Knowledge (XXG)

Adaptive step size

Source đź“ť

22: 1272: 1776: 3100:
The description given above is a simplified procedures used in the stepsize control for explicit RK solvers. A more detailed treatment can be found in Hairer's textbook. The ODE solver in many programming languages uses this procedure as the default strategy for adaptive stepsize control, which adds
2677: 824:. Take the new step size to be one half of the original step size, and apply two steps of Euler's method. This second solution is presumably more accurate. Since we have to apply Euler's method twice, the local error is (in the worst case) twice the original error. 1071: 1065: 2825: 1638: 158:. There, scenarios emerge where one can take large time steps when the spacecraft is far from the Earth and Moon, but if the spacecraft gets close to colliding with one of the planetary bodies, then small time steps are needed. 2902: 923: 146:. Using an adaptive stepsize is of particular importance when there is a large variation in the size of the derivative. For example, when modeling the motion of a satellite about the earth as a standard 1996: 1578: 2250: 665: 2481: 3097:, which has lower order. The above prediction formula is plausible in a sense that it enlarges the step if the estimated local error is smaller than the tolerance and it shrinks the step otherwise. 1368: 279: 1889: 557: 2091:
Similar methods can be developed for higher order methods, such as the 4th-order Runge–Kutta method. Also, a global error tolerance can be achieved by scaling the local error to global scope.
3059: 2421: 1801:
is a safety factor to ensure success on the next try. The minimum and maximum are to prevent extreme changes from the previous stepsize. This should, in principle give an error of about
1267:{\displaystyle \tau _{n+1}^{(1)}=c\left({\frac {h}{2}}\right)^{2}+c\left({\frac {h}{2}}\right)^{2}=2c\left({\frac {h}{2}}\right)^{2}={\frac {1}{2}}ch^{2}={\frac {1}{2}}\tau _{n+1}^{(0)}} 2005:
accurate in the local scope (second order in the global scope), but since there is no error estimate for it, this doesn't help in reducing the number of steps. This technique is called
739: 393: 1827: 929: 2710: 2310: 131: 3095: 2473: 1626: 413: 1471: 2718: 2084:, using an optimal number of steps given a local error tolerance. A drawback is that the step size may become prohibitively small, especially when using the low-order 2959: 2932: 2444: 2340: 2143: 2042: 1799: 822: 775: 154:
may be sufficient. However things are more difficult if one wishes to model the motion of a spacecraft taking into account both the Earth and the Moon as in the
1771:{\displaystyle h\rightarrow 0.9\times h\times \min \left(\max \left(\left({\frac {\text{tol}}{2\left|\tau _{n+1}^{(1)}\right|}}\right)^{1/2},0.3\right),2\right)} 2082: 2062: 1604: 1391: 1429: 2712:
is the unnormalized error. To normalize it, we compare it against a user-defined tolerance, which consists of the absolute tolerance and relative tolerance:
3185: 2831: 830: 2104: 163: 39: 3115: 3170: 3156: 105: 1897: 1479: 2672:{\displaystyle {\textrm {err}}_{n}(h)={\tilde {y}}_{n+1}-y_{n+1}=h({\tilde {\psi }}(t_{n},y_{n},h_{n})-\psi (t_{n},y_{n},h_{n}))} 2151: 86: 568: 58: 2115:
methods. These methods are considered to be more computationally efficient, but have lower accuracy in their error estimates.
43: 1278: 784:
is not known to us. Let us now apply Euler's method again with a different step size to generate a second approximation to
65: 1832: 467: 191: 3190: 2967: 72: 2345: 2112: 32: 2006: 54: 684: 1060:{\displaystyle y_{n+1}^{(1)}=y_{n+{\frac {1}{2}}}+{\frac {h}{2}}f(t_{n+{\frac {1}{2}}},y_{n+{\frac {1}{2}}})} 350: 3134:
E. Hairer, S. P. Norsett G. Wanner, “Solving Ordinary Differential Equations I: Nonstiff Problems”, Sec. II.
2100: 1804: 292:
may denote vectors (in which case this equation represents a system of coupled ODEs in several variables).
2684: 2256: 135: 2108: 3071: 2449: 3110: 678:
is sufficiently smooth) the local truncation error is proportional to the square of the step size:
671: 159: 139: 2820:{\displaystyle {\textrm {tol}}_{n}={\textrm {Atol}}+{\textrm {Rtol}}\cdot \max(|y_{n}|,|y_{n-1}|)} 1609: 398: 155: 123: 1434: 79: 3166: 3152: 2937: 2910: 2429: 2318: 2121: 2015: 1891:, we consider the step successful, and the error estimate is used to improve the solution: 179: 1784: 1606:
should be modified to achieve the desired accuracy. For example, if a local tolerance of
801: 754: 2118:
To illustrate the ideas of embedded method, consider the following scheme which update
2067: 2047: 1589: 1376: 1396: 3179: 2099:
Adaptive stepsize methods that use a so-called 'embedded' error estimate include the
2085: 175: 151: 147: 182:
methods are preferred due to their superior convergence and stability properties.
174:
For simplicity, the following example uses the simplest integration method, the
166:
are examples of a numerical integration methods which use an adaptive stepsize.
143: 119: 21: 3147:
William H. Press, Saul A. Teukolsky, William T. Vetterling, Brian P. Flannery,
2897:{\displaystyle E_{n}={\textrm {norm}}({\textrm {err}}_{n}/{\textrm {tol}}_{n})} 2044:, this theory facilitates our controllable integration of the ODE from point 918:{\displaystyle y_{n+{\frac {1}{2}}}=y_{n}+{\frac {h}{2}}f(t_{n},y_{n})} 448:, the Euler method gives approximations to the corresponding values of 562:
The local truncation error of this approximation is defined by
15: 3101:
other engineering parameters to make the system more stable.
1991:{\displaystyle y_{n+1}^{(2)}=y_{n+1}^{(1)}+\tau _{n+1}^{(1)}} 1573:{\displaystyle y_{n+1}^{(1)}-y_{n+1}^{(0)}=\tau _{n+1}^{(1)}} 138:) in order to control the errors of the method and to ensure 1586:
The local error estimate can be used to decide how stepsize
2245:{\displaystyle y_{n+1}=y_{n}+h_{n}\psi (t_{n},y_{n},h_{n})} 660:{\displaystyle \tau _{n+1}^{(0)}=y(t_{n+1})-y_{n+1}^{(0)}} 1363:{\displaystyle y_{n+1}^{(1)}+\tau _{n+1}^{(1)}=y(t+h)} 3074: 2970: 2940: 2913: 2834: 2721: 2687: 2484: 2452: 2432: 2348: 2321: 2259: 2154: 2124: 2070: 2050: 2018: 1900: 1835: 1807: 1787: 1641: 1612: 1592: 1482: 1437: 1399: 1379: 1281: 1074: 932: 833: 804: 757: 687: 571: 470: 401: 353: 194: 132:
numerical solution of ordinary differential equations
3151:, Second Edition, CAMBRIDGE UNIVERSITY PRESS, 1992. 320:), and we are interested in finding the solution at 1884:{\displaystyle |\tau _{n+1}^{(1)}|<{\text{tol}}} 1583:This local error estimate is third order accurate. 1431:. In reality its rate of change is proportional to 798:). We get a second solution, which we label with a 552:{\displaystyle y_{n+1}^{(0)}=y_{n}+hf(t_{n},y_{n})} 46:. Unsourced material may be challenged and removed. 3089: 3053: 2953: 2926: 2896: 2819: 2704: 2671: 2467: 2438: 2415: 2334: 2304: 2244: 2137: 2076: 2056: 2036: 1990: 1883: 1821: 1793: 1770: 1620: 1598: 1572: 1473:. Subtracting solutions gives the error estimate: 1465: 1423: 1385: 1362: 1266: 1059: 917: 816: 769: 751:We have marked this solution and its error with a 733: 659: 551: 407: 387: 273: 274:{\displaystyle y'(t)=f(t,y(t)),\qquad y(a)=y_{a}} 3054:{\displaystyle h_{n}=h_{n-1}(1/E_{n})^{1/(q+1)}} 2759: 1668: 1660: 3165:, Second Edition, John Wiley & Sons, 1989. 347:denote the solution that we compute. We write 8: 3068:is the order corresponding to the RK method 2416:{\displaystyle h_{n}=g(t_{n},y_{n},h_{n-1})} 178:; in practice, higher-order methods such as 2342:is predicted from the previous information 150:, a fixed time-stepping method such as the 2475:. The error then can be simply written as 3076: 3075: 3073: 3029: 3025: 3015: 3006: 2988: 2975: 2969: 2945: 2939: 2918: 2912: 2885: 2879: 2878: 2872: 2866: 2860: 2859: 2849: 2848: 2839: 2833: 2809: 2797: 2788: 2780: 2774: 2765: 2750: 2749: 2740: 2739: 2730: 2724: 2723: 2720: 2696: 2690: 2689: 2686: 2657: 2644: 2631: 2609: 2596: 2583: 2565: 2564: 2543: 2524: 2513: 2512: 2493: 2487: 2486: 2483: 2454: 2453: 2451: 2431: 2398: 2385: 2372: 2353: 2347: 2326: 2320: 2296: 2283: 2264: 2258: 2233: 2220: 2207: 2191: 2178: 2159: 2153: 2129: 2123: 2069: 2049: 2017: 1976: 1965: 1946: 1935: 1916: 1905: 1899: 1876: 1868: 1856: 1845: 1836: 1834: 1814: 1806: 1786: 1736: 1732: 1709: 1698: 1681: 1640: 1613: 1611: 1591: 1558: 1547: 1528: 1517: 1498: 1487: 1481: 1442: 1436: 1398: 1378: 1327: 1316: 1297: 1286: 1280: 1252: 1241: 1227: 1218: 1201: 1192: 1178: 1158: 1144: 1127: 1113: 1090: 1079: 1073: 1042: 1035: 1016: 1009: 989: 974: 967: 948: 937: 931: 906: 893: 873: 864: 845: 838: 832: 803: 756: 725: 703: 692: 686: 645: 634: 612: 587: 576: 570: 540: 527: 505: 486: 475: 469: 400: 358: 352: 265: 193: 106:Learn how and when to remove this message 734:{\displaystyle \tau _{n+1}^{(0)}=ch^{2}} 415:is the error in the numerical solution. 3127: 2426:For embedded RK method, computation of 388:{\displaystyle y_{b}+\varepsilon =y(b)} 2012:Beginning with an initial stepsize of 1822:{\displaystyle 0.9\times {\text{tol}}} 2907:Then we compare the normalized error 748:is some constant of proportionality. 7: 44:adding citations to reliable sources 2705:{\displaystyle {\textrm {err}}_{n}} 2305:{\displaystyle t_{n+1}=t_{n}+h_{n}} 185:Consider the initial value problem 3116:Adaptive numerical differentiation 14: 2446:includes a lower order RK method 674:, it can be shown that (provided 3186:Numerical differential equations 3090:{\displaystyle {\tilde {\psi }}} 2468:{\displaystyle {\tilde {\psi }}} 130:is used in some methods for the 20: 2934:against 1 to get the predicted 336:) denote the exact solution at 245: 134:(including the special case of 31:needs additional citations for 3081: 3046: 3034: 3022: 3000: 2891: 2855: 2814: 2810: 2789: 2781: 2766: 2762: 2666: 2663: 2624: 2615: 2576: 2570: 2561: 2518: 2505: 2499: 2459: 2410: 2365: 2239: 2200: 1983: 1977: 1953: 1947: 1923: 1917: 1869: 1863: 1857: 1837: 1716: 1710: 1645: 1565: 1559: 1535: 1529: 1505: 1499: 1460: 1454: 1449: 1443: 1418: 1400: 1393:is constant over the interval 1357: 1345: 1334: 1328: 1304: 1298: 1259: 1253: 1097: 1091: 1054: 1002: 955: 949: 912: 886: 811: 805: 764: 758: 710: 704: 652: 646: 624: 605: 594: 588: 546: 520: 493: 487: 382: 376: 307:) and the initial conditions ( 255: 249: 239: 236: 230: 218: 209: 203: 1: 1373:Here, we assume error factor 1621:{\displaystyle {\text{tol}}} 408:{\displaystyle \varepsilon } 3207: 2001:This solution is actually 1466:{\displaystyle y^{(3)}(t)} 295:We are given the function 1628:is allowed, we could let 2095:Embedded error estimates 2007:Richardson extrapolation 3149:Numerical Recipes in C 3091: 3055: 2955: 2928: 2898: 2821: 2706: 2673: 2469: 2440: 2417: 2336: 2306: 2246: 2139: 2078: 2058: 2038: 1992: 1885: 1823: 1795: 1772: 1622: 1600: 1574: 1467: 1425: 1387: 1364: 1268: 1061: 919: 818: 771: 735: 661: 553: 409: 389: 275: 3161:Kendall E. Atkinson, 3092: 3056: 2956: 2954:{\displaystyle h_{n}} 2929: 2927:{\displaystyle E_{n}} 2899: 2822: 2707: 2674: 2470: 2441: 2439:{\displaystyle \psi } 2418: 2337: 2335:{\displaystyle h_{n}} 2307: 2247: 2140: 2138:{\displaystyle y_{n}} 2079: 2059: 2039: 2037:{\displaystyle h=b-a} 1993: 1886: 1824: 1796: 1773: 1623: 1601: 1575: 1468: 1426: 1388: 1365: 1269: 1062: 920: 819: 772: 736: 662: 554: 410: 390: 276: 136:numerical integration 3072: 2968: 2938: 2911: 2832: 2719: 2685: 2482: 2450: 2430: 2346: 2319: 2257: 2152: 2122: 2105:Runge–Kutta–Fehlberg 2068: 2048: 2016: 1898: 1833: 1829:in the next try. If 1805: 1785: 1639: 1610: 1590: 1480: 1435: 1397: 1377: 1279: 1072: 930: 831: 802: 755: 685: 569: 468: 399: 351: 192: 164:Runge–Kutta–Fehlberg 140:stability properties 55:"Adaptive step size" 40:improve this article 3111:Adaptive quadrature 1987: 1957: 1927: 1867: 1794:{\displaystyle 0.9} 1720: 1569: 1539: 1509: 1338: 1308: 1263: 1101: 959: 817:{\displaystyle (1)} 770:{\displaystyle (0)} 714: 656: 598: 497: 3191:Numerical analysis 3163:Numerical Analysis 3087: 3051: 2951: 2924: 2894: 2817: 2702: 2669: 2465: 2436: 2413: 2332: 2302: 2242: 2135: 2074: 2054: 2034: 1988: 1961: 1931: 1901: 1881: 1841: 1819: 1791: 1768: 1694: 1618: 1596: 1570: 1543: 1513: 1483: 1463: 1421: 1383: 1360: 1312: 1282: 1264: 1237: 1075: 1057: 933: 915: 814: 767: 731: 688: 657: 630: 572: 549: 471: 405: 385: 271: 156:Three-body problem 128:adaptive step size 124:numerical analysis 3084: 2882: 2863: 2852: 2753: 2743: 2727: 2693: 2573: 2521: 2490: 2462: 2077:{\displaystyle b} 2057:{\displaystyle a} 1879: 1817: 1726: 1685: 1616: 1599:{\displaystyle h} 1386:{\displaystyle c} 1235: 1209: 1186: 1152: 1121: 1050: 1024: 997: 982: 881: 853: 116: 115: 108: 90: 3198: 3135: 3132: 3096: 3094: 3093: 3088: 3086: 3085: 3077: 3060: 3058: 3057: 3052: 3050: 3049: 3033: 3020: 3019: 3010: 2999: 2998: 2980: 2979: 2960: 2958: 2957: 2952: 2950: 2949: 2933: 2931: 2930: 2925: 2923: 2922: 2903: 2901: 2900: 2895: 2890: 2889: 2884: 2883: 2880: 2876: 2871: 2870: 2865: 2864: 2861: 2854: 2853: 2850: 2844: 2843: 2826: 2824: 2823: 2818: 2813: 2808: 2807: 2792: 2784: 2779: 2778: 2769: 2755: 2754: 2751: 2745: 2744: 2741: 2735: 2734: 2729: 2728: 2725: 2711: 2709: 2708: 2703: 2701: 2700: 2695: 2694: 2691: 2678: 2676: 2675: 2670: 2662: 2661: 2649: 2648: 2636: 2635: 2614: 2613: 2601: 2600: 2588: 2587: 2575: 2574: 2566: 2554: 2553: 2535: 2534: 2523: 2522: 2514: 2498: 2497: 2492: 2491: 2488: 2474: 2472: 2471: 2466: 2464: 2463: 2455: 2445: 2443: 2442: 2437: 2422: 2420: 2419: 2414: 2409: 2408: 2390: 2389: 2377: 2376: 2358: 2357: 2341: 2339: 2338: 2333: 2331: 2330: 2311: 2309: 2308: 2303: 2301: 2300: 2288: 2287: 2275: 2274: 2251: 2249: 2248: 2243: 2238: 2237: 2225: 2224: 2212: 2211: 2196: 2195: 2183: 2182: 2170: 2169: 2144: 2142: 2141: 2136: 2134: 2133: 2101:Bogacki–Shampine 2083: 2081: 2080: 2075: 2063: 2061: 2060: 2055: 2043: 2041: 2040: 2035: 1997: 1995: 1994: 1989: 1986: 1975: 1956: 1945: 1926: 1915: 1890: 1888: 1887: 1882: 1880: 1877: 1872: 1866: 1855: 1840: 1828: 1826: 1825: 1820: 1818: 1815: 1800: 1798: 1797: 1792: 1777: 1775: 1774: 1769: 1767: 1763: 1756: 1752: 1745: 1744: 1740: 1731: 1727: 1725: 1724: 1719: 1708: 1683: 1682: 1627: 1625: 1624: 1619: 1617: 1614: 1605: 1603: 1602: 1597: 1579: 1577: 1576: 1571: 1568: 1557: 1538: 1527: 1508: 1497: 1472: 1470: 1469: 1464: 1453: 1452: 1430: 1428: 1427: 1424:{\displaystyle } 1422: 1392: 1390: 1389: 1384: 1369: 1367: 1366: 1361: 1337: 1326: 1307: 1296: 1273: 1271: 1270: 1265: 1262: 1251: 1236: 1228: 1223: 1222: 1210: 1202: 1197: 1196: 1191: 1187: 1179: 1163: 1162: 1157: 1153: 1145: 1132: 1131: 1126: 1122: 1114: 1100: 1089: 1066: 1064: 1063: 1058: 1053: 1052: 1051: 1043: 1027: 1026: 1025: 1017: 998: 990: 985: 984: 983: 975: 958: 947: 924: 922: 921: 916: 911: 910: 898: 897: 882: 874: 869: 868: 856: 855: 854: 846: 823: 821: 820: 815: 776: 774: 773: 768: 740: 738: 737: 732: 730: 729: 713: 702: 672:Taylor's theorem 666: 664: 663: 658: 655: 644: 623: 622: 597: 586: 558: 556: 555: 550: 545: 544: 532: 531: 510: 509: 496: 485: 418:For a sequence ( 414: 412: 411: 406: 394: 392: 391: 386: 363: 362: 280: 278: 277: 272: 270: 269: 202: 160:Romberg's method 111: 104: 100: 97: 91: 89: 48: 24: 16: 3206: 3205: 3201: 3200: 3199: 3197: 3196: 3195: 3176: 3175: 3144: 3142:Further reading 3139: 3138: 3133: 3129: 3124: 3107: 3070: 3069: 3021: 3011: 2984: 2971: 2966: 2965: 2941: 2936: 2935: 2914: 2909: 2908: 2877: 2858: 2835: 2830: 2829: 2793: 2770: 2722: 2717: 2716: 2688: 2683: 2682: 2653: 2640: 2627: 2605: 2592: 2579: 2539: 2511: 2485: 2480: 2479: 2448: 2447: 2428: 2427: 2394: 2381: 2368: 2349: 2344: 2343: 2322: 2317: 2316: 2292: 2279: 2260: 2255: 2254: 2229: 2216: 2203: 2187: 2174: 2155: 2150: 2149: 2125: 2120: 2119: 2097: 2066: 2065: 2046: 2045: 2014: 2013: 1896: 1895: 1831: 1830: 1803: 1802: 1783: 1782: 1690: 1686: 1677: 1676: 1675: 1671: 1667: 1663: 1637: 1636: 1608: 1607: 1588: 1587: 1478: 1477: 1438: 1433: 1432: 1395: 1394: 1375: 1374: 1277: 1276: 1214: 1174: 1173: 1140: 1139: 1109: 1108: 1070: 1069: 1031: 1005: 963: 928: 927: 902: 889: 860: 834: 829: 828: 800: 799: 797: 753: 752: 721: 683: 682: 608: 567: 566: 536: 523: 501: 466: 465: 460: 439: 427:) of values of 426: 397: 396: 354: 349: 348: 345: 319: 261: 195: 190: 189: 172: 112: 101: 95: 92: 49: 47: 37: 25: 12: 11: 5: 3204: 3202: 3194: 3193: 3188: 3178: 3177: 3174: 3173: 3159: 3143: 3140: 3137: 3136: 3126: 3125: 3123: 3120: 3119: 3118: 3113: 3106: 3103: 3083: 3080: 3064:The parameter 3062: 3061: 3048: 3045: 3042: 3039: 3036: 3032: 3028: 3024: 3018: 3014: 3009: 3005: 3002: 2997: 2994: 2991: 2987: 2983: 2978: 2974: 2948: 2944: 2921: 2917: 2905: 2904: 2893: 2888: 2875: 2869: 2857: 2847: 2842: 2838: 2827: 2816: 2812: 2806: 2803: 2800: 2796: 2791: 2787: 2783: 2777: 2773: 2768: 2764: 2761: 2758: 2748: 2738: 2733: 2699: 2680: 2679: 2668: 2665: 2660: 2656: 2652: 2647: 2643: 2639: 2634: 2630: 2626: 2623: 2620: 2617: 2612: 2608: 2604: 2599: 2595: 2591: 2586: 2582: 2578: 2572: 2569: 2563: 2560: 2557: 2552: 2549: 2546: 2542: 2538: 2533: 2530: 2527: 2520: 2517: 2510: 2507: 2504: 2501: 2496: 2461: 2458: 2435: 2412: 2407: 2404: 2401: 2397: 2393: 2388: 2384: 2380: 2375: 2371: 2367: 2364: 2361: 2356: 2352: 2329: 2325: 2315:The next step 2313: 2312: 2299: 2295: 2291: 2286: 2282: 2278: 2273: 2270: 2267: 2263: 2252: 2241: 2236: 2232: 2228: 2223: 2219: 2215: 2210: 2206: 2202: 2199: 2194: 2190: 2186: 2181: 2177: 2173: 2168: 2165: 2162: 2158: 2132: 2128: 2113:Dormand–Prince 2096: 2093: 2073: 2053: 2033: 2030: 2027: 2024: 2021: 1999: 1998: 1985: 1982: 1979: 1974: 1971: 1968: 1964: 1960: 1955: 1952: 1949: 1944: 1941: 1938: 1934: 1930: 1925: 1922: 1919: 1914: 1911: 1908: 1904: 1875: 1871: 1865: 1862: 1859: 1854: 1851: 1848: 1844: 1839: 1813: 1810: 1790: 1779: 1778: 1766: 1762: 1759: 1755: 1751: 1748: 1743: 1739: 1735: 1730: 1723: 1718: 1715: 1712: 1707: 1704: 1701: 1697: 1693: 1689: 1680: 1674: 1670: 1666: 1662: 1659: 1656: 1653: 1650: 1647: 1644: 1595: 1581: 1580: 1567: 1564: 1561: 1556: 1553: 1550: 1546: 1542: 1537: 1534: 1531: 1526: 1523: 1520: 1516: 1512: 1507: 1504: 1501: 1496: 1493: 1490: 1486: 1462: 1459: 1456: 1451: 1448: 1445: 1441: 1420: 1417: 1414: 1411: 1408: 1405: 1402: 1382: 1371: 1370: 1359: 1356: 1353: 1350: 1347: 1344: 1341: 1336: 1333: 1330: 1325: 1322: 1319: 1315: 1311: 1306: 1303: 1300: 1295: 1292: 1289: 1285: 1274: 1261: 1258: 1255: 1250: 1247: 1244: 1240: 1234: 1231: 1226: 1221: 1217: 1213: 1208: 1205: 1200: 1195: 1190: 1185: 1182: 1177: 1172: 1169: 1166: 1161: 1156: 1151: 1148: 1143: 1138: 1135: 1130: 1125: 1120: 1117: 1112: 1107: 1104: 1099: 1096: 1093: 1088: 1085: 1082: 1078: 1067: 1056: 1049: 1046: 1041: 1038: 1034: 1030: 1023: 1020: 1015: 1012: 1008: 1004: 1001: 996: 993: 988: 981: 978: 973: 970: 966: 962: 957: 954: 951: 946: 943: 940: 936: 925: 914: 909: 905: 901: 896: 892: 888: 885: 880: 877: 872: 867: 863: 859: 852: 849: 844: 841: 837: 813: 810: 807: 792: 766: 763: 760: 742: 741: 728: 724: 720: 717: 712: 709: 706: 701: 698: 695: 691: 668: 667: 654: 651: 648: 643: 640: 637: 633: 629: 626: 621: 618: 615: 611: 607: 604: 601: 596: 593: 590: 585: 582: 579: 575: 560: 559: 548: 543: 539: 535: 530: 526: 522: 519: 516: 513: 508: 504: 500: 495: 492: 489: 484: 481: 478: 474: 456: 435: 422: 404: 384: 381: 378: 375: 372: 369: 366: 361: 357: 343: 315: 282: 281: 268: 264: 260: 257: 254: 251: 248: 244: 241: 238: 235: 232: 229: 226: 223: 220: 217: 214: 211: 208: 205: 201: 198: 171: 168: 114: 113: 28: 26: 19: 13: 10: 9: 6: 4: 3: 2: 3203: 3192: 3189: 3187: 3184: 3183: 3181: 3172: 3171:0-471-62489-6 3168: 3164: 3160: 3158: 3157:0-521-43108-5 3154: 3150: 3146: 3145: 3141: 3131: 3128: 3121: 3117: 3114: 3112: 3109: 3108: 3104: 3102: 3098: 3078: 3067: 3043: 3040: 3037: 3030: 3026: 3016: 3012: 3007: 3003: 2995: 2992: 2989: 2985: 2981: 2976: 2972: 2964: 2963: 2962: 2946: 2942: 2919: 2915: 2886: 2873: 2867: 2845: 2840: 2836: 2828: 2804: 2801: 2798: 2794: 2785: 2775: 2771: 2756: 2746: 2736: 2731: 2715: 2714: 2713: 2697: 2658: 2654: 2650: 2645: 2641: 2637: 2632: 2628: 2621: 2618: 2610: 2606: 2602: 2597: 2593: 2589: 2584: 2580: 2567: 2558: 2555: 2550: 2547: 2544: 2540: 2536: 2531: 2528: 2525: 2515: 2508: 2502: 2494: 2478: 2477: 2476: 2456: 2433: 2424: 2405: 2402: 2399: 2395: 2391: 2386: 2382: 2378: 2373: 2369: 2362: 2359: 2354: 2350: 2327: 2323: 2297: 2293: 2289: 2284: 2280: 2276: 2271: 2268: 2265: 2261: 2253: 2234: 2230: 2226: 2221: 2217: 2213: 2208: 2204: 2197: 2192: 2188: 2184: 2179: 2175: 2171: 2166: 2163: 2160: 2156: 2148: 2147: 2146: 2130: 2126: 2116: 2114: 2110: 2106: 2102: 2094: 2092: 2089: 2087: 2071: 2051: 2031: 2028: 2025: 2022: 2019: 2010: 2008: 2004: 1980: 1972: 1969: 1966: 1962: 1958: 1950: 1942: 1939: 1936: 1932: 1928: 1920: 1912: 1909: 1906: 1902: 1894: 1893: 1892: 1873: 1860: 1852: 1849: 1846: 1842: 1811: 1808: 1788: 1764: 1760: 1757: 1753: 1749: 1746: 1741: 1737: 1733: 1728: 1721: 1713: 1705: 1702: 1699: 1695: 1691: 1687: 1678: 1672: 1664: 1657: 1654: 1651: 1648: 1642: 1635: 1634: 1633: 1632:evolve like: 1631: 1593: 1584: 1562: 1554: 1551: 1548: 1544: 1540: 1532: 1524: 1521: 1518: 1514: 1510: 1502: 1494: 1491: 1488: 1484: 1476: 1475: 1474: 1457: 1446: 1439: 1415: 1412: 1409: 1406: 1403: 1380: 1354: 1351: 1348: 1342: 1339: 1331: 1323: 1320: 1317: 1313: 1309: 1301: 1293: 1290: 1287: 1283: 1275: 1256: 1248: 1245: 1242: 1238: 1232: 1229: 1224: 1219: 1215: 1211: 1206: 1203: 1198: 1193: 1188: 1183: 1180: 1175: 1170: 1167: 1164: 1159: 1154: 1149: 1146: 1141: 1136: 1133: 1128: 1123: 1118: 1115: 1110: 1105: 1102: 1094: 1086: 1083: 1080: 1076: 1068: 1047: 1044: 1039: 1036: 1032: 1028: 1021: 1018: 1013: 1010: 1006: 999: 994: 991: 986: 979: 976: 971: 968: 964: 960: 952: 944: 941: 938: 934: 926: 907: 903: 899: 894: 890: 883: 878: 875: 870: 865: 861: 857: 850: 847: 842: 839: 835: 827: 826: 825: 808: 795: 791: 787: 783: 780:The value of 778: 761: 749: 747: 726: 722: 718: 715: 707: 699: 696: 693: 689: 681: 680: 679: 677: 673: 649: 641: 638: 635: 631: 627: 619: 616: 613: 609: 602: 599: 591: 583: 580: 577: 573: 565: 564: 563: 541: 537: 533: 528: 524: 517: 514: 511: 506: 502: 498: 490: 482: 479: 476: 472: 464: 463: 462: 459: 455: 451: 447: 443: 438: 434: 430: 425: 421: 416: 402: 379: 373: 370: 367: 364: 359: 355: 346: 339: 335: 331: 327: 324: =  323: 318: 314: 310: 306: 302: 298: 293: 291: 287: 266: 262: 258: 252: 246: 242: 233: 227: 224: 221: 215: 212: 206: 199: 196: 188: 187: 186: 183: 181: 177: 169: 167: 165: 161: 157: 153: 149: 145: 141: 137: 133: 129: 125: 121: 110: 107: 99: 88: 85: 81: 78: 74: 71: 67: 64: 60: 57: â€“  56: 52: 51:Find sources: 45: 41: 35: 34: 29:This article 27: 23: 18: 17: 3162: 3148: 3130: 3099: 3065: 3063: 2906: 2681: 2425: 2314: 2117: 2098: 2090: 2086:Euler method 2011: 2002: 2000: 1780: 1629: 1585: 1582: 1372: 793: 789: 785: 781: 779: 750: 745: 743: 675: 669: 561: 457: 453: 449: 445: 441: 436: 432: 428: 423: 419: 417: 341: 337: 333: 329: 325: 321: 316: 312: 308: 304: 300: 296: 294: 289: 285: 283: 184: 176:Euler method 173: 152:Euler method 148:Kepler orbit 127: 117: 102: 96:October 2012 93: 83: 76: 69: 62: 50: 38:Please help 33:verification 30: 2003:third order 180:Runge–Kutta 144:A-stability 120:mathematics 3180:Categories 3122:References 340:, and let 66:newspapers 3082:~ 3079:ψ 2993:− 2802:− 2757:⋅ 2622:ψ 2619:− 2571:~ 2568:ψ 2537:− 2519:~ 2460:~ 2457:ψ 2434:ψ 2403:− 2198:ψ 2109:Cash–Karp 2029:− 1963:τ 1843:τ 1812:× 1696:τ 1658:× 1652:× 1646:→ 1545:τ 1511:− 1314:τ 1239:τ 1077:τ 690:τ 628:− 574:τ 403:ε 368:ε 3105:See also 395:, where 200:′ 142:such as 670:and by 431:, with 170:Example 80:scholar 3169:  3155:  744:where 328:. Let 284:where 126:, an 82:  75:  68:  61:  53:  461:) as 87:JSTOR 73:books 3167:ISBN 3153:ISBN 2851:norm 2752:Rtol 2742:Atol 2111:and 1874:< 1781:The 288:and 162:and 122:and 59:news 2881:tol 2862:err 2760:max 2726:tol 2692:err 2489:err 2064:to 1878:tol 1816:tol 1809:0.9 1789:0.9 1750:0.3 1684:tol 1669:max 1661:min 1649:0.9 1615:tol 118:In 42:by 3182:: 2961:: 2423:. 2145:: 2107:, 2103:, 2088:. 2009:. 796:+1 777:. 446:nh 444:+ 440:= 311:, 3066:q 3047:) 3044:1 3041:+ 3038:q 3035:( 3031:/ 3027:1 3023:) 3017:n 3013:E 3008:/ 3004:1 3001:( 2996:1 2990:n 2986:h 2982:= 2977:n 2973:h 2947:n 2943:h 2920:n 2916:E 2892:) 2887:n 2874:/ 2868:n 2856:( 2846:= 2841:n 2837:E 2815:) 2811:| 2805:1 2799:n 2795:y 2790:| 2786:, 2782:| 2776:n 2772:y 2767:| 2763:( 2747:+ 2737:= 2732:n 2698:n 2667:) 2664:) 2659:n 2655:h 2651:, 2646:n 2642:y 2638:, 2633:n 2629:t 2625:( 2616:) 2611:n 2607:h 2603:, 2598:n 2594:y 2590:, 2585:n 2581:t 2577:( 2562:( 2559:h 2556:= 2551:1 2548:+ 2545:n 2541:y 2532:1 2529:+ 2526:n 2516:y 2509:= 2506:) 2503:h 2500:( 2495:n 2411:) 2406:1 2400:n 2396:h 2392:, 2387:n 2383:y 2379:, 2374:n 2370:t 2366:( 2363:g 2360:= 2355:n 2351:h 2328:n 2324:h 2298:n 2294:h 2290:+ 2285:n 2281:t 2277:= 2272:1 2269:+ 2266:n 2262:t 2240:) 2235:n 2231:h 2227:, 2222:n 2218:y 2214:, 2209:n 2205:t 2201:( 2193:n 2189:h 2185:+ 2180:n 2176:y 2172:= 2167:1 2164:+ 2161:n 2157:y 2131:n 2127:y 2072:b 2052:a 2032:a 2026:b 2023:= 2020:h 1984:) 1981:1 1978:( 1973:1 1970:+ 1967:n 1959:+ 1954:) 1951:1 1948:( 1943:1 1940:+ 1937:n 1933:y 1929:= 1924:) 1921:2 1918:( 1913:1 1910:+ 1907:n 1903:y 1870:| 1864:) 1861:1 1858:( 1853:1 1850:+ 1847:n 1838:| 1765:) 1761:2 1758:, 1754:) 1747:, 1742:2 1738:/ 1734:1 1729:) 1722:| 1717:) 1714:1 1711:( 1706:1 1703:+ 1700:n 1692:| 1688:2 1679:( 1673:( 1665:( 1655:h 1643:h 1630:h 1594:h 1566:) 1563:1 1560:( 1555:1 1552:+ 1549:n 1541:= 1536:) 1533:0 1530:( 1525:1 1522:+ 1519:n 1515:y 1506:) 1503:1 1500:( 1495:1 1492:+ 1489:n 1485:y 1461:) 1458:t 1455:( 1450:) 1447:3 1444:( 1440:y 1419:] 1416:h 1413:+ 1410:t 1407:, 1404:t 1401:[ 1381:c 1358:) 1355:h 1352:+ 1349:t 1346:( 1343:y 1340:= 1335:) 1332:1 1329:( 1324:1 1321:+ 1318:n 1310:+ 1305:) 1302:1 1299:( 1294:1 1291:+ 1288:n 1284:y 1260:) 1257:0 1254:( 1249:1 1246:+ 1243:n 1233:2 1230:1 1225:= 1220:2 1216:h 1212:c 1207:2 1204:1 1199:= 1194:2 1189:) 1184:2 1181:h 1176:( 1171:c 1168:2 1165:= 1160:2 1155:) 1150:2 1147:h 1142:( 1137:c 1134:+ 1129:2 1124:) 1119:2 1116:h 1111:( 1106:c 1103:= 1098:) 1095:1 1092:( 1087:1 1084:+ 1081:n 1055:) 1048:2 1045:1 1040:+ 1037:n 1033:y 1029:, 1022:2 1019:1 1014:+ 1011:n 1007:t 1003:( 1000:f 995:2 992:h 987:+ 980:2 977:1 972:+ 969:n 965:y 961:= 956:) 953:1 950:( 945:1 942:+ 939:n 935:y 913:) 908:n 904:y 900:, 895:n 891:t 887:( 884:f 879:2 876:h 871:+ 866:n 862:y 858:= 851:2 848:1 843:+ 840:n 836:y 812:) 809:1 806:( 794:n 790:t 788:( 786:y 782:c 765:) 762:0 759:( 746:c 727:2 723:h 719:c 716:= 711:) 708:0 705:( 700:1 697:+ 694:n 676:f 653:) 650:0 647:( 642:1 639:+ 636:n 632:y 625:) 620:1 617:+ 614:n 610:t 606:( 603:y 600:= 595:) 592:0 589:( 584:1 581:+ 578:n 547:) 542:n 538:y 534:, 529:n 525:t 521:( 518:f 515:h 512:+ 507:n 503:y 499:= 494:) 491:0 488:( 483:1 480:+ 477:n 473:y 458:n 454:t 452:( 450:y 442:a 437:n 433:t 429:t 424:n 420:t 383:) 380:b 377:( 374:y 371:= 365:+ 360:b 356:y 344:b 342:y 338:b 334:b 332:( 330:y 326:b 322:t 317:a 313:y 309:a 305:y 303:, 301:t 299:( 297:f 290:f 286:y 267:a 263:y 259:= 256:) 253:a 250:( 247:y 243:, 240:) 237:) 234:t 231:( 228:y 225:, 222:t 219:( 216:f 213:= 210:) 207:t 204:( 197:y 109:) 103:( 98:) 94:( 84:· 77:· 70:· 63:· 36:.

Index


verification
improve this article
adding citations to reliable sources
"Adaptive step size"
news
newspapers
books
scholar
JSTOR
Learn how and when to remove this message
mathematics
numerical analysis
numerical solution of ordinary differential equations
numerical integration
stability properties
A-stability
Kepler orbit
Euler method
Three-body problem
Romberg's method
Runge–Kutta–Fehlberg
Euler method
Runge–Kutta
Taylor's theorem
Richardson extrapolation
Euler method
Bogacki–Shampine
Runge–Kutta–Fehlberg
Cash–Karp

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.

↑