A data indicator-based deep belief networks to detect multiple faults in axial piston pumps

A novel wavenets long short term memory paradigm for wind power prediction

This paper describes the application of wavelet transform (WT) to detect internal leakage in hydraulic actuators, caused by seal damage. The method analyzes the pressure signal at one side of the actuator in response to periodic step inputs to the control valve. It is shown that the detailed version of decomposed pressure signal, using discrete WT, establishes feature patterns that can effectively detect internal leakage and its severity. The proposed scheme requires a baseline (threshold) value, predetermined first by analyzing the pressure signal of a healthy actuator. Once the root mean square (rms) of the level-two detail coefficient values, obtained from the measured pressure signals in subsequent offline tests, fall below this baseline, a fault alarm is triggered. Furthermore, the degree of changes of the rms value from the one obtained under normal operating condition indicates the severity of fault. Experimental tests show promising results for detecting internal leakages as low as 0.124 L/min, representing approximately 2.6% reduction of flow rate available to move the actuator. This is done without a need to model the actuator or leakage. Other methods of leakage fault diagnosis require the model of the actuator or leakage fault. Furthermore, no other method reported the internal leakage detection of magnitude as low as the one reported in this paper.

A Wavelet-Based Approach to Internal Seal Damage Diagnosis in Hydraulic Actuators

Layered clustering multi-fault diagnosis for hydraulic piston pump

The fault detection and diagnosis (FDD) along with condition monitoring (CM) and of rotating machinery (RM) have critical importance for early diagnosis to prevent severe damage of infrastructure in industrial environments. Importantly, valuable industrial equipment needs continuous monitoring to enhance the safety, reliability, and availability and to decrease the cost of maintenance of modern industrial systems and applications. However, induction motor (IM) has been extensively used in several industrial processes because it is cheap, reliable, and robust. Rolling bearings are considered to be the main component of IM. Undoubtedly, any failure of this basic component can lead to a serious breakdown of IM and for whole industrial system. Thus, many current methods based on different techniques are employed as a fault prognosis and diagnosis of rolling elements bearing of IM. Moreover, these techniques include signal/image processing, intelligent diagnostics, data fusion, data mining, and expert systems for time and frequency as well as time-frequency domains. Artificial intelligence (AI) techniques have proven their significance in every field of digital technology. Industrial machines, automation, and processes are the net frontiers of AI adaptation. There are quite developed literatures that have been approaching the issues using signals and data processing techniques. However, the key contribution of this work is to present an extensive review of CM and FDD of the IM, especially for rolling elements bearings, based on artificial intelligent (AI) methods. This study highlights the advantages and performance limitations of each method. Finally, challenges and future trends are also highlighted.

/pdf/a-review-of-artificial-intelligence-methods-for-condition-2w1dbx3w3x.pdf

A Review of Artificial Intelligence Methods for Condition Monitoring and Fault Diagnosis of Rolling Element Bearings for Induction Motor

The diagnosis of hydraulic pump health in real-time is important for increasing hydraulic system reliability and performance. Because of high noise levels in the pump pulsation pressure signal, many existing health diagnosis methods, such as limit checking, spectrum analysis, and logic reasoning, cannot effectively perform a reliable on-line health diagnosis for hydraulic pumps. Wavelet analysis, a waveform signal analysis method performed by breaking up an evaluating signal into shifted and scaled versions of a standard wavelet, can identify feature signals in multiple decomposed band windows of the original signal. The methodology for applying this wavelet analysis in real-time health diagnosis for hydraulic pumps was investigated in this study. Results obtained from both simulation analysis and on-line experimental validation verified that the wavelet analysis method can improve the capability of diagnosing the health conditions of piston pumps, and more importantly can identify the types of pump defects based on the patterns and the amplitudes of obtained wavelet coefficients.

Q. Zhang

Papers

Wavelet--based pressure analysis for hydraulic pump health diagnosis