Search Results (77)

This study focuses on the optimization of valve assemblies in downhole rod pumping units (DRPUs), which remain the predominant artificial lift technology in oil production worldwide. The research addresses the critical issue of premature failures in DRPUs caused by leakage in valve pairs, i.e., a problem that accounts for approximately 15% of all failures, as identified in a statistical analysis of the 2022 operational data from the Uzen oilfield in Kazakhstan. The leakage is primarily attributed to the accumulation of mechanical impurities and paraffin deposits between the valve ball and seat, leading to concentrated surface wear and compromised sealing. To mitigate this issue, a novel valve assembly design was developed featuring a flow turbulizer positioned beneath the valve seat. The turbulizer generates controlled vortex motion in the fluid flow, which increases the rotational frequency of the valve ball during operation. This motion promotes more uniform wear across the contact surfaces and reduces the risk of localized degradation. The turbulizers were manufactured using additive FDM technology, and several design variants were tested in a full-scale laboratory setup simulating downhole conditions. Experimental results revealed that the most effective configuration was a spiral plate turbulizer with a 7.5 mm width, installed without axis deviation from the vertical, which achieved the highest ball rotation frequency and enhanced lapping effect between the ball and the seat. Subsequent field trials using valves with duralumin-based turbulizers demonstrated increased operational lifespans compared to standard valves, confirming the viability of the proposed solution. However, cases of abrasive wear were observed under conditions of high mechanical impurity concentration, indicating the need for more durable materials. To address this, the study recommends transitioning to 316 L stainless steel for turbulizer fabrication due to its superior tensile strength, corrosion resistance, and wear resistance. Implementing this design improvement can significantly reduce maintenance intervals, improve pump reliability, and lower operating costs in mature oilfields with high water cut and solid content. The findings of this research contribute to the broader efforts in petroleum engineering to enhance the longevity and performance of artificial lift systems through targeted mechanical design improvements and material innovation. Full article

Hydropressor systems are of paramount importance in keeping water supplies running properly. A typical such device consists of two (or more) identical electropumps operating alternately, so as to avoid downtime as much as possible. We consider a dual pump configuration to identify the ideal usage proportion of each pump (from 0%-100%, meaning interchange only upon failure, to 50%-50%, where each pump works half the time) in order to improve availability, accounting solely for corrective maintenance. We also address the possibility of improving the availability of a single pump under the hazard of failure in three different ways (with their own occurrence frequencies), while also accounting for preventive maintenance. Both settings are tackled through Monte Carlo simulation and the models are implemented with the Python 3.12 programming language. The results indicate that significant improvements to standard industry practices can be made. Full article

Plasticized polyvinyl chloride (PVC-P) geomembranes (GMBs) are applied as anti-seepage materials in membrane-faced rockfill dams and pumped storage power stations. Assessing their lifetime to ensure durability during operation is crucial. This study conducted accelerated aging tests on three PVC-P GMBs immersed in water, along with axial tensile tests to investigate the degradation of mechanical properties. The degradation model was constructed using the Arrhenius equation, and the time to nominal failure (TNF) was predicted based on this model and failure criterion. The prediction model’s accuracy was verified using test data collected over 180 days at 20 °C. The results demonstrate that the TNF of PVC-P GMBs is influenced by water temperature, plasticizer content, and thickness of GMBs. Elevated temperatures accelerate the loss rate of plasticizers. Specifically, at 20 °C in a water environment, the estimated TNFs of Materials A and B with identical thicknesses were 49.05 and 153.76 years, respectively. This suggests that increasing the initial plasticizer content and enhancing its structural stability can significantly extend the TNF. Furthermore, Material C, which has a composition similar to Material B but with increased thickness, exhibited a predicted TNF of 181.30 years, indicating that greater thickness can effectively reduce the migration rate of plasticizers. The findings provide a theoretical basis for evaluating the TNF of PVC-P GMBs in reservoir bottom and below dead water level applications during operation. Full article

In the present paper, the subject of investigation is the reliability assessment of the single-stage reversible Hydropower Unit No. 3 (HU3) in the Bulgarian Pumped Hydro-Electric Storage (PHES) plant “Chaira”, which processes the waters of the “Belmeken” dam and “Chaira” dam. Preceding the destruction of HU4 and its virtual simulation, an analysis and its conclusions for rehabilitation and safety provided the information required for the reliability assessment of HU3. Detailed analysis of the consequences of the prolonged use of HU3 was carried out. The Supervisory Control and Data Acquisition (SCADA) system records were studied. Fault Tree Analysis (FTA) was applied to determine the component relationships and subsystem failures that can lead to an undesired primary event. A Failure Modes and Effect Analysis methodology was proposed for the large-scale hydraulic units and PHES. Based on the data of the virtual simulation and the investigations of the HU4 and its damages, as well as on the failures in the stay vanes of HU3, it is recommended to organize the monitoring of crucial elements of the structure and of water ingress into the drainage holes, which will allow for detecting failures in a timely manner. Full article

This study investigates groundwater flow patterns in a landslide area above the settlement of Koroška Bela in NW Slovenia using a series of tracer tests with sodium chloride (NaCl) and fluorescein (uranine). The tracer experiments, using a combination of pumping tests and continuous groundwater observations, reveal two distinct groundwater flow horizons within the landslide body: a prevailing shallower flow within highly permeable gravel layers and a slower deep flow in the weathered low-permeability clastic layers. Uranine injections suggest longer retentions, indicating complex hydrogeological conditions. Groundwater is recharged by the infiltration of precipitation and subsurface inflow from the upper-lying carbonate rocks. In the upper landslide, highly permeable gravel layers accelerate flow, especially during heavy rainfall, while downstream interactions between permeable gravel and less permeable clastic materials create local aquifers and springs. These groundwater dynamics significantly influence landslide stability, as rapid infiltration during intense precipitation events can lead to transient increases in pore water pressure, reducing shear strength and potentially triggering slope movement. Meanwhile, slow deep flows contribute to prolonged saturation of critical failure surfaces, which may weaken the landslide structure over time. The study emphasizes the region’s geological heterogeneity and landslide stability, providing valuable insights into the groundwater dynamics of this challenging environment. By integrating hydrogeological assessments with engineering measures, the study provides supportive information for mitigating landslide risks and improving groundwater management strategies. Full article

To investigate the motion patterns of flexible fibers inside a sewage pump and their impact on internal flow characteristics, visualization experiments were conducted to compare the pump flow when transporting water—0.3% CMC solution and 0.3% CMC solution containing flexible fibers under different operating conditions. The results showed that changes in the rheological properties of the 0.3% CMC solution primarily affected fluid viscous dissipation. Under the same rotational speed, the flow rate increased by only 2.4%, but power consumption decreased by 9.1%, resulting in a 6.4% improvement in efficiency. The curvature and distribution of fibers within the impeller flow channel remained stable. Their impact on the flow was characterized by an overall reduction in velocity within the impeller region, with the peak velocity decreasing by up to 26.3%. The primary cause of pump failure due to fibers was their tendency to repeatedly accumulate and detach at the tongue, leading to blockages. Fiber length had a more significant impact on the blockage rate than mass concentration. Full article

In fourth-generation district heating systems (DHSs), the supply temperature of modern heat sources such as heat pumps and waste heat can potentially be reduced by mixing in hot water from combustion-based producers, thereby increasing efficiency and facilitating integration into networks with unrenovated buildings. However, this approach introduces the risk of thermo-hydraulic oscillations driven by mixing dynamics, transport delays, and mass flow adjustments by consumers. These oscillations can increase wear and cost and may potentially lead to system failure. This study addresses the asymptotic stability of multi-supplier DHSs by combining theoretical analysis and practical validation. Through linearization and Laplace transformation, we derive the transfer function of a system with two suppliers. Using pole-zero analysis, we show that transport delay can cause instability. We identify a new control law, demonstrating that persisting oscillations depend on network temperatures and low thermal inertia and enabling stabilization through careful temperature selection, thorough choice of the slack supplier, or installation of buffer tanks. We validate our findings using dynamic simulations of a nonlinear delayed system in Modelica, highlighting the applicability of such systems to real-world DHSs. These results provide actionable insights for designing robust DHSs and mitigating challenges in multi-supplier configurations by relying on thoughtful system design rather than complex control strategies. Full article

Information and communication technology (ICT) components, especially actuators in automated irrigation systems, are essential for managing precise irrigation and optimal soil moisture, enhancing orchard growth and yield. However, actuator malfunctions can lead to inefficient irrigation, resulting in water imbalances that impact crop health and reduce productivity. The objective of this study was to develop a signal processing technique to detect potential malfunctions based on the power consumption level and operating status of actuators for an automated orchard irrigation system. A demonstration orchard with four apple trees was set up in a 3 m × 3 m soil test bench inside a greenhouse, divided into two sections to enable independent irrigation schedules and management. The irrigation system consisted of a single pump and two solenoid valves controlled by a Python-programmed microcontroller. The microcontroller managed the pump cycling ‘On’ and ‘Off’ states every 60 s and solenoid valves while storing and transmitting sensor data to a smartphone application for remote monitoring. Commercial current sensors measured actuator power consumption, enabling the identification of normal and abnormal operations by applying threshold values to distinguish activation and deactivation states. Analysis of power consumption, control commands, and operating states effectively detected actuator operations, confirming reliability in identifying pump and solenoid valve failures. For the second solenoid valve in channel 2, with 333 actual instances of normal operation and 60 actual instances of abnormal operation, the model accurately detected 316 normal and 58 abnormal instances. The proposed method achieved a mean average precision of 99.9% for detecting abnormal control operation of the pump and solenoid valve of channel 1 and a precision of 99.7% for the solenoid valve of channel 2. The proposed approach effectively detects actuator malfunctions, demonstrating the potential to enhance irrigation management and crop productivity. Future research will integrate advanced machine learning with signal processing to improve fault detection accuracy and evaluate the scalability and adaptability of the system for larger orchards and diverse agricultural applications. Full article

To investigate the mechanisms underlying the continuous failure of deep foundation pits in tropical water-rich sandy strata, this study comprehensively examines a foundation pit project in Haikou city, China. Using the PLAXIS^3D 24.1 software, a three-dimensional finite element numerical model was developed. The analysis integrates design schemes, field investigations, monitoring data, and other relevant information to elucidate the mechanisms of disaster damage, such as foundation pit water inrush, floor collapse, and sidewall failure. The results indicate that the water barrier layer is the thinnest at the elevator shaft foundation pit, with a rapid shortening of seepage paths following the extraction of steel sheet piles; the seepage velocity increases by approximately 120%, leading to groundwater breaching both the water barrier and cushion layers. The inadequate length of the suspended impervious curtain in the confined aquifer results in a maximum seepage velocity at the defect site that is 40 times greater than that at other locations, facilitating groundwater influx into the foundation pit. As the excavation deepens, significant alterations occur in the groundwater seepage field at the defect location in the water-resisting curtain, with the seepage velocity increasing from 6.4 mm/day outside the pit to 78.8 mm/day inside the pit, thereby threatening the stability of the pit foundation. Additionally, construction quality defects arising from the three-axis mixing method in the silty sand layer cause a downward shift in the maximum horizontal displacement of the supporting structure, with displacement increments near the defects reaching 63%. Unreasonable emergency pumping measures can lead to floor collapses and sidewall damage. The soil in the pit significantly affects the back pressure, but it is also affected by the distance, and the increase in seepage velocity in the elevator shaft remains under 1% and does not significantly impact the damaging incident. Full article

Rainwater pumping stations located near urban centers or agricultural areas help prevent flooding by activating an appropriate number of pumps with varying capacities based on real-time rainwater inflow. However, relying solely on rule-based pump operations that monitor only basin water levels is often insufficient for effective control. In addition to maintaining a low maximum water level to prevent flooding, pump operation at rainwater stations also requires minimizing the number of pump on/off switches. Reducing pump switch frequency lowers the likelihood of mechanical failure and thus decreases maintenance costs. This paper proposes a real-time pump operation method for rainwater pumping stations using Deep Reinforcement Learning (DRL) to meet these operational requirements simultaneously, based only on currently observable information such as rainfall, inflow, storage volume, basin water level, and outflow. Simulated rainfall data with various return periods and durations were generated using the Huff method to train the model. The Storm Water Management Model (SWMM), configured to simulate the Gasan rainwater pumping station located in Geumcheon-gu, Seoul, South Korea, was used to conduct experiments. The performance of the proposed DRL model was then compared with that of the rule-based pump operation currently used at the station. Full article

This paper presents a comparative analysis of deep learning techniques for anomaly detection and failure prediction. We explore various deep learning architectures on an IoT dataset, including recurrent neural networks (RNNs, LSTMs and GRUs), convolutional neural networks (CNNs) and transformers, to assess their effectiveness in anomaly detection and failure prediction. It was found that the hybrid transformer-GRU configuration delivers the highest accuracy, albeit at the cost of requiring the longest computational time for training. Furthermore, we employ explainability techniques to elucidate the decision-making processes of these black box models and evaluate their behaviour. By analysing the inner workings of the models, we aim at providing insights into the factors influencing failure predictions. Through comprehensive experimentation and analysis on sensor data collected from a water pump, this study contributes to the understanding of deep learning methodologies for anomaly detection and failure prediction and underscores the importance of model interpretability in critical applications such as prognostics and health management. Additionally, we specify the architecture for deploying these models in a real environment using the RAI4.0 metamodel, meant for designing, configuring and automatically deploying distributed stream-based industrial applications. Our findings will offer valuable guidance for practitioners seeking to deploy deep learning techniques effectively in predictive maintenance systems, facilitating informed decision-making and enhancing reliability and efficiency in industrial operations. Full article

The circulation pump in a distillation column is a core device in a material circulation system, and its stable operation is crucial for the production process. The impeller of the circulation pump is prone to failure due to long-term contact with corrosive media, and subjected to a large amount of material erosion, which severely challenges the safety control of the distillation reaction system. Focusing on the corrosion failure phenomenon of circulation pump impellers, the failure mechanism was studied by means of macroscopic inspection, chemical composition analysis, metallographic examination, scanning electron microscopy (SEM), and energy dispersive spectrometer (EDS). Results indicated that the corrosion of circulation pump impellers was the result of the combined effects of surface wear, cavitation, and halogen element corrosion. The medium in contact with the impeller contained chloride ions, fluoride ions, and solid particles. During circulation pump operation, a low-pressure zone formed at the inlet, generating numerous water vapor bubbles. These bubbles burst in the high-pressure zone, creating highly localized impact forces. Combined with the abrasive action of solid particles on the impeller surface, this led to the destruction of the passivation film and the formation of numerous small pits. These corrosion pits and the surrounding environment formed micro-galvanic corrosion cells with small anodes and large cathodes. Under the accelerated corrosion caused by fluoride and chloride ions, the corrosion process towards the inner wall of the impeller intensified, ultimately leading to impeller failure. This study clarified the corrosion failure mechanism and its root causes in the 2507 duplex stainless steel circulation pump impeller and proposes corresponding improvement recommendations, providing a scientific basis for preventing similar issues from occurring in the future. Full article

The mud pump water sealing system (MPWSS) is important in the efficient operation and prolonged service life of the cutter suction dredger’s (CSD) mud pump. Considering that the underwater pump operates underwater and the shaft seal water pressure sensor is prone to failure, a hybrid deep learning model MCNN transformer is proposed to predict the underwater pump shaft seal water pressure in the event of sensor failure. This paper uses big data from the dredging project to deeply excavate the relationship between the shaft end sealing water pressure and other construction data by combining experience and artificial intelligence, and then uses multi-scale convolutional neural network (MCNN) to reconstruct the data, highlighting the time series characteristics of the multi-scale data were then input into the transformer model for prediction, and compared with a single MCNN, transformer model and four other neural networks. Finally, the cutter suction dredger “Hua An Long” was selected as an application research case; experimental comparisons were conducted on seven different models to verify the accuracy and applicability of the MCNN-transformer model. Full article

Future global sustainability depends heavily on the development of renewable energy. The object of this study is a system of two plants (Kaunas hydropower plant (HP) and Kruonis pumped-storage hydropower plant) and upper and lower reservoirs. A possible dam failure accident in such an important system can endanger the population of Kaunas City. The methodology for estimating dam-failure-induced flood wave uncertainty included scenarios of the upper reservoir embankment failure hydrographs, modeling flood wave spreading (MIKE 21 hydrodynamic model), and estimating wave heights. The GRS methodology was selected to assess the uncertainty of flood wave modeling results and the sensitivity of hydrodynamic model parameters. The findings revealed that the discharge values of the Nemunas inflow and outflow through the HP outlets are the most important parameters determining the greatest height of the flood wave. Therefore, by correctly managing the amount of water in the upper reservoir, it would be possible to prevent the lower reservoir dam from breaking. Full article

The centrifugal pump is the workhorse of many industrial and domestic applications, such as water supply, wastewater treatment and heating. While modern pumps are reliable, their unexpected failures may jeopardise safety or lead to significant financial losses. Consequently, there is a strong demand for early fault diagnosis, detection and predictive monitoring systems. Most prior work on machine learning-based centrifugal pump fault detection is based on either synthetic data, simulations or data from test rigs in controlled laboratory conditions. In this research, we attempted to detect centrifugal pump faults using data collected from real operational pumps deployed in various places in collaboration with a specialist pump engineering company. The detection was done by the binary classification of visual features of DQ/Concordia patterns with residual networks. Besides using a real dataset, this study employed transfer learning from the image detection domain to systematically solve a real-life problem in the engineering domain. By feeding DQ image data into a popular and high-performance residual network (e.g., ResNet-34), the proposed approach achieved up to 85.51% classification accuracy. Full article