# US 11,976,817 B2 · Method for detecting a dioxin emission concentration of a municipal solid waste incineration process based on multi-level feature selection

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,976,817 B2
- Document ID: US11976817B2
- Status: Granted
- Kind: B2
- Application number: 17/038,723
- Filed: 2020-10-26
- Published: 2021-02-04
- Granted: 2024-05-07
- Earliest claimed: 2019-05-14
- Assignees: Beijing University of Technology
- Applicants: Beijing University of Technology
- Inventors: Junfei Qiao; Zihao Guo; Jian Tang
- Web page: https://patentgrep.com/patent/US11976817B2

## CPC

- F23G 5/50: Control or safety arrangements
- G01N 33/0036: {specially adapted to detect a particular component (physical analysis of gaseous biological material G01N33/497)}
- F23G 2207/10: Arrangement of sensing devices
- F23G 2208/00: Safety aspects
- F23G 2900/55003: Sensing for exhaust gas properties, e.g. O2 content
- Y02E 20/12: Heat utilisation in combustion or incineration of waste
- F23G: CREMATION FURNACES; CONSUMING WASTE PRODUCTS BY COMBUSTION
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- Y02E: REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION

## Abstract

A method for detecting a dioxin emission concentration of a municipal solid waste incineration process based on multi-level feature selection. A grate furnace-based MSWI process is divided into a plurality of sub-processes. A correlation coefficient value, a mutual information value and a comprehensive evaluation value between each of original input features of the sub-processes and the DXN emission concentration are obtained, thereby obtaining first-level features. The first-level features are selected and statistically processed by adopting a GAPLS-based feature selection algorithm and according to redundancy between different features, thereby obtaining second-level features. Third-level features are obtained according to the first-level features and statistical results of the second-level features. A PLS algorithm-based DXN detection model is established based on model prediction performance and the third-level features. The obtained PLS algorithm-based DXN detection model is applied to detect the DXN emission concentration of the MSWI process.

## Claims

### Claim 1

A computer-implemented method for detecting a dioxin (DXN) emission concentration in a municipal solid waste incineration (MSWI) process based on multi-level feature selection, comprising the steps of:
1. dividing a grate furnace-based municipal solid waste incineration (MSWI) process into a plurality of sub-processes based on an incineration process; wherein the plurality of sub-processes comprise an incineration treatment sub-process, a boiler operation sub-process, a flue gas treatment sub-process, a steam electric power generation sub-process, a stack emission sub-process and a common resource supply sub-process;
2. obtaining a correlation coefficient value and a mutual information value between each of original input features of the sub-processes and the DXN emission concentration; and obtaining a comprehensive evaluation value of candidate input features according to the obtained correlation coefficient value and the obtained mutual information value, thereby completing the selection of first-level features of all of the sub-processes;
3. selecting and statistically processing the first-level features by adopting a feature selection algorithm based on genetic algorithm-based partial least squares (GAPLS) and according to redundancy between different features, thereby completing the selection of second-level features of all of the sub-processes;
4. performing a third-level feature selection according to the first-level features and statistical results of the second-level features within a preset threshold range, thereby completing the selection of third-level features of all of the sub-processes; and
5. establishing a partial least squares (PLS) algorithm-based DXN detection model according to model prediction performance and the third-level features; and detecting the DXN emission concentration by the obtained PLS algorithm-based DXN detection model,
wherein the step of establishing the PLS algorithm-based DXN detection model comprises a microprocessor performing the steps of:
increasing values of a third-level feature selection threshold θDXN3rd between θDXNdownlimit and θDXNuplimit one by one, so as to establish a plurality of first temporary PLS algorithm-based DXN detection models, wherein θDXNdownlimit is a lower limit of the threshold θDXN3rd, and θDXNuplimit is a upper limit of the threshold θDXN3rd;
selecting a second temporary PLS algorithm-based DXN detection model from the plurality of first temporary PLS algorithm-based DXN detection models, wherein the second temporary PLS algorithm-based DXN detection model has a minimum value of root mean square error (RMSE); and
checking the input features of the DXN emission concentration detection model to determine whether the input features comprise concentrations of CO, HCl, O2 and NOx emitted from a chimney, and removing features in the common resource supply sub-process; if the input features do not include concentrations of CO, HCl, O2 and NOx, additionally selecting the third-level features to obtain selected three-level features X3rdsel, thereby varying a number of features that are selected and establishing the PLS algorithm-based DXN detection model.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
arranging the first-level features in series after the step of obtaining the first-level features of all of the sub-processes so as to obtain the first-level features based on a single feature correlation.

### Claim 3 (depends on 2)

The method of claim 2, wherein the DXN detection model comprises input data and output data;
wherein the input data is expressed as X∈RN×P and comprises N samples as row data and P variables as column data; the input data is derived from the sub-processes of the MSWI process; monitoring data of an i-th sub-process is obtained by using a programmable logic controller (PLC) device or a distributed control system (DCS) device installed on site and is expressed as Xi∈RN×Pi; and Xi∈RN×Pi is input data from the i-th sub-process and satisfies Equations (1) and (2);
X=[X1,L,Xi,L,XI]={Xi}i=1I (1)
P=P1+L+Pi+L+PI=Σi=1IPi (2)
wherein I represents the number of the sub-processes, and Pi represents the number of input features in the i-th sub-process;
Xi is expressed as:
X
i
=
[
{
(
x
n
1
)
i
}
n
=
1
N
,
…
,
{
(
x
n
p
i
)
i
}
n
=
1
N
,
…
,
{
(
x
n
P
i
)
i
}
n
=
1
N
]
=
[
(
x
1
)
i
,
…
,
(
x
p
i
)
i
,
…
,
(
x
P
i
)
i
]
=
{
(
x
p
i
)
i
}
p
i
=
1
P
i
(
3
)
wherein (xpi)i represents a pi-th input feature of the i-th sub-process; and xpi={xnpi}n=1N represents a column vector; and
wherein the output data is expressed as y={yn}n=1N∈RN×1, and comprises N samples; and ŷ represents a predicted value.

### Claim 4 (depends on 3)

The method of claim 3, wherein the step of obtaining the correlation coefficient value comprises:
1.1) calculating an original correlation coefficient value between each of the original input features and the DXN emission concentration, wherein an original correlation coefficient value between a p-th input feature (xpi)i={(xnpi)i}n=1N of the i-th sub-process and the DXN emission concentration is calculated according to
(
ξ
corr_ori
p
i
)
i
=
∑
n
=
1
N
[
(
(
x
n
p
i
)
i
-
x
_
p
i
)
⁢
(
y
n
-
y
_
)
]
∑
n
=
1
N
(
(
x
n
p
i
)
i
-
x
_
p
i
)
2
⁢
∑
n
=
1
N
(
y
n
-
y
_
)
2
(
4
)
wherein xpi represents an average value of the p-th input feature of the i-th sub-process; and yl represents an average value of N modeling samples of the DXN emission concentration;
1.2) preprocessing the original correlation coefficient value (ξcorr_oripi)i as follows:
(ξcorrpi)=|(ξcorr_oripi)i| (5)
wherein |⋅| represents an absolute value;
1.3) repeating steps (1.1)-(1.2) until correlation coefficient values of all of the original input features are obtained; and recording the obtained correlation coefficient values as {ξcorrpi}pi=1Pi;
1.4) setting a weight factor of the i-th sub-process as ficorr; calculating a threshold θicorr configured to select correlation coefficient-based input features according to
θ
i
corr
=
f
i
corr
·
1
p
i
⁢
∑
p
i
=
1
P
i
(
ξ
corr
p
i
)
i
(
6
)
wherein a maximum value (ficorr)max and a minimum value (ficorr)min of ficorr are calculated according to Equation (7):
{
(
f
i
corr
)
max
⁢
max
⁡
(
(
ξ
corr
1
)
i
,
…
,
(
ξ
corr
p
i
)
i
,
…
,
(
ξ
corr
P
i
)
i
)
1
p
i
⁢
∑
p
i
=
1
P
i
(
ξ
corr
p
i
)
i
(
f
i
corr
)
min
=
min
⁡
(
(
ξ
corr
1
)
i
,
…
,
(
ξ
corr
p
i
)
i
,
…
,
(
ξ
corr
P
i
)
i
)
1
p
i
⁢
∑
p
i
=
1
P
i
(
ξ
corr
p
i
)
i
(
7
)
wherein max(⋅) is a function for finding a maximum value; and min(⋅) is a function for finding a minimum value;
1.5) selecting the p-th input feature of the i-th sub-process according to rules as follows:
α
i
p
i
=
{
1
,
if
⁢
(
ξ
corr
p
i
)
i
≥
θ
i
corr
0
,
else
⁢
(
ξ
corr
p
i
)
i
<
θ
i
corr
(
8
)
wherein θicorr is taken as a threshold;
1.6) selecting a feature (xpi)i in αipi=1 as a correlation coefficient-selected candidate feature; and recording the correlation coefficient-selected candidate feature as
(
x
(
p
i
)
corr
sel
)
i
;
1.7) performing steps (1.1)-(1.6) for all of the original input features of the i-th sub-process; and recording the selected candidate features as:
(
X
corr
sel
)
i
=
[
(
x
1
)
i
,
…
,
(
x
(
p
i
)
corr
sel
)
i
,
…
,
(
x
(
P
i
)
corr
sel
)
i
]
(
9
)
wherein (Pi)corrsel represents the number of correlation coefficient-selected process variables of the i-th sub-process; and (Xcorrsel)i represents a correlation coefficient-selected candidate feature set selected from the input features of the i-th sub-process; and
1.8) repeating steps (1.1)-(1.7) for all the sub-processes; and recording correlation coefficient measurement-selected features as {(Xcorrsel)i}i=1I.

### Claim 5 (depends on 4)

The method of claim 4, wherein the step of obtaining the mutual information value comprises:
2.1) calculating the mutual information value between each of the original input features and the DXN emission concentration, wherein the mutual information value between the p-th input feature (xpi)i of the i-th sub-process and the DXN emission concentration is calculated according to
(
ξ
mi
p
i
)
1
=
∑
n
=
1
N
∑
n
=
1
N
{
p
rob
(
(
x
n
p
i
)
i
,
y
n
)
⁢
log
⁡
(
p
o
⁢
b
(
(
x
n
p
i
)
i
,
y
n
)
p
rob
(
(
x
n
p
i
)
i
)
⁢
p
rob
(
y
n
)
)
}
(
10
)
wherein prob((xnpi)i,yn) represents a joint probability density; and prob((xnpi)i) and prob(yn) each represent a marginal probability density;
2.2) repeating step (2.1) until mutual information values of all of the original input features are obtained; and recording the obtained mutual information values as {ξmipi}pi=1Pi;
2.3) setting a weight factor of the i-th sub-process as fimi, and calculating a threshold θimi configured to select the input features based on the mutual information value according to
θ
i
mi
=
f
i
mi
·
1
p
i
⁢
∑
p
i
=
1
P
i
(
ξ
m
⁢
i
p
i
)
i
(
11
)
wherein a maximum value (fimi)max and a minimum value (fimi)min of fimi are calculated according to
{
(
f
i
mi
)
max
⁢
max
⁡
(
(
ξ
mi
1
)
i
,
…
,
(
ξ
mi
p
i
)
i
,
…
,
(
ξ
mi
P
i
)
i
)
1
p
i
⁢
∑
p
i
=
1
P
i
(
ξ
mi
p
i
)
i
(
f
i
mi
)
min
=
min
⁡
(
(
ξ
mi
1
)
i
,
…
,
(
ξ
mi
p
i
)
i
,
…
,
(
ξ
mi
P
i
)
i
)
1
p
i
⁢
∑
p
i
=
1
P
i
(
ξ
mi
p
i
)
i
(
1
⁢
2
)
wherein max(⋅) is a function for finding a maximum value; and min(⋅) is a function for finding a minimum value;
2.4) selecting the p-th input feature of the i-th sub-process according to rules as follows:
β
i
p
i
=
{
1
,
if
⁢
(
ξ
mi
p
i
)
i
≥
θ
i
mi
0
,
else
⁢
(
ξ
mi
p
i
)
i
<
θ
i
mi
(
13
)
wherein θimi is taken as a threshold;
2.5) selecting a feature (xpi)i of βipi=1 as a mutual information value-selected candidate feature; recording the mutual information value-selected candidate feature as
(
x
(
p
i
)
mi
sel
)
i
;
2.6) performing steps (2.1)-(2.5) for all of the input features of the i-th sub-process; and recording the selected candidate features as:
(
X
mi
sel
)
i
=
[
(
x
1
)
i
,
…
,
(
x
(
p
i
)
mi
sel
)
i
,
…
,
(
x
(
P
i
)
mi
sel
)
i
]
(
14
)
wherein (Pi)misel represents the number of mutual information value-selected features in the i-th sub-process; and (Xmisel)i represents a candidate feature set selected based on a mutual information value measurement from the input features of the i-th sub-process; and
2.7) repeating steps (2.1)-(2.6) for all the sub-processes; and recording mutual information value measurement-selected features as {(Xmisel)i}i=1I.

### Claim 6 (depends on 5)

The method of claim 5, wherein the step of obtaining the comprehensive evaluation value comprises:
3.1) for the i-th sub-process, taking the intersection of the mutual information-selected features (Xmisel)i and the correlation coefficient-selected features (Xcorrsel)i according to Equation (15), thereby obtaining a comprehensive evaluation value-selected candidate feature set,
(
X
corr
⁢
_
⁢
mi
sel
)
i
=
(
X
mi
sel
)
i
⋂
(
X
corr
sel
)
i
=
[
(
x
1
)
i
,
…
,
(
x
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
,
…
,
(
x
(
P
i
)
corr
⁢
_
⁢
mi
sel
)
i
]
(
15
)
wherein | represents the intersection;
x
i
(
p
i
)
corr
⁢
_
⁢
mi
sel
represents a (pi)corr_misel-th candidate feature of the i-th sub-process; and a correlation coefficient value of the (pi)corr_misel-th candidate feature is
(
ξ
corr
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
;
and a mutual information value of the (pi)corr_misel-th candidate feature is
(
ξ
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
;
3.2) performing normalization according to Equations (16) and (17) so as to eliminate size differences of the correlation coefficient value and mutual information value of the different input features;
(
ζ
corr
⁢
_
⁢
norm
(
p
i
)
corr
⁢
_
⁢
m
⁢
i
sel
)
i
=
(
ζ
corr
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
∑
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
(
ζ
corr
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
(
16
)
(
ζ
mi
⁢
_
⁢
norm
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
=
(
ζ
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
∑
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
(
ζ
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
(
17
)
wherein
(
ζ
corr
⁢
_
⁢
norm
p
corr
⁢
_
⁢
mi
sel
)
i
represents a standardized correlation coefficient value of the pcorr_misel-th candidate feature of the i-th sub-process; and
(
ζ
mi
⁢
_
⁢
norm
p
corr
⁢
_
⁢
mi
sel
)
i
represents a standardized mutual information value of the pcorr_misel-th candidate feature of the i-th sub-process;
3.3) defining the comprehensive evaluation value of the candidate input features as
ζ
i
(
p
i
)
corr
⁢
_
⁢
mi
sel
,
expressing
ζ
i
(
p
i
)
corr
⁢
_
⁢
mi
sel
as
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
k
i
corr
·
ζ
corr
⁢
_
⁢
norm
(
p
i
)
corr
⁢
_
⁢
mi
sel
+
k
i
mi
·
ζ
mi
⁢
_
⁢
norm
(
p
i
)
corr
⁢
_
⁢
mi
sel
(
18
)
wherein kicorr and kimi each represent a proportional coefficient; and
kicorr+kimi=1; and
3.4) repeating steps (3.1)-(3.3) until comprehensive evaluation values of all of the candidate input features are obtained; and recording the obtained comprehensive evaluation values as
{
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
}
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
.

### Claim 7 (depends on 6)

The method of claim 6, wherein kicorr is equal to 0.5; and kimi is equal to 0.5.

### Claim 8 (depends on 6)

The method of claim 6, wherein the step of obtaining the comprehensive evaluation value further comprises:
4.1) setting a weight factor of the i-th sub-process as ficorr_mi; calculating a threshold θi1stsel configured to select the input features based on the comprehensive evaluation value according to
θ
i
1
⁢
stsel
=
f
i
corr
⁢
_
⁢
mi
·
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
⁢
∑
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
(
19
)
wherein a maximum value (ficorr_mi)max and a minimum value (ficorr_mi)min of ficorr_mi are calculated according to
{
(
f
i
corr
⁢
_
⁢
mi
)
max
=
max
⁡
(
(
ζ
corr
⁢
_
⁢
mi
1
)
i
,
…
,
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
,
…
,
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
)
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
⁢
∑
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
(
f
i
corr
⁢
_
⁢
mi
)
min
=
min
⁡
(
(
ζ
corr
⁢
_
⁢
mi
1
)
i
,
…
,
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
,
…
,
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
)
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
⁢
∑
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
(
P
i
)
corr
⁢
_
⁢
mi
sel
(
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
)
i
(
20
)
4.2) selecting a (pi)corr_misel-th candidate input feature of the i-th sub-process according to rules as follows:
γ
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
{
1
,
if
⁢
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
≥
θ
i
1
⁢
stsel
0
,
else
⁢
ζ
corr
⁢
_
⁢
mi
(
p
i
)
corr
⁢
_
⁢
mi
sel
<
θ
i
1
⁢
stsel
(
21
)
wherein θi1stsel is taken as a threshold;
4.3) performing steps (4.1)-(4.2) for all the original candidate input features; selecting variables of
γ
(
p
i
)
corr
⁢
_
⁢
mi
sel
=
1
as comprehensive evaluation value-selected input features; and expressing the variables as:
(
X
1
⁢
st
sel
)
i
=
[
(
x
1
)
i
,
…
,
(
x
p
i
sel
)
i
,
…
,
(
x
P
i
sel
)
i
]
(
22
)
wherein (X1stsel)i represents first-level features of the i-th sub-process selected using a comprehensive evaluation value measurement from the candidate feature set selected by a correlation coefficient method and a mutual information method; and
4.4) repeating steps (4.1)-(4.3) until the first-level features of all the sub-processes is obtained.

### Claim 9 (depends on 8)

The method of claim 8, wherein the step of arranging the first-level features in series comprises:
arranging the first-level features in series to obtain the first-level features X1stsel based on the single feature correlation;
X
1
⁢
st
sel
=
[
(
X
1
⁢
st
sel
)
1
,
…
,
(
X
1
⁢
st
sel
)
i
,
…
,
(
X
1
⁢
st
sel
)
I
]
=
[
x
1
1
⁢
st
sel
,
…
,
x
p
1
⁢
st
sel
,
…
,
x
P
1
⁢
st
sel
]
(
23
)
wherein
x
p
1
⁢
st
sel
represents a p1stsel-th feature in a first-level feature selection set;
P
1
⁢
st
sel
=
∑
i
=
1
I
P
i
sel
represents the number of all of the first-level features; and X1stsel represents single feature correlation-based first-level feature obtained by serially combining the first-level features of all of the sub-processes.

### Claim 10 (depends on 8)

The method of claim 8, wherein a strategy of second-level feature selection comprises:
inputting the first-level features X1stsel; running the GAPLS algorithm times;
outputting the second-level features (X2ndsel)j and then outputting the number of times that the respective first-level input features are selected; and statistically processing the second-level features that are selected Jsel times, wherein when a GAPLS model prediction error is smaller than a prediction error average obtained by running the GAPLS algorithm J times, a second-level feature is selected;
recording the number of times that a p1stsel-th feature is selected as
f
num
p
1
⁢
st
sel
;
accordingly, recording all P1stsel-th features of the first-level features as
{
f
num
p
1
⁢
st
sel
}
p
1
⁢
st
sel
=
1
P
1
⁢
st
sel
;
wherein J is the number of times that the GAPLS algorithm runs; Jsel is the number of GAPLS models prediction errors of which are smaller than a prediction error average; and (X2ndsel)j represents multiple feature redundancy-based second-level features selected by jth run of the GAPLS algorithm.

### Claim 11 (depends on 10)

The method of claim 10, wherein the step of the second-level feature selection comprises:
5.1) setting the number of times that the GAPLS algorithm runs as J; setting GAPLS algorithm parameters; initializing a population size, maximum genetic algebra, mutation probability, a crossover method and a number of latent variables of the PLS algorithm; and setting j=1 and starting the selection of the second-level features;
5.2) determining whether the GAPLS algorithm runs J times; if yes, proceeding to step (5.11); if no, proceeding to step (5.3);
5.3) performing binary encoding for features, wherein a length of a chromosome is the number of input features; 1 implies that a feature is selected; and 0 implies that no feature is selected;
5.4) performing random initialization on population;
5.5) evaluating the fitness of the population; and calculating a root mean square error of cross-validation (RMSECV) using a leave-one-out cross-validation method;
5.6) determining whether a termination condition of the maximum genetic algebra is reached, if no, proceeding to step (5.7); if yes, proceeding to step (5.9);
5.7) performing genetic operations comprising selection, crossover and variation, wherein the selection is performed through an elite substitution strategy, that is, individuals with poor fitness are replaced with individuals with good fitness; the crossover is performed through single point crossover; and the genetic variation is performed through single point mutation;
5.8) obtaining a new population and proceeding to step (5.5);
5.9) obtaining an optimal individual after running the GAPLS algorithm times; and performing decoding to obtain selected second-level features and recording the selected second-level features as (X2ndsel)j;
5.10) setting j=j+1; and proceeding to step (5.2);
5.11) calculating an average value of root mean square errors (RMSE) of a prediction model obtained by running the GAPLS algorithm J times; recording the number of the root mean square errors of the GAPLS model that are larger than the average value as Jsel; processing the second-level features that are selected Jsel times by counting the number of times that the P1stsel-th feature in the first-level features is selected,
{
(
x
2
⁢
nd
sel
)
j
}
j
=
1
J
sel

{
f
num
1
1
⁢
st
sel
,
…
,
f
num
p
1
⁢
st
sel
,
…
,
f
num
P
1
⁢
st
sel
}
=
{
f
num
p
1
⁢
st
sel
}
p
1
⁢
st
sel
=
1
P
1
⁢
st
sel
,
(
24
)
1
≤
f
num
p
1
⁢
st
sel
≤
J
sel
wherein
f
num
p
1
⁢
st
sel
is the number of times that the p1stsel-th feature in the first-level features is selected.

### Claim 12 (depends on 11)

The method of claim 11, wherein the population size is 20; the maximum genetic algebra is 40; a maximum number of latent variables of the PLS algorithm is 6; and the mutation probability is 0.005.

### Claim 13 (depends on 11)

The method of claim 11, wherein the step of the third-level feature selection comprises:
according to the number of times
{
f
num
p
1
⁢
st
sel
}
p
1
⁢
st
sel
=
1
P
1
⁢
st
sel
that all the p1stsel-th features in the first-level features are selected, setting a scale factor as fDXNRMSE; determining the lower limit of the threshold θDXN3rd configured to select the third-level features as θDXNdownlimit; calculating θDXNdownlimit according to:
θ
DXN
downlimit
=
floor
⁢
{
f
DXN
RMSE
·
1
P
1
⁢
st
sel
⁢
∑
p
1
⁢
st
sel
=
1
P
1
⁢
st
sel
f
num
p
1
⁢
st
sel
)
(
25
)
wherein floor(⋅) represents a function that returns integers;
calculating a maximum value (fDXNRMSE)max and a minimum value (fDXNRMSE)min of fDXNRMSE according to
{
(
f
DXN
RMSE
)
max
=
max
⁡
(
f
num
1
1
⁢
st
sel
,
…
,
f
num
p
1
⁢
st
sel
,
…
,
f
num
P
1
⁢
st
sel
)
1
P
1
⁢
st
sel
⁢
∑
p
1
⁢
st
sel
=
1
P
1
⁢
st
sel
f
num
p
1
⁢
st
sel
(
f
DXN
RMSE
)
min
=
min
⁡
(
f
num
1
1
⁢
st
sel
,
…
,
f
num
p
1
⁢
st
sel
,
…
,
f
num
P
1
⁢
st
sel
)
1
P
1
⁢
st
sel
⁢
∑
p
1
⁢
st
sel
=
1
P
1
⁢
st
sel
f
num
p
1
⁢
st
sel
(
26
)
finding a maximum value of the number of times that all the p1stsel-th features in the first-level features are selected based on the upper limit θDXNuplimit of the threshold configured to select the third-level features,
θ
DXN
uplimit
=
max
⁡
(
f
num
1
1
⁢
st
sel
,
…
,
f
num
p
1
⁢
st
sel
,
…
,
f
num
P
1
⁢
st
sel
)
(
27
)
recording the threshold θDXN3rd between θDXNdownlimit and θDXNuplimit; and performing the third-level feature selection according to
μ
p
=
{
1
,
if
f
num
p
1
⁢
st
sel
≥
θ
DXN
3
⁢
rd
0
,
else
f
num
p
1
⁢
st
sel
<
θ
DXN
3
⁢
rd
(
28
)
wherein
f
num
p
1
⁢
st
sel
represents the number of times that the p1stsel-th feature in the first-level features is selected by running the GAPLS algorithm J times; μp represents a threshold selection criterion for selecting the third-level features;
sequentially storing feature variables of μp=1 in X3rdsel_temp and calculating the RMSE, wherein X3rdsel_temp serves as input variables in the establishment of the PLS algorithm-based DXN detection model; and X3rdsel represents the third-level features selected from X1stsel based on a feature selection threshold θ3rd and prior knowledge.

### Claim 14 (depends on 1)

The method of claim 1, wherein variables of the PLS algorithm-based DXN detection model have 287 dimensions.

### Claim 15 (depends on 1)

The method of claim 1, wherein weight factors ficorr, fimi and ficorr_mi of feature selection of the correlation coefficient value and the mutual information value of the first-level features are 0.8.

### Claim 16 (depends on 1)

The method of claim 1, wherein there are 132 feature variables selected by the comprehensive evaluation value; for the selected 132 process variables based on the single feature correlation, an optimal process variable combination is determined using the GAPLS algorithm so as to remove redundant features.