Detection and identification of faults in a District Heating Network

Bahlawan, Hilal; Ferraro, Niccolò; Gambarotta, Agostino; Losi, Enzo; Manservigi, Lucrezia; Morini, Mirko; Saletti, Costanza; Spina, Pier Ruggero; Venturini, Mauro

doi:10.1016/j.enconman.2022.115837

Cited by 18 publications

(14 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To this purpose, twenty-three datasets were generated by a digital twin of the DHN under analysis, in which different fault causes and magnitudes were implanted. As discussed in [9], the diagnostic approach correctly detected all datasets. In particular, Bahlawan et al [9] focused on the diagnosis of eight out of twenty-three datasets, i.e., one healthy dataset and seven faulty datasets.…”

Section: Introductionmentioning

confidence: 75%

“…As discussed in [9], the diagnostic approach correctly detected all datasets. In particular, Bahlawan et al [9] focused on the diagnosis of eight out of twenty-three datasets, i.e., one healthy dataset and seven faulty datasets. In [9], the diagnostic approach detected and identified all faults, by also evaluating the correct health index of each pipe of the DHN.…”

Section: Introductionmentioning

confidence: 75%

“…As outlined in [8] and [9], the efficiency of DHNs depends on pipe health state, which can be compromised by three fault types, i.e., leakages, anomalous heat and pressure losses. Water leakages are caused by pipe corrosion [10] and failures in welded joints.…”

Section: Introductionmentioning

confidence: 99%

“…In the literature, only a few diagnostic methodologies tackled this goal, by simultaneously detecting water leakages along pipes, and anomalous heat and pressure losses. This purpose has been recently pursued by Manservigi et al [8] and Bahlawan et al [9], who developed a novel diagnostic approach for detecting and identifying faults affecting DHN pipes. In [8] and [9], the DHN diagnosis was performed by coupling a DHN simulation model with an optimization algorithm.…”

Section: Introductionmentioning

confidence: 99%

“…This purpose has been recently pursued by Manservigi et al [8] and Bahlawan et al [9], who developed a novel diagnostic approach for detecting and identifying faults affecting DHN pipes. In [8] and [9], the DHN diagnosis was performed by coupling a DHN simulation model with an optimization algorithm. The DHN simulation model calculated the DHN variables, i.e., mass flow rate, temperature and pressure at each pipe and node of the DHN, while the optimization algorithm identified pipe health indices in such a manner that the error between measurement and simulated variables was minimized.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Fault diagnosis in district heating networks

Bahlawan

Gambarotta

Losi

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

District Heating Networks (DHNs), which dispatch thermal energy from a heat source to end-users by means of a heat carrier, are composed of pipes that can be affected by faults that endanger system reliability. Thus, reliable diagnostic approaches have to be employed to evaluate the health state of the DHN. In the framework of the ENERGYNIUS research project, the authors of this paper developed a diagnostic approach aimed at detecting and identifying the most frequent faults that affect DHN pipes, i.e., water leakages, heat losses and pressure losses. The diagnostic approach detects and identifies pipe faults by coupling a DHN model with an optimization algorithm. As a result, the health indices of each pipe of the DHN, the fault position, its type and magnitude are provided. This study investigates the capability of the diagnostic approach by using two datasets, in which challenging faults were hypothetically implanted in the DHN of the campus of the University of Parma. The diagnostic approach successfully detected and identified both faults, by also accurately assessing fault magnitude. In addition, the relative error with which each DHN variable is predicted is lower than 0.06 %.

show abstract

Section: Introductionmentioning

confidence: 75%