Vataman V. Improved PWR automated control system by using an approximation model of internal disturbances

The development of conditions for safe operation of a nuclear power unit in cyclic load conditions by improving methods and models of the automated control system of a NPP with a VVER reactor, which use an approximation model of internal core disturbances calculated by the method of two-dimensional identification of the core, whose properties change over time and are nonlinear. Where a mathematical 3D model of a PWR type nuclear reactor has been developed, which includes the distribution of the active zone into elementary cells by the number of the layer (y), the number of the fuel rod row (x), and the number of the cell in the row (z). The spatially distributed mathematical 3D model of the PWR core was improved, which, unlike others, was used as an element of the automated real-time control condition, which allowed to calculate a multicoordinate interconnected current control system that depends on the arising internal and external disturbances on the temperature field of the reactor core, which changes the neutron field and further allows to constant balance of production and energy consumption between the equipment circuits. Developed method of two-dimensional identification of internal disturbances in the core of PWR that affect the adjustable parameter, which are nonlinear and vary over time and depend on operating conditions. An approximation model of the transfer function has been developed, which corresponds to the results of solving a system of nonlinear differential equations that calculate the properties of the core and realizes the complex interpretation inverse problem of the adjustable parameter. A control system is developed to which a polynomial model is integrated, which during the current day performs functions for solving control problems, the coefficients to which are supplied from a mathematical 3D model, which are calculated in parallel during the operation of the NPU. Calculation according to the approximation model makes it possible to predict the processes in the core in advance for an interval of about 24 hours or for 3 shifts of operational personnel. Such a forecast is enough to perform a power maneuver, which is performed by the operator within two hours. The consequences of the power maneuver are manifested for about 30 hours. Therefore, after 18‒20 hours of operation of the nuclear power unit, the control subsystem receives updated calculation data for the 3D model and recalculates the coefficients of the approximation model. The developed model, which is based on the inverse problem of complex interpretation, can be used in existing automatic control systems with a rule base of NPP of the VVER. In the developed computer automation system, the axial offset is kept constant and doesn’t affect the subsequent characteristics of the cyclic power change in a multicoordinate interconnected distributed system. The computerized automation system of NPP of the VVER was improved for timely correction of adjustable parameters when increasing or decreasing the nuclear reactor load, which ensures stable and controlled power generation throughout the reactor core at the required parameters constant balance of production and energy consumption in the first and second circuits of NPP of the VVER, due to the fact that an approximation model is additionally integrated into the computer simulation mathematical model, which during the current day performs functions for solving control tasks, the coefficients to which are input from the mathematical model in 3D representation, which are calculated in parallel during operation. The approximation model of NPU control was verified by comparing changes in technological parameters using two research methods: during shockless programmer switching and using the approximation model. The upgraded automatic control system with rule base made it possible to remove fluctuations in energy production in the core due to fluctuations in xenon concentration and maintain a constant value of axial offset. However, the method requires preliminary synthesis of an approximation model as an inverse problem of complex interpretation based on the results of neutron-physical calculations of the core and subsequent calculation of the settings to obtain adjustable parameters.