CN103884625B - Data-fusion-based reservoir bottom mud density detection method and system - Google Patents

Abstract

The invention discloses a data-fusion-based reservoir bottom mud density detection method and system. The system comprises a sampling device, a testing device and an analyzing device, wherein the sampling device is used for acquiring a bottom mud sample; the testing device is used for acquiring the relationship between the wave speed and depth of the bottom mud sample, the relationship between density and depth of the bottom mud sample, and the relationship between a reflection signal main frequency and the depth of the bottom mud sample; the analyzing device obtains the density of the bottom mud sample by analyzing and calculating the relationship between the density and the depth, the wave speed and the reflection signal main frequency; the testing device comprises a sub-bottom profiler, a transmitting transducer, a receiving transducer and a density testing instrument. The reservoir bottom ground layer reflection signal is recognized automatically by the method or the system provided by the invention; parameters of density and the like of a reservoir bottom ground layer are determined with combination of a low-disturbance sampling testing result, so that the bottom mud sampling is accurate, and the density detection is efficient.

Description

A kind of Reservoir Sediment density detection method and system based on data fusion

Technical field

The invention belongs to Reservoir Sediment Detection Techniques field, it is based especially on the Reservoir Sediment density detection side of data fusion Method and system.

Background technology

The identification of Reservoir Sediment feature is related to multiple detection methods, and data volume is larger, if different detections and measurement are tied Fruit integrates, and using certain data fusion method, then can preferably identify bed mud feature, use manpower and material resources sparingly, and reaches fast Speed accurately identifies.

Reservoir Sediment detection method includes multiple methods such as bed mud sampling, acoustic detection, prospecting by boring at present.Reservoir Sediment The sampling equipment that sampling uses has grab type, the box, sampling equipment such as freezing type, gravity type.The equipment of bed mud acoustic detection under water Mainly have: multibeam sounding system, sub-bottom profiler and side scan sonar equipment, they have similar operation principle: by sound Learn transducer and produce reflection and scatter echo, sound to detecting water body transmitting sound wave, sound wave after running into water-bed (seabed) or barrier Learn after transducer receives echo-signal, processing unit according to echo amplitude and phase calculation sound wave hourage, further according to surveying The Sound speed profile surveying water body calculates the actual propagation distance of sound wave, then according to transmitting angle of release, and the attitude ginseng of motion sensor Number calculates position and the water depth value of echo-signal, also records amplitude and the strength information of echo simultaneously.Data fusion technique, bag Include the collection to the useful information that various information sources provide, transmission, synthesis, filtration, correlation and synthesize, so that auxiliary people are carried out Detection, analysis, situation judge, plan, verifying, diagnosing etc..Grab type, the box, sampling equipment such as freezing type, gravity type are mainly deposited In two large problems: perturbed problem and sample leakage problem during sampling to sample during sampling, samples taken testing result can be led to Deviation larger it is impossible to represent the genuine property of bed mud well.

Carry out acoustic detection using sub-bottom profiler, applicable depth of water scope is big, and penetrate reservoir area bed mud thickness greatly, but It is to produce reflection echo after running into water-bed (seabed) or barrier, echo information characterizes and runs into different medium object Interface, but cannot specifically quantify to draw the density data of diverse location bed mud it is impossible to inquire into alluvial mud using information of acoustic wave Husky density feature.

Content of the invention

The technical problem to be solved in the present invention there is provided a kind of Reservoir Sediment density detection method based on data fusion And system, using the method for the present invention or system automatic identification bottom of the reservior stratum reflected signal, in conjunction with low disturbance sample test results, Judge the parameters such as bottom of the reservior density of earth formations, reach bed mud sampling more accurately, density detection is more efficient.

The invention provides a kind of Reservoir Sediment density detection system based on data fusion, described system includes sampling and sets Standby, test equipment and analytical equipment, wherein, described sampling equipment is used for obtaining bed mud sample；Described test equipment is used for obtaining The velocity of wave of bed mud sample and the relation of depth, the relation of the relation, reflected signal dominant frequency and depth of density and depth；Described analysis Equipment passes through the relation between analytical calculation density and depth, velocity of wave, reflected signal dominant frequency, obtains the close of described bed mud sample Degree.

Further, described sampling equipment is the low disturbance sampling equipment based on gravity type piston technology, including guiding device, Triggering device, sampler, the bushing pipe of described sampler is transparent pc pipe, and internal diameter is 90mm, pipe range is 10m.

Further, described test equipment includes sub-bottom profiler, transmitting transducer, receive transducer and density test instrument Device, described transmitting transducer is consistent with the transmission signal parameters of described sub-bottom profiler.

Further, the operating frequency of described sub-bottom profiler is 2-20khz, and the Reservoir Sediment thickness that can detect is 0- 100m, operating depth scope 0-500m.

Present invention also offers a kind of Reservoir Sediment density detection method based on data fusion, wherein, methods described bag Include:

Step 2, test ship carry the measuring point that sampling equipment selectes in reservoir and carry out low disturbance sampling work, obtain Bed mud Sample storage in bushing pipe, send laboratory after bushing pipe is sealed back to；

Step 3, taken bed mud sample depth are sample to be divided into n block, every piece of thickness is dz, carries out bed mud sample Acoustical testing, obtains SVEL data, and velocity of wave data is fitted to the function v of depth z₁Z (), by fitting function v₁Z () exists Depth z ＞ z₁Interval is expanded；

Step 4, above-mentioned bed mud sample carry out density measurement by density test instrument, and density data are fitted to depth The function ρ of degree z₁Z (), by fitting function in depth z ＞ z₁Interval is expanded；

Step 5, sound wave through transmission technique detect the absorption function of every piece of bed mud sample: transmitting transducer transmitting linear frequency modulation letter Number (lfm) s_lfm, this signal is consistent with the transmission signal parameters of sub-bottom profiler, and the receipt signal of receive transducer is(i= 1～n), then the receptance function of bed mud sample isFft is Fourier transformation, because reflected signal passes for round trip Broadcast, do not consider the impact of reflectance factor, then the test block receptance function of round trip travelling is:At depth * dz if There is reflecting layer, do not consider reflective index impacts, then reflected signal frequency-domain function is

$s_{\mathrm{reflect}}^k=\mathrm{fft}\left(s_{\mathrm{lfm}}\right)\·\underset{i=1}{\overset{k}{\mathrm{\π}}}{\mathrm{sd}}_i^c$

TakeDominant frequency, obtains reflected signalDominant frequency with depth z change function；

Receipt signal s of sub-bottom profiler_mea(t),s_meaT () time span is t_mea, by s_meaT () carries out time domain and adds Window, t represents window time length, t₀Represent time delay (0 ＜ t₀＜ t_mea), take Fourier transformation fft () of windowing signal, led Frequency is f, and the corresponding depth of this signal meets function；

Step 6, increase to from 0, if () is unsatisfactory for function f, the silt density of depth is that velocity of longitudinal wave is；As Really () meets function f, then judge there is reflecting layer in depth.

Further, before described step 2, also step one, test ship carries sub-bottom profiler, records one The sonic data of scanning on survey line, the reservoir bottom configuration according to ultra sonic scanner data conversion, phase is selected on survey line To flat position as low disturbance sampling work measuring point.

Further, the acquisition of described bed mud sample is divided into: decentralization process, trigger process, sampling process and removal process four In the individual stage, when sampling equipment is transferred to and also had a segment distance apart from river bed, counterweight block first touches riverbed, and lever loses flat Weighing apparatus, triggers relieving mechanism action, and sampler subject freedom falls, and in insertion drift in the presence of deadweight and inertia, takes Sampling device lower end to be cut off using pawl spring structure and to hold drift；Because the suction that piston produces makes sampler inwall and sample Friction dynamic balance between product is offset, and is just filled with the bed mud sample of low disturbance in sampler.

Further, the thickness l of described bed mud sample_c, add water in water tank, fixed transmission transducer and receive transducer, send out Penetrate transducer transmission signal, record receive transducer receives time t₁, obtain the velocity of wave v of water_w=l₁/t₁；Transmitting transducer and connecing Receive and between transducer, place bed mud sample, with through transmission technique detection, obtain the propagation time t that sound wave penetrates bed mud sample₂, then bed mud The velocity of wave of sample is v_c=l_c/(t₂-l₂/v_w-l₃/v_w).

Compared with prior art, the positive effect of the present invention is:

(1) use low disturbance sampling equipment, longer reservoir area bed mud original state sample can be obtained, counter bored holes exploration takes Sample, gravity type sampling, grab type sampling, improve efficiency, reduce cost, alleviate the disturbance shadow to sample for the sampling work Ring, be obtained in that more accurately bed mud physical characteristic parameter.

(2) carry out ultra sonic scanner using sub-bottom profiler, acoustic emission frequency is relatively low, the sound wave electrical pulse energy of generation Larger, there is stronger penetration power, be capable of the stratum of effective ten meters of the through-fall truth of a matter, and detection efficient is high, accuracy is high.

(3) utilize laboratory test to obtain data and the density data of reservoir area bed mud velocity of wave, two kinds of data are fitted to difference Function curve, find out the dependency relation of the two, function rationally expanded, can inquire into and samples taken depth to go to the bottom The density feature of mud, thus obtain reservoir area bed mud more deeply spend lower density characterisitic parameter.

(4) compared using the sonic data and the curve set up of sub-bottom profiler scanning, special using signal spectrum Levy quick identification formation parameter, be conducive to reservoir area sediment layer reason analysis.

(5) form a set of efficient Reservoir Sediment density detection method, the method is suitable for reservoir scope extensively, can be to be applicable To other numerous reservoirs, improve Reservoir Sediment detection efficient, provide important basic data for reservoir harness and development.

Brief description

Fig. 1 is sampling process schematic diagram；

Fig. 2 is indoor SVEL test schematic diagram.

Specific embodiment

The purpose of the present invention is the data processing method being proposed based on Data fusion technique, and integrated low disturbance sampling equipment takes Sample detection data and sub-bottom profiler acoustic detection data, form a set of efficient Reservoir Sediment detection method such that it is able to fast Speed obtains density and the structure of Reservoir Sediment, can provide basic data for more in depth analysis and research water sand migration rule, right Huge reservoir flood control sand discharge is significant.

Below in conjunction with the accompanying drawings, technical scheme is further described in detail.

Yellow River In Recent Years Institute of Hydro-Technical Research uses for reference deep-sea drift fidelity sampling principle, transforms and develops based on weight The low disturbance column deep water reservoir area sampling equipment of power formula piston technology: Yang Yong, " huge reservoir is low written by Zhang Qingxia, Chen Hao, Zheng Jun The design of disturbance sampling equipment and performance evaluation [c] ". the 5th the Yellow River international symposium paper, 2012.09；Yang Yong, Zheng Jun, Chen Hao Written by " huge reservoir low disturbance sampling equipment Machine Design ". Advances In Science And Technology of Water Resources, 2012,32 (s2): 18-19: design Sampling equipment internal diameter 90mm, pipe range 10m.Principle is: sampler subject freedom falls, and inserts in the presence of deadweight and inertia Enter in drift, sampler lower end to be cut off using pawl spring structure and to hold drift.Because the suction that piston produces makes to take Friction dynamic balance between sampling device inwall and sample is offset, and so, is just filled with the drift of low disturbance in sampler Sample.The deep water sampling preliminary test carrying out in Reservoir Area of Sanmenxia shows, for soft channel aggradation silt, can obtain thickness The original state cylindrical sample of more than 3m, but for hard shoal materials, because soil particle angle of friction is larger, and shoal materials is close Solidity is larger, leads to sampling effect undesirable.

The method of the present invention is to obtain Reservoir Sediment sample (being also called test block) by low disturbance sampling equipment, through shallow Layer section plotter test obtains bed mud physical property with change in depth rule, carries out acoustical testing further across sub-bottom profiler, obtains To the attenuation change function of acoustic propagation characteristic, this function is applied to the analysis of the test data of shallow seismic profile instrument, prediction Weak reflection layer whether there is, and in the case of there is not reflecting layer, its bed mud physical property meets in-situ test prediction of result function.

It is cylindric sample that low disturbance sampling equipment obtains original state sample, and diameter is less than or equal to 90mm, length be less than or Equal to 8m, sample is installed in bushing pipe, and bushing pipe is that transparent pc manages it is easy to cutting section, with the naked eye can clearly distinguish taking-up sample Layering, color and proterties.

Reservoir Sediment acoustic detection scanning work uses sub-bottom profiler, and its operating frequency is 2-20khz, can Detecting reservoir area bed mud thickness is 0-100m, operating depth scope 0-500m.

Based on the Reservoir Sediment detection method of data fusion, mainly comprise the steps:

Step one, test ship carry sub-bottom profiler, and the sonic data of scanning on one survey line of record, according to sound wave The shown reservoir bottom configuration of scan data conversion, selectes relatively flat position as low disturbance sampling work on survey line Measuring point；

Step 2 is as shown in figure 1, low disturbance sampling equipment is mainly by parts such as guiding device, triggering device, samplers Composition.The acquisition of test block is roughly divided into: decentralization process, trigger process, sampling process and removal process four-stage, when sampling sets Standby transfer to apart from river bed also have a segment distance when, counterweight block first touches riverbed, lever disequilibrium, and triggering relieving mechanism moves Make, sampler subject freedom falls, in insertion drift in the presence of deadweight and inertia, sampler lower end adopts pawl spring Structure is cutting off and to hold drift.Because the suction that piston produces makes the friction dynamic balance between sampler inwall and sample Offset, in sampler, so, be just filled with the bed mud sample of low disturbance.Test ship carries sampling equipment and selectes in reservoir Good measuring point carries out low disturbance sampling work, and samples taken is saved in bushing pipe, and bushing pipe good seal is sent back to laboratory；

Step 3 is as shown in Fig. 2 the thickness l of bed mud sample_c, add water in water tank, fixed transmission transducer and receive transducing Device (in test process, test block and transducer are all under water), transmitting transducer transmission signal, when record receive transducer receives Between t₁, obtain the velocity of wave v of water_w=l₁/t₁.Place sample between transmitting transducer and receive transducer, in bed mud sample both sides, use Through transmission technique detects, obtains the propagation time t that sound wave penetrates test block₂, then the velocity of wave of test block is v_c=l_c/(t₂-l₂/v_w-l₃/v_w).Institute Bed mud sample depth is taken to be sample to be divided into n block, every piece of thickness is dz, carries out the acoustical testing of bed mud sample, obtains sound wave Speed data, and velocity of wave data is fitted to the function v of depth z₁Z (), by fitting function v₁Z () is in depth z ＞ z₁Interval is carried out Expand；

Step 4, by density test instrument, bed mud sample rate test is carried out to above-mentioned sample, and density data is intended It is combined into the function ρ of depth z₁Z (), by fitting function in depth z ＞ z₁Interval is expanded；

Step 5, sound wave through transmission technique detect the absorption function of every piece of sample: transmitting transducer transmitting linear FM signal (lfm) s_lfm, this signal is consistent with the transmission signal parameters of sub-bottom profiler, and receipt signal is(i=1～n), then sample The receptance function of test block isFft is Fourier transformation, because reflected signal is propagated for round trip, does not consider anti- Penetrate the impact of coefficient, then the dielectric response function of round trip travelling is:If there are reflecting layer at depth * dz, no Consider reflective index impacts, then reflected signal frequency-domain function is

$s_{\mathrm{reflect}}^k=\mathrm{fft}\left(s_{\mathrm{lfm}}\right)\·\underset{i=1}{\overset{k}{\mathrm{\π}}}{\mathrm{sd}}_i^c$

TakeDominant frequency.Obtain reflected signalDominant frequency with depth z change function.

Receipt signal s of sub-bottom profiler_mea(t), s_meaT () time span is t_mea.By s_meaT () carries out time domain and adds Window, t represents window time length, t₀Represent time delay (0 ＜ t₀＜ t_mea), take Fourier transformation fft () of windowing signal, led Frequency is f, and the corresponding depth of this signal meets function.

Step 6, increase to from 0, if () is unsatisfactory for function f, the silt density of depth is that velocity of longitudinal wave is；As Really () meets function f, then judge there is reflecting layer in depth.

Described step 3, to step 6, can realize standardization by computer program with reference to computer hardware, Effectively said method is promoted.

Compared with prior art, the positive effect of the present invention is:

(1) use low disturbance sampling equipment, longer reservoir area bed mud original state sample can be obtained, counter bored holes exploration takes Sample, gravity type sampling, grab type sampling, improve efficiency, reduce cost, alleviate the disturbance shadow to sample for the sampling work Ring, be obtained in that more accurately bed mud physical characteristic parameter.

(2) carry out ultra sonic scanner using sub-bottom profiler, acoustic emission frequency is relatively low, the sound wave electrical pulse energy of generation Larger, there is stronger penetration power, be capable of the stratum of effective ten meters of the through-fall truth of a matter, and detection efficient is high, accuracy is high.

(3) utilize laboratory test to obtain data and the density data of reservoir area bed mud velocity of wave, two kinds of data are fitted to difference Function curve, find out the dependency relation of the two, function rationally expanded, can inquire into and samples taken depth to go to the bottom The density feature of mud, thus obtain reservoir area bed mud more deeply spend lower density characterisitic parameter.

(4) compared using the sonic data and the curve set up of sub-bottom profiler scanning, special using signal spectrum Levy quick identification formation parameter, be conducive to reservoir area sediment layer reason analysis.

(5) form a set of efficient Reservoir Sediment density detection method, the method is suitable for reservoir scope extensively, can be to be applicable To other numerous reservoirs, improve Reservoir Sediment detection efficient, provide important basic data for reservoir harness and development.

Claims (1)

1. a kind of detection method of the Reservoir Sediment density detection system based on data fusion, described system include sampling equipment, Test equipment and analytical equipment it is characterised in that: described sampling equipment is used for obtaining bed mud sample；Described test equipment is used for obtaining Take the relation of the relation, reflected signal dominant frequency and depth of the velocity of wave of bed mud sample and the relation of depth, density and depth；Described point Desorption device passes through the relation between analytical calculation density and depth, velocity of wave, reflected signal dominant frequency, obtains the close of described bed mud sample Degree；Described sampling equipment is the low disturbance sampling equipment based on gravity type piston technology, including guiding device, triggering device, takes Sampling device, the bushing pipe of described sampler is transparent pc pipe, and internal diameter is 90mm, pipe range is 10m；Described test equipment is included shallowly Layer section plotter, transmitting transducer, receive transducer and density test instrument, described transmitting transducer and described sub-bottom profiler Transmission signal parameters consistent；The operating frequency of described sub-bottom profiler is 2-20khz, the Reservoir Sediment thickness that can detect For 0-100m, operating depth scope 0-500m；

Methods described includes:

Step one, test ship carries sub-bottom profiler, and the sonic data of scanning on one survey line of record, according to ultra sonic scanner Reservoir bottom configuration shown by data conversion, selectes relatively flat position as the survey of low disturbance sampling work on survey line Point；

Step 2, test ship carry the measuring point that sampling equipment selectes in reservoir and carry out low disturbance sampling work, the bottom of acquisition Mud Sample storage, in bushing pipe, sends laboratory back to after bushing pipe is sealed；Wherein, the acquisition of described bed mud sample is divided into: under let slip Journey, trigger process, sampling process and removal process four-stage, transfer to when sampling equipment and also have a segment distance apart from river bed When, counterweight block first touches riverbed, lever disequilibrium, triggers relieving mechanism action, and sampler subject freedom falls, certainly In insertion drift in the presence of weight and inertia, sampler lower end to be cut off using pawl spring structure and to hold drift；Due to The suction that piston produces makes the friction dynamic balance between sampler inwall and sample offset, and is just filled with low in sampler The bed mud sample of disturbance；

Step 3, taken bed mud sample depth are z, sample are divided into n block, every piece of thickness is dz, carries out the sound of bed mud sample Learn test, obtain SVEL data, and velocity of wave data is fitted to function v (z) of depth z, by fitting function v (z) in depth Degree z ＞ z₁Interval is expanded；The assay method of described velocity of wave is, will be equal to bed mud sample, transmitting transducer and receive transducer It is placed in water tank, in test process, bed mud sample, transmitting transducer and receive transducer are all under water；Described bed mud sample Thickness l_c, add water in water tank, fixed transmission transducer and receive transducer, transmitting transducer transmission signal, record receives transducing Device receives time t₁, obtain the velocity of wave v of water_w=l₁/t₁, wherein l₁For the distance between transmitting transducer and receive transducer；Send out Penetrate placement bed mud sample between transducer and receive transducer, with through transmission technique detection, obtain the propagation that sound wave penetrates bed mud sample Time t₂, then the velocity of wave of bed mud sample is v_c=l_c/(t₂-l₂/v_w-l₃/v_w), wherein, l₂For transmitting transducer and bed mud sample it Between distance, l₃For the distance between receive transducer and bed mud sample；

Step 4, above-mentioned bed mud sample carry out density measurement by density test instrument, and density data are fitted to depth z Function ρ (z), by fitting function in depth z ＞ z₁Interval is expanded；

Step 5, sound wave through transmission technique detect the absorption function of every piece of bed mud sample: transmitting transducer transmitting linear FM signal (lfm)s_lfm, this signal is consistent with the transmission signal parameters of sub-bottom profiler, and the receipt signal of receive transducer isThen bottom The receptance function of mud sample isFft is Fourier transformation, because reflected signal is propagated for round trip, does not consider anti- Penetrate the impact of coefficient, then the test block receptance function of round trip travelling is:If there are reflecting layer at depth z, do not consider Reflective index impacts, then reflected signal frequency-domain function beTakeDominant frequency f_k, obtain reflection letter NumberDominant frequency with depth z change function f (z)；

Step 6, receipt signal s of sub-bottom profiler_mea(t), s_mea(t)Time span is t_mea, by s_mea(t)Carry out time domain to add Window, obtains s_winT (), takes the Fourier transformation fft (s of windowing signal_win(t)), obtaining dominant frequency is f, this signal and corresponding depth z Meet certain function；

Step 7, increase to t from 0_meaIf f meets function f (z), judge to have reflecting layer at depth z；If f is unsatisfactory for Function f (z), then the sediment characteristics at depth z meet data matching ρ (z) of density, data matching v (z) of velocity of wave.