CN113062731B - Intelligent identification method for complex underground drilling conditions - Google Patents
Intelligent identification method for complex underground drilling conditions Download PDFInfo
- Publication number
- CN113062731B CN113062731B CN201911404045.4A CN201911404045A CN113062731B CN 113062731 B CN113062731 B CN 113062731B CN 201911404045 A CN201911404045 A CN 201911404045A CN 113062731 B CN113062731 B CN 113062731B
- Authority
- CN
- China
- Prior art keywords
- neural network
- characteristic parameters
- drilling
- logging
- underground
- 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.)
- Active
Links
- 238000005553 drilling Methods 0.000 title claims abstract description 83
- 238000000034 method Methods 0.000 title claims abstract description 59
- 239000002245 particle Substances 0.000 claims abstract description 94
- 238000013528 artificial neural network Methods 0.000 claims abstract description 85
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 50
- 238000005457 optimization Methods 0.000 claims abstract description 13
- 230000005540 biological transmission Effects 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 37
- 239000012530 fluid Substances 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 14
- 239000013598 vector Substances 0.000 claims description 14
- 230000035515 penetration Effects 0.000 claims description 11
- 210000004205 output neuron Anatomy 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 210000002364 input neuron Anatomy 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000005086 pumping Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 4
- 238000005520 cutting process Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 4
- 238000011161 development Methods 0.000 abstract description 2
- 210000002569 neuron Anatomy 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 239000011435 rock Substances 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000004215 Carbon black (E152) Substances 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 229930195733 hydrocarbon Natural products 0.000 description 1
- 150000002430 hydrocarbons Chemical class 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- 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
-
- 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
- E21B21/00—Methods or apparatus for flushing boreholes, e.g. by use of exhaust air from motor
- E21B21/08—Controlling or monitoring pressure or flow of drilling fluid, e.g. automatic filling of boreholes, automatic control of bottom pressure
-
- 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
- E21B44/00—Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions
-
- 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
- E21B45/00—Measuring the drilling time or rate of penetration
-
- 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
- E21B47/00—Survey of boreholes or wells
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Mechanical Engineering (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The invention discloses an intelligent identification method for complex underground drilling conditions, which comprises the following steps: step 100, constructing logging characteristic parameters corresponding to different underground complex working conditions; step 200, constructing an intelligent BP neural network algorithm based on particle swarm optimization; and 300, creating an underground complex working condition early warning model based on the BP neural network intelligent algorithm in the step 200. In the invention, aiming at the defects of the BP neural network, an algorithm for optimizing the BP neural network based on particle swarm is provided; the algorithm based on the particle swarm optimization BP neural network is applied to real-time early warning of underground complex working conditions, so that an underground complex working condition early warning model is established, intelligent and real-time quantitative judgment of underground complex conditions and faults is realized, and the method has great significance for preventing and controlling the development of drilling complex accidents, reducing losses to the maximum extent; meanwhile, a data transmission interface of the comprehensive logging data and the underground complex working condition early warning model is compiled, and real-time data transmission between the comprehensive logging data and the underground complex working condition early warning model is realized.
Description
Technical Field
The invention relates to the technical field of deep shale gas complex stratum drilling, in particular to an intelligent identification method for underground complex working conditions of drilling.
Background
The oil and gas drilling engineering is basic work of oil and gas exploration and development, is a system engineering involving multiple departments, multiple links, large scale, complex technology, large investment and large risk, determines that a plurality of uncertain factors are hidden in the oil and gas drilling construction process, and when some decisions have to be made by means of insufficient, inaccurate or incomplete information, various underground complex conditions such as surge, leakage, collapse, clamping and the like can be possibly caused, so that safe and efficient drilling construction is severely restricted.
At present, an expert knowledge system is mainly adopted, or conventional logging data is utilized to analyze engineering anomalies, only a small amount of measurement and operation work is automatically completed by a computer, most of analysis and judgment are still completed manually, and due to individual knowledge, experience, responsibility and heart difference and other reasons, abnormal and complex drilling situations cannot be found and processed in time, so that drilling risks are increased; it is in fact impractical to require the operator to be attentively aware of the changes in the monitored data and to quickly ascertain the accident. Therefore, the problems of the related technology such as poor comprehensive utilization capability of monitoring information, insufficient time of risk early warning, strong subjectivity and the like are more remarkable.
Disclosure of Invention
In view of the above, the invention provides an intelligent identification method for complex working conditions under well drilling, which can establish an early warning model of complex working conditions under well based on an artificial neural network intelligent algorithm, and realize predictability and controllability of complex conditions under well and faults.
In order to achieve the above purpose, the present invention provides the following technical solutions:
an intelligent identification method for complex underground drilling conditions comprises the following steps:
step 100, constructing logging characteristic parameters corresponding to different underground complex working conditions;
step 200, constructing an intelligent BP neural network algorithm based on particle swarm optimization;
step 300, creating an underground complex working condition early warning model based on the BP neural network intelligent algorithm in step 200, comprising:
Step 310, designing a BP neural network, where the designing the BP neural network includes:
Step 311, designing an input layer, where the input layer includes:
selecting logging characteristic parameters with strong correlation with underground complex working conditions in the step 100 to be set as input neurons;
step 312, designing an output layer, the design output layer including:
setting the number of output neurons according to the number of underground complex working condition types, and representing each output neuron by adopting an expected output vector;
step 313, designing an underlying layer, the underlying layer comprising:
Calculating the number of hidden layers by a classical formula;
step 320, designing a particle swarm algorithm; the design particle swarm algorithm comprises the following steps:
step 321, calculating a connection weight and a threshold value according to the BP neural network structure obtained in the step 310, determining the particle length of a particle swarm algorithm according to the connection weight and the threshold value, and carrying out vector coding on each particle;
Step 322, creating an fitness function, where creating the fitness function includes:
taking root mean square error of BP neural network training as fitness function:
Preferably, in the step 311, the logging characteristic parameters include: the number of input neurons is set to 9 according to the hook load, vertical pressure, torque, mechanical drilling speed, rotating speed of a rotary table, drilling pressure, drilling fluid outlet flow, drilling fluid outlet density and total volume of a drilling fluid pool;
in the step 312, the downhole complex operating condition types include: if the kick, the collapse, the lost circulation, the sticking and the normal operation are performed, the output neurons are set to be 5, and the respectively corresponding expected output vectors are ：q1＝(1,0,0,0,0)、q2＝(0,1,0,0,0)、q3＝(0,0,1,0,0)、q4＝(0,0,0,1,0)、q5＝(0,0,0,0,1);
In the step 313, the calculating the number of hidden layers from the classical formula includes:
Calculating the number of hidden layers by using a classical formula to obtain values of 1-10, and setting the number of hidden layers to be 10;
in the step 321, the calculating the connection weight and the threshold according to the BP neural network structure obtained in the step 310, determining the particle length of the particle swarm algorithm according to the connection weight and the threshold, and performing vector encoding on each particle includes:
according to the obtained BP neural network structure of 9-10-5, the weight values of 9 multiplied by 10+10 multiplied by 5=140 and the threshold values of 10+5=15 are calculated, the number of the particle swarm optimization parameters is 155, and the particle length is 155;
Then particle i is encoded as:
preferably, after the step 320, the method further includes:
step 330, preprocessing the logging characteristic parameters in step 100, where the logging characteristic parameters in the preprocessing step 100 include:
step 331, normalizing the logging characteristic parameters in step 100.
Preferably, in the step 331, the normalizing the logging characteristic parameter in the step 100 includes:
normalization processing is carried out on logging characteristic parameters in the step 100 by adopting a maximum and minimum method:
where x min、xmax is the minimum and maximum value in the sequence, respectively.
Preferably, in the step 100, the constructing logging characteristic parameters corresponding to different downhole complex conditions includes:
Combing and generalizing a qualitative recognition method of underground complex conditions and faults by combining a traditional expert knowledge judgment method; and (3) acquiring logging characteristic parameters corresponding to different underground complex conditions and faults by judging expert knowledge of the underground complex conditions and faults.
Preferably, if the downhole complications and malfunctions include: the well kick, lost circulation, stuck drill and well wall collapse are judged through expert knowledge of underground complex conditions and faults, and the well logging characteristic parameters corresponding to different underground complex conditions and faults are obtained through arrangement, wherein the well logging characteristic parameters comprise:
(1) Constructing logging characteristic parameters related to the kick, wherein the logging characteristic parameters related to the kick comprise: outlet flow, outlet density, outlet temperature, outlet resistivity, total volume of drilling fluid pool, pump pressure, pump stroke, hook load, rate of penetration, and gas measurement component;
(2) Constructing logging characteristic parameters related to lost circulation, wherein the logging characteristic parameters related to lost circulation comprise: outlet flow, drilling fluid pool volume, outlet density, mechanical drilling rate, pumping pressure, rotary table drilling rate and drilling pressure;
(3) Constructing a logging characteristic parameter related to a stuck drill, wherein the logging characteristic parameter related to the stuck drill comprises: hook load, vertical weight, torque, rate of penetration, rotary table rotational speed, weight on bit;
(4) Constructing logging characteristic parameters related to well wall collapse, wherein the logging characteristic parameters related to the well wall collapse comprise: cuttings logging, outlet flow, outlet density, torque, rotary table rotational speed, and rate of penetration.
Preferably, in the step 200, the constructing a BP neural network intelligent algorithm optimized based on a particle swarm algorithm includes:
step 210, initializing parameters, where the initializing parameters include:
Determining a BP neural network topological structure, and initializing a connection weight and a threshold value of the BP neural network topological structure; determining a particle swarm dimension D according to the connection weight and the number of the threshold values; setting a particle swarm population scale M and BP neural network iteration times N; carrying out real number coding on the connection weight and the threshold value to obtain an initial population of the particle swarm; and setting the particle speed s i,d and the particle position z i,d within the allowable range [ s min,smax]、[zmin,zmax ]; setting a learning factor c 1、c2;
Step 220, using the root mean square error obtained by each iteration in the BP neural network as an adaptability function of the particles;
Step 230, according to the learning step of the particle swarm algorithm, solving the global optimal position of the particles;
Step 240, checking whether the iteration termination condition is met, if so, stopping outputting the optimal particles, and performing inverse decoding to obtain an optimal weight and a threshold;
Step 250, training and predicting according to the BP neural network.
Preferably, after the step 300, the method further includes:
And 400, training and optimizing the underground complex working condition early warning model in the step 300.
Preferably, in the step 400, the training and tuning the downhole complex condition early warning model in the step 300 includes:
and (3) selecting comprehensive logging parameters of the well which is subjected to fault drilling as training samples by utilizing the underground complex working condition early warning model constructed in the step 300 and combining with specific shale gas drilling field practice, and training the established model and optimizing the model parameters.
Preferably, the method further comprises:
and 500, establishing a transmission interface between the comprehensive logging data and the underground complex working condition early warning model.
According to the technical scheme, the intelligent identification method for the complex underground drilling working conditions has the following beneficial effects: compared with traditional underground complex working condition prediction, the method is excessively dependent on subjective judgment of experts, and the result is mostly qualitative or semi-quantitative. The invention provides an underground complex working condition early warning method based on a particle swarm optimization BP neural network, and an underground complex working condition early warning model is established, so that the intelligent and real-time quantitative judgment of underground complex faults is realized, and the problems of poor comprehensive utilization capability of monitoring information, insufficient time of risk early warning, strong subjectivity and the like of the existing method are solved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a method for intelligently identifying complex conditions under a well drilling well, which is provided by an embodiment of the invention;
FIG. 2 is a flowchart illustrating steps of a particle swarm optimization BP neural network algorithm according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating steps of a particle swarm algorithm according to an embodiment of the present invention;
Fig. 4 is a flowchart of steps of a BP neural network algorithm according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a BP neural network for underground complex operating mode early warning provided by an embodiment of the invention;
fig. 6 is a graph of a pre-warning model for optimizing a BP neural network under complex conditions based on a particle swarm algorithm according to an embodiment of the present invention.
Detailed Description
The invention discloses an intelligent identification method for underground complex drilling conditions based on a particle swarm optimization BP neural network, which establishes a real-time early warning model for underground complex drilling conditions (such as surge, leakage, collapse, clamping, drilling tool breakage and the like) based on intelligent algorithms such as an artificial neural network and the like, and realizes predictability and controllability of underground complex conditions and faults. The invention has important significance for improving the single well profit level of deep well complex stratum drilling, reducing complex and fault loss and improving the competitive power of teams in a work area.
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The intelligent identification method for the complex underground drilling conditions provided by the embodiment of the invention, as shown in figure 1, comprises the following steps:
step 100, constructing logging characteristic parameters corresponding to different underground complex working conditions;
Step 200, constructing an intelligent BP neural network algorithm based on particle swarm optimization; it should be noted that, the BP neural network has a defect of itself, it is difficult to obtain an optimal value of the connection weight and the threshold value, and a larger error is generated between the predicted result and the actual value of the BP neural network, which affects the accuracy of the result. For this, it is necessary to combine with another optimal solution algorithm to use with the BP neural network, so as to improve the accuracy and precision of prediction. Preferably, the invention utilizes the capacity of global searching optimal solution of the particle swarm algorithm to cooperate with BP neural network for use, so as to finally improve the accuracy of BP neural network prediction;
step 300, creating an underground complex working condition early warning model based on the BP neural network intelligent algorithm in step 200, comprising:
It should be noted that the artificial neural network is an artificial intelligence theory which is relatively fluidic at present, the early warning model is created by adopting the BP neural network intelligent algorithm, the method is mainly based on the outstanding characteristic of real-time learning capability, the defect of the traditional artificial intelligence processing unstructured information based on logical symbols is overcome, and the requirements of prediction and judgment of drilling engineering abnormality and risk are exactly met by utilizing the characteristics and advantages of the artificial neural network; in addition, the invention only provides a method for creating an underground complex working condition early warning model based on the BP neural network intelligent algorithm, and the method also needs to input data for further early warning, and the process mainly comprises the following steps: 1. acquiring comprehensive logging parameters; 2. the comprehensive logging parameters are input into an underground complex working condition early warning model, and working condition results are output by utilizing a BP neural network intelligent algorithm, and are not described in detail herein;
Step 310, designing a BP neural network, where the designing the BP neural network includes:
Step 311, designing an input layer, where the input layer includes:
selecting logging characteristic parameters with strong correlation with underground complex working conditions in the step 100 to be set as input neurons;
step 312, designing an output layer, the design output layer including:
setting the number of output neurons according to the number of underground complex working condition types, and representing each output neuron by adopting an expected output vector;
step 313, designing an underlying layer, the underlying layer comprising:
Calculating the number of hidden layers by a classical formula;
Step 320, designing a particle swarm algorithm, wherein the particle swarm algorithm comprises:
step 321, calculating a connection weight and a threshold value according to the BP neural network structure obtained in the step 310, determining the particle length of a particle swarm algorithm according to the connection weight and the threshold value, and carrying out vector coding on each particle;
Step 322, creating an fitness function, where creating the fitness function includes:
taking root mean square error of BP neural network training as fitness function:
According to the technical scheme, the intelligent identification method for the complex underground drilling working conditions has the following beneficial effects:
compared with traditional underground complex working condition prediction, the method is excessively dependent on subjective judgment of experts, and the result is mostly qualitative or semi-quantitative. The invention provides an underground complex working condition early warning method based on a particle swarm optimization BP neural network, and an underground complex working condition early warning model is established, so that the intelligent and real-time quantitative judgment of underground complex faults is realized, and the problems of poor comprehensive utilization capability of monitoring information, insufficient time of risk early warning, strong subjectivity and the like of the existing method are solved.
In this scenario, in the step 311, the logging characteristic parameters include: the number of input neurons is set to 9 according to the hook load, vertical pressure, torque, mechanical drilling speed, rotating speed of a rotary table, drilling pressure, drilling fluid outlet flow, drilling fluid outlet density and total volume of a drilling fluid pool;
In the step 312, the downhole complex operating condition types include: kick, collapse, lost circulation, sticking and normal, 5 output neurons are set, and the corresponding expected output vectors are:
q1＝(1,0,0,0,0)、q2＝(0,1,0,0,0)、q3＝(0,0,1,0,0)、q4＝(0,0,0,1,0)、q5＝(0,0,0,0,1); It should be noted that, each type of underground complex working condition corresponds to one or several of the above 9 logging feature parameters, and there may be one type of underground complex working condition and another type of underground complex working condition simultaneously corresponds to one of the above 9 logging feature parameters, as detailed below;
In the step 313, the calculating the number of hidden layers from the classical formula includes:
calculating the number of hidden layers by using a classical formula to obtain values of 1-10, and setting the number of hidden layers to be 10; it should be noted that, by researching that the number of hidden layers is preferably 10, the obtained result is most consistent with the actual situation; the BP neural network structure is finally established to be 9-10-5, as shown in fig. 5 and 6;
in the step 321, the calculating the connection weight and the threshold according to the BP neural network structure obtained in the step 310, determining the particle length of the particle swarm algorithm according to the connection weight and the threshold, and performing vector encoding on each particle includes:
According to the obtained BP neural network structure of 9-10-5, the number of the particle swarm optimization parameters is 155 and the length of the particles is 155 when 9×10+10×5=140 connection weights and 10+5=15 thresholds are obtained through calculation;
Then particle i is encoded as:
Specifically, after the step 320, the method further includes:
step 330, preprocessing the logging characteristic parameters in step 100, where the logging characteristic parameters in the preprocessing step 100 include:
Step 331, normalizing the logging characteristic parameters in step 100. The method is used for carrying out data normalization processing on logging data so as to avoid the problem of increased prediction result errors of the BP neural network caused by large difference of the orders of magnitude of input data and output data, and ensure the accuracy of an early warning model created based on the BP neural network.
In order to further optimize the above technical solution, in the step 331, the normalizing the logging characteristic parameters in the step 100 includes:
normalization processing is carried out on logging characteristic parameters in the step 100 by adopting a maximum and minimum method:
where x min、xmax is the minimum and maximum value in the sequence, respectively. It should be noted that, the normalization processing of the data is performed by using the maximum and minimum method, so as to obtain a good data preprocessing effect.
In this scenario, in the step 100, the constructing logging feature parameters corresponding to different downhole complex conditions includes:
Combing and summarizing a qualitative recognition method of underground complex conditions and faults by combining a traditional expert knowledge judgment method, and taking the method as one of important basic works for artificial intelligent risk prediction; expert knowledge of underground complex conditions and faults is judged, logging characteristic parameters corresponding to different underground complex conditions and faults are obtained in an arrangement mode, namely the logging characteristic parameters corresponding to different underground complex conditions and faults are built, and the logging characteristic parameters are used as modeling bases of underground complex working condition early warning models.
Specifically, if the downhole complications and faults include: the well kick, lost circulation, stuck drill and well wall collapse are judged through expert knowledge of underground complex conditions and faults, and the well logging characteristic parameters corresponding to different underground complex conditions and faults are obtained through arrangement, wherein the well logging characteristic parameters comprise:
(1) Constructing logging characteristic parameters related to the kick, wherein the logging characteristic parameters related to the kick comprise: outlet flow, outlet density, outlet temperature, outlet resistivity, total volume of drilling fluid pool, pump pressure, pump stroke, hook load, rate of penetration, and gas measurement component; the expert knowledge of the kick is determined as: ① The outlet flow of the drilling fluid is increased, the volume of the drilling fluid pool is increased, fluid in the well overflows during the tripping process, and the well head drilling fluid overflows after the stopping process; ② The pumping pressure is increased firstly, then decreased and the pumping pressure is increased; ③ The load of the big hook is increased and then decreased; ④ The rate of penetration suddenly increases; ⑤ Drilling fluid density decreases and outlet temperature, resistivity, etc. fluctuate. ⑥ Drilling fluid oil, gas and water invasion. The gas total hydrocarbon and component detection value may increase;
(2) Constructing logging characteristic parameters related to lost circulation, wherein the logging characteristic parameters related to lost circulation comprise: outlet flow, drilling fluid pool volume, outlet density, mechanical drilling rate, pumping pressure, rotary table drilling rate and drilling pressure; it should be noted that, the expert knowledge judgment of lost circulation includes: ① And judging lost circulation in drilling and judging lost circulation in ② tripping processes. In the determination of lost circulation during drilling, the lithology of the stratum is determined by the sand sample which is fetched during drilling, and when the stratum is drilled into permeable stratum, the density is too high or pumping pressure fluctuation exists during drilling, so that lost circulation can be caused. In the case of high rates of mechanical drilling, lost circulation may occur if pump pressure is reduced, the outlet displacement is reduced, and the total pool level is reduced; in the well leakage judgment in the tripping process, whether drilling fluid returns out of the tripping is monitored, and the monitoring can be performed according to a discharge conductivity sensor arranged at a drilling fluid return outlet;
(3) Constructing a logging characteristic parameter related to a stuck drill, wherein the logging characteristic parameter related to the stuck drill comprises: hook load, vertical weight, torque, rate of penetration, rotary table rotational speed, weight on bit; it should be noted that, different types of stuck drills correspond to different judging methods, and the stuck drill judging methods are summarized by combining on-site drilling practice and judging according to expert knowledge, as shown in table 1.
TABLE 1
(4) Constructing logging characteristic parameters related to well wall collapse, wherein the logging characteristic parameters related to the well wall collapse comprise: cuttings logging, outlet flow, outlet density, torque, rotary table rotational speed, and rate of penetration. The method is characterized in that the basis for judging the collapse of the well wall is the condition of returned rock debris in the well, and the returned rock debris has larger volume, more quantity, complex lithology and uneven color compared with the normal rock debris.
Specifically, as shown in fig. 2, in the step 200, the constructing a BP neural network intelligent algorithm optimized based on a particle swarm algorithm includes:
step 210, initializing parameters, where the initializing parameters include:
Determining a BP neural network topological structure, and initializing a connection weight and a threshold value of the BP neural network topological structure; determining a particle swarm dimension D according to the connection weight and the number of the threshold values; setting a particle swarm population scale M and BP neural network iteration times N; carrying out real number coding on the connection weight and the threshold value to obtain an initial population of the particle swarm; and setting the particle speed s i,d and the particle position z i,d within the allowable range [ s min,smax]、[zmin,zmax ]; setting a learning factor c 1、c2;
Step 220, using the root mean square error obtained by each iteration in the BP neural network as an adaptability function of the particles;
Step 230, according to the learning step of the particle swarm algorithm, solving the global optimal position of the particles;
Step 240, checking whether the iteration termination condition is met, if so, stopping outputting the optimal particles, and performing inverse decoding to obtain an optimal weight and a threshold;
Step 250, training and predicting according to the BP neural network.
For the particle swarm algorithm, it is further explained that: setting up a search for an optimal solution in a D-dimensional space, initializing M particles, a "bird group" t= { Z 1,Z2,...ZM }, i=1, 2..m. wherein Z i＝(zi1,zi2,...ziD), i=1, 2,..m represents a position vector of the i-th particle in the D-dimensional space, s i＝(si1,si2,...siD), i=1, 2, … M represents a velocity vector of the i-th particle in the D-dimensional space; the flight motion experience of the particle itself, pbest i＝(Pbesti1,Pbesti2,...,PbestiD), the global optimum position is gbest= (Gbest 1,Gbest2,…GbestD). The recursive formula of the particle algorithm is:
Individual cognition The memory capacity of the particles to the self optimal position is embodied; at the same time, population cognition/>And information sharing among particles in the population is embodied.
Specifically, as shown in fig. 3, the learning step of the particle swarm algorithm in step 230 includes:
Step 231, initializing particle swarm parameters, wherein the parameters include: population size M, particle population dimension number D, iteration number N, particle velocity s i,d and particle position z i,d, learning factor c 1、c2;
232, selecting an fitness function, calculating fitness values of all particles, and setting Pbest i,d and Gbest of initial particles of a population;
step 233, updating the speed and position of the particles;
Step 234, calculating the adaptive value of the updated particles and updating f (Prest i,d); otherwise Pbest i,d is not updated;
Step 235, checking the end condition, and if the requirements are met, ending the optimizing; otherwise, go to step 233 to proceed.
Accordingly, for the BP neural network intelligent algorithm, it needs to be further explained that:
The BP neural network is a typical multi-layer feedforward network, and the main advantage of the BP neural network is the strongest nonlinear mapping capability. The BP neural network is a three-layer forward artificial neural network consisting of an input layer, an implicit layer and an output layer. The output of each neuron of the hidden layer can be obtained according to the calculation formula of the output signals of the neurons as follows:
Wherein: v ij is the connection weight of the input layer neuron i and the hidden layer neuron j; θ j is the threshold for hidden layer neuron j; f () is the activation function of the neuron.
Similarly, the output signals of the output layer of the BP neural network are obtained as follows:
Wherein: mu jk is the connection weight of the hidden layer neuron j and the output layer neuron k; beta k is the threshold for output layer neuron k; f () is the activation function of the neuron.
As shown in fig. 4, the BP neural network intelligent algorithm has the following learning steps:
(1) Initializing various parameter values of the neural network:
Assigning random numbers between [ -1,1] to the connection weights v ij、μjk of the neurons of each layer; setting the learning rate eta to be a decimal between 0 and 1; setting the error E to 0; the error threshold Emin is set to a positive fraction; the node number m of the hidden layer is obtained according to an empirical formula; the sample pattern calculator p and training times q are reset to 1; the training times are preset as M;
(2) Inputting training samples, and calculating output of each layer:
Selecting a pair of samples in the training samples (X p,Qp) to assign values to the input parameters, and calculating the output O, Y by using a calculation formula of the output signals of the neural network;
(3) Calculating the output error of the neural network:
setting the total number of the training samples as P, and obtaining an error of each training as Total output error/>
(4) Calculating error signals of each layer:
The error signal calculation formulas of the output layer and the hidden layer are as follows:
the error signal calculation formulas of the hidden layer and the input layer are as follows:
(5) Adjusting the connection weight and the threshold value of each layer:
The calculation formula of the connection weight value and the threshold value of the output layer and the hidden layer is as follows:
Δμjk＝η(qk-yk)yk(1-yk)Oj (9)
Δβk＝η(qk-yk)yk(1-yk) (10)
the calculation formula of the connection weight value and the threshold value of the hidden layer and the input layer is as follows:
(6) Judging whether the training is finished for all samples or not:
If the sample pattern calculator P < the total number of training samples P, then both the sample pattern calculator P and the training number q are increased by 1, and then the step (2) is returned; otherwise go to step (7);
(7) Checking whether the total error of the neural network meets an error threshold:
If the total output error E RME is less than the error threshold E min or the training times q is greater than the training preset times M, ending; otherwise, the error E is reset to 0, the sample pattern calculator p is set to 1, and then step (2) is returned.
In this embodiment, after the step 300, the method further includes:
And 400, training and optimizing the underground complex working condition early warning model in the step 300. It should be noted that, through training and tuning the underground complex working condition early warning model, the effect of good initialization is achieved, so that the underground complex working condition early warning can be better served.
Specifically, in the step 400, the training and tuning the downhole complex condition early-warning model in the step 300 includes:
and (3) selecting comprehensive logging parameters of the well which is subjected to fault drilling as training samples by utilizing the underground complex working condition early warning model constructed in the step 300 and combining with specific shale gas drilling field practice, and training the established model and optimizing the model parameters.
Step 410, setting control elements of the algorithm:
control elements of the algorithm in the fault early warning model are set as shown in table 2:
TABLE 2
Step 420, training and simulating a model:
In combination with the well logging and the well Shi Ziliao of the Chuan nan shale gas work area which are well-drilled and have faults, comprehensive logging parameters monitored in a period of time before and after the faults occur are selected, normalization processing is carried out according to a maximum and minimum method, and the results are shown in table 3. The 10 groups of samples are selected as training samples to train the constructed fault model for verifying the accuracy of the constructed model.
TABLE 3 Table 3
The training results of the fault model are shown in table 4, and the results are analyzed: the simulation diagnosis result is matched with the underground actual state, so that the neural network obtained through training can well meet the requirement of the dynamic evaluation of the drilling risk of the interval of the block. Notably, are: under different geological environments and different drilling conditions, the change of logging parameters corresponding to underground abnormal states or risks can be distinguished, so that when a training sample is selected, the training sample is selected according to different drilling environments and conditions, and the neural network obtained through training is only applied to abnormal working conditions and risk diagnosis under similar drilling conditions.
TABLE 4 Table 4
In this scheme, the method for intelligently identifying the complex working conditions under the well drilling well provided by the embodiment of the invention further comprises the following steps:
And 500, establishing a transmission interface between the comprehensive logging data and the underground complex working condition early warning model. It should be noted that, this design is so as to realize the real-time dynamic early warning of trouble. Moreover, the well computer system uses a WITS (Wellsite Information TRANSFER STANDARD, well site information transfer Specification) communication format to transfer various data in the field to the well computer system. Specifically, the transmission of comprehensive logging data from the site to fault early warning software (fault early warning model) for data analysis is subject to the following five steps:
Step 510, logging companies acquire on-site real-time data through sensors on the tools;
Step 520, the collected data is transmitted to the surface via WITS, stored or used for other purposes;
step 530, transmitting data to software in real time in WITS file form or TCP/IP protocol form through WITS data transmission software;
step 540, saving the real-time transmitted data to a local database for other purposes;
step 550, the fault software directly reads the data in the local database for dynamic early warning of faults.
Furthermore, the scheme also compiles related interface software, and realizes real-time data transmission between comprehensive logging data and underground complex condition and fault early warning software.
According to the data to be transmitted, finding corresponding parameters in the WITS record: hook load, vertical weight, torque, rate of penetration, rotational speed of the rotary table, weight on bit, drilling fluid outlet flow, drilling fluid outlet density, and total volume of the drilling fluid pool.
In summary, according to the intelligent identification method for the complex underground drilling working condition provided by the invention, firstly, the learning step of the BP neural network is analyzed for the system arrangement, the defect of the BP neural network is analyzed, and an algorithm for optimizing the BP neural network based on particle swarm is provided; then, an algorithm based on particle swarm optimization BP neural network is applied to real-time early warning of underground complex working conditions, so that intelligent and real-time quantitative judgment of underground complex conditions and faults is realized; meanwhile, a data transmission interface of comprehensive logging data and underground complex condition and fault early warning software is compiled, and real-time data transmission between the comprehensive logging data and the underground complex condition early warning software is realized.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. The intelligent identification method for the complex underground working conditions of the well drilling is characterized by comprising the following steps of:
step 100, constructing logging characteristic parameters corresponding to different underground complex working conditions;
step 200, constructing an intelligent BP neural network algorithm based on particle swarm optimization;
step 300, creating an underground complex working condition early warning model based on the BP neural network intelligent algorithm in step 200, comprising:
step 310, designing a BP neural network, including:
Step 311, designing an input layer: selecting logging characteristic parameters with strong correlation with underground complex working conditions in the step 100 to be set as input neurons;
Step 312, designing an output layer:
setting the number of output neurons according to the number of underground complex working condition types, and representing each output neuron by adopting an expected output vector;
step 313, designing hidden layer:
Calculating the number of hidden layers by a classical formula;
step 320, designing a particle swarm algorithm, including:
step 321, calculating a connection weight and a threshold value according to the BP neural network structure obtained in the step 310, determining the particle length of a particle swarm algorithm according to the connection weight and the threshold value, and carrying out vector coding on each particle;
Step 322, creating a fitness function:
taking root mean square error of BP neural network training as fitness function:
In the step 200, the constructing a BP neural network intelligent algorithm optimized based on a particle swarm algorithm includes:
step 210, initializing parameters, where the initializing parameters include:
Determining a BP neural network topological structure, and initializing a connection weight and a threshold value of the BP neural network topological structure; determining a particle swarm dimension D according to the connection weight and the number of the threshold values; setting a particle swarm population scale M and BP neural network iteration times N; carrying out real number coding on the connection weight and the threshold value to obtain an initial population of the particle swarm; and setting the particle speed s i,d and the particle position z i,d within the allowable range [ s min,smax]、[zmin,zmax ]; setting a learning factor c 1、c2;
Step 220, using the root mean square error obtained by each iteration in the BP neural network as an adaptability function of the particles;
Step 230, according to the learning step of the particle swarm algorithm, solving the global optimal position of the particles;
Step 240, checking whether the iteration termination condition is met, if so, stopping outputting the optimal particles, and performing inverse decoding to obtain an optimal weight and a threshold;
Step 250, training and predicting according to the BP neural network.
2. The method for intelligently identifying complex downhole conditions of drilling according to claim 1, wherein in the step 311, the logging characteristic parameters include: the number of input neurons is set to 9 according to the hook load, vertical pressure, torque, mechanical drilling speed, rotating speed of a rotary table, drilling pressure, drilling fluid outlet flow, drilling fluid outlet density and total volume of a drilling fluid pool;
In the step 312, the downhole complex operating condition types include: kick, collapse, lost circulation, sticking and normal, 5 output neurons are set, and the corresponding expected output vectors are:
q1＝(1,0,0,0,0)、q2＝(0,1,0,0,0)、q3＝(0,0,1,0,0)、q4＝(0,0,0,1,0)、q5＝(0,0,0,0,1);
In the step 313, the calculating the number of hidden layers from the classical formula includes:
Calculating the number of hidden layers by using a classical formula to obtain values of 1-10, and setting the number of hidden layers to be 10;
in the step 321, the calculating the connection weight and the threshold according to the BP neural network structure obtained in the step 310, determining the particle length of the particle swarm algorithm according to the connection weight and the threshold, and performing vector encoding on each particle includes:
according to the obtained BP neural network structure of 9-10-5, the weight values of 9 multiplied by 10+10 multiplied by 5=140 and the threshold values of 10+5=15 are calculated, the number of the particle swarm optimization parameters is 155, and the particle length is 155;
Then particle i is encoded as:
3. The method of claim 1, further comprising, after step 320:
step 330, preprocessing the logging characteristic parameters in step 100, where the logging characteristic parameters in the preprocessing step 100 include:
step 331, normalizing the logging characteristic parameters in step 100.
4. The method for intelligently identifying complex downhole conditions according to claim 3, wherein in the step 331, the normalizing the logging characteristic parameters in the step 100 includes:
normalization processing is carried out on logging characteristic parameters in the step 100 by adopting a maximum and minimum method:
where x min、xmax is the minimum and maximum value in the sequence, respectively.
5. The method according to claim 1, wherein in step 100, the constructing logging characteristic parameters corresponding to different downhole complex conditions comprises:
Combing and generalizing a qualitative recognition method of underground complex conditions and faults by combining a traditional expert knowledge judgment method; and (3) acquiring logging characteristic parameters corresponding to different underground complex conditions and faults by judging expert knowledge of the underground complex conditions and faults.
6. The method of claim 5, wherein if the downhole complex conditions and faults include: the well kick, lost circulation, stuck drill and well wall collapse are judged through expert knowledge of underground complex conditions and faults, and the well logging characteristic parameters corresponding to different underground complex conditions and faults are obtained through arrangement, wherein the well logging characteristic parameters comprise:
(1) Constructing logging characteristic parameters related to the kick, wherein the logging characteristic parameters related to the kick comprise: outlet flow, outlet density, outlet temperature, outlet resistivity, total volume of drilling fluid pool, pump pressure, pump stroke, hook load, rate of penetration, and gas measurement component;
(2) Constructing logging characteristic parameters related to lost circulation, wherein the logging characteristic parameters related to lost circulation comprise: outlet flow, drilling fluid pool volume, outlet density, mechanical drilling rate, pumping pressure, rotary table drilling rate and drilling pressure;
(3) Constructing a logging characteristic parameter related to a stuck drill, wherein the logging characteristic parameter related to the stuck drill comprises: hook load, vertical weight, torque, rate of penetration, rotary table rotational speed, weight on bit;
(4) Constructing logging characteristic parameters related to well wall collapse, wherein the logging characteristic parameters related to the well wall collapse comprise: cuttings logging, outlet flow, outlet density, torque, rotary table rotational speed, and rate of penetration.
7. The method for intelligently identifying complex conditions downhole in a well according to claim 1, further comprising, after step 300:
And 400, training and optimizing the underground complex working condition early warning model in the step 300.
8. The method according to claim 7, wherein in the step 400, the training and tuning the downhole complex condition pre-warning model in the step 300 comprises:
and (3) selecting comprehensive logging parameters of the well which is subjected to fault drilling as training samples by utilizing the underground complex working condition early warning model constructed in the step 300 and combining with specific shale gas drilling field practice, and training the established model and optimizing the model parameters.
9. The method for intelligently identifying complex conditions under a well drilling well according to claim 1, further comprising:
and 500, establishing a transmission interface between the comprehensive logging data and the underground complex working condition early warning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911404045.4A CN113062731B (en) | 2019-12-30 | 2019-12-30 | Intelligent identification method for complex underground drilling conditions |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911404045.4A CN113062731B (en) | 2019-12-30 | 2019-12-30 | Intelligent identification method for complex underground drilling conditions |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113062731A CN113062731A (en) | 2021-07-02 |
| CN113062731B true CN113062731B (en) | 2024-05-21 |
Family
ID=76558446
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911404045.4A Active CN113062731B (en) | 2019-12-30 | 2019-12-30 | Intelligent identification method for complex underground drilling conditions |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113062731B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114991746B (en) * | 2021-11-23 | 2024-01-19 | 中国石油天然气集团有限公司 | Intelligent calibration method and system for drilling working conditions |
| CN114112822B (en) * | 2021-11-26 | 2023-09-29 | 中国地质大学（武汉） | Drilling fluid dynamic sand carrying capacity simulation system and prediction method |
| CN114482885B (en) * | 2022-01-25 | 2024-03-29 | 西南石油大学 | Intelligent control system for pressure-controlled drilling |
| CN115324571B (en) * | 2022-08-02 | 2023-10-10 | 西南石油大学 | Complex stratum well wall collapse quantitative prediction method and device based on drilling and logging information |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104121014A (en) * | 2014-06-16 | 2014-10-29 | 西南石油大学 | Method for diagnosing type of leakage of drilled well based on neural network fusion technique |
| CN104453875A (en) * | 2014-10-29 | 2015-03-25 | 中国石油集团川庆钻探工程有限公司 | Shale gas reservoir identification method based on self-organizing competitive neural network |
| CN104675380A (en) * | 2015-01-28 | 2015-06-03 | 扬州大学 | Online oil-drilling drill string monitoring system and fault diagnosis method |
| CN104806226A (en) * | 2015-04-30 | 2015-07-29 | 北京四利通控制技术股份有限公司 | Intelligent drilling expert system |
| CN105089620A (en) * | 2014-05-14 | 2015-11-25 | 中国石油天然气集团公司 | Drilling tool jamming monitoring system, drilling tool jamming monitoring method and drilling tool jamming monitoring device |
| CN107100611A (en) * | 2017-05-27 | 2017-08-29 | 中国地质大学（武汉） | A kind of geological prospecting drilling process Double-layer intelligent drilling speed modeling method |
| CN107194524A (en) * | 2017-07-28 | 2017-09-22 | 合肥工业大学 | A kind of Coal and Gas Outbursts Prediction method based on RBF neural |
| CN109763810A (en) * | 2018-12-25 | 2019-05-17 | 内江师范学院 | A kind of reservoir gas drilling risk recognition methods |
-
2019
- 2019-12-30 CN CN201911404045.4A patent/CN113062731B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105089620A (en) * | 2014-05-14 | 2015-11-25 | 中国石油天然气集团公司 | Drilling tool jamming monitoring system, drilling tool jamming monitoring method and drilling tool jamming monitoring device |
| CN104121014A (en) * | 2014-06-16 | 2014-10-29 | 西南石油大学 | Method for diagnosing type of leakage of drilled well based on neural network fusion technique |
| CN104453875A (en) * | 2014-10-29 | 2015-03-25 | 中国石油集团川庆钻探工程有限公司 | Shale gas reservoir identification method based on self-organizing competitive neural network |
| CN104675380A (en) * | 2015-01-28 | 2015-06-03 | 扬州大学 | Online oil-drilling drill string monitoring system and fault diagnosis method |
| CN104806226A (en) * | 2015-04-30 | 2015-07-29 | 北京四利通控制技术股份有限公司 | Intelligent drilling expert system |
| CN107100611A (en) * | 2017-05-27 | 2017-08-29 | 中国地质大学（武汉） | A kind of geological prospecting drilling process Double-layer intelligent drilling speed modeling method |
| CN107194524A (en) * | 2017-07-28 | 2017-09-22 | 合肥工业大学 | A kind of Coal and Gas Outbursts Prediction method based on RBF neural |
| CN109763810A (en) * | 2018-12-25 | 2019-05-17 | 内江师范学院 | A kind of reservoir gas drilling risk recognition methods |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113062731A (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113062731B (en) | Intelligent identification method for complex underground drilling conditions | |
| CN112901137B (en) | Deep well drilling mechanical drilling speed prediction method based on deep neural network Sequential model | |
| CN110852018B (en) | PSO drilling parameter optimization method based on neural network | |
| CN109508827B (en) | Drilling accident early warning method based on time recursion neural network | |
| CN106121621A (en) | A kind of intelligent drilling specialist system | |
| CN110110740B (en) | Drilling process working condition identification method based on multi-time scale features and neural network | |
| CN110807557A (en) | Drilling rate prediction method based on BP neural network and drilling rate optimization method based on BP neural network and particle swarm optimization | |
| CN112364560B (en) | Intelligent prediction method for working hours of mine rock drilling equipment | |
| Chernikov et al. | Application of artificial intelligence methods for identifying and predicting complications in the construction of oil and gas wells: problems and solutions | |
| CN115234220B (en) | Method and device for identifying downhole stick-slip vibration in real time by utilizing intelligent drill bit | |
| CN116384554A (en) | Method and device for predicting mechanical drilling speed, electronic equipment and computer storage medium | |
| CN112926839A (en) | Overflow risk cooperative monitoring method and system for oil and gas well drilling process | |
| CN115860197A (en) | Data-driven coal bed gas yield prediction method and system | |
| CN116066062A (en) | Drilling sticking real-time early warning method based on parameter change trend abnormity diagnosis | |
| CN117035197B (en) | Intelligent lost circulation prediction method with minimized cost | |
| CN111340275B (en) | Tunnel support mode selection real-time prediction method based on detection while drilling technology | |
| CN117172360A (en) | Drilling mechanical drilling speed optimization method, system, equipment and medium based on MLP and high-efficiency PSO | |
| CN111598366A (en) | Real-time drilling auxiliary decision-making method and system | |
| CN113431557B (en) | Underground borehole track tracking method based on artificial intelligence | |
| CN115640759A (en) | Drill jamming early warning method and system based on machine learning | |
| CN113129157A (en) | Underground stuck-drill fault real-time early warning method suitable for shale gas long water section | |
| CN113685166B (en) | Drilling accident early warning method and system | |
| CN115729910A (en) | Implementation method and system for analyzing drilling speed of drilling machine in real time | |
| Mnati et al. | Prediction of penetration rate and cost with artificial neural network for alhafaya oil field | |
| CN112418489A (en) | Neural network prediction method for dosage of temporary plugging agent for repairing natural gas well |
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 |