photoacoustic signal noise reduction system, including the following steps:

s1: the photoacoustic signal excited in the photoacoustic cell is pre-amplified by the input signal processing unit to obtain a pre-amplified output photoacoustic signal S_i(t)，C is the amplitude of the signal and is,for the signal phase, ω is the angular frequency,

is the initial phase at time t-0; n is_i(t) is disordered white noise;

s2: pre-amplified output photoacoustic signal S_i(t) inputting into the phase-sensitive detection unit, and combining with the reference signal generated by the reference signal generation unit and the phase-shifted reference signal to obtain two detection output signals, which are respectively:

where A (t) is time-varying noise superimposed in amplitude after extracting a signal of known frequency; s_r(t) is a known frequency signal, sin ω t and cos ω t in the formula; nT is the width of the fourier time window; n is the number of fundamental wave cycles; a (t) is much less than n_i(t) andA(t)>C；

s3: the low-pass filter unit filters and outputs two paths of detection output signals generated by the phase-sensitive detection unit to eliminate high-frequency components except cut-off frequency, and the output Y of each path of photoacoustic signal processed by the low-pass filter₁Or Y₂Comprises the following steps:

a' (t) is the low-pass filtered noise component;

s4: and inputting the output Y of the photoacoustic signal processed by the low-pass filter into a Kalman filtering unit for Kalman filtering noise reduction, and continuously predicting and correcting an output value by using the Kalman filtering unit to further reduce the superposed noise: let the effective value after the low-pass filter processing be X' (t),

which is superimposed on a noise ofThen at time t there is a transformation of the output Y of the low pass filter into: x (t) ═ X' (t) + w (t);

let the effective value of the output obtained by measurement be z (t), the measurement error be v (t), and z (t) ═ x (t) + v (t); w (t) and V (t) are both time-varying Gaussian noise portions, the variances of W (t) and V (t) beingCovariance

According to the output X (t) of the low-pass filter at the time t, the estimated value X (t +1) of the output of the low-pass filter at the time t +1 is solved as follows:

and (3) estimating value at the next moment: x (t +1) | (t) ═ X (t);

the covariance matrix is predicted as: p (t +1) | (t) ═ P (t) + w (t);

the state is updated as follows:

x (t +1) ═ X (t +1) | (t) + K (t +1) [ Z (t +1) -X (t +1) | (t) ]; k (t +1) is the Kalman gain;

the covariance matrix is updated as: p (t +1) ═ 1-K (t +1) ] P (t +1) | (t);

the effective value after filtering and noise reduction by the Kalman filtering unit is as follows:

w "(t) is the output noise after Kalman filtering unit filtering noise reduction, W" (t)<W (t); the intensity of noise contained in the effective value is further reduced;

s5: the output signal processing unit calculates the amplitude of the acousto-optic signal through addition according to the result of Kalman filtering denoising, and further calculates the gas concentration of the acousto-optic pool; the relationship between the photoacoustic signal and the gas concentration is: u. of_λ＝aσ_λI_λC_cell；

Wherein u is_λIs the intensity of the photoacoustic signal excited in the photoacoustic cell, a is the proportionality coefficient, σ_λIs the absorption coefficient of the gas at the wavelength λ; i is_λIs the intensity of the incident light at wavelength λ; c_cellIs the cell constant of the photoacoustic cell; the light for excitation is a monochromatic laser or infrared light, and σ is determined according to the change of gas concentration_λWill also follow the changes.

Compared with the prior art, the photoacoustic signal noise reduction system and the photoacoustic signal noise reduction method provided by the invention have the following beneficial effects:

(1) according to the invention, the low-pass filtering unit and the Kalman filtering unit are sequentially added after the processing of the existing lock-in amplifier, so that the noise reduction capability can be enhanced without increasing the time constant, a weak photoacoustic signal can be reliably amplified, and the subsequent detection is facilitated;

(2) compared with the existing signal detection processing system, when the noise is at the same level, the invention can reduce the work load of the phase-locked amplifier by setting higher cut-off frequency of the low-pass filtering unit;

(3) the Kalman filtering unit can prevent signal distortion from generating singular values, and the stability of the online monitoring system is improved;

(4) the mean value and the covariance of the Kalman filtering unit have linear transitivity, an output value at each moment is recorded into a posterior estimation value at the next moment through multiplication, the posterior estimation value at the moment contains information of all previous measurement values through continuous multiplication and iteration, and the system error is reduced.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a schematic diagram of gas concentration detection by photoacoustic spectroscopy;

FIG. 2 is a system block diagram of a photoacoustic signal noise reduction system and a noise reduction method thereof according to the present invention;

FIG. 3 is a schematic diagram illustrating the noise reduction effect of a conventional signal noise reduction method;

FIG. 4 is a schematic diagram of the noise reduction effect of the photoacoustic signal noise reduction system and the noise reduction method thereof according to the present invention;

FIG. 5 is a schematic diagram of the time delay corresponding to a lower cut-off frequency when the prior signal extraction method achieves the same noise reduction effect of the present invention;

fig. 6 is a schematic diagram of the filtering effect of the photoacoustic signal noise reduction system and the noise reduction method thereof according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.

As shown in fig. 2, the present invention provides a photoacoustic signal noise reduction system, including an input signal processing unit, a reference signal generating unit, a phase-sensitive detecting unit, a low-pass filtering unit, a kalman filtering unit, and an output signal processing unit; wherein:

the input signal processing unit is used for pre-amplifying the input photoacoustic signal and then sending the amplified photoacoustic signal to the phase-sensitive detection unit;

the reference signal generating unit is used for generating a reference signal and directly sending the reference signal to the phase-sensitive detection unit, and phase-shifting the reference signal and then sending the reference signal to the phase-sensitive detection unit;

the phase-sensitive detection unit is used for respectively compounding the output photoacoustic signals subjected to pre-amplification with reference signals and phase-shifted reference signals to generate two paths of detection output signals, and the two paths of detection output signals are sent to the low-pass filtering unit;

the low-pass filtering unit respectively filters two paths of detection output signals generated by the phase-sensitive detection unit and outputs the two paths of detection output signals to the Kalman filtering unit;

after receiving the output value of the low-pass filter, the Kalman filtering unit further performs Kalman filtering noise reduction, and outputs a filtering noise reduction result to the output signal processing unit;

the output signal processing unit carries out addition operation according to the results output by the two Kalman filtering units to obtain the amplitude of the photoacoustic signal to be detected, and the gas concentration in the photoacoustic cell is inversely calculated;

the output end of the input signal processing unit and the output end of the reference signal generating unit are respectively in signal connection with the input end of the phase-sensitive detection unit; the output end of the phase-sensitive detection unit is in signal connection with the input end of the low-pass filtering unit, and the output end of the low-pass filtering unit is in signal connection with the input end of the Kalman filtering unit; the output end of the Kalman filtering unit is in signal connection with the input end of the output signal processing unit; the output signal processing unit carries out subsequent processing to obtain the concentration information of the gas, which is convenient for subsequent analysis.

In the invention, a phase-sensitive detection unit performs proportional operation on a pre-amplified input photoacoustic signal and a reference signal and a phase-shifted reference signal respectively to obtain two paths of detection output signals; the reference signal and the phase-shifted reference signal are orthogonal sinusoidal signals. The phase-sensitive detection unit comprises a first digital multiplier and a second digital multiplier, and the pre-amplified input photoacoustic signal and the reference signal are input into the first digital multiplier; the pre-amplified input photoacoustic signal and the phase-shifted reference signal are input into a second digital multiplier, and the outputs of the two paths of digital multipliers are led into a low-pass filtering unit.

Further, the low-pass filtering unit is an FIR filter or an IIR filter.

In addition, the invention also provides a noise reduction method of the photoacoustic signal noise reduction system, which specifically comprises the following steps:

s1: the photoacoustic signal excited in the photoacoustic cell is pre-amplified by the input signal processing unit to obtain a pre-amplified output photoacoustic signal S_i(t)，C is the amplitude of the signal and is,for the signal phase, ω is the angular frequency,is the initial phase at time t-0; n is_i(t) is disordered white noise; the directly input photoacoustic signal to be detected is weak and needs to be amplified, so that the working requirement of the phase-sensitive monitoring unit is met.

S2: pre-amplified output photoacoustic signal S_i(t) inputting into the phase-sensitive detection unit, and combining with the reference signal generated by the reference signal generation unit and the phase-shifted reference signal to obtain two paths of detection output signals, S_r(t) is the known frequency signal, here orthogonal sin ω t and cos ω t; according to the principle of correlation, S_i(t) and S_rThe ideal output result of (t) multiplication is:

however, when the actually detected gas concentration is low, the extracted photoacoustic signal still has strong noise due to too much noise power relative to the weaker photoacoustic signal; s_i(t) and S_rThe actual result of (t) multiplication is:

where A (t) is time-varying noise superimposed in amplitude after extracting a signal of known frequency; s_r(t) is a known frequency signal, sin ω t and cos ω t in the formula; nT is the width of the fourier time window; n is the number of fundamental wave cycles; a (t) is much less than n_i(t) and A (t)>C。

S3: the low-pass filter unit filters and outputs two paths of detection output signals generated by the phase-sensitive detection unit to eliminate high-frequency components except cut-off frequency, and the output Y of each path of photoacoustic signal processed by the low-pass filter₁Or Y₂Comprises the following steps:

a' (t) is the low-pass filtered noise component;

if Y is to be₁And Y₂Directly accumulating and carrying out subsequent calculation to obtain the amplitude of the acousto-optic signal to be measured, wherein A ' (t) can be ignored when C > A ' (t) because the acousto-optic signal to be measured contains an A ' (t) noise part; when the gas concentration is low, A' (t) can cause the amplitude of the acousto-optic signal to be detected to generate large fluctuation to generate errors, which is not beneficial to monitoring and maintenance of the power transformer; therefore, further noise reduction measures need to be introduced; according to the noise reduction principle of the low-pass filter, the signal-to-noise ratio of the output signal can be improved by reducing the cut-off frequency, but the time constant is increased, so that great time delay is caused, when the photoacoustic signal is very weak, the low-pass filter output value meeting the requirement can be obtained by using an extremely low cut-off frequency, and the low-pass filter needs to wait for a long timeThe calculation result can be obtained only by time, which does not meet the real-time requirement of on-line monitoring; meanwhile, the lower cut-off frequency affects the load of the phase-lock amplification of the phase-sensitive detection unit. Therefore, the invention further adopts the Kalman filtering unit to further reduce noise.

S4: and inputting the output Y of the photoacoustic signal processed by the low-pass filter into a Kalman filtering unit for Kalman filtering noise reduction, and continuously predicting and correcting an output value by using the Kalman filtering unit to further reduce the superposed noise: let the effective value after the low-pass filter processing be X' (t),which is superimposed on a noise of

Then at time t there is a transformation of the output Y of the low pass filter into: x (t) ═ X' (t) + w (t);

let the effective value of the output obtained by measurement be z (t), the measurement error be v (t), and z (t) ═ x (t) + v (t); w (t) and V (t) are both time-varying Gaussian noise portions, the variances of W (t) and V (t) being

Covariance

According to the output X (t) of the low-pass filter at the time t, the estimated value X (t +1) of the output of the low-pass filter at the time t +1 is solved as follows:

and (3) estimating value at the next moment: x (t +1) | (t) ═ X (t);

the covariance matrix is predicted as: p (t +1) | (t) ═ P (t) + w (t);

the state is updated as follows:

x (t +1) ═ X (t +1) | (t) + K (t +1) [ Z (t +1) -X (t +1) | (t) ]; k (t +1) is the Kalman gain;

the covariance matrix is updated as: p (t +1) ═ 1-K (t +1) ] P (t +1) | (t);

the effective value after filtering and noise reduction by the Kalman filtering unit is as follows:

w "(t) is the output noise after Kalman filtering unit filtering noise reduction, W" (t)<W (t); the intensity of noise contained in the effective value is further reduced;

the mean value and the covariance of the Kalman filtering unit have linear transitivity, an output value at each moment is recorded into a posterior estimation value at the next moment through multiplication, the posterior estimation value at the moment contains information of all previous measurement values through continuous multiplication and iteration, and the system error is reduced. On the other hand, the invention can prevent the signal distortion from generating singular values, improve the stability of the on-line monitoring system, and save the hardware resources of the system compared with reducing the cut-off frequency of the low-pass filter. The Kalman filtering unit is a Kalman filter.

S5: the output signal processing unit further calculates the gas concentration of the acousto-optic cell according to the result of the Kalman filtering denoising; the relationship between the photoacoustic signal and the gas concentration is: u. of_λ＝aσ_λI_λC_cell；

Wherein S_i(t) intensity of photoacoustic signal excited in the photoacoustic cell, a is proportionality coefficient, σ_λIs the absorption coefficient of the gas at the wavelength λ; i is_λIs the intensity of the incident light at wavelength λ; c_cellIs the cell constant of the photoacoustic cell; the light for excitation is a monochromatic laser or infrared light, and σ is determined according to the change of gas concentration_λWill also follow the changes.

Cell constant C of acousto-optic cell_cellThe photoacoustic conversion capacity of the photoacoustic cell is determined, the smaller the radius of the resonant cavity of the photoacoustic cell is, the larger the length of the resonant cavity is, the larger the cell constant is, but the radius of the resonant cavity cannot be too small, so that the difficulty of light beam collimation can be increased, and the material with good heat conduction performance and small gas adsorption property is actually selected to be used for manufacturing the photoacoustic cell.

As shown in fig. 3-5, fig. 3 shows a schematic diagram of noise reduction effect of a conventional signal noise reduction method; FIG. 4 is a schematic diagram illustrating the noise reduction effect of the signal noise reduction method according to the present invention; fig. 5 shows a schematic time delay diagram corresponding to a lower cut-off frequency when the conventional signal extraction method achieves the same noise reduction effect of the present invention.

FIG. 6 is a pair C₂H₂The concentration monitoring result is schematic, and the concentration comparison of the existing noise reduction method and the noise reduction method of the invention is shown respectively, so that the detection result is smoother and more stable by adopting the noise reduction system and the noise reduction method of the invention.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.