Engine fault diagnosis based on sensor data fusion considering information quality and evidence theory

Abstract

Fault diagnosis is a problem processing variable information obtained from different sources in nature. Evidence theory, efficient to deal with information viewed as evidence, is widely used in fault diagnosis. However, a shortcoming of the existing fault diagnosis methods only gets probability distribution rather than the basic probability assignment. A novel method of generating basic probability assignment that takes information quality into account is proposed. The probability distribution is determined by the preliminary matrix and sampling matrix that are constructed by sensor data. And the quality of probability distribution is taken as the discount factor and the rest of belief is assigned to the universal set. Hence, the basic probability assignment is obtained. Then, basic probability assignment can be combined with Dempster and Shafer evidence theory to determine the status of the engine. An application of engine fault is shown to illustrate the practicability of the proposed method. Then by comparing the result of the method which takes information quality into account (the proposed method) and does not do it, the former is better than the latter. Finally, the reliability analysis shows that the proposed method has strong reliability because performance accuracy is 100% when the error rate is less than 10%.

Keywords

Demspter–Shafer evidence theory belief function data driven sensor data fusion fault diagnosis

Introduction

Fault diagnosis is important to system to correct timely and work smoothly. Up to now, fault diagnosis has been applied extensively to all kinds of profession, such as mechanics,^1–4 chemistry,⁵ nucleus,⁶ and electric.⁷ A number of approaches to optimize the algorithm of fault diagnosis are proposed, such as the average current Park’s vector approach,⁸ a fuzzy approach,^9,10 and optimized threshold de-noising method.²

In real system, a larger quantity of information can be obtained from all kinds of sensors that detect the concrete values. Consequently, it is necessary for decision-makers to get rational result by considering all the complicated information^11–15 whose certainty maybe very high or low. And some methods of information fusion¹⁶ used in fault diagnosis and other fields have been proposed, such as Kalman Filter,^17–19 neural network,^20–22 and fuzzy logic.^23–28

In addition, evidence theory^29,30 is efficient to deal with uncertainty^31–37 and Dempster’s rule can take the advantage of evidence combination from different sources without prior information. Up to now, on the basis of evidence theory, D-number,^38–40 Z-number,^41,42 and so on, have been studied by lots of scholars. As a result, evidence theory is not only used in medical diagnosis,^43–45 dependence assessment,⁴⁶ correlation,^47,48 data stream,⁴⁹ forensic crime investigations,⁵⁰ traffic,⁵¹ and target recognition^52,53 but also in fault diagnosis.^54–59 There are many researches about applying evidence theory into fault diagnosis. For example, the new combination rule⁵⁷ is then built to allocate the conflicted information from multi-sensors based on the support degree of focal element. A novel-weighted evidence combination rule⁶⁰ based on evidence distance and uncertainty measure is proposed. A novel method⁶¹ that comprehensively analyzes vibration and temperature signals to diagnose bearing faults based on improved Dempster–Shafer (D-S) evidence theory is presented. A novel dissolved gas analysis (DGA) method⁶² for power transformer incipient fault diagnosis based on integrated adaptive neuro fuzzy inference system (ANFIS) and Dempster–Shafer theory (DST) is presented. A weak thruster fault detection method⁶³ is developed based on the combination of artificial immune system and single pre-processing. The architecture of an expert system⁶⁴ that uses flame images grabbed during the combustion process in an experimental oil furnace as input parameters is presented. An effective method⁶⁵ for precise fault diagnosis of planetary gearbox based on fusion of vibration and acoustic data using the DST is proposed. A new transformer fault diagnosis method⁶⁶ based on a wavelet neural network optimized by adaptive genetic algorithm (AGA) and an improved D-S evidence theory fusion technique is proposed. In Basir and Yuan,⁶⁷ to make rational decisions, a method is proposed with respect to engine quality and to evaluate the performance of the proposed information fusion system, a criterion is presented. But there are two issues needed to be improved, one is that the mass function obtained by calculating the distance between the measured features and fault prototypes is only probability distribution for singleton and cannot be called as basic probability assignment to some extent. And the other is that information quality is not taken into consideration.

To handle the above issues, in this article, a new method that Shannon entropy^68,69 of each probability obtained from each feature is used as discount factor^70–72 to acquire the basic probability assignment is proposed. All the measurements obtained from sensors can characterize two faults: (1) X₁: exhaust valve fault; (2) X₂: piston ring fault, but not all of the measurements can distinguish fault accurately in some situation, in other words, measurements have different information quality. Therefore, the mass function can be obtained by considering the information quality and on basis of probability distribution. The main advantages of this method are that Shannon entropy as the discount factor that is assigned to the universal set can obtain basic assignment instead of probability distribution and decrease conflicts to fuse efficiently.

This article is organized as follows: in section “Preliminary,” the concepts and rule of evidence theory and notations and formulation of the Shannon entropy are introduced. In section “Proposed method,” we present the frame of discernment and the new evidence combination. In section “Application in fault diagnosis,” an application is used to illustrate efficiency and reliability of the proposed method. In section “Conclusion,” the study is briefly concluded.

Preliminary

Preliminary notion of the D-S evidence theory

Evidence theory is the classical mathematic theory of evidence which is initially based on Dempster’s work concerning lower and upper probability distribution families and expanded by Shafer. In this section, some basic concepts and functions are introduced as follows:

1. The frame of discernment: Let $Θ$ be a finite set of elements; an element can be a hypothesis or a fault in our case. We refer $Θ$ is the frame of discernment. The $Θ$ is denoted as follows^29,30

Θ = {H_{1}, H_{2}, H_{3}, \dots, H_{N}}

(1)

the set consisting of all the subsets of $Θ$ is called as power set of $Θ$ and denoted as follows

2^{Θ} = {\emptyset, {H_{1}}, {H_{2}}, \dots, {H_{1}, H_{2}}, \dots, Θ}

(2)

where $\emptyset$ is null set consisting of nothing.

2. Mass function: A key point of the frame of discernment is the basic probability assignment (BPA), which is a mapping of the power set $2^{Θ}$ to a number between 0 and 1, that is

m : 2^{Θ} \to [0, 1]

(3)

m (\emptyset) = 0, \sum_{A \subseteq Θ} m (A) = 1

(4)

where $m (A)$ represents how strongly the evidence supports A.

3. Rules of evidence combination: Assuming $m_{1}$ and $m_{2}$ are two mass functions, having the discrepancy, obtained from different information sources. The Dempster’s rules of combination, called orthogonal sum and noted by $m = m_{1} \oplus m_{2}$ , are defined as follows^29,30

m (\emptyset) = 0, m (A) = \frac{\sum_{B \cap C \subseteq A} = m_{1} (B) m_{2} (C)}{1 - K}

(5)

K = \sum_{B \cap C \subseteq \emptyset} m_{1} (B) m_{2} (C)

(6)

The $m (\emptyset)$ represents that the evidence gives it little support in the dominance of “closed-world.”K is a coefficient to reflect the conflict between $m_{1}$ and $m_{2}$ , determined by summing the products of the mass function of all sets where the intersection is null.

Shannon entropy

The idea of entropy, of which Shannon entropy is accepted by most people, is an important concept to the probability distribution on the space $X = {x_{1}, x_{2}, \dots, x_{n}}$ . Then the Shannon entropy⁶⁸ is

H (P) = - \sum_{i = 1}^{n} p_{i} \ln (p_{i})

(7)

where $p_{i}$ represents the probability of ith element in vector. Obviously, the entropy will have maximum value $\ln (n)$ when it is uniform distribution, $p_{i} = 1 / n$ for $i = 1, 2, \dots, n$ and minimum value 0 when one $p_{i} = 1$ and all other $p_{i} = 0$ . Hence, intuitively, the larger is $H (P)$ , the less is the information provided by the probability. Conversely, the smaller is $H (P)$ , the more is the information. For example, we cannot decide which elements are best when $p = {0.5, 0.5}$ but we can do it when $p = {1, 0}$ .

Proposed method

In this article, the proposed method is in Figure 1 shown and detailed as follows:

Figure 1.

The flowchart of the proposed method.

Faults in the frame of discernment

To solve fault diagnosis problem by using evidence theory, the frame of discernment is necessary. Each element in the frame of discernment represents relevant fault that machine has when working. For example, we take $X = {X_{0}, X_{1}}$ where $X_{0}$ intimates the absence of all kinds of faults and $X_{1}$ suggests the presence of a fault as the frame of discernment. Thus, N faults are taken account of and the frame of discernment is established as follows

X = {X_{0}, X_{1}, X_{2}, \dots, X_{N}}

where $X_{N}$ represents the presence of ith fault.

In fault diagnosis, possessing information is the prerequisite of all work. Accordingly, various sensors, such as the vibration sensor and the acoustic sensor, are used to detect the corresponding characteristic values that describe some features to determine the status of the engine and decide which fault takes place. Besides information obtained from sensors when the engine is working, the relevant parameter of the specific engine is also significant, or decision made cannot make a comparison. The specific feature value corresponding to the engine being in specific fault is necessary to be obtained. Consequently, a preliminary feature matrix of $(N + 1) \times M$ is obtained as follows

H = [\begin{matrix} h_{01} & h_{02} & \dots & h_{0 M} \\ h_{11} & h_{12} & \dots & h_{1 M} \\ ⋮ & ⋮ & ⋱ & ⋮ \\ h_{N 1} & h_{N 2} & \dots & h_{NM} \end{matrix}]

(8)

where N represents the number of faults taking place in the engine and M represents all feature values that the sensor can obtain.

After preliminary feature matrix is established, the values obtained from all sensors when the engine is working, it is used to construct sampling matrix compared by preliminary feature matrix. Let $S_{k}$ represent the measurement vector obtained from the kth sampling from all sensors

S_{k} = {S_{k 1}, S_{k 2}, \dots, S_{kM}}

where k represents the kth sampling and $j = 1, 2, 3, \dots, M$ represents jth feature value. Then the sampling matrix composed of all the measurement vectors is established.

A key problem is that which method we choose to be used to calculate the basic probability assignment. Obviously, if the more similar the element of measurement vector S _k we get from sensors is to the corresponding row vector {h_j1h_j2 . . . h_jM} of the preliminary matrix, the more probable the corresponding fault X_j occurs. Inversely, the less similar it is, the less probable it does. There are many measures to quantify the distance between the measured feature obtained from sensors and the relevant parameter value. The absolute distance measure is used in this article as follows

d_{kji} = | S_{kj} - h_{ij} |

(9)

where k is the kth sample, i is the ith fault, and j is the jth feature that sensors detect.

The distances between all sensor measurements and the relevant parameter values can be captured in a matrix form⁷³

D = [\begin{matrix} (d_{110}, d_{111}, \dots, d_{11 N}) & (d_{120}, d_{121}, \dots, d_{12 N}) & \dots & (d_{1 M 0}, d_{1 M 1}, \dots, d_{1 MN}) \\ (d_{210}, d_{211}, \dots, d_{21 N}) & (d_{220}, d_{221}, \dots, d_{22 N}) & \dots & (d_{2 M 0}, d_{2 M 1}, \dots, d_{2 MN}) \\ ⋮ & ⋮ & ⋱ & ⋮ \\ (d_{L 10}, d_{L 11}, \dots, d_{L 1 N}) & (d_{L 20}, d_{L 21}, \dots, d_{L 2 N}) & \dots & (d_{LM 0}, d_{LM 1}, \dots, d_{LMN}) \end{matrix}]

(10)

Each third dimension in the matrix represents the distance between measurements obtained from sensors and all fault including non-fault. The smaller the distance $d_{kji}$ is, the more probable the ith fault is. Therefore, defining $p_{kji}$ as

p_{kji} = \frac{1}{d_{kji}}

(11)

After normalization, a probability matrix of $L \times M \times (N + 1)$ is obtained as follows

P = [\begin{matrix} (p_{110}, p_{111}, \dots, p_{11 N}) & (p_{120}, p_{121}, \dots, p_{12 N}) & \dots & (p_{1 M 0}, p_{1 M 1}, \dots, p_{1 MN}) \\ (p_{210}, p_{211}, \dots, p_{21 N}) & (p_{220}, p_{221}, \dots, p_{22 N}) & \dots & (p_{2 M 0}, p_{2 M 1}, \dots, p_{2 MN}) \\ ⋮ & ⋮ & ⋱ & ⋮ \\ (p_{L 10}, p_{L 11}, \dots, p_{L 1 N}) & (p_{L 20}, p_{L 21}, \dots, p_{L 2 N}) & \dots & (p_{LM 0}, p_{LM 1}, \dots, p_{LMN}) \end{matrix}]

(12)

\sum_{i = 0}^{N} p_{kji} = 1

Accordingly, the probability is obtained from the above matrix. Each vector obtained from the measurement and preliminary parameter in the matrix can be used as the probability because each feature can detect the fault. According to the preceding analysis, each fault is determined by many features. However, some features do not correctly distinguish the types of faults in some situation. For example, the feature value is equal to 10 when fault $X_{1}$ occurs, while it is also equal to 10 when fault $X_{2}$ takes place. Nevertheless, the other feature values are different once the faults are not same. It is important to consider the information quality of each probability varying a lot. Therefore, the Shannon entropy $H (P) = - \sum_{i = 1}^{n} p_{i} \ln (p_{i})$ introduced in section “Shannon entropy,” representing information quality, is taken as the discounting factor. Consequently, the basic probability assignment can be obtained by considering information quality and on the basis of the probability distribution. Define $ω_{kj} = H (P)$ as the discount factor

Ω = [\begin{matrix} ω_{11} & ω_{12} & \dots & ω_{1 M} \\ ω_{21} & ω_{22} & \dots & ω_{2 M} \\ ⋮ & ⋮ & ⋱ & ⋮ \\ ω_{L 1} & ω_{L 2} & \dots & ω_{LM} \end{matrix}]

(13)

To make $ω_{kj} \in [0, 1]$ , variance for all $ω_{kj}$ is used to calculate

ω_{kj} = 1 - \frac{ω_{kj}}{\max {ω_{kj}} + variance}

(14)

Then, for $p_{kj} = {p_{kj 0}, p_{kj 1}, \dots, p_{kjN}}$ corresponding to non-fault, 1st fault, …, Nth fault, define $m_{kj} = w_{kj} \times p_{kj}$ as the basic probability assignment. Let $m_{kj} (X) = 1 - \sum_{i = 0}^{N} m_{kj}$ , hence, uncertainty is assigned to the universal set to decrease the conflict. Therefore, the discounting basic probability assignment is obtained as follows

m_{kj} = {m_{kj 1}, m_{kj 2}, \dots, m_{kjN}, m_{kj} (X)}

(15)

Evidence combination in fault diagnosis

There are many rules proposed to fuse information which are based in D-S evidence theory. Although these rules^74,75 measure and decrease conflict or make result be rational intuitively, they also have some drawbacks, increasing computation complexity or lacking associative property. Due to the above basic probability assignments having fewer conflict when information quality is taken consideration, the classic Dempster’s rule, having fewer computation complexity, is the best choice for us to fuse information. Consequently, according to equations (5) and (6), a mass function can be obtained as follows

\begin{matrix} m ({X_{0}}) = p_{0}, m ({X_{1}}) = p_{1}, \dots, m ({X_{N}}) \\ = p_{N}, m (X) = p_{N + 1} \end{matrix}

(16)

The decision is made on basis of the mass function. The rule decision-makers use is maximum support rule, in which the hypothesis with maximum belief function is chosen to represent the state of the engine. Intuitively, if $m ({X_{0}})$ is greater than the others obviously, the engine is free of any fault and if $m ({X_{1}})$ has the largest value, the fault $X_{1}$ occurs.

In addition, a situation where an unknown fault may occur when the engine is working is qualitatively analyzed. If this fault occurs, because we have not added this fault in the frame of discernment, the quality of information obtained by the sensors is reduced. Then, based on $m_{kj} (X) = 1 - \sum_{i = 0}^{N} m_{kj}, m_{kj} (X)$ , increases. Therefore, after fusion, $m (X)$ increases. Because an unknown fault leads to an increase in uncertainty which is assigned to the universal set, we could infer from the value of $m (X)$ whether there is an unknown fault.

Application in fault diagnosis

A practical application of the proposed method

In this section, an application for detecting the state of the engine^76,77 is given to illustrate the preceding proposed fault diagnosis method. Three sensors (two acceleration sensors and one acoustic sensor) are used to detect the state of the engine. One acceleration sensor is mounted on the cylinder cover near the outlet valve. Another is located on the cylinder cover near the inlet valve. Their peak-to-peak value (P-to-P) in the time domain and the frequency of the maximum spectrum $(F_{\max})$ are calculated as four features represented as $α_{1}, α_{2}, α_{3}, and α_{4}$ . In concern of the acoustic signal, its mean pressure level (MPL) and the centrobaric frequency of the spectrum $(F_{c})$ are calculated as another two features, represented as $α_{5}$ and $α_{6}$ . Because only one sensor may be unreliable in some situation, Three sensors are chosen together to get a rational result. Two faults $(X_{1} : exhaust valve fault; X_{2} : piston ring fault)$ are defined when engine is working and the free of fault is also taken into consideration when the engine works smoothly. Therefore, the frame of discernment is defined as follows

X = {X_{0}, X_{1}, X_{2}}

(17)

Then, the power set of X is established as follows

\begin{matrix} 2^{X} = {\emptyset, {X_{0}}, {X_{1}}, {X_{2}}, {X_{0}, X_{1}}, \\ {X_{0}, X_{2}}, {X_{1}, X_{2}}, X} \end{matrix}

(18)

Six features are taken into consideration for each state of the engine and their characteristic values are obtained as follows

H = \begin{matrix} \begin{matrix} α_{1} & α_{2} & α_{3} & α_{4} & α_{5} & α_{6} \end{matrix} \\ \begin{matrix} X_{0} \\ X_{1} \\ X_{2} \end{matrix} & (\begin{matrix} h_{01} & h_{02} & h_{03} & h_{04} & h_{05} & h_{06} \\ h_{11} & h_{12} & h_{13} & h_{14} & h_{15} & h_{16} \\ h_{21} & h_{22} & h_{23} & h_{24} & h_{25} & h_{26} \end{matrix}) \end{matrix}

(19)

The values of H that are from Basir and Yuan’s⁶⁷ research are shown in Table 1.

Table 1.

Features according to previous knowledge.

	$α_{1}$	$α_{2}$	$α_{3}$	$α_{4}$	$α_{5}$	$α_{6}$
${X_{0}}$	313.5	559.6	378.6	557.4	152.9	762.7
${X_{1}}$	1850.7	550.8	1734.5	597.2	152.3	808.2
${X_{2}}$	2669.3	546.6	2567.4	534.8	152.7	724.1

Then, the characteristic values of the features are collected and a feature matrix is constructed by data obtained from sensors after four samplings as follows

S = \begin{matrix} \begin{matrix} α_{1} & α_{2} & α_{3} & α_{4} & α_{5} & α_{6} \end{matrix} \\ \begin{matrix} S_{0} \\ S_{1} \\ S_{2} \\ S_{3} \end{matrix} & (\begin{matrix} S_{01} & S_{02} & S_{03} & S_{04} & S_{05} & S_{06} \\ S_{11} & S_{12} & S_{13} & S_{14} & S_{15} & S_{16} \\ S_{21} & S_{22} & S_{23} & S_{24} & S_{25} & S_{26} \\ S_{31} & S_{32} & S_{33} & S_{34} & S_{35} & S_{36} \end{matrix}) \end{matrix}

(20)

The values of S that are from Song and Jiang’s⁷⁸ research are shown in Table 2.

Table 2.

Features according to samplings.

	$α_{1}$	$α_{2}$	$α_{3}$	$α_{4}$	$α_{5}$	$α_{6}$
$S_{0}$	1830.6	553.9	1780.5	600.2	152.5	780.3
$S_{1}$	1883.5	549.9	1702.4	590.0	151.9	813.6
$S_{2}$	1854.0	551.7	1738.1	595.4	152.1	797.5
$S_{3}$	1882.2	555.2	1757.3	575.5	152.5	802.4

According to equations (9)–(12), the probability matrix can be established as follows

P = [\begin{matrix} (p_{110}, p_{111}, p_{112}) & (p_{120}, p_{121}, p_{122}) & \dots & (p_{160}, p_{161}, p_{162}) \\ (p_{210}, p_{211}, p_{212}) & (p_{220}, p_{221}, p_{222}) & \dots & (p_{260}, p_{261}, p_{262}) \\ (p_{310}, p_{311}, p_{312}) & (p_{320}, p_{321}, p_{322}) & \dots & (p_{360}, p_{361}, p_{362}) \\ (p_{410}, p_{411}, p_{412}) & (p_{420}, p_{421}, p_{422}) & \dots & (p_{460}, p_{461}, p_{462}) \end{matrix}]

(21)

The result of calculation is shown in Table 3.

Table 3.

Probability distribution.

	$α_{1}$	$α_{2}$	$α_{3}$
$S_{0}$	(0.0128, 0.9641, 0.0231)	(0.2763, 0.5080, 0.2157)	(0.0301, 0.9164, 0.0536)
$S_{1}$	(0.0197, 0.9411, 0.0393)	(0.0679, 0.7323, 0.1997)	(0.0228, 0.9422, 0.0350)
$S_{2}$	(0.0021, 0.9938, 0.0040)	(0.0883, 0.7750, 0.1368)	(0.0026, 0.9931, 0.0043)
$S_{3}$	(0.0189, 0.9433, 0.0378)	(0.3981, 0.3981, 0.2037)	(0.0158, 0.9572, 0.0269)
	$α_{4}$	$α_{5}$	$α_{6}$
$S_{0}$	(0.0628, 0.8961, 0.0411)	(0.2000, 0.4000, 0.4000)	(0.5144, 0.3245, 0.1611)
$S_{1}$	(0.1634, 0.7400, 0.0965)	(0.2105, 0.5263, 0.2632)	(0.0910, 0.8573, 0.0517)
$S_{2}$	(0.0440, 0.9284, 0.0276)	(0.1579, 0.6316, 0.2105)	(0.2116, 0.6881, 0.1003)
$S_{3}$	(0.4388, 0.3660, 0.1952)	(0.2000, 0.4000, 0.4000)	(0.1197, 0.8196, 0.0607)

All the measurements obtained from sensors can characterize two faults, but not all of the measurements can distinguish fault accurately in some situation, in other words, measurements have different information quality. The discount factor obtained from the Shannon entropy, representing information quality, is used to construct basic probability assignments. Therefore, $Ω$ , namely the discount factor, can be obtained by equations (7) and (14) as follows

Ω = \begin{matrix} \begin{matrix} α_{1} & α_{2} & α_{3} & α_{4} & α_{5} & α_{6} \end{matrix} \\ \begin{matrix} S_{1} \\ S_{2} \\ S_{3} \\ S_{4} \end{matrix} & (\begin{matrix} ω_{11} & ω_{12} & ω_{13} & ω_{14} & ω_{15} & ω_{16} \\ ω_{21} & ω_{22} & ω_{13} & ω_{14} & ω_{15} & ω_{26} \\ ω_{31} & ω_{32} & ω_{13} & ω_{14} & ω_{15} & ω_{36} \\ ω_{41} & ω_{42} & ω_{13} & ω_{14} & ω_{15} & ω_{46} \end{matrix}) \end{matrix}

(22)

The value of $Ω$ is shown in Table 4.

Table 4.

The discount factor.

	$α_{1}$	$α_{2}$	$α_{3}$	$α_{4}$	$α_{5}$	$α_{6}$
$S_{0}$	0.4259	0.1136	0.2784	0.2466	0.1112	0.1165
$S_{1}$	0.3354	0.1527	0.3370	0.1506	0.1149	0.2078
$S_{2}$	0.7612	0.1618	0.7415	0.3019	0.1267	0.1392
$S_{3}$	0.3421	0.1110	0.3919	0.1119	0.1112	0.1835

After the basic probability assignment matrix constructed by considering information quality is obtained, the Dempster’s rule of evidence combination is used to get mass function in this article. According to equations (5) and (6), we can get the final result as follows

\begin{matrix} m {X_{0}} = 0.000566 \\ m {X_{1}} = 0.998134 \\ m {X_{2}} = 0.000544 \\ m {X} = 0.000756 \end{matrix}

To show the difference of mass function intuitively, the histogram is shown as follows:

The rule decision-makers use is maximum support rule, in which the hypothesis with maximum belief function is chosen to represent the state of the engine. Consequently, from Figure 2, the basic probability assignment (0.9981) of $X_{1}$ of the proposed method is obviously larger than the others. It can prove that the fault is $X_{1}$ , namely, exhaust valve fault.

Figure 2.

The values of mass function for the methods which take information quality into account and do not do it.

In Figure 2, both methods get effective results that the fault is $X_{1}$ . By comparing the result of the method which takes information quality into account (the proposed method) and does not do it, the basic probability assignment of $X_{1}$ of the former is significantly larger than that of the latter. In this article, the proposed method not only includes multiple subsets but also considers information quality. Accordingly, the proposed method is illustrated as practicability.

Reliability analysis and comparison with the typical method

In this section, the reliability of the proposed method is analyzed by the procedure of simulating random errors in actual measurement.⁶⁷ And maximum support rule is used as a decisive standard in this analysis.

When the sensors perform measurement, there is a certain error for data obtained from sensors due to its own measurement error. Therefore, on this basis, we would give a random error rate no more than a certain percent (5%, 10%, 15%, and 20%) to the data obtained from Table 2 for reliability analysis. Use the proposed method to process the new data obtained, then according to maximum support rule, performance accuracy is obtained and it is shown in Table 5. We can find that when the error rate is less than 10%, performance accuracy is 100%, so the proposed method has strong reliability.

Table 5.

Performance accuracy.

The proposed method					The typical method
Random error rate (%)	5	10	15	20	0
Performance accuracy (%)	100	100	95	83	94

Basir and Yuan⁶⁷ had used three sensors (two acceleration sensors and one acoustic sensors) to test the efficiency of the typical method. The performance accuracy is displayed on the left side of the Table 5. By comparing performance of the new method and the typical method, the former is slightly better than the latter.

Conclusion

In this article, a new method that entropy of each probability obtained from each feature is used as the discount factor to get the basic probability assignment is proposed. First, it is important to construct the frame of discernment. Second, the preliminary matrix is established by expert knowledge. Third, the basic probability assignment matrix can be obtained from calculating the distance between the values from samplings and preliminary matrix. Last, the Dempster’ rule is used to combine evidence to get final result. The proposed method could be demonstrated efficient from the preceding example. And the proposed method also has strong reliability. In addition, in this article, there is a limitation of this article which is that multiple subsets are not taken into account because the engine may have multiple faults at the same time. Therefore, the future research direction is to extend the single subset to multiple subsets to more effectively identify one or more possible faults of the engine at the same time.

Footnotes

Acknowledgements

The authors greatly appreciate the reviewers’ suggestions and the editor’s encouragement.

Handling Editor: ZW Zhong

Data availability statement

The authors confirm that the data sources in this paper are public.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partially supported by National Natural Science Foundation of China (Grant Nos 61573290 and 61503237).

ORCID iD

Yong Deng

References

Sharma

Parey

A review of gear fault diagnosis using various condition indicators. Procedia Engineer 2016; 144: 253–263.

Abdelkader

Derouiche

Kaddour

et al . Rolling bearing faults diagnosis based on empirical mode decomposition: optimized threshold de-noising method. In: Proceedings of the international conference on modelling, identification and control, Algiers, Algeria, 15–17 November 2016, pp.186–191. New York: IEEE.

Lei

Lin

et al . A review on empirical mode decomposition in fault diagnosis of rotating machinery. Mech Syst Signal Pr 2013; 35: 108–126.

Jiang

Xie

Zhuang

et al . Failure mode and effects analysis based on a novel fuzzy evidential method. Appl Soft Comput 2017; 57: 672–683.

Majid

NAA

Young

Taylor

et al . K-means clustering pre-analysis for fault diagnosis in an aluminium smelting process. In: Proceedings of the data mining and optimization, Langkawi, Malaysia, 2–4 September 2012, pp.43–46. New York: IEEE.

Gong

Qian

et al . Research on fault diagnosis methods for the reactor coolant system of nuclear power plant based on D-S evidence theory. Ann Nucl Energy 2018; 112: 395–399.

Abbasfard

Baniamerian

Khorasani

. Fault diagnosis of gas turbine engines: a symbolic multiple model approach. In: Proceedings of the control conference, Strasbourg, 24–27 June 2014, pp.944–951. New York: IEEE.

Abadi

Mendes

AMS

Cruz

SMA

. Three-level NPC inverter fault diagnosis by the average current Park’s vector approach. In: Proceedings of the XXth international conference on electrical machines, Marseille, 2–5 September 2012, pp.1893–1898. New York: IEEE.

Lee

Dexter

AL.

A fuzzy approach to fault diagnosis in the presence of sensor bias. Int J Fuzzy Syst 2001; 3: 1585–1588.

10.

Mohamed

Ibrahim

AS.

Model-based fault diagnosis via parameter estimation using knowledge base and fuzzy logic approach. In: Proceedings of the electrotechnical conference, Cairo, Egypt, 7–9 May 2002, pp.505–509. New York: IEEE.

11.

Zhang

Ashuri

Deng

A novel method for forecasting time series based on fuzzy logic and visibility graph. Adv Data Anal Classif 2017; 11: 759–783.

12.

Jiang

Wei

Liu

et al . Intuitionistic fuzzy power aggregation operator based on entropy and its application in decision making. Int J Intell Syst 2017; 1: 49–67.

13.

Deng

Generalized ordered propositions fusion based on belief entropy. Int J Comput Commun 2018; 13: 792–807.

14.

Zheng

Deng

Evaluation method based on fuzzy relations between Dempster–Shafer belief structure. Int J Intell Syst 2018; 33: 1343–1363.

15.

Fei

Wang

Chen

et al . A new vector valued similarity measure for intuitionistic fuzzy sets based on OWA operators. Iran J Fuzzy Syst page 2017; 15: 5.

16.

Jiang

Wei

Xie

et al . An evidential sensor fusion method in fault diagnosis. Adv Mech Eng 2016; 8: 641820.

17.

Foo

GHB

Zhang

Vilathgamuwa

DM.

A sensor fault detection and isolation method in interior permanent-magnet synchronous motor drives based on an extended Kalman filter. IEEE T Ind Electron 2013; 60: 3485–3495.

18.

Tudoroiu

Khorasani

Satellite fault diagnosis using a bank of interacting Kalman filters. IEEE T Aero Elec Syst 2007; 43: 1334–1350.

19.

Huang

Tan

Tong

HL.

Fault diagnosis and fault-tolerant control in linear drives using the Kalman filter. IEEE T Ind Electron 2012; 59: 4285–4292.

20.

Zhang

Ding

Liu

et al . An artificial neural network approach to transformer fault diagnosis. IEEE T Power Deliver 1996; 11: 1836–1841.

21.

Samanta

Al-Balushi

KR.

Artificial neural network based fault diagnostics of rolling element bearings using time-domain features. Mech Syst Signal Pr 2003; 17: 317–328.

22.

Chow

Tipsuwan

et al . Neural-network-based motor rolling bearing fault diagnosis. IEEE T Ind Electron 2000; 47: 1060–1069.

23.

Mechefske

CK.

Objective machinery fault diagnosis using fuzzy logic. Mech Syst Signal Pr 1998; 12: 855–862.

24.

Chin

HC.

Fault section diagnosis of power system using fuzzy logic. IEEE Power Eng Rev 2003; 22: 62.

25.

Dash

Rengaswamy

Venkatasubramanian

Fuzzy-logic based trend classification for fault diagnosis of chemical processes. Comput Chem Eng 2003; 27: 347–362.

26.

Chang

CT.

A fuzzy-logic based fault diagnosis strategy for process control loops. Chem Eng Sci 2003; 58: 3395–3411.

27.

Zhang

Mahfouf

Panoutsos

et al . Knowledge discovery for friction stir welding via data driven approaches—part 2: multiobjective modelling using fuzzy rule based systems. Sci Technol Weld Joi 2012; 17: 681–693.

28.

Zhang

Mahfouf

Panoutsos

et al . Model fusion via a master fuzzy system with special application to engineering materials. IFAC Proc Vol 2011; 44: 11732–11737.

29.

Dempster

AP.

Upper and lower probabilities induced by a multivalued mapping. Ann Math Stat 1967; 38: 325–339.

30.

Shafer

A mathematical theory of evidence. Princeton, NJ: Princeton University Press, 1976.

31.

Yager

RR.

On viewing fuzzy measures as fuzzy subsets. IEEE T Fuzzy Syst 2016; 24: 811–818.

32.

Song

Wang

Lei

et al . Uncertainty measure for interval-valued belief structures. Measurement 2016; 80: 241–250.

33.

Xiao

An improved method for combining conflicting evidences based on the similarity measure and belief function entropy. Int J Fuzzy Syst pages 2018; 20: 1256–1266.

34.

Yager

RR.

Uncertainty modeling using fuzzy measures. Knowl-Based Syst 2016; 92: 1–8.

35.

Jiang

Wang

An uncertainty measure for interval-valued evidences. Int J Comput Commun 2017; 12: 631–644.

36.

Han

Deng

An enhanced fuzzy evidential DEMATEL method with its application to identify critical success factors. Soft Comput 2018; 22: 5073–5090.

37.

Zhang

Deng

Combining conflicting evidence using the EMATEL method. Soft Comput 2018; 2018: 1–10. DOI: 10.1007/s00500-018-3455-8.

38.

Xiao

An intelligent complex event processing with D numbers under fuzzy environment. Math Probl Eng 2016; 2016: 1–10.

39.

Deng

D-AHP method with different credibility of information. Soft Comput 2017; 2017: 1–7. DOI: 10.1007/s00500-017-2993-9.

40.

Deng

A new MADA methodology based on D numbers. Int J Fuzzy Syst 2018; 20(8): 2458–2469. DOI: 10.1007/s40815-018-0514-3.

41.

Kang

Chhipi-Shrestha

Deng

et al . Stable strategies analysis based on the utility of Z-number in the evolutionary games. Appl Math Comput 2018; 324: 202–217.

42.

Jiang

Xie

Zhuang

et al . Sensor data fusion with z-numbers and its application in fault diagnosis. Sensors 2016; 16: 1509.

43.

Xie

Wen

A method for fuzzy soft sets in decision making based on grey relational analysis and D-S theory of evidence: application to medical diagnosis. Comput Math Meth Med 2014; 2014: 581316.

44.

Zhang

Wang

Zhang

. An information fusion approach and its application based on D-S evidence theory and neural network. In: Proceedings of the Chinese control conference, Kunming, China, 16–18 July 2008, pp.623–626. New York: IEEE.

45.

Xiao

Aritsugi

Wang

et al . Efficient processing of multiple nested event pattern queries over multi-dimensional event streams based on a triaxial hierarchical model. Artif Intell Med 2016; 72: 56–71.

46.

Deng

Evidential model validation under epistemic uncertainty. Math Probl Eng 2018; 2018: 6789635.

47.

Deng

Dependent evidence combination based on Shearman coefficient and Pearson coefficient. IEEE Access 2018; 6: 11634–11640.

48.

Yin

Deng

Measuring transferring similarity via local information. Physica A 2018; 498: 102–115.

49.

Xiao

Zhan

Lai

et al . New parallel processing strategies in complex event processing systems with data streams. Int J Distrib Sens Netw 2017; 13: 1–15.

50.

Yager

Elmore

Petry

Soft likelihood functions in combining evidence. Inform Fusion 2017; 36: 185–190.

51.

Jiang

Deng

et al . A modified Physarum-inspired model for the user equilibrium traffic assignment problem. Appl Math Model 2018; 55: 340–353.

52.

Zhu

Zou

Chen

et al . The robot target recognition based on support vector machine and D-S evidence theory. Adv Mater Res 2011; 308–310: 1215–1219.

53.

Pan

Hou

Yang

GS.

An algorithm for target recognition based on infrared/radar double mode fusion combining neuron-fuzzy technique with D-S evidence theory. Chin J Electron 2002; 11: 357–360.

54.

Song

et al . The optimal design of industrial alarm systems based on evidence theory. Control Eng Pract 2016; 46: 142–156.

55.

Liu

Sun

et al . Fault diagnosis based on the updating strategy of interval-valued belief structures. Chin J Electron 2014; 23: 753–760.

56.

Zheng

Yang

et al . Track irregularity fault identification based on evidence reasoning rule. In: Proceedings of the IEEE international conference on intelligent rail transportation, Birmingham, 23–25 August 2016, pp.298–306.

57.

Luo

Yang

et al . Agent oriented intelligent fault diagnosis system using evidence theory. Expert Syst Appl 2012; 39: 2524–2531.

58.

Yuan

Xiao

Fei

et al . Modeling sensor reliability in fault diagnosis based on evidence theory. Sensors 2016; 16: 113.

59.

Xiao

A novel evidence theory and fuzzy preference approach-based multi-sensor data fusion technique for fault diagnosis. Sensors 2017; 17: E2504.

60.

Chen

Diao

Sang

Weighted evidence combination rule based on evidence distance and uncertainty measure: an application in fault diagnosis. Math Probl Eng 2018; 2018: 1–10.

61.

Liu

Chen

Zhao

An intelligent fault diagnosis method for bogie bearings of metro vehicles based on weighted improved D-S evidence theory. Energies 2018; 11: 232.

62.

Kari

Gao

Zhao

et al . An integrated method of ANFIS and Dempster-Shafer theory for fault diagnosis of power transformer. IEEE T Dielect El In 2018; 25: 360–371.

63.

Yao

Wang

Zhang

Weak thruster fault detection for autonomous underwater vehicle based on artificial immune and signal pre-processing. Adv Mech Eng 2018; 10: 758739.

64.

Fleury

Trigo

Pacífico

et al . An inference model for combustion diagnostics in an experimental oil furnace. Expert Syst 2017; 2: e12245.

65.

Khazaee

Ahmadi

Omid

et al . Classifier fusion of vibration and acoustic signals for fault diagnosis and classification of planetary gears based on Dempster–Shafer evidence theory. Proc IMechE, Part E: J Process Mechanical Engineering 2014; 228: 21–32.

66.

Chen

Liang

et al . A fault diagnosis method for power transformers based on wavelet neural network and D-S evidence theory. In: Proceedings of the international conference on high voltage engineering and application, Chongqing, China, 9–12 November 2008, pp.666–671. New York: IEEE.

67.

Basir

Yuan

Engine fault diagnosis based on multi-sensory information fusion using Dempster-Shafer evidence theory. Inform Fusion 2007; 8: 379–386.

68.

Shannon

CE.

A mathematical theory of communication. Bell Syst Tech J 1948; 27: 379–423.

69.

Yager

Petry

An intelligent quality-based approach to fusing multi-source probabilistic information. Inform Fusion 2016; 31: 127–136.

70.

Liu

Dezert

Pan

Combination of sources of evidence with different discounting factors based on a new dissimilarity measure. Decis Support Syst 2011; 52: 133–141.

71.

Fan

Zuo

MJ.

Fault diagnosis of machines based on D-S evidence theory. Part 1: D-S evidence theory and its improvement. Pattern Recogn Lett 2006; 27: 366–376.

72.

Song

Wang

Lei

et al . Credibility decay model in temporal evidence combination. Inform Process Lett 2015; 115: 248–252.

73.

Deng

Shi

Zhu

et al . Combining belief functions based on distance of evidence. Decis Support Syst 2004; 38: 489–493.

74.

Jiang

Wang

Liu

et al . Evidence conflict measure based on OWA operator in open world. PLoS ONE 2017; 12: e0177828.

75.

Jiang

Wei

Qin

et al . Sensor data fusion based on a new conflict measure. Math Probl Eng 2016, 2016: 5769061.

76.

Han

Tao

Multi-sensor data fusion algorithm based on D-S evidence theory and fuzzy mathematics. Chin J Sci Instrum 2000; 21: 644–647.

77.

Wen

Multi-source uncertain information fusion theory and its application: fault diagnosis and reliability evaluation. Alexandria, NSW, Australia: Science Press, 2012.

78.

Song

Jiang

Engine fault diagnosis based on sensor data fusion using evidence theory. Adv Mech Eng 2016; 8: 673219.