All geek questions in one place

Is using BPNN to control excess water quality?

I am developing a freshwater quality management device that can be used for freshwater organisms such as lakes and rivers. The project is distributed in three parts:

  • The first part is devoted to the acquisition of parameters such as pH, turbidity, etc.
  • The second part deals with the adoption of corrective measures based on parameters. For example, if the pH is too low, the device will inject a stock solution to maintain a pH of 7-7.5.
  • Now the third part is devoted to predicting the health status of the lake based on the obtained parameters (pH / turbidity, etc.). The forecasting algorithm should take into account the parameters and develop a correlation between them to explain how long the lake will maintain. To achieve this, I am currently inclined to use the Back Upagation Neural Network (BPNN), as I have found that several other people / institutions prefer NN for water quality management. *

My problem now is whether using BPNN would be redundant for this project? If so, which method / tool should be used?

* 1 , 2 and 3

+4
source share
3

- "" . , , , . , , . , - . , .

?

  • ( , )
  • KNN
  • RBF Networks

- "" . ... .

+8

. , , "", "", "" .., . n (x): = g (Wh (Ax + b) + c), A W - , b c - , h - , , , g - , , . , x , pH, .., , . "" , , u ( pH ..), , . "" A, W, b, c. , A W ( , b c). , n (x) , () , . , A, W, b, c E (A, W, b, c): = (n (x) - t (x)) ^ 2, t (x) - ( , ). E A, W, b, c - E , , .

, , E, . .

, , , ? , , . , , , , . , , , , .

, python, , . , , , . ( ) , . python, sklearn . , . .

- . , , . , , , , .

http://pastebin.com/KrUAX9je - , , "" f (a, b, c) = .

+4

Backpropagation (BP) - . . , BP . , , . , , 10 (pH, ...) , , 2-3 , .

, - BEST . , .

+2

Source: https://habr.com/ru/post/1621105/

More articles:


All Articles