The dynamics of energy and power systems aims to elucidate the dynamic characteristics and regulatory mechanisms of mass and energy transfer, conversion, and storage across multiple scales. However, the increasing integration of renewable energy sources, characterized by its inherent intermittency and volatility, poses significant challenges to the system planning and operational control in this field. The purpose of this review is to understand the evolutionary mechanisms and dynamic behaviors of energy and mass transfer/conversion within the new energy systems, and to promote the disciplinary development of energy and power system dynamics to adapt to the evolution of this new energy systems. First, the developmental history of the discipline was reviewed. Then, from a multi-scale perspective, the dynamic issues of systems centered on traditional thermal objects and new energy objects within the new energy system were systematically examined. Further, the challenges and advances in dynamics at the levels of integrated energy systems and macro-energy systems were discussed. Finally, an outlook on the future research directions and development trends of the discipline was presented.

The integrated energy system (IES) is a multi-source, heterogeneous, and strongly coupled system that covers fossil fuels, electricity, natural gas, hydrogen energy, and renewable energy. Traditional performance evaluation indicators are no longer applicable to the overall and local performance evaluation of IES, and there is an urgent need for a new performance characterization parameter. In IES, the introduction of the exergoeconomic cost coefficient has fundamentally changed the perspective of energy utilization evaluation and plays an important guiding role in the construction of high-efficiency and low-carbon IES in the future. Firstly, based on the origin and development of exergoeconomics, the basic development context of exergoeconomics and its application trend in IES were introduced. Secondly, the concept of the equipment exergoeconomic cost coefficient was proposed, and its characteristics and applicability when applied to IES were analyzed. Finally, the integrated application scenarios of exergoeconomics in IES were analyzed, and its guiding role in key technologies of IES, such as configuration optimization, real-time dispatching, and heterogeneous energy pricing, was presented. Results show that as a characterization for measuring the comprehensive energy utilization level of IES, exergoeconomic evaluation indicators will surely play an important guiding role in the construction, operation, and maintenance of IES.

Based on fuzzy intelligent predictive control, a control scheme for optimizing superheated steam temperature of DC boiler was proposed. The scheme was based on fuzzy set theory and predictive control algorithm, combined with the expert system for optimizing the control decisions. The control signal filtering and non-disruptive switching functions were designed, and a fuzzy intelligent predictive control algorithm was proposed. Results show that under such disturbances as external disturbances, environmental noise and model parameter perturbations, the intelligent predictive control algorithm performs well in controlling the superheated steam temperature of the unit. The intelligent predictive control algorithm can meet the control requirements of the DC boiler superheated steam temperature system with large inertia, large delay and multiple disturbances.

Flame stability at the burner outlet under ultra-low load conditions is a key factor restricting the peak shaving capacity of coal-fired power units. To reveal the correlation mechanism between the dynamic characteristics of pulverized coal combustion and unit operational flexibility, a quantification method targeting the stability of pulverized coal flames at the outlet of swirl burners was proposed. Through coupled modeling of the spectral emission characteristics during pulverized coal combustion and the operating parameters of the distributed control system (DCS), a cross-scale dynamic combustion stability evaluation index was developed. In the experimental verification on a 660 MW coal-fired power unit, this method achieved real-time monitoring of the combustion state, with an image accuracy of 95.36% and a single-frame processing speed of 0.16 s. Compared with traditional detection devices, it exhibits higher accuracy in characterizing extreme combustion conditions. The research results provide dynamic methodological support for the flexible operation of traditional energy power systems in the new-type power system, while proposing a universal technical path for multi-scale dynamic modeling and control optimization of energy power systems.

To address the complex operational characteristics of the low-temperature economizer system of coal-fired units, such as large time delay, significant inertia and strong coupling, a multivariate decoupling control strategy based on fuzzy intelligent predictive control was proposed. First, a multivariate coupled mathematical model was established, taking the flue gas temperature at the outlet of the economizer and the air temperature at the outlet of the heater as the controlled variables, and the water pump frequency converter and the air heater outlet control valve as the control variables. The model parameters were then identified using the partial least squares method optimized by the golden jackal optimization algorithm. Building on this model, a multivariate feed forward decoupling strategy was designed based on the fuzzy intelligent predictive control algorithm to decouple the control for the low-temperature economizer system. The simulation results show that compared with the traditional control method, the multivariate decoupling control based on the fuzzy intelligent predictive algorithm exhibits superior control accuracy and robustness. The practical application results indicate that the multivariate decoupling control strategy using fuzzy intelligent predictive algorithm achieves excellent control performance for both the economizer outlet flue gas temperature and the air heater outlet air temperature. The deviation between the actual outlet flue gas temperature and its setpoint is kept within 5 ℃, and the deviation between the actual outlet air temperature and its setpoint is kept within 3 ℃. This has enhanced the stability of the system and significantly reduced fluctuation amplitude, meeting the requirements of industrial production.

To address the issues of complex NO_x generation mechanisms in coal-fired boilers of thermal power plants, and the difficulty in timely and precise regulation caused by ammonia injection into selective catalytic reduction (SCR) systems, a prediction method for NO_x generation concentration at the SCR system inlet based on an improved long short-term memory network (LSTM) and attention mechanism was proposed. Firstly, the initial feature variable set was determined through mechanistic analysis of NO_x generation during boiler combustion. Following the preprocessing of the raw operational data, the maximum information coefficient was used to ascertain the delay time for each variable, thereby constructing a time-lag compensated data set. Secondly, based on grey relational analysis (GRA), the key variables affecting NO_x formation were ranked to achieve feature selection. Finally, a novel time-series forecasting model ADS-Forecaster integrated the improved LSTM with an attention mechanism was proposed, which used dual-path feature encoder and self-attention fusion decoder to realize real-time prediction of NO_x concentration. The test results based on operational data from a 600 MW subcritical coal-fired unit show that compared with benchmark models (RNN, LSTM and xLSTM), ADS-Forecaster model achieves superior generalization capability and prediction accuracy.

To enhance the active support capability for grid frequency, the deployment of flywheel energy storage systems has become a preferred solution for many thermal power plants, making the coordinated control of flywheel and thermal power units a research hotspot. A coordinated optimal control method for a flywheel-thermal power secondary frequency regulation system based on sequential variational mode decomposition-non-dominated sorting genetic algorithm Ⅱ (SVMD-NSGA Ⅱ) iterative optimization was proposed. First, an automatic generation control (AGC) signal decomposition strategy based on SVMD was designed to allocate control commands in a manner compatible with the slow dynamic characteristics of thermal units. Then, by accounting for frequency regulation performance, the state of charge of the flywheel, and economic efficiency, a multi-objective SVMD parameter optimization method was developed. This formed a collaborative optimization control framework for secondary frequency regulation based on SVMD-NSGA Ⅱ iterative optimization. A case study based on an actual power system was conducted through simulation. Results show that the proposed strategy significantly reduces the fluctuations of fuel consumption and main steam pressure, while markedly improving the secondary frequency regulation performance and economic benefits of the thermal unit. Compared to conventional low-pass filtering methods, the proposed strategy improves the secondary frequency regulation performance index K_P by 17.8% and increases annual frequency regulation economic revenue by 970 800 yuan.

For the gas turbine(GT)-supercritical CO₂ (SCO₂) combined cycle system, the thermodynamic performance of cycle layouts and the influence of key parameters were analyzed for ten cycle configurations. Among the potential layouts, the double-cascade single-split double-expansion layout has a better suitability to utilize the waste heat of gas turbine. Based on the selected layout, a control-oriented dynamic model of GT-SCO₂ combined cycle was established from the mechanism, and researches were carried out on the dynamic simulation and dynamic characterization of the combined cycle system. Through simulation, the coupling characteristics of gas turbine and SCO₂ were investigated, while the influence of gas turbine exhaust temperature and exhaust flow rate on the dynamic characteristics of SCO₂ system and its power output, as well as the relationship between airflow rate rate of gas turbine and the whole system power output, were analyzed. Results indicate that the optimized cycle layout achieves a trade-off between thermal efficiency and system complexity. There are significant differences in the dynamic responses of gas turbine and SCO₂ system. The transient durations of high-temperature turbine inlet temperature caused by step changes in exhaust temperature and exhaust flow rate of gas turbine are approximately 523 and 420 seconds, respectively. The transient duration of the combined cycle output power caused by airflow rate of gas turbine is 516 seconds. Relevant dynamic characteristic studies can provide a support for the design of flexible power regulation strategies in the GT-SCO₂ combined cycle.

A nonlinear model predictive control (NMPC) system was developed to address following challenging control issues in the flue gas-molten salt heat storage process, including significant thermal inertia, significant fluctuations in key parameters, and strong multivariable coupling. Firstly, a dynamic nonlinear model of the flue gas-molten salt heat storage system was established for control-oriented design, and dynamic response characteristics of the system were analyzed. Secondly, a combined scheme (named as NMPC& feedforward scheme) of NMPC-feedforward control framework incorporating an extended Kalman filter (EKF) was proposed. Within this framework, the NMPC is responsible for achieving multi-loop coordinated control, the feedforward compensates for the impact of disturbances on key parameters in the system, and EKF is employed to improve the accuracy of state estimation. Simulation results demonstrate that, in terms of the ability of tracking the set value based on NMPC& feedforward scheme, compared with PID and PID& feedforward schemes, the mean absolute error of outlet flue gas temperature can be reduced by 96.5% and 89.4%, respectively, while the root mean square error can be reduced by 89.8% and 67.8%, respectively. The responsiveness and stability of system can be enhanced by the proposed scheme. In terms of robustness, under the disturbance of inlet flue gas temperature, compared with PID and PID& feedforward schemes, the maximum dynamic deviation of outlet molten salt temperature based on NMPC& feedforward scheme can be reduced by 79.1% and 86.5%, respectively.

To address issues such as difficult heat-power decoupling and limited peak-shaving capacity in coal-fired cogeneration units, a steady-state model of the cogeneration unit was developed, and a molten salt heat storage system was integrated to assist in heat supply. The impact of the heat storage steam supply ratio γ on peak-shaving capability and thermodynamic performance was studied. A comprehensive evaluation was conducted based on the TOPSIS method, and the optimal heating mode was determined. Results show that the γ has a positive promoting effect on the system's peak shaving capacity and energy efficiency. The optimal mode is when molten salt steam supply undertakes the entire heating load. Under this condition, the unit power output decreases from 248.15 MW to 176.34 MW during the heat storage process, achieving a deep peak shaving capacity of 15.66 MW. During the heat discharge process, the output power increases to 345.40 MW, realizing a peak over-generation capacity of 25.40 MW, which bidirectionally expands the power regulation range. This mode features both high efficiency and low carbon properties. Under the heat storage and heat discharge modes, the energy efficiency reaches 48.89% and 60.94% respectively, the exergy efficiency is 35.13% and 37.86% respectively, the round-trip coal consumption rate for power supply is 290.15 g/(kW·h), and the carbon emission amount is 677.14 g/(kW·h).

In response to the safety hazards caused by the high-temperature thermal decomposition of molten salts in molten salt energy storage technology, this study aims to systematically investigate the high-temperature decomposition behavior and mechanism of molten salts by conducting high-precision online monitoring of corresponding key gaseous products. Based on the wavelength modulation-direct absorption spectroscopy (WM-DAS) technique, an experimental system has been designed and constructed for the dynamic, real-time, and high-precision online monitoring of high-temperature reactions of molten salts and corresponding products, enabling the real-time detection of the concentrations of NO and N₂O, which are decomposition products of molten salts. The system achieves lower detection limits of 2.6×10^-6 and 2.1×10^-5 for the volume fractions of NO and N₂O, respectively, with remarkably low standard deviations of the fitting residuals for the absorbance function, at 1.269×10^-3 and 1.416×10^-4, respectively. The experimental results on the high-temperature decomposition of molten salts indicate that the decomposition temperature of pure sodium nitrite molten salt is 256 ℃, while both pure nitric acid molten salt and ternary nitrate molten salt (HTS molten salt) exhibit higher decomposition temperatures, both at 290 ℃. Under identical conditions, pure sodium nitrite molten salt generates higher NO and N₂O emissions compared to HTS molten salt containing an equivalent amount of sodium nitrite. Additionally, reactions between nitric acid molten salt and quartz materials would significantly increase NO emissions at temperatures above 500 ℃.

To solve the problem of deep peak regulation for flexibility faced by thermal power units, a scheme of coupling the adsorption compressed CO₂ energy storage (AB-CCES) system with thermal power units to achieve flexibility retrofit was proposed. The impact of the thermal-electric co-storage/co-supply characteristics of the AB-CCES system on the efficient storage/utilization of electrical and thermal energies in thermal power units was investigated. Thermodynamic research results show that the round-trip efficiency and energy storage density of the AB-CCES system under the constant pressure operation mode are 84.45% and 6.06 kW·h/m³, respectively, while those under sliding pressure operation mode are 89.62% and 5.93 kW·h/m³, respectively. Under the 50% of turbine heat acceptance (THA) load in a thermal power unit, the maximum regulation ranges during load increase/decrease stages of this integrated system are 16.36%P_e (P_e is the rated load) and 10.6%P_e, respectively. However, the thermodynamic performance of the integrated system presents a slight decline compared with that of the thermal power unit.

To addresses the coordinated control of the compressor and outlet regulating valve for managing the charging load in a compressed air energy storage (CAES) system with a wide sliding pressure storage range, specifically under conditions when the pressure in the air storage reservoir falled below the rated/minimum outlet pressure of the final-stage compressor during the energy storage stage, a full working-condition dynamic simulation model of a 150 MW/1 200 MW·h adiabatic compressed air energy storage system was developed. Utilizing the multidisciplinary co-simulation approach, the model enabled real-time simulation of the thermodynamic system's dynamic behavior and the control system on a unified platform. Multiple control strategies for the compressor and the outlet regulating valve were proposed, and the regulatory performance of different operational schemes was verified. Simulation results indicate that the compressor responds more sensitively to flow variations and remains actively involved in regulation during system flow fluctuations. Considering both the stable output of the process system and the average energy efficiency across the entire sliding-pressure range, a combined control strategy which integrates compressor pressure/differential pressure control with outlet valve flow control is recommended when the reservoir pressure is below the final-stage compressor's rated/minimum outlet pressure. Compared to other strategies, this approach keeps the guide vane angles of the first three compressor stages essentially constant, minimizing the throttling energy loss of the outlet regulating valve, and achieving a maximum reduction in the energy consumption of 0.30% during the medium pressure phase of the gas storage interval.

To effectively enhance the stability of power system frequency under high penetration of wind power, and to fully utilize the frequency regulation capabilities of both wind and thermal power sources, a frequency band decomposition-based coordinated frequency regulation optimization control strategy for wind-thermal-storage was proposed. Firstly, at the system level, a wind-fire storage coordinated control model was established, and the thermal power primary frequency regulation signal was decomposed, in which the high-frequency part was responded proportionally by the wind-storage system, serving as power compensation for frequency regulation instructions of each respective response system, while the low-frequency part was responded by the thermal power. Secondly, at the unit level, wind power was designed to adopt low-overspeed deloading combined with comprehensive inertia control, while energy storage to use droop control considering the state of charge to balance their respective frequency regulation capabilities. Finally, simulation verification was carried out in different scenarios. Results show that in different simulation scenarios, the proposed strategy can increase the system minimum frequency point and reduce the output fluctuation of thermal power units.

In order to sufficiently explore the operational flexibility of combined heat and power (CHP) units as well as promote the utilization of renewable energy such as wind power and photovoltaics, the effects of the dynamic characteristics of the heating network on the heat system were considered from the perspective of the integrated energy system to dynamically model the heat system for fully exploring its energy storage potential. The operation strategies of the system under different scenarios were analyzed. Furthermore, the effects of the heating network energy storage and the actual energy storage systems on flexibility were compared, and engineering verification was carried out on the actual units. Results show that after considering the dynamic characteristics, the heating network plays the role of thermal energy storage system, leading to the enhancement of flexibility of combined heat and power units and a 4.7% improvement of renewable consumption rate in the set scenario. The analysis of the heating season data shows that the flexibility and the thermal economy of the units are improved, with an increase of 1.7% in the unit efficiency.

In response to the unclear topology of energy supply network within energy station, the complexity of multi-energy and carbon energy interrelationships, as well as the insufficient modeling of carbon-electric coupling in the interaction process of station network, by fully considering the dynamic impact of charging and discharging process of energy storage equipment on distribution of carbon flow in supply network, a dynamic modeling method for carbon energy flow in grid-connected energy stations with CCHP combined energy supply was proposed. Based on this, a low-carbon interactive optimization model for the power grid was constructed by considering the impact of time-of-use electricity pricing and the internal dynamic carbon potential of the system on energy conservation and carbon reduction of the system. Results show that the energy supply network topology of the grid-connected combined heat, power and cooling energy station presents a star-like structure, and the dynamic charging and discharging process of the energy storage equipment has a significant impact on the carbon flow distribution of the energy supply network. Furthermore, the proposed low-carbon interactive optimization model can reduce the operating costs of energy stations by 3.9% and carbon emissions by 4.4%, while also supporting the power supply and demand balance and low-carbon development of distribution network.

A combined cooling and power supply system equipped with heat pumps and cold storage tanks was developed based on Simulink,and the impact of cooling supply area and coefficient of performance (α_COP) on system performance was analyzed. The operation schedule of the heat pumps was optimized with the goal of maximizing revenue. Results show that the system can fully meet the cooling demand of users across all time periods for an area of 400 000 m² under the current installed capacity of the heat pumps. When the area exceeds 1 300 000 m², the heat pumps operate consistently at peak power during deep peak shaving periods, and the consumption amount of peak shaving electricity reaches its maximum at this moment. In the low α_COP range, improving refrigeration energy efficiency exhibits a significant leverage effect on the cooling supply gap. As the α_COP increases, the marginal improvement in system performance gradually levels off. There exists a critical electricity price that maximizes system revenue so that the heat pumps are activated when the electricity price is below this critical value and stopped when the electricity price is above this critical value. By optimizing the operating schedule of heat pumps, the typical daily revenue of the system can be increased by 1 722 yuan, which is a significant improvement compared to conventional operation modes.

To address the requirements for responding to cascading faults in power systems, a fault evolution model for wind power-integrated systems under typhoon conditions was proposed based on equipment fault modelling and probabilistic power flow modelling. Firstly, a time-series risk inheritance mechanism was constructed to dynamically characterize the risk accumulation effect in each time period according to the temporal characteristics of fault evolution. Furthermore, a dynamic pruning strategy was designed to dynamically identify and eliminate low-risk redundant paths based on real-time accumulated risk values during the evolution process. In addition, a branch isolation caching mechanism was introduced to store intermediate states and reduce computational redundancy. Simulation verification was conducted through using the modified IEEE 39-bus test system. Results show that the constructed model shortens the calculation time of cascading fault evolution to 12.52% of that of the enumeration method, with a high-risk path coverage rate of 91.1%. It effectively balances the capturing accuracy of high-risk paths and computational efficiency, and can provide a feasible computational framework for early warning and defensive dispatching of system cascading faults.

To address the challenges of multi-objective load frequency control in islanded microgrids, a robust dual-mode model predictive control strategy was proposed based on active disturbance application. Firstly, a fully actuated model of the microgrid was developed, in which modeling errors and uncertainties arising from renewable energy sources were aggregated into the generalized disturbance. An extended state observer was designed to estimate this disturbance in real time, and the estimated values were integrated into the receding horizon optimization process of the predictive model. Subsequently, a dual-mode model predictive control method was introduced to ensure input-to-state stability, improve frequency regulation performance, and reduce generation costs, while effectively decoupling robust stability from optimal performance of the frequency regulation system. Finally, the performance of the proposed control strategy was rigorously evaluated through hardware-in-the-loop experiments, comparing with those of existing methods under diverse operating conditions. Experimental results demonstrate that the proposed control strategy reduces frequency regulation time by approximately 20%, decreases overshoot by 10%, and enhances dynamic economic performance by 30%.

With the adjustment of energy structure, the share of renewable energy in integrated energy systems dominated by cold, heat, electricity and gas loads has been steadily increasing gradually. The inherent uncertainties in wind and solar power generation and the frequency stability challenges faced by multi-energy microgrids has become prominent issues. Considering the thermal inertia of temperature propagation, a coordinated scheduling model with wind and solar power integrated into cold, heat, electricity and gas microgrids was constructed. The objective function comprehensively considered both frequency deviation penalties and operational economic costs, with system optimization achieved through mixed integer linear programming (MILP). Simulation results show that the frequency penalty components in operational costs reduce by approximately 95% with consideration of frequency deviations caused by wind and solar power generation and cold, heat, electricity and gas loads, thereby making both system economy and stability better.

To address the issues of high energy consumption and carbon emissions in data centers, an innovative nuclear-wind-solar-storage integrated energy system architecture based on small modular reactors was proposed. By coupling the base load characteristics of nuclear energy with the complementarity of wind, solar, and storage, an electrical-cooling-heating multi-energy collaborative optimization model was constructed. By introducing an absorption cooling priority strategy within a four-objective optimization framework containing economy, energy efficiency, environmental impact, and reliability, solutions were conducted by a multi-objective cat swarm optimization algorithm, and combined with the preferred-sequence graph method and criteria importance through intercriteria correlation (CRITIC) method, an evaluation was carried out based on subjective-objective weighting. Results demonstrate that, compared with the traditional three-objective optimization electrical cooling priority scheme, for the four-objective optimization absorption cooling priority scheme, the net present cost of system is increased by 1.2%, and the carbon emission during life-cycle is increased by 9.9%, the self-sufficiency rate of system is improved by 1.3%, and the approach index reaches 0.675 7. It is validated the effectiveness of the proposed method, and relevant researches can provide a technically feasible pathway for coupling nuclear energy with data centers.

Energy system transition planning face challenges such as difficulties in data processing and high-dimensional solution, which are caused by the large scale of research objects and complex correlations. Traditional technologies are difficult to cope with these challenges, while artificial intelligence (AI) technologies such as AI large models could provide key support for solving this problem by virtue of their strong computing and reasoning capabilities. However, the exploration of applying AI large models in the specific interdisciplinary field of energy system planning is still in its initial stage, and the relevant research results remain relatively weak at present. Therefore, the application potential and development prospects of AI large model technology in the field of energy system planning were discussed, a systematic research and application framework from theory to engineering practice were proposed, and five main tasks of energy system planning were extracted, including planning objects, goals and indicators establishment, data research, demand forecasting, planning method design, and result approval and adjustment, as well as different methods and scales to be selected in the same task. In addition, the coupling relationship between AI large models and energy system planning was studied, the types of large model technologies were divided and their potential roles to adapt to planning research were analyzed. The research and application status, challenges, and gaps of AI large models in energy system planning were analyzed, and the possible research directions and content prospects were discussed. This study can promote the use of AI large models to reconstruct the cognitive framework and decision-making mode of energy system planning, and effectively assist in the transformation, upgrading, and configuration optimization of China's energy structure under the dual carbon goals.

Under the background of the ‘dual-carbon' strategy, the study of dynamic carbon accounting and optimal scheduling of park-level integrated energy systems is crucial. Currently, dynamic carbon accounting faces challenges such as poor data timeliness, fuzzy boundaries, and insufficient monitoring accuracy, and there exists complexity for the coupling with multi-equipment cooperative optimization scheduling, which made it difficult to accurately balance the low-carbon and economic objectives. A multi-objective co-optimization framework for low-carbon economy of park-level integrated energy system was proposed, and a dynamic carbon footprint accounting model was constructed, which solved the traditional static accounting problem, accurately divided the direct carbon emission and indirect carbon emission, and supported the fine carbon management. Based on this, a multi-equipment dynamic scheduling model was established, which integrated cogeneration, combined heat and power (CHP) with flexible adjustment of heat-to-power ratio, cross-domain synergy between power to gas (P2G) and energy storage, and prioritized consumption of renewable energy sources to fully tap the low-carbon potentials of the source-load-storage link. A multi-scenario simulation based on an industrial park showed that the proposed method can significantly reduce carbon emissions and total operating costs compared with the baseline scenario through the triple synergistic optimization of ‘adjustable heat-to-power ratio, P2G/energy storage and stepped carbon trading'.