CN110084518A - A kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model - Google Patents
A kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model Download PDFInfo
- Publication number
- CN110084518A CN110084518A CN201910354795.9A CN201910354795A CN110084518A CN 110084518 A CN110084518 A CN 110084518A CN 201910354795 A CN201910354795 A CN 201910354795A CN 110084518 A CN110084518 A CN 110084518A
- Authority
- CN
- China
- Prior art keywords
- grade
- rule
- slope
- input
- slope stability
- 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.)
- Pending
Links
- 238000013461 design Methods 0.000 title claims abstract description 13
- 238000011156 evaluation Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000004913 activation Effects 0.000 claims abstract description 6
- 238000005065 mining Methods 0.000 claims description 15
- 230000001373 regressive effect Effects 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 10
- 230000014509 gene expression Effects 0.000 claims description 7
- 235000007164 Oryza sativa Nutrition 0.000 claims description 4
- 235000009566 rice Nutrition 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 240000007594 Oryza sativa Species 0.000 claims 1
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 238000010276 construction Methods 0.000 claims 1
- 238000003786 synthesis reaction Methods 0.000 claims 1
- 230000006798 recombination Effects 0.000 abstract description 5
- 238000005215 recombination Methods 0.000 abstract description 5
- 238000011160 research Methods 0.000 description 7
- 241000209094 Oryza Species 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000013097 stability assessment Methods 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/041—Abduction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Animal Husbandry (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Agronomy & Crop Science (AREA)
- Educational Administration (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of slope stability evaluation methods based on Mixed design confidence rule-based reasoning model.The present invention utilizes the non-linear relation model between confidence rule base model foundation input factor and output safety coefficient level, it is excessive that this method is directed to former input pointer type, the problems such as data volume macrooperation time is long, the discrete input of slope stability model is encoded, recombination, the latitude that mode input can be greatly reduced reduces the operation time of data, improves the operational efficiency of system.Then the data after coding are subjected to grade classification, calculate each data and corresponds to the matching degree of its reference value and calculate activation weight, merged strictly all rules are consequent, finally obtain slope stability evaluation result.
Description
Technical field
The present invention relates to a kind of slope stability evaluation methods based on Mixed design confidence rule-based reasoning model, belong to side
Slope engineering field.
Background technique
Method for Slope Stability Analysis is important one of the research direction in slope project field, it is stability of slope Journal of Sex Research
Basis.Analysis of Slope Stability process generally comprises following steps: practical slope monitoring-mechanical model building-mathematical modulo
Type building-stable calculation-Stability Judgement, core therein be the building of mechanical model, the building of mathematical model and
The research of calculation method, the i.e. research of Method for Slope Stability Analysis.It is analyzed, can be slapped in time by Slope Stability
Slope stability variation tendency is held, the destruction being likely to occur is accomplished to find in advance, takes appropriate measures and prevents and treats early, with
Avoid unnecessary loss.
Since there is side slope complicated boundary condition and geographical environment to cause slope project such as the heterogeneity of Rock And Soil
The features such as non-linear of problem.Its stability to be related to face very wide, and degree of stability variation is extremely complex, can be construed as
The stability assessment problem of one complication system.The main research including rock mass mechanics environmental condition of the research of slope stability,
Deformation & damage system research and stable calculation analysis.Analysis of Slope Stability often uses the concept of safety coefficient, that is, is
The various indexs of analyzing influence safety coefficient of uniting variation, and establish the nonlinear dependence between these indexs and safety coefficient grade
System, the safety coefficient of the dynamic identification side slope of variation by inputting each index value, thus Slope Stability give it is objective
Evaluation.
Summary of the invention
The slope stability for the confidence rule based on Mixed design that in view of the deficiencies of the prior art, the present invention proposes a kind of
Evaluation method.
The present invention is directed to the problems such as former input pointer type is excessive, and the data volume macrooperation time is long, by slope stability mould
The discrete input of type is encoded, and recombination can greatly reduce the latitude of mode input, reduce the operation time of data, mention
The high operational efficiency of system.Confidence rule base discrete, under continuous type index Mixed design is constructed based on this, is pushed away by rule
Reason generates slope stability and judges corresponding output.
The present invention the following steps are included:
(1) determine that the index for differentiating that mining area slope stability rank is judged mainly has slope angle (x1), slope height
(x2), mining Intrusion Index (x3), rainfall intensity (y1), stability space interpolation coefficient of colligation (y2), Geological Structure Effect
Degree (y3)；Wherein x1, x2, x3For serial number type index, y1, y2, y3For discrete values type index；Slope angle (x1) range be
[0,80], unit are degree, slope height (x2) range be [0,500], unit is rice, mining Intrusion Index (x3) take
Being worth range is [0,1]；Rainfall intensity y1∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate rainfall intensity grade 1, grade 2,
Grade 3, class 4, are grade 1 when the rainfall of one day (24 hours) is less than 10 millimeters, and one day rainfall is more than or equal to 10
Millimeter and be grade 2 when less than 25 millimeters, one day rainfall is more than or equal to 25 millimeters and less than 50 millimeters when is grade 3, and one
It rainfall is class 4 when being more than or equal to 50 millimeters；Stability space interpolation coefficient of colligation y2∈ { 1,2,3,4 }, wherein 1,
2,3,4 grade 1, grade 2, grade 3, class 4 for respectively indicating stability space interpolation coefficient of colligation, when Regressive method body and avalanche
Be grade 1 when face is less than two, Regressive method body and avalanche face are more than or equal to two and less than 6 when is grade 2, Regressive method body and
Avalanche face is more than or equal to 6 and is grade 3 when less than 10, and Regressive method body and avalanche face are class 4 when being more than or equal to 10；Ground
Texture makes influence degree y3∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate Geological Structure Effect degree grade 1, grade 2,
Grade 3, class 4, when tectonic movement is faint, only a small amount of small-sized fracture when be grade 1, it is only small when tectonic movement is not strong
It is grade 2 when type is broken, when tectonic movement is strong, large-scale fracture belt is grade 3 when being broken more intensive, when tectonic movement is strong
Strong, huge fracture belt is class 4 when being broken intensive.
(2) ψ=(y is defined1y2y3) it is discrete indicator vector, the value of every dimension is all only each comprising 1,2,3,4 this four expressions
From the grade of influence degree, therefore the array that can be formed regards three quaternary numbers, i.e. ψ as1=(111), ψ2=
(112), ψ64=(444)；Then decimal number x is converted by this quaternary number by formula (1)4
x4=(y1-1)×4^2+(y2-1)×4+y3 (1)
So each group of quaternary number can the expression of the decimal number as corresponding to only one phase therewith.
(3) rule base being made of S rule is constructed, for modeling slope angle (x1), slope height (x2), mining shadow
Snap number (x3) and coding after integrated impact index (x4) and slope stability differentiation result Zj(j=1,2,3,4) between
Nonlinear Mapping relationship, the nth rule G in rule basenIt can be described as:
GnIf: input x1It isAnd x2It isAnd x3It isAnd x4It isSo
{(Z1,η1,n),(Z2,η2,n),(Z3,η3,n),(Z4,η4,n)} (2)
Wherein x1~x3For consecutive variations amount, x4It is Discrete Change amount, so rule input is continuous, discrete value mixing
It inputs, here xi(i=1,2,3,4) indicates i-th of input variable of rule,It indicates in nth rule
In i-th of input variable reference grade, and haveMeet herein with reference to valueInput variable x1、x2、x3And x4The number of reference grade be respectively v1、v2、v3With
v4, amount to and generate S=v1×v2×v3×v4Rule.
GnConsequent Z1、Z2、Z3、Z4Respectively indicate grade 1, grade 2, grade 3 and grade that slope stability differentiates result
4；ηj,n(j=1,2,3,4；N=1,2 ..., S) it is to distribute to Zj(j=1,2,3,4) confidence level.
(4) X={ x is enabled1,x2,x3,x4It is the input sample obtained online, slope stability is generated by rule-based reasoning and is commented
Sentence corresponding output, the specific steps are as follows:
(4-1) calculates xiThe matching degree of corresponding reference value, detailed process is as follows:
(a) whenOrWhen, xiIt is rightOrMatching degreeValue is 1, for other references etc.
The matching degree of grade is 0；
(b) whenWhen, xiForWithMatching degreeValue is given by formula (3) and (4) respectively
Out, q=1,2 ..., vi- 1:
At this point, input variable xiMatching degree for other reference grades is 0；
Activation weight φ of the sample of (4-2) input to nth rulenIt can be calculate by the following formula:
Wherein, φn∈ [0,1], n=1,2 ..., S；
(4-3) is obtaining the regular regular weight φ that is activatednAfterwards, it merges strictly all rules are consequent, obtains input X
={ x1,x2,x3,x4Correspond to output slope stability ZjConfidence levelFormula is as follows:
(5) it is maximum to find valueIts corresponding ZjThe slope stability grade as determined.
Beneficial effects of the present invention: the slope stability judge side based on Mixed design confidence rule-based reasoning model of proposition
Method, mainly for the processing of the discrete input quantity in the mixed input system that slope stability is judged, by the latitude of discrete input
Reduced, the latitude of whole system input is effectively reduced, reduces its operation time.
Detailed description of the invention
Fig. 1 is the overall block flow diagram of the method for the present invention.
Fig. 2 is the specific implementation process figure of the method for the present invention.
Specific embodiment
The present invention designs a kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model, including with
Under each step:
(1) determine that the index for differentiating that mining area slope stability rank is judged mainly has slope angle (x1), slope height
(x2), mining Intrusion Index (x3), rainfall intensity (y1), stability space interpolation coefficient of colligation (y2), Geological Structure Effect
Degree (y3)；Wherein x1, x2, x3For serial number type index, y1, y2, y3For discrete values type index；Slope angle (x1) range be
[0,80], unit are degree, slope height (x2) range be [0,500], unit is rice, mining Intrusion Index (x3) take
Being worth range is [0,1]；Rainfall intensity y1∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate rainfall intensity grade 1, grade 2,
Grade 3, class 4, are grade 1 when the rainfall of one day (24 hours) is less than 10 millimeters, and one day rainfall is more than or equal to 10
Millimeter and be grade 2 when less than 25 millimeters, one day rainfall is more than or equal to 25 millimeters and less than 50 millimeters when is grade 3, and one
It rainfall is class 4 when being more than or equal to 50 millimeters；Stability space interpolation coefficient of colligation y2∈ { 1,2,3,4 }, wherein 1,
2,3,4 grade 1, grade 2, grade 3, class 4 for respectively indicating stability space interpolation coefficient of colligation, when Regressive method body and avalanche
Be grade 1 when face is less than two, Regressive method body and avalanche face are more than or equal to two and less than 6 when is grade 2, Regressive method body and
Avalanche face is more than or equal to 6 and is grade 3 when less than 10, and Regressive method body and avalanche face are class 4 when being more than or equal to 10；Ground
Texture makes influence degree y3∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate Geological Structure Effect degree grade 1, grade 2,
Grade 3, class 4, when tectonic movement is faint, only a small amount of small-sized fracture when be grade 1, it is only small when tectonic movement is not strong
It is grade 2 when type is broken, when tectonic movement is strong, large-scale fracture belt is grade 3 when being broken more intensive, and tectonic movement is strong,
Huge fracture belt is class 4 when being broken intensive.
(2) ψ=(y is defined1y2y3) it is discrete indicator vector, the value of every dimension is all only each comprising 1,2,3,4 this four expressions
From the grade of influence degree, therefore the array that can be formed regards three quaternary numbers, i.e. ψ as1=(111), ψ2=
(112), ψ64=(444)；Then decimal number x is converted by this quaternary number by formula (1)4
x4=(y1-1)×4^2+(y2-1)×4+y3 (1)
So each group of quaternary number can the expression of the decimal number as corresponding to only one phase therewith.
For ease of understanding, illustrate how the discrete input for slope stability being judged using the formula (1) in step (2) herein
It encoded, recombinated.For rainfall intensity (y1), stability space interpolation coefficient of colligation (y2), Geological Structure Effect degree
(y3), due to these three influence slope stabilities judge discrete indicators all contain only 1,2,3,4 this four indicate its respectively shadow
The grade of the degree of sound then can produce 64 kinds of permutation and combination altogether, can be recombinated these permutation and combination using formula (1), therefore
Three discrete indicators that slope stability can be will affect are reduced to the discrete indicator of a combined influence, as shown in table 1 below
1 three discrete indicators of table table corresponding with the combined influence index after recombination
(3) rule base being made of S rule is constructed, for modeling slope angle (x1), slope height (x2), mining shadow
Snap number (x3) and coding after integrated impact index (x4) and slope stability differentiation result Zj(j=1,2,3,4) between
Nonlinear Mapping relationship, the nth rule G in rule basenIt can be described as:
GnIf: input x1It isAnd x2It isAnd x3It isAnd x4It isSo
{(Z1,η1,n),(Z2,η2,n),(Z3,η3,n),(Z4,η4,n)} (2)
Wherein x1~x3For consecutive variations amount, x4It is Discrete Change amount, so rule input is continuous, discrete value mixing
It inputs, here xi(i=1,2,3,4) indicates i-th of input variable of rule,It indicates in nth rule
In i-th of input variable reference grade, and haveMeet herein with reference to valueInput variable x1、x2、x3And x4The number of reference grade be respectively v1、v2、v3With
v4, amount to and generate S=v1×v2×v3×v4Rule.
GnConsequent Z1、Z2、Z3、Z4Respectively indicate grade 1, grade 2, grade 3 and grade that slope stability differentiates result
4；ηj,n(j=1,2,3,4；N=1,2 ..., S) it is to distribute to Zj(j=1,2,3,4) confidence level.
(4) X={ x is enabled1,x2,x3,x4It is the input sample obtained online, slope stability is generated by rule-based reasoning and is commented
Sentence corresponding output, the specific steps are as follows:
(4-1) calculates xiThe matching degree of corresponding reference value, detailed process is as follows:
(a) whenOrWhen, xiIt is rightOrMatching degreeValue is 1, for other references etc.
The matching degree of grade is 0；
(b) whenWhen, xiForWithMatching degreeValue is given by formula (3) and (4) respectively
Out, q=1,2 ..., vi- 1:
At this point, input variable xiMatching degree for other reference grades is 0；
Activation weight φ of the sample of (4-2) input to nth rulenIt can be calculate by the following formula:
Wherein, φn∈ [0,1], n=1,2 ..., S；
(4-3) is obtaining the regular regular weight φ that is activatednAfterwards, it merges strictly all rules are consequent, obtains input X
={ x1,x2,x3,x4Correspond to output slope stability ZjConfidence levelFormula is as follows:
(5) it is maximum to find valueIts corresponding ZjThe slope stability grade as determined.
For ease of understanding, it illustrates how to utilize regular row reasoning of the formula (6) to being activated in step (4-3) herein
Fusion, it is assumed that the slope stability evaluation method of Mixed design confidence rule-based reasoning model be one four input one output be
System, and the input/output referencing value of model is provided that
The semantic values and reference value of the input of table 2 and output
S, NS, PM and M respectively represent " small ", " less than normal ", " bigger than normal " and " big " in the semantic values of table 1.
Assuming that initial data by step (2) processing after mode input data be X=14.3,89.1,0.32,
21 }, corresponding reference value section is respectively [0,22], [0,210], [0.22,0.49] and [19,43].By in step (4-1)
Formula (3) and formula (4) known to 16 rules that have activated in rule base be respectively the 6th rule S and S and NS and NS, the 7th article
Regular S and S and NS and PM, the 10th rule S and S and PM and NS, Sub_clause 11 rule S and S and PM and PM, the 22nd rule S and
NS and NS and NS, the 23rd rule S and NS and NS and PM, the 26th rule S and NS and PM and NS, the 27th rule S and NS and
PM and PM, the 70th rule NS and S and NS and NS, the 71st rule NS and S and NS and PM, the 74th rule NS and S and PM and
NS, the 75th rule NS and S and PM and PM, the 86th rule NS and NS and NS and NS, the 87th rule NS and NS and NS and PM,
90th rule NS and NS and PM and NS, the 91st rule NS and NS and PM and PM.
The weight that each rule being activated can be calculated by the formula (5) in step (4-2) is respectively φ6=0.1142,
φ7=0.0104, φ10=0.0698, φ11=0.0063, φ22=0.0841, φ23=0.0076, φ26=0.0514, φ27
=0.0047, φ70=0.2134, φ71=0.0194, φ74=0.1305, φ75=0.0119, φ86=0.1571, φ87=
0.0143, φ90=0.0961, φ91=0.0087.It is maximum by the activation weight of visible 70th rule of data, so intuitively
Can sample estimates data closest to the 70th rule.
The confidence level of fused slope stability output can be directly calculated with the formula (6) in step (4-3):It can be seen thatValue it is maximum, therefore can determine whether out defeated at this
Under entering, slope stability evaluation result is grade 2.
Below in conjunction with attached drawing, specific implementation step of the invention is discussed in detail:
The flow chart of the method for the present invention is as shown in Fig. 2, its core is: being directed to the Mixed design system of slope stability judge
Discrete values type index is encoded, is recombinated, it is defeated that whole system can be effectively reduced in this way by the processing of discrete input quantity in system
The latitude entered reduces its operation time.
1, the index and its output for influencing that slope stability rank is judged are determined
The index for differentiating that slope stability rank in mining area is judged has slope angle (x1), slope height (x2), mining influence refers to
Number (x3), rainfall intensity (y1), stability space interpolation coefficient of colligation (y2), Geological Structure Effect degree (y3)；Slope angle (x1)
Range is [0,80], and unit is degree, slope height (x2) range be [0,500], unit is rice, mining Intrusion Index
(x3) value range be [0,1]；Rainfall intensity, stability space interpolation coefficient of colligation, Geological Structure Effect degree be from
Dissipate numeric type influence factor and value be all 1,2,3,4 this four indicate that it influences the amount of grade.Set four expression side slopes
The parameter of stability grade be respectively grade 1, grade 2, grade 3, class 4 successively indicate to stablize, secondary stabilization, danger, murther.
2, the coding of the discrete values type index of slope stability rank is influenced
Formula (1) according to the method for the present invention the step of (2) carries out the discrete values type index in original sampling data
Coding, recombination, and then obtain the input of confidence rule-based reasoning.
Assuming that raw sample data is { 14.3,89.1,0.32,2,2,1 }, rear three data ψ in this six data=
(2 2 1) it is discrete values type index, therefore is substituted into formula (1): x4=(2-1) × 4^2+ (2-1) × 4+1 can acquire x4=
21.These three discrete type numerical indications are thus converted into the energy corresponding decimal number of only one, this
Decimal number, which is that treated, influences the overall target that slope stability is judged.
3, rule base is constructed
In order to make it easy to understand, determining the reference value variation range of each input and output amount in conjunction with model above；It is wherein defeated
Enter the variation range of reference value:Export the change of reference value
Change range: Z1=1, Z2=2, Z3=3, Z4=4；The input quantity of rule base is respectively provided with v1=v2=v3=v4=4 reference points,
The reference value (semantic values) that each input variable and output variable is specifically arranged is as shown in table 4.
S, NS, PM and M respectively represent " small ", " less than normal ", " bigger than normal " and " big " in the semantic values of table 3.
The semantic values and reference value of the input of table 3 and output
In turn, the nth rule G in confidence rule-based reasoning model can be providednIt can be described as:
GnIf: input x1It isAnd x2It isAnd x3It isAnd x4It is
So { (Z1,η1,n),(Z2,η2,n),(Z3,η3,n),(Z4,η4,n)}
Then amounting to can produce S=v1×v2×v3×v4=256 rules.Confidence rule-based reasoning model is given in table 4
Part rule, reliability assignment η thereinj,nFor initial value.
4 confidence rule-based reasoning model part rule of table
4, sample data X={ x1,x2,x3,x4, after rule-based reasoning corresponding output can be judged with slope stability
In order to make it easy to understand, initial data is in the new number generated after coding, recombination also for above providing
Input according to X={ 14.3,89.1,0.32,21 } as confidence rule-based reasoning model.By the formula (3) and formula in step (4-1)
(4) 16 rules for having activated in rule base known to be respectively the 6th rule S and S and NS and NS, the 7th rule S and S and
NS and PM, the 10th rule S and S and PM and NS, Sub_clause 11 rule S and S and PM and PM, the 22nd rule S and NS and NS and NS,
23rd rule S and NS and NS and PM, the 26th rule S and NS and PM and NS, the 27th rule S and NS and PM and PM, the 70th
Rule NS and S and NS and NS, the 71st rule NS and S and NS and PM, the 74th rule NS and S and PM and NS, the 75th rules and regulations
Then NS and S and PM and PM, the 86th rule NS and NS and NS and NS, the 87th rule NS and NS and NS and PM, the 90th rule
NS and NS and PM and NS, the 91st rule NS and NS and PM and PM.
The weight that each rule being activated can be calculated by the formula (5) in step (4-2) is respectively φ6=0.1142,
φ7=0.0104, φ10=0.0698, φ11=0.0063, φ22=0.0841, φ23=0.0076, φ26=0.0514, φ27
=0.0047, φ70=0.2134, φ71=0.0194, φ74=0.1305, φ75=0.0119, φ86=0.1571, φ87=
0.0143, φ90=0.0961, φ91=0.0087.It is maximum by the activation weight of visible 70th rule of data, so intuitively
Can sample estimates data closest to the 70th rule.
Assuming that the confidence structure of consequent attribute corresponding with the rule being activated are as follows:
The confidence structure for the corresponding consequent attribute of rule that table 5 is activated
The confidence level of fused slope stability output can be directly calculated with the formula (6) in step (4-3):It can be seen thatValue it is maximum, therefore can determine whether out defeated at this
Under entering, slope stability evaluation result is grade 2.
Claims (1)
1. a kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model, it is characterised in that this method packet
Include following steps:
(1) index judged for differentiating mining area slope stability rank, including slope angle x are determined1, slope height x2, mining
Intrusion Index x3, rainfall intensity y1, stability space interpolation coefficient of colligation y2With Geological Structure Effect degree y3；Wherein x1, x2, x3
For serial number type index, y1, y2, y3For discrete values type index；
Slope angle x1Range be [0,80], unit be degree, slope height x2Range be [0,500], unit is rice, mining
Intrusion Index x3Value range be [0,1]；
Rainfall intensity y1∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate the grade 1, grade 2, grade 3, grade of rainfall intensity
4, it is grade 1 when one day rainfall is less than 10 millimeters, one day rainfall is more than or equal to 10 millimeters and when less than 25 millimeters
For grade 2, one day rainfall is more than or equal to 25 millimeters and less than 50 millimeters when is grade 3, and one day rainfall is more than or equal to
It is class 4 at 50 millimeters；
Stability space interpolation coefficient of colligation y2∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate stability space interpolation synthesis
Grade 1, grade 2, the grade 3, class 4 of coefficient, are grade 1 when Regressive method body and avalanche face are less than two, Regressive method body and
Avalanche face is more than or equal to two and is grade 2 when less than 6, and Regressive method body and avalanche face are more than or equal to 6 and when less than 10
For grade 3, Regressive method body and avalanche face are class 4 when being more than or equal to 10；
Geological Structure Effect degree y3∈ { 1,2,3,4 }, wherein 1,2,3,4 respectively indicate Geological Structure Effect degree grade 1,
Grade 2, grade 3, class 4, when tectonic movement is faint, only a small amount of small-sized fracture when be grade 1, when tectonic movement is not strong,
Only small-sized fracture when be grade 2, when tectonic movement is strong, large-scale fracture belt is grade 3 when being broken more intensive, works as construction
Movement is strong, huge fracture belt, is class 4 when being broken intensive；
(2) ψ=(y is defined1 y2 y3) it is discrete indicator vector, the value of every dimension is all only respective comprising 1,2,3,4 this four expressions
The grade of influence degree, therefore the array that can be formed regards three quaternary numbers, i.e. ψ as1=(1 1 1), ψ2=(1 1
2), ψ64=(4 4 4)；Then decimal number x is converted by this quaternary number by formula (1)4
x4=(y1-1)×4^2+(y2-1)×4+y3 (1)
So each group of quaternary number can the expression of the decimal number as corresponding to only one phase therewith；
(3) rule base being made of S rule is constructed, for modeling slope angle x1, slope height x2, mining Intrusion Index x3
And the integrated impact index x after coding4Result Z is differentiated with slope stabilityjBetween Nonlinear Mapping relationship, j=1,2,3,
4, the nth rule G in rule basenDescription are as follows:
GnIf: input x1It isAnd x2It isAnd x3It isAnd x4It isSo
{(Z1,η1,n),(Z2,η2,n),(Z3,η3,n),(Z4,η4,n)} (2)
Wherein x1~x3For consecutive variations amount, x4It is Discrete Change amount, so rule input is continuous, discrete value Mixed design,
Here xiIndicate i-th of input variable of rule,Indicate the reference grade of i-th of input variable in nth rule, and
HaveMeet herein with reference to valueInput variable x1、x2、
x3And x4The number of reference grade be respectively v1、v2、v3And v4, amount to and generate S=v1×v2×v3×v4Rule；
GnConsequent Z1、Z2、Z3、Z4Respectively indicate grade 1, grade 2, grade 3 and class 4 that slope stability differentiates result；ηj,n
To distribute to ZjConfidence level；
(4) X={ x is enabled1,x2,x3,x4It is the input sample obtained online, slope stability judge pair is generated by rule-based reasoning
The output answered, the specific steps are as follows:
(4-1) calculates xiThe matching degree of corresponding reference value, detailed process is as follows:
(a) whenOrWhen, xiIt is rightOrMatching degreeValue is 1, for other reference grades
Matching degree is 0；
(b) whenWhen, xiForWithMatching degreeValue is provided by formula (3) and (4) respectively, q
=1,2 ..., vi- 1:
At this point, input variable xiMatching degree for other reference grades is 0；
Activation weight φ of the sample of (4-2) input to nth rulenIt is calculate by the following formula:
Wherein, φn∈ [0,1], n=1,2 ..., S；
(4-3) is obtaining the regular regular weight φ that is activatednAfterwards, it merges strictly all rules are consequent, obtains input X=
{x1,x2,x3,x4Correspond to output slope stability ZjConfidence levelFormula is as follows:
(5) it is maximum to find valueIts corresponding ZjThe slope stability grade as determined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910354795.9A CN110084518A (en) | 2019-04-29 | 2019-04-29 | A kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910354795.9A CN110084518A (en) | 2019-04-29 | 2019-04-29 | A kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110084518A true CN110084518A (en) | 2019-08-02 |
Family
ID=67417694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910354795.9A Pending CN110084518A (en) | 2019-04-29 | 2019-04-29 | A kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084518A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110850206A (en) * | 2019-11-13 | 2020-02-28 | 武汉理工大学 | Inverter capacitor aging fault diagnosis method based on confidence rule reasoning |
| CN112070399A (en) * | 2020-09-09 | 2020-12-11 | 中国人民解放军火箭军工程大学 | Large-scale engineering structure safety risk assessment method and system |
| CN113034855A (en) * | 2021-03-09 | 2021-06-25 | 杭州电子科技大学 | Slope landslide early warning method based on NPR cable slip force monitoring |
| CN114140987A (en) * | 2021-11-05 | 2022-03-04 | 绍兴文理学院 | Landslide early warning method based on confidence rule base |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101538861A (en) * | 2009-04-21 | 2009-09-23 | 中国科学院武汉岩土力学研究所 | Highway slope stability grading evaluation method |
| CN103955580A (en) * | 2014-05-04 | 2014-07-30 | 杭州电子科技大学 | Integrated circuit parameter yield estimation method based on BRB (Belief Rule Base) ratiocination |
| CN105044774A (en) * | 2015-08-26 | 2015-11-11 | 长安大学 | Side slope stability prediction method under earthquake effect |
| CN105139086A (en) * | 2015-08-13 | 2015-12-09 | 杭州电子科技大学 | Track profile irregularity amplitude estimation method employing optimal belief rules based inference |
| CN106059412A (en) * | 2016-01-28 | 2016-10-26 | 杭州电子科技大学 | Method for controlling rotating speed of separately excited DC motor based on belief rule base reasoning |
| CN106597840A (en) * | 2017-01-16 | 2017-04-26 | 杭州电子科技大学 | PID parameter setting method based on production rule reasoning |
| CN107123058A (en) * | 2017-05-26 | 2017-09-01 | 辽宁工程技术大学 | A kind of Method for Slope Stability Analysis |
| CN108982096A (en) * | 2018-06-01 | 2018-12-11 | 杭州电子科技大学 | Industrial robot crank axle wear detecting method based on heuristic rule system |
| CN109507876A (en) * | 2019-01-25 | 2019-03-22 | 杭州电子科技大学 | A kind of electricity based on reliability reasoning pushes away marine electrical motors pid parameter setting method |
-
2019
- 2019-04-29 CN CN201910354795.9A patent/CN110084518A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101538861A (en) * | 2009-04-21 | 2009-09-23 | 中国科学院武汉岩土力学研究所 | Highway slope stability grading evaluation method |
| CN103955580A (en) * | 2014-05-04 | 2014-07-30 | 杭州电子科技大学 | Integrated circuit parameter yield estimation method based on BRB (Belief Rule Base) ratiocination |
| CN105139086A (en) * | 2015-08-13 | 2015-12-09 | 杭州电子科技大学 | Track profile irregularity amplitude estimation method employing optimal belief rules based inference |
| CN105044774A (en) * | 2015-08-26 | 2015-11-11 | 长安大学 | Side slope stability prediction method under earthquake effect |
| CN106059412A (en) * | 2016-01-28 | 2016-10-26 | 杭州电子科技大学 | Method for controlling rotating speed of separately excited DC motor based on belief rule base reasoning |
| CN106597840A (en) * | 2017-01-16 | 2017-04-26 | 杭州电子科技大学 | PID parameter setting method based on production rule reasoning |
| CN107123058A (en) * | 2017-05-26 | 2017-09-01 | 辽宁工程技术大学 | A kind of Method for Slope Stability Analysis |
| CN108982096A (en) * | 2018-06-01 | 2018-12-11 | 杭州电子科技大学 | Industrial robot crank axle wear detecting method based on heuristic rule system |
| CN109507876A (en) * | 2019-01-25 | 2019-03-22 | 杭州电子科技大学 | A kind of electricity based on reliability reasoning pushes away marine electrical motors pid parameter setting method |
Non-Patent Citations (5)
| Title |
|---|
| 刘佳俊等: "基于证据推理和置信规则库的装备寿命评估", 《控制理论与应用》 * |
| 吕琨等: "模糊综合评判在矿山边坡稳定性评价中的应用", 《河北联合大学学报（自然科学版）》 * |
| 陈汉勇: "冷冻鱼糜新评价方法的建立及混合鱼糜品质改良的研究", 《中国优秀硕士学位论文全文数据库-信息科技I辑》 * |
| 陶志刚等: "南芬露天铁矿高陡边坡危险性区划研究", 《矿冶工程》 * |
| 陶志刚等: "耦合多种成灾因素下的边坡稳定性分析", 《地下空间与工程学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110850206A (en) * | 2019-11-13 | 2020-02-28 | 武汉理工大学 | Inverter capacitor aging fault diagnosis method based on confidence rule reasoning |
| CN112070399A (en) * | 2020-09-09 | 2020-12-11 | 中国人民解放军火箭军工程大学 | Large-scale engineering structure safety risk assessment method and system |
| CN112070399B (en) * | 2020-09-09 | 2024-02-13 | 中国人民解放军火箭军工程大学 | Safety risk assessment method and system for large-scale engineering structure |
| CN113034855A (en) * | 2021-03-09 | 2021-06-25 | 杭州电子科技大学 | Slope landslide early warning method based on NPR cable slip force monitoring |
| CN114140987A (en) * | 2021-11-05 | 2022-03-04 | 绍兴文理学院 | Landslide early warning method based on confidence rule base |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084518A (en) | A kind of slope stability evaluation method based on Mixed design confidence rule-based reasoning model | |
| Li et al. | An approach using random forest intelligent algorithm to construct a monitoring model for dam safety | |
| Lermusiaux et al. | Uncertainties | |
| CN110866974B (en) | Hydraulic monitoring system based on three-dimensional display | |
| CN107886161A (en) | A kind of global sensitivity analysis method for improving Complex Information System efficiency | |
| CN105574685A (en) | Subjective and objective combination-based cloud service evaluation method | |
| CN103235877B (en) | Robot control software's module partition method | |
| CN102456062B (en) | Community similarity calculation method and social network cooperation mode discovery method | |
| CN105740245A (en) | Frequent item set mining method | |
| CN105843829B (en) | A kind of big data creditability measurement method based on hierarchical mode | |
| Xia et al. | Slope stability analysis based on group decision theory and fuzzy comprehensive evaluation | |
| Shi et al. | Development of a surface modeling method for mapping soil properties | |
| CN105760649A (en) | Big-data-oriented creditability measuring method | |
| CN110516907A (en) | A kind of rock burst grade evaluation method based on AHP- entropy weight cloud model | |
| Wang et al. | Roof pressure prediction in coal mine based on grey neural network | |
| CN112529267A (en) | Advanced geological prediction method and system based on fuzzy comprehensive evaluation method | |
| CN109858134A (en) | The method of the more change system architectural model buildings of the auxiliary of big data driving | |
| Zhang et al. | Spatial variability and possible cause analysis of regional precipitation complexity based on optimized sample entropy | |
| CN110716998B (en) | Fine scale population data spatialization method | |
| CN110245370A (en) | A kind of high CFRD multiple target mechanics parameter inversion method | |
| CN106960098A (en) | It is a kind of with method of the mathematical modeling to architectural design comfortableness overall merit | |
| Li et al. | On the probabilistic distribution of wind speeds: theoretical development and comparison with data | |
| Wu | An ecological environmental carrying capacity estimation of tourist attractions based on structural equation | |
| CN105046324A (en) | Height anomaly fitting interpolation calculation method based on mobile neural network | |
| CN109255394A (en) | A kind of forced oscillation recognition methods based on Pattern similarity |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190802 |