The transportation via high-pressure long-tube trailers serves as a crucial method for medium-short distance transfer of flammable and explosive gases such as hydrogen and natural gas. As the core equipment in this system, the trailer filling compressor operates under continuously varying discharge pressures across wide ranges during gas loading processes. Current research and development phases face challenges in fully replicating real-world operating conditions for thermal performance testing. Addressing this requirement, this study proposes a closed-loop experimental system with gas staged recovery and continuous release functions, featuring continuous backpressure regulation capability for trailer filling compressors. The system enables the simulation of actual filling processes by creating both stable and dynamic operating conditions with wide-ranging discharge pressure variations for comprehensive compressor testing. A mathematical model of the testing system was established using zero-dimensional simulation methodology. Systematic investigations were conducted on parameter variations and operational characteristics throughout complete testing procedures, including initial pressurization, compressor startup/shutdown, and wide-range operational testing. Through optimized improvements in system configuration and component matching, critical operational constraints were achieved: gas reservoir temperatures were maintained below 85 °C during testing, and post-recovery system pressures were reduced below 1.5 MPa. These optimizations resulted in the development of a refined and rational testing system and methodology for trailer filling compressors, effectively addressing the technical challenges in simulating actual working conditions during compressor development phases.

Under the background of the energy transition, energy storage, as an important technical means to support a high proportion of renewable energy access and consumption, has gradually become an indispensable part of virtual power plants. Among them, energy storage selection is a key issue to ensure the safe and stable operation of the power grid and improve the scheduling efficiency of virtual power plants. Therefore, an energy storage selection method based on game combination weighting and GRA-MARCOS was proposed for virtual power plants. Firstly, the evaluation indicator system of energy storage adaptability was constructed on the basis of considering the technical, economic, security, and environmental protection indicators of energy storage. Secondly, based on game theory, the comprehensive subjective and objective weights of the indicators were obtained by combining the subjective weights from FAHP(fuzzy analytic hierarchy process) method with the objective weights derived from the CRITIC and MEREC methods. Finally, the utility function of each alternative energy storage technology was calculated using the MARCOS method improved by GRA(grey relational analysis), and this was used to rank and make selection decisions for energy storage. The validity and robustness of the proposed energy storage selection method were verified through examples and sensitivity analysis. This method provides a reasonable and effective decision-making scheme for energy storage selection in different scenarios of virtual power plants.

Power plants utilizing substandard coal for electricity generation often face reduced denitrification efficiency due to high fly ash concentrations in the flue gas, which complicates achieving ultra-low emissions. The flue gas duct from the economizer outlet to the riser duct entrance in the W-shaped flame boiler at Guizhou Chayuan Power Plant was investigated. Numerical simulations were conducted to analyze flue gas flow and fly ash particle trajectories. The results show that the economizer ash hopper is minimally effective, with a collection rate of only 9.99%. Simulation results reveale fly ash deposition on the wall at the SCR riser flue bend at 1.12 kg/s, representing 11.33% of total fly ash. Based on these findings, two design schemes for positioning a denitrification ash hopper below the riser flue are proposed to enhance fly ash collection efficiency. Optimal results are obtained with a 45° angle between the inclined surface and horizontal wall, increasing fly ash collection by 8.38% with a 2.25 Pa pressure drop increase.

The technological demand for distributed wind energy and wind-solar complementary energy utilization on building roofs was addressed. A new type of spatial support frame for wind-solar complementary systems was proposed. The power generation efficiency of small vertical axis wind turbines was enhanced by using flow deflectors with combined wind collection and flow stabilization functions. The wind collection effect and power generation efficiency of the framework were analyzed through theoretical methods. Numerical simulations were conducted to study the impact of flow deflector distancing and width on internal airflow velocity and turbulent kinetic energy at different wind attack angles. Optimization parameters were identified. Wind tunnel experiments were performed to investigate the power generation performance of small wind turbines with the spatial support framework. The results show that the framework significantly increases the airflow velocity entering the wind collection device and reduces turbulent kinetic energy in the internal space. When the flow deflectors have a distance of 0.53 m and a width of 0.12 m, the wind speed increases by 1.21 times and the generator power increases by about 1.77 times.

In recent years, the scale of wind turbine grid connection has been increasing, for the deep learning of wind speed prediction requires a large amount of data, as well as stochastic differential equations for wind power system modeling fail to portray the impact of wind speed correlation on the output power and grid-connection point voltage, a Markov switching stochastic differential equation modeling method considering stochastic factors and wind speed correlation was proposed for power systems containing wind power. The Nataf and LSTM were introduced to construct the wind speed spatio-temporal correlation model, the Markov switching stochastic differential equation was used to segment and linearize the wind power system into various linear segments. Then the effects of wind speed correlation and stochastic excitation strength on the voltage at the grid-connection point were studied, and the critical stable excitation strength of the wind power system was analyzed. Finally, the stochastic simulation of the constructed system model was carried out by numerical analysis methods, and the results show that the system state variable fluctuates in the stable region within the critical value of the random excitation intensity, and the comparison with the stable waveform of voltage in the Simulink simulation circuit verifies the validity of the modeling method in this paper, and provides a theoretical basis for the stability analysis of the new wind farm access to the power system.

The hybrid DC transmission system has problems such as inconsistent boundary components, inconsistent fault response characteristics, difficult resolution of high resistance fault effects, and low accuracy in identifying near end faults, which reduces the reliability of protection schemes. Therefore, the phase characteristics of the regional refractive index of the hybrid DC transmission system were analyzed for the first time, and a single ended protection scheme suitable for hybrid boundaries was proposed based on this. Firstly, establish a hybrid DC transmission system model and analyze the traveling wave transmission characteristics of different fault types. Subsequently, the fault areas of the hybrid DC transmission system were divided, and the refractive index expressions and phase frequency characteristics of the areas were derived separately. Finally, a single ended protection scheme based on a specific frequency refractive index is proposed and its performance is tested. The test results show that the proposed protection scheme not only has the speed of traditional protection schemes, but also has better resistance to high impedance faults, noise interference, and other abilities.

With the rapid development of the energy industry and technological innovation, a large number of professional terms and expressions are constantly updated, and new words continue to emerge. However, traditional neologism discovery methods often rely on dictionaries or rules, and it is difficult to efficiently process and update a large number of specialized terms, especially in the rapidly changing energy field. Therefore, combined with the characteristics of text data in the energy field, a new word discovery method in ENFM(energy field combining N-Gram and multiple attention mechanism) was proposed. Firstly, the N-Gram model was used to process the text data in the field of energy, and the candidate list of new words was generated by statistics and analysis of word frequency. Subsequently, the ERNIE-BiLSTM-CRF model integrating multiple attention mechanism was introduced to further improve the accuracy and efficiency of neologism discovery. Compared with the traditional neologism discovery technology, the accurate identification and overall efficiency of neologism have been significantly improved. The accuracy rate, recall rate and F₁ value of neologism in the data set of policy text in the energy field are 95.71%, 95.56% and 95.63%, respectively. The experimental results show that this method can accurately identify new words in a large number of text data in the field of energy, effectively identify the specific words and expressions in the field of energy, and significantly improve the recognition ability of professional terms in the field of energy in Chinese word segmentation tasks.

To address the limitations of traditional protocol recognition methods caused by the presence of numerous non-standard protocols in IC (industrial control) sector, a method based on edge-distributed deep learning was studied to enhance IC protocol recognition technology. A recognition method based on CNN (convolutional neural networks) was proposed: real IC protocol data from the network was collected and preprocessed, and an appropriate CNN model was selected according to protocol characteristics to implicitly extract the essential features of the protocols. This achieved classification and recognition of seven types of IC protocols with an accuracy of up to 99.92%. Furthermore, the IC protocol recognition model was deployed at the network edge, leveraging a data-parallel distributed strategy for collaborative training within an edge server computing cluster. This improved the training efficiency of the model by 1.87~2.81 times while maintaining high accuracy. The results show that this method significantly improves the accuracy of IC protocol recognition, greatly enhances model training efficiency, and is well-suited for deployment in edge computing environments. It is evident that this method has significant value in optimizing IC protocol recognition performance.

sEMG (surface electromyography) signals are physiological signal closely related to human movement, and the analysis of sEMG signals play an important role in the field of human-machine interaction. Aiming at the difficulty of both efficiency and accuracy of electromyographic signal classification, an upper limb sEMG classification method was innovatively proposed, which combined feature screening with classifier hyperparameter optimization. BPSO (binary particle swarm optimization) algorithm was adopted to screen the features. PSO (particle swarm optimization) algorithm was further utilized to adjust the hyperparameters of the LSSVM (least-squares support vector machine). By collecting sEMG signals from four parts of the human upper body and extracting 48-dimensional features from them, classification experiments were conducted on four common movements of upper limb. The results show that the BPSO-PSO-LSSVM algorithm retains only the 21-dimensional features of the EMG data, and the average classification accuracy obtained reaches 97.54%. It is proved that this method can effectively screen out the optimal combination of features for upper limb motion classification and improve the accuracy of movement classification.

In order to solve the problem of insufficient feature extraction of human dynamic skeleton features in abnormal behavior recognition, an unsupervised abnormal behavior recognition method based on enhanced spatiotemporal graph normalization flow was proposed. Transformer and convolution block attention module were employed to enhance the feature expression capability of the model and the performance of the abnormal behavior recognition algorithm in the global and spatiotemporal domains. Firstly, the Transformer module was incorporated into the affine layer of the normalized flow to augment the efficacy of dynamic skeleton feature information at the global level. Subsequently, the convolution attention was introduced into the convolution module of space and time graphs respectively to effectively enhance the spatial and temporal representation of dynamic skeleton features. Finally, simulation verification was conducted on the ShanghaiTech and UBnormal datasets, and the recognition accuracy attains 86.4% and 70.2% respectively, thereby demonstrating the effectiveness of the method.