Deconvolution Protocols for Melt Crystallization Peaks in Recyclate Blends
Deconvolution of DSC melt crystallization peaks isolates PE and PP fractions in recyclates to verify composition before compounding.
Nonlinear Least Squares
Meaning
Optimization algorithm finds the best fit for a mathematical model when the relationship between independent and dependent variables is not linear. Nonlinear least squares is the standard computational method for extracting material constants from experimental rheology or tensile data. It minimizes the sum of the squares of the differences between the observed values and the predicted model.
In polymer science, this is used to fit complex models to viscosity data measured at various shear rates.
Parameter Estimation
Convergence of the algorithm requires a good initial guess for the variables to avoid falling into a local minimum. For a moulder using flow simulation software, nonlinear least squares determines the coefficients that describe how a resin will flow through a thin gate. If these parameters are inaccurate, the predicted injection pressure will not match the actual pressure required on the factory floor.
Iterative Process
Iterative software repeatedly adjusts the model parameters until the change in the residual sum of squares falls below a predefined limit. This ensures the model converges on the most statistically likely value.
Statistical Reliability
Analysis of the residuals helps to identify if the chosen model correctly describes the physical behavior of the resin. This statistical check ensures that the fitted parameters are reliable for process simulation.