Value-added Plots for models estimated by plm-package in r and interpretation?

Question

If I estimated my model using the lm-function, I would be able to create Value-Added Plots typing the following code:

ols_model <- lm( log(y) ~ log(x1) + log(x2) + log(x3), data = data)
avPlots(ols_model)

However, I estimated my model using the plm-function because I'm using panel data methods.

fe_model <- plm( log(y) ~ log(x1) + log(x2) + log(x3),
                      model = "within", 
                      data = data, 
                      index = c("country","year"), 
                      effect = "twoways")

How can I create Value-Added Plots in this case? My current solution is to replicate the same model with the lm-package (knowing that the standard errors are not correct) and to run the avPlots-command. So, basically I'm doing this:

fe_lm <- lm( log(y) ~ log(x1) + log(x2) + log(x3) + time_dummy1 + time_dummy2 + ... + country_dummy1 + country_dummy2 + ... , data = data)
avPlots(fe_lm)

However, now I'm not sure how to interpret the plots. Do I want to see a linear relationship between each pair of variables or rather a non-pattern?