Open access peer-reviewed chapter

Fault Residuals Based on Distributed Discrete-Time Linear Kalman Filtering

By Dušan Krokavec and Anna Filasová

Submitted: March 31st 2018Reviewed: July 13th 2018Published: November 5th 2018

DOI: 10.5772/intechopen.80296

Downloaded: 235

Abstract

The chapter is concerned with the application of distributed discrete-time linear Kalman filtering with decentralized structure of sensors in fault residual generation. Two variants of distributed Kalman filtering algorithms are introduced, giving the incidence of equivalent functional realization structure of fault residual filters. The obtained solutions use Kalman filter innovations in a nonstandard way to generate residuals with significantly higher dynamic signal range. The obtained results, offering structures for fault detection filter realization, are illustrated with a numerical example to note the effectiveness of the approach.

Keywords

  • linear noisy systems
  • Kalman filtering
  • innovation sequences
  • fault residual filters
  • distributed computing

1. Introduction

The castigatory principal aspect for designing a fault-tolerant control (FTC) structure is a functionality of diagnostic operations that solve the fault detection and isolation (FDI) tasks. These techniques most commonly use residuals generated by fault detection filters (FDF), followed by the residual signal evaluation within decision functions. Guarantying adequate sensitivity to faults, the accessory objective is to create residuals with minimal sensitivity to noises. Kalman filtering is an optimal state estimation process applied to a dynamic system that involves random noises, giving a linear, unbiased, and minimum error variance recursive algorithm to optimally estimate the unknown state of a dynamic system from noisy data taken from sensors [1, 2].

Practically, a bank of Kalman filters is used to achieve sensor and actuator fault detection applied to a steady-state system, while the statistical characteristics of the system are not required to be known after a fault has occurred [3, 4]. In these methods, the faults are assumed to be known, and the Kalman filters are designed for such kind of sensor or actuator faults. Another approach based on Kalman filtering is the analysis of the innovation sequence, since faults displace its zero mean and change its covariance matrix [5]. The associated problem is quick detection of changes in these parameters from their nominal values. Evidently, research in Kalman filter based-FDI is the subject of wide range of other publications (see, e.g., [6, 7, 8, 9] and the reference therein). Other applications can be found in [10].

The state estimation obtained by the Kalman filter prediction-correction equations at every time instant can be solved almost optimally and substantially faster by applying a distributed approach [11, 12, 13, 14]. With this setup can be exploited the fact that the correction error can be decaying exponentially with time instant sequence to reach the optimal values [15, 16, 17, 18].

The chapter exploits a variant of distributed methods to apply the distributed correction stage filtering equations on each sensor level as well as an approach based on quasi-parallel central computation. Benefiting from the distributed Kalman filtering algorithm, two residually equivalent signal structures are presented for the discrete-time linear noisy systems.

The outline of this chapter is as follows: Section 1 delineates the problem and draws the basic starting points of solutions. Dealing with the discrete-time noisy systems description, the equations describing Kalman filters for uncorrelated measurement and system noises are traced out in Section 2, to delineate distributed approaches in Kalman filter design, suitable for supporting the fault residual generation, presented in Section 3. Section 4 gives a numerical example, illustrating the properties of the proposed method, and Section 5 presents some concluding remarks.

Throughout the chapter, the notations are narrowly standard in such a way that xTand XTdenote the transpose of vector xand matrix X, respectively, and diagdenotes a block diagonal matrix—for a square matrix X>0means that Xis a symmetric positive definite matrix. The symbol Inindicates the nth order unit matrix; IRdenotes the set of real numbers; IRnand IRn×rrefer to the set of all n-dimensional real vectors and n×rreal matrices, respectively; and Z+is the set of all positive integers.

2. Discrete-time linear Kalman filter

In this section, one version of the Kalman filtering concept is applied for the discrete-time linear multi inputs and multi outputs (MIMO) plants with the system and output noises of the form

qi+1=Fqi+Gui+vi,E1
yi=Cqi+oi,E2

where qiIRn, uiIRr, and yiIRmare vectors of the system state, input and measurement output variables, respectively; viIRnand oiIRmare vectors of the system and measurement noise; and FIRn×n, GIRn×r, and CIRm×nare conditioned by 1m,rn. Kalman filter is used only for diagnostic purposes. Zero-mean Gaussian white noise processes are considered such that

Evioi=00,E3
EvioivTkoTk=QSSTRδik,E4

where Eis the a statistical averaging operator,

δik=1ifi=k,0ifik,E5

is the Kronecker delta-function and the covariance matrices QIRn×nand RIRm×mare symmetric positive definite matrices.

It is assumed that the deterministic system initial state q0=q0is independent of viand oiin the sense that

Eq0vTi=0,Eq0oTi=0foralliE6

and that the system and measurement noises are uncorrelated, i.e., S=0.

Determining the optimal system state vector estimate, qeii1denotes the predicted estimation of the system state vector qiat the time instant iin the dependency on all noisy output measurement vector sequence yjj=01i1up to time instant i1; qeiiis the corrected estimation of the system state vector qiat the time instant iin the dependency on all noisy output measurement sequence yjj=01iup to time instant i; and eii1=qiqeii1and eii=qiqeiiare prediction and correction errors.

Definition 1. [19] If the Kalman filter, associated with the plant (1), (2) with uncorrelated system and measurement noises, is defined by the set of equations

qeii1=Fqei1i1+Gui1,E7
qeii=qeii1+Jiyiyeii1,E8
yeii1=Cqeii1,E9
yeii=Cqeii,E10

then with qe00=q0, P00=Q, where QIRn×nis a positive definite matrix, yielding

Ji=Pii1CTR+CPii1CT1,E11
Pii1=FPi1i1FT+Q,E12
Pii=IJiCPii1,E13

where

Pii1=Eeii1eTii1,E14
Pii=EeiieTii,E15

are the covariance matrices of prediction and correction errors and JiIRn×mis the Kalman filter gain matrix, all at time instant i.

The discrete-time Kalman filter equations can be algebraically manipulated into a variety of forms [6, 16, 20]. From the point of view of distributed filtration, it is necessary to achieve such form of the equation for calculating the Kalman gain Jithat yields the matrix Cfrom the matrix inversion operation (see (11)). If the system and measurement noises are uncorrelated, then for the Kalman filter gain, one can propose the following:

Lemma 1. If the system and measurement noises are uncorrelated, then the Kalman filter gain and the correction error covariance matrix can be computed using (12) and

Ji=PiiCTR1,E16
P1ii=P1ii1+CTR1C.E17

Proof. Substituting (11) into (13), one can obtain that

Pii=Pii1Pii1CTR+CPii1CT1CPii1.E18

Exploiting the Sherman-Morrison-Woodbury formula of the form [21].

A+BDBT1=A1A1BD1+BTA1B1BTA1,E19

where square invertible matrices A, D, and a matrix Bof appropriate dimensions are such that A+BDBTis invertible, with

A=Pii1,B=Pii1CT,D=R+CPii1CT1,E20

yields, since the covariance matrices are positive definite,

P1ii=P1ii1P1ii1Pii1CTE1CPii1P1ii1,E21

where

E=RCPii1CT+CPii1P1ii1Pii1CT=R.E22

Then, evidently,

P1ii=P1ii1CTE1CE23

and (31) implies (17).

Premultiplying the left side by P1iiand postmultiplying the right side by P1ii1, (13) gives

P1ii1=P1iiP1iiJiCE24

and comparing (17) and (24), it can be seen that

P1iiJiC=CTR1C.E25

Thus, (25) implies (16). This concludes the proof.

Note, since CTR1Cis at least a positive semi-definite matrix, it is evident from (17) that Piiis never larger than Pii1. Moreover, the result is an unbiased filter with the estimates of minimum error variances. More details can be found in [12, 22].

Corollary 1. Considering that qeii1is known and qeiiis the best estimate of qithat minimizes the cost criterion

Ti=qiqeii1TP1ii1qiqeii1+yiCqiTR1yiCqi.E26

Then, evaluating (26) it follows, with the optimal setting of a state vector estimate qi=qii, that the minimum expected cost is given by

dTidqiT=P1ii1qiiqeii1CTR1yiCqii=0,E27

which implies

P1ii1+CTR1Cqeii=P1ii1qeii1+CTR1yi=P1ii1+CTR1Cqeii1+CTR1yiCqeii1.E28

Therefore, using the above relations, at the ith step Eq. (28) gives

qeii=qeii1+P1ii1+CTR1C1CTR1yiCqeii1=qeii1+Pii×CTR1yiCqeii1.E29

Pre-multiplying the left side by Piiand post-multiplying the right side by Pii1then it follows from (17)

Pii1=Pii+PiiCTR1CPii1,E30

which can be proved recursively as follows

Pii=InPiiCTR1CPii1.E31

Comparing (29) with the covariance matrix of the filtering error given by (13), it is evident that

Ji=PiiCTR1E32

which is identical to (16).

On the other side, substituting (11) into (13), one can write

Pii=Pii1Pii1CTR+CPii1CT1CPii1E33

and using the Sherman-Morrison-Woodbury formula, Eq. (27), it follows

P1ii=P1ii1CTRCPii1CT+CPii1CT1CE34

and so, evidently, (34) gives (17).

3. Fault residual generation using distributed Kalman filtering

The obtained equations, Eqs. (16) and (17), allow the use of the open form of the Kalman filter equations if

R=EoioTi=diagR1R2Rm.E35

Writing separately,

yTi=y1iy2iymi,E36
uTi=u1iu2iuri,E37
CT=c1c2cm,G=g1g2gr,E38

then, (7)(11), (16), and (27) imply

qeii1=Fqei1i1+h=1wghuhi1,E39
qeii=qeii1+h=1mjhiyhichTqeii1,E40
yejii1=chTqeii1,E41
yejii=chTqeii,E42
jhi=PiichRh1,E43
P1ii=P1ii1+h=1mchRh1chT,E44
Pii1=FPi1i1FT+Q.E45

It is evident from the above given formulation that the relation of (40) gives the possibility to compute corrections from the data obtained at all sensor nodes.

Theorem 1. Defining the residual vector as

rTi=z1iz2izmi,E46

where

zhi=yhichTqecii1,E47

then

qecii1=Fqeci1i1+h=1rghuhi1,E48
qedii1=Fqedi1i1,E49
qecii=qecii1,E50
qedii=qedii1+Piih=1mchRh1zhichTqedii1,E51

while the filter gain matrices, as well as recurrences of the covariance matrices are given by (43)–(45).

Proof. Considering that there are components of the system state vector estimate that are dependent on the control signal as well as ones that are not dependent on the control signals, since the correction step does not depend on the control inputs, (40) can be rewritten as

qeii=qecii1+qedii1+h=1mjhiyhichTqecii1+qedii1.E52

Prescribing that

qeii=qedii+qecii,E53

Eqs. (52) and (53) can be separated as

qecii=qecii1,E54
qedii=qedii1+h=1mjhiyhichTqecii1+qedii1E55

and using (47), (54) gives (50), and (55) implies (51).

Substituting (53) in (39) yields

qeii1=Fqedi1i1+qeci1i1+h=1wghuhi1E56

and, evidently, (56) implies (48) and (49). This concludes the proof. □

Remark 1. If Eqs. (46)–(51) are analyzed from a computational point of view, it is clear that their structures support autonomous parallel calculations only with a single interaction defined by Eq. (47). However, the cost for this parallelism is additional computation at each step, but the directional properties of the components of the residual vector are advantageous in the case of single sensor faults. The directional sensor residual property derives indirectly from relationship (44). Since every component zhicarries with it the measurement noise ohiif qedii1is used for LQG control, it will be no noise at the state control law input.

In principle, it is possible to define the residue generation by results of the local system state correction at Kalman filtration at the time instant i.

Theorem 2. Defining the residual vector as

rTi=z1iz2izmi,E57

where

zhi=yhichTqecii1,E58

then

qedjii1=Fqedji1i1,E59
qedjii=qedjii1+jhizhichTqedjii1,E60
jhi=PhiichRh1,E61
Ph1ii=P1ii1+chRh1chT,E62

while the predicted system state at the time instant i is computed centrally and the filtered full system state is covered by the equations

qedii=h=1mPiiPh1iiqedjiih=1mPiiP1ii1Fqedji1i1))+PiiP1ii1Fqedi1i1,E63
qecii=qecii1E64

at the time instants iZ+.

Proof. The correction step for the Kalman filter in Eq. (51) can be prescribed locally for the jth node as

qedjii=qedjii1+jhizhizdhii1,E65

where

zhi=yhichTqecii1,E66
zdhii1=chTqedjii1,E67
jhi=PhiichRh1,E68
Ph1ii=P1ii1+chRh1chT.E69

Substituting (66), rearranging and postmultiplying the left side by Ph1ii, (65) implies

Ph1iiqedjiiqedjii1=chRh1zhizdhii1,E70
chRh1zhi=chRh1chTqedjii1+Ph1iiqedjiiqedjii1,E71

respectively. Since (69) gives

chRh1chT=Ph1iiP1ii1,E72

with a simple elimination after inserting (72), (71) gives

chRh1zhi=Ph1iiqedjiiqedjii1+Ph1iiqedjii1P1ii1qedjii1=Ph1iiqedjiiP1ii1qedjii1.E73

Combining (49) and (51) results in

qedii=Fqedi1i1+Piih=1mchRh1(zhichTFqedi1i1,E74

which can be written as

qedii=h=1mPiichRh1zhi+Inh=1nPiichRh1chTFqedi1i1.E75

Pre-multiplying the left side of (44) by Piileads to

Inh=1wPiichRh1chT=PiiP1ii1E76

and considering (76), relation (75) takes the form

qedii=h=1mPiichRh1zhi+PiiP1ii1Fqedi1i1.E77

Thus, the substitution of (73) into (77) gives

qedii=h=1mPiiPh1iiqedjiih=1mPiiP1ii1qedjii1+PiiP1ii1Fqedi1i1E78

and with the notation

qedjii1=Fqedji1i1,E79

(78) implies (62). This concludes the proof.

Remark 2. It is clear that each of Eqs. (59)–(62) is only bound to the jth node and therefore such correction can be done locally for each sensor. Conversely, the system state prediction and the residual vector must be computed globally by Eqs. (39), (53), (57), (58), (63), and (64), respectively.

Remark 3. Obviously, under the above conditions, the distributed realization of the Kalman filter correction step is optimal in the sense of criterion (26), and therefore the structure of the fault residual generator based on distributed Kalman filtration is optimal.

4. Illustrative examples

4.1. Example 1

To eliminate specific system dependencies, the Schur discrete-time linear strictly positive system [23] is used for demonstration of the Kalman filtering technique in residual signals construction. The considered system can be put in the system class (1)(4), with the sampling period ts=0.8s, with uncorrelated system and measurement Gaussian noise and the noise covariance matrices

R=diag0.0030.04,Q=0.002I4

while the system matrix parameters are

F=0.76500.62670.60580.05100.10480.10830.08130.00980.14840.14190.11710.01500.17090.22860.16030.1998,G=0.02410.01390.01510.00130.01090.00560.01420.0032,C=0.00010100.0000001.

Since the discrete-time stochastic linear strictly positive system is stable, the system control law in simulations is defined for the forced mode control as ui=Wwo, where

W=117.384179.3124280.8078187.1829,wo=0.60.8

and the initial conditions for the Kalman filter are

qe00=0,P00=I4.

Using the given initial conditions, Figures 1 and 2 display the residuals obtained by the residual filter generated by the distributed Kalman filter defined in (46)(51), reflecting single actuator and sensor faults, starting at the time instant t=50s. The time scale is discrete with the sampling period T=0.8s.

Figure 1.

Residual responses to single faults: (a) the first actuator and (b) the second actuator.

Figure 2.

Residual responses to single faults: (a) the first sensor and (b) the second sensor.

Evidently, the residual trajectories indicate that the proposed residual filter generates directional signals in the event of single sensor faults, and has a significantly higher dynamic signal range in the event of single faults of the actuators, as compared to the residual presented using the standard Kalman filter.

4.2. Example 2

To produce another example that demonstrates achievable performances of the presented design method, the sign-indefinite Schur discrete-time linear system is used, where [24]

F=1.10390.23600.05630.02290.10630.79710.05750.01090.01000.02110.94010.04760.05990.08430.01110.9633,G=0.19570.28780.09760.19210.09690.09390.00120.0982,C=00010100,

ts=0.05s, and the Gaussian noise covariance matrices are R=diag0.0030.04and Q=0.002I4. To force the desired system output values, it is prescribed

W=2.12500.93751.87500.5625,wo=0.60.8,qe00=0,P00=I4.

Figures 3 and 4 present the residual responses of the residual filter based on distributed Kalman filtering, from which it is clear that the used principle, especially when compared to the achievable responses with the alternative system presented in Example 1, is operational. Evidently, the residual filter behavior is also acceptable for the system parameters in this example and the system noise environment. The step-like single faults start and continue from the time instant t=25s, the time scale is discrete with the sampling period T=0.05s.

Figure 3.

Residual responses to single faults: (a) the first actuator and (b) the second actuator.

Figure 4.

Residual responses to single faults: (a) the first sensor and (b) the second sensor.

4.3. Example 3

Following the above-mentioned procedures, Example 3 verifies their effectiveness for the linear discrete-time system with the parameters

F=0.10.20.30.40.7000.10.20.80.20.3000.50.2,g=0.100.30,gf=000.30,cT=1001,cfT=0001,

where Fis a left-stochastic matrix [25], ts=0.05s, and the Gaussian noise covariance matrices are R=0.003and Q=0.002I4. The behavior of the system is changed by the state-feedback control

ui=kTqi+Wwo

where

kT=0.29821.07310.37110.6412,W=1.1834,wo=0.6,

which provides Schur matrix Fcas follows

Fc=FgkT=0.07020.09270.26290.33590.7000000.10000.11050.47810.08870.1076000.50000.2000

Note, the steady states of Fcare absorbing states.

The single fault effects in residuals, when using the proposed algorithm of distributed Kalman filtering (46)(51) with setting F=Fc, qe00=0and P00=I4are shown in Figure 5. The time scale is discrete with the time sample interval T = 0.1s.

Figure 5.

Residual responses to single faults: (a) the actuator and (b) the sensor.

From the figures, we find that the fault responses are satisfactory by using the proposed method also for this system and noise environment.

Analyzing all examples, exactly the same responses are reached using the same parameters as before and assuming the same fault patterns if the residuals are evaluated exploiting formulas (57)(64) or (46)(51). It is given by the equivalent principles of distributed computing that are bound by the equivalent relationships (44) and (62), respectively. As a result, in this particular point of view, the proposed distributed algorithm has only one common matrix component, P1ii1, which has to be transmitted to every separated sensor before carrying out the state correction filtering step at every time instant. Since the correction step at time instant iis done in dependency on the measured value at the same time instant yi, it is clear that the shorter the computation at the correction step, the smaller the time-delay introduced into the fault detection system responses.

5. Concluding remarks

Realization forms for fault detection residual structures, based on distributed Kalman filtering destined for noisy discrete-time linear systems, were derived in this chapter. The main idea deals with introducing a distributed sensor measurement noise corrector step of a Kalman filter, applied in such a way to be locally uncorrelated with other sensor measurements. Two different algorithmic supports, a parallel decentralized Kalman filter and a locally distributed Kalman filter, are constructed to generate fault residuals. Both solutions are discussed in detail to demonstrate the condition of their equivalency. The problem accomplishes the manipulation in the manner giving guaranty of asymptotic stability of a local fault residual detection filter. Simulated example is included to illustrate the applicability of the proposed methods, encouraging the results that are obtained. Note, since the Kalman filter is based on the nominal system parameters G and C, it cannot estimate system states and outputs starting for faulty regimes with modified matrices Gf and Cf, respectively.

From the point of cloud-based distributed systems, to combine appropriately the network and computational resources, a locally distributed Kalman filter seems to be naturally adaptable, also with cross-correlated sensor noises. Of course, no theoretical justification for this affirmation is presented in the chapter. This is seen as an area for future research by the authors.

Acknowledgments

The work presented in this chapter was supported by VEGA, the Grant Agency of the Ministry of Education and the Academy of Science of Slovak Republic, under Grant No. 1/0608/17. This support is very gratefully acknowledged.

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite and reference

Link to this chapter Copy to clipboard

Cite this chapter Copy to clipboard

Dušan Krokavec and Anna Filasová (November 5th 2018). Fault Residuals Based on Distributed Discrete-Time Linear Kalman Filtering, Fault Detection and Diagnosis, Constantin Volosencu, IntechOpen, DOI: 10.5772/intechopen.80296. Available from:

chapter statistics

235total chapter downloads

More statistics for editors and authors

Login to your personal dashboard for more detailed statistics on your publications.

Access personal reporting

Related Content

This Book

Next chapter

Fault Diagnosis of Hybrid Computing Systems Using Chaotic-Map Method

By Nageswara S. V. Rao and Bobby Philip

Related Book

First chapter

Microassembly Using Water Drop

By Taksehi Mizuno

We are IntechOpen, the world's leading publisher of Open Access books. Built by scientists, for scientists. Our readership spans scientists, professors, researchers, librarians, and students, as well as business professionals. We share our knowledge and peer-reveiwed research papers with libraries, scientific and engineering societies, and also work with corporate R&D departments and government entities.

More About Us