Systems and methods are provided for predictive control of brown stock handling at a
pulp mill. Various online sensors produce output signals representative of actual values of respective process characteristics, each of which is directly or indirectly affected by adjustments to a corresponding
process variable. A controller uses the output signals or related measurement data to dynamically set target values for the process characteristics based on predicted effects of control responses to the corresponding process variables. The controller further produces control signals to actuators associated with the corresponding process variables based on detected differences between the respective actual values and the target values. Exemplary brown stock washing control systems can optimize various types of brown stock washing configurations, including, for example, vacuum drum washers, compaction baffle washers, chemical washers, direct displacement washers, and wash presses. Cloud-based analysis and
machine learning can also be implemented to improve control algorithms over time.