Showing papers by "Shantanu Das published in 2007"

Fractional divergence for neutron flux profile in nuclear reactors

Abstract: The concept of fractional divergence is developed for application to the constitutive neutron diffusion equation for describing the neutron flux profile. The analytical solutions for bare reactors in multiplying media and neutron diffusion in non-multiplying media are obtained for fractional-order differential equations. This definition of fractional divergence describes the neutron flux by not considering it as a classical point quantity. Fractional calculus gives a better representation for a distributed system, and takes 'initialisation-function' (instead of a constant of initialisation) in its solution; thereby encompassing the process history. Therefore, this fractional divergence really describes the reactor flux profile, and having a measurement system based on this concept will generate efficient reactor control. The analytical solutions obtained herein may be verified by standard numerical regression methods in a working reactor to evaluate the order of fractional divergence. This paper discusses the practical application of 300 years of fractional calculus theory to describe reactor systems.

Ratio control with logarithmic logics in a new P&P control algorithm for a true fuel-efficient reactor

Abstract: The concept of expending minimum control energy while stabilising or manoeuvring reactor power for a nuclear reactor is the key for a fuel-efficient reactor control. This concept results from treating the reactor close to its natural characteristics by a new Power and Period (P&P) control scheme. This is proven by spending only the required rate of reactivity (not more) to govern the power by setting together a power error and period error, thus through the characteristic reactor equation (inverse-hour rule), defining the absolute step reactivity required at each power points. However, the use of logarithmic logic, in the new P&P method does add another concept of ratio control, where the effective power error is generated by ratio of demanded power exponential to the observed power exponential curves. This method indeed thus compares the shape of the exponential curve demanded with the observed trajectory, thus close to physics.

Embedding nuclear physics rules and fuel chemistry limits in control algorithms for a fuel-efficient nuclear reactor

Abstract: In a nuclear reactor, the energy which is stored in the fuel is proportional to the amount of fissile material. The consumption of that amount, after a set burn-up, should and must result in delivering a set energy, barring leakages (losses). The loss of energy due to an inefficient control algorithm, after a set burn-up, is a problem that no one has yet addressed. Nuclear fuel chemists and metallurgists may have a given amount of fuel with a set energy content, but control system engineers, treating a nuclear reactor as a conventional PID (Proportional Integral Derivative) control system, use lots of control energy to stabilise the nuclear reactor power and also thereby consume 'unnecessary' fuel energy. The aim of this paper is to draw attention to basic neutronic behaviour of a nuclear reactor and to control it in its natural physics instead of treating it as a conventional PID.