这是用户在 2024-6-3 20:58 为 https://app.immersivetranslate.com/pdf-pro/620e989a-df68-41be-a1a7-f70924a6f08c 保存的双语快照页面,由 沉浸式翻译 提供双语支持。了解如何保存?
2024_06_03_c300222b9052e0d247aag

Basic Mathematics  基本数学

Contents 目录

1 Basic Skills ..... 2
1 基本技能 ..... 2

1.1 Practice Questions ..... 2
1.1 练习题 ..... 2

2 Linear Algebra ..... 3
2 线性代数 ..... 3

2.1 Matrices and Vectors ..... 3
2.1 矩阵和向量 ..... 3

2.1.1 Definitions ..... 3
2.1.1 定义 ..... 3

2.1.2 Notation ..... 4
2.1.2 符号 ..... 4

2.1.3 Addition ..... 4
2.1.3 加法 ..... 4

2.1.4 Subtraction ..... 5
2.1.4 减法 ..... 5

2.1.5 Multiplication by a scalar . ..... 5
2.1.5 标量乘法。..... 5

2.1.6 Multiplication of two matrices ..... 5
2.1.6 两个矩阵的乘法 ..... 5

2.1.7 Motivation for matrix-matrix multiplication ..... 7
2.1.7 矩阵矩阵乘法的动机 ..... 7

2.1.8 Matrix-vector multiplication ..... 8
2.1.8 矩阵-向量乘法 ..... 8

2.1.9 Special Matrices ..... 8
2.1.9 特殊矩阵 ..... 8

2.1.10 Scalar products and orthogonality ..... 10
2.1.10 标量积和正交性 ..... 10

2.2 Linear Systems ..... 11
2.2 线性系统 ..... 11

2.3 Determinants ..... 12
2.3 行列式 ..... 12

2.3.1 Using determinants to invert a matrix ..... 14
2.3.1 使用行列式来求逆 矩阵..... 14

2.4 Eigenvalues and Eigenvectors ..... 15
2.4 特征值和特征向量 ..... 15

3 Differentiation and Integration ..... 21
3 差异化和整合 ..... 21

3.1 Differentiation ..... 21
3.1 差异化 ..... 21

3.1.1 Notation ..... 22
3.1.1 符号 ..... 22

3.1.2 Standard Results ..... 23
3.1.2 标准结果 ..... 23

3.1.3 Product rule ..... 23
3.1.3 产品规则 ..... 23

3.1.4 Chain rule ..... 23
3.1.4 链式法则 ..... 23

3.1.5 Quotient rule ..... 24
3.1.5 商规则 ..... 24

3.1.6 Stationary points in 1D ..... 24
3.1.6 1D 中的静止点 ..... 24

3.1.7 Partial derivatives ..... 25
3.1.7 偏导数 ..... 25

3.1.8 Stationary points in 2 dimensions ..... 25
3.1.8 二维空间中的静止点 ..... 25

3.1.9 Taylor Series ..... 26
3.1.9 泰勒级数 ..... 26

3.2 Integration ..... 27
3.2 集成 ..... 27

3.2.1 Finding Integrals ..... 29
3.2.1 寻找积分 ..... 29

4 Complex Numbers ..... 32
4 复数 ..... 32

4.1 Motivation ..... 32
4.1 动机 ..... 32

4.1.1 Graphical concept ..... 32
4.1.1 图形概念 ..... 32

4.2 Definition ..... 33
4.2 定义 ..... 33

4.3 Complex Plane ..... 34
4.3 复平面 ..... 34

4.4 Addition/Subtraction ..... 34
4.4 加法/减法 ..... 34

4.5 Multiplication ..... 35
4.5 乘法 ..... 35

4.6 Conjugates ..... 35
4.6 结合物 ..... 35

4.7 Division ..... 37
4.7 分割 ..... 37

4.8 Polar Form ..... 38
4.8 极坐标形式 ..... 38

4.9 Exponential Notation ..... 39
4.9 指数表示法 ..... 39

4.10 Application to waves ..... 40
4.10 应用于波浪 ..... 40

4.10.1 Amplitude and phase ..... 41
4.10.1 幅度和相位 ..... 41

4.10.2 Complex solution to the wave equation ..... 43
4.10.2 波动方程的复杂解..... 43

5 Error analysis ..... 45
5 错误分析 ..... 45

5.1 Plus/Minus Notation ..... 46
5.1 正负符号表示法 ..... 46

5.2 Propagation of errors ..... 46
5.2 误差的传播.....46

5.3 Comparison with "worst case" scenario? ..... 47
5.3 与“最坏情况”进行比较?..... 47

5.4 Normal Distribution ..... 47
5.4 正态分布 ..... 47

5.5 Central limit theorem ..... 48
5.5 中心极限定理 ..... 48

5.6 Confidence Intervals ..... 49
5.6 置信区间 ..... 49

1 Basic Skills 1 基本技能

This document contains notes on basic mathematics. There are links to the corresponding Leeds University Library skills@Leeds page, in which there are subject notes, videos and examples.
本文档包含有关基本数学的笔记。其中包含指向对应的利兹大学图书馆技能@利兹页面的链接,页面中包含主题笔记、视频和示例。
If you require more in-depth explanations of these concepts, you can visit the Wolfram Mathworld website:
如果您需要对这些概念进行更深入的解释,您可以访问沃尔夫勒姆数学世界网站:
Wolfram link (http://mathworld.wolfram.com/)
Wolfram 链接 (http://mathworld.wolfram.com/)
  • Algebra (Expanding brackets, Factorising) :
    代数(展开括号,因式分解):
Library link  图书馆链接
(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/algebra/).
http://library.leeds.ac.uk/tutorials/maths-solutions/pages/algebra/。

- Fractions : - 分数:

(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/numeracy/fractions.html).
http://library.leeds.ac.uk/tutorials/maths-solutions/pages/numeracy/fractions.html.
  • Indices and Powers :
    指数和幂:
(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/numeracy/indices.html).
http://library.leeds.ac.uk/tutorials/maths-solutions/pages/numeracy/indices.html.

- Vectors : - 向量:

(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/mechanics/vectors.html).
http://library.leeds.ac.uk/tutorials/maths-solutions/pages/mechanics/vectors.html.
  • Trigonometry and geometry :
    三角学和几何学:
(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/trig_geom/).
http://library.leeds.ac.uk/tutorials/maths-solutions/pages/trig_geom/。
  • Differentiation and Integration :
    微积分:差异化和整合
Library link  图书馆链接
(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/calculus/).
http://library.leeds.ac.uk/tutorials/maths-solutions/pages/calculus/。

1.1 Practice Questions 1.1 练习题

There are practice equations available online to accompany these notes.
这些笔记配有在线练习方程。

2 Linear Algebra 2 线性代数

Wolfram link (http://mathworld.wolfram.com/LinearAlgebra.html)
Wolfram 链接(http://mathworld.wolfram.com/LinearAlgebra.html)

2.1 Matrices and Vectors
2.1 矩阵和向量

Library link (http://library.leeds.ac.uk/tutorials/maths-solutions/pages/mechanics/vectors.html)
图书馆链接 (http://library.leeds.ac.uk/tutorials/maths-solutions/pages/mechanics/vectors.html)

2.1.1 Definitions 2.1.1 定义

A matrix is a rectangular array of numbers enclosed in brackets. These numbers are called
矩阵是一组用括号括起来的数字的矩形数组。这些数字被称为
e.g. 例如。
Matrix has 2 rows and 3 columns.
矩阵 有 2 行 3 列。
A row vector is a matrix with a single row:
一行向量是具有单行的矩阵:
e.g. 例如。
Whereas a column vector is a matrix with a single column:
一个列向量是一个只有一列的矩阵:
e.g. 例如。
The size of a matrix is defined by where is the number of rows and is the number of columns. Matrix , as defined in equation 1 , is a matrix.
矩阵的大小由 定义,其中 是行数, 是列数。矩阵 ,如方程 1 中定义的,是一个 矩阵。
An element of a matrix can be described by its row position and column position. For ex-
矩阵的一个元素可以通过其行位置和列位置来描述。例如-

ample: the top left element in matrix , equal to 1 , is in row 1 and column 1 and can be labelled as element ; the element in the column of row 1 , equal to 3 , is labelled as . A general element is located in row and column (see equation 4 for a further example).
在矩阵 中,左上角的元素等于 1,位于第 1 行第 1 列,可以标记为元素 ;第 1 行的 列元素等于 3,标记为 。一般元素 位于第 行和第 列(参见方程 4 以获取更多示例)。

2.1.2 Notation 2.1.2 符号

There are different types of notation for matrices and vectors that you may encounter in text books. Below are some examples:
矩阵和向量的表示法有不同类型,你可能在教科书中遇到。以下是一些示例:
Matrix 矩阵
italics 斜体
bold, italics 粗体,斜体
double underline, italics
双下划线,斜体
Vector 矢量
italics 斜体
top arrow, italics 顶部箭头,斜体
single underline, italics
单下划线,斜体
bold 粗体

2.1.3 Addition 2.1.3 加法

Wolfram link (http://mathworld.wolfram.com/MatrixAddition.html)
Wolfram 链接(http://mathworld.wolfram.com/MatrixAddition.html)
Video link (http://www.youtube.com/watch?v=FX4C-JpTFgY)
视频链接 (http://www.youtube.com/watch?v=FX4C-JpTFgY)
Two matrices (or vectors) of the same size may be added together, element by element. For instance, if we have two matrices and :
两个相同大小的矩阵(或向量) 可以逐个元素相加。例如,如果我们有两个矩阵
then, 那么,

2.1.4 Subtraction 2.1.4 减法

Similar to addition, corresponding elements in and are subtracted from each other:
与加法类似, 中对应的元素相互相减:

2.1.5 Multiplication by a scalar
2.1.5 标量乘法

If is a number (i.e. a scalar) and is a matrix, then is also a matrix with entries
如果 是一个数字(即标量),而 是一个矩阵,则 也是一个具有条目的矩阵

2.1.6 Multiplication of two matrices
两个矩阵的乘法

Wolfram link (http://mathworld.wolfram.com/MatrixMultiplication.html)
Wolfram 链接(http://mathworld.wolfram.com/MatrixMultiplication.html)
This is non-trivial and is governed by a special rule. Two matrices , where is of size , and of size , can only be multiplied if , i.e. the number of columns in must match the number of rows in . The matrix produced has size , with each entry being the dot (or scalar) product (see section 2.1.10) of a whole row in by a whole column in .
这是一个非平凡的问题,受特殊规则约束。两个矩阵 ,其中 的大小为 的大小为 ,只有在 的情况下才能相乘,即 中的列数必须与 中的行数相匹配。生成的矩阵大小为 ,每个条目都是 中整行与 中整列的点(或标量)积(参见第 2.1.10 节)。

e.g. if 例如
then 然后
Formally, if 正式地,如果

Aside 一旁

When using Matlab (or octave), two matrices can be multiplied in an element-wise sense. This is NOT the same as described above.
在使用 Matlab(或 Octave)时,可以以逐元素方式相乘两个矩阵。这与上面描述的不同。

2.1.7 Motivation for matrix-matrix multiplication
2.1.7 矩阵矩阵乘法的动机

To understand why we may need to perform matrix-matrix multiplication, consider two customers of a repair garage, Peter and Alex, who require a number of car parts for each of their vehicles. Peter requires litre engine and 2 doors, whereas Alex requires litre engine and 4 doors. All the parts require a certain number of screws and bolts. But how many total screws and bolts do Peter and Alex need?
为了理解为什么我们可能需要执行矩阵-矩阵乘法,请考虑修车厂的两位顾客 Peter 和 Alex,他们各自的车辆需要一些汽车零件。Peter 需要 升引擎和 2 扇门,而 Alex 需要 升引擎和 4 扇门。所有零件都需要一定数量的螺丝和螺栓。但 Peter 和 Alex 总共需要多少螺丝和螺栓?
We can present the quantity of each car part that Peter and Alex need in a table:
我们可以用表格来展示彼得和亚历克斯需要的每个汽车零件的数量:
3 litre engine 3 升引擎 5 litre engine 5 升引擎 Doors 
Peter 彼得 1 0 2
Alex 亚历克斯 0 1 4
or as the matrix, :
或者作为矩阵,
The number of screws and bolts for each car part are expressed in the table:
每个汽车零件的螺丝和螺栓数量在表中表示:
bolts 螺栓 screws 螺丝钉
3 litre 3 升 3 4
5 litre 5 升 1 8
doors  2 1
or can be expressed as matrix, :
或者可以表示为矩阵,
Using simple addition we can find out how many screws and bolts are needed.
使用简单的加法,我们可以找出需要多少螺丝和螺栓。
  1. How many bolts are needed for Peter's car parts?
    彼得的汽车零件需要多少个螺栓?
.
  1. How many bolts are needed for Alex's car parts?
    Alex 的汽车零件需要多少个螺栓?
  1. How many screws are needed for Peter's car parts?
    彼得的汽车零件需要多少颗螺丝?
Or we can use matrix multiplication to get all four scenarios:
或者我们可以使用矩阵乘法得到所有四种情况:

2.1.8 Matrix-vector multiplication
2.1.8 矩阵-向量乘法

Since a vector is a special case of a matrix, this is simply a special case of the matrix-matrix multiplication we have already discussed. Consider multiplying a column vector of length by a matrix of size ,
由于向量是矩阵的特殊情况,这只是我们已经讨论过的矩阵乘法的特殊情况。考虑将长度为 的列向量乘以大小为 的矩阵,
e.g. 例如。
which results in a column vector of length and in this case .
这导致一个长度为 的列向量,在这种情况下

2.1.9 Special Matrices 2.1.9 特殊矩阵

Identity Matrix, The identity matrix, , of size , is defined in equation 12 .
单位矩阵, 大小为 的单位矩阵, ,在方程式 12 中定义。
i.e. if  即如果
This is a special case of a diagonal matrix possessing non-zero entries only on its diagonal e.g.
这是对角矩阵的一个特殊情况,只在对角线上具有非零条目,例如。
If is a square matrix, then the identity matrix has the special property that:
如果 是一个方 矩阵,则单位矩阵 具有特殊性质,即:
NB: is the equivalent of 1 in scalar arithmetic i.e. .
NB: 是标量算术中的 1 的等价物,即
Transpose, : If is a matrix then the transpose of , denoted , is a matrix found by swapping rows and columns of ,
转置, :如果 是一个 矩阵,那么 的转置,记为 ,是通过交换 的行和列找到的 矩阵
e.g. 例如。
Inverse matrix, If is an matrix, sometimes (see later) there exists another matrix called the inverse of , written , such that
逆矩阵,如果 是一个 矩阵,有时(见后文)存在另一个矩阵称为 的逆矩阵,记作 ,使得
NB: For scalar numbers, is the inverse of when considering multiplication, since
NB:对于标量数字,在考虑乘法时, 的倒数
Clearly when this breaks down and has no inverse - this is also true when dealing with some matrices.
明显地崩溃并且 没有逆时 - 处理某些矩阵时也是如此。

2.1.10 Scalar products and orthogonality
2.1.10 标量积和正交

The scalar product (or dot product, or inner product) of two column vectors of length , where and , is
两个长度为 的列向量(或点积,或内积)的数量积,其中 ,是
This can also be written as ; that is, the product of a row vector of length with a column vector of length . Two vectors are said to be orthogonal if their scalar product is zero.
这也可以写成 ;也就是说,长度为 的行向量与长度为 的列向量的乘积。如果两个向量的数量积为零,则称它们是正交的。

2.2 Linear Systems 2.2 线性系统

Wolfram link (http://mathworld.wolfram.com/LinearSystemofEquations.html)
Wolfram 链接(http://mathworld.wolfram.com/LinearSystemofEquations.html)
Video link (http://www.youtube.com/watch? )
视频链接(http://www.youtube.com/watch?
A linear system of equations such as
线性方程组如
can be written as
可以写成
as can be verified by multiplying out the left hand side. When solving the linear system , (where is a matrix, is the vector and is a vector of numbers) two cases can arise:
左侧展开后可以验证。解线性系统 时(其中 是矩阵, 是向量 是数字向量),可能出现两种情况:

i) exists. i) 存在。
ii) doesn't exist. There is then no solution in general.
ii) 不存在。通常情况下没有解决方案。

2.3 Determinants 2.3 行列式

Wolfram link (http://mathworld.wolfram.com/Determinant.html)
Wolfram 链接(http://mathworld.wolfram.com/Determinant.html)
Video link (http://www.youtube.com/watch?v=23LLB9mNJvc)
视频链接(http://www.youtube.com/watch?v=23LLB9mNJvc)
How do we know when exists? One method is to calculate the determinant of , written det or . The determinant is a single number that contains enough information about to determine whether it is invertible.
我们如何知道 存在?一种方法是计算 的行列式,写作 det 。行列式是一个包含足够关于 的信息的单个数字,可以确定它是否可逆。
determinants: In the case, if
决定因素:在 情况下,如果
then . For example, for the linear system given by
然后 。例如,对于给定的线性系统
the determinant of the coefficient matrix is
系数矩阵的行列式是
The matrix is therefore invertible (see section 2.3.1) and so the solution exists.
矩阵 是可逆的(见第 2.3.1 节),因此解存在。
As another example, consider
作为另一个例子,请考虑
The determinant of the coefficient matrix is
系数矩阵的行列式是
Therefore, has no inverse and so no solution exists. This can also be seen in the fact that it is not possible that simultaneously equal both 2 and 3 .
因此, 没有逆元,因此不存在解决方案。这也可以从这样一个事实中看出,即不可能同时等于 2 和 3。
determinants: Determinants can be generalised to matrices. For the matrix A,
确定因素: 确定因素可以推广到 矩阵。对于矩阵 A,
the determinant of is:
的行列式是:
or equivalently, 或者等价地,
e.g. 例如。
or
Do whichever is easier!
做任何更容易的事情!

2.3.1 Using determinants to invert a matrix
2.3.1 使用行列式来求逆 矩阵

The determinant can be used in finding the inverse of a matrix.
行列式可以用于找到 矩阵的逆。
For example: find the inverse of matrix .
例如: 找到矩阵 的逆。

2.4 Eigenvalues and Eigenvectors
2.4 特征值和特征向量

Wolfram link 1 (http://mathworld.wolfram.com/Eigenvalue.html)
Wolfram 链接 1 (http://mathworld.wolfram.com/Eigenvalue.html)
(http://mathworld.wolfram.com/Eigenvector.html)
(http://mathworld.wolfram.com/Eigenvector.html)
Video link (http://www.youtube.com/watch? lXNXrLcoerU)
视频链接(http://www.youtube.com/watch? lXNXrLcoerU)
Often we are interested in whether a matrix can stretch a vector. In such a case:
通常我们对一个矩阵是否能拉伸一个向量感兴趣。在这种情况下:
where is the "stretch factor". The scalar is called an eigenvalue (from the German: eigen meaning same) and is an eigenvector. is equivalent to:
其中 是“拉伸因子”。标量 称为特征值(来自德语:eigen 意为相同), 是特征向量。 等同于:
If then the system can be solved to find . If we want non-zero vectors , then we require .
如果 ,那么系统可以被解决以找到 。如果我们想要非零向量 ,那么我们需要
To find the eigenvectors and eigenvalues, we use a two stage process:
为了找到特征向量和特征值,我们使用两阶段过程:
i) Solve , i) 解决
ii) Find . ii) 找到
For example: 例如:
i) The eigenvalues are such that
i) 特征值 是这样的
ii) Now to find the eigenvectors, :
ii) 现在要找到特征向量,
If then 如果 那么
Hence the eigenvector is: , or , or just , (see below).
因此特征向量是: ,或 ,或只是 ,(见下文)。
If , then 如果 ,那么
giving the eigenvector of or just .
给出特征向量 或者只是
The eigenvectors provide a direction and so we can ignore common factors. This is because if is an eigenvector, then so is for any value of .
特征向量提供了一个方向,因此我们可以忽略公共因素。这是因为如果 是一个特征向量,那么对于任何值 也是一个特征向量。
i.e. . 
For example: if above, then are all eigenvectors. We typically choose the simplest!
例如:如果 以上,则 都是特征向量。通常我们选择最简单的!
Matrix diagonalisation Suppose is a matrix. Form a new matrix , each column of which is an eigenvector of . If exists then,
矩阵对角化 假设 是一个 矩阵。形成一个新矩阵 ,其中每一列都是 的特征向量。如果 存在,则,
the eigenvalues of on the diagonal.
对角线上的特征值
E.g. 例如。
- Eigenvector 1 is: , eigenvector 2 is:
- 特征向量 1 是: ,特征向量 2 是:
It turns out that exists and is:
原来 存在并且是:
So then the matrix product
那么矩阵乘积
is equal to the matrix .
等于矩阵
Special case: Symmetric matrix A matrix is symmetric if . It turns out that, in this case, all the eigenvectors are orthogonal, i.e. if and are different eigenvectors, then
特殊情况:对称矩阵 A 矩阵 如果 。事实证明,在这种情况下,所有的特征向量都是正交的,即如果 是不同的特征向量,则
If each eigenvector is normalised such that , then,
如果每个特征向量都被归一化,使得 ,那么,
(This is an easy way to find ).
(这是一个找到 的简单方法)。
Eigenvectors are given by:
特征向量 由以下给出:
Hence, eigenvector is: .
因此,特征向量是:
But this can be normalised (since it's magnitude is arbitrary) by any number : .
但这可以通过任何数字 来归一化(因为它的大小是任意的)。
Let's choose such that: where .
让我们选择 ,使得: 其中
Hence, or  因此,
Our normalised eigenvector is then .
我们的归一化特征向量是
Hence, eigenvector is .
因此,特征向量是
Normalise such that , so
使得 正规化,因此
The normalised eigenvector is then .
归一化的特征向量是
Define: 定义:
Then we can check that
然后我们可以检查一下
Why is this useful?
为什么这个有用?
Example: What is ? Using the matrix diagonalisation,
示例:什么是 ?使用矩阵对角化,
and is an easy matrix to raise to a power:
是一个容易提高到幂次的矩阵:
This method is much easier than multiplying the matrix by itself 8 times.
这种方法比将矩阵乘以自身 8 次要容易得多。

3 Differentiation and Integration
3 差异化和整合

Library link (http://library.leeds.ac.uk/tutorials/maths-solutions/pages/calculus/)
图书馆链接(http://library.leeds.ac.uk/tutorials/maths-solutions/pages/calculus/)

3.1 Differentiation 3.1 差异化

Wolfram link (http://mathworld.wolfram.com/Derivative.html)
Wolfram 链接(http://mathworld.wolfram.com/Derivative.html)
Suppose the function gives distance as a function of time as shown in Figure 1.
假设函数 给出了时间的函数距离,如图 1 所示。
Figure 1 图 1
At point A, the sketched tangent gives the instantaneous rate of change of distance with time or speed.
在点 A 处,所画的切线给出了距离随时间或速度的瞬时变化率。
To calculate the speed (or gradient function) at any time, we approximate the tangent by connecting two neighbouring points: and (see Figure 2).
为了计算任何时刻的速度(或梯度函数),我们通过连接两个相邻点 来近似切线(见图 2)。
An estimate of the gradient at time is then
时间 的梯度估计是
As , this estimate becomes more accurate.
随着 ,这个估计变得更加准确。
Figure 2 图 2
For example Suppose , then
例如假设 ,那么
Hence the gradient of at any point is .
因此,在任何点 处的 的梯度是

3.1.1 Notation 3.1.1 符号

The gradient or derivative of a function can be written:
函数 的梯度或导数可以写成:
We can also have higher derivatives. Consider the gradient of a gradient function. If represents distance, then is the speed and is the acceleration.
我们也可以有更高阶导数。考虑一个梯度函数的梯度。如果 代表距离,那么 是速度, 是加速度。

3.1.2 Standard Results 3.1.2 标准结果

1 0
1
Differentiation is linear: e.g.
差异化是线性的:例如。

3.1.3 Product rule 3.1.3 产品规则

If we need to take the time derivative of the product of two functions and then we use the product rule.
如果我们需要对两个函数 的乘积 进行时间导数,则我们使用乘积法则。

3.1.4 Chain rule 3.1.4 链式法则

The chain rule can be used to differentiate more complicated functions.
链式法则可用于对更复杂的函数进行求导。
The chain rule is defined as:
链式法则被定义为:
For example: 例如:
  1. How do we differentiate the function ?
    我们如何区分函数
We know how to differentiate , so let's define . Then we simply need to assemble the ingredients for the chain rule: and . It then follows that
我们知道如何区分 ,所以让我们定义 。然后我们只需要组装链式法则的成分: 。接着就是
  1. How do we differentiate the function ?
    我们如何区分函数
We know how to differentiate , so let's go ahead and use the chain rule, the key ingredient we need being . Then
我们知道如何区分 ,所以让我们继续使用链式法则,我们需要的关键要素是 。然后

3.1.5 Quotient rule 3.1.5 商规

If we need to take the derivative of a quotient of functions: , then we use the quotient rule.
如果我们需要对函数的商进行求导: ,那么我们使用商规则。

3.1.6 Stationary points in 1D
3.1.6 1D 中的静止点

Wolfram link (http://mathworld.wolfram.com/StationaryPoint.html)
Wolfram 链接(http://mathworld.wolfram.com/StationaryPoint.html)
For a function , a stationary point is where the gradient, vanishes. To decide whether
对于一个函数 ,一个稳定点是梯度 消失的地方。要决定

it represents a local maximum or minimum, we need the 2 nd derivative. If at the stationary point, then it is a minimum; conversely if then it's a maximum.
它代表了一个局部极大值或极小值,我们需要第二阶导数。如果在稳定点处为 0,则为极小值;相反,如果为 1,则为极大值。
For example, find and classify the stationary points of on . The gradient function is , which on the range in question is zero at and . Evaluating the 2nd derivative, at these points gives and . Hence is a local maximum, and is a local minimum.
例如,在 上找到并分类 的静止点。梯度函数是 ,在问题范围内在 处为零。在这些点上评估二阶导数 得到 。因此 是局部最大值, 是局部最小值。

3.1.7 Partial derivatives
3.1.7 偏导数

Wolfram link (http://mathworld.wolfram.com/PartialDerivative.html)
Wolfram 链接(http://mathworld.wolfram.com/PartialDerivative.html)
Suppose is a function of more than one independent variable, e.g. . We can make sense of the gradient by varying each variable, one at a time:
假设 是一个关于多个自变量的函数,例如 。我们可以通过逐个改变每个变量来理解梯度:
hold constant 保持 不变
hold constant 保持 不变
The notation means a "partial" derivative with respect to and regards all other variables as constant.
符号 表示对 的“偏导数”,并将所有其他变量视为常数。
e.g. . 例如
We can take and higher order derivatives:
我们可以取 及更高阶导数:
e.g. 例如。

3.1.8 Stationary points in 2 dimensions
3.1.8 二维空间中的静止点

One mathematical description of a surface in is a function , giving the height as a function of coordinates and . Just as in 1D, stationary points are where the gradient in all directions is
一个表面在 的数学描述是一个函数 ,给出高度作为坐标 的函数。就像在 1D 中一样,静止点是所有方向的梯度的地方。

zero (corresponding to local maxima, minima or saddle points) and are given by the conditions
零(对应于局部极大值、极小值或鞍点)由条件给出

3.1.9 Taylor Series 3.1.9 泰勒级数

We can approximate the behaviour of the function at a point using knowledge of its derivatives. For example, suppose a car is at a distance of and is travelling at . Where will it be in 30 minutes?
我们可以利用对函数导数的知识来近似描述某一点的行为。例如,假设一辆汽车距离 ,并以 的速度行驶。30 分钟后它会在哪里?
If its speed is constant, . But if its speed changes, then we need a correction term.
如果它的速度是恒定的, 。但如果它的速度改变了,那么我们就需要一个修正项。
A function is related to its behaviour at by,
一个函数 与其在 处的行为相关,由此可知,
For example: If and ,
例如:如果
We can generalise this idea to functions of several variables. If is expanded near :
我们可以将这个想法推广到多个变量的函数。如果 附近展开:
e.g. If , how does behave close to ?
例如,如果 附近会表现如何?
Hence, to order.
因此, 订单。

3.2 Integration 3.2 集成

The integral of a function between and is the area under the curve of : . To find it, we often use the "fundamental theorem" of calculus:
函数 之间的积分是 曲线下的面积: 。为了找到它,我们经常使用微积分的“基本定理”:
Figure 3 图 3
where  在哪里
That is, is the anti-derivative of .
的反导数。
An indefinite integral is an integral without limits and gives a function that is the anti-derivative of (including an arbitrary constant):
一个不定积分是一个没有限制的积分,得到的函数是 的反导数(包括一个任意常数):
Standard Results 标准结果
1

3.2.1 Finding Integrals 3.2.1 寻找积分

  1. By parts 逐部
We have already seen that: .
我们已经看到:
If we integrate this:
如果我们整合这个:
e.g. Find: 例如:查找:
Let: 请让:
Therefore: 因此:

2. By substitution 2. 通过替代

For example, suppose we want to find
例如,假设我们想要找到
Write 
so that or rearranging to give . Substituting all ' ' variables for 'u' variables gives
这样 或重新排列以给出 。用'u'变量替换所有' '变量。
As another example, suppose we want to find:
作为另一个例子,假设我们想要找到:
Choose , so that . The integral then becomes,
选择 ,以便 。然后积分变为,

4 Complex Numbers 4 复数

4.1 Motivation 4.1 动机

It was not so long ago that equations like or were considered absurd - how can you have -5 or cows? Nevertheless, these mathematical representations of negative and fractional numbers are extremely useful.
不久前,像 这样的方程被认为是荒谬的 - 你怎么可能有-5 头牛或 头牛呢?然而,这些负数和分数的数学表示法非常有用。
Along similar lines, it wasn't very long ago that the equation: , was considered ridiculous, for how can a squared number be negative? But if we proceed anyway and define a solution, it turns out to be very useful indeed.
沿着类似的思路,不久之前,方程式: ,被认为是荒谬的,因为一个平方数怎么可能是负数呢?但如果我们继续前行并定义一个解,结果会发现它确实非常有用。

4.1.1 Graphical concept 4.1.1 图形概念

How might we visualise the solution to: ? That is, what transformation applied twice to 1 , gives 9 ? The answer of course is 3 or -3 .
我们如何将解决方案可视化: ?也就是说,对 1 应用两次的转换,得到 9?答案当然是 3 或-3。
But let's consider, . What transformation, applied twice to 1 , gives -1 ?
但让我们考虑, 。应用两次于 1 的转换,得到-1?
We can express this as a rotation of , (see Figure 4).
我们可以将这个表示为一个旋转 ,(见图 4)。
A rotation of also works as in Figure 5,
的旋转也可以像图 5 中那样工作
To make sense of this, we need to define a new "imaginary dimension"; or is what becomes of 1 or -1 after a rotation of or .
为了理解这一点,我们需要定义一个新的“虚拟维度”; 是在旋转 后成为 1 或 -1 的结果。
What happens if we keep multiplying by ?
如果我们不断乘以 会发生什么?
Figure 4 图 4
Figure 5 图 5
as depicted below in Figure 6 .
如下图所示

4.2 Definition 4.2 定义

A complex number is written or where is the real part of and is the imaginary part. The number satisfies . For any complex number , its real part is or , and the complex (or imaginary) part , is or . If ,
一个复数 写成 ,其中 的实部, 是虚部。数 满足 。对于任何复数 ,它的实部 ,虚部 。如果
Figure 6 图 6
then is purely real; if , then is purely imaginary.
那么 是纯实数;如果 ,那么 是纯虚数。

4.3 Complex Plane 4.3 复平面

We can plot a complex number in an domain called the complex plane or the Argand Diagram. For example, is displayed in the Argand diagram in Figure 7.
我们可以在一个称为复平面或阿贡图的 域中绘制一个复数。例如, 在图 7 中显示在阿贡图中。

4.4 Addition/Subtraction
4.4 加法/减法

Two complex numbers and can be added or subtracted by adding or subtracting their real and imaginary parts separately.
两个复数 可以通过分别添加或减去它们的实部和虚部来相加或相减。
For example: 例如:
Figure 7: An argand diagram showing the complex plane. The complex number has real part and complex part .
图 7:阿尔干图显示复平面。复数 的实部为 ,虚部为

4.5 Multiplication 4.5 乘法

For two complex numbers and , find their product by multiplying out in full.
对于两个复数 ,通过完全展开来找到它们的乘积。
For example: 例如:
using . 使用

4.6 Conjugates 4.6 结合物

If is a complex number , then the conjugate to is: .
如果 是一个复数 ,那么 的共轭是:
This is useful because:
这是有用的,因为:
is a purely real number.
是一个纯粹的实数。
Geometrically, , where is the "magnitude" of , see Figure 8 below.
几何上, ,其中 的“大小”,请参见下面的图 8。
Figure 8: An argand diagram showing the magnitude of the complex number .
图 8:阿尔干图显示复数 的大小

4.7 Division 4.7 分割

It is not immediately obvious how to divide two complex numbers and . However, we do know how to divide a complex number by a real number. For example:
如何将两个复数 相除并不是显而易见的。但是,我们知道如何将一个复数除以一个实数。例如:
To divide by we need to use :
要将 除以 ,我们需要使用
For example: 例如:

4.8 Polar Form 4.8 极坐标形式

A useful representation of a complex number is in polar coordinates.
一个复数的有用表示是极坐标。
For example: the complex number: as shown in Figure 9 can also be represented by , where is the magnitude of (the distance from 0 ) and is the angle with the horizontal.
例如:如图 9 所示的复数: 也可以用 表示,其中 的大小(距离 0 的距离), 是与水平线的角度。
Figure 9: An argand diagram showing the magnitude (or modulus) and angle (or argument) of a complex number.
图 9:阿尔干图显示了复数的幅度(或模量) 和角度(或参数)
In this notation, 在这种表示法中,
and  
so, . 所以,
The magnitude of is also called the modulus, is called the argument and denoted:
的大小也被称为模量, 被称为幅角,表示为:
In order that is unique, is often used.
为了使 是独一无二的,通常会使用

4.9 Exponential Notation
4.9 指数表示法

It turns out that:
原来如此:
where  在哪里
This makes it easy to multiply and divide in polar form:
这使得在极坐标形式中进行乘法和除法变得容易:
and compute powers, e.g.
计算幂,例如。

4.10 Application to waves
4.10 应用于波浪

Some links about waves as a refresher:
一些关于波浪的链接作为复习:
Movie link Sine and cosine waves
电影链接 正弦和余弦波
(http://videos.kightleys.com/Science/Maths/23131008_CsD3fs/1880848370_
http://videos.kightleys.com/Science/Maths/23131008_CsD3fs/1880848370_
VMGSWd3#!
Movie link Superposition of waves
电影链接 波的叠加
(http://www.acs.psu.edu/drussell/demos/superposition/superposition.html)
http://www.acs.psu.edu/drussell/demos/superposition/superposition.html
Waves can be described (in real form) as
波浪可以被描述为(以实际形式)
where is the amplitude, is the angular frequency and is the phase.
其中 是振幅, 是角频率, 是相位。
However, it has been established that there is a close relationship between complex numbers and sine/cosine functions - it can be useful to use complex numbers to express wave characteristics. Since waves are a physical phenomenon, we would like any representation to be real. We can therefore speak of "complex waves", subject to the expectation that we are only really interested in the real part (we often just throw away the imaginary part). Therefore, the complex wave
然而,已经确定复数与正弦/余弦函数之间存在密切关系 - 使用复数来表达波特性是有用的。由于波是一种物理现象,我们希望任何表示都是真实的。因此,我们可以谈论“复波”,但我们只对实部感兴趣(通常会丢弃虚部)。因此,复波
has real part and so the two representations: (38) and (39) are essentially equivalent.
具有实部 ,因此两种表示:(38)和(39)本质上是等价的。
If a wave is described by where is a complex number, then in fact contains information about both amplitude and phase. This is because we can write , so that
如果一个波被描述为 ,其中 是一个复数,那么实际上 包含有关振幅和相位的信息。这是因为我们可以写成 ,这样
The amplitude of the wave is and the phase is .
波的振幅为 ,相位为
Examples of two types of waves are shown in Figure 10. Both waves have amplitude of 2 but the phase is shifted. Visually you can see that one wave ( wave shown as dashed lines) can be shifted by to be equal to the wave.
图 10 显示了两种类型的波的示例。这两种波的振幅都为 2,但相位发生了移位。从视觉上可以看出,一种波( 波以虚线显示)可以通过移动 来等于 波。
Figure 10 图 10

Aside: 旁白:

Both and have been used as angles in this chapter on complex numbers and the general convention is to use:
本章复数和一般约定使用的角度为
  • for angles (in the polar plane);
    用于角度(在极坐标平面中);
  • for phase of waves.
    波的相位

4.10.1 Amplitude and phase
4.10.1 幅度和相位

Links: 链接:
Wiki link Phase (http://en.wikipedia.org/wiki/Phase_(waves))
Wiki 链接相位(http://en.wikipedia.org/wiki/Phase_(waves))
Wolfram link Amplitude (http://mathworld.wolfram.com/Amplitude.html)
Wolfram 链接振幅(http://mathworld.wolfram.com/Amplitude.html)
To find the amplitude and phase of a wave function we convert the function into the exponential complex form.
要找到波函数的幅度和相位,我们将函数转换为指数复数形式。
For example: 例如:
  • is the real part of and has amplitude 1 and phase 0 .
    的实部,振幅为 1,相位为 0。
  • is the real part of and has amplitude 2 and phase .
    的实部,振幅为 2,相位为
  • is the real part of and has amplitude 1 and phase .
    的实部,振幅为 1,相位为
is the real part of and has amplitude 5 and phase .
的实部,振幅为 5,相位为
Example: What is the amplitude and phase of:
示例:以下是振幅和相位是多少:
  1. Convert each part into a complex form:
    将每个部分转换为复数形式:
  • is the real part of .
    的实部。
  • is the real part of .
    的实部。
Thus 因此
so can be represented by the complex wave
因此可以用复波表示
  1. Now compare to the complex wave solution of the form (where here). Dividing through by gives
    现在与形式 的复杂波解进行比较(其中 在这里)。 通过 除以得到
Taking the modulus: .
取模:
Real part : Imaginary part :
实部: 虚部:
So, . 那么,
Figure 11 图 11
In Figure 11, .
在图 11 中,
NB - you need to take care with finding the phase , since the inverse sine and cosine functions may give the correct answer in the wrong range. i.e. your calculator will tell you that
NB - 你需要小心找到相位 ,因为反正弦和反余弦函数可能会在错误的范围内给出正确答案。即你的计算器会告诉你
and
These are correct, but you must interpret them correctly! To be sure, draw a picture.
这些是正确的,但你必须正确解释它们!要确保,画一幅图片。

4.10.2 Complex solution to the wave equation
4.10.2 波动方程的复杂解

A problem in time-series analysis might be to find the solution to the equation:
时间序列分析中的一个问题可能是找到方程的解决方案:
The answer turns out to be:
答案原来是:
Equation 42 is an expression for a wave. But how might we discover this solution? One method is to attempt a trail solution of the form:
方程 42 是一个波的表达式。但我们如何发现这个解决方案呢?一种方法是尝试一个形式的试验解:
and then try to find and .
然后尝试找到
Another is to find a "complex wave" solution of the form,
另一种方法是找到一个形式为“复波”解
where is a complex number and as stated at the beginning of this section, we are only interested in the real component, , as the physical solution. In this case, , so that:
其中 是一个复数,并且正如本节开头所述,我们只关注实部 作为物理解。在这种情况下, ,以便:
The advantage of this method is that we need only find a single unknown number , rather than two and ).
这种方法的优点是我们只需要找到一个未知数 ,而不是两个 )。

5 Error analysis 5 错误分析

Suppose we make an observation of a quantity , called , which we repeat times to get the set: .
假设我们观察一个名为 的数量 ,我们重复 次以获得集合:
The mean, , is:
均值, ,是:
The variance, , is:
方差, ,是:
.
The standard deviation, , is:
标准差, ,是:
.
The mode is: 模式是:
the most common value.
最常见的值。
The median is: 中位数是:
the middle value if the values of are written in numerical order.
如果 的值按数字顺序编写,则中间值。
Suppose is an observation of the true value .
假设 是真值 的观测。
The absolute error is: .
绝对误差为:
The relative error is: .
相对误差为:
Accuracy is how close a measured value is to the true value (i.e. absolute error).
准确性是指测量值与真实值(即绝对误差)之间的接近程度。
Precision is how close the (repeated) measure values are to each other.
精度是指(重复)测量值彼此之间的接近程度。
We can have: 我们可以有:
  • high precision yet low accuracy,
    高精度但低准确性
  • high accuracy yet low precision,
    高准确性但低精度
  • high accuracy and high precision.
    高准确性和高精度。
Accuracy and precision may differ if there is bias in the measurement. For example if a set of high precision digital scales read " " instead of " 0 " with nothing on the scales, then all measurements of weight are out. Therefore, the scales may be precise yet not accurate.
如果测量中存在偏差,准确性和精度可能会有所不同。例如,如果一组高精度数字秤读取“ ”而不是“0”,而秤上什么也没有,那么所有的重量测量都会 。因此,秤可能是精确的,但不准确。

5.1 Plus/Minus Notation 5.1 正负号表示

Errors are often quoted in the form: value standard deviation units.
错误通常以以下形式引用:值 标准偏差单位。
For example: , where 1 here is the standard deviation. With this terminology, it is possible that is equal to 98 or , although very unlikely.
例如: ,其中 1 是标准差。使用这个术语,可能 等于 98 或 ,尽管可能性很小。
Sometimes, the absolute range of values are given. The range of values of may be: . With this terminology, it is NOT possible for to be .
有时候,会给出数值的绝对范围。 的数值范围可能是: 。使用这个术语, 不可能是

5.2 Propagation of errors
5.2 误差的传播

Suppose we estimate the mass of a sphere using the formula: , where and have a measurement error. What is the error in ?
假设我们使用公式估计球体的质量: ,其中 存在测量误差。 的误差是多少?
We can use the formula:
我们可以使用这个公式:
where is the error in mass, .
质量误差为
e.g. if and , then
例如,如果 ,那么
Hence, i.e. relative error.
因此, 相对误差。
This formula extends to any number of variables. If , then,
此公式适用于任意数量的变量。如果 ,那么,
Note: or gives the standard deviation of or , calculated as a function of the standard deviation of its dependant variables.
注意: 给出了 的标准偏差,该标准偏差是其相关变量的标准偏差的函数。
What about adding quantities? For example, if , where both and have an error, what is the error in ?
如何添加数量?例如,如果 ,其中 都有错误,那么 中的错误是什么?
This is actually a special case of the general formula (equation 44):
这实际上是一般公式(方程 44)的一个特殊情况:

5.3 Comparison with "worst case" scenario?
5.3 与“最坏情况”进行比较?

An alternative to using equation 44 is to compute the worst case error. For example, if the density measurement was incorrect by a single standard deviation above its mean, and the radius too high by a single standard deviation, then the mass calculated would be , compared to the mean value of . This is above the mean, or , a lot higher than that calculated above. But this is not a fair calculation, as it is very unlikely that both the density and radius conspire together in this way.
一种替代使用方程 44 的方法是计算最坏情况误差。例如,如果密度测量比其平均值高一个标准偏差,半径也比一个标准偏差高,那么计算出的质量将为 ,而不是 的平均值。这比平均值高 ,或 ,比上面计算的要高得多。但这不是一个公平的计算,因为密度和半径很少会以这种方式共同作用。

5.4 Normal Distribution 5.4 正态分布

If a variable is normally distributed with mean and standard deviation, , it has a probabilitydensity function (or likelihood function) as shown in Figure 12.
如果一个变量 符合均值为 ,标准差为 的正态分布,它的概率密度函数(或似然函数)如图 12 所示。
The most probable value of is . The probability of other values are also known:
的最可能值是 。其他值的概率也是已知的:
occurs with a probability of .
的概率为
: occurs with a probability of .
的概率为
: occurs with a probability of .
的概率为
In Geophysics, many variables are assumed to be normally distributed. The formula that
在地球物理学中,许多变量被假定为正态分布。这个公式是
Figure 12 图 12
gives the likelihood of is:
给出 的可能性是:
which is normalised such that
规范化,使得

5.5 Central limit theorem
5.5 中心极限定理

Normal distributions often arise out of other non-normal distributions. If , etc ... are identical variables then we can define .
正态分布经常源于其他非正态分布。如果 ,等等...是相同的变量,那么我们可以定义
It turns out that is approximately normally distributed with mean and standard deviation , where each of the has mean and standard deviation , i.e.
原来 大致上呈正态分布,均值为 ,标准差为 ,其中每个 的均值为 ,标准差为 ,即。
For example: a die shows a random number from 1 to 6 with equal probability for each.
例如:一个骰子以相等的概率显示从 1 到 6 的随机数字。
The mean of the value shown by a single die is and the standard deviation is found by .
单个骰子显示的值的平均值为 ,标准偏差由 确定。
A single die is not normally distributed, however, if we throw 10 dice and add their scores, the totals will be approximately normally distributed. For 10 dice the mean is , and the standard deviation is .
一个骰子通常不是正态分布的,但是,如果我们投掷 10 个骰子并将它们的分数相加,总分将近似正态分布。对于 10 个骰子,平均值为 ,标准差为
When considering the average of the ten dice, i.e. , then has mean , the same mean as each of the . The standard deviation of is , so the standard deviation of is . This occurs because the standard deviation is linear in any factor applied to the variables (here, the factor ).
考虑十个骰子的平均值,即 ,那么 的平均值为 ,与每个 的平均值相同。 的标准偏差为 ,因此 的标准偏差为 。这是因为标准偏差在应用于变量的任何因子(这里是因子 )时是线性的。

5.6 Confidence Intervals
5.6 置信区间

Suppose a variable is distributed normally with mean and standard deviation . Then the confidence interval for is as there is a probability of taking a value in this range. Similarly the confidence interval is (approximately) .
假设一个变量 符合正态分布,均值为 ,标准差为 。那么 置信区间为 ,因为有 的概率取值在这个范围内。类似地, 置信区间为(大约)
For example: Suppose are measurements all with mean and standard deviation . What is the confidence interval for ?
例如:假设 是所有具有平均值 和标准偏差 的测量值。 置信区间是多少?

Solution: 解决方案:

is distributed normally with mean and standard deviation . The confidence interval is then of the form .
符合均值为 和标准差为 的正态分布。然后 置信区间为 ,形式为
Additional material: 附加材料:
Video link (http://ocw.mit.edu/courses/mathematics/18-05-introduction-to-probabilityand-statistics-spring-2005/)
视频链接(http://ocw.mit.edu/courses/mathematics/18-05-introduction-to-probabilityand-statistics-spring-2005/)