阅读说明：本技术 一种淡水水体气泡的声学信号识别方法 (A kind of acoustic signal recognition methods of poisons in freshwater bubble ) 是由 武智 李捷 李新辉 朱书礼 杨计平 李跃飞 于 2019-09-04 设计创作，主要内容包括：本发明涉及淡水水体气泡的声学信号识别技术领域,且公开了一种淡水水体气泡的声学信号识别方法,包括以下步骤：S1、声学数据获取；S2、声学数据分析；S3、信号识别；S4、结果验证；其中S1还包括以下步骤：A1、选取水库或湖泊作为研究水体,将鱼探仪换能器垂直向下固定于船舷,吃水深度约0.5m,船速≤3km/h,A2、航线走向为垂直于气泡的密度梯度线,采用“之”字型或平行断面两种方式。本发明通过采集并分析原始声学数据,探讨水下气体的信号特征、运动规律、识别方法,为分析河流、湖泊、水库底质与演变提供一种思路,同时在资源评估中,可以减少信号误判,提高资源评估准确度。(The present invention relates to the acoustic signal identification technology fields of poisons in freshwater bubble, and disclose a kind of acoustic signal recognition methods of poisons in freshwater bubble, comprising the following steps: S1, acoustic data acquisition；S2, acoustic data analysis；S3, signal identification；S4, result verification；Wherein S1 is further comprising the steps of: A1, choosing reservoir or lake as research water body, energy transducer of fish finder is fixed on ship side vertically downward, draft about 0.5m, ship's speed≤3km/h, A2, course line trend are the density gradient line perpendicular to bubble, using " it " font or parallel cutting surfaces two ways.The present invention is by acquiring and analyzing original acoustic data, signal characteristic, the characteristics of motion, recognition methods of underwater gas are inquired into, provides a kind of thinking for analysis river, lake, reservoir substrate and differentiation, while in stock assessment, signal erroneous judgement can be reduced, stock assessment accuracy is improved.)

1. a kind of acoustic signal recognition methods of poisons in freshwater bubble, it is characterised in that: the following steps are included:

S1, acoustic data acquisition；

S2, acoustic data analysis；

S3, signal identification；

S4, result verification；

Wherein S1 is further comprising the steps of:

A1, reservoir or lake are chosen as research water body, energy transducer of fish finder is fixed on ship side, draft vertically downward 0.5m, ship's speed≤3km/h；

A2, course line trend are the density gradient line perpendicular to bubble, using " it " font or parallel cutting surfaces two ways；

Wherein S2 is further comprising the steps of:

B1, objective monomer detection method (STD, single target is used to original acoustic data (.raw format) Detection) echo image is analyzed；

B2, the distance (R, m) according to target, pass through 40log₁₀(R) TVG compensates the echo-signal (voltage) received；

B3, trajectory track is carried out to the monomer signal extracted after STD using fish body method for tracing (FT, fish tracking)；

B4, output echo signal；

Wherein S3 is further comprising the steps of:

C1, as ship's speed≤3km/h, same target continuously appears in more ping (pulse) in being greater than 5m water layer, opposite fish body Signal, bubble are constantly in propradation, and horizontal displacement very little, pass through the distinguishable fish body of TS reflectogram and bubble；

C2, by variables such as the echo forms, TS value, running track of feature and signal in C1, manually choose and to determine very much Classical bubble signal and fish body signal are used as " training sample ", to determine the initial center of each group, by discriminant analysis come really Determine to reject useless variable to contributive variable is differentiated；

C3, using principal component analysis (PCA) to differentiating that contributive variable analyzes, the variable using characteristic value greater than 1 as The significant variable of the ingredient；

C4, all data by output, variable of the characteristic value greater than 1 chosen in principal component analysis carry out discriminant analysis again, To determine the bubble signals in data；

Wherein S4 is further comprising the steps of:

D1, it will identify that the bubble signals come are calculated by the following formula out its radius, TS, speed, verified by statistical analysis Whether meet in uphill process, radius be gradually reduced or in uphill process TS be gradually reduced or in uphill process speed by These decrescence slow conditions:

V=h/t

In formula, v is bobble rise velocity, and h is the vertical change (cm) in wave beam, and t is the time (s) in wave beam, and r is gas It steeps radius (cm), δ_bsAcoustic scattering intensity.

2. a kind of acoustic signal recognition methods of poisons in freshwater bubble according to claim 1, it is characterised in that: step A1 In reservoir or lake depth be greater than 5m.

3. a kind of acoustic signal recognition methods of poisons in freshwater bubble according to claim 1, it is characterised in that: step A1 Middle energy transducer of fish finder wage frequency is 120kHz, 7.0 ° of power 200W, -3dB wave beam wide (beam width), pulse width (pulse duration)256μs。

4. a kind of acoustic signal recognition methods of poisons in freshwater bubble according to claim 1, it is characterised in that: minimum sound It learns threshold value and is set as -75dB, maximum wave beam compensation is set as 8dB, and calibration pulse length is (0.5-1.5), and pulse length determines Level is 6dB.

5. a kind of acoustic signal recognition methods of poisons in freshwater bubble according to claim 1, it is characterised in that: original sound It learns data (.raw) and uses acoustics the poster processing softProcessing is used according to the distance (R, m) of target 40log₁₀(R) TVG compensates the echo-signal (voltage) received automatically.

6. a kind of acoustic signal recognition methods of poisons in freshwater bubble according to claim 1, it is characterised in that: bubble and Resonant cavity is similar, makees homogeneous deformation in sound wave effect lower aprons, other variables calculate following formula:

σ_s=4 π σ_bs

In formula, r is bubble radius (cm), and ρ is ambient density, P₀It is the pressure for acting on bubble, γ is gas constant pressure specific heat Ratio, for the air under standard state, γ=1.41；δ_sFor bubble surface area, δ_bsAcoustic scattering intensity, V₀For gas Foam product, f₀For the resonance frequency of minute bubbles, P₀It is the pressure for acting on bubble.

Technical field

The present invention relates to the acoustic signal identification technology field of poisons in freshwater bubble, specially a kind of poisons in freshwater bubble Acoustic signal recognition methods.

Background technique

A series of Ecological and environmental problems that atmospheric greenhouse gas concentration increases and global warming causes have caused people To the extensive concern in greenhouse gases " source/remittance ".The aquatic ecosystem Yuan Hehui important as greenhouse gases, to atmospheric environment The variation of percent of greenhouse gases concentration has great influence.Greenhouse gases, which discharge process, has height temporal-spatial heterogeneity, and delivery mode is more Sample, and release process is influenced by many, leads to the uncertain increase for assessing its release.Currently, mainly there are following three kinds of methods To measure gas discharge: gas collector, video/picture and supersonic sounding technology.First two method can be in the region of restriction Interior quantization gas burst size.In water-borne sediment(s), due in gas discharge process, the temporal-spatial heterogeneity of height, and macroscopic view With microcosmic discontinuity, cause first two method that can not quantify gas release iron in big waters, and acoustic technique is then with it Quickly, the advantages such as area coverage is big allow to carry out noiseless assessment to the special heterogeneity of bubble.But using acoustic technique into When the release investigation of row water body bubble, since bubble target strength distribution is overlapped with water body Mesichthyes height, signal will lead to Erroneous judgement causes assessment result inaccurate.

Summary of the invention

It is an object of that present invention to provide a kind of acoustic signal recognition methods of poisons in freshwater bubble, can be improved in water body The accuracy of bubble identification reduces signal erroneous judgement, improves stock assessment accuracy.

To achieve the above object, the invention provides the following technical scheme: a kind of acoustic signal of poisons in freshwater bubble identifies Method, it is characterised in that: the following steps are included:

S1, acoustic data acquisition；

S2, acoustic data analysis；

S3, signal identification；

S4, result verification；

Wherein S1 is further comprising the steps of:

A1, reservoir or lake are chosen as research water body, energy transducer of fish finder is fixed on ship side vertically downward, drinking water is deep Spend 0.5m, ship's speed≤3km/h；

A2, course line trend are the density gradient line perpendicular to bubble, using " it " font or parallel cutting surfaces two ways；

Wherein S2 is further comprising the steps of:

B1, objective monomer detection method (STD, single target is used to original acoustic data (.raw format) Detection) echo image is analyzed；

B2, the distance (R, m) according to target, pass through 40log₁₀(R) TVG compensates the echo-signal (voltage) received；

B3, the monomer signal progress track extracted after STD is chased after using fish body method for tracing (FT, fish tracking) Track；

B4, output echo signal；

Wherein S3 is further comprising the steps of

C1, as ship's speed≤3km/h, same target continuously appears in more ping in being greater than 5m water layer, and opposite fish body is believed Number, bubble is constantly in propradation, and horizontal displacement very little, passes through the distinguishable fish body of TS reflectogram and bubble；

C2, by variables such as the echo forms, TS value, running track of feature and signal in C1, choose manually very true Fixed classical bubble signal and fish body signal are used as " training sample ", to determine the initial center of each group, pass through discriminant analysis It determines to contributive variable is differentiated, rejects useless variable；

C3, the contributive variable of differentiation is analyzed using principal component analysis (PCA), characteristic value is greater than to 1 variable Significant variable as the ingredient；

C4, all data by output choose variable of the characteristic value greater than 1 in principal component analysis and carry out differentiating again and divide Analysis, to determine the bubble signals in data；

Wherein S4 is further comprising the steps of:

D1, it will identify that the bubble signals come are calculated by the following formula out its radius, TS, speed, pass through statistical analysis It verifies whether to meet in uphill process, radius is gradually reduced or TS is gradually reduced or in uphill process middling speed in uphill process Spend these conditions that slow down gradually:

V=h/t

In formula, v is bobble rise velocity, and h is the vertical change (cm) in wave beam, and t is the time (s) in wave beam, r For bubble radius (cm), δ_bsAcoustic scattering intensity.

Preferably, the reservoir in step A1 or lake depth are greater than 5m.

Preferably, energy transducer of fish finder wage frequency is 120kHz, the wide (beam of power 200W, -3dB wave beam in step A1 Width) 7.0 °, 256 μ s of pulse width (pulse duration).

Preferably, minimal acoustic threshold value is set as -75dB, and maximum wave beam compensation is set as 8dB, and calibration pulse length is (0.5-1.5), pulse length determine that horizontal is 6dB.

Preferably, original acoustic data (.raw) use acoustics the poster processing soft Echoview4.Processing, according to target Distance (R, m), compensate the echo-signal (voltage) received automatically by time-varying gain (TVG).Echo, which counts, to be used 40log₁₀(R)TVG。

Preferably, bubble resonant cavity is similar, makees homogeneous deformation in sound wave effect lower aprons, other variables calculate as follows Formula:

σ_s=4 π σ_bs

In formula, r is bubble radius (cm), and ρ is ambient density, P₀It is the pressure for acting on bubble, γ is gas etc. The ratio of pressure ratio heat, for the air under standard state, γ=1.41；δ_sFor bubble surface area, δ_bsAcoustic scattering intensity, V₀For bubble volume.f₀For the resonance frequency of minute bubbles.P₀It is the pressure for acting on bubble.

The present invention provides a kind of acoustic signal recognition methods of poisons in freshwater bubble.Have it is following the utility model has the advantages that

The present invention inquires into signal characteristic, the characteristics of motion, identification of underwater gas by acquiring and analyzing original acoustic data Method provides a kind of thinking for analysis river, lake, reservoir substrate and differentiation, while in stock assessment, it is possible to reduce signal Erroneous judgement improves stock assessment accuracy.

Detailed description of the invention

Fig. 1 is bubble acoustic echo image map of the present invention；

Fig. 2 be bubble of the present invention in uphill process TS with the depth of water variation tendency；

Fig. 3 is the relational graph of bubble radius of the present invention and the depth of water；

Fig. 4 is different type bubble TS frequency distribution of the present invention；

Fig. 5 is the case figure of fish and bubble signals difference variable.

Specific embodiment

Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.

Examples of the embodiments are shown in the accompanying drawings, and in which the same or similar labels are throughly indicated identical or classes As element or element with the same or similar functions.The embodiments described below with reference to the accompanying drawings are exemplary, purport It is being used to explain the present invention, and is being not considered as limiting the invention.

In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", " length ", " width ", " thickness ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom" "inner", "outside", " up time The orientation or positional relationship of the instructions such as needle ", " counterclockwise ", " axial direction ", " radial direction ", " circumferential direction " be orientation based on the figure or Positional relationship is merely for convenience of description of the present invention and simplification of the description, rather than the device or element of indication or suggestion meaning must There must be specific orientation, be constructed and operated in a specific orientation, therefore be not considered as limiting the invention.

In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc. Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect It connects, is also possible to be electrically connected；It can be directly connected, can also can be in two elements indirectly connected through an intermediary The interaction relationship of the connection in portion or two elements.It for the ordinary skill in the art, can be according to specific feelings Condition understands the concrete meaning of above-mentioned term in the present invention.

The present invention provides a kind of technical solution: a kind of acoustic signal recognition methods of poisons in freshwater bubble, including following step It is rapid:

S1, acoustic data acquisition.

S2, acoustic data analysis.

S3, signal identification.

S4, result verification.

Wherein S1 is further comprising the steps of:

A1, reservoir or lake are chosen as research water body, energy transducer of fish finder is fixed on ship side vertically downward, drinking water is deep About 0.5m, ship's speed≤3km/h are spent, reservoir or lake depth are greater than 5m, and energy transducer of fish finder wage frequency is 120kHz, power 7.0 ° of 200W, -3dB wave beam wide (beam width), 256 μ s of pulse width (pulse duration),

A2, course line trend are the density gradient line perpendicular to bubble, using " it " font or parallel cutting surfaces two ways.

Wherein S2 is further comprising the steps of:

B1, objective monomer detection method (STD, single target is used to original acoustic data (.raw format) Detection) echo image is analyzed, original acoustic data (.raw) use acoustics the poster processing soft Echoview4.Processing.

B2, the distance (R, m) according to target, using 40log₁₀(R) TVG compensates the echo-signal (voltage) received.Most Small acoustics threshold value is set as -75dB, and maximum wave beam compensation is set as 8dB, and calibration pulse length is (0.5-1.5), pulse length Determine that horizontal is 6dB.

B3, the monomer signal progress track extracted after STD is chased after using fish body method for tracing (FT, fish tracking) Track.

B4, output echo signal.

Wherein S3 is further comprising the steps of:

C1, as ship's speed≤3km/h, same target continuously appears in more ping in being greater than 5m water layer, and opposite fish body is believed Number, bubble is constantly in propradation, and horizontal displacement very little, passes through the distinguishable fish body of TS reflectogram and bubble.

C2, by variables such as the echo forms, TS value, running track of feature and signal in C1, choose manually very true Fixed classical bubble signal and fish body signal are used as " training sample ", to determine the initial center of each group, pass through discriminant analysis It determines to contributive variable is differentiated, rejects useless variable.

C3, the contributive variable of differentiation is analyzed using principal component analysis (PCA), characteristic value is greater than to 1 variable Significant variable as the ingredient.

C4, all data by output choose variable of the characteristic value greater than 1 in principal component analysis and carry out differentiating again and divide Analysis, to determine the bubble signals in data.

Wherein S4 is further comprising the steps of:

D1, it will identify that the bubble signals come are calculated by the following formula out its radius, TS, speed, pass through statistical analysis It verifies whether to meet in uphill process, radius is gradually reduced or TS is gradually reduced or in uphill process middling speed in uphill process Spend these conditions that slow down gradually:

V=h/t

In formula, v is bobble rise velocity, and h is the vertical change (cm) in wave beam, and t is the time (s) in wave beam, r For bubble radius (cm), δ_bsAcoustic scattering intensity.

Bubble resonant cavity is similar, makees homogeneous deformation in sound wave effect lower aprons, other variables calculate following formula:

σ_s=4 π σ_bs

In formula, r is bubble radius (cm), and ρ is ambient density, P₀It is the pressure for acting on bubble, γ is gas etc. The ratio of pressure ratio heat, for the air under standard state, γ=1.41.δ_sFor bubble surface area, δ_bsAcoustic scattering intensity, V₀For bubble volume.f₀For the resonance frequency of minute bubbles.P₀It is the pressure for acting on bubble.

Such as Fig. 1, (a) bubble column in figure, ship's speed is about 5km/h.(b) is monomer signal TS vertical distribution in figure a in figure. (c) distinguishable bubble stream.(d) bubble divides in uphill process, and ship's speed is about 2km/h, and the present invention mainly has found to have following Several bubble types: (1) bubble column (bubble flares or bubble plume), shape and true bubble more phase Like (Fig. 1 a).(2) distinguishable single string bubble stream, when water flow and contrary research vessel navigation, bubble is in uphill process The echo image map such as Fig. 1 c will be generated.If the two direction is identical, bubble rising can generate the image being tilted to the right. (3) when ship's speed is slow enough, can see in reflectogram, bubble can occur to divide (see Fig. 1 d) in uphill process, this can It can be the result that air pocket is split into minute bubbles, it is also possible to which bubble similar in several positions comes from bottom storehouse rise simultaneously, but nothing Method is distinguished by sound system.

From Figure 2 it can be seen that bubble, in uphill process, TS is gradually reduced (Fig. 1 b also shows that this trend).In order to calculate gas The size of bubble, it is assumed that region bubble is near acoustic equipment frequency without resonance.It under this assumption, can according to formula conversion Know, mean radius 1.657mm, radius 0.203-21.649mm, 95% confidence interval is (0.294,3.019), is passed through Above-mentioned formula calculate bubble average external volume be 0.017mL, range be (0.00014,38.7) mL.The depth of water is divided one every 2m Water layer averages to each layer of bubble signals frequency distribution, as seen from Figure 3, splits data into different water levels in Fig. 3, and every layer It is spaced 2m, bubble radius in uphill process is gradually reduced, significant related (R2=0.953) to the depth of water.According to formula it is found that Under the energy converter working frequency of 120KHz, the resonance radius of investigation waters bubble is 0.027mm~0.055mm,

Fig. 4 is two kinds of different type bubble TS frequency distributions, and distribution moves towards almost the same, is distributed in single peak type, SBF is averaged TS as -60.29dB, and peak value is about -58dB.BFS is averaged TS as -56.97dB, and peak value is about -52dB.

When ship's speed is less than 3km/h, fish body signal and bubble have significant difference in acoustic echo image.Artificially have chosen 165 A typical acoustic signal (72 bubble signals, 93 fish body signals) is analyzed, and the variable exported to Echoview is passed through It is handled manually, rejects descriptive variable, variance analysis is carried out to remaining 22 variables using SPSS, the results show that Wherein between 12 variables there are significant difference (P=0), in conjunction with the meaning that the feature of two kinds of signals and each variable characterize, choose Out relative velocity, vertical direction variation, target vertical variation, distance and come embody bubble signals (BS) and fish signal (FS) it Between difference, each variable detail statistics result See Figure:

Relative velocity is the opposite swimming rate in target level direction in upper figure, 50% fish relative velocity 1.0~ Between 1.5m/s, 50% bubble relative velocity is between 0.8~1.2m/s.Vertical direction variation is primarily used to description signal Migratory direction (- 90,90), fish vertical direction variation have just have it is negative, 50% variation range be -3 °~5 ° between, and bubble hang down Straight direction change is all positive, and 50% variation range is 12 °~18 °, shows that bubble is rising always, and fish are then without obvious rule Rule.The vertical change of target trajectory is the difference of first signal and the last one signal in track, and bubble signals are all negative, Fish body then have just have it is negative.Distance and refer to the sum of the distance between adjacent monomer signal.Due to bubble horizontal displacement in wave beam It is smaller, and fish are then without evident regularity.

It can to sum up obtain, the present invention inquires into signal characteristic, the movement of underwater gas by acquiring and analyzing original acoustic data Rule, recognition methods, can bubble signals in automatic discrimination water-outlet body, provided for analysis river, lake, reservoir substrate and differentiation A kind of thinking, while in stock assessment, it is possible to reduce signal erroneous judgement improves stock assessment accuracy.