Accurately predicting solar irradiation (SI) is crucial for power scheduling and photovoltaic site selection. With the development of high-performance computing and large-capacity storage devices, data-driven deep learning models have gained widespread attentions in the SI prediction domain. However, the lack of physical interpretability due to the “black-box” nature of deep learning models restricts their credibility in specific scenarios. To enhance the interpretability of the model on the premise of maintaining prediction accuracy and keeping the model structure unchanged, and without increasing computational complexity, a model based on long short-term memory (LSTM) neural network is constructed, demonstrating an 8.07% performance improvement over the conventional neural networks and showing superior outlier handling capabilities. By employing layer-wise relevance propagation (LRP) algorithm, factors influencing the model output are scored from both temporal and spatial dimensions, enhancing the model’s interpretability. The research results indicate that the model possesses good interpretability under the premise of ensuring performance, with historical solar irradiation, time-related features (such as hour, day, week, month), solar altitude information (such as sunrise and sunset times), cloud cover, radiation time, temperature, and dew point temperature being the main factors influencing SI prediction.

In order to realize efficient and accurate detection of the pitch bearing tooth surface of wind turbines in service under the adhesion of high viscosity lubricating grease, profiling array eddy current technology is used to study the effects of different coil arrangement methods and lifting distances on the detection results of bearing tooth surface defects. It can be concluded that, the edge effect range of Z-shaped and composite coil layout is different, and the composite profiling probe is significantly shorter than the Z-shaped probe. By detecting the artificial groove defects of 10.00 mm×0.50 mm×1.00 mm (slot length × slot width × slot depth), it indicates that the lifting distance limits for inner and outer toothed bearings are 1.04 mm and 1.43 mm, respectively. Experimental verification is conducted on the in-service fan bearings under the condition of high viscosity lubricating grease on the surface, and the results of defect size and position detection are accurate, with an error of less than 5%. This provides technical support and new ideas for supervision and inspection of the pitch bearing gear tooth surface in service.

The accurate measurement of solar absorber coating absorption is of great significance to the evaluation and optimization of the absorber performance. At present, the experimental researches of absorber coating absorption are mostly limited to the flat metal substrate. By taking the cylindrical tube of tower solar molten salt absorber, two widely used measurement methods for testing the absorption of curved surface coatings are proposed. The equipment preparation of Blu-Tack method is simple but the operation is complex, and the equipment preparation of diffuse reflection box method is complex but the operation is simple. The two methods isolate the interference of ambient light during the measurement process, and obtain the measured absorption of various coatings attached to the surface of plate and tube. The fitting curve is then established by the corresponding relationship between the absorption of different coatings on the plate and tube, so as to modify the measured data to obtain the absorption of the absorber tube coating. The results show that, the accuracy of the fitting curve R²≥0.995, the maximum relative error is 13.39% when the absorption of 2.5 cm diameter tube is measured directly under unshaded conditions, and the relative error reduces to 0.27% and 0.45% by using Blu-Tack method and diffuse reflection box method, respectively. The difference of the relative error of each point of the two methods is less than 0.30%. The greater the coating absorption, the less the influence of each factor on the measurement results. In the test method of Blu-Tack, the larger the tube diameter, the smaller the measured absorption.

In recent years, numerous incidents of harmonic resonance accidents involving new energy power stations have occurred both domestically and internationally. The frequency scanning method, characterized by its simplicity of operation and clear physical significance, has been widely employed in engineering for the assessment of system resonance. When establishing electromagnetic transient simulation models for large-scale new energy power stations and conducting frequency scans, it is common to substitute a single or multiple generators for the actual power station to reduce modeling complexity. However, the applicability of this approach in addressing high-frequency harmonic resonance issues has not been effectively revealed. To address this gap, this study takes a large-scale photovoltaic power station as an example and employs a bottom-up modeling approach to establish a detailed impedance model for the photovoltaic power station. Building upon this, a dynamic equivalent model is developed using the principle of equal power loss. The two models are then thoroughly compared and analyzed based on actual field station parameters. The research findings indicate that the series impedance of the collector lines has a minimal effect on the harmonic model. As a result, the study proposes a simplified dynamic equivalent model replacing the π-type circuit with a ground capacitor. The results demonstrate that the equivalent model established through the equal power loss method accurately reflects the harmonic resonance characteristics of the photovoltaic power station. Additionally, the model neglecting the inductance of the collector lines proves applicable for analyzing the mid-to-low frequency harmonic resonance in photovoltaic power stations.

An organic Rankine cycle (ORC) unit is designed through thermodynamic analysis and equipment type selection under geothermal conditions. Then, this unit is simulated to study its dynamic operating characteristics and investigate the influences of four parameters, such as heat source temperature, mass flow rate, cooling water temperature and working fluid mass flow rate, on key operating parameters and performance of the ORC unit. The results show that, the output performance of the unit is primarily affected by the cooling water temperature. An increase of 10 ℃ in the heat and cold source temperatures results in a decrease of 16% and an increase of 7% in the output power, respectively. The increased heat source temperature and mass flow rate will make the degree of superheat, evaporating pressure and shaft work increase significantly, and the system thermal efficiency decrease slightly. However, the effect of increasing heat source temperature and mass flow rate on the heat transfer of the cold side is limited, that on the hot side heat transfer is also neglectable, so the vapor superheat and evaporating pressure remains constant. Compared with the change of cold and heat source parameters, the change of working fluid mass flow rate has the lowest overshoot of parameters during dynamic operation of the unit, and the unit can reach the next steady state rapidly.

A new kilowatt-class methane reforming hydrogen production reactor is designed, using solid oxide fuel cell exhaust gas for heat supply. The system can make full use of the waste heat and combustible components in the exhaust gas to form a compact and efficient natural gas power generation system. Computational fluid dynamics was used to numerically simulate the combustion and reforming reactions in the reactor. The results show that the solid oxide fuel cell anode and cathode exhaust gases can be stably burned in the reactor to form a high-temperature flame of 1 486 ℃ to provide heat for the methane steam reforming reaction. In the reaction tube, the concentrations of H₂O and CH₄ continue to decrease along the way. Due to excess water vapor, the H₂O volume concentration at the outlet is 35%, the hydrogen concentration volume fraction is 45%, and the methane conversion rate reaches 90%. Nickel catalyst has a high thermal conductivity, so the temperature difference between the inside and outside of the reaction tube is less than 15 ℃. At the same time, experimental research was used to obtain data such as temperature, methane concentration and methane conversion rate in the reactor. The simulation results were compared to verify the accuracy of the numerical simulation.

The health status of gearbox directly affects the power generation of wind turbine. In order to achieve early warning of gearbox fault status in engineering practice, a K-means clustering algorithm based on improved lion swarm optimization was proposed. The supervision mechanism and the sine and cosine optimization algorithm considering nonlinear weights are introduced into the lion swarm algorithm, and then the optimized lion swarm algorithm is used to iterate the lion king position. By selecting the optimal solution as the clustering center of the K-means algorithm, the problem of strong dependence of conventional clustering algorithms on the selection of initial clustering centers is solved. The UCI data are selected for comparative verification of the algorithm, and the results show that, the K-means clustering algorithm based on the improved lion swarm optimization has achieved a better improvement in classification accuracy and stability. This algorithm is then applied to comparative test of gearbox vibration acceleration effective value for four wind turbines of the same type in a wind farm. It is found that the distribution of classification centers determined by this algorithm is consistent with the actual operating status of the gearbox, and agrees well with the vibration energy distribution corresponding to different states of the gearbox specified in the standard, indicating that the algorithm can realize early fault warning of wind turbine gearbox.

The mathematical and physical models of different semicircular channels are established, and the accuracy of the numerical models is verified by comparing with the experimental data. The thermal and hydraulic heat transfer performance of supercritical carbon dioxide (S-CO₂) in uniform cross-section semicircular tube, diverging tube and converging tube is studied, and the influence of different channels and pressures on the thermal and hydraulic performance of S-CO₂ in a semi-circular tube with variable cross-section is calculated and analyzed. The results show that, compared with the uniform cross-section semicircular tube, the diverging tube deteriorates heat transfer, the converging tube enhances heat transfer. The overall heat transfer coefficient of converging semicircular tube with the inlet and outlet radius ratio of 1.0:0.5 increases by 39.93%, and the maximum evaluation factor PEC of flow heat transfer comprehensive performance is 1.346. When the pressure is closer to the critical pressure or the heat flux is low, the heat transfer performance is higher. Finally, the reason why the converging tube with variable cross-section can enhance heat transfer is explained from the perspective of field coordination and turbulent kinetic energy distribution. The research results can provide new ideas and theoretical guidance for the design and optimization of coolers in S-CO₂ circulation system.

To address the mismatch between electricity supply and demand caused by the intermittency and fluctuation of renewable energy sources, a combined cycle energy storage and power generation system incorporating a closed supercritical carbon dioxide (S-CO₂) cycle and a high-temperature heat pump is proposed, which is an innovative exploration of the Carnot battery form. Through energy exchange via molten salt heat storage and water cold storage devices, this system efficiently integrates the heating process of the heat pump cycle with power generation process of the S-CO₂ cycle, which achieves a favorable round-trip efficiency for the energy storage power generation system. Simulations are performed to calculate the typical operational parameters and thermodynamic performance of the combined cycle, and to analyze the influence of main parameters of the S-CO₂ cycle on the overall efficiency of the system. The results indicate that, increasing the inlet temperature of the expander aids in enhancing the overall cycle efficiency, achieving an optimal electrical-to-electrical efficiency of 62.8%, while reducing the demand for heat storage molten salt. Elevating the inlet gas parameters of the main compressor will lead the system efficiency to reach a peak value, beyond which the overall cycle efficiency no longer increases. The optimal bypass ratio for the main recompressor is 0.35, which allows the system to achieve optimal efficiency. The optimal operating conditions of the S-CO₂ cycle system are identified, offering an electrical-to-electrical efficiency that is 7.98% higher than a reversible Brayton system under the same conditions.

The gas cooler, as an essential heat exchange device in Brayton cycle system, has a significant influence on structural compactness and operational efficiency of the cycle system. The performance and influencing factors of a cross flow printed circuit heat exchanger (PCHE)-plate-fin gas cooler are analyzed. A calculation model is established for this type of heat exchanger, and a MATLAB program is written to verify its reliability. Based on this, the coolers are designed, and the power density is above 1 MW/m³, indicating the cooler is compact heat exchanger. Moreover, the design and performance analysis of the heat exchanger are carried out under varying working conditions, and the change law of the pressure drop and heat transfer performance of the gas cooler with the inlet state of the circulating working medium and cooling air is given. The pressure drop and heat transfer performance are compared when the working medium of the Brayton cycle is supercritical carbon dioxide, nitrogen and air. The results show that, the change of cold and hot fluid mass flow has the most obvious influence on the heat transfer performance. The research has reference significance for the design and operation of Brayton circulating air cooling heat exchangers.