Solution of metrological water-ecological problems using fuzzy logic methods

In order to reduce the number of erroneous water management decisions, it is necessary to have sufficiently strict metrological support for studies of the composition and properties of natural waters. The use of standard methods requires expanding the scope of hydromonitoring and increasing the accuracy of the data obtained in order to ensure their representativeness, including the ability to refl ect general trends and transfer the results of the study to a wider range of objects. As a possible alternative to standard methods, it is proposed to analyze the accumulated measurement information using fuzzy logic. A methodology for applying the methods and mathematical apparatus of fuzzy (multi-valued) logic to solve metrological water-ecological problems has been developed and tested using the example of water quality assessment. Using fuzzy logic methods, the infl uence of four cause factors “Leaching”, “Weathering and sedimentation”, “Anthropogenic discharges”, “Self-purification” on the effect factor “Decrease in water quality against background” in the fi ve-level Harrington scale adopted in expert statistical assessment was studied. Using the software package of fuzzy logic MatLab Fuzzy Logic, forecasts of changes in water quality depending on four factors were obtained. The method of assessing the quality of natural water was tested on a specific example of setting up a fuzzy system for assessing water quality. It was found that the risks of errors still exist, but they were significantly reduced by taking into account poorly formalized linguistic information from expert hydrologists. The possibility of using the method for an a priori assessment of the probable consequences of changes in factors infl uencing the decline in water quality and taking preventive measures to optimize the operation of the water use system was shown.