In order to investigate the influence of structural parameters on the flow characteristics in the ejector and to evaluate the performance of the ejector heat pump system under different working conditions, this study constructed a three-dimensional model of the ejector and a mathematical model of the ejector heat pump system, investigated the influence of geometrical parameters, such as the ejector throat diameter (D_t) and the throat-orifice position (Nozzle Exit Position, NXP), on the performance of the ejector and analyzed the influence of the different working conditions (The effects of different operating conditions (evaporating temperature T_e, condensing temperature T_c, and generating temperature T_g) on the system performance and injector characteristics were analyzed. It is found that increasing D_t can significantly improve the pressure rise ratio of the injector, but it will lead to the reduction of the elicitation coefficient, and the pressure rise ratio decreases with the increase of NXP, and the elicitation coefficient reaches the peak value at NXP=7 mm. When T_c is 40 ℃ and T_g is 70 ℃, the system cooling coefficient (Coefficient of Performance (COP)) increases from 0.94 to 2.64 as T_e increases from 0 ℃ to 15 ℃, the pressure rise ratio of the injector decreases from 1.61 to 1.31, and the priming coefficient increases from 0.51 to 0.89. Among them, the optimization of D_t and NXP can significantly improve the pressure-to-lift ratio and the elicitation efficiency, while the reasonable matching of T_e, T_g and T_c plays a decisive role in the system performance.

The intermittence and volatility of photovoltaic power generation threaten the stability of the power grid, and high-precision prediction can provide key support for power system scheduling. Solar irradiance prediction methods based on meteorological data and sky images have been widely studied, but the mining of multi-scale temporal features and sky image features is still very limited, and the correlation between features is not fully utilized. Therefore, this paper proposes an ultra-short-term solar irradiance prediction model based on multi-source data and Transformer framework. First, the long-term and short-term memory neural network-Transformer is used to learn the long-term and short-term temporal variation characteristics of multivariate meteorological sequences. Secondly, the convolution visual Transformer is used to model the global information and local structure of the sky image, and the deep image features are extracted. Finally, a cross-modal feature fusion module based on attention mechanism is proposed, which uses image features to represent temporal features, realizes cross-modal feature fusion, and outputs prediction results by decoder. Experimental results show that the proposed method has better prediction performance than the state-of-the-art methods. Compared with the case of only using meteorological data, the NRMSE of the model is reduced by 7.4% on average. The model significantly improves the prediction accuracy by mining the features of multi-source data (meteorological time series and sky images) and using the cross-modal attention mechanism to effectively mine their correlation.

In response to the weak generalization of cross-domain fault diagnosis for rolling bearings in wind turbine units, a Wind Turbine Bearing Fault Diagnosis Method via Combined Features Adaptation is proposed. Firstly, a data selection method based on one-dimensional convolutional neural networks is introduced to filter data from the source domain that is similar to the target domain, forming a new source domain to reconstruct the training dataset and mitigate the negative transfer effect. Secondly, the continuous wavelet transform is applied to obtain two-dimension time-frequency features of the training data, these features are afforded for the residual neural network, which generates one-dimensional feature vectors as the initial features for transfer learning, ensuring comprehensive extraction of the training data's features. Thirdly, a new combined feature adaptation loss function is presented. In it, not only the feature transfer loss of the source domain and the target domain, but also the classification loss of the source domain is considered. This enables effective feature transfer from the source domain to the target domain. Finally, the softmax classifier of the constructed residual neural network is used for the diagnosis classification of the fault. By the simulation under different kinds of migration tasks between source domain and target domain, the proposed model is of higher diagnosis performance in the cross-domain environment.

For large floating wind turbines, the influence of blade deformation on wind turbine performance cannot be ignored. Based on the free wake model of lift line and the geometrically accurate beam model, the effects of aerodynamic coupling effects on the load characteristics of EI-15 MW wind turbine under rolling conditions are studied in this paper. The results show that the additional velocity brought by the rolling motion of the floating platform will change the inflow characteristics at the section of the airfoil, so that the wind turbine load will fluctuate, and the wind turbine load will decrease after considering the flexible deformation of the blade. Compared with the fixed condition, the asymmetry of the load distribution on the rotor plane is more obvious in the rolling condition, which leads to the fluctuation amplitude of the yaw and pitching moment of the rotor. In addition, the shape variable of the blade will fluctuate periodically under rolling conditions, and the shape variable gradually increases from blade root to blade tip. The effective Angle of attack decreases due to torsional deformation, and the Angle of attack of the flexible blade at the same profile section is smaller than that of the rigid blade, which is the direct reason why the load of the flexible blade is lower than that of the rigid blade.

Windmills are important facilities that convert wind energy into electrical energy. This paper addresses the issues of missed detections and false detections in the process of identifying small-scale windmills in remote sensing images by constructing a new GF-2 windmill dataset. It proposes the YOLO-WM algorithm for the small target detection task of windmills, adding a tiny object detection layer based on the YOLOv8 object detection algorithm, and incorporates a BSAM dynamic attention mechanism module in the backbone section. It replaces the C2f module with the lightweight C2f_RepGhost module and finally utilizes the SAHI slicing algorithm in the prediction stage to complete the windmill target detection task. Experimental results show that the YOLO-WM algorithm can better retain shallow convolutional features while reducing algorithm complexity, allowing for more accurate capture of small-scale target information. This method compensates for the shortcomings of small object detection datasets, and it has been verified that the improved algorithm performs better in detecting small-scale windmills than other algorithms. This research provides valuable insights for the rational optimization and allocation of wind energy resources and the joint analysis of geographical elements.

Due to the current-magnitude limit, controlled phase angle, and absence of zero-sequence current of converters, fault-location errors occur when two-phase short-circuit faults happen in offshore wind-farm collector cables. This paper analyzes the composite-sequence network under such faults and exploits the capability of both the VSC-HVDC link and the wind-turbine converters to suppress negative-sequence current during the fault. By using the locally measured negative-sequence voltage, the positive-sequence voltage at the fault point is computed on-line, a fault-distance equation is established, and the fault position is solved in real time. The proposed on-line algorithm remains accurate for large and non-resistive fault resistances and is immune to the reactive-power support strategy of the offshore wind farm, thus enhancing the safe and stable operation of offshore wind power systems.

Diode rectifier unit (DRU) based offshore wind power delivery scheme can reduce the volume and weight of offshore platforms and thus save a lot of operation and maintenance costs. However, the DRU does not have commutation controllability, and its flexibility is poor when power fluctuations occur in the wind farm. This paper first establishes a topological model of a wind farm composed of three-phase permanent magnet synchronous generators connected in parallel with a DRU and an auxiliary voltage-source converter (VSC) for power transmission.) Secondly, the grid-froming control strategy of the grid side converter and the grid-following control strategy of VSC are adopted respectively. Finally, a RT-LAB semi-physical simulation system is built based on Matab/simulink. The results show that the proposed control strategy realizes the auxiliary VSC control of reactive power and DRU output power of wind turbines.

With the large-scale access of wind turbines with frequency regulation functions to the power system, the influence of frequency regulation control on the transient power angle stability of the system is increasingly prominent. In this paper, an extended two-machine system with a wind turbine connected to the sending end area and a three-phase short-circuit fault at the intermediate node of the tie line is taken as the research background. Aiming at the influence mechanism of the droop control of the wind turbine on the stability of the first swing of the power angle of the system, firstly, based on the equal area rule, the relationship between the speed difference of the synchronous machine and the transient energy of the system is established, and the WDPAS index, including the influence factors such as the output of the wind turbine, the grid-connected position, and the inertia distribution of the system, is proposed to analyze the influence of the droop control of the wind turbine on the speed difference of the synchronous machine. Then, based on the system state of the maximum power point tracking (MPPT) control of the wind turbine, the dynamic changes of the system transient energy during the fault and after the fault clearance are studied. Finally, the first swing stability of the system power angle is evaluated by the maximum offset of the system power angle first pendulum, and an extended two-machine system simulation model is built in the DIgSILENT/PowerFactory simulation software to verify the rationality of the index and the correctness of the conclusion.

More and more renewable energy is incorporated into the power system through power electronic equipment, resulting in reduced system strength and physical inertia. This paper proposes a grid-forming control strategy for modular multilevel converter based high voltage direct current transmission system (MMC-HVDC) with MMC internal energy control which is connected with a weak AC system. Firstly, based on the droop grid-forming control,, inertia link and PI control are added, so that MMC can automatically track the load fluctuation and adjust its own power output, and realize zero steady-state error regulation of AC voltage and frequency. Then, aiming at the problem of internal energy fluctuation of MMC controlled by grid-forming control, a MMC internal energy compensation controller is designed, which can effectively reduce fluctuation of DC voltage of submodule capacitance. Finally, based on the PSCAD/EMTDC electromagnetic transient simulation platform, a MMC-HVDC transmission system was established, and two operating conditions of AC grid load increase and AC grid load reduction are selected to verify the feasibility and effectiveness of the proposed control strategy.

To address the power fluctuations in the distribution network caused by the integration of distributed new energy, this article proposes an economic dispatch strategy for the distribution network that considers the frequency regulation capability and source load correlation of electric vehicle battery swapping stations (BSS). This strategy aims to minimize the total system cost, coordinate and schedule frequency modulation resources including BSS, and constrain BSS charging and discharging behavior while maintaining power balance. Firstly, a frequency modulation signal prediction model is constructed: a long short-term memory neural network is used to predict new energy and load power, and the Copula function is used to characterize their correlation, deriving the conditional probability distribution of the frequency modulation signal. Secondly, a prediction driven economic dispatch method based on source load correlation is proposed: Wasserstein distance is used to process the predicted data and construct a fuzzy set, combined with distribution robust optimization and opportunity constraints to establish an economic dispatch model considering the cost of secondary frequency modulation; Further transform the opportunity constraint through conditional value at risk (CVaR) approximation, and use Wasserstein duality to transform the model into a linear programming problem for solution. The simulation results show that the proposed strategy is economically superior to traditional methods.