Professional Documents
Culture Documents
Fault Detection in A Grid-Connected Photovoltaic System Using Adaptive
Fault Detection in A Grid-Connected Photovoltaic System Using Adaptive
Fault Detection in A Grid-Connected Photovoltaic System Using Adaptive
Solar Energy
journal homepage: www.elsevier.com/locate/solener
A R T I C LE I N FO A B S T R A C T
Keywords: In this paper, an adaptive monitoring scheme with Fuzzy Logic Filter (FLF) is developed and applied to monitor a
MWPCA Grid-Connected Photovoltaic System (GCPVS). This method is based on Principal Component Analysis (PCA) and
PCA Moving Window Principal Component Analysis (MWPCA). It is designed to generate adaptive thresholds for its
Grid-connected PV systems monitoring indices. The FLF filters the monitoring indices to reduce the number of False Alarms (FA) and in-
Process monitoring
crease the Fault Detection Rate (FDR). The application is carried out on the GCPVS of the Power Electronics and
Data-driven methods
Renewable Energy Research Laboratory (PEARL) of Malaya University. The proposed technique is compared
Adaptive thresholds
against PCA method in terms of FAR reduction. The detection ability of the adaptive thresholding with FLF
monitoring scheme is tested first on simulated faults then it is applied to detect a real abnormal behaviour. The
results show that the proposed method is effective in reducing the number of false alarms and in detecting
different types of faults with high accuracy.
https://doi.org/10.1016/j.solener.2018.09.024
Received 23 March 2018; Received in revised form 7 July 2018; Accepted 12 September 2018
0038-092X/ © 2018 Elsevier Ltd. All rights reserved.
M. Ammiche et al. Solar Energy 174 (2018) 762–769
handle the dynamic behaviour in both steady and unsteady states, Lou Once the number of principal components, a, is selected, Xs be-
et al. (2017) proposed a two-step PCA. There are other techniques be- comes as
sides PCA have been also extensively developed and employed for s + E
Xs = X (4)
system monitoring such that Canonical Variate Analysis (CVA) (Russell
et al., 2000, Shang et al., 2017), Partial Least Squares (PLS) (Wold et al., Such that
2001) and Independent Canonical Analysis (ICA) (Zhong and Deng,
X ̂ T
s = TP (5)
2016).
The main problem that one can face when applying linear data- ∼∼T
E = TP (6)
driven techniques, such as PCA, to a process monitoring with fixed
∼ ∼
control limits is constantly the high number of false alarms. These false where P ∈ Rmxa , P ∈ Rm x(m − a) , T ̂ ∈ Rnxa and T ∈ Rn x(m − a)
alarms may be due to the presence of random noise in the system,
dynamics, nonlinear behaviours, the operation setpoint changes and the 2.1.2. The number of principal components
mathematical formulas by which the control limits are calculated. To In the literature, many methods are used for the number of principal
address this problem, adaptive thresholding technique with Fuzzy Logic components, a, selection. These methods mainly include the parallel
Filter (FLF) is developed. The monitoring scheme combines both PCA analysis, cumulative percentage of variances, scree plot and cross-va-
and MWPCA to result in an adaptive and robust technique to false lidation (Chiang et al., 2001, Sheriff et al., 2017). In this work, the
alarms. The use of PCA allows the construction of a statistical model parallel analysis criterion has been used. So, the number of retained
that properly fits the process. This model is kept constant during the principal components, a, is defined as the point at which the eigenvalue
monitoring task. The constructed model is then used to evaluate the profile of the original data crosses the eigenvalue profile of generated
monitoring indices of each new testing sample. A moving window of data assuming independent observations (Chiang et al., 2001).
fixed size allows the generation of the adaptive thresholds for the
evaluated monitoring indices after they have been filtered by FLF. Both 2.1.3. PCA-based fault detection
adaptive thresholds and FLF involve in the number of false alarms re- Two monitoring indices are associated with PCA-based fault de-
duction besides to the PCA performance enhancements. The application tection method, the Hotelling’s T2 and Q statistics
of the proposed method has been performed on the grid-connected PV (Ramahaleomiarantsoa et al., 2012). These indices are provided by
system of the Power Electronic and Renewable Energy Research
̂ −1P T̂ x
T 2 = xT PΛ (7)
Laboratory (PEARL) at Malaya University, Malaysia. The adaptive
scheme with FLF has been tested first to evaluate its detection ability of T 2
Q = (I −PP̂ ̂ ) x = xT (I −PP̂ ̂ ) x
T 2
(8)
random sudden changes, and incipient fault. These faults are associated
with short circuits, an abrupt increase or decrease of currents and The control limits, which correspond to each monitoring index, are
voltages, sensors failure, and slow temperature augmentation. The given by
monitoring scheme has been then used to detect a real abnormal op-
a (n−1)(n + 1)
eration of the GCPVS. The obtained results demonstrate the effective- Tδ2 = Fδ (a, n−a)
n (n−a) (9)
ness of the proposed technique in promptly and correctly detecting
different types of faults. 1
2 h0
The remaining sections of this paper are organized as follows. ⎛ Cδ 2θ2 h 0 θ h (h −1) ⎞
Qδ = θ1 + 1 + 2 0 20
Section 2 is dedicated to the mathematical formulation of the PCA- ⎜ θ1 θ1 ⎟
⎝ ⎠ (10)
based fault detection method. In Section 3, the proposed adaptive
thresholding scheme with FLF has been provided and explained. where n is the number of observations and a is the number PCs.
Section 4 shows the application results and discussions. The last section, Fδ (a, n−a) is an F-distribution of a, n-a degree of freedom evaluated at
Section 5, is restricted to some conclusions and findings. given confidence level (1−δ ) .
m
2. Backgrounds θi = ∑ λij , i = 1, 2, 3
j=a+1 (11)
2.1. Principal component analysis 2θ1 θ2
h 0 = 1−
3θ22 (12)
2.1.1. Definition and mathematical formulation
Principal Component Analysis (PCA) is a dimensionality reduction Cδ is the normal deviate corresponding to (1−δ ) percentile.
technique able to capture the most dominant variances in the data
(Chiang et al., 2001). PCA defines the principal component subspace 3. Proposed adaptive monitoring scheme with FLF
and the residual subspace by means of a linear transformation (Chiang
et al., 2001). Given a process data of m measured variables x1, x2, …, xm The performance of PCA based fault detection with fixed control
and n observations. These data can be stored in a zero mean and unity limit is poor in terms of false alarms rate reduction due to the random
variance matrix Xs = [XsT1 , XsT2 , ⋯, Xsm
T
]. The linear transformation, that noise in the process measured variables and the process dynamics. In
projects the data onto the subspaces, is provided by the following addition, the mathematical formulas used to develop the thresholds
equation always allow some normal samples to violet the limits which corre-
spond to the type I error. The false alarms rate with this technique can
T = Xs P (1)
be reduced by increasing the confidence interval, δ . However, high
Where T is the score matrix. P is an m by m orthonormal matrix values of δ decreases the PCA monitoring indices sensitivity and hence
consists of the covariance matrix of Xs eigenvectors. The covariance some of the faults may not be detectable. The proposed monitoring
matrix of Xs is given by the Singular Value Decomposition (SVD) as scheme aims to overcome this problem and improve the PCA based
fault detection technique. It is a combination of both PCA and Moving
cov (Xs ) = PT ΛP (2)
Window PCA (MWPCA) schemes. The monitoring indices are being
Eq. (1) permits to rewrite Xs as evaluated using a fixed constructed PCA model while the adaptive
thresholds are being updated through a fixed length moving window.
Xs = TPT (3)
The sensitivity enhancement is handled by the adaptive thresholds
763
M. Ammiche et al. Solar Energy 174 (2018) 762–769
whereas the FAR reduction is handled by both the adaptive thresholds a statistical model by determining the necessary parameters of the PCA
and the FLF. The FLF filters the Q and T2 before making a decision model (means, standard deviation, the number of principal components
about the tested samples and then the filtered values are considered for and the initial thresholds). This task is performed on the training data-
the adaption process of the thresholds. set that is collected while the process is not exhibiting an abnormal
behaviour. Once PCA model is obtained, it is validated using a testing
3.1. Fuzzy logic filter data-set that is also acquired when the process is under its normal op-
eration. During the validation tasks, the window length, the FLF para-
Fuzzy logic filters have been widely used to remove noise from meters (the membership function and the fuzzy rules) and the fixed
different measured signals (Ferahtia et al., 2012; Naso et al., 2006). In thresholds, that will be used when a fault is detected, are selected.
this paper, the goal of the Fuzzy Logic Filter (FLF) is to remove the spike These parameters are determined according to the obtained FAR and
noise from the monitoring indices to reduce the number of alarms. The resulted in fault recovery delay for this data-set. The offline measure-
implemented FLF has been inspired from (Naso et al., 2006). The fuzzy ments are taken from the inputs and the outputs of the GCPVS. At the
system inputs are u1, u2 and u3 while the fuzzy system output is K . For end of this phase, the adaptive thresholding scheme with FLF is ready to
instance, the vector Z = [ z1 z2 ⋯ z n ] is to be filtered, then the inputs, monitor new testing samples. At each time a new observation is
in a specific moving window length of three samples, are given in Eqs. available, this sample is shifted and scaled using the mean and standard
(13)–(15) deviation of the constructed PCA model then its monitoring indices are
computed. These indices are passed through the implemented FLF to
u1 (j ) = zj−f (13)
obtain the filtered Q and T2 which will be fed to the decision block
u2 (j ) = zj + 1−zj (14) where they are compared with their initial thresholds. If the filtered
monitoring indices are greater than the initial thresholds, then the
u3 (j ) = zj + 2−zj + 1 (15) sample is considered to be faulty; as a result, an alarm is generated via a
binary variable, fault indicator. The adaptive thresholds are then set to
where f represents the filtered sample of the last (j − 1)th moving
the predetermined fixed thresholds while the initial control limits are
window
set to the adaptive thresholds for the next sample monitoring. In the
wj − 1 = [ zj − 1 ≡ f zj zj + 1] (16) case where the filtered monitoring indices are both less than their as-
sociated initial thresholds, the fault indicator generates no alarm and
These inputs will be transformed to linguistic variables. The eva-
the adaptation mechanism will be activated to update the thresholds
luation of pre-determined fuzzy rules yields to obtain the membership
where the new sample is included in the moving window and the oldest
information K (j ) . Consequently, the filtered sample is obtained by
one is excluded. The adaptation procedures are given by Eqs. (18), (19),
zjfiltered = f + K (j )(z j−f ) (17) (10), and (12).
The following case study represents a numerical example to illus-
The output of the filter, z jfiltered , depends on the value of K (j ) .
trate how the adaptive thresholding scheme with FLF can be applied to
For this work, the vector Z represents either the Q or T2.
monitor multivariate processes.
3.2. Adaptive thresholds generation
3.3. Case study
The monitoring indices thresholds depend on the number of re-
tained principal components a and the number of observations (The Consider the following system
length of the sliding window) (refer to Eqs. (9)–(11)). However, with
the developed technique, these two parameters are kept unchanged. ⎧ X1 = u1 + ε1
The Q statistic threshold can be updated by using the linear projection ⎪ X2 = u1 + ε2
⎪
defined in equation Eq. (1). Within a selected window of fixed length, X3 = u1 + ε3
⎨
the corresponding variances of the residual subspace from one time to ⎪ X 4 = u2 + ε4
another serve to update the control limit. If wn is the moving window ⎪ X5 = u1 + 3u2 + ε5 (20)
∼ ⎩
over the data of length H and P ∈ Rm × (m − a) is the matrix of the ei-
genvectors that correspond to smallest eigenvalues, then the projected
data vector onto the residual subspace is given by εi is an independent white noise variable uniformly distributed over
[0.02, 2].
∼
Bj = wn∗Pj (18) ui is a random variable which follows a normal distribution with
zero mean and 0.1 variance.
wn is first shifted and scaled using the mean and standard deviation
∼ ∼
values of the constructed PCA model. Pj is the jth vector in P where
The training data-set is stored in the data matrix
j = 1, 2, ⋯, m−a
X = [ X1 X2 X3 X 4 X5 ] ∈ R70 × 5 .
θi of Eq. (11) can be calculated by
The mean and standard deviation values of the generated training
m−a
data-set are the vectors Mean ∈ R1 × 5 and Strd ∈ R1 × 5, respectively
θi = ∑ (var (Bj ))i i = 1, 2, 3
j=1 (19) Mean = [0.9603 1.0724 0.9312 0.8945 0.8832] (21)
where var (·) stands for the variance value.
Using Eqs. (10) and (12) and after substituting the values of θi , the Strd = [0.6020 0.5859 0.5972 0.5943 0.6422] (22)
new threshold of the index Q can be computed.
The control limit for the T2 statistic is updated by fitting the em- These two vectors are used to scale and shift the data matrix X to
pirical distribution over the window of the past evaluated values. zero mean and unity variance. The new matrix obtained throughout this
Fig. 1 represents the schematic diagram of the proposed adaptive operation is denoted by Xs where its (5 × 5) covariance matrix can be
monitoring scheme with FLF. manipulated accordingly to Eq. (2) by two matrices Λ and P . These
In this figure, the offline phase includes two important tasks in matrices are respectively,
process monitoring which are the process training and the model va-
Λ = diag(1.3288 1.2142 1.0284 0.8500 0.5786) (23)
lidation. The process training is used to train the process and construct
764
M. Ammiche et al. Solar Energy 174 (2018) 762–769
Fig. 1. The schematic diagram of the adaptive monitoring scheme with FLF.
Table 1 Using the parallel analysis criterion, three principal components are
Process description. retained. The corresponding eigenvectors of principal and residual
∼
Specifications Polycrystalline Monocrystalline Thin film subspaces P and P are given in Eq. (24).
The initial thresholds for 99% confidence level are
Number of modules 16 25 20
DC power per module (Wp) 125 75 135 Qthin = 7.9291 (25)
Voltage at max. power (V) 17.3 47.0 17.0
Current at max. power (A) 7.9 2.9 4.4 2
Tthin = 14.6636 (26)
DC/AC inverter capacity (W) 1600 1600 2500
Inverter max. DC voltage (V) 600 600 700
Inverter max. AC voltage (V) 220–240 220–240 220–240 A testing data-set, collected from the fault-free process, is employed
to validate this PCA model (Mean, Stdrd, number of principal compo-
nents), set the FLF parameters (the membership function parameters
⎡ ⎤ and the fuzzy rules), determine the appropriate length of the moving
⎢− 0.6019 0.4698 − 0.0454 0.1759 − 0.6197 ⎥ window and select the fixed thresholds. Suppose that the optimal
⎢− 0.0126 0.7517 0.3706 − 0.0384 0.5441 ⎥ window length that corresponds to the minimum FAR for the testing
P = ⎢− 0.3144 − 0.2752 0.6665 − 0.6049 − 0.1238 ⎥
⎢− 0.5199 − 0.3715 0.2382 data-set is 30 samples, then the last 30 observations of the training
0.6233 0.3829 ⎥
⎢ − 0.5181 − 0.0244 − 0.5997 − 0.4618 0.3976 ⎥ data-set initiate the window wn ∈ R30 × 5 . The new testing sample
⎢ ⏟ ⏟
∼ ⎥ xT = [1.9468 1.0006 2.0161 0.1681 1.6616] is scaled using the mean
⎣ P P ⎦ (24)
and standard deviation values. The standardized observation is
765
M. Ammiche et al. Solar Energy 174 (2018) 762–769
AC line system
Printer
CPU CPU
Computer 1 Computer 2
Training Data-Set
0
Q Qth
10
766
M. Ammiche et al. Solar Energy 174 (2018) 762–769
Q Qth Q Qth
0 0
10 10
500 1000 1500 2000 2500 200 400 600 800 1000 1200 1400
2 2 T2 T2 th
T T th
0 0
10 10
500 1000 1500 2000 2500 200 400 600 800 1000 1200 1400
Fault Indicator Fault Indicator
1 1
0 0
500 1000 1500 2000 2500 200 400 600 800 1000 1200 1400
Samples Samples
Q Qth Q Qth
0
10 0
10
200 400 600 800 1000 1200 1400 5 200 400 600 800 1000 1200 1400
10
T2 T2th T2 T2th
0
10 0
10
200 400 600 800 1000 1200 1400 200 400 600 800 1000 1200 1400
Fault Indicator Fault Indicator
1 1
0 0
200 400 600 800 1000 1200 1400 200 400 600 800 1000 1200 1400
Samples Samples
Fig. 7. Testing data-set. Fig. 9. Detection of random fault.
Table 3 Fault 3
FAR of adaptive thresholding scheme with FLF and PCA for the training and Q Qth
testing data-sets.
Method Adaptive thresholding scheme with FLF PCA
0
Statistics Q T2 Q T2 10
Training data-set 0 0.61 7.99 4.45 200 400 600 800 1000 1200 1400
Testing data-set 0 0 7.14 5.94 T2 T2th
767
M. Ammiche et al. Solar Energy 174 (2018) 762–769
0
10 The resulted monitoring indices of the adaptive thresholding
scheme with FLF along with their corresponding adaptive thresholds for
500 1000 1500 2000 the GCPVS are shown in Figs. 4 and 5.
T2 T2th
It can be seen from these figures that all the samples of the training
and testing data-sets are under the developed thresholds as it is de-
monstrated by the fault indicator signals. These results are obtained
0 because the GCPVS is operating under its normal circumstances. To
10
illustrate the improvements achieved with the proposed method, the
500 1000 1500 2000 PCA technique is also applied to these data-sets. The monitoring indices
Fault Indicator with their corresponding thresholds for the PCA are depicted in the
1 Figs. 6 and 7. Form the fault indicator signal of these figures, it can be
0
500 1000 1500 2000 seen that a large number of false alarms is resulted. These false alarms
Samples affect the monitoring accuracy where the process status is not clearly
Fig. 11. Detection of a real fault. identified. The obtained FARs are tabulated in Table 3. The resulted
FAR in the Q is reduced to 0% for both training and testing data-sets.
The adaptive thresholding with FLF decreases also the FAR to 0% and
4. Application
0.61% in the T2. According to the results, the detection accuracy has
been noticeably improved. These results confirm the ability of the
4.1. Process description
adaptive scheme with FLF to reduce the effect of false alarms in a
GCPVS monitoring.
The GCPVS from which the measurements are taken is installed at
The detection ability of the monitoring proposal has been tested on
the Power Electronic and Renewable Energy Research Laboratory
simulated faults of Table 2. The results of the simulated faults detection
(PEARL) of Malaya University (Hasan and Mekhilef, 2017). This system
are shown in Figs. 8–10. The monitoring indices start violet the adap-
contains three inverters: INV 01, INV 02 and INV 03 which are con-
tive thresholds at the time index at which the fault is introduced sig-
nected to 16 modules of polycrystalline silicon (POLY), 25 modules of
nifying that an abnormality is occurred except for the drift fault, in
monocrystalline (MONO) and 20 modules of thin film, respectively.
which the detection is delayed. This delay is due to the slope of the fault
Several sensors are installed within the system; these sensors measure
that is taken very small to simulate the worst case. Q and T2 return
solar radiation, wind speed, panel and ambient temperatures, voltages
under the adaptive thresholds when the fault is disappeared. These
and currents at different stages of the GCPVS. This PV system is con-
figures show that all the faults have been successfully and accurately
nected to supply power to the laboratory building. The description of
detected.
the process is shown in Table 1 (Hossain et al., 2017). Fig. 2 represents
New data have been measured for online monitoring. The number of
the three types of solar panels. The schematic diagram of the system is
observations of this set is n = 2465 sample recorded every 5 min. The
shown in Fig. 3.
monitoring results are shown in Fig. 11. This figure illustrates that some
of the samples are above the adaptive thresholds which indicate the
4.2. Application procedures occurrence of a fault. The fault indicator signal confirms the occurrence
of the abnormality since it switches from zero to one for some samples.
For monitoring purposes, two data-sets have been collected from the The occurrence interval of the fault, from this figure, appears to be
GCPVS described above. The first data-set consists of m = 29 measured small; this is because the sampling interval is high.
variables and the n = 2766 observations with a sampling interval of 1
observation per 5 min. These data are used as a training data-set to 5. Conclusion
construct the PCA model (computing the mean, standard deviation, and
the number of PCs). These data are employed also to compute the initial In this paper, an adaptive monitoring scheme with fuzzy logic filter
thresholds. Before applying PCA, the data have been processed where has been applied to monitor a Grid-Connected Photovoltaic System
all the observations exceeding μ ± 3σ are removed, such that μ and σ (GCPVS). The applied technique is able to detect different types of
represent respectively the mean and standard deviation values of each faults that may occur in a GCPVS; hence it permits to increase the en-
measured variable. This data processing is required for constructing a ergy production with high quality. It serves also to avoid damages of
quite representative PCA model. The remaining data-set is the testing GCPVS equipment, economic losses and accidents. This method has
data-set; it is used to validate the PCA model and appropriately de- been presented and demonstrated how it can be used in monitoring. It is
termine the length of the moving window as well as the FLF parameters a data-driven technique based on PCA and MWPCA that needs only
(the membership function parameters). This data-set consists of m = 29 collected data from a GCPVS. The generated adaptive thresholds in-
measured variables and n = 1500 observations in which one sample is crease the monitoring indices sensitivity besides the number of false
recorded every 5 min. The fixed control limits, which are utilized when alarms reduction. The use of the filter involves also in the number of
a fault is detected, are determined by trial and error. The condition false alarms reduction since it removes noise and outliers. The appli-
under which these limits are selected is that they don’t affect the fault cation results show that no false alarm is signalled for the testing data-
recovery. To test the detection ability of the proposed method and set. Furthermore, a real fault has been successfully and accurately de-
evaluate its monitoring performance, three types of faults have been tected.
injected in the data: abrupt, random and drift faults. The simulated
faults description is shown in Table 2. 8 principal components have References
been retained using parallel analysis technique. These components ex-
plain 79.53% of the total variance in the data. The best window length Chiang, L.H., Russell, E.L., Braatz, R.D., 2001. Fault Detection and Diagnosis in Industrial
is found to be 142 samples. The time required to complete the com- Systems. Springer, London.
Deng, X., Tian, X., Chen, S., Harris, C.J., 2017. Fault discriminant enhanced kernel
putations and return the state of the monitored data for the proposed
768
M. Ammiche et al. Solar Energy 174 (2018) 762–769
principal component analysis incorporating prior fault information for monitoring processes monitoring. Can. J. Chem. Eng. https://doi.org/10.1002/cjce.22855.
nonlinear processes. Chemom. Intell. Lab. Syst. 162, 21–34. Madeti, S.R., Singh, S.N., 2017. Monitoring system for photovoltaic plants: a review.
Dhimish, M., Holmes, V., 2016. Fault detection algorithm for grid-connected photovoltaic Renew. Sustain. Energy Rev. 67, 1180–1207.
plants. Sol. Energy 137, 236–245. Naso, D., Scalera, A., Aurisicchio, G., Turchiano, B., 2006. Removing spike noise from
Elshenawy, L.M., Yin, S., Naik, A.S., Ding, S.X., 2010. Efficient recursive principal com- railway geometry measures with a fuzzy filter. IEEE Trans. Syst. Man Cybernetics 36,
ponent analysis algorithms for process monitoring. Ind. Eng. Chem. Res. 49, 252–259. 485–494.
Fazai, R., Taouali, O., Harkat, M.F., Bouguila, N., 2016. A new fault detection method for Ramahaleomiarantsoa, J., Heraud, N., Sambatra, E.J.R., Razafimahenina, J.M., 2012. On
nonlinear process monitoring. Int. J. Adv. Manuf. Technol. 87, 3425–3436. the sensitivity of principal components analysis applied in wound rotor induction
Ferahtia, J., Djarfour, N., Baddari, K., Kheldoun, A., 2012. A fuzzy logic-based filter for machines faults detection and localization. Int. J. Energy Sci. 2, 262–271.
the removal of spike noise from 2D electrical resistivity data. J. Appl. Geophys. 87, Rato, T.J., Reis, M.S., 2013. Defining the structure of DPCA models and its impact on
19–27. process monitoring and prediction activities. Chemom. Intell. Lab. Syst. 125, 74–86.
Firth, S., Klomas, K.J., Rees, S.J., 2010. A simple model of PV system performance and its Rato, T.J., Reis, M., Schmitt, E., Hubert, M., De Ketelaere, Bart, 2016. A systematic
use in fault detection. Sol. Energy 84, 624–635. comparison of PCA-based statistical process monitoring methods for high-dimen-
Ge, Z., Chen, J., 2016. Plant-wide industrial process monitoring: a distributed modeling sional, time-dependent processes. Process Syst. Eng., AIChE 62, 1478–1493.
framework. IEEE Trans Ind. Inf. 12, 310–321. Russell, E.L., Chiang, L.H., Braatz, R.D., 2000. Fault detection in industrial processes
Harrou, F., Suna, Y., Taghezouit, B., Saidi, A., Hamlati, M.-E., 2018. Reliable fault de- using canonical variate analysis and dynamic principal component analysis.
tection and diagnosis of photovoltaic systems based on statistical monitoring ap- Chemom. Intell. Lab. Syst. 51, 81–93.
proaches. Renew. Energy. https://doi.org/10.1016/j.renene.2017.09.048. Shang, L., Liu, J., Zhang, Y., 2017. Efficient recursive kernel canonical variate analysis for
Hasan, R., Mekhilef, S., 2017. Highly efficient flyback microinverter for grid-connected monitoring nonlinear time-varying processes. Can. J. Chem. Eng. https://doi.org/10.
rooftop PV system. Sol. Energy 146, 511–522. 1002/cjce.22897.
Hassani, S., Saidur, R., Mekhilef, S., Taylor, R.A., 2016. Environmental and exergy benefit Sheriff, M.Z., Mansouri, M., Karim, M.N., Nounou, H., Nounou, M., 2017. Fault detection
of nano fluid-based hybrid PV/T systems. Energy Convers. Manage. 123, 431–444. using multiscale PCA-based moving window GLRT. J. Process Control 54, 47–64.
Hossain, M., Mekhilef, M., Danesh, M., Olatomiwa, L., Shamshirband, S., 2017. Silvestre, S., López, L.M., Kichou, S., Pacheco, F.S., Pumar, M.D., 2016. Remote super-
Application of extreme learning machine for short term output power forecasting of vision and fault detection on OPC monitored PV systems. Sol. Energy 137, 424–433.
three grid-connected PV systems. J. Clean. Prod. 167, 395–405. Spataru, S., Sera, D., Kerekes, T., Teodorescu, R., 2015. Diagnostic method for photo-
Kim, il. S., 2016. On-line fault detection algorithm of a photovoltaic system using wavelet voltaic systems based on light I-V measurements. Sol. Energy 119, 29–44.
transform. Sol. Energy 126, 137–145. Wold, S., Trygg, J., Berglund, A., Antti, H., 2001. Some recent developments in PLS
Ku, W., Storer, R.H., Georgakis, C., 1995. Disturbance detection and isolation by dynamic modeling. Chemom. Intell. Lab. Syst. 58, 131–150.
principal component analysis. Chemom. Intell. Lab. Syst. 30, 179–196. Zhong, N., Deng, X., 2016. Multimode non-Gaussian process monitoring based on local
Liu, X., Li, K., McAfee, M., Irwin, G.W., 2011. Improved nonlinear PCA for process entropy independent component analysis. http://doi.org/10.1002/cjce.22651.
monitoring using support vector data description. J. Process Control 21, 1306–1317. Zhu, J., Geand, Z., Song, Z., 2017. Distributed parallel PCA for modeling an monitoring of
Lou, Z., Shen, D., Wang, Y., 2017. Two-step principal component analysis for dynamic large-scale plant-wide processes with big data. IEEE Trans. Ind. Inf. 13, 1877–1885.
769