I'm currently working on power plant time series data and my main objective is finding out the optimal combination of independent variables which would keep "SO2 concentration (dependent variable) below a threshold of 200mg/m3. For some clarity, some of the independent variables I'm working on include: Power, Number of Pumps, Heat Value, Dust, % Calcium in Coal etc. So, a good insight for example would be the optimal amount of power (in combination with other variables) required to keep "SO2 conc" under the above threshold.
I've been working on fitting regression models to the data and examining variable importance to find out how influential the predictors are to "SO2 conc". However, this doesn't really address my objective.
I'm using R to work on this but I'm not really sure what statistical problem this is as it seems to require more than just regression modelling. Does anyone have any ideas on how to approach this?
