【考研数学:高数3】一元函数微分学的概念

目录

前言

一、导数

1.概念

2.例题

二、导数的几何意义

三、高阶导数

四、微分的概念

1.概念

2.例题

结尾


ID:HL_5461

前言

本文为张宇老师《基础三十讲》高数第三讲的自用笔记。不做商用,侵删致歉!

例题的序号,以1.3为例,意思是第一个的第3个例题,总之从标题一(一、二、三酱紫的)往里数就是。

从本章开始是连续五章的关于微分学的内容,分概念、计算和应用。其中应用最重要分三章。本章先说应用

这一章没什么概念上的难点,就是第一部分的例题真的很多还很重要


一、导数

1.概念

eq?f%27%28x_0%29%3D%5Cfrac%7Bdy%7D%7Bdx%7D%7C_%7Bx%3Dx_0%7D%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7Bf%28x_0+%5CDelta%20x%29-f%28x_0%29%7D%7B%5CDelta%20x%7D%3D%5Clim_%7Bx%5Crightarrow%20x_0%7D%5Cfrac%7Bf%28x%29-f%28x_0%29%7D%7Bx-x_0%7D

①函数一点可导的充要条件:左右导数存在且相等

②函数一点可导的必要条件:函数在该点连续

证:

eq?f%27%28x_0%29%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7Bf%28x_0+%5CDelta%20x%29-f%28x_0%29%7D%7B%5CDelta%20x%7D%3DA

eq?%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Bf%28x_0+%5CDelta%20x%29-f%28x_0%29%5D%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7DA%5Ccdot%20%5CDelta%20x%3D0

所以函数在eq?x_0连续

③求一次导奇变偶偶变奇

证:

eq?f%28x%29为奇函数,则eq?f%28-x%29%3D-f%28x%29

eq?f%27%28-x%29%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7Bf%28-x+%5CDelta%20x%29-f%28-x%29%7D%7B%5CDelta%20x%7D%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7Bf%28x-%5CDelta%20x%29-f%28x%29%7D%7B-%5CDelta%20x%7D%3Df%27%28x%29

偶函数同理

④以T为周期的函数,导函数也以T为周期

证:

eq?f%27%28x+T%29%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7Bf%28x+T+%5CDelta%20x%29-f%28x+T%29%7D%7B%5CDelta%20x%7D%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7Bf%28x+%5CDelta%20x%29-f%28x%29%7D%7B%5CDelta%20x%7D%3Df%27%28x%29

⑤一点可导推不出邻域可导(参照连续性)

2.例题

例1.1:设eq?f%28x%29%3D%5Cfrac%7B1%7D%7B2%5Ex+1%7D%2Cx%5Cin%20R,则eq?f%5E%7B%282024%29%7D%280%29%3D

思路:

大于三阶一般不考虑求了吧……见高数一壹、五、3.例1.5.3.1,eq?%5Cfrac%7B1%7D%7Ba%5Ex+1%7D-%5Cfrac%7B1%7D%7B2%7D是奇函数,所以原式构造,然后奇函数求偶数阶导仍是奇函数,奇函数在0处值为0

解:

eq?g%28x%29%3D%5Cfrac%7B1%7D%7Ba%5Ex+1%7D-%5Cfrac%7B1%7D%7B2%7D

eq?f%28x%29%3Dg%28x%29+%5Cfrac%7B1%7D%7B2%7D%5CRightarrow%20f%5E%7B%282024%29%7D%28x%29%3Dg%5E%7B%282024%29%7D%28x%29

eq?g%5E%7B%282024%29%7D%28x%29为奇函数,所以eq?g%5E%7B%282024%29%7D%280%29%3D0

eq?%5Ctherefore%20f%5E%7B%282024%29%7D%280%29%3D0

例1.2:设eq?f%28x%29eq?x%3D0的某邻域内有定义,并且eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%5Cleqslant%201-cos%20%5C%20x,则eq?f%28x%29eq?x%3D0处连续吗?可导吗?如果可导求导数

思路:

连续性,

绝对值加小于等于可以考虑夹逼准则然后用eq?%5Clim_%7Bx%5Crightarrow%20x_0%20%7Df%28x%29%3D0%5CLeftrightarrow%20%5Clim_%7Bx%5Crightarrow%20x_0%20%7D%5Cleft%20%7Cf%28x%29%20%5Cright%20%7C%3D0(见高数2,二、③)去掉绝对值,算0处极限。再原不等式直接代入0算0处函数

可导,

先假设可导,原不等式修改一下表示导数公式,算的出就可导,顺便也把该点导数值也算了,算不出就不可导

解:

eq?0%5Cleqslant%20%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%5Cleqslant%201-cos%20%5C%20x,由夹逼准则,eq?%5Clim_%7Bx%5Crightarrow0%20%7D%5Cleft%20%7Cf%28x%29%20%5Cright%20%7C%3D0%5CRightarrow%20%5Clim_%7Bx%5Crightarrow0%20%7Df%28x%29%20%3D0

eq?0%5Cleqslant%20%5Cleft%20%7C%20f%280%29%20%5Cright%20%7C%5Cleqslant%201-cos%20%5C%200%3D0%5CRightarrow%20f%280%29%3D0

eq?%5Cbecause%20f%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7Df%28x%29%3D0%20%5C%20%5Ctherefore%20f%28x%29连续

eq?f%27%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7Bf%28x%29-f%280%29%7D%7Bx-0%7D%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7Bf%28x%29%7D%7Bx%7D

eq?0%5Cleqslant%20%5Cleft%20%7C%20%5Cfrac%7Bf%28x%29%7D%7Bx%7D%20%5Cright%20%7C%5Cleqslant%20%5Cfrac%7B1-cos%20%5C%20x%7D%7Bx%7D,由夹逼准则,eq?%5Clim_%7Bx%5Crightarrow0%20%7D%5Cleft%20%7C%20%5Cfrac%7Bf%28x%29%7D%7Bx%7D%20%5Cright%20%7C%3D0%5CRightarrow%20%5Clim_%7Bx%5Crightarrow0%20%7D%20%5Cfrac%7Bf%28x%29%7D%7Bx%7D%20%3D0%5CRightarrow%20f%27%280%29%3D0

例1.3:设eq?f%28x%29eq?x%3Da处连续,eq?F%28x%29%3Df%28x%29%5Cleft%20%7C%20x-a%20%5Cright%20%7C,则当eq?F%28x%29eq?x%3Da处可导的充要条件是eq?f%28a%29等于几?

思路:

可导的充要条件是左右导数存在且相等,顺着这个思路下去,写出a点左右导数然后令它们相等就可以了

解:

eq?F%28x%29%3D%5Cleft%5C%7B%5Cbegin%7Bmatrix%7D%20%28x-a%29f%28x%29%2C%20%26%20x%3Ea%26%20%5C%5C%20%28a-x%29f%28x%29%2C%26%20x%3Ca%26%20%5C%5C%200%2C%26%20x%3Da%26%20%5Cend%7Bmatrix%7D%5Cright.

eq?F%27_%7B-%7D%28x%29%3Df%28a%29eq?F%27_%7B+%7D%28x%29%3D-f%28a%29

eq?F%27_%7B-%7D%28x%29%3DF_%7B+%7D%28x%29%5CRightarrow%20f%28a%29%3D0

eq?f%28a%29%3D0%5CLeftrightarrow%20F%28x%29%3Df%28x%29%5Cleft%20%7C%20x-a%20%5Cright%20%7Cx=a处可导。是一个很重要的结论,嗯,记住就好

例1.4:设eq?f_1%28x%29%3D%28x%5E2-1%29%5Cleft%20%7C%20x%5E3+x%5E2-2x-2%20%5Cright%20%7Ceq?f_2%28x%29%3D%28x%5E2-1%29%5Cleft%20%7C%20x%5E3-2x%5E2-x+2%20%5Cright%20%7Ceq?f_3%28x%29%3D%28x%5E2-1%29%5Cleft%20%7C%20x%5E3+3x%5E2-2x-6%20%5Cright%20%7Ceq?f_i%28x%29的不可导点数为eq?n_i,则eq?n_i的大小关系为?

思路:

这题用的是上一题的结论,所以看不可导点数直接将绝对值内的因式与前面的平方因式匹配起来就好了

解:

eq?f_1%28x%29%3D%28x-1%29%28x+1%29%5Cleft%20%7C%20%28x+%5Csqrt%7B2%7D%29%28x-%5Csqrt%7B2%7D%29%28x+1%29%5Cright%20%7C

eq?f_2%28x%29%3D%28x-1%29%28x+1%29%5Cleft%20%7C%20%28x+1%29%28x-1%29%28x-2%29%5Cright%20%7C

eq?f_3%28x%29%3D%28x-1%29%28x+1%29%5Cleft%20%7C%20%28x+%5Csqrt%7B2%7D%29%28x-%5Csqrt%7B2%7D%29%28x+3%29%5Cright%20%7C

根据上一题的结论:

eq?n_1%3D2(绝对值内,-1能和前面匹配,eq?%5Cpm%20%5Csqrt%7B2%7D不行)

eq?n_2%3D1(绝对值内,eq?%5Cpm%201能和前面匹配,2不行)

eq?n_3%3D3(绝对值内,eq?%5Cpm%20%5Csqrt%7B2%7D和3都不能与前面匹配)

eq?n_3%3En_1%3En_2

例1.5: 设eq?f%28x%29eq?x%3D0处可导,eq?f%28%5Cfrac%7B1%7D%7Bn%7D%29%3D%5Cfrac%7B2%7D%7Bn%7D%2Cn%3D1%2C2%2C...,求eq?f%27%280%29

思路:

这个导数肯定得列导数定义式来求,eq?f%27%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7Bf%28x%29-f%280%29%7D%7Bx-0%7D。然后就会发现差一个eq?f%280%29

这个直接代入是肯定求不出来的,因为eq?%5Cfrac%7B1%7D%7Bn%7D不可能等于0。但随后就想到虽然不能等于0但是可以趋于0,可不可以用极限等于eq?f%280%29?由于在0处可导,由可导必连续,连续函数在一点处的极限等于该点函数值,很显然,答案是肯定的。

随后的问题是,这个是数列而不是函数,联系数列和函数的那就只有海涅定理了,由海涅定理,可以用数列极限表示函数极限求eq?f%280%29

代入以后又回到了原本的公式eq?f%27%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7Bf%28x%29-f%280%29%7D%7Bx-0%7D,没有eq?f%28x%29表达式,还是得用eq?f%28%5Cfrac%7B1%7D%7Bn%7D%29,还是数列极限与函数极限的问题,再用一次海涅定理完美解决。

解:

eq?f%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7Df%28x%29%3D%5Clim_%7B%5Cfrac%7B1%7D%7Bn%7D%5Crightarrow%200%7Df%28%5Cfrac%7B1%7D%7Bn%7D%29%3D%5Clim_%7B%5Cfrac%7B1%7D%7Bn%7D%5Crightarrow%200%7D%5Cfrac%7B2%7D%7Bn%7D%3D0

eq?f%27%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7Bf%28x%29-f%280%29%7D%7Bx-0%7D%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7Bf%28x%29%7D%7Bx%7D%3D%5Clim_%7B%5Cfrac%7B1%7D%7Bn%7D%5Crightarrow%200%7D%5Cfrac%7B%5Cfrac%7B2%7D%7Bn%7D%7D%7B%5Cfrac%7B1%7D%7Bn%7D%7D%3D2

例1.6: 设eq?f%28x%29可导,eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7Ceq?x%3D0处不可导,则eq?f%280%29eq?f%27%280%29等不等于0?

思路:

老实说第一眼看到这题时我想多了呃呃,我想到了例1.4的结论,结果发现用不上去。然后我就试着写eq?f%27%280%29的定义式,因为我以为第一句话可以扩展成eq?f%28x%29在0处可导,然后根据定义式里的eq?f%280%29看它能不能等于0。emmm白忙活一场,毕竟人家公式里还有个eq?f%28x%29是吧(我也不清楚我当时怎么就脑抽了)……

总之以上以失败告罄以后就开始折腾eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C,控制住准备写定义式的蠢蠢欲动的小手……虽然现在感觉当时这个思路才应该是正常的,毕竟不可导我应该想到左右导不等才是……果然脑抽。然后就直接求导数表达式了,eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%27%3D%28%5Csqrt%7Bf%5E2%28x%29%7D%29%27%3D%5Cfrac%7Bf%28x%29%5Ccdot%20f%27%28x%29%7D%7B%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%7D,再代入0,总算有点眉目了,eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%27%7C_%7Bx%3D0%7D%3D%5Cfrac%7Bf%280%29%5Ccdot%20f%27%280%29%7D%7B%5Cleft%20%7C%20f%280%29%20%5Cright%20%7C%7D不存在,很明显eq?f%280%29%3D0

但后面就做不下去了,算是钻了牛角尖,坚信这个条件已经用完了,然后拼命在“eq?f%28x%29可导”里面试图求eq?f%27%280%29(谁给我的自信哦)

总之最后答案是绝对值导数不存在还要继续用,应该用定义式然后左右导数不等,嗯,酱紫

但是这题也可以从几何上理解,如下表:

eq?f%280%29%3D0eq?f%27%280%29%3D0d7058bba1904413a96af06b5d7a4b6bd.png
eq?f%27%280%29%5Cneq%200a4e8294b3c54402ea800e02f77b043bd.png
eq?f%280%29%5Cneq%200519343e4cd9f40e6b6cb52d11266e35a.png

解:

eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%27%3D%28%5Csqrt%7Bf%5E2%28x%29%7D%29%27%3D%5Cfrac%7Bf%28x%29%5Ccdot%20f%27%28x%29%7D%7B%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%7D

eq?%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C%27%7C_%7Bx%3D0%7D%3D%5Cfrac%7Bf%280%29%5Ccdot%20f%27%280%29%7D%7B%5Cleft%20%7C%20f%280%29%20%5Cright%20%7C%7D%5CRightarrow%20f%280%29%3D0

eq?g%28x%29%3D%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C

eq?g%27_+%280%29%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cfrac%7B%5Cleft%20%7C%20f%28x%29%20%5Cright%20%7C-%5Cleft%20%7Cf%280%29%20%5Cright%20%7C%7D%7Bx-0%7D%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cleft%20%7C%20%5Cfrac%7Bf%28x%29%7D%7Bx%7D%20%5Cright%20%7C%3D%5Clim_%7Bx%5Crightarrow%200%7D%5Cleft%20%7C%20%5Cfrac%7Bf%28x%29-f%280%29%7D%7Bx-0%7D%20%5Cright%20%7C%3D%5Cleft%20%7C%20f%27%280%29%20%5Cright%20%7C

(这步变化重点)

同理eq?g%27_-%280%29%3D-%5Cleft%20%7C%20f%27%280%29%20%5Cright%20%7C

eq?%5Cbecause%20g%27_+%280%29%5Cneq%20g%27_-%280%29%5C%20%5Ctherefore%20%5Cleft%20%7C%20f%27%280%29%20%5Cright%20%7C%5Cneq%20-%5Cleft%20%7C%20f%27%280%29%20%5Cright%20%7C%5CRightarrow%20f%27%280%29%5Cneq%200


二、导数的几何意义

①切线方程:eq?y-y_0%3Df%27%28x_0%29%28x-x_0%29

②法线方程:eq?y-y_0%3D-%5Cfrac%7B1%7D%7Bf%27%28x_0%29%7D%28x-x_0%29%28f%27%28x_0%29%5Cneq%200%29

③在一点出现两条单侧切线,则称该点为一个角点

如图

4b5322573d8746a992ae1773fdb2e468.png

点O为一个“角点”

eq?f%27%28x_0%29%3D+%5Cinftyeq?-%5Cinfty,称为无穷导数,在高等数学中视作导数不存在

如图

741530d0e2f04cec8beee0ef0c9b696c.png


三、高阶导数

eq?f%28x%29eq?x_0有n阶导数,则在eq?x_0有1~(n-1)阶的各阶导数

eq?f%5E%7B%28n%29%7D%28x_0%29存在,则eq?f%5E%7B%28n-1%29%7D%28x_0%29eq?x_0的邻域内有定义,在eq?x_0处连续

③当eq?f%27%28x_0%29%3D0eq?f%27%27%28x_0%29%5Cneq%200时,eq?f%27%27%28x_0%29%3C0%28%3E0%29,则eq?f%28x%29eq?x_0处取得极大(极小)值

证:

以大于0为例,

eq?f%27%27%28x_0%29%3D%5Clim_%7Bx%5Crightarrow%20x_0%7D%5Cfrac%7Bf%27%28x%29-f%27%28x_0%29%7D%7Bx-x_0%7D%3E0

由极限保号性,在eq?x_0的去心邻域内,eq?%5Cfrac%7Bf%27%28x%29-f%27%28x_0%29%7D%7Bx-x_0%7D%3E0

eq?x%5Cin%20%28%5Cdelta%20%2Cx_0%29时,eq?x-x_0%3C0%5CRightarrow%20f%27%28x%29-f%27%28x_0%29%3C0%5CRightarrow%20f%27%28x%29%3Cf%27%28x_0%29%3D0

eq?x%5Cin%20%28x_0%2C%5Cdelta%29时,eq?x-x_0%3C0%5CRightarrow%20f%27%28x%29-f%27%28x_0%29%3E0%5CRightarrow%20f%27%28x%29%3Ef%27%28x_0%29%3D0

eq?x_0的去心邻域内函数先减后增,所以eq?x_0为极小值


四、微分的概念

1.概念

“可微”即在一点处可以用切线增量代替曲线增量,即eq?%5CDelta%20y%3DA%5CDelta%20x+o%28%5CDelta%20x%29,其中eq?%5Cleft%5C%7B%5Cbegin%7Bmatrix%7D%20dy%7C_%7Bx%3Dx_0%7D%3DA%5CDelta%20x%3Df%27%28x_0%29%5CDelta%20x%5C%5C%20dx%3D%5CDelta%20x%20%5Cend%7Bmatrix%7D%5Cright.eq?o%28%5CDelta%20x%29eq?%5CDelta%20x%5Crightarrow%200时比eq?%5CDelta%20x更高阶的无穷小

如图:

e17afe5c5b42447d91b2e60cc8fc9d4a.jpeg

①线性主部,高阶无穷小

281c4205fc8949e5a962d69be33da4e4.png

②通过可导判断可微

③作极限证明是eq?%5CDelta%20x%5Crightarrow%200时比eq?%5CDelta%20x更高阶的无穷小判断可微

eq?%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7B%5CDelta%20y-A%5CDelta%20x%7D%7B%5CDelta%20x%7D若为0则可微

其中,eq?%5Cleft%5C%7B%5Cbegin%7Bmatrix%7D%20%5CDelta%20y%3Df%28x_0+%5CDelta%20x%29-f%28x_0%29%5C%5C%20A%5CDelta%20x%3Df%27%28x_0%29%5CDelta%20x%20%5Cend%7Bmatrix%7D%5Cright.

2.例题

例4.1:设函数eq?f%28u%29可导,且eq?y%3Df%28x%5E2%29,当自变量eq?xeq?x%3D-1处取得增量eq?%5CDelta%20x%3D-0.1时,相应的函数增量eq?%5CDelta%20y的线性主部为0.1,求eq?f%27%281%29

思路:

这题最重要的是线性主部的定义,eq?dy%7C_%7Bx%3Dx_0%7D%3DA%5CDelta%20x%3Df%27%28x_0%29%5CDelta%20x,这个地方要注意是y对x求导,而这题的y是一个复合函数,所以需要用复合函数求导表示线性主部

解:

eq?f%27%28-1%29%3Df%27%28x%5E2%29%5Ccdot%202x%7C_%7Bx%3D-1%7D

eq?dy%7C_%7Bx%3D-1%7D%3Df%27%28x%5E2%29%5Ccdot%202x%7C_%7Bx%3D-1%7D%5Ccdot%20%5CDelta%20x

eq?%5Ctherefore%200.1%3D-2f%281%29%5Ccdot%20%28-0.1%29%5CRightarrow%20f%27%281%29%3D0.5

例4.2:设函数eq?f%28x%29eq?x%3D1处可导,且eq?%5CDelta%20f%281%29eq?f%28x%29在增量为eq?%5CDelta%20x时的函数值增量,求eq?%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7B%5CDelta%20f%281%29-df%281%29%7D%7B%5CDelta%20x%7D

思路:

其实这题直接概念,eq?%5CDelta%20y%3Ddy+o%28%5CDelta%20x%29秒了,嗯……但还是可以列个式子严谨一点

解:

eq?%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7B%5CDelta%20f%281%29-df%281%29%7D%7B%5CDelta%20x%7D%3D%5Clim_%7B%5CDelta%20x%5Crightarrow%200%7D%5Cfrac%7B%5CDelta%20y%3Ddf%281%29+o%28%5CDelta%20x%29-df%281%29%7D%7B%5CDelta%20x%7D%3D0


结尾

这部分概念超简单的说,但是这几题可以好好看看,嗯,酱紫

评论 1
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

是兰兰呀~

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付:¥1
获取中
扫码支付

您的余额不足,请更换扫码支付或充值

打赏作者

实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值