CN105952377A - Method for controlling path of coal mine underground directional drilling - Google Patents

Abstract

The invention relates to a method for controlling the path of coal mine underground directional drilling. The method comprises following steps: step 1. determining a predetermined path; step 2. determining input and output, establishing a prediction model based on two output least squares support vector machine, and predicting next testing point path parameters by means of practical drilling path; step 3. comparing the path predicted by practical drilling path with the predetermine path, and offering the optimal tool face angle through the minimum offset, and controlling the following practical path drilling and performing correction to realize effective control on drilling path. Therefore, the provided method has easy operation, reliable results and engineering use value which can be a real time guidance for on-site drilling construction.

Description

Underground coal mine directional drilling method for controlling trajectory

Technical field

The present invention relates to the technical field of geological prospecting, particularly relate to a kind of underground coal mine directional drilling TRAJECTORY CONTROL side Method.

Background technology

Safety of Coal Mine Production proposes the highest wanting to the drilling efficiency during coal mine underground construction, drilling track etc. Ask.Technology of Directional Drilling possesses that rate of penetration is fast, orientation accuracy is high, the advantage such as " a hole multiple-limb ", it has also become high-yield and high-efficiency coal The effective technological means of ore deposit underground boring construction, is widely used in the field such as control of coalmine gas, geological prospecting.

During directional drilling, drilling measuring technology is the most ripe, but the control to drilling track also lacks row Effective means.Owing to drilling track is easily affected by factors such as formation geology condition, coal seam distribution, drilling rod gravity, tool-face To angle, inclination angle, azimuthal effect tendency are relatively easily obtained, but between them, the relation of quantification is also difficult to determine.Existing The drilling track of row controls to be mostly to carry out by artificial experience, has the strongest empirical and uncertain.Work progress The lacking experience of middle technical staff can frequently result in drilling trajectory deviation desired trajectory, it is impossible to ensure construction drill in coal seam Length is passed through, and causes gas pumping poor effect, causes drilling tool relatively big at hole internal friction resistance, and then affects the safety of drilling construction With quality, the accident such as cave in even causing burying, hole, cause great economic loss.

To this end, the designer of the present invention is because drawbacks described above, by concentrating on studies and designing, comprehensively it is engaged in for many years for a long time The experience of related industry and achievement, research design goes out a kind of underground coal mine directional drilling method for controlling trajectory, to overcome above-mentioned lacking Fall into.

Summary of the invention

It is an object of the invention to provide a kind of underground coal mine directional drilling method for controlling trajectory, its step is simple, can be real Existing drilling trajectory approaches desired trajectory, it is ensured that construction drill is the longest in coal seam to be passed through, and improves gas pumping efficiency.

For solving the problems referred to above, the invention discloses a kind of underground coal mine directional drilling method for controlling trajectory, comprise as follows Step:

Step one: determine desired trajectory；

Step 2: determine input and output, builds forecast model based on two output least square method supporting vector machines, utilizes The next measuring point trajectory parameters of drilling trajectory prediction；

Step 3: contrast the track and desired trajectory predicted by drilling trajectory, provided the work of optimum by minimum deflection Tool, towards angle, controls follow-up drilling trajectory and is modified, it is achieved control effectively drilling track.

Wherein: step 2 includes following sub-step:

1) determine input and the output of forecast model, input the hole inclination angle theta for just completing n boring measuring point record_i, orientation Angle α_i, tools for angle ω_i, i=1,2 ..., n, it is output as the inclination angle of next measuring pointAzimuth

2) structure forecast model based on two output least square method supporting vector machines, as shown in formula (1),

$min\frac{1}{2}\underset{k=1}{\overset{2}{\Σ}}\left|\right|w^{\left(k\right)}|{|}^2+\frac{C}{2}\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\left({\mathrm{\ξ}}_i^{\left(k\right)}\right)}^2$

$\begin{array}{cc}s.t.& B_i^{\left(k\right)}={\left(w^{\left(k\right)}\right)}^T\mathrm{\φ}\left(A_i\right)+b^{\left(k\right)}+{\mathrm{\ξ}}_i^{\left(k\right)}\end{array}---\left(1\right)$

Wherein,It is optimization aim,It is Optimal conditions, i=1,2 ..., n represents the dimension of input parameter, k=1, and 2 represent two outputs, and the normal vector of w hyperplane, b is phase The side-play amount answered, C is penalty factor, and ξ represents error between predictive value and actual value, and A represents input vector, B with A is corresponding Label value, φ (A_i) represent nonlinear mapping function, nonlinear sample set is mapped in high-dimensional feature space and is converted into line Sex chromosome mosaicism solves,

Then introduce Lagrange multiplier γ, it is thus achieved that Lagrangian, as shown in formula (2), optimal conditions is added In optimization aim

$L=\frac{1}{2}\underset{k=1}{\overset{2}{\Σ}}\left|\right|w^{\left(k\right)}|{|}^2+\frac{C}{2}\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\left({\mathrm{\ξ}}_i^{\left(k\right)}\right)}^2-\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\mathrm{\γ}}_i^{\left(k\right)}\[{\left(w^{\left(k\right)}\right)}^T\mathrm{\φ}\left(A_i\right)+b+{\mathrm{\ξ}}_i^{\left(k\right)}-B_i^{\left(k\right)}\]---\left(2\right)$

Utilizing LagrangianL derivation at saddle point is zero, can solve formula (2), followed by formula (3) The inclination angle of the next measuring point of predictionAzimuth

$f^{\left(k\right)}\left(A\right)=\underset{i=1}{\overset{n}{\Σ}}{\mathrm{\γ}}_i^{\left(k\right)}{\mathrm{\φ}}^T\left(A\right)\mathrm{\φ}\left(A_i\right)+b^{\left(k\right)}---\left(3\right)$

Wherein

3) by the inclination angle of predictionAnd azimuthIt is converted into 3 d space coordinate: horizontal displacementPosition, left and right MoveUpper and lower displacement

$X_{n+1}^P=X_n+{\mathrm{\ΔL}}_{n+1}\×cos\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)*cos(\frac{{\mathrm{\α}}_n+{\mathrm{\α}}_{n+1}^P}{2}-\mathrm{\λ})$

$Y_{n+1}^P=Y_n+{\mathrm{\ΔL}}_{n+1}\×cos\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)*sin(\frac{{\mathrm{\α}}_n+{\mathrm{\α}}_{n+1}^P}{2}-\mathrm{\λ})$

$Z_{n+1}^P=Z_n+{\mathrm{\ΔL}}_{n+1}\×sin\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)$

Wherein, X_n、Y_n、Z_nRepresent the three-dimensional coordinate of n measuring point, Δ L_n+1Represent and increase from n measuring point to the drilling depth of n+1 measuring point Amount, λ is predetermined party parallactic angle.

Wherein: step 3 includes following sub-step:

1) predictive value of calculating drilling track: horizontal displacementLeft and right displacementUpper and lower displacementWith boring rail The predetermined value of mark: horizontal displacementLeft and right displacementUpper and lower displacementBetween deviation

$d=\sqrt{{(X_{n+1}^P-X_{n+1}^S)}^2+{(Y_{n+1}^P-Y_{n+1}^S)}^2+{(Z_{n+1}^P-Z_{n+1}^S)}^2}$

2) select different tools for angle ω, repeat the above steps, obtain a different set of deviation d, obtain deviation d with Relation curve between tools for angle ω, according to minimum deflection d_minObtain the tools for angle ω of optimum_opt, it is achieved to boring Track control effectively.

By said structure, the underground coal mine directional drilling method for controlling trajectory of the present invention has the effect that

1, utilize existing real brill data, build perfect drilling tracks based on two output least square method supporting vector machines pre- Survey mechanism, next step drilling track of Accurate Prediction；

2, the tools for angle of optimum is given, it is achieved drilling track is automatically controlled, changes over and utilize artificial warp Test the present situation controlling track, reduce construction risk；

3, simple to operate, reliable results, can real-time instruction scene drilling construction, there is good engineering use value.

The detailed content of the present invention can be obtained by explanation described later and institute's accompanying drawing.

Accompanying drawing explanation

Fig. 1 is the schematic three dimensional views of the drilling track Controlling model of the present invention.

Fig. 2 is drilling trajectory in the embodiment of the present invention, desired trajectory and TRAJECTORY CONTROL plane graph, and wherein Fig. 2 (a) is XOY Drilling trajectory, desired trajectory and TRAJECTORY CONTROL in plane；Fig. 2 (b) is the drilling trajectory in XOZ plane, desired trajectory and rail Mark controls.

Fig. 3 is optimum tool-face at 99m in the embodiment of the present invention.

Detailed description of the invention

See Fig. 1 to Fig. 3, it is shown that the underground coal mine directional drilling method for controlling trajectory of the present invention.

As it is shown in figure 1, set up 3-D walls and floor OX, OY, OZ, in figureFor desired trajectory,For real brill Track, dotted lineFor the extended line of drilling trajectory,For prediction locus,ForIn XOY plane Projection, Parallel with OX, ∠ A_iA_i-1D isInclination angle theta, ∠ B^XB_i-1B_iIt isAzimuth angle alpha,Angle of bend β for screw motor.

With reference to shown in Fig. 2, solid dot setting-out is desired trajectory, and open triangles frame setting-out is drilling trajectory, open circle setting-out For TRAJECTORY CONTROL.

With reference to shown in Fig. 3, abscissa is tools for angle, and unit is degree；Vertical coordinate is that drilling track controls and pre-orbit determination Deviation between mark, unit m, the some abscissa marked out is optimum tools for angle, and vertical coordinate is minimum deflection.

The control method of the present invention mainly comprises the steps of:

Step one: determine desired trajectory；

Step 2: determine input and output, builds forecast model based on two output least square method supporting vector machines, utilizes The next measuring point trajectory parameters of drilling trajectory prediction；

Step 3: contrast the track and desired trajectory predicted by drilling trajectory, provided the work of optimum by minimum deflection Tool, towards angle, controls follow-up drilling trajectory and is modified, it is achieved control effectively drilling track.

Wherein, build the continuous bend model of angle of bend based on helicoid hydraulic motor and tools for angle, utilize real brill rail In the step 2 of the next measuring point track of mark prediction, it may include following sub-step:

1) determine input and output, input the hole inclination angle theta for just completing n boring measuring point record_i, azimuth angle alpha_i, tool-face To angle ω_i, i=1,2 ..., n, it is output as the inclination angle of next measuring pointAzimuth

2) structure forecast model based on two output least square method supporting vector machines, as shown in formula (1).

$min\frac{1}{2}\underset{k=1}{\overset{2}{\Σ}}\left|\right|w^{\left(k\right)}|{|}^2+\frac{C}{2}\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\left({\mathrm{\ξ}}_i^{\left(k\right)}\right)}^2$

$\begin{array}{cc}s.t.& B_i^{\left(k\right)}={\left(w^{\left(k\right)}\right)}^T\mathrm{\φ}\left(A_i\right)+b^{\left(k\right)}+{\mathrm{\ξ}}_i^{\left(k\right)}\end{array}---\left(1\right)$

Wherein,It is optimization aim,It is Optimal conditions, i=1,2 ..., n represents the dimension of input parameter.K=1,2 represents two outputs.The normal vector of w hyperplane, b is phase The side-play amount answered.C is penalty factor, and ξ represents error between predictive value and actual value.A represents input vector, B with A is corresponding Label value.φ(A_i) represent nonlinear mapping function, nonlinear sample set is mapped in high-dimensional feature space and is converted into line Sex chromosome mosaicism solves.

Then introduce Lagrange multiplier γ, it is thus achieved that Lagrangian, as shown in formula (2), optimal conditions is added In optimization aim

$L=\frac{1}{2}\underset{k=1}{\overset{2}{\Σ}}\left|\right|w^{\left(k\right)}|{|}^2+\frac{C}{2}\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\left({\mathrm{\ξ}}_i^{\left(k\right)}\right)}^2-\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\mathrm{\γ}}_i^{\left(k\right)}\[{\left(w^{\left(k\right)}\right)}^T\mathrm{\φ}\left(A_i\right)+b+{\mathrm{\ξ}}_i^{\left(k\right)}-B_i^{\left(k\right)}\]---\left(2\right)$

Utilizing LagrangianL derivation at saddle point is zero, can solve formula (2).Followed by formula (3) The inclination angle of the next measuring point of predictionAzimuth

$f^{\left(k\right)}\left(A\right)=\underset{i=1}{\overset{n}{\Σ}}{\mathrm{\γ}}_i^{\left(k\right)}{\mathrm{\φ}}^T\left(A\right)\mathrm{\φ}\left(A_i\right)+b^{\left(k\right)}---\left(3\right)$

Wherein

3) by the inclination angle of predictionAnd azimuthIt is converted into 3 d space coordinate: horizontal displacementPosition, left and right MoveUpper and lower displacement

$X_{n+1}^P=X_n+{\mathrm{\ΔL}}_{n+1}\×cos\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)*cos(\frac{{\mathrm{\α}}_n+{\mathrm{\α}}_{n+1}^P}{2}-\mathrm{\λ})$

$Y_{n+1}^P=Y_n+{\mathrm{\ΔL}}_{n+1}\×cos\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)*sin(\frac{{\mathrm{\α}}_n+{\mathrm{\α}}_{n+1}^P}{2}-\mathrm{\λ})$

$Z_{n+1}^P=Z_n+{\mathrm{\ΔL}}_{n+1}\×sin\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)$

Wherein, X_n、Y_n、Z_nRepresent the three-dimensional coordinate of n measuring point, Δ L_n+1Represent and increase from n measuring point to the drilling depth of n+1 measuring point Amount, λ is predetermined party parallactic angle.

Contrasted with desired trajectory by prediction locus, provide the tools for angle of optimum according to minimum deflection, it is achieved to brill In the step 2 that hole track control effectively, may particularly include following sub-step:

1) predictive value of calculating drilling track: horizontal displacementLeft and right displacementUpper and lower displacementWith boring rail The predetermined value of mark: horizontal displacementLeft and right displacementUpper and lower displacementBetween deviation

$d=\sqrt{{(X_{n+1}^P-X_{n+1}^S)}^2+{(Y_{n+1}^P-Y_{n+1}^S)}^2+{(Z_{n+1}^P-Z_{n+1}^S)}^2}$

2) select different tools for angle ω, repeat the above steps, obtain a different set of deviation d, obtain deviation d with Relation curve between tools for angle ω, according to minimum deflection d_minObtain the tools for angle ω of optimum_opt, it is achieved to boring Track control effectively.

Thus, the present invention is determined by input and the output of prediction, and builds based on two output least squares support vectors The forecast model of machine, utilizes the next measuring point trajectory parameters of drilling trajectory prediction, then by prediction locus and desired trajectory pair Ratio, provides the tools for angle of optimum, it is achieved control effectively drilling track according to minimum deflection.The method operation letter Single, reliable results, can real-time instruction scene drilling construction, there is engineering use value.

The said method of the present invention is applied in concrete operations:

Specific embodiment 1

In certain coal mine underground directional drilling work progress, predetermined 200m, construction requirement is constructed according to desired trajectory, and is guaranteed Boring is positioned at predetermined party parallactic angle λ=184.26 ° in coal seam, screw motor angle of bend β=1.25 °.Send out when drilling depth 57m Existing drilling trajectory is 2.21m with desired trajectory deviation, now proceeds by TRAJECTORY CONTROL, and drilling trajectory is for manually by virtue of experience to enter Row controls, and the deviation between drilling trajectory and desired trajectory is controlled within 0.5m at 135m through 14 steps, and by this Bright method, it is only necessary to 8 steps, can control the deviation between drilling track and desired trajectory within 0.5m when 99m.Track Control result as in figure 2 it is shown, corresponding error is as shown in table 1.It can also be seen that utilize artificial experience track at 93m from Fig. 2 During 105m, although in XOY plane, drilling trajectory is close to desired trajectory, but drilling trajectory and pre-orbit determination in XOZ plane Between mark, deviation is but increasing.The inventive method can process this problem well, allows drilling track approach pre-in space Fixed track.Fig. 3 is the optimal tools for angle ω in the output of 99m TRAJECTORY CONTROL_opt=39 °, corresponding minimum deflection d_min= 0.4038m。

Deviation between table 1 drilling trajectory and desired trajectory, TRAJECTORY CONTROL and desired trajectory

By above-described embodiment, the method for the present invention have also been obtained sufficient effect in actual applications and embodies, its Trajector deviation is the least, it is achieved that drilling trajectory and the actually active control of desired trajectory.

It is readily apparent that above description and record are only to illustrate rather than in order to limit in disclosure of the invention Hold, apply or use.Although describing and be described in the drawings embodiment the most in an embodiment, but the present invention being not intended to By accompanying drawing example and be described as in an embodiment it is now recognized that optimal mode with implement the teachings of the present invention particular case Son, the scope of the present invention will include any embodiment falling into description and appended claims above.

Claims (3)

1. a underground coal mine directional drilling method for controlling trajectory, comprises the steps of:

Step one: determine desired trajectory；

Step 2: determine input and output, builds forecast model based on two output least square method supporting vector machines, utilizes real boring Trajectory predictions next one measuring point trajectory parameters；

Step 3: contrast the track and desired trajectory predicted by drilling trajectory, provided the tool-face of optimum by minimum deflection To angle, control follow-up drilling trajectory and be modified, it is achieved drilling track is control effectively.

2. underground coal mine directional drilling method for controlling trajectory as claimed in claim 1, it is characterised in that: step 2 includes as follows Sub-step:

1) determine input and the output of forecast model, input the hole inclination angle theta for just completing n boring measuring point record_i, azimuth angle alpha_i、 Tools for angle ω_i, i=1,2 ..., n, it is output as the inclination angle of next measuring pointAzimuth

2) structure forecast model based on two output least square method supporting vector machines, as shown in formula (1),

$min\frac{1}{2}\underset{k=1}{\overset{2}{\Σ}}\left|\right|w^{\left(k\right)}|{|}^2+\frac{C}{2}\underset{i=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{2}{\Σ}}{\left({\mathrm{\ξ}}_i^{\left(k\right)}\right)}^2$

$\begin{array}{cc}s.t.& B_i^{\left(k\right)}={\left(w^{\left(k\right)}\right)}^T\mathrm{\φ}\left(A_i\right)+b^{\left(k\right)}+{\mathrm{\ξ}}_i^{\left(k\right)}\end{array}---\left(1\right)$

Wherein,It is optimization aim,It is Optimal conditions, i=1,2 ..., n represents the dimension of input parameter, k=1, and 2 represent two outputs, and the normal vector of w hyperplane, b is phase The side-play amount answered, C is penalty factor, and ξ represents error between predictive value and actual value, and A represents input vector, B with A is corresponding Label value, φ (A_i) represent nonlinear mapping function, nonlinear sample set is mapped in high-dimensional feature space and is converted into line Sex chromosome mosaicism solves,

Then introduce Lagrange multiplier γ, it is thus achieved that Lagrangian, as shown in formula (2), optimal conditions is joined excellent Change in target

$L=\frac{1}{2}\underset{k=1}{\overset{2}{\mathrm{\Σ}}}\left|\right|w^{\left(k\right)}|{|}^2+\frac{C}{2}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{k=1}{\overset{2}{\mathrm{\Σ}}}{\left({\mathrm{\ξ}}_i^{\left(k\right)}\right)}^2-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{k=1}{\overset{2}{\mathrm{\Σ}}}{\mathrm{\γ}}_i^{\left(k\right)}\[{\left(w^{\left(k\right)}\right)}^T\mathrm{\φ}\left(A_i\right)+b+{\mathrm{\ξ}}_i^{\left(k\right)}-B_i^{\left(k\right)}\]---\left(2\right)$

Utilizing LagrangianL derivation at saddle point is zero, can solve formula (2), predicts followed by formula (3) The inclination angle of next measuring pointAzimuth

$f^{\left(k\right)}\left(A\right)=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\γ}}_i^{\left(k\right)}{\mathrm{\φ}}^T\left(A\right)\mathrm{\φ}\left(A_i\right)+b^{\left(k\right)}---\left(3\right)$

Wherein

3) by the inclination angle of predictionAnd azimuthIt is converted into 3 d space coordinate: horizontal displacementLeft and right displacement Upper and lower displacement

$X_{n+1}^P=X_n+{\mathrm{\ΔL}}_{n+1}\×cos\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)*cos(\frac{{\mathrm{\α}}_n+{\mathrm{\α}}_{n+1}^P}{2}-\mathrm{\λ})$

$Y_{n+1}^P=Y_n+{\mathrm{\ΔL}}_{n+1}\×cos\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)*sin(\frac{{\mathrm{\α}}_n+{\mathrm{\α}}_{n+1}^P}{2}-\mathrm{\λ})$

$Z_{n+1}^P=Z_n+{\mathrm{\ΔL}}_{n+1}\×sin\left(\frac{{\mathrm{\θ}}_n+{\mathrm{\θ}}_{n+1}^P}{2}\right)$

Wherein, X_n、Y_n、Z_nRepresent the three-dimensional coordinate of n measuring point, Δ L_n+1Represent the drilling depth increment from n measuring point to n+1 measuring point, λ For predetermined party parallactic angle.

3. underground coal mine directional drilling method for controlling trajectory as claimed in claim 1 or 2, it is characterised in that: step 3 includes Following sub-step:

1) predictive value of calculating drilling track: horizontal displacementLeft and right displacementUpper and lower displacementWith drilling track Predetermined value: horizontal displacementLeft and right displacementUpper and lower displacementBetween deviation

$d=\sqrt{{\left(X_{n+1}^P-X_{n+1}^S\right)}^2+{\left(Y_{n+1}^P-Y_{n+1}^S\right)}^2+{\left(Z_{n+1}^P-Z_{n+1}^S\right)}^2}$

2) select different tools for angle ω, repeat the above steps, obtain a different set of deviation d, obtain deviation d and instrument Relation curve between the ω of angle, according to minimum deflection d_minObtain the tools for angle ω of optimum_opt, it is achieved to drilling track It control effectively.