Demo Mixtures.

Examples

### Expreiment #1: test all combinations of mixtures for three analytes A, B and C

set1 <- c("A 0.05", "B 0.05", "C 1", # pure analytes
  "A 0.05, B 0.05", "A 0.05, C 1", "B 0.05, C 1", # binary mixtures
  "A 0.05, B 0.05, C 1") # a ternary mixture

# data model 'plsr' leads to a visually nice distribution of gas classes via PCA scoreplot
sa1 <- SensorArray(model = "plsr", num = 3:5, dsd = 0)

# look at the level of signal in reponse to pure analytes and mixtures
# - the highest and the lowest levels of signal correspond to two pure analytes, 
#   A and B, respectively (that will result in a nice triangle-bounded distribution of PCA scores)
p1 <- plotSignal(sa1, set = set1)
p1



# If air samples are included, (in most cases) PCA shows the signal magnitudes 
# across gas classes in respect to air-level (zero-level) 
p2 <- plotPCA(sa1, set = rep(set1, 3))
p2 



p3 <- plotPCA(sa1, set = rep(set1, 3), air = FALSE)
p3 



### Experiment #2: two analytes A and C, and their binary mixture AC 
#   at differenct concentration levels

set2 <-  c("A 0.01", "A 0.05", "C 0.1", "C 1", "A 0.01, C 0.1", "A 0.05, C  1")

# default data model 'ispline'
sa2 <- SensorArray(num = 3:5, dsd = 0)

p4 <- plotSignal(sa2, set = set2)
p4



p5 <- plotPCA(sa2, set = rep(set2, 3), air = FALSE)
p5