**6. Conclusions**

This paper proposes a new IHMP method to locate PMs for SE purposes in radial distribution systems. In more detail, the proposed approach is focused on branch power flow estimation, and it allows the determination of the minimum number and best position of load power meters in order to achieve a target accuracy in the branch power flow estimation. The proposed measurement technique assumes that power meters are located in a limited number of nodes, whereas in the remaining grid nodes the load powers are estimated as pseudo-measurements, with a given uncertainty. A sensitivity index is defined, which takes into account the branch power flow uncertainty all over the

network; this ensures that, at any branch, the power flow is evaluated with an uncertainty value below a prefixed threshold.

The IHMP procedure was implemented for a backward/forward load flow algorithm proposed by the authors, which allows the evaluation of MV branch power flows starting from LV load power measurements. The integration of the proposed IHMP method with this load flow solution allows the obtainment of an improvement in power flow estimation accuracy, with limited costs for PMs. In fact, the proposed solution can be developed by using a low-cost measurement infrastructure that exploits PMs at LV level instead of those at MV level. The proposed method was implemented and tested in real case of the MV radial distribution network of the Island of Ustica, under different load conditions. The obtained results confirm that the method is useful to find the measurement configuration that guarantees a low uncertainty all over the network. In the case under study, the proposed meter placement procedure ensures an adequate power flow uncertainty level when PMs are installed in about half of the grid nodes.

The proposed IHMP technique is incremental, thus it allows not only for the best measurement configuration to be found in a new distribution network, but also allows the determination of how to obtain an enhancement of an existing measurement configuration. Both applications can be of practical interest for DSOs, as they may be interested in either installing a new measurement infrastructure or enhancing an already owned one. In both cases, the possibility of using measurements at LV level can increase the cost-effectiveness of the measurement infrastructure. In fact, thanks to the use of LV measurements instead of MV ones and the use of the proposed IHMP method, a good tradeoff can be reached between low cost and accuracy: the cost is reduced because of the choice to install PMs at the low-voltage side of the power transformer, while the IHMP technique allows suitable PM placement in order to obtain the target estimation accuracy.

**Author Contributions:** Conceptualization, A.C., V.C., D.D.C. and G.T.; Data curation, G.A., S.G., N.P. and E.T.; Investigation, A.C., V.C., D.D.C. and G.T.; Methodology, G.A., A.C., V.C., D.D.C., S.G., N.P., E.T. and G.T.; Software, G.A., S.G., N.P. and E.T.; Supervision, A.C., V.C., D.D.C. and G.T.; Validation, G.A., A.C., V.C., D.D.C., S.G., N.P., E.T. and G.T.; Writing—original draft, G.A., S.G., N.P. and E.T.; Writing—review & editing, A.C., V.C., D.D.C. and G.T.

**Funding:** This research received no external funding

**Acknowledgments:** The authors wish to thank the local DSO of Ustica (Impresa Elettrica D'Anna e Bonaccorsi S.R.L.) for their support and the collected network parameters and measurement data.

**Conflicts of Interest:** The authors declare no conflict of interest.
