Search Results (20)

A reinforcement learning control method for a solid attitude and divert propulsion system is proposed. The system in this study includes four divert thrust nozzles, six attitude thrust nozzles, and a common combustion chamber. To achieve the required thrust, the pressure in the combustion chamber is first adjusted by controlling the total opening of the nozzles to generate the gas source. Next, by controlling the opening of nozzles at different positions, the required thrust is produced in the five-axis direction. Finally, the motor speed is regulated to drive the valve core to the specified position, completing the closed-loop control of the nozzle opening. The control algorithm used is the Proximal Policy Optimization (PPO) reinforcement learning algorithm. Through system identification and numerical modeling, the training environment for the intelligent agent is created. To accommodate different training objectives, multiple reward functions are implemented. Ultimately, through training, a multi-layer intelligent agent architecture for pressure, thrust, and nozzle opening is established, achieving effective system pressure and thrust control. Full article

The Make-up Air Unit (MAU) is an air conditioning system which plays an important role in serving semiconductor cleanrooms. It provides constant temperature and humidity for fresh air through various sections, including fresh air filtration, preheating, precooling, humidification, recooling, reheating, air supply, and high-efficiency filtration. However, the commonly used PID control method of the MAU indicates a deficiency in energy consumption. Hence, this research introduces a proactive energy-saving optimization control method based on machine learning and intelligent optimization algorithms. Firstly, the machine learning methods are used to train historical data of the MAU, resulting in a data-driven prediction model of energy consumption for the system. Subsequently, the customized genetic algorithm (GA) is used to optimize energy in cold and hot water systems. It facilitates the dynamic adjustment of the regulating valve opening for the cold and hot water coil in the fresh air unit, responding to real-time variations in outdoor air conditions. Meanwhile, it ensures that the supply air temperature and humidification adhere to specified requirements, thereby reducing the energy consumption associated with cold and hot water usage in the MAU. The experimental results indicate that the proposed algorithm can provide significant energy conservation in the MAU. Full article

Short-circuit blowing is a crucial technical approach for ensuring the rapid surfacing of submersibles. In order to investigate the law, L₁₈(3⁷) orthogonal experiments based on a proportional short-circuit blowing model test bench were conducted. Subsequently, a Back Propagation Neural Network (BPNN) and Pearson correlation analysis were employed to train the experimental data; further examination of the correlation between individual factors and blowing served as an enhancement to the orthogonal experiments. It has been proved that both multi-factor combinations and personal factors, including blowing duration, sea tank back pressure, gas blowing pressure from the cylinder group, and sea valve flowing area, exert significant influence with Pearson correlation coefficients of 0.6535, 0.8105, 0.5569, and 0.5373, respectively. Notably, the F-ratio for blowing duration exceeds the critical value of 3.24. The statistical evaluation metrics for the BPNN ranged from 10⁻¹ to 10⁻¹², with relative errors below 3%, and achieving a prediction accuracy rate of 100%. Based on these findings, a robust predictive methodology for submersible short-circuit blowing has been established along with recommendations for engineering design and operational strategies that highlight its advantages as well as its initial condition settings. Full article

As an electric multiple unit (EMU) train passes through an extreme tunnel characterized by high altitude, steep gradient, and extended lengths, the pressure waves generated by the train–tunnel aerodynamic coupling combine with the baseline pressure variations within the tunnel. This interaction results in rapid fluctuations and extreme external pressure with higher amplitudes, which are transmitted into the carriage, causing pressure fluctuations that can adversely affect passenger comfort. These waves interact with multiple factors within the carriage, such as air ducts, airtight gaps, carbody deformation, oxygen supply systems, and temperature, creating a highly nonlinear internal pressure transmission system. This study first establishes a single-factor internal pressure fluctuation model. Subsequently, a multi-factor coupled internal pressure fluctuation model is constructed based on the ideal gas polytropic process assumption and the law of mass conservation. The model parameters are corrected and the model’s effectiveness and accuracy are validated using experimental data to predict and summarize the internal pressure variation patterns of the EMU train during dynamic operation in such tunnels, ensuring safe train operation and meeting the pressure comfort requirements of passengers. Finally, to address the challenges of maintaining and regulating multi-physical variable comfort under extreme tunnel conditions, this study investigates the impact of partial oxygen pressure and temperature on pressure fluctuations and comfort. The study finds that higher oxygen pressure and temperature significantly increase internal pressure fluctuation amplitude, with the oxygen supply system contributing 18.11% and temperature 5.74% of total variation. Thus, setting appropriate standards for oxygen supply, temperature, and valve operation is crucial for mitigating internal pressure fluctuations and enhancing safety and comfort. This research provides a theoretical foundation for developing a comprehensive comfort evaluation and regulation system under harsh environments. Full article

Flow processes onboard ships are common in order to transport fluids like oil, gas, and water. These processes are controlled by PID controllers, acting on the regulation valves as actuators. In case of a malfunction or refitting, a PID controller needs to be re-adjusted for the optimal control of the process. To avoid experimenting on operational real systems, models are convenient alternatives. When real-time information is needed, digital twin (DT) concepts become highly valuable. The aim of this paper is to analyze and determine the optimal NARX model architecture in order to achieve a higher-accuracy model of a ship’s flow process. An artificial neural network (ANN) was used to model the process in MATLAB. The experiments were performed using a multi-start approach to prevent overtraining. To prove the thesis, statistical analysis of the experimental results was performed. Models were evaluated for generalization using mean squared error (MSE), best fit, and goodness of fit (GoF) measures on two independent datasets. The results indicate the correlation between the number of input delays and the performance of the model. A permuted k-fold cross-validation analysis was used to determine the optimal number of voltage and flow delays, thus defining the number of model inputs. Permutations of training, test, and validation datasets were applied to examine bias due to the data arrangement during training. Full article

The last decade has seen the development and application of data-driven methods taking off in nuclear engineering research, aiming to improve the safety and reliability of nuclear power. This work focuses on developing a reinforcement learning-based control sequence optimization framework for advanced nuclear systems, which not only aims to enhance flexible operations, promoting the economics of advanced nuclear technology, but also prioritizing safety during normal operation. At its core, the framework allows the sequence of operational actions to be learned and optimized by an agent to facilitate smooth transitions between the modes of operations (i.e., load-following), while ensuring that all safety significant system parameters remain within their respective limits. To generate dynamic system responses, facilitate control strategy development, and demonstrate the effectiveness of the framework, a simulation environment of a pebble-bed high-temperature gas-cooled reactor was utilized. The soft actor-critic algorithm was adopted to train a reinforcement learning agent, which can generate control sequences to maneuver plant power output in the range between 100% and 50% of the nameplate power through sufficient training. It was shown in the performance validation that the agent successfully generated control actions that maintained electrical output within a tight tolerance of 0.5% from the demand while satisfying all safety constraints. During the mode transition, the agent can maintain the reactor outlet temperature within $\pm 1.5$ °C and steam pressure within 0.1 MPa of their setpoints, respectively, by dynamically adjusting control rod positions, control valve openings, and pump speeds. The results demonstrate the effectiveness of the optimization framework and the feasibility of reinforcement learning in designing control strategies for advanced reactor systems. Full article

In this work, an artificial neural network (ANN)-based model is proposed to describe the input–output relationships in a Limaçon-To-Circular (L2C) gas expander with an inlet valve. The L2C gas expander is a type of energy converter that has great potential to be used in organic Rankine cycle (ORC)-based small-scale power plants. The proposed model predicts the different performance indices of a limaçon gas expander for different input pressures, rotor velocities, and valve cutoff angles. A network model is constructed and optimized for different model parameters to achieve the best prediction performance compared to the classic mathematical model of the system. An overall normalized mean square error of 0.0014, coefficient of determination ($R^2$) of 0.98, and mean average error of 0.0114 are reported. This implies that the surrogate model can effectively mimic the actual model with high precision. The model performance is also compared to a linear interpolation (LI) method. It is found that the proposed ANN model predictions are about 96.53% accurate for a given error threshold, compared to about 91.46% accuracy of the LI method. Thus the proposed model can effectively predict different output parameters of a limaçon gas expander such as energy, filling factor, isentropic efficiency, and mass flow for different operating conditions. Of note, the model is only trained by a set of input and target values; thus, the performance of the model is not affected by the internal complex mathematical models of the overall valved-expander system. This neural network-based approach is highly suitable for optimization, as the alternative iterative analysis of the complex analytical model is time-consuming and requires higher computational resources. A similar modeling approach with some modifications could also be utilized to design controllers for these types of systems that are difficult to model mathematically. Full article

This work is devoted to the study of gas-dynamic processes in the operation of climate control systems in the cabins of vehicles (HVAC), focusing on pressure values. This research examines the issue of assessing the required values of air overpressure inside the locomotive cabin, which is necessary to prevent gas exchange between the interior of the cabin and the outside air through leaks in the cabin, including protection against the penetration of harmful substances. The pressure boost in the cabin depends, among other things, on the external air pressure on the locomotive body, the power of the climate system fan, and the ratio of the input and output deflectors. To determine the external air pressure, the problem of train movement in a wind tunnel is considered, the internal and external fluids domain is considered, and the air pressure on the cabin skin is determined using numerical methods CFD based on the Navier–Stokes equations, depending on the speed of movement. The finite-volume modeling package Ansys CFD (Fluent) was used as an implementation. The values of excess internal pressure, which ensures the operation of the climate system under different operating modes, were studied numerically and on the basis of an approximate applied formula. In particular, studies were carried out depending on the speed and movement of transport, on the airflow of the climate system, and on the ratio of the areas of input and output parameters. During a numerical experiment, it was found that for a train speed of 100 km/h, the required excess pressure is 560 kPa, and the most energy-efficient way to increase pressure is to regulate the area of the outlet valves. Full article

In order to reduce carbon emissions, which are currently a problem in the shipping and offshore plant sectors, the international community is strengthening regulations such as the Energy Efficiency Design Index (EEDI) and Energy Efficiency Existing Ship Index (EEXI). To cope with this, eco-friendly fuel propulsion technology is being developed, and the development of an ammonia fuel supply system is in progress. Among them, fuel valve train (FVT) technology was researched for the final supply and cutoff of fuel and purging through nitrogen for ammonia engines. In this paper, we analyzed the change in ammonia supply due to FVT opening and the change in nitrogen supply due to closure. In addition, a plan to minimize risk factors was presented by applying a control method to remove residual fuel in FVT. According to the presented FVT model, the difference in the flow rate of supplied fuel was as much as 17.8 kg/s. Additionally, by opening the gas bleed valve at intervals during the closing process and purging about 0.28 kg of nitrogen, the internal fuel could be completely discharged. This is expected to have an impact on improving the marine environment through the application of eco-friendly fuels and the development of fuel supply system technology. Full article

Screening methods available for colorectal cancer (CRC) to date are burdened by poor reliability and low patient adherence and compliance. An altered pattern of volatile organic compounds (VOCs) in exhaled breath has been proposed as a non-invasive potential diagnostic tool for distinguishing CRC patients from healthy controls (HC). The aim of this study was to evaluate the reliability of an innovative portable device containing a micro-gas chromatograph in enabling rapid, on-site CRC diagnosis through analysis of patients’ exhaled breath. In this prospective trial, breath samples were collected in a tertiary referral center of colorectal surgery, and analysis of the chromatograms was performed by the Biomedical Engineering Department. The breath of patients with CRC and HC was collected into Tedlar bags through a Nafion filter and mouthpiece with a one-way valve. The breath samples were analyzed by an automated portable gas chromatography device. Relevant volatile biomarkers and discriminant chromatographic peaks were identified through machine learning, linear discriminant analysis and principal component analysis. A total of 68 subjects, 36 patients affected by histologically proven CRC with no evidence of metastases and 32 HC with negative colonoscopies, were enrolled. After testing a training set (18 CRC and 18 HC) and a testing set (18 CRC and 14 HC), an overall specificity of 87.5%, sensitivity of 94.4% and accuracy of 91.2% in identifying CRC patients was found based on three VOCs. Breath biopsy may represent a promising non-invasive method of discriminating CRC patients from HC. Full article

Downhole safety valves are essential equipment for oil and gas extraction, and it is crucial to carry out a downhole safety valve failure risk evaluation and reliability analysis to ensure the safety of oil and gas production. In order to improve the operation and maintenance management level of downhole safety valves and explore the key failure risk factors of downhole safety valves, this study firstly carries out a Failure Mode and Criticality Analysis of downhole safety valves; identifies the causes of failure of downhole safety valves and the consequences of accidents through the Bow-tie method; and quantitatively evaluates the failure risk factors based on the improved Decision-making Trial and Evaluation Laboratory method and obtains the influence and importance ranking of 14 types of failure risk factors. Specific preventive measures for key failure risk factors are proposed in several aspects: optimising the structural design of downhole safety valves, improving the processing and manufacturing process, setting up an efficient field management team, carrying out equipment operation and maintenance management training, establishing a field failure response mechanism, and setting up an intelligent O&M management platform for downhole safety valves. The research results of this study are conducive to improving the reliability of downhole safety valves, ensuring the safety and integrity of on-site operation and maintenance management, and providing theoretical guidance for the analysis of the risk of failure and operation and maintenance management of downhole safety valves. Full article

Nowadays, more frequently, it is necessary to perform underwater operations such as surveying an area or inspecting and intervening on industrial infrastructures such as offshore oil and gas rigs or pipeline networks. Recently, the use of Autonomous Underwater Vehicles (AUV) has grown as a way to automate these tasks, reducing risks and execution time. One of the used sensing modalities is vision, providing RGB high-quality information in the mid to low range, making it appropriate for manipulation or detail inspection tasks. This work presents the use of a deep neural network to perform pixel-wise 3D segmentation of pipes and valves on underwater point clouds generated using a stereo pair of cameras. In addition, two novel algorithms are built to extract information from the detected instances, providing pipe vectors, gripping points, the position of structural elements such as elbows or connections, and valve type and orientation. The information extracted on spatially referenced point clouds can be unified to form an information map of an inspected area. Results show outstanding performance on the network segmentation task, achieving a mean F1-score value of 88.0% at a pixel-wise level and of 95.3% at an instance level. The information extraction algorithm also showcased excellent metrics when extracting information from pipe instances and their structural elements and good enough metrics when extracting data from valves. Finally, the neural network and information algorithms are implemented on an AUV and executed in real-time, validating that the output information stream frame rate of 0.72 fps is high enough to perform manipulation tasks and to ensure full seabed coverage during inspection tasks. The used dataset, along with a trained model and the information algorithms, are provided to the scientific community. Full article

The energy transition with transformation into predominantly renewable sources requires technology development to secure power production at all times, despite the intermittent nature of the renewables. Micro gas turbines (MGTs) are small heat and power generation units with fast startup and load-following capability and are thereby suitable backup for the future’s decentralized power generation systems. Due to MGTs’ fuel flexibility, a range of fuels from high-heat to low-heat content could be utilized, with different greenhouse gas generation. Developing micro gas turbines that can operate with carbon-free fuels will guarantee carbon-free power production with zero CO₂ emission and will contribute to the alleviation of the global warming problem. In this paper, the redevelopment of a standard 100-kW micro gas turbine to run with methane/hydrogen blended fuel is presented. Enabling micro gas turbines to run with hydrogen blended fuels has been pursued by researchers for decades. The first micro gas turbine running with pure hydrogen was developed in Stavanger, Norway, and launched in May 2022. This was achieved through a collaboration between the University of Stavanger (UiS) and the German Aerospace Centre (DLR). This paper provides an overview of the project and reports the experimental results from the engine operating with methane/hydrogen blended fuel, with various hydrogen content up to 100%. During the development process, the MGT’s original combustor was replaced with an innovative design to deal with the challenges of burning hydrogen. The fuel train was replaced with a mixing unit, new fuel valves, and an additional controller that enables the required energy input to maintain the maximum power output, independent of the fuel blend specification. This paper presents the test rig setup and the preliminary results of the test campaign, which verifies the capability of the MGT unit to support intermittent renewable generation with minimum greenhouse gas production. Results from the MGT operating with blended methane/hydrogen fuel are provided in the paper. The hydrogen content varied from 50% to 100% (volume-based) and power outputs between 35kW to 100kW were tested. The modifications of the engine, mainly the new combustor, fuel train, valve settings, and controller, resulted in a stable operation of the MGT with NOx emissions below the allowed limits. Running the engine with pure hydrogen at full load has resulted in less than 25 ppm of NOx emissions, with zero carbon-based greenhouse gas production. Full article

Accurate simulations of gas turbines’ dynamic performance are essential for improvements in their practical performance and advancements in sustainable energy production. This paper presents models with extremely accurate simulations for a real dual-fuel gas turbine using two state-of-the-art techniques of neural networks: the dynamic neural network and deep neural network. The dynamic neural network has been realized via a nonlinear autoregressive network with exogenous inputs (NARX) artificial neural network (ANN), and the deep neural network has been based on a convolutional neural network (CNN). The outputs selected for simulations are: the output power, the exhausted temperature and the turbine speed or system frequency, whereas the inputs are the natural gas (NG) control valve, the pilot gas control valve and the compressor variables. The data-sets have been prepared in three essential formats for the training and validation of the networks: normalized data, standardized data and SI units’ data. Rigorous effort has been carried out for wide-range trials regarding tweaking the network structures and hyper-parameters, which leads to highly satisfactory results for both models (overall, the minimum recorded MSE in the training of the MISO NARX was 6.2626 × 10⁻⁹ and the maximum MSE that was recorded for the MISO CNN was 2.9210 × 10⁻⁴, for more than 15 h of GT operation). The results have shown a comparable satisfactory performance for both dynamic NARX ANN and the CNN with a slight superiority of NARX. It can be newly argued that the dynamic ANN is better than the deep learning ANN for the time-based performance simulation of gas turbines (GTs). Full article

Spark-ignited internal combustion engines are known to exhibit a decreased brake efficiency in part-load operation. Similarly to cylinder deactivation, the x-stroke operation presented in this paper is an adjustable form of skip-cycle operation. It is an effective measure to increase the efficiency of an internal combustion engine, which has to be equipped with a variable valve train to enable this feature. This paper presents an optimization procedure for the exhaust valve timings applicable to any valid stroke operation number greater than four. In the first part, the gas spring operation, during which all gas exchange valves are closed, is explained, as well as how it affects the indicated efficiency and the blow-by mass flow. In the second part, a simulation model with variable valve timings, parameterized with measurement data obtained on the engine test, is used to find the optimal valve timings. We show that in 12-stroke operation and with a cylinder load of 5 Nm, an indicated efficiency of 34.3% is achieved. Preloading the gas spring with residual gas prevents oil suction and thus helps to reduce hydrocarbon emissions. Measurements of load variations in 4-, 8-, and 12-stroke operations show that by applying an x-stroke operation, the indicated efficiency remains high and the center of combustion remains optimal in the range of significantly lower torque outputs. Full article