It is crucial to improve the dynamic performance of the yaw system of wind turbines in multiple operating scenarios. Therefore, a predictive control strategy for wind turbine yaw system model based on reinforcement learning is proposed, which achieves multi-objective parameter dynamic optimization through the dual-delay depth deterministic policy gradient (TD3) algorithm. Firstly, a multi-step model predictive controller for the yaw system (YMPC) is established to address the conflicting control objectives of power loss rate and yaw actuator utilization rate. Secondly, based on the optimization objectives and wind conditions of the yaw system, a dual-delay depth deterministic strategy gradient (TD3) intelligent agent is designed to determine the input state, action, and reward mechanism of the YMPC. The TD3 intelligent agent is then used to tune the weight coefficients and control step size of the YMPC. Finally, the effectiveness of this method was validated using typical daily data from wind farms in northern China. The results indicate that the proposed strategy significantly improves the overall performance of the yaw system compared with the YMPC with fixed control parameters.

In light of the intricate nature of surface defects in wind turbine blades, conventional convolutional neural networks face problems such as threshold screening and non-maximum suppression processes, which increase computational complexity and are not conducive to model deployment. A novel defect detection model that integrates real-time-detection transformer (RT-DETR) with YOLOv5 algorithm is proposed. Firstly, the backbone network of YOLOv5 is redesigned based on RepVGG and FasterNet to reduce the computational complexity of the model. Recognizing the presence of small-sized targets within the detection tasks, an efficient channel attention (ECA) mechanism is integrated into the neck network’s feature fusion component, thereby augmenting the expressiveness of the output features. Finally, the detection head of original network is reconstructed with the Decoder from RT-DETR, minimizing the effect of non-maximum suppression on the model’s performance. The experimental results show that, the average detection accuracy and accuracy of YOLO-RT are 87.2% and 92.7%, respectively, on a self-constructed dataset of wind turbine blade surface defects, reflecting improvements of 4.4 and 8.0 percentage points over the original YOLOv5 model. The detection rate reaches 118.3 frames per second, surpassing that of alternative detection models. The enhancements introduced in this algorithm significantly improve both detection accuracy and speed, making it highly suitable for practical applications in detecting surface defects on wind turbine blades.

A resin-based solid amine adsorbent was prepared based on in-situ synthesis technology. The effects of air humidity (30%~90%), adsorption temperature (30~90 ℃) and adsorption time on the adsorption performance of CO₂ were investigated. Moreover, the adsorption kinetic characteristics of the adsorbents at different air humidities were studied. The results showed that, the maximum CO₂ adsorption capacity of the resin-based solid amine adsorbents in the air reached 2.38 mmol/g, and the air humidity and adsorption temperature had significant effects on the adsorption rate. The optimal adsorption efficiency was obtained when the air humidity was higher than 50% and the adsorption temperature was 25~50 ℃. The adsorbent exhibits very good cycle stability due to its excellent high temperature resistance.

For a supercritical carbon dioxide (S-CO₂) recompression Brayton (RB) system with two-stage compression and intercooling process, two system models with different layouts are constructed. The effects of key parameters such as low-pressure stage pressure ratio and split ratio on the system performance are explored. The results indicate that, the minimum and optimum splitting ratios exist for the RB cycle, the two-stage compression cycle of the main compressor (TCIP-RB), and the two-stage compression cycle of the recompressor (RTCIP-RB) under the design conditions. Moreover, the thermal efficiency of the TCIP-RB cycle is higher than that of the other two cycles within a certain range of split ratios. When the above three systems adopt the optimal split ratios, the maximum efficiency of the TCIP-RB cycle is 50.95%, which surpasses that of the RB and RTCIP-RB cycle by 3.20% and 3.98%, respectively. At different low-pressure stage pressure ratios, TCIP-RB and RTCIP-RB cycles have an optimal split ratio to maximize the thermal efficiency of the system, and the maximum thermal efficiency decreases with the increase of the low-pressure stage pressure ratio.

Based on the application of micro-channel printed circuit heat exchangers in fields such as thermoelectric power generation and aerospace, a high-efficiency, low-resistance, and easy-to-manufacture transverse slotted channel is proposed using the theory of boundary layer re-development, and the heat transfer is enhanced. Numerical simulations are employed to study the flow and heat transfer characteristics of both straight and slotted channels. The mechanisms of heat transfer enhancement and flow resistance reduction in the transverse slotted channel are investigated. The results show that the entrance effect can significantly enhance heat transfer with a minimal increase in flow resistance. The transverse slotted channel creates multiple entrance effects in the slotted regions by inducing flow separation, which leads to periodic boundary layer redevelopment, thereby greatly enhancing local convective heat transfer. Additionally, due to the relatively small velocity gradient in the slotted regions, local resistance is effectively reduced. As a result, the proposed transverse slotted channel improves the heat transfer capability of the channel by 2.24%~2.59%, reduces the resistance by 6.66%~7.91%, and increases the overall heat transfer performance by 9.87%~11.02%.

Computational fluid dynamics-chemical reactor network (CFD-CRN) simulation is a suitable method for predicting NO_x emissions from gas turbines. A universal CRN automatic partitioning/solving program was developed and then applied and verified on a natural gas micro-mixing combustor. Through analysis of flow and combustion characteristics in the micro-mixing combustor based on CFD simulation, CRN partitioning criteria are established: firstly, the air and fuel zones are extracted, then major zones along the axial direction are divided, and further the zones are subdivided radially/circumferentially according to fuel-staging locations. The results indicate that, the CRN automatic partitioning/solving program enhances generality by using an XML standardized information interface and is suitable for complex combustor structures. The relative error between the predicted and experimental NO_x emissions under different operating conditions of the micro-mixing combustor is less than 11%, and the influence of CFD grid number on the NO_x prediction by CRN is relatively small. The effect of fuel distribution ratio on NO_x emissions from micro-mixing combustor is analyzed, and a suitable adjustment range is given. The proposed CRN automatic partitioning/solving algorithm has potential applications in predicting NO_x emissions from gas turbines.

An analytical model for a combined heat and power (CHP) system driven by deep geothermal energy based on heat pipes was developed. The dynamic heat extraction characteristics of the heat pipes are obtained through numerical calculations based on the heat pipe-geothermal rock layer model. By analyzing the thermodynamic and thermo-economic performance of the direct expansion CHP system, the effects of heat pipe structure (heat pipe diameter, length, and insulation layer length), operating time, and geothermal temperature gradient on the performance of the system are investigated. The results show that, lower steam condensation temperature of the heat pipes leads to greater heat extraction, which helps shorten the investment recovery of the system. However, reducing the condensation temperature also decreases thermal efficiency of the CHP system. Moreover, there exists an optimal steam condensation temperature that minimizes the system’s levelized cost of electricity (LCOE). The heat extraction rate from the heat pipes declines rapidly in the first five years, and then gradually stabilizes. To maintain stable heat extraction over long term (30 years) and avoid interference between adjacent heat pipes, the center distance between any two heat pipes should be kept above 80 meters. The economic performance of the CHP system is closely related to the structural parameters of the heat pipes. At an optimal steam condensation temperature, increasing the heat pipe diameter and length, and selecting target zones with higher geothermal gradients can effectively reduce both the investment payback period and the LCOE.

Low-temperature adsorption technology for coal-fired flue gas pollutants can synergistically remove various pollutants and achieve near-zero emission. Focusing on the key equipment of this technology, flue gas spray cooling tower, ANSYS Fluent software is used to simulate the inside of the tower, and the impacts of various parameters are analyzed. The results indicate that, increasing the spray height effectively extends the contact time between flue gas and cooling water, thus significantly enhances heat exchange. Reducing the temperature of the cooling water strengthens the tower’s cooling capacity. Additionally, moderately reducing the inlet flue gas velocity increases its residence time in the tower, promoting more thorough heat exchange. Reducing the droplet diameter of the cooling water enhances the heat transfer efficiency by increasing the contact area. Enlarging the spray angle extends the residence time of cooling water within the tower and lengthens the contact duration with flue gas, boosting heat exchange. Increasing the cooling water flow rate expands the heat exchange area, further improving the heat transfer performance. The addition of packing material improves the heat exchange capacity of the tower while conserving cooling water. Comprehensively optimizing these parameters can substantially reduce the temperature of cooled flue gas, providing theoretical support for the design, manufacturing, and optimization of spray cooling towers in the low-temperature adsorption technology for coal-fired flue gas pollutants.

The formation and emission of SO₃ in coal-fired flue gas pose serious threats to both the safe and economical operation of power plants and the atmospheric environment. To solve this problem, the SO₃ removal performance of sodium-based, calcium-based and magnesium-based absorbents is investigated, and the performance variations of Na₂CO₃, Ca(OH)₂ and CaO under different operating conditions are studied. The results indicate that, under the coexistence of SO₂ and SO₃, the absorbers would react with SO₃ in the flue gas preferentially, and the effectiveness of SO₃ removal by various absorbents ranked from highest to lowest is as follows: Na₂CO₃>NaHCO₃>Mg(OH)₂>MgO>Ca(OH)₂>CaO. Under certain experimental conditions, the SO₃ removal efficiencies of all the absorbents could reach higher than 80% when the chemical equivalent ratio of absorbent to SO₂ reached 2:1. Pre-calcination treatment for the absorbents enhanced their pore structures, facilitating SO₃ diffusion into the absorbent and improving the SO₃ absorption efficiency. Increasing reaction temperature, chemical equivalent ratio, and initial SO₃ mass concentration can promote the SO₃ removal. Additionally, a moderate increase in H₂O volume fraction aided SO₃ removal. When the absorbent is significantly excessive, external diffusion is the main controlling step affecting the chemical reaction rate, while the type of absorbent has a relatively minor impact on it.

Effective monitoring of chloride ion indicators in precision treated effluent plays a crucial role in ensuring the quality of the effluent, adjusting operational processes, and extending operating cycles. According to the theory of ion exchange equilibrium, the operating characteristics, effluent quality, and resin regeneration requirements of hydrogen type and ammonium type operation modes for precision treatment were analyzed and compared. Through laboratory simulation experiments and tracking experiments of the operating cycle of a precision treated mixed bed in a certain power plant, the migration characteristics of chloride ions in the effluent of precision treatment were mainly studied. The results show that, the requirement for regeneration degree of resin in the hydrogen stage is low, and the main focus of this stage is desalination, which is not prone to chlorine leakage. As the pH value of the effluent increases during the conversion stage, the requirement for resin regeneration also increases. When the condensate contains chloride ions, chloride ion displacement is prone to occur. During ammonium type operation, the mixed bed no longer focuses on desalination, and the mass concentration of chloride ions in the effluent is equivalent to that in the condensate. Therefore, effective monitoring of chloride ions is necessary for ensuring the quality of effluent water, especially during ammonium type operation. At the same time, using chloride ions as one of the monitoring indicators can not only ensure the quality of effluent water effectively, but also guide the adjustment of precision treatment operation process, and significantly extend the precision treatment operation cycle while ensuring the safe operation of the unit.