How to reduce the lines of code with multi variable equations?

Question

I am trying to reduce lines of code because I realized that I am repeating the same equations every time. I am programming a contour map and putting several sources of intensity into it. Until now I put 3 sources, but in the future I want to put more, and that will increase the lines a lot. So I want to see if it is possible to reduce the lines of "source positions" and "Intensity equations". As you can see the last equation is a logaritmic summation of z1, z2 and z3, is it possible to reduce that, any idea?

import numpy as np
import matplotlib.pyplot as plt

def run():
    # mesh
    N = 100
    x = np.linspace(-N, N)
    y = np.linspace(-N, N)
    X,Y = np.meshgrid(x,y)


    # sources coordinates
    p1 = [-5,-1]
    p2 = [6 , 0]
    p3 = [-10 , -8]


    # sources postions

    R1 = np.sqrt((((p1[0]-X)**2)+((p1[1]-Y)**2)))
    R2 = np.sqrt((((p2[0]-X)**2)+((p2[1]-Y)**2)))
    R3 = np.sqrt((((p3[0]-X)**2)+((p3[1]-Y)**2)))

    # # Intensity Equation

    z1 = np.round(10 * np.log10((((((R1/(1000**(1/3)))**-1.45) * 516)/(20**(-3)))**2)),0)
    z2 = np.round(10 * np.log10((((((R2/(1000**(1/3)))**-1.45) * 516)/(20**(-3)))**2)),0)
    z3 = np.round(10 * np.log10((((((R3/(4000**(1/3)))**-1.45) * 516)/(20**(-3)))**2)),0) 

    z_total = 10*np.log10((10**(z1/10))+(10**(z2/10)+(10**(z3/10))))

    # # Ploting

    Z = np.array(z_total).reshape(len(y),len(x))
    cs = plt.contourf(X,Y,Z,10)
    plt.colorbar()
    plt.show()

if __name__ =='__main__':
    run()

Answer 1

You could iterate over certain parts in a loop.

I tried to keep the same format overall and just rearranged the code to show how you might do it.

import numpy as np
import matplotlib.pyplot as plt

def run():
    # mesh
    N = 100
    x = np.linspace(-N, N)
    y = np.linspace(-N, N)
    X,Y = np.meshgrid(x,y)

    # sources coordinates
    p = [
        [-5,-1],
        [6 , 0],
        [-10 , -8],
    ]

    # sources postions + intensity Equation
    z_log = 0
    for px, py in p:
        r = np.sqrt((((px-X)**2)+((py-Y)**2)))
        z = np.round(10 * np.log10((((((r/(1000**(1/3)))**-1.45) * 516)/(20**(-3)))**2)),0)
        z_log += 10**(z/10)
    z_total = 10*np.log10(z_log)

    # # Ploting
    Z = np.array(z_total).reshape(len(y),len(x))
    cs = plt.contourf(X,Y,Z,10)
    plt.colorbar()
    plt.show()

if __name__ =='__main__':
    run()

A recommendation I'd have is spread out operators (x + y * 2 is preferred over x+y*2), give your variables descriptive names (eg. coordinates and not p), and split your calculations into more parts. There are so many brackets in z for example that I'd hate to try and debug if something was not working correctly.