An heuristic approach has been adopted to obtain empirical scaling formulas for the dependence of neutron emission on predefined machine parameters, such as applied auxiliary heating power, plasma current IP, and toroidal magnetic field BT. The results are intended to be used to predict the neutron emission before each discharge. These formulas are of interest in their own right but more practically can be used as input to control software to pre-set the optimum precollimator apertures for neutron diagnostics such as spectrometers. Formulas have been obtained for data from plasma pulses during the years 1992 and (1994 + 1995), i.e., before and after the divertor installation and major modifications of the Joint European Torus (JET) vessel.

Obtained scaling formulas for moderate beam power PNB show neutron emission after installation of the divertor to be lower than before. However, for high beam power, the scaling laws predict as large and even larger neutron emissions for the (1994 + 1995) neutral beam (NB)-heated plasmas as compared to 1992 plasmas. The dependence on radio-frequency (rf) heating power Prf is significantly larger in the scaling laws deduced prior to the divertor phase than after, which implies more efficient heating in 1992. With the exception of combined NB- and rf-heated plasma pulses, the dependence on plasma current has increased moderately after the modifications of JET. For all observations with combined NB and rf heating, the dependence on Prf is quite small, and the neutron production for this category of discharges is dominated by NB heating.

A set of scaling laws is found that predicts the neutron emission within a factor of 2, which is consistent with our objective.