To solve the highly non-linear problem of sludge conditioning and dewatering model, the Support Vector Machines (SVM), was used to model the sludge conditioning process and the sample data like organic matter content, dosage ratio of reagent A and reagent B was set as input while the filter index was set as out put. The simulation results indicated that the filtering performance was basically consistent with the actual value of dosage and the model prediction result, and the target value could be well predicted. It shown that the model had a certain value for industrial application.

Topics
Theoretical computer science
1.
Suhua
 
Wu
,
Ping
 
Li
,
Jie
 
Xu
,
Kaiyi
 
Xu
,
Lan
 
Xiao
. (
2016
).
Screening of Flocculants Used in Sludge Deep Dewatering and Optimization of Sludge Conditioning System Applied with Microparticles as Flocculant Aid
.
Water Purification Technology.
 
35
(
2
),
64
–
70
.
Google Scholar
 
2.
M. E.
 
Tipping
. (
2001
).
Sparse Bayesian learning and the relevance vector machine
.
Jonrnal of Machine Learning Research.
 
1
:
211
–
244
.
Google Scholar
 
3.
Taian
 
Liu
,
Xin
 
Xue
,
Wenxu
 
Feng
,
Xinying
 
Liu
. (
2008
).
Research on application of successive overrelaxation for support vector regression arithmetic
.
Computer Engineering and Design.
 
29
(
6
),
1489
–
1490
.
Google Scholar
 
4.
Guixiang
 
Zhang
,
Hongwei
 
Chen
. (
2008
).
Soft--Sening Modal of the AC Asynchronous Electrical Dynamometer Based on SVR
.
China Mechanical Engineering.
 
19
(
2
).
Google Scholar
 
This content is only available via PDF.
© 2019 Author(s).
2019
Author(s)
You do not currently have access to this content.