An improved grey wolf optimizer (MGWO) is used to optimize BiLSTM to predict water wall temperature. The improved algorithm adopts nonlinear factor adjustment strategy, adaptive position update strategy and dynamic weight modification strategy to improve the global optimization ability of the GWO. The improved grey wolf optimizer is used to optimize the number of hidden layers, learning rate and regularization parameters of the BiLSTM model to improve the prediction accuracy of the model. The data of a power plant in Xinjiang are used for prediction simulation. The results show that, the improved optimizer has higher prediction accuracy, and can predict the change trend of wall temperature when the unit is lifting and lowering load. Compared with the LSTM and BiLSTM models, the average root mean square error of the model reduces by 9.86% and 3.69%, respectively, and the overtemperature of water wall temperature can be predicted in advance, which is of great significance for the prevention of overtemperature of water wall.

As a fundamental component of the organic Rankine cycle (ORC), the scroll expander's operating characteristics critically influence the ORC's overall performance. Initially, we establish a three-dimensional transient simulation model of the scroll expander. This allows us to systematically analyze the effects of various operating conditions on aspects such as suction pressure, exhaust pressure, rotational speed, and the output power and isentropic efficiency of the scroll expander, using numerical simulation. Following this, we study the effect of different operating conditions on the transient performance and mechanical properties of the scroll expander, and achieve a more comprehensive understanding of the mechanisms involved. Ultimately, we verify the accuracy of our numerical model using a laboratory-built test bench of the ORC low-temperature waste heat oil-free power generation system. The close correlation between the experimental results and numerical simulation outcomes authenticates the reliability and applicability of our numerical simulation method. In conclusion, this research's findings offer significant referential value for the design and optimization of the scroll expander.

In order to further investigate the loss mechanism in supercritical carbon dioxide turbines, the flow characteristics in a turbine stage were studied by numerical method. The losses in the passage of stator and rotor blades were decomposed, and the various loss values and their proportions were quantitatively calculated. The loss sequence of the supercritical carbon dioxide turbine stage was clarified. The results show that, the high density of supercritical carbon dioxide and low blade height result in a very large leakage loss in turbine stage. When the stage load coefficient is 0.93, the relative height of the stator clearance is 0.012 and the relative height of the rotor clearance is 0.010, the leakage loss accounts for 38.23% of the total loss, including 21.94% of the diaphragm seal leakage loss and 16.29% of the tip seal leakage loss. Except for leakage loss, when the average maximum thickness divided by chord length is 0.33 and the aspect ratio is 2.07 in stator passage, the profile loss is much higher than the endwall loss and trailing edge loss, accounting for 9.68% of the total loss. In rotor passage, when the average maximum thickness divided by chord length is 0.28 and the aspect ratio is 1.73, the difference among endwall loss, profile loss, and trailing edge loss is not significant, and the profile loss has the highest proportion, accounting for 15.39% of the total loss. The influence range of secondary flow in rotor is even wider, and its endwall loss is much higher than that of the stator. The main sources of endwall loss are viscous dissipation of fluid near the end wall and secondary loss caused by horseshoe vortices, passage vortices, etc. The research results will provide direction guidance and data support for the design and optimization of supercritical carbon dioxide turbines.

To establish an accurate and effective dynamic model of cogeneration units, a modeling method based on digital twin technology is proposed using unit operation data. Firstly, the historical data stored in the unit data server is extracted, it is then clustered using the improved genetic simulated annealing fuzzy C-means method to establish a historical data clustering library. Then, during the operation of the unit, real-time operational data is collected and transmitted, and a multi-level similarity recognition strategy is used to retrieve the historical data closest to real-time operational data in the historical data clustering library. Then, based on the optimization, the extreme learning machine will use the searched historical data for unit modeling. Finally, a twin model of a cogeneration unit in Hangzhou is established and comparative experiments are conducted. The results show that, the built model meets the accuracy requirements and can track the real-time state response of the unit. The model accuracy can be further optimized by flexibly changing the parameter settings during the modeling process.

Under the strategic goal of “carbon peaking and carbon neutrality”, the penetration rate of new energy sources is increasing, and coal power is gradually transforming into a flexibly adjustable power supply. While the flexible operation of coal-fired generating units will lead to energy efficiency losses and additional pollutant emissions, which have a certain degree of impact on the cost of power generation and are not conducive to the further exploitation of coal power flexibility resources. In this paper, a method for grading the flexible operating performance of coal-fired units is constructed. By calculating the performance indicators to assess the flexibility performance of coal-fired units under specific operating conditions, the relationship between the unit coal consumption rate and the generation of some air pollutants that affect the cost of power generation and the operational flexibility is investigated. Meanwhile, the kilowatt-hour cost of coal-fired power plants at the second level is estimated. Finally, second-scale operating data collected from a 330 MW unit are used to evaluate the impact of coal-fired unit flexibility on the cost of electricity generation, and build least squares support vector machine (LSSVM) model to realize the prediction of unit power cost under different flexibility levels. The research results indicate that prediction error in specific conditions can be reduced by 50% compared with that in full conditions.

The photovoltaic array will produce multi-peak P-U characteristics under partial shading conditions. Aiming at the problem of how to quickly and accurately realize maximum power point tracking (MPPT) to avoid a large amount of energy loss, this paper proposes an improved aquila optimization (AO) algorithm, which uses Circle chaotic mapping and reverse learning strategy to reasonably allocate the initial population position, so as to shorten the optimization time of the algorithm. At the same time, spiral optimization is carried out for the short gliding attack in aquila optimization algorithm. The whale optimization algorithm is combined to improve local optimal stagnation and convergence speed. Simulations and experiments demonstrate that, in comparison to particle swarm optimization (PSO), whale optimization algorithm (WAO) and aquila optimization algorithm, the algorithm can search the global maximum power point with greater speed, accuracy and suppleness under both static and dynamic partial shading conditions.

For the purpose of absorbing renewable energy, the coal-fired power units are required to operate flexibly, and the resulting variable operating environment will lead to large nonlinearity and uncertainties of the wet desulphurization process. Particularly, the time delay will make the control even harder. Therefore, in order to achieve a more flexible control structure, a new desulphurization control strategy based on frequency retrofit is proposed. Based on the strategy, a dynamic model is obtained through field experiments. Meanwhile, in order to deal with uncertainty well and achieve safe compensation of uncertainty in the control process, an updated Gaussian process model predictive control method for time-delayed objects is proposed, and the performance of the method is demonstrated by parameter analysis and simulation experiments. Finally, the effectiveness of the proposed control strategy and control method is verified by field application.

To solve the problem of excessive air leakage from the four-compartment air preheater, a four-compartment rotary air preheater of a 1 000 MW power plant is used as the study object to investigate the thermal characteristics, rotor thermal deformation and air leakage of the air preheater, by combining theoretical analysis with numerical simulation. The numerical simulation results show that, for every 1% increase in air leakage rate of the 1 000 MW unit, the heat transfer efficiency of air preheater will be reduced by 1.13%, and the power plant will consume 3 604 tons of standard coal more on average per year. There are large temperature gradients in the axial and circumferential directions inside the air preheater, and most of the corrosion and deposition areas are distributed in the low temperature section. There are large differences in rotor thermal deformation during different operating conditions of the air preheater, and the maximum thermal deformation occurs on the secondary air side. It is necessary to make corresponding adjustments according to the thermal deformation characteristics of different positions of the rotor when setting the sealing system.

In response to the instability problem of distributed new energy as an independent power source supplying energy to household energy systems, a capacity configuration method for distributed home energy systems based on hydrogen energy is proposed. This method is based on the characteristics of hydrogen energy that can suppress wind and solar fluctuations and flexibly convert load demands, and constructs a hydrogen coupled distributed energy family terminal energy system structure. Combining with the characteristics of renewable resources in the region, a hydrogen based distributed household energy system capacity allocation model is established with the goal of minimizing the net total cost. Taking the wind and solar resources and typical household energy data in Xinjiang region as an example, the impact of the configuration capacity of distributed energy and hydrogen energy systems on the system is simulated and analyzed, and the optimal capacity configuration of hydrogen energy systems under the optimal wind and solar ratio conditions is obtained. From the simulation results, it can be seen that the proposed model can effectively reduce the total cost of the household energy system while achieving reliable energy supply off the grid, and promote the consumption of new energy, which provides suggestions for the design of distributed household energy based on hydrogen energy.

A Python model was established for thermal performance of the heavy-duty gas-steam combined cycle triple-pressure heat recovery steam generator. The model calculates the detailed heat recovery steam generator parameters under the condition of changing unit load, including main steam pressure and flow rate, heat and heat transfer coefficient of each heat exchanger, as well as power output and efficiency. As the effect of exhaust gas temperature and flow rate on the heat recovery steam generator is analyzed, how the ambient temperature, humidity and fuel heating affect the heat recovery steam generator output is also discussed when the unit is in part-load. It is verified that the model has good simulation accuracy and calculation efficiency. Simulations for one certain frame gas turbine combined cycle show that: When the unit load is reduced from full load 650 MW to partial load 250 MW, the triple main steam pressure and feed water flow rate of heat recovery steam generator decrease, the steam turbine power output decreases from 219.1 MW to 130.4 MW, and the efficiency of heat recovery steam generator increases from 89.3% to 92.1%; main-steam flow increases as the flue gas flow rate and temperature increase at the inlet of heat recovery steam generator; As the load of the random group decreases, the heat transfer coefficient and heat transfer amount of each heat exchange surface of the waste heat boiler decreases, but the proportion of heat exchanger between the flue gas and the high-temperature section heat exchanger in the total heat increases, and the heat transfer between the flue gas and the low-temperature section increases. The proportion of heat exchanger heat transfer in the total heat is reduced; when the unit load is reduced from 650 MW to 300 MW, the proportion of steam turbine shaft power increases by 1.67 percentage points, the proportion of heat loss in the chimney flue gas decreases by 2.63 percentage points.