An Integrated Deep Learning Method towards Fault Diagnosis of Hydraulic Axial Piston Pump

A few-shot sample augmentation algorithm based on SCAM and DEPS for pump fault diagnosis

In recent years, pumps have become critical components in agriculture, industry, and the military, necessitating extensive development and implementation of the fault diagnosis method. In the majority of existing fault classification models, the connection between performance improvement and the amount of training data remains high, yet real-world samples are difficult to obtain. Combining domain migration theory and sample expansion method, this paper introduces a few-shot learning fault diagnosis method. Employing the T-SNE visualization algorithm, we examine the validity of the self-calibration attention mechanism (SCAM) and distribution edge prediction strategy (DEPS). The accomplishment demonstrated that the proposed algorithm could effectively map the expanded sample space within a separate interval, thereby avoiding the problem of feature aliasing caused by the overlap of sample features among similar categories and significantly enhancing the quality and quantity of training samples. The experimental analysis indicates that the proposed methodology can effectively increase the accuracy of few-shot tasks, especially in the 9way-15shot task, where it maintains a performance of 72 %, which leading the mean accuracy calculated from the others of about 30%. It is believed that much of the work has superior applicability to other few-shot diagnosis cases.

A Transfer-Based Convolutional Neural Network Model with Multi-Signal Fusion and Hyperparameter Optimization for Pump Fault Diagnosis

Pumps are one of the core components of drilling equipment, and their fault diagnosis is of great significance. The data-driven approach has made remarkable achievements in the field of pump fault diagnosis; however, most of them are easily affected by complex background conditions and usually suffer from data scarcity problems in real-industrial scenarios, which limit their application in practical engineering. To overcome the above shortcoming, a novel framework for a model named Hyperparameter Optimization Multiple-Signal Fusion Transfer Convolution Neural Network is proposed in this paper. A convolutional neural network model based on transfer learning is built to promote well-learned knowledge transfer over different background conditions, improve robustness, and generalize the model to cross-domain diagnosis tasks. The multi-signal fusion strategy is involved in capturing system state information for establishing the mapping relationship between the raw signal and fault pattern by integrating the multi-physical signal with the weight allocation protocol. The hyperparameter optimization method is explored in conjunction with the transfer-based model by integrating Grid Search with the Gradient Descent algorithm for further improvement of diagnosis performance. Results show that the proposed model can effectively realize the fault diagnosis of pumps under different background conditions, achieving 95% accuracy.

Current Status and Applications for Hydraulic Pump Fault Diagnosis: A Review

To implement Prognostics Health Management (PHM) for hydraulic pumps, it is very important to study the faults of hydraulic pumps to ensure the stability and reliability of the whole life cycle. The research on fault diagnosis has been very active, but there is a lack of systematic analysis and summary of the developed methods. To make up for this gap, this paper systematically summarizes the relevant methods from the two aspects of fault diagnosis and health management. In addition, in order to further facilitate researchers and practitioners, statistical and comparative analysis of the reviewed methods is carried out, and a future development direction is prospected.

Piston Wear Detection and Feature Selection Based on Vibration Signals Using the Improved Spare Support Vector Machine for Axial Piston Pumps

A piston wear fault is a major failure mode of axial piston pumps, which may decrease their volumetric efficiency and service life. Although fault detection based on machine learning theory can achieve high accuracy, the performance mainly depends on the detection model and feature selection. Feature selection in learning has recently emerged as a crucial issue. Therefore, piston wear detection and feature selection are essential and urgent. In this paper, we propose a vibration signal-based methodology using the improved spare support vector machine, which can integrate the feature selection into the piston wear detection learning process. Forty features are defined to capture the piston wear signature in the time domain, frequency domain, and time-frequency domain. The relevance and impact of sparsity in 40 features are illustrated through the single and multiple statistical feature analysis. Model performance is assessed and the sparse features are discovered. The maximum model testing and training accuracy are 97.50% and 96.60%, respectively. Spare features s₁₀, s₁₂, E_w(8), x₇, E_e(5), and E_e(4) are selected and validated. Results show that the proposed methodology is applicable for piston wear detection and feature selection, with high model accuracy and good feature sparsity.

Research on Prediction Method of Gear Pump Remaining Useful Life Based on DCAE and Bi-LSTM

As a hydraulic pump is the power source of a hydraulic system, predicting its remaining useful life (RUL) can effectively improve the operating efficiency of the hydraulic system and reduce the incidence of failure. This paper presents a scheme for predicting the RUL of a hydraulic pump (gear pump) through a combination of a deep convolutional autoencoder (DCAE) and a bidirectional long short-term memory (Bi-LSTM) network. The vibration data were characterized by the DCAE, and a health indicator (HI) was constructed and modeled to determine the degradation state of the gear pump. The DCAE is a typical symmetric neural network, which can effectively extract characteristics from the data by using the symmetry of the encoding network and decoding network. After processing the original vibration data segment, health indicators were entered as a label into the RUL prediction model based on the Bi-LSTM network, and model training was carried out to achieve the RUL prediction of the gear pump. To verify the validity of the methodology, a gear pump accelerated life experiment was carried out, and whole life cycle data were obtained for method validation. The results show that the constructed HI can effectively characterize the degenerative state of the gear pump, and the proposed RUL prediction method can effectively predict the degeneration trend of the gear pump.

A New Fusion Fault Diagnosis Method for Fiber Optic Gyroscopes

The fiber optic gyroscope (FOG) is a high precision inertial navigation device, and it is necessary to ensure its reliability for effective use. However, the extracted fault features are easily distorted due to the interference of vibrations when the FOG is in operation. In order to minimize the influence of vibrations to the greatest extent, a fusion diagnosis method was proposed in this paper. It extracted features from fault data with Fast Fourier Transform (FFT) and wavelet packet decomposition (WPD), and built a strong diagnostic classifier with a sparse auto encoder (SAE) and a neural network (NN). Then, a fusion neural network model was established based on the diagnostic output probabilities of the two primary classifiers, which improved the diagnostic accuracy and the anti-vibration capability. Then, five fault types of the FOG under random vibration conditions were established. Fault data sets were collected and generated for experimental comparison with other methods. The results showed that the proposed fusion fault diagnosis method could perform effective and robust fault diagnosis for the FOG under vibration conditions with a high diagnostic accuracy.

Big Data Analysis and Prediction System Based on Improved Convolutional Neural Network

This paper presents a big data analysis and prediction system based on convolutional neural networks. Continuous template matching technology is used to analyze the distributed data structure of big data, and the information fusion processing of cloud service combination big data is combined with matching related detection methods, frequent item detection, and association rule feature extraction of high-dimensional fusion data. A clustering method is adopted to realize the classification and mining of cloud service portfolio big data. The hardware equipment of the car to detect the surrounding environment is complicated, and the combination of the convolutional neural network and the camera to detect the surrounding environment has become a research hotspot. However, simply using the convolutional neural network to process the camera data to control the turning angle of the car has the problems of long training time and low accuracy. An improved convolutional neural network is proposed. The experimental results show that the accuracy of data mining by this method is 12.43% and 21.76% higher than that of traditional methods, and the number of iteration steps is shorter, indicating that the timeliness of mining is higher. This network structure can effectively improve the training speed of the network and improve the accuracy of the network. It is proven that the convolutional neural network has faster training speed and higher accuracy.

Wear Analysis of Additively Manufactured Slipper-Retainer in the Axial Piston Pump

Additive manufacturing (AM) of spare parts is going to become more and more common. In the case of hydraulic solutions, there are also some applications of AM technology related to topological optimization, anti-cavitation improvements, etc. An examination of all available research results shows that authors are using specialized tools and machines to properly prepare AM spare parts. The main aim of this paper is to analyze the influence of quick repair of the damaged slipper-retainer from an axial piston pump by using an AM spare part. Hence, it was prepared with a 100-h test campaign of the AM spare part, which covers the time between damage and supply of the new pump. The material of the slipper-retainer has been identified and replaced by another material—available as a powder for AM, with similar properties as the original. The obtained spare part had been subjected to sandblasting only to simulate extremely rough conditions, directly after the AM process and an analysis of the influence of the high surface roughness of AM part on wear measurements. The whole test campaign has been divided into nine stages. After each stage, microscopic measurements of the pump parts’ surface roughness were made. To determine roughness with proper measurements, a microscopical investigation was conducted. The final results revealed that it is possible to replace parts in hydraulic pumps with the use of AM. The whole test campaign caused a significant increase in the surface roughness of the pump’s original parts, which was worked with the AM spare slipper-retainer: (1) from Ra = 0.54 µm to Ra = 3.84 µm in the case of two tested pistons; (2) from Ra = 0.33 µm to Ra = 1.98 µm in the case of the slipper-retainer. Despite significant increases in the surface roughness of the pump’s parts, the whole test campaign has been successfully finished without any damages to the other important parts of the whole hydraulic test rig.

Application of the MEMS Accelerometer as the Position Sensor in Linear Electrohydraulic Drive

Various distance sensors are used as measuring elements for positioning linear electrohydraulic drives. The most common are magnetostrictive transducers or linear variable differential transformer (LVDT) sensors mounted inside the cylinder. The displacement of the actuator’s piston rod is proportional to the change in the value of the current or voltage at the output from the sensor. They are characterized by relatively low measurement noise. The disadvantage of presented sensors is the need to mount them inside the cylinders and the high price. The article presents preliminary research on the replacement of following sensors and the use of a microelectromechanical system (MEMS) accelerometer as a measuring element in the electrohydraulic drive control system. The control consisted of two phases: at first, the signal from the acceleration sensor was analyzed during the actuator movement, based on the value determined from the simplified model implemented on the controller. In the range of motion in which the dynamics were the lowest, the signal was integrated and the obtained value was used in the second phase of motion. In the correction phase, a new set point was determined. Conducting the research required building a dedicated research stand. The author conducted the simulation and experimental research.

Intelligent Fault Diagnosis of Hydraulic Piston Pump Based on Wavelet Analysis and Improved AlexNet

Hydraulic piston pump is the heart of hydraulic transmission system. On account of the limitations of traditional fault diagnosis in the dependence on expert experience knowledge and the extraction of fault features, it is of great meaning to explore the intelligent diagnosis methods of hydraulic piston pump. Motivated by deep learning theory, a novel intelligent fault diagnosis method for hydraulic piston pump is proposed via combining wavelet analysis with improved convolutional neural network (CNN). Compared with the classic AlexNet, the proposed method decreases the number of parameters and computational complexity by means of modifying the structure of network. The constructed model fully integrates the ability of wavelet analysis in feature extraction and the ability of CNN in deep learning. The proposed method is employed to extract the fault features from the measured vibration signals of the piston pump and realize the fault classification. The fault data are mainly from five different health states: central spring failure, sliding slipper wear, swash plate wear, loose slipper, and normal state, respectively. The results show that the proposed method can extract the characteristics of the vibration signals of the piston pump in multiple states, and effectively realize intelligent fault recognition. To further demonstrate the recognition property of the proposed model, different CNN models are used for comparisons, involving standard LeNet-5, improved 2D LeNet-5, and standard AlexNet. Compared with the models for contrastive analysis, the proposed method has the highest recognition accuracy, and the proposed model is more robust.