A novel healthy state monitoring method for offshore booster station platforms is proposed to enhance the damage detection capabilities under complex operating conditions. Using a deep learning framework based on memory unit autoencoders, the method magnifies fault-relevant features via denoising and angular domain resampling high-frequency vibration data from offshore boosting stations. The model employs a deep convolutional neural network to learn historical data patterns, constructs a hidden state memory bank, and achieves sparse matching between sample encoded features and the memory bank. Finally, a Gaussian mixture probability model is employed to model the generated membership scores to assess the health status of the booster station. A case study of the offshore booster station in Rudong, Jiangsu, validates the approach, achieving an anomaly recall rate and accuracy of over 98%, outperforming other comparison algorithms.

The Brayton cycle is widely recognized as a key power cycle in the third-generation solar thermal power generation technology. Leveraging the strengths of artificial neural network methods for importance evaluation and quantitative analysis, this research employs a control variable approach to identify critical parameters, including turbine inlet temperature and compression ratio, from a range of operating parameters. In this method, the significance of parameters increases as the R² value decreases. Notably, when excluding these key parameters, the R² values fall to 0.57 and 0.64, respectively, both are lower than other operating parameters. Furthermore, the quantitative analysis of output power in the Brayton cycle yields exceptional results, achieving an R² value exceeding 0.999. The R² values for thermal efficiency and input heat are 0.992 and 0.988, respectively. Finally, the multi-objective optimization results suggest optimal settings of 500 ℃ for turbine inlet temperature and 2.19 for the compression ratio, corresponding to a thermal efficiency of 46.58%, output power of 100.97 kJ/kg, and input heat of –176.5 kJ/kg. This study offers valuable insights for the operational efficiency and performance assessment of the Brayton cycle in solar thermal power plants.

Due to the significant volatility and randomness of wind power data, low prediction accuracy is often observed with a single model in wind power prediction. To overcome this, an ultra-short-term wind power prediction method is introduced, based on modal decomposition and a combined neural network model. Firstly, the wind power data are processed based on the improved fully integrated empirical modal decomposition and sample entropy, which decomposes the unsteady series into smoother sub-sequences and reconstructs the high-frequency oscillatory component and low-frequency smooth component synchronously. Secondly, a hybrid prediction model for wind power based on an adaptive sparse self-attention mechanism is constructed. For the high-frequency oscillatory component with high complexity, the adaptive sparse Transformer model is used to fully explore the fluctuation information. For the low-frequency stationary components, the sequence features are fully extracted by the bidirectional gated recurrent unit model. Finally, the final prediction outcomes are derived by overlaying the forecast results of each component. Test was performed with actual data from a wind farm in Shandong, and the results show that, compared with other commonly used models, the proposed model’s root mean square error and average absolute error has decreased by 2.644 MW and 2.42 MW, and the coefficient of determination has a notable 18.2% increase, implying it has a good prediction performance.

The heat transfer characteristics and safety of a single tank thermal storage system during charging and discharging cycle are important indicators affecting the performance of the thermal storage tank. By coupling finite volume method and finite element method, a comprehensive model of a multi-layer wall structure molten salt single tank system is established, and the effects of inlet flow velocity and inlet/outlet temperature difference on the dynamic thermal characteristics and mechanical properties of the thermal storage tank during continuous charging and discharging cycling process are explored. The results indicate that, increasing the inlet flow rate will reduce the heat storage and improve the thermal efficiency, but will also increase the equivalent stress on the tank wall. Increasing the temperature difference between the inlet and outlet will increase the heat storage and reduce the thermal efficiency, and also increase the equivalent stress on the tank wall. To ensure the heat storage and thermal efficiency of the single tank heat storage system, as well as the safety of the system, for the single tank system with a heat storage capacity of 40 MW·h, the inlet flow rate of molten salt should be controlled within 0.002 60~0.003 46 m/s, and the temperature difference between the inlet and outlet of molten salt needs to be controlled within 200~250 K.

Supercritical carbon dioxide (S-CO₂) in horizontal tube with circumferential heating and semicircle heating is investigated numerically based on pseudo-boiling theory. The phase distribution of supercritical fluid in the tube is obtained. It is found that the heat transfer performance of supercritical fluid is determined by the thickness of vapor-like film on tube, which can be characterized by supercritical K number, involving the balance between evaporation momentum force and inertia force. The increasing thickness of local vapor-like film can trigger heat transfer deterioration. There are both overshoot wall temperature along the flow direction and non-uniform wall temperature in the circumferential direction. The emerging condition of heat transfer deterioration can be accurately predicted by supercritical boiling number SBO. Under working conditions where the pressure p is 8~20 MPa, and the range of mass flux G and heat flux density q_w is 300~1 300 kg/(m²·s) and 42~500 kW/m² respectively, compared with the circumferential heating tube, the semicircle heating tube behaves thinner vapor-like film to enhance the heat transfer performance. The critical SBO to heat transfer deterioration rises from 6.179×10^–4 to 9.798×10^–4. Furthermore, the semicircle heating tube keeps more uniform vapor-like film, the maximum temperature difference between the top and bottom generatrix of tube wall changes from 116.3 K to 57.1 K. Due to the ability to repress heat transfer deterioration and non-uniform wall temperature, semicircle heating is recommended to ensure the safe operation of horizontal heat exchangers in advanced supercritical CO₂ system.

Data quality is a key factor affecting the application effectiveness of optimization models of denitrification system operation. In response to the problems of lagging and poor representativeness of monitoring parameters in denitrification system operation, a denitrification performance parameter dimension reduction technology suitable for real-time performance monitoring is developed. The utilization rate of reducing agents that can reflect the denitrification ability of the denitrification system itself is set as the monitoring and evaluation parameter for denitrification system operation status, to improve the efficiency of data generation. Based on this, an optimization method for denitrification system operation that can eliminate adjustment delays is established, and an identification technology for typical abnormalities in denitrification system operation is constructed to guide the economic, safe, stable, and standard operation of the denitrification system. This technology has been implemented and applied in a 1 000 MW coal-fired unit at different loads. The results show that, the denitrification system operation guided by the utilization rate of reducing agents reduces the unit consumption of urea solution by 1.5%~8.4% and the ammonia escape at the denitrification system outlet by 10.7%~27.0%, and all of the ammonia escape at different loads meets the general control value of ammonia escape rate. The variation range of NO_x emission mass concentration in the exhaust reduces from 44.5~58.3 mg/m³ to 9.2~10.6 mg/m³, and the distribution deviation significantly decreases from 59.2%~75.2% to 21.4%~25.1%, which is more conducive to the automatic and stable control of the denitrification system.

A 300 MW supercritical carbon dioxide (S-CO₂) Brayton cycle oxyfuel power generation system is designed, and a simulation model is constructed using process simulation methods to study the effects of key operating parameters (such as the primary dry cycle flue gas ratio, the economizer side split ratio, the cold primary air temperature and the high-pressure turbine inlet pressure) on the system’s thermal performance indexes. The effects of key operating parameters such as primary dry cycle flue gas ratio, economizer side split ratio, cold primary air temperature and high-pressure turbine inlet pressure on the system thermodynamics were investigated, and the thermal characteristics of the generating unit were revealed. The results show that, the boiler efficiency decreases with the increase of the proportion of primary dry-cycle flue gas, and when the proportion of primary dry-cycle flue gas reaches 50%, the net electric efficiency of the system is the highest, which is 42.93%. The boiler efficiency rises at first and then decreases with the increase of the coal economizer-side shunt ratio, and the net electric efficiency of the system reaches the highest (42.86%) when the coal economizer-side shunt ratio is 11%. With the increase of cold primary air temperature, the boiler efficiency keeps increasing, and the rising trend slows down and stabilizes at 99.29% at temperatures higher than 65 ℃, and the net electric efficiency of the system keeps increasing and reaches 43.06% at 95 ℃. With the increase of high-pressure turbine inlet pressure, the boiler efficiency firstly rises and then decreases, and reaches the maximum value of 99.35% at 29.5 MPa, and the net electric efficiency of the system reaches the optimal value (43.66%) at 29.0 MPa.

The current coal-fired unit coupled with molten salt heat storage system technology has the problems of limited peak shifting capacity and poor peak heat economy. To address these issues, a new system of coal-fired unit coupled with compressed steam and molten salt heat storage is proposed, specifically including single molten salt heat storage scheme and double-molten-salt-heat-storage scheme. The system performs multi-stage compression of extracted steam through a multi-stage compressor and uses molten salt for heat storage. The compressed steam is eventually condensed to water so that its latent heat of condensation will be utilized. The simulation model of the coupled system scheme is established by EBSILON software. The research results indicate that, compared with the conventional molten salt heat storage technology scheme, the compressed steam and molten salt heat storage system can effectively reduce the effect of steam extraction and heat storage on the thermal economy of the system, and expand the peaking range of the unit. Specifically, the round-trip efficiencies of the single molten salt and double-molten-salt scheme are improved from 27.43%~38.03% to 62.13%~64.56% and 65.69%~66.93%, respectively, and the minimum outputs are reduced from 20.91%Pe to 19.84%Pe and 19.28%Pe, respectively, compared with the conventional scheme. Considering the thermodynamic and economic performance of the system, the single molten salt scheme is the best choice.

The sterilization effect of chlorine-containing disinfectants does not hinge on the total available chlorine concentration, but rather on the concentration of hypochlorous acid (HClO) molecules. To reduce the cost of circulating water sterilization in thermal power plants, HClO solution was used to sterilize circulating water. The HClO solution was prepared from sodium hypochlorite (NaClO) solution, CO₂ and pure water by non-electrolytic method, and experimental study on application performance (including stability and sterilization effect) of the HClO solution was conducted. The results showed that, the mass concentration of HClO in NaClO solution was extremely low, the mass fraction of HClO to available chlorine was only 0.690%~0.012% in NaClO solution with mass concentration of available chlorine in the range of 100~2 000 mg/L, which could be increased to 94.91% by reacting with CO₂, thereby improving the sterilization efficiency. When 30, 60, 80, and 90 mg/L stabilizer were added, the decomposition rate of HClO solution with mass concentrations of 500, 1 000, 1 500 and 2 000 mg/L could meet the requirements for a shelf life of one year of the Disinfection Technical Specification. When the dosage of NaClO and HClO (calculated by available chlorine) was 5 mg/L and 0.06 mg/L，the sterilizing time was 120 min and 15 min, the sterilizing rate could reach 90%, and it can be seen that HClO was more economical and efficient for sterilization. HClO solution can reduce the cost of circulating water bactericides in thermal power plants by more than 55% compared with that of conventional NaClO.

Proton exchange membrane (PEM) electrolysis for hydrogen production has broad application prospects, but it still has disadvantages such as high equipment cost and insufficient durability. Optimizing the flow channel can improve the uniformity of water and heat distribution in PEM and extend the service life of the electrolyzer. To enhance the performance of PEM, a three-dimensional wavy flow channel model was designed and simulated using COMSOL simulation software. The polarization curves, distribution of reactants and products, and temperature distribution of electrolyzers with different frequency wavy structures were studied, and the influence of flow rate changes due to the addition of wavy structures on the performance of the electrolyzer was explored. The results show that, compared with the conventional rectangular flow channel, the electrolyzer with wavy flow channels has significantly enhanced the heat and mass transfer effects and got better polarization performance. When the wavy frequency is 1.0, the current density of the electrolyzer increases by 2.1%, and the average gas volume fraction in the anode catalyst layer decreases by 3.7%, achieving the best overall performance. This study can provide certain references for the flow channel design of PEM electrolyzers.