CN103196424A - Early warning method for settlement of oil gas gob area - Google Patents

Early warning method for settlement of oil gas gob area Download PDF

Info

Publication number
CN103196424A
CN103196424A CN2013101456110A CN201310145611A CN103196424A CN 103196424 A CN103196424 A CN 103196424A CN 2013101456110 A CN2013101456110 A CN 2013101456110A CN 201310145611 A CN201310145611 A CN 201310145611A CN 103196424 A CN103196424 A CN 103196424A
Authority
CN
China
Prior art keywords
value
oil gas
early warning
formula
settlement
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
Application number
CN2013101456110A
Other languages
Chinese (zh)
Other versions
CN103196424B (en
Inventor
韩连福
付长凤
高宇飞
卢召红
刘超
赵冬岩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Linquan Lingcui Technical Service Co Ltd
Original Assignee
Northeast Petroleum University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northeast Petroleum University filed Critical Northeast Petroleum University
Priority to CN201310145611.0A priority Critical patent/CN103196424B/en
Publication of CN103196424A publication Critical patent/CN103196424A/en
Application granted granted Critical
Publication of CN103196424B publication Critical patent/CN103196424B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Emergency Alarm Devices (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to an early warning method for settlement of an oil gas gob area. The method comprises the steps that 1, settlement data is acquired and address settlement data is recorded; 2, the measured data of a settlement sensor is subjected to recognition and correction of an abnormal value; 3, the smoothness of the settlement data is improved; 4, one time of accumulation (1-AGO) is carried out to form a sequence; 5, prediction is carried out through an NDGM (Non-homogenous Discrete Grey Model) (1,1) to obtain a predicted value of the oil gas gob area; 6, inverse accumulation and inverse smoothness variations are carried out to obtain a predicted value of an oil gas settlement area; 7, an early warning value is determined by adopting a BP (Back Propagation) neural network method; and 8, an oil gas gob area settlement early warning model based on grey relation is established; after the settlement of the gob area exceeds a certain critical value, the settlement area has the risk of collapse, and the critical value is referred to as the early warning value; and if the settlement volume of some point exceeds the critical value, early warning is required to be sent out. The method overcomes the defect that the prediction precision is reduced due to the rough data, and the prediction precision is improved.

Description

Oil gas goaf sedimentation method for early warning
One, technical field:
What the present invention relates to is field of measuring technique, and what be specifically related to is oil gas goaf sedimentation method for early warning.
Two, background technology:
Oil reservoir or gas-bearing formation can be at dummy sections of underground formation after exploitation, and the ground that is positioned at this top, zone is called as the oil gas goaf.Because the oil gas goaf is empty zone, geology balance originally is destroyed, ground sedimentation always is up to realizing new balance, especially precipitation continuously, mulching material physique amount heightens suddenly or the interference of factor such as small-sized earthquake under, the goaf sedimentation is more obvious.
The light meeting of the sedimentation in goaf causes the distortion of face of land building, the fission of body of wall, destroys the bearing capacity of buildings, thereby can cause building collapsing, causes casualties.The sedimentation in goaf also can cause road the crack to occur, and serious meeting forms karst topography, causes the major traffic accidents of car crash.In the planning process of actual building, accomplish that building is avoided sedimentation seriously and settling velocity is regional faster, to guarantee the safety of buildings.Thereby need come the following settling amount in predicting oil goaf according to the settling amount in existing oil gas goaf, when predicting settling amount greater than early warning value, send the early warning announcement in advance and avoid personnel and property loss.
Settlement prediction disposal route in goaf mainly is methods such as support vector machine, Kalman filtering and Hilbert transform at present, but there is following shortcoming in existing settlement prediction method and can't be applied to the oil gas goaf:
One: the modeling of existing settlement prediction method needs a large amount of measurement data, and the measurement data in oil gas goaf is less.
Two: existing settlement prediction method requires reliable measuring data strong, but the measurement data in oil gas goaf is to obtain in open relatively environment, inevitably can occur measuring abnormal data in the measurement data owing to be subjected to influencing of external interference factor.
Three: existing settlement prediction method requires measurement data measured value changes in amplitude bigger, but changes characteristics more slowly because oil gas goaf measurement data has, thereby existing settlement prediction method can't be used in the measurement in oil gas goaf.
It is few, low and to advantages such as data changes in amplitude no requirement (NR)s to the data quality requirements that gray system theory has a modeling desired data, therefore can adopt specific method to be applied to during the sedimentation early warning of oil gas goaf handles.Because oil gas goaf measurement data form is ever-changing, the classical model of gray system--GM (1,1) model can't be directly applied in the processing of oil gas goaf settlement prediction inadequately because of precision, and the settling data in goaf meets weak nonhomogeneous index law, thereby oil gas goaf settling data is adopted gray system nonhomogeneous index discrete model--NDGM (1,1) handles.
Three, summary of the invention:
An object of the present invention is to provide oil gas goaf sedimentation method for early warning, this oil gas goaf sedimentation method for early warning has the problem of being strict with and can't being applicable to the settlement prediction of oil gas goaf for solving quantity, quality and the changes in amplitude of existing settlement prediction method to measurement data.
The technical solution adopted for the present invention to solve the technical problems is: this oil gas goaf sedimentation method for early warning is as follows:
Step 1: adopt the sky negative area at oil gas and lay settlement sensor, settlement sensor is connected with server, settlement sensor is gathered the friendship of settling data signal and is transferred to server, and server is handled settling data, at first, and the recording address settling data , wherein
Figure 518912DEST_PATH_IMAGE002
,
Figure 657769DEST_PATH_IMAGE003
,
Figure 428148DEST_PATH_IMAGE004
Be the measurement sequence number, the Be measurement point position sequence number;
Step 2: the measurement data of settlement sensor is carried out identification and the correction of exceptional value, identify exceptional value
Figure 600820DEST_PATH_IMAGE006
, the adjacent difference ratio of NDGM (1,1) model Be constant, so can calculate according to adjacent difference
Figure 669719DEST_PATH_IMAGE006
Modified value
Figure 660808DEST_PATH_IMAGE008
, its expression formula is as follows:
Figure 500588DEST_PATH_IMAGE009
；
In the formula,
Figure 168199DEST_PATH_IMAGE010
Be
Figure 460640DEST_PATH_IMAGE005
The adjacent difference ratio of sequence is measured in individual measuring position;
Step 3: strengthen the slickness of settling data, right
Figure 255421DEST_PATH_IMAGE001
Carry out pre-service one by one, generate
Figure 638123DEST_PATH_IMAGE006
, its expression formula is:
Figure 289684DEST_PATH_IMAGE011
；
Step 4: for strengthening the adaptability of oil gas goaf settling data modeling, will
Figure 272683DEST_PATH_IMAGE006
Carry out one-accumulate (1-AGO) and form sequence
Figure 667893DEST_PATH_IMAGE012
Step 5: right
Figure 403636DEST_PATH_IMAGE012
Adopt NDGM (1,1) model to predict the predicted value that obtains the oil gas goaf
Figure 491678DEST_PATH_IMAGE013
, its expression formula is:
Figure 696395DEST_PATH_IMAGE014
In the formula,
Figure 895295DEST_PATH_IMAGE015
,
Figure 987010DEST_PATH_IMAGE016
,
Figure 183636DEST_PATH_IMAGE017
With
Figure 937965DEST_PATH_IMAGE018
It is the parameter of NDGM (1,1) model;
Step 6: counter adding up obtained the predicted value of oil gas negative area with reflective sliding the variation
Figure 861928DEST_PATH_IMAGE019
,
Figure 385313DEST_PATH_IMAGE013
Be that original predicted data strengthens smooth ratio and the later predicted value that adds up for 1 time through 1 time
Figure 752840DEST_PATH_IMAGE020
, for obtaining the predicted value of oil gas negative area , it is right to need
Figure 946286DEST_PATH_IMAGE013
Carry out inverse transformation, its expression formula is as follows:
Figure 589757DEST_PATH_IMAGE021
Figure 862607DEST_PATH_IMAGE022
Figure 591528DEST_PATH_IMAGE023
Figure 857293DEST_PATH_IMAGE024
；
Step 7: adopt the BP neural network method to determine early warning value according to quantity, geologic structure, the quality of above ground structure, the excess surface water situation of sampled point
Figure 292954DEST_PATH_IMAGE025
Step 8: set up based on the related oil gas goaf sedimentation Early-warning Model of ash, when the goaf sedimentation reach surpass certain critical value after, just there is the danger of subsiding the negative area, this critical value
Figure 799022DEST_PATH_IMAGE025
Be called early warning value, if the settling amount of certain point surpasses critical value, then need to send early warning.
Further, for the early warning that prevents from sending in the such scheme is possible be subjected to extraneous interference should not report to the police and produce warning, need to rely on the measurement data of whole measured zone to judge whether to report to the police, thereby set up following oil gas goaf sedimentation Early-warning Model:
Figure 703655DEST_PATH_IMAGE026
，
In the formula, It is measurement point iThe measurement data predicted value With
Figure 818875DEST_PATH_IMAGE029
Between related coefficient, its expression formula is as follows:
In the formula
Figure 864378DEST_PATH_IMAGE031
Expression
Figure 274630DEST_PATH_IMAGE028
With
Figure 810916DEST_PATH_IMAGE029
Between minor increment,
Figure 736147DEST_PATH_IMAGE032
Figure 898138DEST_PATH_IMAGE028
With
Figure 225214DEST_PATH_IMAGE029
Between ultimate range,
Figure 696516DEST_PATH_IMAGE028
With
Figure 109042DEST_PATH_IMAGE029
Between minor increment,
Figure 809145DEST_PATH_IMAGE033
It is coefficient of autocorrelation.
The recognition methods of in the such scheme measurement data of settlement sensor being carried out exceptional value is that the exceptional value criterion is as follows:
Figure 990728DEST_PATH_IMAGE034
In the formula
Figure 134395DEST_PATH_IMAGE035
Be exceptional value identification coefficient, general value is 3,
Figure 34218DEST_PATH_IMAGE036
Be
Figure 538012DEST_PATH_IMAGE005
The mean value of individual measurement point sedimentation;
If following formula is false, so
Figure 574101DEST_PATH_IMAGE006
It or not exceptional value; If following formula is set up, so
Figure 387205DEST_PATH_IMAGE006
It is exceptional value.
In the such scheme in the step 5
Figure 712007DEST_PATH_IMAGE015
,
Figure 816230DEST_PATH_IMAGE016
,
Figure 660820DEST_PATH_IMAGE017
With
Figure 457875DEST_PATH_IMAGE018
Expression formula as follows:
Figure 4394DEST_PATH_IMAGE037
Figure 364837DEST_PATH_IMAGE039
In the formula, matrix
Figure 375518DEST_PATH_IMAGE040
Be neighbour's matrix,
Figure 281157DEST_PATH_IMAGE041
Be grey background matrix, its expression formula is respectively:
Figure 377289DEST_PATH_IMAGE042
，
Figure 511729DEST_PATH_IMAGE043
’
Adopt the BP neural network method to determine early warning value in the such scheme
Figure 580180DEST_PATH_IMAGE025
Method be:
With quality d, excess surface water situation water, underground oil and gas total production p, gas production q and the petroleum production t of sampled point number number, geologic structure mv, above ground structure, as the parameter of BP neural network input layer;
The middle layer of BP neural network is hidden layer, and adopts 8 neurons;
With early warning value
Figure 719037DEST_PATH_IMAGE025
As the output parameter of BP neural network output layer, set up sample set A=(a 1, a 2..., a n), a wherein i=(number i , mv i , d i , water i , p i , q i , t i ,
Figure 489416DEST_PATH_IMAGE044
), wherein
Figure 739131DEST_PATH_IMAGE045
Adopt sample set A to train, obtain input layer to hidden layer weight matrix w1 IjArrive output layer weight matrix w2 with hidden layer Ij, w1 wherein IjMiddle i=8 is the hidden neuron number, and j=7 is the number of input layer parameter, w2 IjIn wherein i=1 be output layer neuron number, j=8 is the number of input layer parameter, and learning and memory, obtains early warning value
Beneficial effect:
1, the present invention adopts the smooth treatment technology that original oil gas goaf settling amount has been carried out pre-service, has overcome because the defective that the rough precision of prediction that causes of data reduces has improved accuracy of predicting.
2, the present invention adopts the exceptional value treatment technology that original oil gas settling amount has been carried out exceptional value must to identify and revise, and has got rid of the sedimentation false alarm that the existence owing to exceptional value causes, and has improved accuracy of predicting.
3, the present invention has adopted the grey correlation analysis technology to obtain the information of forecasting in whole oil gas goaf, thereby has got rid of the possibility of single-point false alarm, has improved accuracy of predicting.
4, the present invention adopts nerual network technique to determine the sedimentation early warning value in the oil gas goaf under different regions, the different situations
Figure 660262DEST_PATH_IMAGE025
, guaranteed early warning value
Figure 730986DEST_PATH_IMAGE025
The specific aim of value has improved the precision of early warning.
5, the present invention is not only applicable to the sedimentation early warning of oil gas goaf, and is applicable to the goaf sedimentation early warning after other mineral resources such as colliery, iron ore are adopted sky.
Four, description of drawings:
The BP neural network model figure that Fig. 1 adopts for the inventive method; Fig. 2 adopts oil gas goaf survey sensor layout in the embodiment two, wherein rhombus represents the oil gas goaf, pentagram representative sensor layout; Fig. 3 contains the oil gas goaf Sensor section measurement data of measuring exceptional value in the embodiment two, horizontal ordinate represents the collection sequence number of settling amount, ordinate represents sedimentation value, the value of measurement data 3 is 2 microns among the figure, and other measured values are several microns at zero point, so measurement data 3 is for measuring exceptional value; Oil gas goaf Sensor section measurement data in Fig. 4 embodiment two after the correction exceptional value, horizontal ordinate represents the collection sequence number of settling amount, and ordinate represents sedimentation value, and the value of measurement data 3 is 0.15 micron among the figure, be more or less the same with other measured value, so the modified value of measurement data 3 is reasonable; Fig. 5 is oil gas negative area early warning curve, and the curve upper area represents the hazardous location to be needed to report to the police, and the curve lower zone represents the safety zone not to be needed to report to the police.
Five, embodiment:
The present invention is described further below in conjunction with accompanying drawing:
Embodiment 1:
As shown in Figure 1, this oil gas goaf sedimentation method for early warning is as follows:
Step 1: adopt the sky negative area at oil gas and lay settlement sensor, each settlement sensor is connected with server respectively, each settlement sensor was gathered the settling data signal automatically every 12 hours And being transferred to server, server is handled settling data, at first, the recording address settling data
Figure 561856DEST_PATH_IMAGE001
, in the formula
Figure 229467DEST_PATH_IMAGE002
,
Figure 459591DEST_PATH_IMAGE003
,
Figure 316689DEST_PATH_IMAGE004
Be the measurement sequence number, the
Figure 699391DEST_PATH_IMAGE005
Be measurement point position sequence number.
Step 2: the identification of exceptional value and correction, because the measurement data in oil gas goaf obtains at open area, measurement result will inevitably be subjected to the influence of external interference and produce exceptional value so, the existence of exceptional value can reduce the goaf precision of prediction, even warning error can appear, need exceptional value is picked out for this reason, exceptional value is picked out the back and is formed a hole in corresponding position, the existence in hole also can cause the reduction of precision of prediction, thereby need to give the rational substitution value in hole, the exceptional value criterion is as follows:
Formula one:
Figure 350952DEST_PATH_IMAGE034
In the formula
Figure 333951DEST_PATH_IMAGE035
Be exceptional value identification coefficient, general value is 3,
Figure 729161DEST_PATH_IMAGE036
Be
Figure 464904DEST_PATH_IMAGE005
The mean value of individual measurement point sedimentation;
If formula one is false, so
Figure 552946DEST_PATH_IMAGE006
It or not exceptional value; If formula one is set up, so
Figure 757662DEST_PATH_IMAGE006
Be exceptional value, need give A suitable modified value, the adjacent difference ratio of NDGM (1,1) model
Figure 59996DEST_PATH_IMAGE007
Be constant, so can calculate according to adjacent difference Modified value
Figure 745373DEST_PATH_IMAGE008
, its expression formula is as follows:
Formula two:
Figure 934914DEST_PATH_IMAGE009
In the formula,
Figure 458300DEST_PATH_IMAGE010
Be
Figure 560248DEST_PATH_IMAGE005
The adjacent difference ratio of sequence is measured in individual measuring position;
Step 3: strengthen the slickness of settling data, right
Figure 801873DEST_PATH_IMAGE001
Carry out pre-service one by one, generate
Figure 30992DEST_PATH_IMAGE006
, its expression formula is as follows:
Formula three:
Figure 408883DEST_PATH_IMAGE011
Step 4: for strengthening the adaptability of oil gas goaf settling data modeling, will
Figure 947312DEST_PATH_IMAGE006
Carry out one-accumulate (1-AGO) and form sequence
Figure 410654DEST_PATH_IMAGE012
, its expression formula is as follows:
Formula four:
Figure 941999DEST_PATH_IMAGE046
Step 5: right
Figure 439976DEST_PATH_IMAGE012
Adopt NDGM (1,1) model to predict the predicted value that obtains the oil gas goaf
Figure 883727DEST_PATH_IMAGE013
, its expression formula is as follows:
Formula five:
Figure 788360DEST_PATH_IMAGE014
In the formula,
Figure 670866DEST_PATH_IMAGE015
,
Figure 961033DEST_PATH_IMAGE016
, With Be the parameter of NDGM (1,1) model, its expression formula is as follows:
Formula six:
Figure 949083DEST_PATH_IMAGE037
Formula seven:
In the formula, matrix
Figure 207206DEST_PATH_IMAGE040
Be neighbour's matrix,
Figure 820852DEST_PATH_IMAGE041
Be grey background matrix, its expression formula is respectively:
Formula eight:
Figure 45160DEST_PATH_IMAGE042
,
Formula nine:
Figure 309920DEST_PATH_IMAGE043
,
Step 6: counter adding up obtained the predicted value of oil gas negative area with reflective sliding the variation
Figure 594270DEST_PATH_IMAGE019
Figure 928169DEST_PATH_IMAGE013
Be that original predicted data strengthens smooth ratio and the later predicted value that adds up for 1 time through 1 time
Figure 956167DEST_PATH_IMAGE047
, for obtaining the predicted value of oil gas negative area , it is right to need Carry out inverse transformation, its expression formula is as follows:
Formula ten:
Figure 118924DEST_PATH_IMAGE048
Figure 685034DEST_PATH_IMAGE022
Formula 11:
Figure 658807DEST_PATH_IMAGE023
Step 7: adopt the BP neural network method to determine early warning value according to quality, excess surface water situation, underground oil and gas total production, gas production and the petroleum production of the quantity of sampled point, geologic structure, above ground structure The early warning value that adopts the BP neural network method to determine
Figure 900935DEST_PATH_IMAGE025
Acquisition methods be:
With quality d, excess surface water situation water, underground oil and gas total production p, gas production q and the petroleum production t of sampled point number number, geologic structure mv, above ground structure, as the parameter of BP neural network input layer;
The middle layer of BP neural network is hidden layer, and adopts 8 neurons;
With early warning value
Figure 745525DEST_PATH_IMAGE025
Output parameter as BP neural network output layer; Obtain early warning value through training
Figure 542580DEST_PATH_IMAGE025
The training process of described BP neural network is: set up sample set A=(a 1, a 2..., a n), a wherein i=(number i , mv i , d i , water i , p i , q i , t i ,
Figure 89099DEST_PATH_IMAGE044
), in the formula
Adopt sample set A to train, obtain input layer to hidden layer weight matrix w1 IjArrive output layer weight matrix w2 with hidden layer Ij, w1 wherein IjMiddle i=8 is the hidden neuron number, and j=7 is the number of input layer parameter, w2 IjIn wherein i=1 be output layer neuron number, j=8 is the number of input layer parameter, and learning and memory, obtains early warning value
Figure 194644DEST_PATH_IMAGE025
Step 8: set up oil gas goaf sedimentation Early-warning Model.When the goaf sedimentation reach surpass certain critical value after, just there is the danger of subsiding the negative area, this critical value
Figure 100283DEST_PATH_IMAGE025
Be called early warning value.If the settling amount of certain point satisfies following formula, then need to send early warning:
Formula 12:
Figure 196415DEST_PATH_IMAGE049
,
Certain point in oil gas goaf satisfies formula 13 and just should send warning in theory, but because the sedimentation in oil gas goaf is to obtain in open relatively bad border, certain a bit is subjected to extraneous interference probably and should report to the police and produce warning, need to rely on the measurement data of whole measured zone to judge whether to report to the police for this reason, thereby set up following oil gas goaf sedimentation Early-warning Model:
Formula 13:
Figure 596435DEST_PATH_IMAGE026
,
In the formula,
Figure 727202DEST_PATH_IMAGE027
It is measurement point iThe measurement data predicted value
Figure 803742DEST_PATH_IMAGE028
With Between related coefficient, its expression formula is as follows:
Formula 14:
Figure 823837DEST_PATH_IMAGE050
In the formula
Figure 809110DEST_PATH_IMAGE031
Expression With
Figure 861697DEST_PATH_IMAGE029
Between minor increment,
Figure 380982DEST_PATH_IMAGE028
With
Figure 861642DEST_PATH_IMAGE029
Between ultimate range, With
Figure 198132DEST_PATH_IMAGE029
Between minor increment, It is coefficient of autocorrelation.
Embodiment two:
Below in conjunction with Fig. 1, Fig. 2, Fig. 3, Fig. 4, Fig. 5 present embodiment is described, this oil gas goaf sedimentation method for early warning is as follows:
Adopt the sky negative area at oil gas and lay settlement sensor, each settlement sensor is connected with server respectively, each settlement sensor was gathered the settling data signal automatically every 12 hours And being transferred to server, server is handled settling data, at first, the recording address settling data
Figure 141359DEST_PATH_IMAGE001
, in the formula
Figure 536568DEST_PATH_IMAGE002
,
Figure 537891DEST_PATH_IMAGE003
,
Figure 360353DEST_PATH_IMAGE004
Be the measurement sequence number, the
Figure 565070DEST_PATH_IMAGE005
Be measurement point position sequence number; The record measurement data.
Then, the measurement data of sensor is carried out automatic identification and the correction of exceptional value.Continuous 4 days measurement data data1 of certain position transducer, the data data2 that picks out after the exceptional value are as shown in table 1, and unit is micron.
Table 1
1 2 3 4 5 6 7 8
data1 0.14 0.16 2 0.13 0.16 0.12 0.16 0.15
data2 0.14 0.16 0.15 0.13 0.16 0.12 0.16 0.15
As shown in Table 1, the 3rd measurement data of measurement data data1 occurs unusual, because its other measured values of numerical value 2 substantial deviations, adopting the modified value of the exceptional value of the present invention's acquisition is 0.15, does not impact predicting the outcome.
Then, measurement data is carried out smooth treatment and obtain measurement data
Figure 763970DEST_PATH_IMAGE006
Then, strengthen the slickness of settling data, right
Figure 121264DEST_PATH_IMAGE001
Carry out pre-service one by one, generate
Figure 114628DEST_PATH_IMAGE006
Then, for strengthening the adaptability of oil gas goaf settling data modeling, will
Figure 806640DEST_PATH_IMAGE006
Carry out one-accumulate (1-AGO) and form sequence
Figure 996182DEST_PATH_IMAGE012
Then, right
Figure 519567DEST_PATH_IMAGE012
Adopt NDGM (1,1) model to predict the predicted value that obtains the oil gas goaf
Figure 309931DEST_PATH_IMAGE013
Then,
Figure 551557DEST_PATH_IMAGE013
Be that original predicted data strengthens smooth ratio and the later predicted value that adds up for 1 time through 1 time , need be right for obtaining raw data
Figure 407834DEST_PATH_IMAGE013
Carry out inverse transformation;
Then, quality, excess surface water situation, underground oil and gas total production, gas production and the petroleum production of the quantity of sampling BP neural network method sampled point, geologic structure, above ground structure are determined early warning value
Figure 195531DEST_PATH_IMAGE025
At last, set up based on the related oil gas goaf sedimentation Early-warning Model of ash.

Claims (5)

1. oil gas goaf sedimentation method for early warning, it is characterized in that: this oil gas goaf sedimentation method for early warning is as follows:
Step 1: adopt the sky negative area at oil gas and lay settlement sensor, settlement sensor is connected with server, settlement sensor is gathered the settling data signal and is transferred to server, and server is handled settling data, at first, and the recording address settling data
Figure 456494DEST_PATH_IMAGE001
, in the formula
Figure 373634DEST_PATH_IMAGE002
,
Figure 91054DEST_PATH_IMAGE003
,
Figure 486264DEST_PATH_IMAGE004
Be the measurement sequence number, the
Figure 487586DEST_PATH_IMAGE005
Be measurement point position sequence number;
Step 2: the measurement data of settlement sensor is carried out identification and the correction of exceptional value, identify exceptional value
Figure 310049DEST_PATH_IMAGE006
, the adjacent difference ratio of NDGM (1,1) model
Figure 514765DEST_PATH_IMAGE007
Be constant, so can calculate according to adjacent difference
Figure 713665DEST_PATH_IMAGE006
Modified value
Figure 70960DEST_PATH_IMAGE008
, its expression formula is as follows:
；
In the formula,
Figure 756336DEST_PATH_IMAGE010
Be
Figure 493348DEST_PATH_IMAGE005
The adjacent difference ratio of sequence is measured in individual measuring position;
Step 3: strengthen the slickness of settling data, right
Figure 203684DEST_PATH_IMAGE001
Carry out pre-service one by one, generate
Figure 633528DEST_PATH_IMAGE006
, its expression formula is:
Figure 812837DEST_PATH_IMAGE011
；
Step 4: for strengthening the adaptability of oil gas goaf settling data modeling, will
Figure 773446DEST_PATH_IMAGE006
Carry out one-accumulate (1-AGO) and form sequence
Figure 416917DEST_PATH_IMAGE012
Step 5: right
Figure 689767DEST_PATH_IMAGE012
Adopt NDGM (1,1) model to predict the predicted value that obtains the oil gas goaf
Figure 418688DEST_PATH_IMAGE013
, its expression formula is:
Figure 684453DEST_PATH_IMAGE014
In the formula,
Figure 182431DEST_PATH_IMAGE015
, ,
Figure 842399DEST_PATH_IMAGE017
With It is the parameter of NDGM (1,1) model;
Step 6: counter adding up obtained the predicted value of oil gas negative area with reflective sliding the variation
Figure 765804DEST_PATH_IMAGE019
,
Figure 646035DEST_PATH_IMAGE013
Be that original predicted data strengthens smooth ratio and the later predicted value that adds up for 1 time through 1 time
Figure 83970DEST_PATH_IMAGE020
, for obtaining the predicted value of oil gas negative area
Figure 691538DEST_PATH_IMAGE019
, it is right to need
Figure 164107DEST_PATH_IMAGE013
Carry out inverse transformation, its expression formula is as follows:
Figure 140471DEST_PATH_IMAGE022
Figure 364779DEST_PATH_IMAGE023
Figure 111761DEST_PATH_IMAGE024
；
Step 7: adopt the BP neural network method to determine early warning value according to quantity, geologic structure, the quality of above ground structure, the excess surface water situation of sampled point
Figure 396112DEST_PATH_IMAGE025
Step 8: set up based on the related oil gas goaf sedimentation Early-warning Model of ash, when the goaf sedimentation reach surpass certain critical value after, just there is the danger of subsiding the negative area, this critical value
Figure 746322DEST_PATH_IMAGE025
Be called early warning value, if the settling amount of certain point surpasses critical value, then need to send early warning.
2. oil gas goaf sedimentation method for early warning according to claim 1, it is characterized in that: for preventing that described to send early warning be possible be subjected to extraneous interference should not report to the police and produce warning, need to rely on the measurement data of whole measured zone to judge whether to report to the police, thereby set up following oil gas goaf sedimentation Early-warning Model:
Figure 774321DEST_PATH_IMAGE026
，
In the formula,
Figure 142854DEST_PATH_IMAGE027
It is measurement point iThe measurement data predicted value
Figure 332527DEST_PATH_IMAGE028
With
Figure 170033DEST_PATH_IMAGE029
Between related coefficient, its expression formula is as follows:
In the formula
Figure 726227DEST_PATH_IMAGE032
Expression With
Figure 411604DEST_PATH_IMAGE029
Between minor increment,
Figure 781405DEST_PATH_IMAGE033
Figure 858952DEST_PATH_IMAGE028
With
Figure 656006DEST_PATH_IMAGE029
Between ultimate range,
Figure 202525DEST_PATH_IMAGE028
With
Figure 376018DEST_PATH_IMAGE029
Between minor increment,
Figure 72185DEST_PATH_IMAGE034
It is coefficient of autocorrelation.
3. oil gas goaf sedimentation method for early warning according to claim 2, it is characterized in that: the recognition methods that described measurement data to settlement sensor is carried out exceptional value is that the exceptional value criterion is as follows:
Figure 40141DEST_PATH_IMAGE035
In the formula
Figure 73956DEST_PATH_IMAGE036
Be exceptional value identification coefficient, general value is 3,
Figure 785560DEST_PATH_IMAGE037
Be
Figure 837698DEST_PATH_IMAGE005
The mean value of individual measurement point sedimentation;
If following formula is false, so
Figure 976556DEST_PATH_IMAGE038
It or not exceptional value; If following formula is set up, so It is exceptional value.
4. oil gas goaf sedimentation method for early warning according to claim 3 is characterized in that: in the described step 5
Figure 747383DEST_PATH_IMAGE015
,
Figure 732656DEST_PATH_IMAGE016
,
Figure 730830DEST_PATH_IMAGE017
With
Figure 801555DEST_PATH_IMAGE018
Expression formula as follows:
Figure 792644DEST_PATH_IMAGE040
In the formula, matrix
Figure 300035DEST_PATH_IMAGE042
Be neighbour's matrix,
Figure 592476DEST_PATH_IMAGE043
Be grey background matrix, its expression formula is respectively:
Figure 387257DEST_PATH_IMAGE044
，
Figure 81543DEST_PATH_IMAGE045
。
5. oil gas goaf sedimentation method for early warning according to claim 4, it is characterized in that: described employing BP neural network method is determined early warning value Method be:
With quality d, excess surface water situation water, underground oil and gas total production p, gas production q and the petroleum production t of sampled point number number, geologic structure mv, above ground structure, as the parameter of BP neural network input layer;
The middle layer of BP neural network is hidden layer, and adopts 8 neurons;
With early warning value
Figure 487344DEST_PATH_IMAGE025
As the output parameter of BP neural network output layer, set up sample set A=(a 1, a 2..., a n), a wherein i=(number i , mv i , d i , water i , p i , q i , t i ,
Figure 820236DEST_PATH_IMAGE046
), in the formula
Adopt sample set A to train, obtain input layer to hidden layer weight matrix w1 IjArrive output layer weight matrix w2 with hidden layer Ij, w1 wherein IjMiddle i=8 is the hidden neuron number, and j=7 is the number of input layer parameter, w2 IjIn wherein i=1 be output layer neuron number, j=8 is the number of input layer parameter, and learning and memory, obtains early warning value
Figure 644022DEST_PATH_IMAGE025
CN201310145611.0A 2013-04-25 2013-04-25 Early warning method for settlement of oil gas gob area Active CN103196424B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310145611.0A CN103196424B (en) 2013-04-25 2013-04-25 Early warning method for settlement of oil gas gob area

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310145611.0A CN103196424B (en) 2013-04-25 2013-04-25 Early warning method for settlement of oil gas gob area

Publications (2)

Publication Number Publication Date
CN103196424A true CN103196424A (en) 2013-07-10
CN103196424B CN103196424B (en) 2015-02-18

Family

ID=48719129

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310145611.0A Active CN103196424B (en) 2013-04-25 2013-04-25 Early warning method for settlement of oil gas gob area

Country Status (1)

Country Link
CN (1) CN103196424B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107767625A (en) * 2017-10-23 2018-03-06 山西省交通科学研究院 A kind of highway goaf subgrade stability monitoring and pre-warning system and method
CN110348568A (en) * 2019-07-16 2019-10-18 山东科技大学 A kind of deep Mined-Out Areas method suitable for strong electromagnetic area

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000048159A2 (en) * 1999-02-01 2000-08-17 Honeywell International Inc. Apparatus, methods and computer program product for the generation and use of a runway field clearance floor envelope about a selected runway
KR20100041996A (en) * 2008-10-15 2010-04-23 박종진 Apparatus and method for measuring ground settlement of whole section
CN102778893A (en) * 2012-08-07 2012-11-14 中铁二十三局集团有限公司 Precise locating detecting method for truss girder

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000048159A2 (en) * 1999-02-01 2000-08-17 Honeywell International Inc. Apparatus, methods and computer program product for the generation and use of a runway field clearance floor envelope about a selected runway
KR20100041996A (en) * 2008-10-15 2010-04-23 박종진 Apparatus and method for measuring ground settlement of whole section
CN102778893A (en) * 2012-08-07 2012-11-14 中铁二十三局集团有限公司 Precise locating detecting method for truss girder

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘春玲等: "基于ArcGIS Engine的溢油环境预警软件(OSEA* )开发", 《环境科学与管理》 *
龚士良等: "地面沉降自动化测控与预警预报系统", 《计算机测量与控制》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107767625A (en) * 2017-10-23 2018-03-06 山西省交通科学研究院 A kind of highway goaf subgrade stability monitoring and pre-warning system and method
CN110348568A (en) * 2019-07-16 2019-10-18 山东科技大学 A kind of deep Mined-Out Areas method suitable for strong electromagnetic area
CN110348568B (en) * 2019-07-16 2021-02-23 山东科技大学 Deep goaf detection method suitable for strong electromagnetic interference area

Also Published As

Publication number Publication date
CN103196424B (en) 2015-02-18

Similar Documents

Publication Publication Date Title
CN104730596B (en) A kind of discrete fractures modeling method based on multiple dimensioned factor constraints
CN106205060B (en) For outdoor pit Tailings Dam slope and land slide early alarming and forecasting method
CN112100727B (en) Early warning prevention and control method for water burst of water-rich tunnel under influence of fault fracture zone
CN115311821B (en) Geological disaster intelligent monitoring cloud platform based on digital twin technology
CN113570226A (en) Method for evaluating occurrence probability grade of tunnel water inrush disaster in fault fracture zone
CN116597616B (en) Intelligent monitoring and early warning system for geological disasters in mining area
CN110298107B (en) Working face impact risk evaluation method based on incremental stacking
CN115270527B (en) Real-time assessment method, equipment and storage medium for road collapse risk
CN111022119B (en) Shield tunnel structure anti-seismic safety evaluation method based on pattern recognition
CN103810524A (en) Method for predicting ground subsidence in underground metro construction process
Zhang et al. Subway tunnel damage detection based on in-service train dynamic response, variational mode decomposition, convolutional neural networks and long short-term memory
CN112308306A (en) Multi-mode input coal and gas outburst risk prediction method
CN112131748A (en) Deformation prediction method and system for composite stratum in urban tunnel construction
CN103244187A (en) Early-warning comprehensive index system of coal and gas outburst inoculation and development process
CN104899928A (en) Three-directional geology modeling method based on sparse borehole points
CN103196424B (en) Early warning method for settlement of oil gas gob area
CN107784437B (en) Stress concentration-based coal roadway driving face outburst danger identification method
KR20200058258A (en) System and method for predicting ground layer information, and a recording medium having computer readable program for executing the method
CN108776854B (en) Method for evaluating stability precision of large surface mine side slope
CN111026790A (en) Structure safety assessment and forecasting method based on data mining
CN103953024B (en) Foundation ditch automatic monitoring disorder data recognition method
Gaida et al. Development and application of a reliability-based approach to slope design acceptance criteria at Bingham Canyon Mine
CN110188439A (en) The subway work ground settlement method for early warning of case-based reasioning and system dynamics
CN112285773B (en) Quantitative characterization method for binary structure of buried fault
Zhou et al. Developing a cloud evidence method for dynamic early warning of tunnel construction safety risk in undersea environment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20201125

Address after: 25-1 Wuzhuang village, Banqiao neighborhood committee, weizhai Town, Linquan County, Fuyang City, Anhui Province

Patentee after: Linquan lingcui Technical Service Co., Ltd

Address before: 163319 No. 199, development road, hi tech Development Zone, Heilongjiang, Daqing

Patentee before: NORTHEAST PETROLEUM University