In-core neutron detectors are used in large nuclear reactors for the purpose of acquiring knowledge about the spatial flux distribution, which helps in control, monitoring and protection of the reactors. Random errors and faults, which cause degradations of the signals from the true values, invalidate the in-core detector measurement data. A scheme utilizing the analytical redundancy among physically redundant in-core detectors greatly facilitates removing the ill-effects of these errors. Particularly, data-based or process-history-based schemes are suitable when there are a large number of variables as in the case of in-core detectors. In this paper, application of a Data Reconciliation (DR) scheme (with the aim of minimizing the effect of random errors) coupled with a Fault Detection and Isolation (FDI) technique (which makes the data free from faults) to the in-core detector measurement data of the Advanced Heavy Water Reactor (AHWR) is presented. This reactor uses Vanadium Self-Powered Neutron Detectors (VSPNDs) as in-core neutron detectors. Since the constraint model developed for all the in-core neutron detectors as a whole cannot yield satisfactory performance from the DR and FDI, they are developed for the individual clusters of VSPNDs using the Iterative Principal Component Analysis (IPCA), a multi-variate statistical technique. The FDI technique used is the Iterative Principal Component Test (IPCT) discussed in Sagar et al. (2015c). The effectiveness of this scheme is tested as follows: (i) the mathematical model of the AHWR is used to simulate measurement data from VSPNDs; (ii) the data from VSPNDs are grouped as clusters based on correlations among the variables; and (iii) the effectiveness of the DR-based FDI of VSPNDs in some clusters is established based on some performance indices, when the spatial control of the reactor is active.