Intelligent monitoring method and apparatus for abnormal working conditions in heavy metal wastewater treatment process based on transfer learning and storage medium

Patent metadata

Field	Value
Publication number	US-12578694-B2
Application number	US-202117908922-A
Country	US
Kind code	B2
Filing date	Feb 25, 2021
Priority date	Mar 19, 2020
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

An intelligent monitoring method and apparatus for abnormal working conditions in a heavy metal wastewater treatment process based on transfer learning and a storage medium are provided. During an intelligent monitoring, the abnormal working conditions can be automatically and intelligently recognized by fusing data in the treatment process of the heavy metal wastewater different in source; specifically, a normal sample Y SD in the treatment process of the heavy metal wastewater with fixed sources and a small number of normal samples Y TD in the treatment process of the heavy metal wastewater with unknown sources are utilized; and first, a data representation dictionary D SD of Y SD is obtained through learning on Y SD , and then considering different distribution of Y SD and Y TD , a transfer learning method is adopted to fuse characters of Y TD into a dictionary learning process to obtain a dictionary D TD with higher generalization ability.

First claim

What is claimed is:

1 . A monitoring method for abnormal working conditions in a heavy metal wastewater treatment process based on transfer learning, comprising:
1) constructing an offline dictionary for heavy metal wastewater treatment data samples with fixed sources according to historically-collected heavy metal wastewater treatment data samples with fixed sources; 2) acquiring an augmented dictionary D TD corresponding to effective heavy metal wastewater treatment data samples with unknown sources by utilizing historically-collected effective heavy metal wastewater treatment data samples with unknown sources to perform the transfer learning on the offline dictionary; 3) calculating a reconstruction error of the effective heavy metal wastewater treatment data samples with unknown sources through the augmented dictionary D TD , and acquiring a control limit under a working condition in the heavy metal wastewater treatment process through a kernel density estimation based on the reconstruction error; and 4) calculating a reconstruction error of to-be-monitored data y t under the augmented dictionary D TD , wherein if the reconstruction error of to-be-monitored data y t calculated is less than the control limit, it is considered that a current heavy metal wastewater treatment process is normal, otherwise, it is considered that the current heavy metal wastewater treatment process is abnormal, a calculation of the reconstruction error refers to a 2-norm calculation of a sample collection value of a to-be-calculated reconstruction error and an expression value of the augmented dictionary D TD for samples and corresponding sparse coding: 5) setting direction selection vectors sequentially by setting an abnormal wastewater index positioning objective function, and determining a reconstruction error of each wastewater index under the augmented dictionary D TD until abnormal amplitudes on abnormal samples are converged to determine abnormal wastewater indexes; and 6) in response to determining that the abnormal wastewater indexes include PH, changing reagent amount to achieve pH stability.

2 . The monitoring method according to claim 1 , wherein the constructing the offline dictionary for the heavy metal wastewater treatment data samples with fixed sources comprises the following steps:
step 1.1: collecting historical samples, wherein a sensor is utilized for collecting heavy metal wastewater treatment historical samples with fixed sources, and a sample historical set with fixed sources is Y SD ; Y SD =[y 1 ,y 2 , . . . , y N s ] ∈R m×Ns , y j represents an ith heavy metal wastewater treatment historical sample with fixed source, 1≤j≤N s , each sample comprises m wastewater indexes {pH value, current density, conductivity, initial heavy metal concentration and flow}, the m is 5, and N s represents a number of samples in Y SD ; and step 1.2: representing Y SD through a dictionary D 1 and a sparse coding X based on a sparse representation principle, constructing an objective function of offline dictionary learning, and acquiring an optimal initial dictionary D SD corresponding to Y SD and a sparse coding X SD corresponding to D SD by solving the objective function of the offline dictionary learning, wherein the objective function of the offline dictionary learning constructed is specifically expressed as Formula (1):

〈

D
SD

,

X
SD

〉

=

arg

min

D
1

,
X



Y
SD

-

D
1

⁢
X



2
2

(
1
)

s
.
t
.

∀
i

,



x
i



0

≤
T

,

wherein, an initial value of the dictionary D 1 is a matrix formed by in-column arranging of K samples randomly selected from Y SD , K=10*m, is a set value of a number of nonzero elements in each column of a sparse coding matrix, and


•


2
2

and ∥•∥ 0 represent 2-norm and 0-norm correspondingly; and x i represents an ith column in X.

3 . The monitoring method according to claim 2 , wherein the objective function of the offline dictionary learning is solved through a K-SVD method, and the dictionary D 1 and the sparse coding X are constantly updated until the optimal initial dictionary D SD corresponding to Y SD is obtained.

4 . The monitoring method according to claim 1 , wherein the acquiring the augmented dictionary D TD by utilizing the historically-collected effective heavy metal wastewater treatment data samples with unknown sources to perform the transfer learning on the offline dictionary comprises the following steps:
utilizing a sensor for collecting effective heavy metal wastewater treatment historical samples with unknown sources, wherein an effective sample set with unknown sources is Y SD ; and utilizing an initial dictionary D SD and a corresponding sparse coding X for representing the Y TD based on a sparse representation principle, constructing an objective function of sparse coding corresponding to heavy metal wastewater treatment data samples with unknown sources, solving an optimal sparse coding X P corresponding to the effective sample set Y TD with unknown sources through the transfer learning, and then acquiring a corresponding optimal dictionary through X P ; wherein the objective function of sparse coding is specifically expressed as Formula (2):

Assignees

• Univ Central South

Inventors

• Huang Keke
• Wen Haofei
• Yang Chunhua
• Zhu Hongqiu
• Li Yonggang

Classifications

• C02F1/00

  Treatment of water, waste water, or sewage (C02F3/00 - C02F9/00 take precedence) · CPC title

• C02F1/463

  by electrocoagulation · CPC title

• C02F2209/40

  Liquid flow rate · CPC title

• C02F2209/05

  Conductivity or salinity · CPC title

• C02F2209/006

  comprising a software program or a logic diagram · CPC title