In order to improve the accuracy of ultrashortterm power prediction of wind turbines, this paper proposes a CNNBiLSTM ultrashortterm power prediction method considering the health status of wind turbines and dual attention mechanism. Firstly, considering the influence of the interaction between the environmental factors and the components of the wind turbine on the output power of the wind turbine, he relative error of the normal operation of each component of the wind turbine is used as the deterioration degree of the monitoring index. Secondly, the fuzzy comprehensive evaluation method assesses the health of wind turbines, and the historical data set is categorized based on the evaluation results. Finally, the dual attention mechanism CNN BiLSTM model is used to construct an ultrashortterm power prediction model for the classified data set. The experimental results show that the RMSE and MAE considering the health status of wind turbines are reduced by 17.3% and 20.5% respectively compared with the RSME and MSE without considering the health status of wind turbines.

Aiming at the unbalanced load that the wind turbine is subjected to when it operates above the rated wind speed, an independent pitch control strategy that combines the Radial Basis Function (RBF) neural network and Model Predictive Control (MPC) is proposed. A meanperiod statespace model suitable for controller design is established by means of wind turbine dynamics equations and coordinate transformations. On the basis of Kalman state observer, the model predictive control is used to adjust the pitch angle of the wind turbine instantaneously and the RBF controller to suppress the loads, and then the required independent pitch controller is designed. Taking the NERL 5 MW wind turbine platform as an example, the load characteristics of the independent pitch control strategies based on Proportional Integral (PI), MPC, and MPCRBF are analysed under turbulent winds, as well as their operating characteristics. Simulation results indicate that the method can reduce the load efficiently, improve the operating life of the wind turbine, and have a better suppression effect on the power fluctuation.

In the security and stability analysis of largescale power grid, the wind farm models are usually simplified with the dynamic equivalent modeling method. For a largescale wind farm, due to the large number of wind turbines and the divergences in their characteristics, the wind farm is generally aggregated into multiple equivalent wind turbines. When estimating the parameters of the equivalent wind turbines, in order to avoid identification of a large number of parameters at the same time, the existing method only selects the key parameters with large sensitivity for identification, and the remaining nonkey parameters are not identified by giving the theoretical values. Therefore, the accuracy and robustness of the equivalent model are greatly affected by the accuracy of the assignment of nonkey parameters. In order to solve this problem, the paper proposes a dynamic equivalent modeling method for wind farms based on multistep parameter identification. Firstly, the clustering method is used to group the wind turbines, and the wind turbines within each subgroup are aggregated into one equivalent wind turbine to establish a simplified wind farm model. Secondly, based on the hybrid dynamic simulation technology, the external system of each equivalent wind turbine is replaced with a variable impedance to realize the independent identification of each equivalent wind turbine. Finally, the equivalent parameters of wind turbine are classified based on the trajectory sensitivity, and the classified parameters are identified with a multistep identification method. The effectiveness of the proposed method is verified in a modified IEEE 39 bus system.

In transmission lines, due to the influence of line impedance, it is difficult for energy storage power station systems to allocate power reasonably according to capacity. In order to better promote the stable operation of black start, this article proposes a battery power distribution scheme for energy storage power plants based on improved Virtual Synchronous Generator (VSG). This scheme first introduces virtual impedance to eliminate bus voltage fluctuations caused by line impedance, in order to improve the accuracy of power allocation in energy storage power station systems; Then, in response to the problem of uneven power distribution among multiple energy storage units in parallel in an energy storage power station due to the influence of line impedance, the power transmission and circulating characteristics of multiple energy storage devices in parallel were analyzed to continuously improve the black start system. Finally, experimental analysis was conducted using Matlab/Simulink and semi simulation platforms to verify the effectiveness of the proposed strategy, which can improve the stability and economy of system operation.

When the doublyfed machine is connected to the grid, it can support or raise the frequency of the grid by compensating the active power, but the rotor speed decreases quickly and is not controllable, the time of active power compensation and frequency support is limited, and the stability of motor can not be guaranteed. This paper presents a frequency support technique for doublyfed machine (DFIG) phase modulation system with flywheel energy storage based on virtual synchronization control and dynamic speed limit. Firstly, the flywheel is hung on the rotor shaft of the doubly fed machine to increase the inertia of the system, and the mechanical energy storage, and delay the rotor speed decline rate in the process of frequency drop. Secondly, when the frequency of power grid drops, the realtime compensation of active power is carried out by means of virtual synchronization control strategy, and the compensation time of active power is regulated on demand based on the dynamic control of the lower speed limit of rotor. Finally, the validity of the proposed frequency support technology is verified by Matlab/Simulink simulating.

With the construction of the energy internet and the proposal of the dual carbon goal, the quantitative assessment of carbon emissions of multienergy systems is a key link to achieve lowcarbon operation of the system. Therefore, an optimal operation model of multienergy system considering wind power and PV development trend and carbon emission assessment is proposed in this paper. Firstly, by analyzing the energy input and output characteristics of thermal power units, gas turbines, P2G, and multienergy storage equipment, the topology and multienergy power balance model of the multienergy system are established; Secondly, the generalized Bass model is used to quantify the development trend of wind power and photovoltaic. On this basis, a carbon emission assessment model of the multi energy system considering the development trends of wind power and photovoltaic is established; Then, with the goal of maximizing the total revenue including energy regulation revenue, carbon quota trading revenue and operating cost of multienergy system, an optimal operation model of multienergy system considering wind power and PV development trend and carbon emission assessment is proposed. Finally, the results of the example show that the optimization operation model of multienergy system proposed in this paper can effectively improve the operational benefits and reduce carbon emissions of the multi energy systems.

This article proposes an intelligent charging station energy scheduling system based on machine learning, which is applied to public fast charging station microgrids equipped with photovoltaic systems and energy storage systems using secondary life electric vehicle batteries. The energy dispatch system can be used to address the uncertainty of energy demand for electric vehicles and the power gap between grid connection and fast charging services. In addition, this article uses machine learning methods to automatically synthesize suitable energy scheduling systems based on fuzzy rules. The energy dispatch system proposed in this article considers different electric vehicle fleets and photovoltaic scales, providing a reference for the optimal scale of photovoltaic systems and the effectiveness of nanogrid systems. Finally, in the experiment, a mixed deterministic stochastic process was used to simulate the energy demand of electric vehicles, which showed an improvement in performance compared to the optimal benchmark solution. This indicates that the system can more effectively handle the energy demand uncertainty of electric vehicles and the power gap between grid connection and fast charging services.

The transient power angle stability and voltage stability issues of new energy grid connected systems like wind power are combined, and suffer the risk of short circuit current level. Current research mostly focuses on improving the stability based on single factor, without considering the multiple factors to develop optimization methods for wind turbine control parameters. To solve the issue, an optimization method for key parameters of wind turbine short circuit current and reactive power support is proposed in this paper considering transient power angle stability. Firstly, based on the simplified model of windthermal combined system, the mechanism of power angle stability problem and the influence of active power output of the wind turbine on power angle stability are analyzed. Then, the analytical expression for the shortcircuit current of the directdriven wind turbine is derived, and the key factors that influence the levels of active and reactive currents are analyzed. Finally, the influence of key parameters of short circuit current on the reactive power voltage support capacity of wind turbines are studied by simulation, based on which the optimization principles and method for key parameters of wind turbine short circuit current and reactive power support considering transient power angle stability are developed. The simulation analysis results based on actual power grids show that the proposed method can improve the reactive voltage support capacity of wind turbines while ensuring transient power angle stability margin and short circuit current level.

A model of cement calciner using forestry biomass as alternative fuel was established based on Aspen Plus. Poplar wood was selected as the representative forestry biomass to replace 40% of the coal fuel, and the effects of gasification temperature, equivalence ratio, and biomass moisture content on the gas production components and calorific value, the outlet temperature of the cement calciner, and NO emission concentration was investigated under the conditions of lowtemperature gasification in a fluidized bed. Then NO emission concentration was taken as the response value, and the central combination test was carried out to establish the response surface and analyzed. The results showed that at a biomass feed rate of 11 000 kg/h, with the gasification temperature increasing from 550 °C to 700 °C, the content of each combustible component in the produced gas firstly decreased and then increased, and the increase in the content of the combustible component had a promotional effect on the generation of NO; at the equivalence ratio of more than 0.15, coke was mostly consumed by the gasification reaction, and the reduction of NO was weakened, which led to the increase in NO emission concentration; the increase in the moisture content of biomass led to the increase in the emission concentration of NO; the increase in the moisture content of biomass leads to a decrease in the content of each combustible component in the produced gas, and the outlet temperature of the cement calciner showed a decreasing trend; there was an interaction between the three factors, and the significance level of the effect of the equivalence ratio on the NO emission concentration was the highest, and the strongest interaction between the gasification temperature and the equivalence ratio was observed; when the gasification temperature was 550 °C, the equivalence ratio was 0.2, and the biomass water content was 12.5%, the NO emission concentration(442.91 mg/m³) was minimized

In this paper, a swirl staged burner with pyrolysis volatiles as fuel was designed. The flow field of the burner combustion process was simulated by ANSYS software, and the flow and combustion characteristics of the burner were studied. The results show that the swirling air of the burner effectively breaks up the pyrolysis volatiles, and achieves the effect of sufficient mixing, stable and uniform flow field and temperature field. The edge flow field generated by swirling air reduces the temperature of the inner wall of the burner (up to 815.69 K), slows down the high temperature corrosion of the wall and reduces the formation of NO₂; the NO₄ concentration in the flue gas of the burner is 43.33 mg/m³, which is significantly lower than the standard limit. The content of pyrolysis volatiles at the outlet of the burner is less than 0.2%.