HOME
Full Articles List
Search by Author
Title: AN ANALYTICAL DECISION SUPPORT SYSTEM IN PROGNOSTICATION OF SURFACE WATER POLLUTION INDICATORS
BibTex 
Open this article in the author's DB |
|
AN ANALYTICAL DECISION SUPPORT SYSTEM IN PROGNOSTICATION OF SURFACE WATER POLLUTION INDICATORS
|
|
|
|
|
|
|
|
||
|
|
|
|
|
|
1314-2704
|
|
|
|
||
|
|
English
|
|
|
|
19
|
|
|
|
2.1
|
|
|
|
|
|
|
|
||
|
|
Today, under the conditions of negative effects of the influence of natural and, especially, anthropogenic factors on the natural environment, automated systems for monitoring the state of the environment are spreading.
The purpose of the work is to develop an analytical decision support system based on the results of forecasting indicators of surface water pollution of the city Lviv. To analyze and forecast the indicators of surface water pollution in Lviv, a trained neural network was created using the Python programming language and the Keras library. The Keras library is a high-level interface for creating neural networks. Keras is written in Python and has powerful and compact program code. Experimental data were obtained from the results of monitoring of the state of the environment of Lviv region during 2012-2016. Observation of the state of surface water in Lviv is carried out by the Communal Enterprise "Administrative and Technical Administration" of the Lviv City Council. The greatest number of excesses was recorded for the following pollutants: suspended matter, total iron, BOD5, ammonia, phosphates, SPAR, COD. In the set of experimental data there are 9 parameters: suspended matter, total iron, ammonia, phosphates, SPAR, nitrates, chlorides, BOD5. The latter, parameter 9 shows the change in the value of the chemical oxygen demand (COD) and it is defined as the output parameter of the model. The Keras library allows you to create neural networks with a minimum number of operations. For the model of the neural network, a sequential Sequental network from the module keras.models with defined layers of keras.layers type Dense is taken. The received neural network enables to synthesize the system of intellectual decision support based on the results of forecasting indicators of surface water pollution in the city of Lviv. The problem solving system and its structure are presented. The software of the analytical system should use the already existing code of the software script in the language of Python, based on the model of forecasting indicators of surface water pollution in the city of Lviv. |
|
|
|
conference
|
|
|
|
||
|
|
||
|
|
19th International Multidisciplinary Scientific GeoConference SGEM 2019
|
|
|
|
19th International Multidisciplinary Scientific GeoConference SGEM 2019, 30 June - 6 July, 2019
|
|
|
|
Proceedings Paper
|
|
|
|
STEF92 Technology
|
|
|
|
International Multidisciplinary Scientific GeoConference-SGEM
|
|
|
|
Bulgarian Acad Sci; Acad Sci Czech Republ; Latvian Acad Sci; Polish Acad Sci; Russian Acad Sci; Serbian Acad Sci & Arts; Slovak Acad Sci; Natl Acad Sci Ukraine; Natl Acad Sci Armenia; Sci Council Japan; World Acad Sci; European Acad Sci, Arts & Letters; Ac
|
|
|
|
49-56
|
|
|
|
30 June - 6 July, 2019
|
|
|
|
website
|
|
|
|
cdrom
|
|
|
|
5329
|
|
|
|
neural network; Python programming language; Keras library; analytical decision support system; indicators of surface water pollution.
|
|
Download PDF