CN107357966A - A kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure - Google Patents
A kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure Download PDFInfo
- Publication number
- CN107357966A CN107357966A CN201710472314.5A CN201710472314A CN107357966A CN 107357966 A CN107357966 A CN 107357966A CN 201710472314 A CN201710472314 A CN 201710472314A CN 107357966 A CN107357966 A CN 107357966A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- sample
- kernel function
- surrounding rock
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure, and it mainly includes the measured data of the influence factor of surrounding rock of actual mining roadway stability inputting computer；Then sample data is transferred, establishes monokaryon function SVM prediction model；With selected global kernel function and karyomerite combination of function into mixed kernel function, mixed kernel function SVM prediction model is established；Recycle particle cluster algorithm to optimize the penalty factor in forecast model, nuclear parameter and monokaryon function coefficients, establish the particle cluster algorithm mixed kernel function SVM prediction model of surrounding rock of actual mining roadway stability；Treat the precision of prediction of forecast sample and reliability is calculated and models mature degree is carried out the step such as to assess.The present invention has simple to operate, methodological science, and surrounding rock of actual mining roadway stability grade forecast result and models mature degree assessment result are accurate, reliable, have the characteristics that the higher goodness of fit with actual conditions.
Description
Technical field
The present invention relates to a kind of surrounding rock of actual mining roadway safety prediction and appraisal procedure, more particularly to a kind of stope drift active workings to enclose
Rock stability prediction and appraisal procedure, belong to technical field of mine safety.
Background technology
Underground coal mine stope drift active workings are Coal Transport, mine ventilation, the important channel of pedestrian's haul, and surrounding rock of actual mining roadway is steady
Qualitatively Accurate Prediction or assessment, be supporting scheme design premise and basis, it is significant concerning safety in production.In recent years,
The research of the Forecasting Methodology of surrounding rock of actual mining roadway stability, have become coal mining safety production technique field one are important
Scientific research task.
In the prior art, the Forecasting Methodology of surrounding rock of actual mining roadway stability has a variety of, mainly includes single factor test index method, shows
Field measurement method, neural network and fuzzy clustering algorithm.
But there is different degrees of limitation or inborn weak point in these above-mentioned methods：
Single factor test index method, stability assessment is carried out on the basis of single influence factor due to establishing, although method
Simply, it is easy to operate, quick.
But because surrounding rock of actual mining roadway stability is the coefficient result of factors, and between multiple influence factors
Mutual restriction, connect each other, influence each other, it is this foundation on the basis of single influence factor carry out stability assessment side
Method, inevitably, larger limitation and one-sidedness be present.
Field measurement method is the most accurate, reliable method.
But its quantities is big, wastes time and energy, cost of labor is high, plus factors such as the geographical complicated and condition of construction in underground
Restriction, many times, extensive or comprehensive field measurement can not be carried out.
Neural network, it is built upon the influence factor of analysis surrounding rock of actual mining roadway stability and the base of field measurement data
On plinth, the neural network prediction model of surrounding rock of actual mining roadway stability is built.
But neural network prediction model carries out easily being absorbed in locally optimal solution, predictablity rate when weights update with threshold value
It is relatively not high, and Generalization Ability is poor.
Fuzzy clustering algorithm is to construct fuzzy matrix according to the influence factor of surrounding rock of actual mining roadway stability, and basic herein
Upper basis certain degree of membership determines clustering relationships.
But this method agriculture products weight and during degree of membership in the presence of stronger subjectivity, when sample size is larger
Obtaining cluster conclusion has certain difficulty.
As can be seen here, prior art is needed further improvement and improved, can be accurately and effectively pre- there is an urgent need to one kind
The method that survey time adopts improving stability of surrounding rocks in roadway, and the maturity of this method can be assessed.
The content of the invention
It is an object of the present invention to provide a kind of prediction of surrounding rock of actual mining roadway stability and appraisal procedure, and it has operation
Simply, prediction and appraisal procedure science, surrounding rock of actual mining roadway stability grade forecast result and models mature degree assessment result are accurate
Really, reliably, the features such as being identical with actual conditions.
The present invention to achieve the above object, the technical scheme adopted is that a kind of surrounding rock of actual mining roadway stability prediction with
Appraisal procedure, it is characterised in that comprise the following steps：
The first step, with the ratio between the country rock weighed intensities σ of stope drift active workings, tunnel buried depth H, immediate roof thickness and mining height N,
The ratio between width of chain pillar X, coal seam and lane height K and first roof caving step pitch L, the shadow as surrounding rock of actual mining roadway stability
The factor of sound, field measurement is carried out to surrounding rock of actual mining roadway stability grade and its influence factor, measured data is inputted into computer,
Establish sample database；
Second step, some groups of sample datas are transferred from sample database, the influence factor of each group of sample data is made
For input vector, and by the surrounding rock of actual mining roadway stability grade corresponding to each group of sample data, as output vector, establish
Monokaryon function SVM prediction model, show that training sample rolls over the consensus forecast accuracy rate under cross validation in M；
Using the influence factor of the surrounding rock of actual mining roadway stability as during the input vector of forecast model, it is necessary to first press
Following formula (1) is normalized：
In above formula (1)：
xjFor sample value after normalization；
xiFor training sample input value；
xminFor training sample minimum value；
xmaxFor training sample maximum；
3rd step, the higher global kernel function of consensus forecast accuracy rate and local kernel function are chosen, is combined into mixed nucleus letter
Number, establishes mixed kernel function SVM prediction model；
4th step, using particle cluster algorithm to the penalty factor in above-mentioned mixed kernel function SVM prediction model,
Nuclear parameter and monokaryon function coefficients optimize, and establish particle cluster algorithm-mixed kernel function of surrounding rock of actual mining roadway stability
SVM prediction model；
5th step, the sample to be predicted after normalization is input to the particle cluster algorithm-mixed kernel function branch trained
Hold in vector machine forecast model, draw the precision of prediction and reliability of sample to be predicted；
6th step, the appraisal procedure of particle cluster algorithm-mixed kernel function SVM prediction model maturity is built, it is right
Models mature degree is assessed.
The technical effect directly brought by the technical proposal is that to the prediction result of surrounding rock of actual mining roadway stability grade with
The assessment result of models mature degree is accurate, reliable, is identical with actual conditions.
Above-mentioned technical proposal, the prediction of accurate, the reliable surrounding rock of actual mining roadway stability grade of prediction result was both can obtain
Model, the assessment of forecast model maturity can be completed simultaneously again, there is good practicality.
Also, the surrounding rock of actual mining roadway stability prediction and appraisal procedure of above-mentioned technical proposal, whole flow process design science
Rationally, it is easy to operate.
Preferably, above-mentioned monokaryon function includes linear kernel function K (x, xj)=xTxj+ a, Polynomial kernel function K (x, xj)=
(βxTxj+r)d, Radial basis kernel function K (x, xj)=exp (- γ | | x-xj||2) and Sigmoid kernel function K (x, xj)=tanh (μ
xTxj+h)；
The linear kernel function is local kernel function, Polynomial kernel function and Sigmoid kernel functions with Radial basis kernel function
For global kernel function.
What the optimal technical scheme was directly brought has the technical effect that, passes through more each monokaryon function SVM prediction
The predictablity rate of model, it can more accurately and effectively select the higher global kernel function of predictablity rate and karyomerite letter
Number.Further preferably, the consensus forecast accuracy rate calculating process of above-mentioned monokaryon function SVM prediction model includes following
Step：
(1) training sample is divided into S parts, 2~S parts sample is used to train, and the 1st part of sample is used to test, record the 1st
The predictablity rate of part sample；
(2) it is repeated in step (1) and obtains the predictablity rate of S part samples, its average is monokaryon function supporting vector
The consensus forecast accuracy rate of machine forecast model.
What the optimal technical scheme was directly brought has the technical effect that, it is pre- to obtain reliable and stable single kernel function support vector machine
Model is surveyed, and whole process is convenient and simple, easy to operation.
Further preferably, above-mentioned mixed kernel function is that monokaryon function SVM prediction model is accurate in higher consensus forecast
Under true rate, the combination of global kernel function and local kernel function, (2) are carried out as the following formula for combination：
KM(x,xj)=λ1K1(x,xj)+λ2K2(x,xj) (2)；
In above formula (2)：
xjIt is characterized vector, x, xj∈Rn, (x, xj) it is inner product；
λ1With λ2For monokaryon function coefficients, λ1,λ2>0, and λ1+λ2=1；
K1(x,xj) it is global kernel function, K2(x,xj) it is local kernel function.
What the optimal technical scheme was directly brought has the technical effect that, ensures that mixed kernel function meets Mercer conditions, passes through
Adjust λ1With λ2So that mixed kernel function, which holds vector machine forecast model, possesses stronger learning ability and generalization ability, have good
Practicality.
Further preferably, above-mentioned particle cluster algorithm optimization mixed kernel function SVM prediction model is as follows
Establish, comprise the following steps：
1st step, initialize population
Set population initialization cognitive learning factor c1With social learning factor c2, i-th of particle initial position is xi=
(xi1,xi2, xid), i-th of particle initial velocity is vi=(vi1,vi2,···,vid)T, i-th of particle individual is most
Excellent position is pi=(pi1,pi2, pid), i-th of particle global optimum position is gi=(gi1,gi2,···,gid)。
2nd step, determines fitness function
Using the consensus forecast accuracy rate of mixed kernel function SVM prediction model as the adaptation for evaluating each particle
Function is spent, fitness function expression formula is following formula (3)：
In above formula (3)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NciFor the correctly predicted sample number of the i-th grade；
NeiFor the error prediction sample number of the i-th grade.
3rd step, calculate fitness value
The fitness value of each particle is calculated using fitness function, and adjusts the individual optimal position of particle according to the following steps
Put and population optimal location：
The first fitness value of more each particle and its optimal location of process, obtain the personal best particle of particle, such as
Fruit currency plBetter than the individual extreme value p of particlebest, then p is setl=pbest, and set currency plPosition for individual it is optimal
Position；
Then, the fitness value of more whole population and its optimal location of process, the optimal position of whole population is obtained
Put, if currency glBetter than the global extremum g of populationbest, then g is setl=gbest, and set currency glPosition be population
Optimal location；
4th step, iteration renewal
(4) are iterated renewal to the speed of each particle with position as the following formula：
In above formula (4)：
D is population population；
ω is inertia weight；
r1,r2∈(0,1)；
T is iterations；
5th step, whether error in judgement condition, which meets, requires, or whether reaches maximum iteration.If so, terminate iteration；
Otherwise, it is back to the 3rd step.
What the optimal technical scheme was directly brought has the technical effect that, finds model optimized parameter by particle cluster algorithm, makes
It is high to obtain the prediction result accuracy rate of mixed kernel function SVM prediction model, is coincide with actual conditions.
Further preferably, (5) calculate the precision of prediction of above-mentioned sample to be predicted as the following formula：
In above formula (5)：
SeiFor the sensitivity function of the i-th grade sample to be predicted；
SpiFor the selectivity function of the i-th grade sample to be predicted；
Nc is correctly predicted sample number；
Ne is error prediction sample number；
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
N is total sample number to be predicted, N=Nc+Ne.
What the optimal technical scheme was directly brought has the technical effect that, it is objective, reasonably evaluate mixed kernel function supporting vector
The precision of prediction of machine forecast model, help in next step to assess models mature degree.
Further preferably, shown in above-mentioned total sample number N to be predicted such as following formulas (6)：
In above formula (6)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NiFor the sample number to be predicted of the i-th grade；
Above-mentioned sensitivity function SeiAs shown in following formula (7)：
In above formula (7)：
NciFor the correctly predicted sample number of the i-th grade；
NiFor the sample number to be predicted of the i-th grade；
Above-mentioned selectivity function SpiAs shown in following formula (8)：
In above formula (8)：
NeiFor the error prediction sample number of the i-th grade；
N is total sample number to be predicted；
NiFor the sample number to be predicted of the i-th grade；
Shown in above-mentioned correctly predicted sample number Nc such as following formulas (9)：
In above formula (9)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NciFor the correctly predicted sample number of the i-th grade；
Shown in above-mentioned error prediction sample number Ne such as following formulas (10)：
In above formula (10)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NeiFor the error prediction sample number of the i-th grade；
What the optimal technical scheme was directly brought has the technical effect that, sensitiveness that is objective, reasonably defining prediction result with
Selectivity, help to calculate precision of prediction.
Further preferably, (11) are calculated the predicting reliability of above-mentioned sample to be predicted as the following formula：
In above formula (11)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
rijReliability for sample to be predicted by the i-th grade forecast into jth grade, rij∈[0,1]；
αijIncidence for sample to be predicted by the i-th grade forecast into jth grade sample, αij∈[0,1]；
What the optimal technical scheme was directly brought has the technical effect that, it is objective, reasonably evaluate mixed kernel function supporting vector
The predicting reliability of machine forecast model, help in next step to assess models mature degree.
Further preferably, above-mentioned reliability rijAssignment, the incidence α are completed by expert gradedij(12) as the following formula
It is calculated：
In above formula (12)：
NciFor the correctly predicted sample number of the i-th grade；
NjFor the sample number to be predicted of jth grade；
What the optimal technical scheme was directly brought has the technical effect that, objective reasonably to define reliability and incidence, helps
Calculated in predicting reliability.
Further preferably, above-mentioned particle cluster algorithm-mixed kernel function SVM prediction model maturity is as the following formula
(13) it is calculated：
Ma=PR (13)；
In above formula (13)：
P is the precision of prediction of sample to be predicted；
R is the predicting reliability of sample to be predicted.
What the optimal technical scheme was directly brought has the technical effect that, pre- according to mixed kernel function SVM prediction model
Precision of prediction and the reliability for surveying result accurately and reliably evaluate the models mature degree, and whole flow process design science rationally,
It is easy to operate.
In summary, sample data can be inputted computer by the present invention, be established by Matlab programmings on stope drift active workings
The forecast model and model of surrounding rock stability, and models mature degree is assessed, whole flow process design science is reasonable, operation letter
Just.The present invention is relative to prior art, with to surrounding rock of actual mining roadway stability grade forecast and the assessment of models mature degree
As a result it is accurate, objective, reasonable and reliable, there is the beneficial effects such as the higher goodness of fit with actual conditions.
Brief description of the drawings
Fig. 1 is the flow chart of the present invention；
Fig. 2 is the flow chart of the mixed kernel function SVM prediction model construction step of the present invention；
Fig. 3 is that the particle cluster algorithm of the present invention optimizes the logic diagram of mixed kernel function SVM prediction model；
Fig. 4 is particle cluster algorithm-mixed kernel function SVM prediction model maturity appraisal procedure of the present invention
Flow chart；
Fig. 5 is the particle cluster algorithm optimizing curve map of the embodiment of the present invention 1；
Fig. 6 is the prediction result of the embodiment of the present invention；
Fig. 7 is particle cluster algorithm-mixed kernel function SVM prediction model maturity change of the embodiment of the present invention 1
Curve map.
Embodiment
With reference to the accompanying drawings and examples, the present invention is described in detail.
Embodiment
With the exploiting field of Berlin ore deposit 042, the exploiting field of Bai Laping ore deposits 032, iron mountain Nan Kuang112 exploiting fields, the exploiting field of Buddha's warrior attendant ore deposit 411 and refined youth
Exemplified by 88 groups of surrounding rock of actual mining roadway stability grades and its influence factor that the exploiting field of ore deposit 421 is collected into, it is illustrated.
As shown in Figures 1 to 5, surrounding rock of actual mining roadway stability grade includes stable, moderate stable and unstable three
Grade, represented respectively with 1,2 with 3, corresponding influence factor includes the country rock weighed intensities σ of stope drift active workings, tunnel buried depth
H, the ratio between the ratio between immediate roof thickness and mining height N, width of chain pillar X, coal seam and lane height K and first roof caving step pitch L, is shown in
Table 1：
Table 1：Surrounding rock of actual mining roadway stability measured data
The surrounding rock of actual mining roadway stability grade is predicted and assesses the maturity of the model, by such as Fig. 1-Fig. 5 institutes
The step of showing is carried out：
The first step, by the country rock weighed intensities of surrounding rock of actual mining roadway stability grade in table 1, and its corresponding stope drift active workings
The ratio between the ratio between σ, tunnel buried depth H, immediate roof thickness and mining height N, width of chain pillar X, coal seam and lane height K and directly top are just
Secondary caving angle L inputs computer, establishes sample database；
Second step, 63 groups of sample datas are transferred from sample database, by the influence factor of each group of sample data, as
Input vector, and by the surrounding rock of actual mining roadway stability grade corresponding to each group of sample data, as output vector, build respectively
The single kernel function support vector machine for linear kernel function, Polynomial kernel function, Radial basis kernel function and the Sigmoid kernel functions of being based on
Forecast model, and show that consensus forecast of each monokaryon function SVM prediction model under 10 folding cross validations is accurate
Rate, it see the table below 2：
Table 2：Monokaryon function SVM prediction model consensus forecast accuracy rate
Using the influence factor of surrounding rock of actual mining roadway stability as during the input vector of forecast model, it is necessary to first as the following formula
(1) normalize to [0,1]：
In above formula (1)：
xjFor sample value after normalization；
xiFor training sample input value；
xminFor training sample minimum value；
xmaxFor training sample maximum；
3rd step, the higher global kernel function-Sigmoid kernel functions of selection consensus forecast accuracy rate and local kernel function-
Radial basis kernel function, as the following formula (2) be combined into mixed kernel function, establish mixed kernel function SVM prediction model；
KM(x,xj)=λ1tanh(βxTxj+r)+λ2exp(-γ||x-xj||2) (2)
In above formula (2)：
λ1For Sigmoid kernel function coefficients, λ2For Radial basis kernel function coefficient, λ1,λ2>0, and λ1+λ2=1；
β is Sigmoid kernel function slopes, β=1/N；
R is Sigmoid kernel function constants, r=1；
γ is Radial basis kernel function slope, γ=1/N；
4th step, using particle cluster algorithm to the penalty factor in above-mentioned mixed kernel function SVM prediction model,
Nuclear parameter and monokaryon function coefficients optimize, and export optimal penalty parameter c=0.79088, nuclear parameter g=1.0697,
Sigmoid kernel function coefficient lambdas1=0.2 with Radial basis kernel function coefficient lambda2=0.8, establish the particle of surrounding rock of actual mining roadway stability
Group's algorithm-mixed kernel function SVM prediction model；
5th step, using 25 groups of sample datas of residue after normalization as test sample, it is input to the particle trained
In group's algorithm-mixed kernel function SVM prediction model, the precision of prediction P=1 and reliability R=of sample to be predicted are drawn
1；
6th step, the appraisal procedure of particle cluster algorithm-mixed kernel function SVM prediction model maturity is built, is commented
Estimate and model maturity M=1；
Monokaryon function includes linear kernel function K (x, xj)=xTxj+ a, Polynomial kernel function K (x, xj)=(β xTxj+r)d、
Radial basis kernel function K (x, xj)=exp (- γ | | x-xj||2) and Sigmoid kernel function K (x, xj)=tanh (μ xTxj+h)；
Linear kernel function is local kernel function with Radial basis kernel function, and Polynomial kernel function and Sigmoid kernel functions are complete
Office's kernel function.
The consensus forecast accuracy rate calculating process of monokaryon function SVM prediction model comprises the following steps：
(1) training sample is divided into 10 parts, the 2nd~10 part of sample is used to train, and the 1st part of sample is used to test, and records
The predictablity rate of 1st part of sample；
(2) it is repeated in step (1) and obtains the predictablity rate of 10 parts of samples, its average is monokaryon function supporting vector
The consensus forecast accuracy rate of machine forecast model, draw each monokaryon function SVM prediction model under 10 folding cross validations
Consensus forecast accuracy rate, see the above table 2.
Mixed kernel function be monokaryon function SVM prediction model under higher consensus forecast accuracy rate, global core letter
Number and the combination of local kernel function, (3) are carried out as the following formula for combination：
KM(x,xj)=λ1K1(x,xj)+λ2K2(x,xj) (3)；
In above formula (3)：
xjIt is characterized vector, x, xj∈Rn, (x, xj) it is inner product；
λ1With λ2For monokaryon function coefficients, λ1,λ2>0, and λ1+λ2=1；
K1(x,xj) it is global kernel function, K2(x,xj) it is local kernel function.
Particle cluster algorithm optimization mixed kernel function SVM prediction model is established as follows, including following
Step：
1st step, initialize population
Set population initialization cognitive learning factor c1With social learning factor c2, i-th of particle initial position is xi=
(xi1,xi2, xid), i-th of particle initial velocity is vi=(vi1,vi2,···,vid)T, i-th of particle individual is most
Excellent position is pi=(pi1,pi2, pid), i-th of particle global optimum position is gi=(gi1,gi2,···,gid)。
2nd step, determines fitness function
Using the consensus forecast accuracy rate of mixed kernel function SVM prediction model as the adaptation for evaluating each particle
Function is spent, fitness function expression formula is following formula (4)：
In above formula (4)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NciFor the correctly predicted sample number of the i-th grade；
NeiFor the error prediction sample number of the i-th grade.
3rd step, calculate fitness value
The fitness value of each particle is calculated using fitness function, and adjusts the individual optimal position of particle according to the following steps
Put and population optimal location：
The first fitness value of more each particle and its optimal location of process, obtain the personal best particle of particle, such as
Fruit currency plBetter than the individual extreme value p of particlebest, then p is setl=pbest, and set currency plPosition for individual it is optimal
Position；
Then, the fitness value of more whole population and its optimal location of process, the optimal position of whole population is obtained
Put, if currency glBetter than the global extremum g of populationbest, then g is setl=gbest, and set currency glPosition be population
Optimal location；
4th step, iteration renewal
(5) are iterated renewal to the speed of each particle with position as the following formula：
In above formula (5)：
D is population population；
ω is inertia weight；
r1,r2∈(0,1)；
T is iterations；
5th step, whether error in judgement condition, which meets, requires, or whether reaches maximum iteration.If so, terminate iteration；
Otherwise, it is back to the 3rd step.
As shown in fig. 6, the sample to be predicted after normalization is input to the particle cluster algorithm-mixed kernel function trained
SVM prediction model, draw prediction result.
Work as λ1When=0.2, the precision of prediction of sample to be predicted is calculated in (6) as the following formula：
(6) calculate the precision of prediction of sample to be predicted as the following formula：
In above formula (6)：
SeiFor the sensitivity function of the i-th grade sample to be predicted, Sei=1；
SpiFor the selectivity function of the i-th grade sample to be predicted, Spi=0；
Nc is correctly predicted sample number, Nc=25；
Ne is error prediction sample number, Ne=0；
N is total sample number to be predicted, N=25；
N is surrounding rock of actual mining roadway stability rating-type number, n=3.
Shown in total sample number N to be predicted such as following formulas (7)：
In above formula (7)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NiFor the sample number to be predicted of the i-th grade；
Sensitivity function SeiAs shown in following formula (8)：
In above formula (8)：
NciFor the correctly predicted sample number of the i-th grade；
NiFor the sample number to be predicted of the i-th grade；
Selectivity function SpiAs shown in following formula (9)：
In above formula (9)：
NeiFor the error prediction sample number of the i-th grade；
N is total sample number to be predicted；
NiFor the sample number to be predicted of the i-th grade；
Shown in correctly predicted sample number Nc such as following formulas (10)：
In above formula (10)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NciFor the correctly predicted sample number of the i-th grade；
Shown in error prediction sample number Ne such as following formulas (11)：
In above formula (11)：
NeiFor the error prediction sample number of the i-th grade；
The reliability of sample to be predicted is calculated in (12) as the following formula：
In above formula (12)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
rijReliability for sample to be predicted by the i-th grade forecast into jth grade, rij∈[0,1]；
αijIncidence for sample to be predicted by the i-th grade forecast into jth grade sample, αij∈[0,1]；
Particle cluster algorithm-mixed kernel function SVM prediction model maturity is calculated in (13) as the following formula：
Ma=PR=1 (13)；
In above formula (13)：
P be sample to be predicted precision of prediction, P=1；
R be sample to be predicted predicting reliability, R=1.
As shown in fig. 7, except in λ1=0.2 models mature degree Ma can reach outside maximum, can produce multigroup λ1With λ2Make
Obtain models mature degree and reach maximum.
Claims (10)
1. a kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure, it is characterised in that comprise the following steps：
The first step, with the ratio between the country rock weighed intensities σ of stope drift active workings, tunnel buried depth H, immediate roof thickness and mining height N, Hu Xiang
The ratio between coal pillar width X, coal seam and lane height K and first roof caving step pitch L, as surrounding rock of actual mining roadway stability influence because
Element, field measurement is carried out to surrounding rock of actual mining roadway stability grade and its influence factor, measured data is inputted into computer, is established
Sample database；
Second step, some groups of sample datas are transferred from sample database, by the influence factor of each group of sample data, as defeated
Incoming vector, and by the surrounding rock of actual mining roadway stability grade corresponding to each group of sample data, as output vector, establish monokaryon
Function SVM prediction model, show that training sample rolls over the consensus forecast accuracy rate under cross validation in M；
Using the influence factor of the surrounding rock of actual mining roadway stability as during the input vector of forecast model, it is necessary to first as the following formula
(1) it is normalized：
<mrow>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msub>
<mi>x</mi>
<mi>max</mi>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (1)：
xjFor sample value after normalization；
xiFor training sample input value；
xminFor training sample minimum value；
xmaxFor training sample maximum；
3rd step, the higher global kernel function of consensus forecast accuracy rate and local kernel function are chosen, mixed kernel function is combined into, builds
Vertical mixed kernel function SVM prediction model；
4th step, the penalty factor in above-mentioned mixed kernel function SVM prediction model, core are joined using particle cluster algorithm
Number and monokaryon function coefficients are optimized, and the particle cluster algorithm-mixed kernel function for establishing surrounding rock of actual mining roadway stability is supported
Vector machine forecast model；
5th step, the sample to be predicted after normalization is input to the particle cluster algorithm-mixed kernel function trained support to
In amount machine forecast model, the precision of prediction and reliability of sample to be predicted are drawn；
6th step, the appraisal procedure of particle cluster algorithm-mixed kernel function SVM prediction model maturity is built, to model
Maturity is assessed.
2. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that the monokaryon
Function includes linear kernel function K (x, xj)=xTxj+ a, Polynomial kernel function K (x, xj)=(β xTxj+r)d, Radial basis kernel function K
(x,xj)=exp (- γ | | x-xj||2) and Sigmoid kernel function K (x, xj)=tanh (μ xTxj+h)；
The linear kernel function is local kernel function with Radial basis kernel function, and Polynomial kernel function and Sigmoid kernel functions are complete
Office's kernel function.
3. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that the monokaryon
The consensus forecast accuracy rate calculating process of function SVM prediction model comprises the following steps：
(1) training sample is divided into S parts, 2~S parts sample is used to train, and the 1st part of sample is used to test, and records the 1st part of sample
This predictablity rate；
(2) it is repeated in step (1) and obtains the predictablity rate of S part samples, its average is that single kernel function support vector machine is pre-
Survey the consensus forecast accuracy rate of model.
4. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that the mixing
Kernel function be monokaryon function SVM prediction model under higher consensus forecast accuracy rate, global kernel function and karyomerite letter
Several combinations, (2) are carried out as the following formula for combination：
KM(x,xj)=λ1K1(x,xj)+λ2K2(x,xj) (2)；
In above formula (2)：
xjIt is characterized vector, x, xj∈Rn, (x, xj) it is inner product；
λ1With λ2For monokaryon function coefficients, λ1,λ2>0, and λ1+λ2=1；
K1(x,xj) it is global kernel function, K2(x,xj) it is local kernel function.
5. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that the particle
Group's algorithm optimization mixed kernel function SVM prediction model is established as follows, is comprised the following steps：
1st step, initialize population
Set population initialization cognitive learning factor c1With social learning factor c2, i-th of particle initial position is xi=(xi1,
xi2..., xid), i-th of particle initial velocity is vi=(vi1,vi2,…,vid)T, i-th of particle personal best particle is pi=
(pi1,pi2..., pid), i-th of particle global optimum position is gi=(gi1,gi2,…,gid)。
2nd step, determines fitness function
Using the consensus forecast accuracy rate of mixed kernel function SVM prediction model as the fitness letter for evaluating each particle
Number, fitness function expression formula is following formula (3)：
<mrow>
<mi>f</mi>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>Nc</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>Nc</mi>
<mi>i</mi>
</msub>
<mo>+</mo>
<msub>
<mi>Ne</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>&times;</mo>
<mn>100</mn>
<mi>%</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (3)：
N is surrounding rock of actual mining roadway stability number of degrees, n≤5；
NciFor the correctly predicted sample number of the i-th grade；
NeiFor the error prediction sample number of the i-th grade.
3rd step, calculate fitness value
The fitness value of each particle is calculated using fitness function, and adjust according to the following steps particle personal best particle and
Population optimal location：
The first fitness value of more each particle and its optimal location of process, obtain the personal best particle of particle, if worked as
Preceding value plBetter than the individual extreme value p of particlebest, then p is setl=pbest, and set currency plPosition be personal best particle；
Then, the fitness value of more whole population and its optimal location of process, obtain the optimal location of whole population, such as
Fruit currency glBetter than the global extremum g of populationbest, then g is setl=gbest, and set currency glPosition for population it is optimal
Position；
4th step, iteration renewal
(4) are iterated renewal to the speed of each particle with position as the following formula：
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<msup>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msup>
<msub>
<mi>&omega;v</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>+</mo>
<msub>
<mi>c</mi>
<mn>1</mn>
</msub>
<msub>
<mi>r</mi>
<mn>1</mn>
</msub>
<mrow>
<mo>(</mo>
<msup>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>-</mo>
<msup>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>)</mo>
</mrow>
<mo>+</mo>
<msub>
<mi>c</mi>
<mn>2</mn>
</msub>
<msub>
<mi>r</mi>
<mn>2</mn>
</msub>
<mrow>
<mo>(</mo>
<msup>
<msub>
<mi>p</mi>
<mrow>
<mi>g</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>-</mo>
<msup>
<msub>
<mi>x</mi>
<mrow>
<mi>g</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msup>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msup>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>+</mo>
<msup>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mi>d</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msup>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (4)：
D is population population；
ω is inertia weight；
r1,r2∈(0,1)；
T is iterations；
5th step, whether error in judgement condition, which meets, requires, or whether reaches maximum iteration.If so, terminate iteration；Otherwise,
It is back to the 3rd step.
6. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that it is described treat it is pre-
(5) are calculated the precision of prediction of test sample sheet as the following formula：
<mrow>
<mi>P</mi>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<msub>
<mi>Se</mi>
<mi>i</mi>
</msub>
<mo>&CenterDot;</mo>
<mi>N</mi>
<mi>c</mi>
<mo>+</mo>
<msub>
<mi>Sp</mi>
<mi>i</mi>
</msub>
<mo>&CenterDot;</mo>
<mi>N</mi>
<mi>e</mi>
</mrow>
<mi>N</mi>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
<mi>n</mi>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (5)：
SeiFor the sensitivity function of the i-th grade sample to be predicted；
SpiFor the selectivity function of the i-th grade sample to be predicted；
Nc is correctly predicted sample number；
Ne is error prediction sample number；
N is total sample number to be predicted, N=Nc+Ne.
7. surrounding rock of actual mining roadway stability prediction according to claim 6 and appraisal procedure, it is characterised in that it is described treat it is pre-
Survey shown in total sample number N such as following formulas (6)：
<mrow>
<mi>N</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>N</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (6)：
NiFor the sample number to be predicted of the i-th grade；
The sensitivity function SeiAs shown in following formula (7)：
<mrow>
<msub>
<mi>Se</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>Nc</mi>
<mi>i</mi>
</msub>
</mrow>
<msub>
<mi>N</mi>
<mi>i</mi>
</msub>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>7</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (7)：
NciFor the correctly predicted sample number of the i-th grade；
NiFor the sample number to be predicted of the i-th grade；
The selectivity function SpiAs shown in following formula (8)：
<mrow>
<msub>
<mi>Sp</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>Ne</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<mi>N</mi>
<mo>-</mo>
<msub>
<mi>N</mi>
<mi>i</mi>
</msub>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (8)：
NeiFor the error prediction sample number of the i-th grade；
N is total sample number to be predicted；
NiFor the sample number to be predicted of the i-th grade；
Shown in the correctly predicted sample number Nc such as following formulas (9)：
<mrow>
<mi>N</mi>
<mi>c</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>Nc</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>9</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (9)：
NciFor the correctly predicted sample number of the i-th grade；
Shown in the error prediction sample number Ne such as following formulas (10)：
<mrow>
<mi>N</mi>
<mi>e</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>Ne</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>10</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (10)：
NeiFor the error prediction sample number of the i-th grade.
8. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that it is described treat it is pre-
(11) are calculated the predicting reliability of test sample sheet as the following formula：
<mrow>
<mi>R</mi>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>&CenterDot;</mo>
<msub>
<mi>&alpha;</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mi>n</mi>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>11</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (11)：
rijReliability for sample to be predicted by the i-th grade forecast into jth grade, rij∈[0,1]；
αijIncidence for sample to be predicted by the i-th grade forecast into jth grade sample, αij∈[0,1]。
9. surrounding rock of actual mining roadway stability prediction according to claim 8 and appraisal procedure, it is characterised in that described reliable
Spend rijAssignment, the incidence α are completed by expert gradedij(12) are calculated as the following formula：
<mrow>
<msub>
<mi>&alpha;</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>Nc</mi>
<mi>i</mi>
</msub>
</mrow>
<msub>
<mi>N</mi>
<mi>j</mi>
</msub>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>12</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
In above formula (12)：
NciFor the correctly predicted sample number of the i-th grade；
NjFor the sample number to be predicted of jth grade.
10. surrounding rock of actual mining roadway stability prediction according to claim 1 and appraisal procedure, it is characterised in that the grain
(13) are calculated swarm optimization-mixed kernel function SVM prediction model maturity as the following formula：
Ma=PR (13)；
In above formula (13)：
P is the precision of prediction of sample to be predicted；
R is the predicting reliability of sample to be predicted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710472314.5A CN107357966B (en) | 2017-06-21 | 2017-06-21 | Evaluation method for stability prediction of surrounding rock of stoping roadway |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710472314.5A CN107357966B (en) | 2017-06-21 | 2017-06-21 | Evaluation method for stability prediction of surrounding rock of stoping roadway |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107357966A true CN107357966A (en) | 2017-11-17 |
| CN107357966B CN107357966B (en) | 2020-07-03 |
Family
ID=60273998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710472314.5A Active CN107357966B (en) | 2017-06-21 | 2017-06-21 | Evaluation method for stability prediction of surrounding rock of stoping roadway |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107357966B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564117A (en) * | 2018-03-30 | 2018-09-21 | 华南理工大学 | A kind of poor student's auxiliary identification based on SVM |
| CN109165406A (en) * | 2018-07-20 | 2019-01-08 | 大连海事大学 | The quick dynamic classification prediction technique of country rock during a kind of constructing tunnel |
| CN109635994A (en) * | 2018-10-23 | 2019-04-16 | 广东精点数据科技股份有限公司 | A kind of crop yield prediction technique for realizing the multi-source heterogeneous fusion of influence factor |
| CN109740183A (en) * | 2018-12-06 | 2019-05-10 | 武汉理工大学 | Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method |
| CN110097069A (en) * | 2019-03-11 | 2019-08-06 | 西安科技大学 | A kind of support vector machines Lithofacies Identification method and device based on depth Multiple Kernel Learning |
| CN110175194A (en) * | 2019-04-19 | 2019-08-27 | 中国矿业大学 | A kind of coal mine roadway surrouding rock deformation rupture discrimination method based on association rule mining |
| CN110399668A (en) * | 2019-07-17 | 2019-11-01 | 西安工业大学 | A method of rapidly and accurately solving calutron output characteristics |
| CN111079810A (en) * | 2019-12-06 | 2020-04-28 | 中国铁路设计集团有限公司 | Tunnel surrounding rock grade prediction method based on support vector machine |
| CN111222736A (en) * | 2019-09-29 | 2020-06-02 | 中国人民解放军空军工程大学 | Ammunition storage reliability evaluation method based on mixed correlation vector machine model |
| CN111291934A (en) * | 2020-02-18 | 2020-06-16 | 山东科技大学 | Surrounding rock real-time grading prediction and self-checking method in tunnel construction process |
| CN111400886A (en) * | 2020-03-05 | 2020-07-10 | 河海大学 | Displacement calculation method for enclosed pressure arch circular tunnel with unevenly distributed surrounding rock space |
| CN112884198A (en) * | 2021-01-13 | 2021-06-01 | 西安理工大学 | Dam crest settlement prediction method combining threshold regression and improved support vector machine panel dam |
| CN113657515A (en) * | 2021-08-20 | 2021-11-16 | 盾构及掘进技术国家重点实验室 | Classification method for judging and improving tunnel surrounding rock grade of FMC model based on rock sensitivity parameters |
| CN113673136A (en) * | 2021-09-06 | 2021-11-19 | 长安大学 | Method, system and equipment for predicting safe thickness of top plate of karst tunnel |
| CN115929406A (en) * | 2022-12-07 | 2023-04-07 | 河海大学 | Neural network-based surrounding rock stability prediction method, system and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1395808A4 (en) * | 2001-05-11 | 2007-01-03 | Transform Pharmaceuticals Inc | Method and system for planning, performing, and assessing high-throughput screening of multicomponent chemical compositions and solid forms of compounds |
| CN102005135A (en) * | 2010-12-09 | 2011-04-06 | 上海海事大学 | Genetic algorithm-based support vector regression shipping traffic flow prediction method |
| CN102608285A (en) * | 2012-02-21 | 2012-07-25 | 南京工业大学 | Prediction method for explosion characteristics of organic mixture based on support vector machine |
| CN103631137A (en) * | 2013-10-08 | 2014-03-12 | 江苏大学 | Five-freedom-degree alternating current active magnetic bearing mixed kernel function support vector machine detecting method |
| CN103631138A (en) * | 2013-10-08 | 2014-03-12 | 江苏大学 | Three-DOF (Degree of Freedom) hybrid magnetic bearing mixed kernel function support vector machine displacement detection method |
| US20140136168A1 (en) * | 2012-11-13 | 2014-05-15 | Baker Hughes Incorporated | Drill bit simulation and optimization |
| US20150161303A1 (en) * | 2013-12-11 | 2015-06-11 | Jerome Lewandowski | Method for Automated Rock Physics Modeling |
| US20160110481A1 (en) * | 2014-10-16 | 2016-04-21 | Baker Hughes Incorporated | Modeling and simulation of drill strings with adaptive systems |
| CN105938116A (en) * | 2016-06-20 | 2016-09-14 | 吉林大学 | Gas sensor array concentration detection method based on fuzzy division and model integration |
| CN106339357A (en) * | 2016-08-12 | 2017-01-18 | 北京同余科技有限公司 | Optimizing parameterized network information security detection system and method |
-
2017
- 2017-06-21 CN CN201710472314.5A patent/CN107357966B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1395808A4 (en) * | 2001-05-11 | 2007-01-03 | Transform Pharmaceuticals Inc | Method and system for planning, performing, and assessing high-throughput screening of multicomponent chemical compositions and solid forms of compounds |
| CN102005135A (en) * | 2010-12-09 | 2011-04-06 | 上海海事大学 | Genetic algorithm-based support vector regression shipping traffic flow prediction method |
| CN102608285A (en) * | 2012-02-21 | 2012-07-25 | 南京工业大学 | Prediction method for explosion characteristics of organic mixture based on support vector machine |
| US20140136168A1 (en) * | 2012-11-13 | 2014-05-15 | Baker Hughes Incorporated | Drill bit simulation and optimization |
| CN103631137A (en) * | 2013-10-08 | 2014-03-12 | 江苏大学 | Five-freedom-degree alternating current active magnetic bearing mixed kernel function support vector machine detecting method |
| CN103631138A (en) * | 2013-10-08 | 2014-03-12 | 江苏大学 | Three-DOF (Degree of Freedom) hybrid magnetic bearing mixed kernel function support vector machine displacement detection method |
| US20150161303A1 (en) * | 2013-12-11 | 2015-06-11 | Jerome Lewandowski | Method for Automated Rock Physics Modeling |
| US20160110481A1 (en) * | 2014-10-16 | 2016-04-21 | Baker Hughes Incorporated | Modeling and simulation of drill strings with adaptive systems |
| CN105938116A (en) * | 2016-06-20 | 2016-09-14 | 吉林大学 | Gas sensor array concentration detection method based on fuzzy division and model integration |
| CN106339357A (en) * | 2016-08-12 | 2017-01-18 | 北京同余科技有限公司 | Optimizing parameterized network information security detection system and method |
Non-Patent Citations (6)
| Title |
|---|
| SHUXIONG QIU等: "Modified Particle Swarm Optimization for Multi-Scale Kernel Function in SVM", 《IEEE XPLORE》 * |
| 刘阳: "基于随机森林的回采巷道围岩稳定性预测", 《黑龙江科技信息》 * |
| 张风达: "深部煤层底板变形破坏机理及突水评价方法研究", 《中国博士学位论文全文数据库 工程科技Ⅰ辑》 * |
| 朱一丁等: "回采巷道围岩分类的支持向量机方法", 《采矿与安全工程学报》 * |
| 焦雪峰: "厚煤层旧采区复采矿压规律及其应用研究", 《中国博士学位论文全文数据库 工程科技Ⅰ辑》 * |
| 石永奎等: "回采巷道围岩稳定性的分类预测方法分析", 《煤炭技术》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564117A (en) * | 2018-03-30 | 2018-09-21 | 华南理工大学 | A kind of poor student's auxiliary identification based on SVM |
| CN108564117B (en) * | 2018-03-30 | 2022-03-29 | 华南理工大学 | SVM-based poverty and life assisting identification method |
| CN109165406A (en) * | 2018-07-20 | 2019-01-08 | 大连海事大学 | The quick dynamic classification prediction technique of country rock during a kind of constructing tunnel |
| CN109635994A (en) * | 2018-10-23 | 2019-04-16 | 广东精点数据科技股份有限公司 | A kind of crop yield prediction technique for realizing the multi-source heterogeneous fusion of influence factor |
| CN109740183A (en) * | 2018-12-06 | 2019-05-10 | 武汉理工大学 | Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method |
| CN110097069A (en) * | 2019-03-11 | 2019-08-06 | 西安科技大学 | A kind of support vector machines Lithofacies Identification method and device based on depth Multiple Kernel Learning |
| CN110175194A (en) * | 2019-04-19 | 2019-08-27 | 中国矿业大学 | A kind of coal mine roadway surrouding rock deformation rupture discrimination method based on association rule mining |
| CN110399668A (en) * | 2019-07-17 | 2019-11-01 | 西安工业大学 | A method of rapidly and accurately solving calutron output characteristics |
| CN110399668B (en) * | 2019-07-17 | 2023-05-16 | 西安工业大学 | Method for rapidly and accurately solving output characteristics of electromagnetic device |
| CN111222736A (en) * | 2019-09-29 | 2020-06-02 | 中国人民解放军空军工程大学 | Ammunition storage reliability evaluation method based on mixed correlation vector machine model |
| CN111079810A (en) * | 2019-12-06 | 2020-04-28 | 中国铁路设计集团有限公司 | Tunnel surrounding rock grade prediction method based on support vector machine |
| CN111291934A (en) * | 2020-02-18 | 2020-06-16 | 山东科技大学 | Surrounding rock real-time grading prediction and self-checking method in tunnel construction process |
| CN111291934B (en) * | 2020-02-18 | 2023-04-07 | 山东科技大学 | Surrounding rock real-time grading prediction and self-checking method in tunnel construction process |
| CN111400886A (en) * | 2020-03-05 | 2020-07-10 | 河海大学 | Displacement calculation method for enclosed pressure arch circular tunnel with unevenly distributed surrounding rock space |
| CN112884198A (en) * | 2021-01-13 | 2021-06-01 | 西安理工大学 | Dam crest settlement prediction method combining threshold regression and improved support vector machine panel dam |
| CN112884198B (en) * | 2021-01-13 | 2023-06-09 | 西安理工大学 | Method for predicting dam crest settlement of panel dam by combining threshold regression and improved support vector machine |
| CN113657515A (en) * | 2021-08-20 | 2021-11-16 | 盾构及掘进技术国家重点实验室 | Classification method for judging and improving tunnel surrounding rock grade of FMC model based on rock sensitivity parameters |
| CN113673136A (en) * | 2021-09-06 | 2021-11-19 | 长安大学 | Method, system and equipment for predicting safe thickness of top plate of karst tunnel |
| CN115929406A (en) * | 2022-12-07 | 2023-04-07 | 河海大学 | Neural network-based surrounding rock stability prediction method, system and storage medium |
| CN115929406B (en) * | 2022-12-07 | 2023-08-18 | 河海大学 | Surrounding rock stability prediction method, system and storage medium based on neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107357966B (en) | 2020-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107357966A (en) | A kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure | |
| Zhou et al. | Forecasting of TBM advance rate in hard rock condition based on artificial neural network and genetic programming techniques | |
| CN104182622B (en) | Based on feedback analysis method and its device in the constructing tunnel of extreme learning machine | |
| CN105654414A (en) | Urban multi-disaster risk loss evaluation system based on open source system framework and building spatial database and method thereof | |
| CN106372748A (en) | Hard-rock tunnel boring machine boring efficiency prediction method | |
| Babapour et al. | Forest road profile optimization using meta-heuristic techniques | |
| CN109948198B (en) | Surrounding rock grading reliability evaluation method based on nonlinear function | |
| CN110322118A (en) | Geological disaster space distribution rule and assessment of easy generation method | |
| CN113570226A (en) | Method for evaluating occurrence probability grade of tunnel water inrush disaster in fault fracture zone | |
| CN112214867A (en) | Strip mine mining boundary and mining program collaborative optimization method under complex coal seam condition | |
| CN103984788A (en) | Automatic intelligent design and optimization system for anchor bolt support of coal tunnel | |
| CN112487695A (en) | Multi-target intelligent comprehensive line selection method for railway in complex environment | |
| CN106991509A (en) | Log Forecasting Methodology based on radial basis function neural network model | |
| CN111882244A (en) | Construction method of multi-source homeland development risk assessment system based on hierarchical framework | |
| Ghajar et al. | An adaptive network-based fuzzy inference system for rock share estimation in forest road construction | |
| CN104361255A (en) | Simulation method for urban expansion through modified cellular automaton | |
| CN107766999A (en) | A kind of integrated evaluating method of sandrock-type uranium deposit utilization | |
| Li et al. | An engineering site suitability index (ESSI) for the evaluation of geological situations based on a multi-factor interaction matrix | |
| CN117745466B (en) | Tailing pond counting intelligent operation and maintenance system | |
| CN115049124A (en) | Deep and long tunnel water inrush prediction method based on Bayesian network | |
| CN105891442B (en) | A kind of soil texture granule content Forecasting Methodology | |
| CN104533406A (en) | Method for determining petroleum entrapment probability | |
| CN112668771B (en) | Comprehensive evaluation site selection method and system for submarine tunnel | |
| CN103559389A (en) | County cultivated land level change monitoring sampling point layout method | |
| Cui et al. | An optimized neural network with a hybrid GA-ResNN training algorithm: applications in foundation pit |
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 |