scholarly journals Multi-Factor Operating Condition Recognition Using 1D Convolutional Long Short-Term Network

Sensors ◽  
2019 ◽  
Vol 19 (24) ◽  
pp. 5488 ◽  
Author(s):  
Zhinong Jiang ◽  
Yuehua Lai ◽  
Jinjie Zhang ◽  
Haipeng Zhao ◽  
Zhiwei Mao
Keyword(s):  
Neural Network ◽  
Diesel Engine ◽  
Mean Value ◽  
Fault Detection And Diagnosis ◽  
Operating Conditions ◽  
Short Term ◽  
Vibration Signals ◽  
Operating Condition ◽  
Model Training ◽  
Condition Recognition

For a diesel engine, operating conditions have extreme importance in fault detection and diagnosis. Limited to various special circumstances, the multi-factor operating conditions of a diesel engine are difficult to measure, and the demand of automatic condition recognition based on vibration signals is urgent. In this paper, multi-factor operating condition recognition using a one-dimensional (1D) convolutional long short-term network (1D-CLSTM) is proposed. Firstly, a deep neural network framework is proposed based on a 1D convolutional neural network (CNN) and long short-Term network (LSTM). According to the characteristics of vibration signals of a diesel engine, batch normalization is introduced to regulate the input of each convolutional layer by fixing the mean value and variance. Subsequently, adaptive dropout is proposed to improve the model sparsity and prevent overfitting in model training. Moreover, the vibration signals measured under 12 operating conditions were used to verify the performance of the trained 1D-CLSTM classifier. Lastly, the vibration signals measured from another kind of diesel engine were applied to verify the generalizability of the proposed approach. Experimental results show that the proposed method is an effective approach for multi-factor operating condition recognition. In addition, the adaptive dropout can achieve better training performance than the constant dropout ratio. Compared with some state-of-the-art methods, the trained 1D-CLSTM classifier can predict new data with higher generalization accuracy.

Selection of Mother Wavelet for Effective Wavelet Transform of Bearing Vibration Signals

2014 ◽  
Vol 1039 ◽  
pp. 169-176 ◽  
Author(s):  
H.S. Kumar ◽  
P. Srinivasa Pai ◽  
N.S. Sriram ◽  
G.S. Vijay
Keyword(s):  
Wavelet Transform ◽  
Fault Detection And Diagnosis ◽  
Operating Conditions ◽  
Variable Load ◽  
Mother Wavelet ◽  
Vibration Signals ◽  
Time Frequency ◽  
Deep Groove ◽  
Wide Range ◽  
Bearing Vibration

Condition monitoring (CM) and fault diagnosis of equipments has gained greater attention in recent years, due to the need to reduce the down time and enhance the life/ condition of the equipments. The rolling element bearings (REB) are the most critical components in rotary machines. Hence, bearing fault detection and diagnosis is an integral part of the preventive maintenance activity. Vibration signal analysis provides wide range of information for analysis. So in this paper, vibration signals for four conditions of a deep groove ball bearing namely Normal (N), bearing with defect on inner race (IR), bearing with defect on ball (B), and bearing with defect on outer race (OR) have been acquired from a customized bearing test rig under maximum speed and variable load conditions. Depending on the machinery operating conditions and the extent of bearing defect severity, the measured vibration signals are non-stationary in nature. Non-stationary signals are effectively analyzed by wavelet transform technique, which is a popular and widely used time-frequency technique. The focus of this paper is to select a best possible mother wavelet for applying WT on bearing vibration signals. The two selection criteria includes minimum Shannon entropycriteria(MSEC) and Maximum Energy to Shannon Entropy Ratio criteriaR(s). This helps in effective bearing CM using WT.

Study on the Use of Blind ICA in Diesel Engine Vibration Processing

2012 ◽  
Vol 239-240 ◽  
pp. 430-433 ◽  
Author(s):  
Chao Ming Huang ◽  
Ping Zhang ◽  
Hong Liang Yu
Keyword(s):  
Diesel Engine ◽  
Operating Conditions ◽  
Characteristic Parameters ◽  
Vibration Signals ◽  
Engine Vibration ◽  
Processing Step ◽  
Fault State ◽  
Different Characteristics ◽  
Mechanical Movement ◽  
Fault Characteristic

A method which can be used to analyze and separate the vibration signals of diesel engine is proposed. The vibration signals contain a great deal of information about the engine’s fault state, and it is hard to obtain the fault characteristic parameters because of the complex mechanical movement and operating conditions. Study on vibration by fourth order blind identification is carried out in this paper. And FOBI model that estimate the separation matrix by independent component analysis is established and applied to diesel engine vibration to separate the different signals. The results show that signals of different characteristics can be separated perfectly. This method can be used as the pre-processing step to obtain the fault characteristic parameters.

Study on the Mean Value of Particle Concentration Signal in Silicon Power Fluidized Bed

2012 ◽  
Vol 550-553 ◽  
pp. 2936-2940
Author(s):  
Xing Yong Liu ◽  
Hu Yang ◽  
You Cheng Wang ◽  
Zhuo Xu Deng
Keyword(s):  
Neural Network ◽  
Fluidized Bed ◽  
Particle Concentration ◽  
Rbf Neural Network ◽  
Radial Distance ◽  
Mean Value ◽  
Operating Conditions ◽  
Axial Distance ◽  
The Mean ◽  
Concentration Signal

The particle concentration signals of silicon powder in the fluidizing gas i.e. air under different operating conditions were determined. The diameter of silicon particles, operating velocity, radial distance and axial distance are used as input vector; the mean value of particle concentration signal in the silicon power fluidized bed is used as a target vector. The RBF neural network is applied to build the predicted model of the mean value in silicon power fluidized bed. The result shows that the prediction of mean value through the RBF neural network is prior to that by BP neural network, and its error is less than 0.2%.

scholarly journals Real Time Monitoring and Neuro-Fuzzy Based Fault Diagnosis of Flow Process in Hybrid System

2012 ◽  
pp. 171-177
Author(s):  
Revathi. P ◽  
Pallikonda Rajasekaran. M ◽  
Babiyola. D ◽  
Aruna. R
Keyword(s):  
Neural Network ◽  
Fault Detection ◽  
Real Time ◽  
Fault Detection And Diagnosis ◽  
Operating Conditions ◽  
Economic Losses ◽  
Monitoring Systems ◽  
Back Propagation Algorithm ◽  
System Failures ◽  
Detection And Diagnosis

Process variables vary with time in certain applications. Monitoring systems let us avoid severe economic losses resulting from unexpected electric system failures by improving the system reliability and maintainability The installation and maintenance of such monitoring systems is easy when it is implemented using wireless techniques. ZigBee protocol, that is a wireless technology developed as open global standard to address the low-cost, low-power wireless sensor networks. The goal is to monitor the parameters and to classify the parameters in normal and abnormal conditions to detect fault in the process as early as possible by using artificial intelligent techniques. A key issue is to prevent local faults to be developed into system failures that may cause safety hazards, stop temporarily the production and possible detrimental environment impact. Several techniques are being investigated as an extension to the traditional fault detection and diagnosis. Computational intelligence techniques are being investigated as an extension to the traditional fault detection and diagnosis methods. This paper proposes ANFIS (Adaptive Neural Fuzzy Inference System) for fault detection and diagnosis. In ANFIS, the fuzzy logic will create the rules and membership functions whereas the neural network trains the membership function to get the best output. The output of ANFIS is compared with Back Propagation Algorithm (BPN) algorithm of neural network. The training and testing data required to develop the ANFIS model were generated at different operating conditions by running the process and by creating various faults in real time in a laboratory experimental model.

Optimization and prediction of engine block vibration using micro-electro-mechanical systems capacitive accelerometer, fueled with diesel-bioethanol (water-hyacinth) blends by response surface methodology and artificial neural network

2021 ◽  
pp. 095440622110528
Author(s):  
Akhilesh Kumar Choudhary ◽  
H Chelladurai ◽  
Hitesh Panchal
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Response Surface Methodology ◽  
Diesel Engine ◽  
Response Surface ◽  
Water Hyacinth ◽  
Engine Block ◽  
Micro Electro Mechanical Systems ◽  
Vibration Signals ◽  
Capacitive Accelerometer

The current investigation is focused on the vibration signals analysis for health status diagnosis of the single-cylinder diesel engine fueled with bioethanol diesel mixture. The water hyacinth (WH) plants (Eichhornia crassipes) are used as raw materials for bioethanol production. The bio-ethanol obtained from WH has been mixed with diesel fuel (WBED) to various extent. Systematically designed experiments were conducted with different working parameters like load, fuel injection pressure (FIP), and compression ratio (CR) in a diesel engine. The Micro-Electro-Mechanical Systems (MEMS) capacitive accelerometer was used to get vibration signals from the engine while operating with blended fuels. The obtained experimental vibrations data have been used to predict the engine vibration by using Response Surface Methodology (RSM) technique and Artificial Neural Network (ANN). The experimental results have been compared with RSM and ANN prediction results. From results, it is elicited that the acceleration declines with the increase in load and CR. At all tested blends, FIP produces a significant effect on the engine block vibration. Among all blends, WBED 5 and WBED 10 produce less vibration as compared to other diesel bioethanol blends. At optimized operating condition the engine block vibration for WBED 5; the experimental acceleration is 0.016962 m/s2 and the predicted acceleration by RSM and ANN is 0.016182 m/s2 and 0.0166 m/s2, respectively. For WBED 10, the acceleration is 0.0172604 m/s2 and the predicted acceleration by RSM and ANN 0.016207 m/s2 0.017 m/s2, respectively, has been found.

scholarly journals Fault Detection of Inline Reciprocating Diesel Engine: A Mass and Gas-Torque Approach

2012 ◽  
Vol 2012 ◽  
pp. 1-6 ◽  
Author(s):  
S. H. Gawande ◽  
L. G. Navale ◽  
M. R. Nandgaonkar ◽  
D. S. Butala ◽  
S. Kunamalla
Keyword(s):  
Diesel Engine ◽  
Fault Detection ◽  
Constant Load ◽  
Gas Pressure ◽  
Fault Detection And Diagnosis ◽  
Operating Conditions ◽  
Medium Speed ◽  
Di Diesel Engine ◽  
Detection And Diagnosis ◽  
Early Fault Detection

Early fault detection and diagnosis for medium-speed diesel engines are important to ensure reliable operation throughout the course of their service. This work presents an investigation of the diesel engine combustion-related fault detection capability of crankshaft torsional vibrations. Proposed methodology state the way of early fault detection in the operating six-cylinder diesel engine. The model of six cylinders DI Diesel engine is developed appropriately. As per the earlier work by the same author the torsional vibration amplitudes are used to superimpose the mass and gas torque. Further mass and gas torque analysis is used to detect fault in the operating engine. The DFT of the measured crankshaft’s speed, under steady-state operating conditions at constant load shows significant variation of the amplitude of the lowest major harmonic order. This is valid both for uniform operating and faulty conditions and the lowest harmonic orders may be used to correlate its amplitude to the gas pressure torque and mass torque for a given engine. The amplitudes of the lowest harmonic orders (0.5, 1, and 1.5) of the gas pressure torque and mass torque are used to map the fault. A method capable to detect faulty cylinder of operating Kirloskar diesel engine of SL90 Engine-SL8800TA type is developed, based on the phases of the lowest three harmonic orders.

Incipient Fault Detection and Diagnosis in Turbine Engines Using Hidden Markov Models

2003 ◽  
Author(s):  
Sunil Menon ◽  
O¨nder Uluyol ◽  
Kyusung Kim ◽  
Emmanuel O. Nwadiogbu
Keyword(s):  
Neural Network ◽  
Fault Detection ◽  
Markov Models ◽  
Hidden Markov ◽  
Fault Detection And Diagnosis ◽  
Operating Conditions ◽  
Turbine Engines ◽  
Incipient Fault ◽  
Detection And Diagnosis ◽  
Incipient Fault Detection

Incipient fault detection and diagnosis in turbine engines is key to effective maintenance and improved availability of systems dependent on these engines. In this paper, we present a novel method for incipient fault detection and diagnosis using Hidden Markov Models (HMMs). In particular, we focus on engine faults that are manifest in transient operating conditions such as engine startup and acceleration. HMMs are stochastic signal models that are effective in modeling transient signals. They are developed with engine data collected under nominal operating conditions. Engine data representing different fault conditions are used to develop the fault HMMs; a separate model is developed for each of the faults. Once the nominal and fault HMMs are developed, new engine data collected from the engine are evaluated against the HMMs and a determination is made whether a fault is indicated. Here, we demonstrate our HMM-based fault detection and diagnosis approach on engine speed profiles taken from a real engine. Further, the effectiveness of the HMM-based approach is compared with a neural-network-based approach and a method based on using principal component analysis in conjunction with a neural network approach.

The Engine Fault Diagnosis with RBFNN Based on AIC

2012 ◽  
Vol 468-471 ◽  
pp. 1066-1069
Author(s):  
Qiang Huang ◽  
Xiao Zhuo Ouyang ◽  
Cheng Wang
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Diesel Engine ◽  
Rbf Neural Network ◽  
Information Criterion ◽  
Vibration Signals ◽  
The Neural Network ◽  
Rotary Machines ◽  
Diagnosis Method ◽  
Diagnosis Model

In this paper, an engine diagnosis method with high precision and quickly response is proposed. Firstly, the Akaike Information Criterion (AIC) is used to improve the performance of the neural network to build the fault diagnosis model. Then the vibration signals are analyzed to estimate the states of the diesel engine. Finally, the five states of diesel engine are set to validate the veracity of diagnosis method. According to experiment and simulation researches, it indicates that the diagnosis method with RBF neural network based on AIC is effective. The veracity of identification is 100% to the single fault. It is a valuable reference to the vibration diagnosis for other complex rotary machines.

Sign in / Sign up

Export Citation Format

Share Document