A novel control method with an automatic tuning of PID controller parameters has been developed for a better regulation of dissolved oxygen concentration in fed-batch fermentations of E. coli. Agitation speed and the oxygen partial pressure in the inlet gas stream were manipulated. In this method, heuristic reasoning allows improved tuning decisions to be made based upon the supervision of certain control performance indices in the same cognitive manner as in expert control, which obviates the need for assumptions about the process or disturbance patterns. The disturbing effect of feed rate changes is compensated for by using a feedforward scheme.
When the oxygen demand of the cells is low, the dissolved oxygen concentration is controlled by manipulating agitation speed. Once the agitation speed reaches its maximum value as the cell concentration increases, the control variable is switched to the oxygen partial pressure in the inlet gas stream. The control input consists of feedback and feedforward parts. The feedback part is determined by using the PID control method and the feedforward part is determined in proportion to the change in the feed rate. The proportional gain is updated on-line by using a set of heuristic rules based on supervision of three performance indices. These indices are the output error covariance, the average value of output error, and the input covariance, which are calculated on-line using a moving window. The integral and derivative time constants are determined from the period of output response, which resembles the procedure of Ziegler-Nichols' method. The specific growth rate is maintained at a low level to avoid acetic acid accumulation and thus to achive a high cell density. The specific growth rate is estimated from the carbon dioxide evolution rate.
In fed-batch fermentations the performance was satisfactory for the entire period of the culture in spite of the changes in the feed rate and the switching of the control input.