CN115291288A - Iron neutron mark-based while-drilling pulse neutron porosity intelligent processing method - Google Patents
Iron neutron mark-based while-drilling pulse neutron porosity intelligent processing method Download PDFInfo
- Publication number
- CN115291288A CN115291288A CN202211194628.0A CN202211194628A CN115291288A CN 115291288 A CN115291288 A CN 115291288A CN 202211194628 A CN202211194628 A CN 202211194628A CN 115291288 A CN115291288 A CN 115291288A
- Authority
- CN
- China
- Prior art keywords
- drilling
- porosity
- neutron
- pulse
- neutrons
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005553 drilling Methods 0.000 title claims abstract description 221
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 title claims abstract description 99
- 229910052742 iron Inorganic materials 0.000 title claims abstract description 38
- 238000003672 processing method Methods 0.000 title claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 78
- 238000004364 calculation method Methods 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 24
- 238000004088 simulation Methods 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 7
- 230000015572 biosynthetic process Effects 0.000 claims description 83
- 230000008447 perception Effects 0.000 claims description 44
- ZAMOUSCENKQFHK-UHFFFAOYSA-N Chlorine atom Chemical compound [Cl] ZAMOUSCENKQFHK-UHFFFAOYSA-N 0.000 claims description 27
- 239000000460 chlorine Substances 0.000 claims description 27
- 229910052801 chlorine Inorganic materials 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 26
- 238000005259 measurement Methods 0.000 claims description 18
- 150000001875 compounds Chemical class 0.000 claims description 17
- 235000019738 Limestone Nutrition 0.000 claims description 12
- 239000006028 limestone Substances 0.000 claims description 12
- 239000010459 dolomite Substances 0.000 claims description 11
- 229910000514 dolomite Inorganic materials 0.000 claims description 11
- 239000011435 rock Substances 0.000 claims description 11
- 230000033558 biomineral tissue development Effects 0.000 claims description 10
- 238000012795 verification Methods 0.000 claims description 9
- 239000003550 marker Substances 0.000 claims description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 6
- 239000004571 lime Substances 0.000 claims description 4
- 230000003213 activating effect Effects 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 241001637516 Polygonia c-album Species 0.000 claims 3
- 238000002372 labelling Methods 0.000 claims 1
- 238000010200 validation analysis Methods 0.000 claims 1
- 238000005755 formation reaction Methods 0.000 description 73
- 238000001228 spectrum Methods 0.000 description 20
- 230000035945 sensitivity Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 7
- 239000008398 formation water Substances 0.000 description 6
- 239000000126 substance Substances 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 239000003921 oil Substances 0.000 description 4
- OWUGOENUEKACGV-UHFFFAOYSA-N [Fe].[Ni].[W] Chemical compound [Fe].[Ni].[W] OWUGOENUEKACGV-UHFFFAOYSA-N 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000002002 slurry Substances 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000004907 flux Effects 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- OYPRJOBELJOOCE-UHFFFAOYSA-N Calcium Chemical compound [Ca] OYPRJOBELJOOCE-UHFFFAOYSA-N 0.000 description 1
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052790 beryllium Inorganic materials 0.000 description 1
- 229910052791 calcium Inorganic materials 0.000 description 1
- 239000011575 calcium Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V5/00—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity
- G01V5/04—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity specially adapted for well-logging
- G01V5/08—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity specially adapted for well-logging using primary nuclear radiation sources or X-rays
- G01V5/10—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity specially adapted for well-logging using primary nuclear radiation sources or X-rays using neutron sources
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
Landscapes
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- High Energy & Nuclear Physics (AREA)
- Mining & Mineral Resources (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Fluid Mechanics (AREA)
- Environmental & Geological Engineering (AREA)
- Geochemistry & Mineralogy (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention discloses an intelligent processing method for porosity of pulse neutrons while drilling based on iron neutron marking, and relates to the technical field of geophysical logging in a mine field. The method includes the steps of measuring by a pulse-while-drilling neutron porosity logging instrument, measuring a standard graduated well by the pulse-while-drilling neutron porosity logging instrument, simulating by a Monte Carlo numerical simulation method, establishing a graduation relation between a simulation value and an actually-measured value of the pulse-while-drilling neutron porosity logging instrument, simulating by the Monte Carlo numerical simulation method to obtain pulse-while-drilling neutron porosity samples in different logging-while-drilling environments, constructing a sample database and a multilayer sensing neural network, training the multilayer sensing neural network by the sample database to calculate the pulse-while-drilling neutron porosity and verify the accuracy of a calculation result, inputting measured data of the pulse-while-drilling neutron porosity logging instrument after graduation processing into the multilayer sensing neural network, calculating the pulse-while-drilling neutron porosity, and improving the calculation accuracy of the pulse-while-drilling neutron porosity.
Description
Technical Field
The invention relates to the technical field of geophysical logging in a mine field, in particular to an intelligent processing method for porosity of pulse neutrons while drilling based on iron neutron marking.
Background
The accurate measurement of the formation porosity is taken as the key of high-precision formation evaluation, because the neutron porosity logging instrument adopting the traditional chemical source is limited by a series of problems of health, safety, environment and the like in development and application, the neutron porosity logging based on the pulse neutron source becomes the development direction of the porosity logging technology, the emission of neutrons can be controlled by adopting the D-T pulse neutron source, and the radiation risk of the traditional chemical source in the aspects of transportation, construction, storage and the like is avoided.
Compared with the traditional chemical source, the high-energy fast neutrons released by the pulse neutron source have the energy of 14MeV, and the neutron energy-space-time field distribution in the process of acting with the formation medium is different from that of the traditional chemical source, so that the porosity measurement sensitivity and the influence factors of the high-energy fast neutrons are greatly different from the porosity of the conventional compensation neutrons.
The logging-while-drilling technology is used as an important means applied to exploration and development of unconventional reservoirs, deep oil gas and deep water oil gas, the drill collar is always in a vibration state in the measurement-while-drilling process, so that a gap exists between the pulsed neutron porosity instrument and a well wall, and meanwhile mud, formation water mineralization and the like adopted in the measurement-while-drilling process can also influence the measurement result of the pulsed neutron porosity while drilling. The traditional neutron porosity processing method corrects the influence of the environment on the neutron porosity by establishing a plate, but in the actual measurement process, the instrument gap and the formation condition are usually difficult to directly obtain. Therefore, it is desirable to provide a method for correcting the influence of environmental factors by using the measurement information of the apparatus itself, so as to improve the measurement accuracy of the porosity of the pulsed neutrons while drilling.
Disclosure of Invention
Aiming at the problem that the measurement precision of a pulse neutron porosity while drilling result is low due to the influence of a measurement-while-drilling environment, the invention discloses an intelligent processing method for the pulse neutron porosity while drilling based on iron neutron marking.
In order to realize the purpose, the invention adopts the following technical scheme:
an intelligent processing method for porosity of pulse neutrons while drilling based on iron neutron marking adopts a pulse neutron porosity while drilling logging instrument to measure, and specifically comprises the following steps:
step 1), placing a pulse neutron porosity logging-while-drilling instrument in a standard calibration well for measurement, obtaining a thermal neutron count and a neutron time spectrum of a near neutron detector, a thermal neutron count and a neutron time spectrum of a far neutron detector and a gamma energy spectrum of a gamma detector, and calculating to obtain measured values of the pulse neutron porosity logging-while-drilling instrument, wherein the measured values comprise a measured value of a macroscopic capture cross section of a borehole medium, a measured value of a macroscopic capture cross section of a formation medium, a measured value of a thermal neutron count ratio, a measured value of a non-bomb gamma count, a measured value of a capture gamma count, a measured value of an iron element characteristic gamma count and a measured value of a chlorine element characteristic gamma count;
step 2), according to the instrument structure of the while-drilling pulse neutron porosity logging instrument, in combination with the structural parameters of a standard graduated well, establishing a while-drilling neutron porosity numerical calculation model by using a Monte Carlo numerical simulation method, and simulating to obtain a simulation value of the while-drilling pulse neutron porosity logging instrument, wherein the simulation value comprises a simulation value of a macroscopic capture cross section of a borehole medium, a simulation value of a macroscopic capture cross section of a formation medium, a simulation value of a thermal neutron count ratio, a simulation value of a non-elastic gamma counting, a simulation value of a capture gamma counting, a simulation value of an iron element characteristic gamma counting and a simulation value of a chlorine element characteristic gamma counting;
step 3), establishing a scale relation between the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument based on the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument;
step 4), changing the size of a drill collar of the pulse neutron porosity logging-while-drilling instrument according to the instrument structure of the pulse neutron porosity logging-while-drilling instrument, sequentially changing the lithology, the formation water mineralization, the drill collar gap, the mud type and the formation porosity of a formation medium under different drill collar sizes, reestablishing a neutron porosity numerical calculation model while drilling, simulating and obtaining the thermal neutron count and the neutron time spectrum of a near neutron detector, the thermal neutron count and the neutron time spectrum of a far neutron detector and the gamma energy spectrum of a gamma detector under different logging-while-drilling environments, and forming a plurality of groups of pulse neutron porosity samples while drilling according to the drill collar sizes, the macro capture section of a borehole medium, the macro capture section of the formation medium, the counting ratio of iron-marked thermal neutrons, the non-elastic gamma count, the gamma capture count and the gamma counting of chlorine element characteristics under different logging-while-drilling environments;
step 5), constructing a sample database based on the multiple groups of while-drilling pulse neutron porosity samples obtained in the step 4), and dividing the sample database into a training set and a verification set;
step 6), constructing a multilayer perception neural network, and calculating the porosity of the while-drilling pulse neutrons by using the training set to train the multilayer perception neural network;
step 7), verifying the accuracy of the porosity of the pulse-while-drilling neutrons by using the multi-layer perceptive neural network after training by using a verification set;
and 8) based on the scale relation between the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument established in the step 3), performing scale processing on a borehole medium macroscopic capture section, a formation medium macroscopic capture section, a thermal neutron count ratio, a non-ballistic gamma count, a capture gamma count, an iron element characteristic gamma count and a chlorine element characteristic gamma count which are actually measured by the pulse-while-drilling neutron porosity logging instrument, marking the thermal neutron count ratio by using the iron element characteristic gamma count, inputting the borehole medium macroscopic capture section, the formation medium macroscopic capture section, the non-ballistic gamma count, the capture gamma count, the chlorine element characteristic gamma count, the iron-marked thermal neutron count ratio and the drill collar size of the pulse-while-drilling neutron porosity logging instrument after scale processing into a multilayer sensing neural network, and calculating by using the multilayer sensing neural network to obtain the pulse-while-drilling neutron porosity.
Preferably, the pulse-while-drilling neutron porosity logging instrument is arranged in a drill collar, a slurry flow guide channel is arranged in the drill collar, the outer wall of a shell of the pulse-while-drilling neutron porosity logging instrument is attached to the inner wall of the drill collar, and a far neutron detector, a gamma detector, a near neutron detector, a tungsten-nickel-iron shield and a pulse neutron source are arranged in the shell at intervals from top to bottom.
Preferably, the near neutron detector and the far neutron detector are both in a cylinder structure, the diameter of the near neutron detector is set to be 4cm, the length of the near neutron detector is set to be 5cm, the diameter of the far neutron detector is set to be 6cm, the length of the far neutron detector is set to be 10cm, the diameter of the gamma detector is set to be 4cm, the length of the gamma detector is set to be 4cm, the source distance of the near neutron detector is set to be 35cm, the source distance of the far neutron detector is set to be 70cm, and the source distance of the gamma detector is set to be 50cm.
Preferably, the step 3) includes a calibration relation between the borehole medium macroscopic capture section simulation value and the measured value, a calibration relation between the formation medium macroscopic capture section simulation value and the measured value, a calibration relation between the thermal neutron count ratio simulation value and the measured value, a calibration relation between the non-bomb-gamma-count simulation value and the measured value, a calibration relation between the capture gamma-count simulation value and the measured value, a calibration relation between the iron element characteristic gamma-count simulation value and the measured value, and a calibration relation between the chlorine element characteristic gamma-count simulation value and the measured value.
Preferably, in the step 4), the drill collar sizes of the logging-while-drilling pulsed neutron porosity logging instrument are sequentially set to be 12.065cm, 17.145cm and 20.32cm, the lithology of the formation medium is sequentially set to be sandstone, limestone, dolomite, sand-ash mixed rock and sand-dolomite mixed rock, wherein the sand-ash mixed rock is formed by mixing sandstone and limestone, the sand-dolomite mixed rock is formed by mixing sandstone and dolomite, the mineralization degree of formation water is sequentially set to be 0kppm, 5kppm, 10kppm, 20kppm, 50kppm, 100kppm and 200kppm, the drill collar gaps are sequentially set to be 0cm, 0.5cm, 1cm, 1.5cm and 2cm, the mud types comprise oil-based mud and water-based mud, and the formation porosity is sequentially set to be 0, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% and 50%.
Preferably, in the step 4), the thermal neutron count ratio is marked by using the characteristic gamma count of the iron element, so as to obtain an iron-marked thermal neutron count ratio, as shown in formula (1):
in the formula (I), the compound is shown in the specification,the ratio of the counts of the thermal neutrons to the iron marker,is the counting ratio of the thermal neutrons,and (4) performing gamma counting for the characteristic iron element after normalization treatment.
Preferably, the step 6) specifically comprises the following steps:
step 6.1, constructing a multilayer perception neural network, wherein the multilayer perception neural network is provided with an input layer, a first hidden layer, a second hidden layer and an output layer, the input layer, the first hidden layer, the second hidden layer and the output layer are sequentially connected, 7 nodes are arranged in the input layer and are used for obtaining input parameters of the multilayer perception neural network, the input parameters comprise drill collar size, borehole medium macroscopic capture cross section, formation medium macroscopic capture cross section, iron marker thermal neutron count ratio, non-bomb gamma counting, capture gamma counting and chlorine element characteristic gamma counting in a while-drilling pulse neutron porosity sample, 10 nodes are arranged in the first hidden layer, 15 nodes are arranged in the second hidden layer, the first hidden layer and the second hidden layer are connected through a ReLU activation function, the second hidden layer and the output layer are connected through a tanh activation function, and 1 node is arranged in the output layer and is used for outputting calculated while-drilling pulse neutron porosity;
step 6.2, setting a porosity precision value of the pulse neutrons while drilling;
step 6.3, randomly selecting a pulse neutron porosity while drilling sample in the training set, inputting the pulse neutron porosity while drilling sample into the multilayer perception neural network, enabling nodes in the input layer of the multilayer perception neural network to obtain input parameters in the pulse neutron porosity while drilling sample, activating the input parameters through a ReLU function, and transmitting the activated input parameters to the first hidden layer, wherein the output result of the first hidden layer is shown as a formula (2):
in the formula (I), the compound is shown in the specification,is the output result of the first hidden layer,is an input parameter of the multi-layer perceptive neural network,is the size of a drill collar,the ratio of the counts of the thermal neutrons to the iron marker,the counting is a non-bomb gamma counting method,to capture gammaThe number of the counting is counted,is the gamma counting of the characteristic of the chlorine element,for the macroscopic capture cross-section of the wellbore medium,is a macroscopic trapping cross section of the formation medium,is the weight of the first hidden layer and,a bias for the first hidden layer;
the output result of the first hidden layer is activated by the ReLU function and then transmitted to the second hidden layer, and the output result of the second hidden layer is shown in formula (3):
in the formula (I), the compound is shown in the specification,as a result of the output of the second hidden layer,is the weight of the second hidden layer and,a bias for the second hidden layer;
and the output result of the second hidden layer is activated by the tanh function and then transmitted to the output layer, and the porosity of the pulse-while-drilling neutrons output by the output layer is as follows:
in the formula (I), the compound is shown in the specification,the porosity of the while-drilling pulse neutrons output by the output layer,as a weight of the output layer,is the bias of the output layer;
step 6.4, calculating the formation porosity of the model by combining the while-drilling neutron porosity value corresponding to the while-drilling pulse neutron porosity sample according to the while-drilling pulse neutron porosity output by the output layer, calculating the while-drilling pulse neutron porosity error value of the multilayer perception neural network, and comparing the while-drilling pulse neutron porosity error value with a preset while-drilling pulse neutron porosity precision value; if the while-drilling pulse neutron porosity error value of the multilayer perception neural network is larger than the set while-drilling pulse neutron porosity precision value, updating the weight and bias of each layer in the multilayer perception neural network by using a logcosh loss function and adopting a back propagation and random gradient descent method, returning to the step 6.3, and continuing to train the multilayer perception neural network; and if the while-drilling pulse neutron porosity error value of the multilayer perception neural network is not less than the set while-drilling pulse neutron porosity precision value, stopping training the multilayer perception neural network, and entering the step 7).
Preferably, in the step 6.4, the update formula of the weights and the biases of each layer in the multi-layer perceptive neural network is:
in the formula (I), the compound is shown in the specification,to minimize the value of the logcosh loss function,is the logcosh loss function.
Preferably, in the step 7), all the pulse-while-drilling neutron porosities in the verification set are input into the trained multilayer perceptive neural network, a calculation value of the pulse-while-drilling neutron porosity is calculated by using the multilayer perceptive neural network for each pulse-while-drilling neutron porosity sample, the formation porosity of the model is calculated by combining the corresponding pulse-while-drilling neutron porosity value of the pulse-while-drilling neutron porosity sample, an absolute error value of the pulse-while-drilling neutron porosity is calculated by using the multilayer perceptive neural network by using an absolute error function, and if the absolute error value is less than 1%, the calculation result of the pulse-while-drilling neutron porosity of the multilayer perceptive neural network is proved to be accurate;
the calculation formula of the absolute error value of the porosity of the while-drilling pulse neutrons is as follows:
in the formula (I), the compound is shown in the specification,is the absolute error value of the porosity of the while-drilling pulse neutrons,to verify the number of concentrated while-drilling pulsed neutron porosity samples,in order to verify the serial number of the porosity sample of the concentrated while-drilling pulse neutron,for multi-layer perceptive neural networks according toA pulse-while-drilling neutron porosity value calculated for each pulse-while-drilling neutron porosity sample,is as followsAnd calculating the formation porosity of the model according to the while-drilling neutron porosity value corresponding to the pulse-while-drilling neutron porosity sample.
The invention has the beneficial effects that:
according to the intelligent processing method for the porosity of the pulse neutrons while drilling based on the iron neutron marker, provided by the invention, the iron element characteristic gamma counting is used as a marking factor to process the thermal neutron counting ratio, so that the influence of the energy of a pulse neutron source on the distribution of a neutron field is eliminated, and the measurement sensitivity of a pulse neutron porosity logging-while-drilling instrument is improved.
Meanwhile, the invention trains a multilayer perception neural network by using a pulse neutron porosity while drilling sample obtained by Monte Carlo simulation, inputs the size of a drill collar of a pulse neutron porosity while drilling instrument, a measured macroscopic capture cross section of a borehole medium, a measured macroscopic capture cross section of a stratum medium, a counting ratio of iron-marked thermal neutrons, a non-elastic gamma count, a captured gamma count and a gamma count of chlorine element characteristics into the multilayer perception neural network as input parameters, obtains the true pulse neutron porosity while drilling by using the multilayer perception neural network calculation, automatically eliminates errors caused by measurement while drilling environments such as a borehole environment and the drill collar, improves the calculation precision of the pulse neutron porosity while drilling, ensures the accuracy of the calculation result of the pulse neutron porosity while drilling, and lays a foundation for high-precision stratum evaluation.
Drawings
FIG. 1 is a schematic structural diagram of a pulse-while-drilling neutron porosity logging instrument according to the present invention. In the figure, 1 is a stratum, 2 is a drill collar, 3 is a slurry diversion channel, 4 is a far neutron detector, 5 is a gamma detector, 6 is a near neutron detector, 7 is a tungsten-nickel-iron shield, and 8 is a pulse neutron source.
FIG. 2 is a graph of capture gamma energy spectra of limestone formations of different formation porosities.
FIG. 3 is a graph of characteristic gamma counts of iron as a function of formation porosity.
FIG. 4 is a graph of the response of thermal neutron count ratio and iron-labeled thermal neutron count ratio with formation porosity.
FIG. 5 is a plot of formation porosity sensitivity versus thermal neutron count ratio and iron-labeled thermal neutron count ratio.
Fig. 6 is a curve of logcosh loss function value with training times in the process of multi-layer perceptive neural network training.
FIG. 7 is a distribution frequency diagram for verifying absolute error values of concentrated while-drilling pulsed neutron porosity samples.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in 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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
After entering the stratum, fast neutrons emitted by the pulse neutron source react with the nuclei of stratum substances through inelastic scattering, elastic scattering, radiation capture and the like, and the energy of the neutrons is reduced to become epithermal neutrons and thermal neutrons. The distribution of thermal neutron flux in the formation depends on the distance traveled by the neutronsAnd the deceleration length of the formation mediumThe formation density also has negligible effect on the thermal neutron distribution, so the distribution of thermal neutrons in the formation is affected by both the hydrogen index and the formation density.
According to the theory of double-group diffusion, the ratio of the counts of thermal neutrons recorded by two neutron detectors in an infinite homogeneous mediumThe approximate description is:
in the formula (I), the compound is shown in the specification,is the thermal neutron count rate of the near neutron detector,is the thermal neutron count rate of the far neutron detector,is the source distance of the near-neutron detector,is the source distance of the far neutron detector,is the deceleration length of the formation medium.
Because the energy of the pulse neutron source is higher than that of an americium-beryllium neutron source, high-energy neutrons emitted by the pulse neutron source mainly generate inelastic scattering action with a rock framework after entering a stratum and are greatly influenced by the stratum density, and the porosity sensitivity measured by the pulse neutron porosity logging method in a high-porosity stratum is lower than that of an Am-Be neutron source.
The method comprises the following steps of enabling thermal neutrons to further react with atomic nuclei in a stratum to generate characteristic gamma rays, recording the characteristic gamma rays by a detector, and regarding a specific while-drilling pulse neutron porosity instrument, enabling the characteristic gamma counting of iron elements to be only related to the thermal neutron flux distribution, enabling the characteristic gamma counting of silicon elements and the characteristic gamma counting of calcium elements to be mainly related to the lithology of the stratum, and enabling the characteristic gamma counting of chlorine elements and a macroscopic capture cross section to be mainly related to the water mineralization of the stratum.
The invention provides an intelligent processing method of pulse-while-drilling neutron porosity based on iron neutron marking, which is characterized in that a pulse-while-drilling neutron porosity logging instrument is adopted to measure in a stratum, as shown in figure 1, the pulse-while-drilling neutron porosity logging instrument is arranged in a drill collar 2, a mud diversion channel 3 is arranged in the drill collar 2, the outer wall of a shell of the pulse-while-drilling neutron porosity logging instrument is attached to the inner wall of the drill collar 2, and a far neutron detector 4, a gamma detector 5, a near neutron detector 6, a tungsten-nickel-iron shield 7 and a pulse neutron source 8 are arranged in the shell at intervals from top to bottom.
In this embodiment, the pulse neutron source 8 is a D-T pulse neutron generator, the D-T pulse neutron generator emits fast neutrons into the formation 1 in a pulse form, the near neutron detector and the far neutron detector are both in a cylinder structure, the diameter of the near neutron detector is set to 4cm, the length of the near neutron detector is set to 5cm, the diameter of the far neutron detector is set to 6cm, the length of the far neutron detector is set to 10cm, the diameter of the gamma detector is set to 4cm, the length of the gamma detector is set to 4cm, the source distance of the near neutron detector is set to 35cm, the source distance of the far neutron detector is set to 70cm, and the source distance of the gamma detector is set to 50cm.
The invention provides an intelligent processing method for porosity of pulse neutrons while drilling based on iron neutron marking, which specifically comprises the following steps:
the method comprises the following steps of 1, placing a drilling pulse neutron porosity logging instrument in a standard graduated well for measurement, setting the iron element energy window range of a gamma detector to be 6.4-7.8MeV, setting the chlorine element energy window range to be 1.02-1.23MeV, setting the non-ballistic gamma recording time period of the gamma detector to be 0-20 mus and setting the capture gamma recording time period to be 60-160 mus, obtaining the thermal neutron count and the neutron time spectrum of a near neutron detector, the thermal neutron count and the neutron time spectrum of a far neutron detector and the gamma energy spectrum of the gamma detector, and calculating measured values of the drilling pulse neutron porosity logging instrument, wherein the measured values comprise measured values of a macroscopic capture section of a borehole medium, measured values of a capture section of a formation medium, measured values of a thermal neutron count ratio, measured values of a non-ballistic gamma count, measured values of a capture gamma count, measured values of an iron element characteristic gamma count and measured values of a chlorine element characteristic gamma count, wherein the macroscopic capture section of the formation medium, the formation neutron count, the thermal neutron count and the non-ballistic gamma energy count are determined by using the time spectrum of the near neutron detector and the far neutron detector, and the thermal neutron count, the thermal neutron capture macroscopic capture section of the formation medium, the chlorine element, and the non-gamma energy count, and the chlorine element count, and the non-gamma energy count of the thermal neutron detector are determined by using the near neutron detector.
And 2, establishing a neutron porosity while drilling numerical calculation model by using a Monte Carlo numerical simulation method according to the instrument structure of the pulse neutron porosity while drilling logging instrument and combining the structural parameters of a standard graduated well, and simulating to obtain the simulated value of the pulse neutron porosity while drilling logging instrument, wherein the simulated value comprises the simulated value of a macroscopic capture section of a borehole medium, the simulated value of a macroscopic capture section of a formation medium, the simulated value of a thermal neutron count ratio, the simulated value of a non-elastic gamma count, the simulated value of a capture gamma count, the simulated value of an iron element characteristic gamma count and the simulated value of a chlorine element characteristic gamma count. FIG. 2 is a capture gamma energy spectrum of a limestone formation under different formation porosity conditions, wherein the capture gamma energy spectrum of the limestone formation with the formation porosity of 5%, the capture gamma energy spectrum of the limestone formation with the formation porosity of 20% and the capture gamma energy spectrum of the limestone formation with the porosity of 40% are shown, and the capture gamma energy spectra under different formation porosity conditions are compared, so that the capture gamma counts of the capture gamma energy spectra under the energy range of 6.5 to 7.8MeV are reduced along with the increase of the formation porosity, and the difference between the capture gamma counts in other energy sections is smaller. Fig. 3 shows the change rule of the characteristic gamma count of the iron element with the porosity of the formation, which can be obtained from fig. 3, and even in a high-porosity formation, the characteristic gamma count of the iron element still changes with the porosity of the formation, and by analyzing fig. 2 and fig. 3, the characteristic gamma count of the iron element can reflect the distribution of the thermal neutron field in the formation. Therefore, the influence of the neutron source type on the neutron field distribution can be eliminated by marking the thermal neutron counting ratio by utilizing the characteristic gamma counting of the iron element, and the sensitivity of the instrument on the porosity measurement of the pulse neutrons is improved. Marking the thermal neutron count ratio by utilizing iron element characteristic gamma counting to obtain the iron-marked thermal neutron count ratio as shown in a formula (1):
in the formula (I), the compound is shown in the specification,the ratio of the counts of the thermal neutrons to the iron marker,is the counting ratio of the thermal neutrons,and (4) performing gamma counting for the characteristic iron element after normalization treatment.
FIG. 4 is a graph showing the response law of the thermal neutron count ratio and the iron-labeled thermal neutron count ratio with the formation porosity, which can be obtained from FIG. 4, wherein the thermal neutron count ratios all increase with the increase of the formation porosity, the thermal neutron count ratio remains unchanged after the formation porosity exceeds 20%, and the iron-labeled thermal neutron count ratio has an obvious change law in the whole formation porosity distribution range, a change curve of the formation porosity relative sensitivity of the thermal neutron count ratio with the formation porosity and a change curve of the formation porosity relative sensitivity of the iron-labeled thermal neutron count ratio with the formation porosity are obtained through calculation, as shown in FIG. 5, the formation porosity relative sensitivity of the iron-labeled thermal neutron count ratio is higher than that of the thermal neutron count ratio, and when the formation porosity is 30%, the formation porosity relative sensitivity of the iron-labeled thermal neutron count ratio is about 2.7 times that of the thermal neutron count ratio.
And 3, establishing a scale relation between the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument based on the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument, wherein the scale relation between the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument comprises a scale relation between a macroscopic capturing cross section analog value and the measured value of the borehole medium, a scale relation between a macroscopic capturing cross section analog value and the measured value of the formation medium, a scale relation between a thermal neutron count ratio analog value and the measured value, a scale relation between a non-bomb gamma counting analog value and the measured value, a scale relation between a capturing gamma counting analog value and the measured value, a scale relation between an iron element characteristic gamma counting analog value and the measured value and a scale relation between a chlorine element characteristic gamma counting analog value and the measured value.
And 4, in addition to the source energy difference, the distribution difference of the neutron field in the stratum can be caused by environmental factors such as lithology, stratum water mineralization and the like, and the problem that iron marking factors are difficult to obtain exists in the actual measurement process, so that the measurement sensitivity and the measurement accuracy of the pulse neutron porosity logging-while-drilling instrument under different measurement-while-drilling environments are improved.
According to the instrument structure of the pulse-while-drilling neutron porosity logging instrument, the size of a drill collar of the pulse-while-drilling neutron porosity logging instrument is changed, the lithology, the formation water mineralization, the drill collar gap, the mud type and the formation porosity of a formation medium are sequentially changed under the condition of different drill collar sizes, a neutron porosity numerical calculation model while drilling is reestablished, the thermal neutron count and the neutron time spectrum of a near neutron detector, the thermal neutron count and the neutron time spectrum of a far neutron detector and the gamma energy spectrum of a gamma detector under different logging-while-drilling environments are obtained in a simulated mode, and a plurality of groups of pulse-while-drilling neutron porosity samples are formed according to the drill collar sizes, the macro capture cross section of a borehole medium, the macro capture cross section of the formation medium, the thermal neutron count ratio of iron markers, the non-elastic gamma count, the capture gamma count and the gamma characteristic count gamma of chlorine elements under different logging-while-drilling environments.
In the embodiment, the sizes of drill collars of a pulse neutron porosity logging-while-drilling instrument are sequentially set to be 12.065cm, 17.145cm and 20.32cm, after the sizes of the drill collars are fixed, the lithology of a formation medium, the formation water mineralization, the drill collar gap, the mud type and the formation porosity in a neutron porosity numerical calculation model while drilling are sequentially modified, wherein the lithology of the formation medium is sequentially set to be sandstone, limestone, dolomite and sand-lime mixed rock (formed by mixing sandstone and limestone, the proportion of the limestone is increased from 0 to 100%, 10% is increased each time) and sand-dolomite mixed rock (formed by mixing sandstone and dolomite, the proportion of the dolomite is increased from 0 to 100%, 10% is increased each time), the formation water mineralization is sequentially set to be 0kppm, 5kppm, 10kppm, 20kppm, 50kppm, 100kppm and 200kppm, the drill collar gap is sequentially set to be 0cm, 0.5cm, 1cm, 1.5cm and 2cm, the types of oil-based mud, water-based mud, 25% slurry, 25% and 35% mud. Different measurement-while-drilling environments are simulated by modifying the neutron porosity-while-drilling numerical calculation model, so that the sizes of drill collars, the macroscopic capture cross section of a borehole medium, the macroscopic capture cross section of a formation medium, the iron-labeled thermal neutron count ratio, the non-elastic gamma count, the capture gamma count and the chlorine element characteristic gamma count in the different measurement-while-drilling environments are obtained, and 10000 groups of pulse-while-drilling neutron porosity samples are obtained in the embodiment.
And 5, constructing a sample database based on the multiple groups of while-drilling pulse neutron porosity samples obtained in the step 4, and dividing the sample database into a training set and a verification set, wherein the number of the while-drilling neutron porosity samples in the training set accounts for 85% of the total number of the while-drilling pulse neutron porosity samples in the sample database, and the number of the while-drilling neutron porosity samples in the verification set accounts for 15% of the total number of the while-drilling pulse neutron porosity samples in the sample database.
6.1, constructing a multilayer perception neural network, wherein the multilayer perception neural network is provided with an input layer, a first hidden layer, a second hidden layer and an output layer, the input layer, the first hidden layer, the second hidden layer and the output layer are sequentially connected, 7 nodes are arranged in the input layer and used for obtaining input parameters of the multilayer perception neural network, the input parameters comprise drill collar size, borehole medium macroscopic capture cross section, formation medium macroscopic capture cross section, iron marker thermal neutron count ratio, non-bomb gamma counting, capture gamma counting and chlorine element characteristic gamma counting in a while-drilling pulse neutron porosity sample, 10 nodes are arranged in the first hidden layer, 15 nodes are arranged in the second hidden layer, the first hidden layer and the second hidden layer are connected through a ReLU activation function, the second hidden layer and the output layer are connected through a tanh activation function, and 1 node is arranged in the output layer and used for outputting calculated while-drilling pulse neutron porosity.
And 6.2, setting a porosity precision value of the pulse neutrons while drilling.
Step 6.3, randomly selecting a pulse neutron porosity while drilling sample in the training set, inputting the pulse neutron porosity while drilling sample into the multilayer perception neural network, enabling nodes in the input layer of the multilayer perception neural network to obtain input parameters in the pulse neutron porosity while drilling sample, activating the input parameters through a ReLU function, and transmitting the activated input parameters to the first hidden layer, wherein the output result of the first hidden layer is shown as a formula (2):
in the formula (I), the compound is shown in the specification,as a result of the output of the first hidden layer,is an input parameter of the multi-layer perceptive neural network,is the size of a drill collar,the counting ratio of the thermal neutrons is marked for the iron,the counting is a non-bomb gamma counting method,in order to capture the gamma counts,is the gamma counting of the characteristic of the chlorine element,for the macroscopic capture cross-section of the wellbore medium,is a macroscopic trapping cross section of the formation medium,is the weight of the first hidden layer and,is the bias of the first hidden layer.
The output result of the first hidden layer is activated by the ReLU function and then transmitted to the second hidden layer, and the output result of the second hidden layer is shown in formula (3):
in the formula (I), the compound is shown in the specification,as a result of the output of the second hidden layer,is the weight of the second hidden layer,is a bias of the second hidden layer.
The output result of the second hidden layer is activated by the tanh function and then transmitted to the output layer, and the porosity of the while-drilling pulse neutrons output by the output layer is as follows:
in the formula (I), the compound is shown in the specification,the porosity of the while-drilling pulse neutrons output by the output layer,as a weight of the output layer,is the bias of the output layer.
Step 6.4, calculating the stratum porosity of the model by combining the while-drilling neutron porosity value corresponding to the while-drilling pulse neutron porosity sample according to the while-drilling pulse neutron porosity output by the output layer, and calculating the while-drilling pulse neutron porosity error value of the multilayer perception neural network and comparing the error value with the preset while-drilling pulse neutron porosity precision value; if the while-drilling pulse neutron porosity error value of the multilayer perception neural network is larger than the set while-drilling pulse neutron porosity precision value, updating the weight and bias of each layer in the multilayer perception neural network by using a formula (5) based on a logcosh loss function through a back propagation and random gradient descent method, returning to the step 6.3, and continuing to train the multilayer perception neural network; and if the porosity error value of the while-drilling pulse neutrons of the multilayer perception neural network is not less than the set porosity precision value of the while-drilling pulse neutrons, stopping training the multilayer perception neural network, and entering the step 7.
In this embodiment, as shown in fig. 6, when the training frequency of the multilayer perceptual neural network reaches 400 times, the loss function value is stable, the error value of the porosity of the pulse-while-drilling neutron of the multilayer perceptual neural network is not less than the set porosity precision value of the pulse-while-drilling neutron, and the calculation accuracy of the porosity of the pulse-while-drilling neutron of the multilayer perceptual neural network is stable.
And 7, verifying the accuracy of the porosity of the pulse-while-drilling neutrons by using the multi-layer perceptive neural network after training by using a verification set, inputting all the porosity of the pulse-while-drilling neutrons in the verification set into the multi-layer perceptive neural network after training, respectively calculating the porosity of the pulse-while-drilling neutrons by using the multi-layer perceptive neural network for each sample of the porosity of the pulse-while-drilling neutrons to obtain a calculated value of the porosity of the pulse-while-drilling neutrons, calculating an absolute error value of the porosity of the pulse-while-drilling neutrons by using a formula (6) in combination with the porosity of the pulse-while-drilling neutrons corresponding to the sample of the porosity of the pulse-while-drilling neutrons, and proving that the calculation result of the porosity of the pulse-while-drilling neutrons of the multi-layer perceptive neural network is accurate if the absolute error value is less than 1%.
In the embodiment, the distribution frequency diagram of the absolute error value of the concentrated while-drilling pulse neutron porosity sample is verified, as shown in fig. 7, and can be obtained from fig. 7, the actual value of the while-drilling pulse neutron porosity close to the while-drilling pulse neutron porosity is obtained through calculation of the multilayer perceptive neural network trained by the method, the absolute error between the while-drilling pulse neutron porosity calculated by the multilayer perceptive neural network after training and the actual value of the formation porosity is basically less than 1%, and the accuracy of the multilayer perceptive neural network used for calculating the while-drilling pulse neutron porosity after training by the method is verified.
And 8, based on the scale relation between the simulated value and the measured value of the pulse-while-drilling neutron porosity logging instrument established in the step 3, performing scale processing on a borehole medium macroscopic capture section, a formation medium macroscopic capture section, a thermal neutron count ratio, a non-elastic gamma count, a capture gamma count, an iron element characteristic gamma count and a chlorine element characteristic gamma count which are actually measured by the pulse-while-drilling neutron porosity logging instrument, marking the thermal neutron count ratio by using the iron element characteristic gamma count, inputting the borehole medium macroscopic capture section, the formation medium macroscopic capture section, the non-elastic gamma count, the capture gamma count, the chlorine element characteristic gamma count, the iron mark thermal neutron count ratio and the drill collar size of the pulse-while-drilling neutron porosity logging instrument after scale processing into a multilayer perception neural network, and calculating by using the multilayer perception neural network to obtain the pulse-while-drilling neutron porosity.
It is to be understood that the above description is not intended to limit the present invention, and the present invention is not limited to the above examples, and those skilled in the art may make various changes, modifications, additions and substitutions within the spirit and scope of the present invention.
Claims (9)
1. An intelligent processing method for porosity of pulse neutrons while drilling based on iron neutron marking is characterized in that a pulse neutron porosity logging while drilling instrument is adopted for measurement, and comprises the following steps:
1) Placing a pulse neutron porosity logging-while-drilling instrument in a standard calibration well for measurement to obtain an actual measurement value of the pulse neutron porosity logging-while-drilling instrument;
2) Establishing a neutron porosity value calculation model while drilling by using a Monte Carlo numerical simulation method to obtain a simulation value of a pulse neutron porosity logging instrument while drilling;
3) Establishing a scale relation between the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument based on the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument;
4) Changing the size of a drill collar of a pulse-while-drilling neutron porosity logging instrument, sequentially changing the lithology, the water mineralization of the stratum, the gap of the drill collar, the mud type and the porosity of the stratum under different drill collar sizes, reestablishing a neutron porosity while-drilling numerical calculation model, and simulating to obtain the size of the drill collar, the macroscopic capture section of the borehole medium, the macroscopic capture section of the stratum medium, the counting number of iron-marked thermal neutrons, the counting number of non-elastic gammas, the counting number of capture gammas and the counting number of chlorine element characteristics gammas under different logging-while-drilling environments to form a plurality of groups of pulse-while-drilling neutron porosity samples;
5) Constructing a sample database based on a plurality of groups of while-drilling pulse neutron porosity samples, and dividing the sample database into a training set and a verification set;
6) Constructing a multilayer perception neural network, and calculating the porosity of the pulse neutrons while drilling by using a training set to train the multilayer perception neural network;
7) Verifying the accuracy of the porosity of the neutrons in the while-drilling pulse by using the multi-layer perception neural network after training by using a verification set;
8) Based on the scale relation between the analog value and the measured value of the pulse-while-drilling neutron porosity logging instrument, after the macroscopic capture cross section of a borehole medium, the macroscopic capture cross section of a formation medium, the nonelastic gamma count, the capture gamma count, the chlorine element characteristic gamma count and the iron-marked thermal neutron count ratio which are actually measured by the pulse-while-drilling neutron porosity logging instrument are obtained through scale processing, the obtained values and the size of the drill collar are input into the multilayer perception neural network together, and the pulse-while-drilling neutron porosity is obtained through calculation.
2. The intelligent processing method for porosity of pulse-while-drilling neutrons based on iron neutron marking as claimed in claim 1, wherein the pulse-while-drilling neutron porosity logging instrument is arranged in a drill collar, a mud diversion channel is arranged in the drill collar, the outer wall of a shell of the pulse-while-drilling neutron porosity logging instrument is closely attached to the inner wall of the drill collar, and a far neutron detector, a gamma detector, a near neutron detector, a wolfram-nickel-iron shield and a pulse neutron source are arranged in the shell at intervals from top to bottom.
3. The intelligent processing method for porosity of pulse-while-drilling neutrons based on iron neutron marking as claimed in claim 2, wherein the near neutron detector and the far neutron detector are both in a cylinder structure, the diameter of the near neutron detector is set to be 4cm, the length of the near neutron detector is set to be 5cm, the diameter of the far neutron detector is set to be 6cm, the length of the far neutron detector is set to be 10cm, the diameter of the gamma detector is set to be 4cm, the length of the gamma detector is set to be 4cm, the source distance of the near neutron detector is set to be 35cm, the source distance of the far neutron detector is set to be 70cm, and the source distance of the gamma detector is set to be 50cm.
4. The method as claimed in claim 1, wherein in the step 3), the calibration relationship between the simulated value and the measured value of the LWD tool includes a calibration relationship between the simulated value and the measured value of the macroscopic capture cross-section of the borehole medium, a calibration relationship between the simulated value and the measured value of the macroscopic capture cross-section of the formation medium, a calibration relationship between the simulated value and the measured value of the thermal neutron count ratio, a calibration relationship between the simulated value and the measured value of the non-bomb gamma count, a calibration relationship between the simulated value and the measured value of the capture gamma count, a calibration relationship between the simulated value and the measured value of the iron element gamma count, and a calibration relationship between the simulated value and the measured value of the chlorine element gamma count.
5. The intelligent iron neutron marker-based pulsed neutron porosity while drilling processing method as claimed in claim 1, wherein in step 4), the drill collar sizes of the pulsed neutron porosity while drilling logging tool are set to be 12.065cm, 17.145cm and 20.32cm in sequence, and the lithology of the formation medium is set to be sandstone, limestone, dolomite, sand-lime mixed rock and sand-dolomite mixed rock in sequence, wherein the sand-lime mixed rock is formed by mixing sandstone and limestone, the sand-lime mixed rock is formed by mixing sandstone and dolomite, the water mineralization of the formation is set to be 0kppm, 5kppm, 10kppm, 20kppm, 50kppm, 100kppm and 200kppm in sequence, the drill collar gaps are set to be 0cm, 0.5cm, 1cm, 1.5cm and 2cm in sequence, the mud types comprise oil-based mud and water-based mud, and the formation porosity is set to be 0, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% and 50% in sequence.
6. The intelligent processing method for porosity of pulse-while-drilling neutrons based on iron neutron marking as described in claim 1, wherein in the step 4), an iron-marked thermal neutron count ratio is obtained by marking a thermal neutron count ratio by using an iron element characteristic gamma count, as shown in formula (1):
7. The intelligent processing method for porosity of pulsed neutrons while drilling based on iron neutron marking according to claim 1, wherein the step 6) comprises:
6.1, constructing a multilayer perception neural network, wherein the multilayer perception neural network is provided with an input layer, a first hidden layer, a second hidden layer and an output layer, the input layer, the first hidden layer, the second hidden layer and the output layer are sequentially connected, 7 nodes are arranged in the input layer and are used for obtaining input parameters of the multilayer perception neural network, the input parameters comprise the size of a drill collar in a pulse neutron porosity while drilling sample, a macroscopic capture section of a borehole medium, a macroscopic capture section of a formation medium, a counting ratio of iron-marked thermal neutrons, a non-elastic gamma count, a capture gamma count and a chlorine element characteristic gamma count, 10 nodes are arranged in the first hidden layer, 15 nodes are arranged in the second hidden layer, the first hidden layer and the second hidden layer are connected through a ReLU activation function, the second hidden layer and the output layer are connected through a tanh activation function, and 1 node is arranged in the output layer and is used for outputting the calculated pulse neutron porosity while drilling;
step 6.2, setting a porosity precision value of the pulse neutrons while drilling;
step 6.3, randomly selecting a while-drilling pulse neutron porosity sample in the training set, inputting the sample into the multilayer perceptive neural network, enabling nodes in the input layer of the multilayer perceptive neural network to obtain input parameters in the while-drilling pulse neutron porosity sample, activating the input parameters by a ReLU function, and transmitting the activated input parameters to the first hiding layer, wherein the output result of the first hiding layer is shown as a formula (2):
in the formula (I), the compound is shown in the specification,as a result of the output of the first hidden layer,is an input parameter of the multi-layer perceptive neural network,is the size of a drill collar,the ratio of the counts of the thermal neutrons to the iron marker,the counting is carried out for the non-bullet gamma counting,in order to capture the gamma counts,is the gamma counting of the characteristic of the chlorine element,is a macroscopic capture cross-section for the wellbore medium,is a macroscopic trapping cross section of the formation medium,is the weight of the first hidden layer,is a firstA hidden layer bias;
the output result of the first hidden layer is activated by the ReLU function and then transmitted to the second hidden layer, and the output result of the second hidden layer is shown in formula (3):
in the formula (I), the compound is shown in the specification,is the output result of the second hidden layer,is the weight of the second hidden layer and,a bias for the second hidden layer;
and the output result of the second hidden layer is activated by the tanh function and then transmitted to the output layer, and the porosity of the pulse-while-drilling neutrons output by the output layer is as follows:
in the formula (I), the compound is shown in the specification,the porosity of the pulse neutron output by the output layer while drilling,as a weight of the output layer(s),is the bias of the output layer;
step 6.4, calculating the formation porosity of the model by combining the while-drilling neutron porosity value corresponding to the while-drilling pulse neutron porosity sample according to the while-drilling pulse neutron porosity output by the output layer, calculating the while-drilling pulse neutron porosity error value of the multilayer perception neural network, and comparing the while-drilling pulse neutron porosity error value with a preset while-drilling pulse neutron porosity precision value; if the porosity error value of the while-drilling pulse neutrons of the multilayer perception neural network is larger than the set porosity precision value of the while-drilling pulse neutrons, updating the weight and bias of each layer in the multilayer perception neural network by using a logcosh loss function and adopting a back propagation and random gradient descent method, returning to the step 6.3, and continuing to train the multilayer perception neural network; and if the porosity error value of the while-drilling pulse neutrons of the multilayer perception neural network is not less than the set porosity precision value of the while-drilling pulse neutrons, stopping training the multilayer perception neural network, and entering the step 7).
8. The intelligent processing method for porosity of pulsed neutrons while drilling based on iron neutron marking as claimed in claim 7, wherein in step 6.4, the weight and bias updating formula of each layer in the multilayer perceptive neural network is as follows:
9. The method for intelligently processing porosity of pulse-while-drilling neutrons based on iron neutron labeling according to claim 1, wherein in step 7), all porosity of pulse-while-drilling neutrons in the validation set are input into the trained multilayer perceptive neural network, a calculation value of the porosity of the pulse-while-drilling neutrons is obtained by using the multilayer perceptive neural network for each sample of the porosity of the pulse-while-drilling neutrons, and an absolute error value of the porosity of the pulse-while-drilling neutrons is calculated by using the multilayer perceptive neural network by combining the formation porosity of the calculation model of the porosity value of the neutron-while-drilling neutrons corresponding to the sample of the porosity of the pulse-while-drilling neutrons, and if the absolute error value is less than 1%, the calculation result of the porosity of the pulse-while-drilling neutrons in the multilayer perceptive neural network is proved to be accurate;
the calculation formula of the absolute error value of the porosity of the while-drilling pulse neutrons is as follows:
in the formula (I), the compound is shown in the specification,is the absolute error value of the porosity of the while-drilling pulse neutrons,to verify the number of concentrated while-drilling pulsed neutron porosity samples,to verify the serial number of the concentrated while-drilling pulsed neutron porosity samples,for multi-layer perceptive neural networks according toA while-drilling pulsed neutron porosity value calculated for each while-drilling pulsed neutron porosity sample,is as followsAnd calculating the formation porosity of the model according to the while-drilling neutron porosity value corresponding to the pulse-while-drilling neutron porosity sample.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211194628.0A CN115291288B (en) | 2022-09-29 | 2022-09-29 | While-drilling pulse neutron porosity intelligent processing method based on iron neutron marking |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211194628.0A CN115291288B (en) | 2022-09-29 | 2022-09-29 | While-drilling pulse neutron porosity intelligent processing method based on iron neutron marking |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115291288A true CN115291288A (en) | 2022-11-04 |
CN115291288B CN115291288B (en) | 2022-12-30 |
Family
ID=83834251
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211194628.0A Active CN115291288B (en) | 2022-09-29 | 2022-09-29 | While-drilling pulse neutron porosity intelligent processing method based on iron neutron marking |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115291288B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116500694A (en) * | 2023-06-28 | 2023-07-28 | 中海油田服务股份有限公司 | Post-sleeve physical quantity inversion method, post-sleeve physical quantity inversion device, computing equipment and storage medium |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4625111A (en) * | 1985-02-04 | 1986-11-25 | Halliburton Company | Method and apparatus for pulsed neutron capture measurement of borehole conditions and formation hydrogen index |
US4737636A (en) * | 1986-11-17 | 1988-04-12 | Halliburton Company | Simultaneous neutron lifetime and oxygen activation and fluid sampling methods and apparatus to locate fluid flow in a cased well borehole |
US5345077A (en) * | 1991-07-24 | 1994-09-06 | Mobil Oil Corporation | Method and apparatus for producing a porosity log of a subsurface formation corrected for detector standoff |
WO2012012120A2 (en) * | 2010-06-30 | 2012-01-26 | Schlumberger Canada Limited | Identification of neutron capture from a pulsed neutron logging tool |
CN103696765A (en) * | 2013-11-06 | 2014-04-02 | 中国石油大学(华东) | Double-LaBr3 detector element energy spectrum logger based on controllable neutron source and logging method |
US20140339410A1 (en) * | 2013-05-15 | 2014-11-20 | Schlumberger Technology Corporation | Borehole Fluid Effect Correction For Pulsed Neutron Porosity Measurements |
CN104747179A (en) * | 2013-12-31 | 2015-07-01 | 中国石油化工集团公司 | Stratum density measuring while drilling instrument based on deuterium-tritium accelerator neutron source |
CN107462929A (en) * | 2017-07-25 | 2017-12-12 | 中国石油大学(华东) | Cupro-nickel mineral products level measuring arrangement and method in a kind of well |
CN110439545A (en) * | 2019-08-02 | 2019-11-12 | 中国石油天然气集团有限公司 | One kind is with brill controllable source neutron porosity log instrument environmental correction method |
CN112904436A (en) * | 2021-04-14 | 2021-06-04 | 中国石油大学(华东) | Porosity measurement method combining element yield and thermal neutron count ratio |
CN115012920A (en) * | 2022-06-08 | 2022-09-06 | 中国石油大学(华东) | Controllable neutron source multi-spectrum logging instrument and method based on double CLYC double-particle detectors |
CN115061219A (en) * | 2022-08-17 | 2022-09-16 | 北京派特杰奥科技有限公司 | Fracture type reservoir prediction and identification method and system based on petroleum and natural gas detection |
-
2022
- 2022-09-29 CN CN202211194628.0A patent/CN115291288B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4625111A (en) * | 1985-02-04 | 1986-11-25 | Halliburton Company | Method and apparatus for pulsed neutron capture measurement of borehole conditions and formation hydrogen index |
US4737636A (en) * | 1986-11-17 | 1988-04-12 | Halliburton Company | Simultaneous neutron lifetime and oxygen activation and fluid sampling methods and apparatus to locate fluid flow in a cased well borehole |
US5345077A (en) * | 1991-07-24 | 1994-09-06 | Mobil Oil Corporation | Method and apparatus for producing a porosity log of a subsurface formation corrected for detector standoff |
WO2012012120A2 (en) * | 2010-06-30 | 2012-01-26 | Schlumberger Canada Limited | Identification of neutron capture from a pulsed neutron logging tool |
US20140339410A1 (en) * | 2013-05-15 | 2014-11-20 | Schlumberger Technology Corporation | Borehole Fluid Effect Correction For Pulsed Neutron Porosity Measurements |
CN103696765A (en) * | 2013-11-06 | 2014-04-02 | 中国石油大学(华东) | Double-LaBr3 detector element energy spectrum logger based on controllable neutron source and logging method |
CN104747179A (en) * | 2013-12-31 | 2015-07-01 | 中国石油化工集团公司 | Stratum density measuring while drilling instrument based on deuterium-tritium accelerator neutron source |
CN107462929A (en) * | 2017-07-25 | 2017-12-12 | 中国石油大学(华东) | Cupro-nickel mineral products level measuring arrangement and method in a kind of well |
CN110439545A (en) * | 2019-08-02 | 2019-11-12 | 中国石油天然气集团有限公司 | One kind is with brill controllable source neutron porosity log instrument environmental correction method |
CN112904436A (en) * | 2021-04-14 | 2021-06-04 | 中国石油大学(华东) | Porosity measurement method combining element yield and thermal neutron count ratio |
CN115012920A (en) * | 2022-06-08 | 2022-09-06 | 中国石油大学(华东) | Controllable neutron source multi-spectrum logging instrument and method based on double CLYC double-particle detectors |
CN115061219A (en) * | 2022-08-17 | 2022-09-16 | 北京派特杰奥科技有限公司 | Fracture type reservoir prediction and identification method and system based on petroleum and natural gas detection |
Non-Patent Citations (5)
Title |
---|
LILI TIANA ET AL.: "A new diffusion effect correction method in pulsed neutron capture logging", 《JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING》 * |
RICHARD C. ODOM ET AL.: "Improvements in a Through-Casing Pulsed-Neutron Density Log", 《2001 SPE ANNUAL TECHNICAL CONFERENCE AND EXHIBITION》 * |
王虎: "脉冲中子孔隙度测井密度校正研究", 《测井技术》 * |
袁超等: "随钻中子孔隙度测井响应特性数值模拟", 《地球科学-中国地质大学学报》 * |
赵靓: "随钻D-T源中子孔隙度校正方法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116500694A (en) * | 2023-06-28 | 2023-07-28 | 中海油田服务股份有限公司 | Post-sleeve physical quantity inversion method, post-sleeve physical quantity inversion device, computing equipment and storage medium |
CN116500694B (en) * | 2023-06-28 | 2023-09-01 | 中海油田服务股份有限公司 | Post-sleeve physical quantity inversion method, post-sleeve physical quantity inversion device, computing equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN115291288B (en) | 2022-12-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8476584B2 (en) | Methods for sourceless density downhole measurement using pulsed neutron generator | |
US6032102A (en) | Method and apparatus for measuring well characteristics and formation properties | |
US9268056B2 (en) | Neutron porosity based on one or more gamma ray detectors and a pulsed neutron source | |
CN105874354A (en) | System and method for making downhole measurements | |
US10725201B2 (en) | Compensated neutron correction for contributions outside the petrophysical model | |
US6150655A (en) | Inferential measurement of photoelectric absorption cross-section of geologic formations from neutron-induced, gamma-ray spectroscopy | |
Wahl et al. | The thermal neutron decay time log | |
CN109521487B (en) | Method for identifying gas layer by using element gamma energy spectrum logging | |
CN106250619B (en) | Method and device for determining mineral content of stratum | |
CN115291288B (en) | While-drilling pulse neutron porosity intelligent processing method based on iron neutron marking | |
CN102084271A (en) | Absolute elemental concentrations from nuclear spectroscopy | |
Wang et al. | Neutron transport correction and density calculation in the neutron-gamma density logging | |
Liu et al. | A method to improve the sensitivity of neutron porosity measurement based on DT source | |
Eyvazzadeh et al. | Modern carbon/oxygen logging methodologies: comparing hydrocarbon saturation determination techniques | |
Fu et al. | A new method of gas reservoir evaluation based on neutron cross section logging | |
CN110469324A (en) | A kind of calculating density of earth formations method based on pulsed neutron log | |
CN115267930A (en) | High-sensitivity neutron porosity measurement method based on D-T pulse neutron source | |
CN100492055C (en) | A chlorine spectrometry logging method | |
US11143786B2 (en) | Intrinsic geological formation carbon to oxygen ratio measurements | |
US5272629A (en) | Method for determining the slowing down length and the porosity of a formation surrounding a borehole | |
Ge et al. | Numerical study of an alternative to a deuterium-tritium source in gas saturation logging based on the inelastic gamma spectrum | |
CN111638559B (en) | Fast neutron scattering cross section characterization method based on pulse neutron logging | |
Zhang et al. | An integrated density correction method of four-detector density logging in cased holes | |
CN117780332A (en) | Method for evaluating oil saturation after sleeving by comprehensively distributing secondary gamma rays and thermal neutrons | |
RU2727091C2 (en) | Method for simultaneous determination of density and porosity of rock |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |