HG 3104-1989 Rubber Test Data Distribution Type Inspection Regulations HG3104-1989 Standard Download Decompression Password: www.bzxz.net

UDC 678. 4. 01
National Standard of the People's Republic of China
GB1117989
Adjusted to: HGh 3104-f8
Rubber-test rule for distribution type of testing data1989-03-31 Issued
State Administration of Technical Supervision
1990.01-01 Implementation
National Standard of the People's Republic of China
Rubber-test rule for distribution type of testing data1 Title Content and Scope of Application
GB 11179—89
This standard specifies the basic procedures for testing the distribution type of rubber test data by methods. According to this standard, the distribution type of product quality indicators and economic indicators can be determined, and the distribution type of rubber or rubber product test data can also be determined. This standard is applicable to the test of rubber test observations, data after function transformation of observations and random data empirical distribution and the goodness of fit of theoretical distribution types.
2 Reference standards
GB 528 Determination of tensile properties of vulcanized rubber GB3187 Basic terms and definitions of reliability GB3358 Statistical terms and symbols
GB4882 Statistical processing and interpretation of data
Normality test
3 Symbols
H…-null hypothesis;
n——sample size;
,X. —n observed values ​​of order statistics of the sample, R(X,)——the function value of the theoretical distribution when the observed value is,; the significance level of the test;
-statistics used in the test;
G(a)—distribution function of the value. Its value is shown in Appendix A. 4 Application conditions
4-1 When it is suspected that the sample comes from a known population distribution, the corresponding distribution type test should be carried out. 4.2 The test method can be used to test various distributions. Sampling must ensure randomness, and the observed values ​​of random variables should be obtained under the same conditions. The observed samples must come from a population. 4.3 The sample size must not be less than 50.
4.4 When it is necessary to estimate the mode, weighted mean or weighted standard deviation from the sample, the samples must be grouped with equal distances. The number of groups depends on the sample size. The relationship with sample size is:
Rw 50~99
#- 140-199
Approved by the Ministry of Chemical Industry of the People's Republic of China on March 10, 1989 710-15
r=1518
Implementation on January 1, 1990
5 Implementation steps
n—1 000
GB 11179—89
r— 1820
7= 35~～40
5.1 Given Ho: The sample size comes from a known theoretical distribution. 5.2 Arrange the observations into an order statistic
X+X..*,X.
5.3 Calculate the theoretical distribution function value
(j-1,2.**)
5.4 Calculate the product according to formula (2):
f(x,)dx
22iInF(X) +
1-2j-1
In(—F(X,))
（2）
Fill in the data according to Table 1 during calculation. The intermediate results in Table 1 are calculated to five significant figures, and the final results are retained to three significant figures. Add up the 10th column of Table 1 and calculate according to formula (2) to obtain the required. 5.5 In Appendix A and the corresponding distribution function value (@). 5.6 Take the significance level at =0.2.
5.7 Make a judgment:
If Q()≥1, then reject H, that is, reject the hypothesis that the empirical distribution is consistent with the theoretical distribution. If ()<1, then accept this hypothesis.
For an example of testing the normal distribution according to o, see Appendix B; for the BASIC program, see Appendix C. 2
GB1117989
( (-t)+
cexa-tia
)+ex)u
(6)+()
(ca-(et)
(x)-(-t)
Eex)&-utkxd-i
GB 11179~89
Appendix A
G(magnetic) value
(supplement)
0. 09
GH11179
Continued Table A1
Appendix B
(reference)
Select a natural rubber vulcanized rubber specimen and obtain a sample with a tensile strength of &=57 (unit: MPa) according to the No. 4 cutting knife of GB528. Use the member method to test whether the population follows a normal distribution. B1 gives I: the sample comes from a normal distribution,
B2 arranges the observations into an order statistic (see Table B1). B3 Calculate the parameter estimates X, 8 and all theoretical distribution function values ​​of the normal distribution according to the observed data: meat
S--1. 1237
(j-1,2,-,57)
Fill the order statistic number; and the theoretical distribution value F(X) of the order statistic X, into the first and third columns of Table 1 respectively, and then calculate according to the calculation formulas in the second and fourth to tenth columns of Table 1, and fill in the corresponding results to form Table 32, where the values ​​in the tenth column are added to get -28.715.
— 57 — 2 × (- 28. 715)
B4 Give the significance level a-0.2
B5 Set Appendix A to get Q(o%)-G(0.13)=D.1771—= 0. 8
G(1-
Accept.: The original sample comes from a normal distribution population. 5
.0. 044
0·482
0. 002 99
0. 006 53
0. 020 81
e. (55 61
Table B1
GB 11179—89
Arranged into the original data of the order statistic
Use the data in Table B1 according to the calculation formulas in Table 1 to calculate the results 5
5. 811 99
--5. 030'87
—3.872 46
—0.050 98
:0.169 84
-2, 889 44 0. 177 42
—2. 713 39
0. 956 11
0. 938 60
—0.214-22
—2. 225 36|| tt|| - 0. 214 73
0. 125 45
--.2. 075 840, 236 72
0. 165 94
0. 189 05
0. 214 04
0.903 - 51 |
0. 993 47
0. 933 69
0. 891 97
0. 874 55
心.810 95
—0.003 00
—0.00297
—0. 006 38
0.053 95.
—0. 138 77
— 0. 020 10 : —0. 189 95
0. 021 03
0. 057 210. 053 700. 231 12
0. 61 - 0. 063 20 — 0. 277 41 0. 114 32
- 0. 103 29
—0. 318 02
0. 134 050. 118 76
—0.35548
-0.181 46bZxz.net
0. 157 58
—0.39391
0. 209 55
-0.178 30
0. 785 96
5 -0. 200 710. 457 64
0. 240 851. 423 57 0. 262 240. 816 79 0. 759 150.240 85
0. 269 40
0. 269 40
1. 423 57-0. 287 21
1. 423 57 0. 312 19
1. 311 57 0. 310 64
1. 311 57 0. 333 65
1. 311 57 -- 0. 356 66
1. 311 57 :0. 379 67
0. 745 61
0. 331 13 m 1. 105 25 0. 339 330.363 96
1. 010 700. 328 03||tt || ~1. 010 70-0. 345 77
0. 363 96
0. 432 49
0. 432 49
0. 467 67
0. 467 67
0. 467 67
1.010 70-0.363 50||tt ||—1. 010 70 :0. 381 23
. 0. 330 87
0. 838 200. 345 57
- 0. 838 200. 360 28||tt ||—0. 759 99
0. 759 99
—0.75999
—0. 339 99
—0.353 33
—0. 366 66
0. 675 44
0. 640 35.
0. 622 81||tt ||0. 605 26
0. 587 72.
0. 552 63
0. 759 15
.0.730 60
0. 668 87
0. 636 04
0. 636 04
0. 567 51
0. 567 51
:0.532 33
0:53233
—0.275560.224 80.-0.487-04
- 0. 275 56
..- 0. 219 96||tt| |—0. 507 18
-0. 275 56 —0. 215 13
0. 319 88-- 0. 239 54
-0.550 18
: 0. 234 04
—0.228 53
-0. 313 880. 223 02
-0.567 68
-- 0. 602 69
— 0. 402 16 -- 0. 278 69
- 0.618 02
.*0. 452 50--0 . 305 64
0. 452 50 0. 297 70
:-0. 452 500.289 76
0. 452 50— 0. 281 82
0. 566 50 0. 342 88
0. 566 50-- 0. 332 94
0. 633 67
0. 643 46
—0.65326
0. 663 05
—0. 673 75
0. 678 51
0. 566 50: 0. 323 00
D. 630 50 0. 348 43
— 0. 337 37
-- 0. 326 31
—0 . 688 43
一心.69心70
0. 692 97
0. 503 11
0. 573 64
0. 573 64
0: 641 88
0. 674 51
0. 674 51
0. 815 14||tt| |0. 815 14
0: 815 14
0. 815 14
0. 837 91
0. 837 91
GB 11179-: 89
Continued Table B2
— 0. 686 94
—0.343 17
— 0. 618 92
-0. 618 92
— 0. 618 92
0. 500 00
-0. 320 32
—0.33118
0. 342 04||tt| |—0.55576
-0. 316 88
0. 443 36 0. 268 35
0. 393 770. 245 24
- 0. 393 770. 252 15
— 0. 259 06
-0. 393 77
—0.23529
—0. 348 31
0. 306 89
— 0.212 67
0. 269 22
— 0. 191 29
0. 235 12||tt| |—0,20440
0. 152 40
0. 204 40-0. 155 99
0. 204 40 --0. 159 57
—0.16316
- 0. 176 B4 - 0. 141 26
-0. 176 ±4 -0.147 37
— 0. 152 25
—0. 129 54
-- 0. 152 25 0, 132 21
0. 447 37
0. 359 65
0 . 342 11
0. 324 56
0. 289 47
.0. t84 21
0. 149 12
0. 131 58| |tt||-- U. 152 250. 134 89
0. 858 78
—0. 137 56
0. 152 25
-0.130 41
0. 120 11
0. 894 81 : -0. 111 11
— 0. 104 29
U. 955 18
0. 962 97
-0. 045 86
-- 0. 037 74:
0. 096 49
0. 078 95
0. 061 40
Note: In addition to this method, GB4882 can also be used for normal distribution test. 7
0. 496 89
0 . 461 48
0. 461 48
0. 358 12
0. 326 49
心.32549
:0.294 14
0. 184 86
0. 184 86
0. 184 85
0. 162 09
0. 141 22||tt ||0. 141 22
0. 1-41 22
0. 122 26
0. 044 82
(0. 037 03
Appendix C
\Tested BASIC program
(reference)
- 0. 699 39
— 0. 349 70
0.773 330. 373 10
0. 773 33
—0.359 53
—0. 773 33
0. 345 96
0. 852. 46 -0. 366 41
-0.852 46
-0.351 45
-1. 026 88 0. 405 35
0. 603 17
-- 0. 690 71
0. 688 00
0. 683 29
0. 678 09
1. 122 42-0. 423 37
0.668 61
1.122 42 0, 403 68 0, 655 831. 122 420. 383 99
—1. 223 69—0. 397 16
1. 330 79 0. 408 57
- 0. 621 25
— 0. 417 95
1. 443·83
1. 562 92 -0. 425 01
1, 688 16 0. 429 41
—-0.596 19
1. 688 16. - 0. 399 83
—0. 555 8 1
1, 688 16..0. $70 21
..0 . 529 78
—1. 688 16 -0. 340 59
—0.503 75
1. 819 62 --0. 335 19 —0, 479 46-1. 819 62 - 0. 303 27 - 0. 450 64-1. 957 41
-0. 291 89
-- 0. 421 44
1. 957 41 0. 267 55 ~ 0. 389 771. 957 41 0. 223 21 ( 0. 358 10 1. 957 41
0. 188 87
— 0. 165 92
—0.32643
—0. 286 03
—0.212.59
- 0. 136 19 -- 0. 180 03||tt| |—3.105 06
—3. 105 06
0. 081 :71
0. 126 36
0. 066 32
This program uses the method to test the normal distribution and double exponential distribution of the sample. Statements 1480 to 1490 are reserved for filling in other distribution calculations. 7
GB 1117989
Calculate the position of the statement. In addition, this program gives a histogram after grouping the input in a human-computer dialogue style. The original data is given in a sequential file with TXT as the backing. The data to be checked in the appendix It is also a data file and is named A,DA. The program can automatically find the corresponding table value. When the data exceeds the range in the table, the program can automatically calculate the value and print it out. 10 WJDTH LPRINT 132
20 CLS:PRINT TAB(20)| |tt||30 PRINTTAB(20)
40 PRINT
“Data distribution type test total menu”
50PRINTTAB(5)“1. Input original data\,: PRINTTAB(46 ) "2. Construct order statistics" 60 PRINTTAB (5) * 3. Print histogram"; 70 PRTNTTAB (46) "4.2 Test\,: PRINTTAB (5) 80 PRINT\.PRINT
90INPUT"Please select the number to perform the corresponding operation", BH100ONBHGOSUB120,250,310,710,1600110 GOTO 20
"6.End"
"Enter the number of data", K: INPUT "Input TXT file name", Q$120 CLS.INPUT
130 DIM X(K),A(K,11).:OPEN \C.\+Q $ +\.TXT FOR INPUT AS I140FORI= 1TOK:INPLT41,X(I).NEXTI.CLOSE#1150 XBA=0:SIGMA=0
160 FOR I=1 TO K
170 XBA=XBA+X(1)||tt ||180 NEXT1
190XBA-XBA/K
200 FOR1--1TOK
210SIGMA=SIGMA+(X(I)-XBA)*(X(1)-XBA)220 NFXTI|| tt||230 SIGMA=SQR(SIGMA/(K-1))
240 RETURN
250CLS, PRINT "Sorting"
260FORKK=1TOK-1
270 FOR NN=I TO K-1
280 IF X(NN)) X(NN+1) THEN SWAP X(NN),X(NN + 1)290 NEXT NN,KK
300 RETURN
310 CLS:INPUT
320 NH-1000
"Input the number of groups\,R:H#一(X(K)一X(1))/R330 IF QI$=\Y\| |tt||OR BH=3 THEN ERASE S,ZZ
340 DIM S(R),ZZ(R)
350 FOR RR -1 TO R
360 S(RR )=0
370 ZZ(RR)-X(1)+(RR-0. 5) *H *380 NEXT RR
390 FOR KK-1 TO K
400 FOR RR=1 TO R
GB 11179-- 89
410 FX(KK)<--X(1)+RR *H#+0. 001 AND X(KK)) =X(1)+(RR-1) *H# 0. 001THFN 420 ELSE 440||tt| |420 S(RR)-S(RR)-+ 1
430 GOrO 450
410 NEXT RR
450 MEXT KK
460 Fs(1)|| tt||470 FORRR-1TOR
480IFS(RR))=FTHEN49DELSF510
490 F=S(RR)
500 R1--RR
610 NEXT RR| |tt||520 M0=X(1)(R1-1)*H# +H# * (S(R1)-S(R1-1))/(2+S(RI)-S(R1-1)—S(R1 +1))530MZ=X(1)+(R—0.5)*H#
540 MZZ=2. 5/(MZ-M0)
550 FOR RR=1 TO R|| tt||“####.###\,X(I)+(RR—1)*H#;.PRINT.560PRINTUSING
“####.###\,X(I1)+RR*H#:PRINT USING "###\,USING
:PRINT
)0THEN580ELSEPRINTUSING"#########",ZZ(RR):GOTO570 IF S(RR)<||tt| |580 PRINT USING *# ## ##. ####\;ZZ(RR);590 FOR NN=1 TO S(RR)
600 PRINT TAB(40+2* (NN.-- 1)) \\670 IF NN-S(RR) THEN PRINT \620 NEXT NN
630 NEXT RR
640INPUT"Print this histogram? Y/N",Q3$650 IF Q3 $ -\y\ THEN Q3$ -\y\660IFQ3$-"Y\THENGOSUB1500
670INPUT' Regroup? Y/N,Q1$680 IF Q1 $=\y\ THEN Q1 S -\Y\690 IF Q1 $ -*Yn THEN 310
700 Q1$--“Y\.RETURN|| tt||710 CLS
720PRINT\Using\ method to test the distribution type of rubber physical test data is generally effective. Here we only give the test of the static distribution and double exponential distribution, and leave a Check the interface of other distribution types"730 PRINT*
740 PRINT"1 Normal distribution"
750PRINT"2Double exponential distribution"
760FRINT"3Other distributions\
770PRINT"4Test end"
780PRINT"\,INPUT\Please Select by serial number”, BH2790 F BH2=4 THEN 1600
800FOR1-1TOK
GB 1117989
810 A(1,1)=-I:A(T,2)= (2*1-1)/(2*K)820 0N BH2 G0SUB 1350,1450,1480,1600830 A(T,3)=A1:A(I,4)-L0G(A(1,3)):A(1,5)=A(I,2)*A(I,4),A(I ,6)=IA(I,2)840 A(I,7)-1—A(I,3):A(I,8)-L0G(A(1,7)):A(I,9) -A(I,6) A(I,8):A(1,10)=A(I,5)+A(I,9):A(I,1)-X(0)||tt| |850 NEXT 1
860 SUM-D
870FOMM-1TOK
880SUM=SUM+A(M,10)
890 NEXT M
900OMEGA=INT((-K2*SUM+0.005)*100/100!910 FOR I-1 TO K
920PRINTUSING“###”,A(1,1);: PRINTUSING“##.###\;A(I,2);930 PRINT USING“##. #####\;A(I,3),+PRINT USING *###.#### #\,A(I,4);940 FRINT USING “###.#####\;A(I,5);:PRINT USING *##.#####\,A(I, 5);950 PRINT USING *###, ####*,A(I,?);:PRINT USING *###. #####\A(I,8)
960PRINTUSING“###.#####”;A(I,9)：PRINTUSING“###，#####”A (1,10) ;
970 NEXT I
980LPRINT"?一"LPRINTUSING"###,##"OMEGA990 CODE% = 100+DMEGA
1000 IF CODE%) 295 THEN 1010 ELSE 12601010 SS=0
1020 FOR JJ=0 TO 6
1030 IF JI=D THEN A1-1;GOTO 11301040FJJ=ITHENAI-0.5:GOTO11301050 FZ-JJ-0.5
1060 FM-JJ
1070 A1=F7./FM
1080FORI-JJTHEN2STEP-1
1090 FZ-FZ—1||tt ||1100 FM--FM-1
1110 A1=A1+FZ/FM
1120 NEXT 11
1130B1=(—1)^JJ*（4*JJ+1 )*EXP((—9.87202324#*（4*JJ+1)/(8*OMEGA)))1140 IF ABS(B1) <1E-D9 THEN 1310I150 DBF FNPL(Y)--EXP(OMEGA/(8 * (Y * Y+1))-*OMEGA))
1160 N=1000
1170 H-500/N
1180 S0. 5 * (FNPL(0)—FNPL (1000))10
-((4 * JJ+1) * 3. 1415926# * Y) ~ 2/(81190FORL=1TON
GB 11179--89
1200 S=$+2*FNPL((2*L—1) *H)+FNPL(2*L* H)1210 NEXT I,
1220 CI=1000 *S* 2. 506628/(3 * N* 0MEGA)1230SSSS+AI*BiC1
1240 NEXT JJ
1250 GOTO1310
1260 OPEN“A;DA\AS 2 LEM=241270TIELD#2,8ASN$.8ASA$||tt ||1280 GEI #2,CODE%
1290 SS=VaL(A$)
1300 CLOSE
I3I0 LPRINT \Gα?)=\, :LPRINT USING “#, # ##\,SS1320INPUT "Still need to do the inspection? Y/N”,Q2$ 1330 F Q2$ -*y\ Then Q2$ --\y\1340 IF Q2 $ -\Y\ THEN. 710 FISE RETURN1350 B=(X(I)—XBA)/SIGMA|| tt||1360 C-—6
1370 NI=50
.1380H--(BC)/(2*NI)
1390 S=0.5* (EXP(—C *2/2))(EXP(—B~2/2))/2.5066282#1400 FOR L- 1 TO NI
: 1410 S=S + 2 *EXP(-(C+(2*L- 1) * H) ~ 2/2)/2. 5066282# +EXP(-(C+2 * L *H) 2/2)/2.5066282#
1420 NEXT L
1430 A1=(B—C)*S/(3*N1)| |tt||1440RETURN
1450B=
-EXP(MZZ8(X(I)-M0))
1460 A1=I--EXP(B) :IH A11 THEN AI -- 0. 999991470 RETURN
1480REMA1-Probability value of a certain distribution to be tested1490 REMRETURN
1500 FORRR-1TOR
1510 LPRINTUSING“####.##±\,X(1)+(RR-1)*H#：.LPRINT“####.###\,X(1)+RR*H#;:LPRINT USING“###”LPRINT USING
1520 IF S(RR) < >0 THEN 153D ELSE LPRINT USING “# ####.####*;ZZ(RR).GOTO1580
1S30LPRINTSING"####,####"ZZ(RR)
1540 FOR NN=1 TO S(RR)
1550 LPRINT TAB(40+ 2 * (NN-1));\\1560IFNN-S(RR)THENI.PRINT| |tt||1570 NEXT NN
1580 NEXT RR
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.