Numerical Representations, Input and Output, and Graphics

Engineering design is usually heavily supported by numerical calculations. One of the first and enduring uses of computers is to automatically perform these calculations for engineers; in fact, the first “computers” were humans who performed numerical calculations by hand, as shown in fig. ¿fig:nasa-human-computers?.

We call engineering numerical calculations numerical analysis. Many programming languages and software packages have been used for numerical analysis, but by far the most popular these days are MATLAB and Python. Python’s built-in data types (e.g., list) and functions (e.g., sum) can be used directly for numerical analysis; however, for most engineering problems it is advantageous to use the ubiquitous package NumPy (Harris et al. 2020). The primary reasons this is preferred are that NumPy provides data types, functions, and methods optimized for numerical calculations, which go far beyond Python’s built-in modules. In the first several sections of this chapter, we will explore NumPy’s data types (most notably the array) and some of its basic functions and methods.

The numerical data represented in NumPy often originates as data from outside the program (e.g., from sensor data gathered via an experiment). Stored in files of various formats, the data must be read from computer memory¹ into the program. This is the most common kind of a program’s inputs. On the other end, a program can have outputs, frequently data files written to computer memory. In this chapter, we will learn how to load input data from files and write output data to files.

Another important kind of program output is a graphic—usually a graph, a plot, or a chart. A graphic is often a very important result of a numerical analysis, data visualization being a key component of engineering decision making. In this chapter, we will learn how to use the Python package Matplotlib (Hunter 2007) to generate graphics from data.