Search

A feedforward/feedback version of a single input/single output self-tuning controller has been developed and tested both via simulation studies and experiments on a section of an absorption/desorption pilot plant. The algorithm is based on the recursive least squares estimation of parameters for the linear models relating the output to the controlled input and to the disturbances; adaptation is achieved using a variable forgetting factor. The control input at each time interval is calculated using one of several single-step extended-horizon control strategies. The results show that the performance of the algorithm is insensitive to the choice of initial parameters, all of which have a readily identifiable intuitive basis. The algorithm is especially robust against deterministic disturbances (measurable and unmeasurable) and unknown and varying time delays. Computational load beyond that of a feedback-only version is minimal.

A study of the identification of the dynamics of a pilot scale absorption-desorption system and the subsequent design of a multivariable control scheme are reported. A multivariable transfer function model was obtained from experimental input-output data in which the inputs (steam supply and liquid circulation rate) were perturbed simultaneously using uncorrelated pseudo random sequences. A two-step, least-squares estimator was employed for the identification of the model. A multivariable controller was then designed on the basis of the direct Nyquist array method. The implementation of the controller design indicated the utility of the design procedure as compared to independent single input, single output control loops; there is, however, some scope for additional tuning.

A robust version of the self-tuning regulator is developed. The regulator, which requires relatively little knowledge of system characteristics (estimated order of transfer function polynomials and an upper bound for transportation delays), has been shown to yield stable control and convergence for linear, time-invariant systems. Simulations and practical tests on a large pilot-scale process have shown that the inclusion of a variable forgetting factor and an “extended horizon” control criterion provides the regulator with a sufficient degree of robustness and flexibility to perform well in a nonlinear time-varying environment. The regulator makes use of intuitively easy-to-understand concepts and leaves few degrees of freedom for the potential user. Furthermore, extensive experiments and simulation studies have shown it to be insensitive to choice of initial conditions and dynamic characteristics set by the user.

A method was devised to yield chemical reaction kinetic parameters from nonisothermal, nonisobaric flow experiments. The system studied was the pyrolysis of propane at high temperatures (800° to 1,000°C.). At these temperatures the rates of the various reactions are so high that a batch or even an isothermal flow experiment is impossible. To keep the conversions low so that the initial stages of decomposition could be studied, the feed gas was diluted with varying amounts of nitrogen. Residence times in the reactor were in the millisecond range. The reactor exit gas was analyzed by mass spectrometry. The method developed in this work is not limited to simple kinetic studies, but can be useful in complicated series and parallel reactions which often require nonisothermal conditions.