Figure - uploaded by Panagiotis Loukopoulos
Content may be subject to copyright.
Source publication
Reciprocating compressors are critical components in the oil and gas sector, though their maintenance cost is known to be relatively high. Compressor valves are the weakest component, being the most frequent failure mode, accounting for almost half the maintenance cost. One of the major targets in industry is minimisation of downtime and cost, whil...
Similar publications
One of the major targets in industry is minimisation of downtime and cost, and maximisation of availability and safety, with maintenance considered a key aspect in achieving this objective. The concept of Condition Based Maintenance and Prognostics and Health Management (CBM/PHM), which is founded on the principles of diagnostics, and prognostics,...
Citations
... Principal Component Analysis (PCA) has been used as a methodology to create vHIs. Loukopoulos et al. [31] and Loutas et al. [22] used PCA, more specifically Q index and T 2 , as HIs to monitor the degradation of reciprocating compressors. Multiple ML models were then used to predict the RUL of the compressors using these two HIs. ...
... Physical indicators HI 3 and HI 4 are an advancement of HI 1 and HI 2 respectively, where HI 1 measures the strain deviation from the reference state and HI 2 shows the effect of each sensor's measured strain at the mean strain at the sensor's foot. Virtual indicators such as vHI 1 , vHI 2 and T 2 do not have an immediate correlation to physical measures, though they have demonstrated good prognostic potential in [31,32]. ...
... x i (t) and x ri (t) are the original and reconstructed data, extracted from PCA at time t and λ i and τ i , are the variance and score of the i th principal component. For more information regarding the HIs the interested reader is referred to [31][32][33][34]. ...
... Principal component analysis (PCA) is frequently used as a means to create vHIs. Loukopoulos et al. [30] applied PCA, specifically the T 2 and Q index, as HIs to predict the RUL of reciprocating compressors. Zhang et al. [31] used wavelet packet decomposition to extract features in combination with PCA to reduce the dimensionality of the data without losing information. ...
... The second vHI proposed in this section is the so-called Q index or the sum of reconstructed squared residuals of PCA. This was also presented as a feature of diag-nostic/prognostic potential in addition to Hotelling's T 2 in [30]. The methodology for calculating vHI 2 is as follows: ...
The development of health indicators (HI) of diagnostic and prognostic potential from generally uninformative raw sensor data is both a challenge and an essential feature for data-driven diagnostics and prognostics of composite structures. In this study, new damage-sensitive features, developed from strains acquired with Fiber Bragg Grating (FBG) and acoustic emission (AE) data, were investigated for their suitability as HIs. Two original fatigue test campaigns (constant and variable amplitude) were conducted on single-stringer composite panels using appropriate sensors. After an initial damage introduction in the form of either impact damage or artificial disbond, the panels were subjected to constant and variable amplitude compression-compression fatigue tests. Strain sensing using FBGs and AE was employed to monitor the damage growth, which was further verified by phased array ultrasound. Several FBGs were incorporated in special SMARTapes TM , which were bonded along the stiffener's feet to measure the strain field, whereas the AE sensors were strategically placed on the panels' skin to record the acoustic emission activity. HIs were developed from FBG and AE raw data with promising behaviors for health monitoring of composite structures during service. A correlation with actual damage was attempted by leveraging the measurements from a phased array camera at several time instances throughout the experiments. The developed HIs displayed highly monotonic behaviors while damage accumulated on the composite panel, with moderate prognosability.
... These HIs were used to predict the RUL of aircraft engines. Loukopoulos et al. [10] used Principal Component Analysis' (PCA) metrics as HIs, more specifically Q index and hotelling's T2, to predict RUL in reciprocating compressors. Zhang et al. [11] also used PCA to reduce the dimensionality of a wavelet decomposition analysis in rotating machinery. ...
... A vHI based on PCA is introduced in this subsection. The HI is Q index, namely the squared sum of residual reconstructed error, and has been previously used in [10]. A PCA model is constructed from a portion of the data X ref , and the transformation coefficients P are then used to transform the entire data X into a new PCA space. ...
... Cabrera et al. [17] used a time series of vibration signals collected from the compressor to train a set of long short-term memory (LSTM) models, which is suitable for diagnosing the valve failure of reciprocating compressors. Other fault-diagnosis techniques such as the k-nearest neighbors (KNN) for temperature data [18], artificial neural networks (ANN) or a genetic algorithm for vibration data [19,20] and hybrid deep belief network (HDBN) for pressure, motor current and vibration data [21], ref. [22] have been demonstrated by other researchers. These methods all require a large number of learning samples, and the acquisition of samples, especially the acquisition of unknown faults, is the main dilemma of such methods. ...
As an essential mechanical device in many industrial applications, reciprocating compressors may be subject to thermal performance failures, mechanical function failures and motor faults resulting in extremely severe catastrophic collapses. Generally, the presence of such faults affects the temperature field distribution of the device. Infrared thermography technology can detect the thermal radiation signal of an object and converts it into images, which is sensitive and reliable to monitor the condition of reciprocating compressor systems. In this paper, three kinds of faults are simulated in an uncontrolled temperature environment. The temperature distribution signal of a reciprocating compressor is captured by a remote infrared camera in the form of a heat map during the experimental process. A slight shaking window is employed to crop the photographed range of experimental equipment, and 30% of each type of images are flipped to prevent the image position information from affecting the classification results. A convolutional neural networks (CNN) is involved for evaluating the monitoring by classifying three common faulty operations. The results demonstrate that thermal images contains the full information and can be a promising technique to diagnose the faults of reciprocating compressors under various operating conditions with a classification accuracy of more than 98.59%.
... Cabrera et al. [17] used a time series of vibration signals collected from the compressor to train a set of long short-term memory (LSTM) models, which is suitable for diagnosing the valve failure of reciprocating compressors. Other fault-diagnosis techniques such as the k-nearest neighbors (KNN) for temperature data [18], artificial neural networks (ANN) or a genetic algorithm for vibration data [19,20] and hybrid deep belief network (HDBN) for pressure, motor current and vibration data [21], ref. [22] have been demonstrated by other researchers. These methods all require a large number of learning samples, and the acquisition of samples, especially the acquisition of unknown faults, is the main dilemma of such methods. ...
As an essential mechanical device in many industrial applications, reciprocating compressors have a high demand for operating efficiency and availability. Because the temperature of each part of a reciprocating compressor depends considerably on operating conditions, faults in any parts will cause the variation of the temperature distribution, which provides the possibility to distinguish the fault type of reciprocating compressors by differentiating the distribution using infrared thermal imaging. In this paper, three types of common fault are laboratory experimented in an uncontrolled temperature environment. The temperature distribution signals of a reciprocating compressor are captured by a non-contact infrared camera remotely in the form of heat maps during the experimental process. Based on the temperature distribution under baseline condition, temperature fields of six main components were selected via Hue-Saturation-Value (HSV) image as diagnostic features. During the experiment, the average grayscale values of each component were calculated to form 6-dimension vectors to represent the variation of the temperature distribution. A computational efficient multiclass support vector machine (SVM) model is then used for classifying the differences of the distributions, and the classification results demonstrate that the average temperatures of six main components aided by SVM is a promising technique to diagnose the faults of reciprocating compressors under various operating conditions with a classification accuracy of more than 99%.
... At present, vibration signals are mostly used as sample data for fault diagnosis of reciprocating compressor valves. However, according to the thermal performance parameter fault diagnosis method of reciprocating compressors established by Qi [19], it is known that the most direct response of a gas valve when a leak occurs is the pressure ratio imbalance, and the temperature and pressure responses are the most obvious [20][21][22][23][24]. As mentioned earlier, the CNNs' ability to learn to extract the optimal features and the system can achieve the optimal fault classification and fault detection accuracy through proper training. ...
Reciprocating compressors are important equipment in oil and gas industries which closely relate with the healthy development of the enterprise. It is essential to detect the valve fault because valve failures account for 60% in total failures. For this field, an artificial neural network (ANN) is widely used, but a complex network is not suitable for its low accuracy and easy overfitting. This paper proposes a fault diagnosis model of a reciprocating compressor valve based on a one-dimensional convolutional neural network (1DCNN). This method takes the differential pressure and differential temperature of each compressor stage as the input of 1DCNN, using the characteristics of the CNN to extract the features and finally using Softmax to classify the fault. In order to verify this method, it is compared with LM-BP, RBF, and BP neural networks. The results show that the fault recognition rate of 1DCNN reaches 100%, which proves the effectiveness and feasibility of the proposed method.
... Previous studies attempted to simulate the mathematical model of compressor valves through dynamics and thermodynamics approach, [3][4][5] while recent researchers include the fluid dynamics and fluid-structure interaction to simulate the valve motions numerically. 6 Due to the complexity of these numerical valve models, several diagnostic methods are developed to monitor the condition of valves, namely the temperature measurements, 7,8 pressure-volume (PV) analysis, 9,10 dynamic pressure and instantaneous angular speed (IAS), [11][12][13] vibration, 14,15 and acoustic emission (AE) technique. This study employs AE as a nondestructive sensing method for valve diagnosis. ...
Acoustic emission technique is often employed to detect valve abnormalities. With the development of technology, machine learning-based fault diagnosis methods are prevalent in the nondestructive testing industry as they can automatically detect valve problems without any human intervention. Nevertheless, feeding in all possible input parameters into the learning algorithm without any prior assessment may result in high computational cost and time, while adding to the risk of having false alarms. This study intended to obtain characteristics of acoustic emission signal for various valve conditions and compressor speeds by examining the four most commonly used parameters, namely the acoustic emission root mean square, acoustic emission crest factor, acoustic emission variance, and acoustic emission kurtosis. The study begins with time–frequency analysis of one revolution acoustic emission signal acquired from a faulty suction valve through discrete wavelet transform to obtain the signal characteristics of valve events. To associate signals with valve movements, the reconstructed discrete wavelet transform signals are further segregated into six time segments, and the four acoustic emission parameters are computed from each of the time segments. These parameters are analyzed through statistical analysis namely the two-way analysis of variance, followed by the Tukey test to obtain the best parameter which can differentiate each valve condition clearly at all speeds. The results revealed that acoustic emission root mean square is the best parameter especially in identification of heavy grease valve condition during suction valve opening event while acoustic emission crest factor is capable to detect leaky valve during the suction valve closing event at all speeds. It is believed that effective valve diagnosis strategy can be delivered by referring to the features of parameters and the characteristic valve event timing corresponding to each valve condition and speed.
... The diagnosis effect of data-driven method mainly depends on the quantity and quality of data and the conditions of collecting data, and it has low requirements for experience knowledge and fault mechanism. Therefore, it has been actively studied in the field of RC fault diagnosis, among which local mean decomposition [1,16], deep confidence network and back-propagation neural network [17][18][19], support vector machine (SVM) [9,20], k approximate regression [18,21], Bayesian estimation algorithm [10,22], big data [23], and other technologies have been successfully applied. The method of combining model and data-driven is to diagnose the system fault by fusing the system operation data with the system fault model. ...
High impact and strong noise complicate the response of reciprocating compressor (RC). It requires a complex signal processing method that is a single response-based or excitation-based fault diagnosis method applied to RC valve leakage fault diagnosis. This paper proposes a quantitative diagnosis method of RC valve leakage that is based on system characteristic diagnosis method. First, the current signal of the RC induction motor and the cylinder vibration signal are introduced as the excitation and response signals, the mathematical model of the RC motor current is established, and the influence mechanism of the valve leakage on the RC vibration is analyzed. Subsequently, the ensemble empirical mode decomposition and comb filter are respectively used to extract the fault characteristic information of excitation signal and response signal to obtain the excitation condition indicators (CIs), response CIs, and system CIs. Finally, the support vector machine based on the obtained CIs classified the valve leakage failure patterns of different severity, and a fault diagnoser was constructed for the quantitative diagnosis of valve leakage fault. The results of experiment and application proved that the proposed method could realize the quantitative diagnosis of RC valve leakage fault while using simple signal processing technology.
... The compressor's valves are the weakest component in the compressor and being the most frequent failing element and accounting for almost half of the maintenance cost. Study used data for valve temperature and presented the combination of algorithms analysis output by using several methods such as multiple linear regression, polynomial regression, K-Nearest Neighbors Regression (KNNR) along with remain useful life method (RUL) [11].The algorithm analysis was used to assess the valve failure conditions. The result showed that all performed methods compared well in qualitative (graphs) and quantitative (metric) analysis. ...
The gas compression plant is one of the major unit in gas industries. Gas needs to be compressed either for process handling, power generation, gas injection or transferring either treated or raw gas as products. Gas compression location always classifies as critical and hazard zone due to high process parameters mainly pressure, temperature and due to gas properties. In addition, gas plants are critical due to suspected impact to human health, safety and environment at particular level of incidents or leaks may happen. In addition, to consider the cost of the compression equipment which are unique and customized based on particular design conditions. Therefore, some studies are conducted mainly for gas equipment condition monitoring (CM) and focused to repair the common frequent failures in gas compression equipment, but limited studies were focused in gas plant maintenance management or to develop the applied maintenance system, strategic plans, risk based inspection (RBI) for the gas compression plant in order to maximize the integrity and reliability levels and the overall equipment effectiveness (OEE) with resources optimization.
This paper presents the literature reviewing of recent relevant contexts and studies in the maintenance systems and condition monitoring for gas compression plants use reciprocating compressors.
http://irphouse.com/volume/ijertv12n12.htm
... It is studied in [10,11] that piston compressors are critical components in the oil and gas sector, although their operating costs are known to be rather high. Compressor valves are the weakest components, being the most frequent type of failure, and they account for nearly half of maintenance costs. ...
To increase the efficiency of gas-lift oil and gas production, it is necessary to improve the operation of compressor stations, namely, to increase the reliability of the gas-motor piston compressor units installed in them. It is found that one of the vulnerable units affecting the reliability and efficiency of the gas-motor piston compressor are direct-flow self-acting valves.
In the process of preparing gas for transportation, as well as to ensure the reliability and efficiency of the entire production process, it is necessary to eliminate all gas leaks, prevent liquid hydrocarbon components and solids from entering the valve plates.
To solve this problem, associated petroleum gas must be cleaned from solids, heavy hydrocarbon components and moisture. To this end, it is recommended to install an additional new design horizontal gas separator on the suction line of gas-motor piston compressors.
The usefulness and importance of the new gas separator lie in a more efficient cleaning of associated petroleum gas supplied to the suction of the 1st stage compressor cylinders, which improves compressor performance, minimizing the leakage of valve plates.
The new design separator is used to clean gas from coarse and fine-grained dropping liquid, partially liquid in the vapor-phase state and solids. The new separator can also be used in various sectors of the oil and gas industry.
The purpose of installing a new gas separator is to increase the efficiency of gas cleaning from liquid and solid impurities
