CN101792607A - Method for selecting fire retardant based on grey correlation analysis - Google Patents

Abstract

The invention discloses a method for selecting a fire retardant based on grey correlation analysis, which mainly comprises the following steps of: designing an orthogonal test plan; setting three fire retardants; determining test indexes including six indexes including softening points, ductility, needle penetration, needle penetration indexes, equivalent softening points and equivalent breaking points of fire-retardant asphalts; determining test factors including variety of fire retardants, admixture of the fire retardants (mass percentage ratio of fire retardants to asphalts), shearing time and shearing temperature; testing each sample after three fire retardants are mixed; carrying out interval value processing on original data; solving a different sequence; solving two-stage maximum difference and minimum difference; calculating an association coefficient; calculating an association degree; arranging an association sequence; calculating proportions occupied by each factor based on the significance of the factors; comprehensively scoring; and finally optimizing a proper and optimal process for preparing the fire retardants and the fire-retardant asphalts, and testing oxygen indexes on the fire-retardant asphalts prepared by the preparation process.

Description

Selection of fire retardant method based on grey correlation analysis

Technical field

The present invention relates to a kind of method that is used for selecting the flame-retardant pitch flame-proof modifier, relate in particular to a kind of selection of fire retardant method based on Gray Association Analysis.

Background technology

At present, the flame-proof modifier that is used for adding to flame-retardant pitch has many kinds, and the method for flame-proof modifier adopts orthogonal test oxygen index method usually in the selection flame-retardant pitch, the weak point that this method exists is: because the testing standard of bituminous material flame retardant properties so far formulates yet, and the domestic bituminous material oxygen index test method more complicated of normal reference.

Summary of the invention

At above-mentioned prior art, the invention provides a kind of selection of fire retardant method based on grey correlation analysis.In general, " in vain " refers to that information knows fully, and " deceiving " refers to that information is uncertain of fully, and " ash " refers to that then message part knows, part is uncertain of, or perhaps INFORMATION OF INCOMPLETE, this be " ash " and basic meaning.Grey correlation analysis is a kind of analytical procedure that proposes in the gray system theory.It is quantitative description and the method relatively that a system development is changed situation, the comparison of developing state, Fundamentals of Mathematics according to Space Theory, according to these four principles of standardization, couple symmetry, globality and proximity, determine reference sequence (female ordered series of numbers) and some relatively incidence coefficient and relational degree between the ordered series of numbers (subnumber row).The purpose of grey correlation analysis is to seek the main relation between each factor in the system, finds out the important factor that influences target value, thereby grasps the principal character of things, and promotion and guidance system develop rapidly and effectively.

In order to solve the problems of the technologies described above, a kind of technical scheme that is achieved based on the selection of fire retardant method of grey correlation analysis of the present invention is: this system of selection may further comprise the steps:

Step 1, flame-retardant pitch orthogonal test conceptual design, its design procedure comprises: (1-1) confirmed test index: the softening temperature of flame-retardant pitch, ductility, penetration degree, penetrative index, equivalent softening temperature and six indexs of equivalent brittle point; (1-2) confirmed test factor: fire retardant kind, fire retardant volume (the fire retardant quality accounts for the per-cent of asphalt quality), shear time, shear temperature;

The index of flame-retardant pitch sample in step 2, above-mentioned each the orthogonal test scheme of test: testing program is tested, and test index comprises softening temperature, ductility and the penetration degree of this flame-retardant pitch;

Step 3, the association of utilization ash are analyzed above-mentioned flame-retardant pitch sample, thereby select suitable fire retardant, comprising:

(3-1) gray system is set:

Definition X ₀=[x ₀(1), x ₀(2), x ₀(3) ..., x ₀(n)] (1)

Formula (1) is the system features sequence;

$\left\{\begin{array}{c}{X}_1=[x_1\left(1\right),x_1\left(2\right),\·\·\·\·\·\·,x_1\left(n\right)]\\ X_2=[x_2\left(1\right),x_2\left(2\right),\·\·\·\·\·\·,x_2\left(n\right)]\\ \·\\ \·\\ \·\\ X_m=[x_m\left(1\right),x_m\left(2\right),\·\·\·\·\·\·,x_m\left(n\right)]\end{array}\right.---\left(2\right)$

Formula (2) is the correlative factor sequence; In formula (1) or the formula (2):

x ₀(n) be female factor series;

x _m(n) be sub-factor series;

Data rows according to following each index of formula (2) definition:

Volume row: x (0)={ x ₁(0), x ₂(0), x ₃(0) ..., x _n(0) }

Softening point range: x (1)={ x ₁(1), x ₂(1), x ₃(1) ... x _n(1) }

Ductility row: x (2)={ x ₁(2), x ₂(2), x ₃(2) ... x _n(2) }

(3-2) adopt interval value treatment process to carry out value just and handle, the starting point of each sequence is overlapped the test index that draws by the orthogonal test scheme; The sequence calculation formula that value is just handled is as follows:

X _i＝X _i/x _i(1)＝[x _i(1)，x _i(2)，…，x _i(n)]

(i＝0，1，2，…m)?????????????????(3)

(3-3) carry out grey incidence coefficient and grey relational degree and calculate, that is: note is { x through female ordered series of numbers of data conversion _o(t) }, subnumber is classified { x as _i(t) }, then when moment t=k, wherein:

{ x _o(k) } with { x _i(k) } incidence coefficient ξ _Oi(k) calculate with following formula (4) formula:

${\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right)=\frac{{\mathrm{\Δ}}_{\min }+{\mathrm{\ρ\Δ}}_{\max }}{{\mathrm{\Δ}}_{\mathrm{oi}}\left(k\right)+{\mathrm{\ρ\Δ}}_{\max }}---\left(4\right)$

In the formula (4): Δ _Oi(k) be the k absolute difference of two sequences, that is: Δ constantly _Oi(k)=| x _o(k)-x _i(k) |;

Δ _Min, Δ _MaxBe respectively maximum value and minimum value in each absolute difference constantly;

ρ is the explanation coefficient, and ρ ∈ (0,1), ρ get 0.1～0.5;

Relational degree r _Oi, calculate with following formula (5):

$r_{\mathrm{oi}}=\frac{1}{N}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right)---\left(5\right)$

In the formula (5): r _OiRelational degree for subsequence i and auxiliary sequence 0;

N is that the length of sequence is the data numbers;

(3-4) discharge the related preface that each achievement data is listed as according to relational degree;

Relational degree is female ordered series of numbers { x _oAnd subnumber row { x (t) } _i(t) } the measuring of cognation size between, if both change basically identicals relatively in evolution, then both relational degree are big; Otherwise both relational degree are just little;

(3-5) relational degree of calculating each index respectively accounts for the ratio of the relational degree of all indexs;

Step 4, utilize the range analysis method of orthogonal test that above-mentioned each index is carried out analysis-by-synthesis, that is:

Many index comprehensives are converted into single index, for the single index Y after transforming _iExpression, Y _iCalculation formula is as follows:

In the formula (6): z is a test piece number (Test pc No.); L is the number of index;

Step 5, according to index Y _iY draws _iWith the graph of a relation of fire retardant kind, fire retardant volume, shear time and each factor of shear temperature, thereby optimize the best preparation technology of a suitable fire retardant and flame-retardant pitch;

Step 6, the flame-retardant pitch that above-mentioned preparation technology is drawn carry out the oxygen index test.

Compared with prior art, the invention has the beneficial effects as follows:

Because the testing standard of bituminous material flame retardant properties is not formulated yet at present, and the domestic bituminous material oxygen index test method more complicated of normal reference, and the inventive method can be selected the better and reasonable fire retardant of flame retardant effect with bitumen compatibility soon under the less situation of cost.The contradiction that produces when meanwhile, the inventive method can also overcome each index and estimates separately.And can not reduce the pavement performance of pitch itself.

Description of drawings

Fig. 1 is to the different volume penetration test of three kinds of fire retardants result schematic diagram according to the orthogonal test scheme;

Fig. 2 is to the different volume softening temperature of three kinds of fire retardants degree test-results synoptic diagram according to the orthogonal test scheme;

Fig. 3 is to the different volume ductility test of three kinds of fire retardants result schematic diagram according to the orthogonal test scheme;

Fig. 4 be shown in Fig. 1 in the method to the synoptic diagram that concerns between the comprehensive grading of fire retardant and each factor;

Fig. 5 is the A that utilizes the inventive method to select ₁Fire retardant volume and oxygen index concern synoptic diagram.

Embodiment

Below in conjunction with the drawings and specific embodiments the present invention is done to describe in further detail.

In order to narrate convenience, clear, fire retardant in three kinds of flame-retardant pitchs in the present embodiment is numbered fire retardant A respectively ₁, fire retardant A ₂, fire retardant A ₃Three kinds of fire retardant technical indicators see Table 1, table 2, table 3.

Table 1.A ₁The fire retardant physical and chemical indexes

Project	Index	Project	Index
				Fire retardant effective content (%)	??70	Appearance color	The red-purple particle
Melt temperature (℃)	??＞120	Physical dimension	3mm * 5mm cylinder
				Point of ignition (℃)	??＞260	Density (g/cm 3)	??1.0～2.0

Table 2.A ₂The flame-proof modifier physical and chemical indexes

Project	Index	Project	Index
				??P 20 5(％)	??＞2.0	Appearance color	White powder
??N(％)	??＞8.0	Particle diameter	??＜10um
				??Al2O3(％)	??＞18	Density (g/cm3)	??＞2.0
??MgO(％)	??35～50	Endothermic temperature (℃)	??＞250
				??PH	??5.5～7.0	Decomposition temperature (℃)	??＞270

Table 3.A ₃The flame-proof modifier physical and chemical indexes

Project	Index	Project	Index
				??P 20 5(％)	??＞34	Appearance color	White powder
??N(％)	??＞15	Particle diameter	??＜15um
				??PH	??5.5～7.0	Density (g/cm 3)	??＞1.74
Decomposition temperature (℃)	??＞300

Adopt the preparation of flame-retardant pitch sample to finish among the present invention by high-speed shearing equipment.

Prepare the flame-retardant pitch sample by above-mentioned three kinds of fire retardants: fire retardant A ₁, fire retardant A ₂With fire retardant A ₃All be very easy to add, directly in pitch, be sprinkled into fire retardant and get final product.Above-mentioned three kinds of flame-retardant pitchs are as the sample in the embodiment of the invention.

The selection of fire retardant method that the present invention is based on grey correlation analysis is as follows:

Step 1, flame-retardant pitch orthogonal test conceptual design, its design procedure comprises:

(1-1) confirmed test index: the softening temperature of flame-retardant pitch, ductility, penetration degree, penetrative index, equivalent softening temperature and six indexs of equivalent brittle point;

(1-2) confirmed test factor: fire retardant kind, fire retardant volume (the fire retardant quality accounts for the per-cent of asphalt quality), shear time and shear temperature;

Level: each factor is all got three levels.List the horizontal factor table in view of the above, as shown in table 4.

Table 4. horizontal factor table

As can be seen from Table 4, test belongs to the orthogonal test of 4 factors, 3 levels, and according to the principle of selecting orthogonal table, promptly the number of Shi Yan level should equal the columns of orthogonal table, therefore can select L ₉(3 ⁴) orthogonal table inserts this orthogonal table to experimental factor and level and promptly form testing program, sees Table 5.

Table 5. orthogonal test scheme

The index of flame-retardant pitch sample in step 2, above-mentioned each the orthogonal test scheme of test,

Utilize orthogonal test to carry out the testing program design of flame-retardant pitch, determine and testing program is tested, test index comprises softening temperature, ductility, penetration degree, penetrative index, equivalent softening temperature and the equivalent brittle point of this flame-retardant pitch; Step 3, the association of utilization ash are analyzed above-mentioned flame-retardant pitch sample, thereby select suitable fire retardant, comprising:

(3-1) gray system is set:

Definition X ₀=[x ₀(1), x ₀(2), x ₀(3) ..., x ₀(n)] (1)

Formula (1) is the system features sequence;

$\left\{\begin{array}{c}{X}_1=[x_1\left(1\right),x_1\left(2\right),\·\·\·\·\·\·,x_1\left(n\right)]\\ X_2=[x_2\left(1\right),x_2\left(2\right),\·\·\·\·\·\·,x_2\left(n\right)]\\ \·\\ \·\\ \·\\ X_m=[x_m\left(1\right),x_m\left(2\right),\·\·\·\·\·\·,x_m\left(n\right)]\end{array}\right.---\left(2\right)$

Formula (2) is the correlative factor sequence; In formula (1) or the formula (2):

x ₀(n) be female factor series;

x _m(n) be sub-factor series;

Data rows according to following each index of formula (2) definition:

Volume row: x (0)={ x ₁(0), x ₂(0), x ₃(0), x ₄(0), x ₅(0), x ₆(0), x ₇(0), x ₈(0), x ₉(0) }

＝{5，10，15，5，10，15，5，10，15}

Softening point range: x (1)={ x ₁(1), x ₂(1), x ₃(1), x ₄(1), x ₅(1), x ₆(1), x ₇(1), x ₈(1), x ₉(1) }

＝{85，87.7，84.5，84，85.4，80.2，84.7，86.7，79.6}

Ductility row: x (2)={ x ₁(2), x ₂(2), x ₃(2), x ₄(2), x ₅(2), x ₆(2), x ₇(2), x ₈(2), x ₉(2) }

＝{40，37.4，30，31.6，24.6，20.6，32.5，20.9，16.9}

Penetrative index row: x (3)={ x ₁(3), x ₂(3), x ₃(3), x ₄(3), x ₅(3), x ₆(3), x ₇(3), x ₈(3), x ₉(3) }

＝{0.47，0.13，-0.04，-0.32，0.24，-0.03，-0.22，-0.45，-0.93}

Equivalent softens point range: x (4)={ x ₁(5), x ₂(5), x ₃(2), x ₄(5, x ₅(5), x ₆(5), x ₇(5), x ₈(5), x ₉(5) }

＝{53.38，52.5，52.2，50.47，52.6，52.1，51.23，51.6，49.5}

Equivalent brittle point row: x (5)={ x ₁(5), x ₂(5), x ₃(5), x ₄(5), x ₅(5), x ₆(5), x ₇(5), x ₈(5), x ₉(5) }

＝{-22.3，-19.5，-18.0，-16.8，-20.5，-18.2，-17.1，-14.3，-11.6}

(3-2) adopt interval value treatment process to carry out value just and handle, the starting point of each sequence is overlapped the test index that draws by the orthogonal test scheme; The sequence calculation formula that value is just handled is as follows:

X _i＝X _i/x _i(1)＝[x _i(1)，x _i(2)，…，x _i(n)]

(i＝0，1，2，…m)?????????????????????????????(3)

Because: min _ix _i(1)=x ₉(1)=79.6; Max _ix _i(1)=x ₂(1)=87.7

$y_i\left(1\right)=\frac{x_j\left(1\right)-{\min }_i{x}_i\left(1\right)}{{\max }_i{x}_i\left(1\right)-{\min }_i{x}_i\left(1\right)}=\frac{x_j\left(1\right)-79.6}{87.7-79.6}=\frac{x_j-79.6}{8.1}$

That is: $y_1\left(1\right)=\frac{x_1\left(1\right)-79.6}{8.1}=\frac{85-79.6}{8.1}=0.6667;$ $y_2\left(1\right)=\frac{x_2\left(1\right)-79.6}{8.1}=\frac{87.7-79.6}{8.1}=1;$

$y_3\left(1\right)=\frac{x_{31}\left(1\right)-79.6}{8.1}=\frac{84.5-79.6}{8.1}=0.6049;$ $y_4\left(1\right)=\frac{x_4\left(1\right)-79.6}{8.1}=\frac{84-79.6}{8.1}=0.5432;$

$y_5\left(1\right)=\frac{x_5\left(1\right)-79.6}{8.1}=\frac{85.4-79.6}{8.1}=0.7160;$ $y_6\left(1\right)=\frac{x_6\left(1\right)-79.6}{8.1}=\frac{80.2-79.6}{8.1}=0.0741;$

$y_7\left(1\right)=\frac{x_7\left(1\right)-79.6}{8.1}=\frac{84.7-79.6}{8.1}=0.6296;$ $y_8\left(1\right)=\frac{x_8\left(1\right)-79.6}{8.1}=\frac{86.7-79.6}{8.1}=0.8765;$

$y_9\left(1\right)=\frac{x_9\left(1\right)-79.6}{8}=\frac{79.6-79.6}{8}=0.$

Thus the vertically interval value softening temperature ordered series of numbers after handling:

y(1)＝{y ₁(1)，y ₂(1)，y ₃(1)，y ₄(1)，y ₅(1)，y ₆(1)，y ₇(1)，y ₈(1)，y ₉(1)}

＝{0.6667，1，0.6049，0.5432，0.7160，0.0741，0.6296，0.8765，0}

Same method can be obtained other ordered series of numbers after interval value is handled, and is as shown in table 6.

Ordered series of numbers after table 6. interval value is handled gathers

??y(0)	??y(1)	??y(2)	??y(3)	??y(4)	??y(5)
						??0	??0.6667	??1	??1	??1	??0
??0.5	??1	??0.9721	??0.7579	??0.7789	??0.2672
						??1	??0.6049	??0.6279	??0.6364	??0.6923	??0.4032
??0	??0.5432	??0.7023	??0.4338	??0.24113	??0.5201
						??0.5	??0.7160	??0.3767	??0.8346	??0.80243	??0.1672
??1	??0.0741	??0.1906	??0.6452	??0.6629	??0.3829
						??0	??0.6296	??0.7442	??0.5104	??0.4399	??0.4879

??y(0)	??y(1)	??y(2)	??y(3)	??y(4)	??y(5)
						??0.5	??0.8765	??0.2046	??0.3429	??0.5416	??0.7494
??1	??0	??0	??0	??0	??1

(3-3) carrying out grey incidence coefficient and grey relational degree calculates;

At first, ask difference sequence:

Δ _0j＝|y ₀(k)-y _j(k)|；j＝1，2，3，4，5；k＝1，2，…，9

That is: Δ ₀₁=(Δ ₀₁(1), Δ ₀₁(2), Δ ₀₁(3), Δ ₀₁(4), Δ ₀₁(5), Δ ₀₁(6), Δ ₀₁(7), Δ ₀₁(8), Δ ₀₁(9))

＝(0.6667，0.5，0.3951，0.5432，0.2160，0.9259，0.6296，0.3765)

Same method can be tried to achieve difference sequence as shown in table 7.

Table 7. difference sequence gathers

??Δ 01	??Δ 02	??Δ 03	??Δ 04	??Δ 05
					??0.6667	??1	??1	??1	??0
??0.5000	??0.4721	??0.2579	??0.2790	??0.2328
					??0.3951	??0.3721	??0.3636	??0.3078	??0.5967
??0.5432	??0.7023	??0.4338	??0.2411	??0.5201
					??0.2160	??0.1233	??0.3346	??0.3024	??0.3328
??0.9259	??0.8093	??0.3548	??0.3371	??0.6171
					??0.6296	??0.7442	??0.5104	??0.4400	??0.4879
??0.3765	??0.2953	??0.1571	??0.0416	??0.2494
					??1	??1	??1	??1	??0

Ask two-stage maximum difference and lowest difference:

Because: max _kΔ ₀₁=1; Min _kΔ ₀₁=0.2160;

max _kΔ ₀₂＝1；min _kΔ ₀₁＝0.1233；

max _kΔ ₀₃＝1；min _kΔ ₀₁＝0.1571；

max _kΔ ₀₄＝1；min _kΔ ₀₁＝0.0416；

max _kΔ ₀₅＝0.6171；min _kΔ ₀₁＝0

So: Δ _Max=max _jMax _k| Δ _0i|=1; Δ _Min=min _jMin _k| Δ _0i|=0

Note is { x through female ordered series of numbers of data conversion _o(t) }, subnumber is classified { x as _i(t) }, then when moment t=k, wherein:

{ x _o(k) } with { x _i(k) } incidence coefficient ξ _Oi(k) calculate with following formula (4) formula:

${\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right)=\frac{{\mathrm{\Δ}}_{\min }+{\mathrm{\ρ\Δ}}_{\max }}{{\mathrm{\Δ}}_{\mathrm{oi}}\left(k\right)+{\mathrm{\ρ\Δ}}_{\max }}---\left(4\right)$

In the formula (4): Δ _Oi(k) be the k absolute difference of two sequences, that is: Δ constantly _Oi(k)=| x _o(k)-x _i(k) |;

Δ _Min, Δ _MaxBe respectively maximum value and minimum value in each absolute difference constantly;

ρ is explanation coefficient, ρ ∈ (0,1), ρ=0.5;

Relational degree r _Oi, calculate with following formula (5):

$r_{\mathrm{oi}}=\frac{1}{N}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right)---\left(5\right)$

In the formula (5): r _OiRelational degree for subsequence i and auxiliary sequence 0;

N is that the length of sequence is the data numbers;

The compute associations coefficient:

According to:

Get ρ=0.5,

It is as shown in table 8 to get incidence coefficient.

Five index incidence coefficients of table 8. calculation result

The compute associations degree

According to: $r_{\mathrm{oi}}=\frac{1}{N}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right),$

Then:

$r_{o1}=\frac{1}{9}\underset{K=1}{\overset{9}{\mathrm{\Σ}}}{\mathrm{\ξ}}_{o1}\left(k\right)=\frac{1}{9}({\mathrm{\ξ}}_{01}\left(1\right)+{\mathrm{\ξ}}_{01}\left(2\right)+{\mathrm{\ξ}}_{01}\left(3\right)+{\mathrm{\ξ}}_{01}\left(4\right)+{\mathrm{\ξ}}_{01}\left(5\right){\mathrm{\ξ}}_{01}\left(6\right)+{\mathrm{\ξ}}_{01}\left(7\right)+{\mathrm{\ξ}}_{01}\left(8\right)+{\mathrm{\ξ}}_{01}\left(9\right))$

$=\frac{1}{9}(0.429+0.5+0.559+0.479+0.698+0.351+0.443+0.570+0.333)=0.485$

In like manner, can get relational degree as shown in table 9.

Table 9. relational degree gathers

??r o1	??r o2	??r o3	??r o4	??r o5
					??0.485	??0.487	??0.542	??0.586	??0.650

(3-4) discharge the related preface that each achievement data is listed as according to relational degree;

Relational degree is female ordered series of numbers { x _oAnd subnumber row { x (t) } _i(t) } the measuring of cognation size between, if both change basically identicals relatively in evolution, then both relational degree are big; Otherwise both relational degree are just little;

Cause: 0.650＞0.586＞0.542＞0.487＞0.485

So: y _jTo x ₀Related preface be: r _O5＞r _O4＞r _O3＞r _O2＞r _O1

(3-5) relational degree of calculating each index respectively accounts for the ratio of the relational degree of all indexs;

Softening temperature:

In like manner, can get the shared ratio of other factors importance, as shown in table 10.

Each factor proportion of table 10. gathers

Softening temperature	Ductility	??PI	??T 800	??T 1.2
					??17.6	??17.7	??19.7	??21.3	??23.3

Step 4, utilize the range analysis method of orthogonal test that above-mentioned each index is carried out analysis-by-synthesis,

According to the orthogonal test conceptual design, each sample to be tested, concrete test-results sees Table 11.

Three kinds of fire retardant orthogonal experiments of table 11.

Simultaneously, in order further to compare different flame retardant, different volume fire retardants have further carried out different volumes, the test of different flame retardant situation to the influence of pitch three big indexs.

The different volume penetration test of three kinds of fire retardants results as shown in Figure 1, the penetration degree of flame-retardant pitch and fire retardant volume are stepped change and concern; When volume when 0% is increased to 5%, penetration degree descends very fast.When volume when 5% is increased to 15%, penetration degree descends slowly.From Fig. 1 equally as can be seen: the fire retardant kind is the most remarkable to the influence of penetration degree, shows that the fire retardant that chooses can suitably improve the penetration degree of flame-retardant pitch; Under same temperature, A ₁The penetration degree of other two kinds of flame-retardant pitchs of the penetration ratio of flame-retardant pitch is big slightly; The analysis of causes that The above results is produced as can be known, A ₂Number and A ₃Number fire retardant all is Powdered, this fire retardant can not form more stable compatible structure with pitch, when very low volume, can be dispersed in pitch mutually in, and after the volume increase, make the stability decreases of fire-retardant modified asphalt, fire retardant dispersive ununiformity makes the bituminous thermotolerance sharply descend, thereby penetration degree is downward trend.A ₁Number fire retardant is particulate state, such fire retardant and bituminous consistency are poorer, more after a little while, be equivalent to filler at volume, the bituminous denseness is increased, penetration degree reduces, when volume reach a certain amount of after, particulate fire retardant and pitch contact area increase, compatible ability strengthens, denseness reduces, and penetration degree has increase to a certain degree.

The different volume softening temperature of three kinds of fire retardants degree test-results as shown in Figure 2, after fire retardant was incorporated into pitch, the bituminous softening temperature is parabolic type to be changed, and can improve the bituminous softening temperature after mixing fire retardant.Simultaneously as can be seen: the fire retardant kind is more remarkable to the influence of flame-retardant pitch softening temperature, under the same volume, and A ₁The flame-retardant pitch softening temperature is all than other two kinds of flame-retardant pitch softening temperature height.And for A ₁, A ₂And A ₃Fire retardant all have when the fire retardant volume hour, softening temperature raises gradually, when the fire retardant volume reached certain numerical value (10%), softening temperature no longer increased, even the trend that reduces is arranged; Show when the fire retardant volume hour, add fire retardant and make the pitch hardening, high-temperature behavior is strengthened; After the fire retardant volume reached certain value, unnecessary fire retardant only can play the effect of filler, and is little to the high-temperature behavior influence of asphalt, even plays the effect of obstruction.

The different volume ductility test of three kinds of fire retardants results as shown in Figure 3, after fire retardant joined pitch, the bituminous ductility was broken line and changes, volume be 5% o'clock ductility to reach maximum value be 38cm, descend rapidly greater than 15% o'clock ductility at volume.Illustrate that fire retardant is bigger to the influence of bituminous non-deformability, along with the increase of flame retardant amount, the ability of pitch viscous deformation is more and more littler.

From The above results as can be seen, for three kinds of fire retardants, along with the increase of fire retardant consumption, the bituminous ductility is on a declining curve, but the lowering speed difference, generally speaking, the bituminous ductility is very responsive to admixture, even addition seldom just makes ductility generation considerable change, this is because the blend process on the physical significance just between fire retardant and the pitch, interfacial tension is very big, is easy to rupture owing to stress concentration appears in defective.

By above-mentioned analytical results and table 11 testing data as can be seen, the effect of estimating fire-retardant modified asphalt with single index can not draw the fire retardant of flame-retardant modified best results, must carry out comprehensive evaluation to five indexs (softening temperature, ductility, penetrative index, equivalent softening temperature, equivalent brittle point), so just can draw suitable flame-retardant modified technology, and then the flame-retardant modified effect that obtains envisioning.The dimension of these five indexs is different simultaneously, influence degree to flame-retardant modified effect also is not quite similar, so need to adopt Gray Association Analysis to try to achieve these several indexs shared ratio (the front grey correlation analysis is finished) when the flame-retardant modified effect importance of influence, can carry out next step evaluation.

Comprehensive evaluation:

For reducing the influence of various factors in the process of the test, overcome the not comprehensive and indefinite defective of single evaluation index, and take into account each index, according to the afore-mentioned test result, carry out comprehensive evaluation.Method commonly used has two kinds: overall equilbrium method and comprehensive grading method.The weighted comprehensive point system in the comprehensive grading method is selected in this research for use.

Comprehensive grading is determined standards of grading according to particular case after promptly many indexs being tested one by one, to these index comprehensive scorings, many index comprehensives is converted into single index, thereby obtains the conclusion of many index tests.Calculate the weighted comprehensive point system and just be equivalent to, thereby be converted into single index to every group of test-results scoring.Before calculating, should make between each index numerically all to be on the same less order of magnitude.In statistical method, each desired value of a certain index is all taken advantage of or with removing a numerical value, perhaps all with adding or subtracting a numerical value together, its trend rule is constant.Calculating weighted comprehensive point system value Y in addition _iThe time, if investigate index require trend identical, then symbol is got identical; The trend difference, then symbol is different, and for example, three indexs all are the smaller the better, and other has the 4th index to be the bigger the better, if former three is just got, then should get negative sign for the 4th.According to Gray Association Analysis, calculate each evaluation index proportion in estimating modified effect, adopt weighting that a plurality of indexs of each sample are carried out comprehensive grading then, many indexs are converted into single index, and then adopt the range analysis method in the orthogonal experimental design to obtain the conclusion that many indexs are tested again.

For the single index Y after transforming _iExpression, Y _iCalculation formula is as follows:

In the formula (6):

With Y ₁For example is calculated as follows:

In like manner:

$b_{i2}=\frac{17.7}{40-16.5}=\frac{17.7}{23.5}=0.75$ $b_{i3}=\frac{19.7}{0.46-(-0.93)}=\frac{19.7}{1.40}=14.1$ $b_{i4}=\frac{21.3}{53.4-49.5}=\frac{21.3}{3.83}=5.56,$

$b_{i5}=\frac{23.3}{-11.6-(-22.3)}=\frac{\mathrm{2.3.3}}{10.68}=2.21$

Then: Y ₁=2.18 * 85+0.75 * 40+14.1 * 0.469+5.56 * 53.38-(2.21 *-22.3)=568.00

In like manner can get:

Y ₂=556.11, Y ₃=536.11, Y ₄=519.86, Y ₅=545.84, Y ₆=519.70, Y ₇=528.57Y ₈=516.86, Y ₉=473.87; Thus, draw flame-retardant pitch technical indicator comprehensive grading value, as shown in table 12.

Table 12. flame-retardant pitch technical indicator comprehensive grading value gathers

Step 5, according to index Y _iY draws _iWith the graph of a relation of fire retardant kind, fire retardant volume, shear time and each factor of shear temperature, as shown in Figure 4, following result can be arranged by table 12 and Fig. 4:

(1) from direct analysis, first group of test score value is the highest, and its optimum level is combined as A ₁B ₁C ₁D ₁

(2) the factor primary and secondary of the flame-retardant modified effect of influence is (main → inferior) ABCD in proper order.That is: the fire retardant kind is to asphalt modification influential effect maximum; Next is the volume of fire retardant; Moreover shear time when being flame-retardant modified; Be that flame-proof modifier adds fashionable temperature at last;

(3) according to the even comparable character of orthogonal experimental design, drawing optimum combination according to Fig. 3 .9 is A ₁B ₂C ₃D ₂

(4) comprehensively select excellently, carry out second batch of test.

Comparative analysis is tested in two combinations that draw above again, obtained data shown in table 3.19.

The contrast of table 13. optimum test combined index

By comprehensive grading as can be known: testing program A ₁B ₂C ₃D ₂Be better than A ₁B ₁C ₁D ₁So, A ₁Be preferred optimum resistance combustion agent, optimum process scheme is: be 10% A with volume ₁It is in 160 ℃ the pitch that fire retardant joins temperature, high speed shear 60 minutes.

Step 6, the flame-retardant pitch that above-mentioned preparation technology is drawn carry out the oxygen index test.

The bituminous flame retardant resistance adopts oxygen index (Oxygen Index is called for short OI) test to estimate.It is generally acknowledged that when OI was 20%～27%, material was a combustible matl, when OI less than 20% the time, material is a red label goods, when OI greater than 27% the time, this material is the material from horizontal blanking in fire.The calculation formula of OI is:

$\mathrm{OI}=\frac{\left[O_2\right]}{[O_2+N_2]}\×100\%$

In the formula: [O ₂]---the volumetric flow rate of oxygen in mixed gas during critical dissolved oxygen concentration;

[N ₂]---the volumetric flow rate of nitrogen in mixed gas during critical dissolved oxygen concentration.

In order further to analyze of the influence of fire retardant volume, to the A that optimizes to flame retardant property of asphaltum ₁Fire retardant adopts different volumes and the best technology of shearing to carry out the flame retardant properties test.Test-results is seen Fig. 5.

Fig. 5 shows A ₁Fire retardant volume and oxygen index graph of a relation, fire retardant is incorporated in the pitch, can improve the bituminous oxygen index, and the oxygen index of matrix pitch is approximately 20.After fire retardant added, the bituminous flame retardant properties can be improved.Add after 10% fire retardant, oxygen index reaches more than 24, and the oxygen index change curve is tending towards level, illustrates to add the effect of fire retardant again after certain volume and the flame retardant effect of volume 10% is more or less the same, from economic angle, 10% fire retardant volume is an optimum addition.

Although top invention has been described in conjunction with figure; but the present invention is not limited to above-mentioned embodiment; above-mentioned embodiment only is schematic; rather than it is restrictive; those of ordinary skill in the art is under enlightenment of the present invention; under the situation that does not break away from aim of the present invention, can also make a lot of distortion, these all belong within the protection of the present invention.

Claims (3)

1. selection of fire retardant method based on grey correlation analysis, it is characterized in that: this method may further comprise the steps:

Step 1, flame-retardant pitch orthogonal test conceptual design, its design procedure comprises:

(1-1) confirmed test index: the softening temperature of flame-retardant pitch, ductility and penetration degree;

(1-2) confirmed test factor: fire retardant kind, fire retardant volume, that is: the fire retardant quality accounts for per-cent, shear time and the shear temperature of asphalt quality;

The index of flame-retardant pitch sample in step 2, above-mentioned each the orthogonal test scheme of test,

Testing program is tested, and test index comprises softening temperature, ductility and the penetration degree of this flame-retardant pitch;

Step 3, utilize Gray Association Analysis that above-mentioned flame-retardant pitch sample is analyzed, thereby select suitable fire retardant, comprising:

(3-1) gray system is set:

Definition X ₀=[x ₀(1), x ₀(2), x ₀(3) ..., x ₀(n)] (1)

Formula (1) is the system features sequence;

$\left\{\begin{array}{c}{X}_1=[x_1\left(1\right),x_1\left(2\right),\·\·\·\·\·\·,x_1\left(n\right)]\\ X_2=[x_2\left(1\right),x_2\left(2\right),\·\·\·\·\·\·,x_2\left(n\right)]\\ \·\\ \·\\ \·\\ X_m=[x_m\left(1\right),x_m\left(2\right),\·\·\·\·\·\·,x_m\left(n\right)]\end{array}\right.---\left(2\right)$

Formula (2) is the correlative factor sequence; In formula (1) or the formula (2):

x ₀(n) be female factor series;

x _m(n) be sub-factor series;

Data rows according to following each index of formula (2) definition:

Volume row: x (0)={ x ₁(0), x ₂(0), x ₃(0) ..., x _n(0) }

Softening point range: x (1)={ x ₁(1), x ₂(1), x ₃(1) ... x _n(1) }

Ductility row: x (2)={ x ₁(2), x ₂(2), x ₃(2) ... x _n(2) }

(3-2) adopt interval value treatment process to carry out value just and handle, the starting point of each sequence is overlapped the test index that draws by the orthogonal test scheme; The sequence calculation formula that value is just handled is as follows:

X _i＝X _i/x _i(1)＝[x _i(1)，x _i(2)，…，x _i(n)]

(i＝0，1，2，…m)????(3)

(3-3) carry out grey incidence coefficient and grey relational degree and calculate, that is: note is { x through female ordered series of numbers of data conversion _o(t) }, subnumber is classified { x as _i(t) }, then when moment t=k, wherein:

{ x _o(k) } with { x _i(k) } incidence coefficient ξ _Oi(k) calculate with following formula (4) formula:

${\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right)=\frac{{\mathrm{\Δ}}_{\min }+{\mathrm{\ρ\Δ}}_{\max }}{{\mathrm{\Δ}}_{\mathrm{oi}}\left(k\right)+{\mathrm{\ρ\Δ}}_{\max }}---\left(4\right)$

In the formula (4): Δ _Oi(k) be the k absolute difference of two sequences, that is: Δ constantly _Oi(k)=| x _o(k)-x _i(k) |;

Δ _Min, Δ _MaxBe respectively maximum value and minimum value in each absolute difference constantly;

ρ is the explanation coefficient, and ρ ∈ (0,1), ρ get 0.1～0.5;

Relational degree r _Oi, calculate with following formula (5):

$r_{\mathrm{oi}}=\frac{1}{N}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ξ}}_{\mathrm{oi}}\left(k\right)---\left(5\right)$

In the formula (5): r _OiRelational degree for subsequence i and auxiliary sequence 0;

N is that the length of sequence is the data numbers;

(3-4) discharge the related preface that each achievement data is listed as according to relational degree;

Relational degree is female ordered series of numbers { x _oAnd subnumber row { x (t) } _i(t) } the measuring of cognation size between, if both change basically identicals relatively in evolution, then both relational degree are big; Otherwise both relational degree are just little;

(3-5) relational degree of calculating each index respectively accounts for the ratio of the relational degree of all indexs;

Step 4, utilize the range analysis method of orthogonal test that above-mentioned each index is carried out analysis-by-synthesis, that is:

Many index comprehensives are converted into single index, for the single index Y after transforming _iExpression, Y _iCalculation formula is as follows:

In the formula (6): z is a test piece number (Test pc No.); L is the number of index;

The index Y that step 5, basis are calculated by formula (6) _iY draws _iWith the graph of a relation of fire retardant kind, fire retardant volume, shear time and each factor of shear temperature, thereby optimize the best preparation technology of a suitable fire retardant and flame-retardant pitch;

Step 6, the flame-retardant pitch that above-mentioned preparation technology is drawn carry out the oxygen index test.

2. according to the described selection of fire retardant method based on grey correlation analysis of claim 1, it is characterized in that: the test index in the step 1 also includes penetrative index, equivalent softening temperature, equivalent brittle point.

3. according to the described selection of fire retardant method of claim 1, it is characterized in that: the penetrative index, equivalent softening temperature and the equivalent brittle point that also calculate this flame-retardant pitch in the step 2 simultaneously based on grey correlation analysis; Also define the data rows of following each index in the step 3 according to formula (2):

Penetrative index row: x (3)={ x ₁(3), x ₂(3), x ₃(3) ..., x _n(3) }

Equivalent softens point range: x (4)={ x ₁(4), x ₂(4), x ₃(4) ..., x _n(4) }

Equivalent brittle point row: x (5)={ x ₁(5), x ₂(5), x ₃(5) ..., x _n(5) }.

Images