FUZZY-LOGIC MODELING OF TEMPERATURE REGIMES IN THE PROCESSES OF RADICAL SUSPENSION POLYMERIZATION

An approach to the synthesis of neuro-fuzzy models of temperature regimes of industrial reactors for polymer synthesis is proposed for consideration. The expediency of developing such a model is justified by its subsequent application in intelligent robust-adaptive automatic control sys-tems as a reference model for generating control signals. The basis for constructing a fuzzy model of the reactor is an array of data obtained as a result of conducting many experiments on a physical model of a polymer synthesis reactor. Structurally fuzzy model includes three blocks of fuzzy logic. The first block is a fuzzy model of the kinetics of the polymerization process. The output of this block is a monomer conversion curve. The second block of the fuzzy model generates a curve of change in the temperature of the mixture in the reactor normalized to the range [0;1]. This is pos-sible due to the fact that the shape of the temperature change curve depends only on the moment of exhibition of the gel effect, which is inherent in the processes of radical polymerization. This point can be unambiguously determined from the monomer conversion curve. The third block of the fuzzy model calculates the value of the scaling factor based on the current flow rate of the coolant, as well as the set value of the water modulus and the temperature regime in the reactor. After calculating this coefficient, it is multiplied by a normalized curve and the value of the speci-fied temperature regime is added, obtaining at the output the final curve of temperature change in the reactor. The block structure of the proposed fuzzy model is its essential difference from possible analogues and its key advantage. Due to this structure of the model, it is possible to replace, if necessary, some of its elements in order to obtain the possibility of modeling the processes of radical polymerization of different monomers.