There is a concise explanation on Wikipedia:

https://en.wikipedia.org/wiki/Powder_diffraction#Size_and_strain_broadening

Follow the following article and their related references:

Conference Paper X-ray Diffraction Analysis of Cu2+ Doped Zn1-xCuxFe2O4 Spine...

The crystallite size and microstrain can be easily calculated by employing Williamson-Hall (W-H) technique, using relation. In this technique crystallite size and lattice strain dependant broadening is fitted by a curve for all the peak width as a function of diffracting angle and FWHM for a XRD peak. A W-H graphs of the samples thus plotted gives an equation of straight line of the form, y = mx + c with slope equal to micro-structural strain and intercept from which average crystallite size can be obtained.

you may refer following article for more info.

Mote, V., Purushotham, Y. & Dole, B. J Theor Appl Phys (2012) 6: 6. https://doi.org/10.1186/2251-7235-6-6

Williamson–Hall’s formula determines the particle size of the NPs, as shown below β cosθ= 0.89λ/d+ 4εsinθ . where, λ, β, θ, ε and d ascertain the wavelength of Cu Kα radiation, the FWHM, the Bragg angle, the strain and the mean size of the particles respectively. The expression 4sinθ along the x-axis and cosθ along the y-axis are usually plotted for NPs. According to the linear fit of the data, the crystalline size was estimated from the y-intercept

Mohammed Mesrar The Williamson-Hall Plot. W-H plot is used to calculate the crystallite size and microstrain from complex XRD data. That's when both the crystallite size and microstrain vary as a function of the Bragg's angle, we can only calculate these parameters from XRD data using W-H plot. I have provided the practice file (Origin file) as well as the calculation file (Excel file) in the video description. Thanks

https://youtu.be/ipX3iUJ5VcU