Analysis of uncertainty of an automated system for testing irrigation sprinklers

Author(s)

Ezequiel Saretta

• Document type: Doctoral Thesis
• Press: Piracicaba.
• Institution: Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC)
• Defense date: 2016-05-17

Examining board members

Tarlei Arriel Botrel; Antonio Pires de Camargo; Jose Antonio Frizzone; Marinaldo Ferreira Pinto; Osvaldo Rettore Neto

• Advisor: Tarlei Arriel Botrel

Abstract

Sprinkler water distribution tests are usually done manually, requiring time and trained labor. Automation of these tests allows saving such resources, and offers potential to minimize failures and/or procedure deviations. Nowadays, calibration and testing laboratories accredited by legal bodies have to present the uncertainty of their instruments and measurement systems. Besides, testing and calibration standards specify acceptable uncertainty, as the standard for sprinkler water distribution test, ISO 15886-3 (2012), which requires an expanded uncertainty up to 3% for 80% of the collectors. The purposes of this work were to develop an automated system for testing sprinkler at laboratory, and perform the analysis uncertainty of measurement, to quantify its magnitude on the testing results, and to support the sizing of collection tubes. The automatic system consisted of a management system, by the supervisory program, a pressurization system, and a collection system, by microprocessed electronic modules developed. According to instructions of the management system, the pressurization system adjusted pressure at the sprinkler by controlling pump rotation, and the collection system measured water application rate through the sprinkler radius. The water from each collector drained to a collecting tube, which was connected to a manifold of valves that contained a pressure transmitter. Each valve was individually triggered in sequence to allow measuring water level within each collection tube, by the transmitter. The analysis of uncertainty showed that the lower diameters of collection tubes presented the lower uncertainty, hence lower diameters should be used when possible. For the collection time, uncertainty decreased as time increased, and should be a minimum value to reach the target uncertainty. Although each application rate requires a minimum time to reach uncertainty, the difference of pressure to be measured stayed the same. Thus, for the tests addressing uncertainty, the difference of pressure was monitored, facilitating the test. Another test condition considered the collection time for 30 laps of the sprinkler, also required by ISO 15886-3 (2012). The third condition considered 1 h of collection, as traditionally done. Water distribution curves obtained from the automated system were similar as those of the conventional tests, for the three situations evaluated. For durations of 1 h or 30 laps of the sprinkler, the automated system required lower total time than the conventional test. However, the system developed demanded more time to perform the test to reach the maximum uncertainty, which is a limitation, even been automatized. In any case, a technician was required just to set parameters and then start the system, so he could allocate his time for other activities. (AU)

• FAPESP's process: 13/14370-4 - Automation for indoor single-radius sprinkler tests
• Grantee: Ezequiel Saretta
• Support Opportunities: Scholarships in Brazil - Doctorate