With the continuous increase of electrification in seaports, the single energy supply mode of seaport microgrid is evolving towards multi-energy integration. Aimed to achieve the goals of peak carbon and carbon neutrality, an optimal carbon trading mechanism-based allocation scheme of hybrid electric and thermal storage system is proposed to further maximize the economic and environmental benefits. First, the integrated energy system model of a seaport is established, incorporating a scheme within the carbon trading market. Then, a bi-level optimization framework is proposed, in which the upper layer is utilized to optimize the allocation of the hybrid energy storage system and the lower layer is employed to optimize the operation. Afterwards, a combination algorithm of the mesh adaptive direct search and the adaptive chaotic particle swarm optimization is developed to solve the proposed problem. Finally, the real-world data of Tianjing port is utilized to verify the method. The numerical results demonstrate that with the help of the proposed method, both the cost and carbon emissions are dramatically reduced.

In view of the fact that the current integrated port energy system (IPES) considers neither the time scale difference of refrigerated containers in port scheduling nor the impact of renewable energy and load uncertainty, this paper proposes a day-ahead and intra-day two-stage rolling optimization scheduling method for a container IPES. In day-ahead scheduling, based on the temperature rise process of refrigerated containers, a port cold chain energy demand model is established, which is combined with the logistics process after the arrival of refrigerated containers. Then, the day-ahead output values of each unit in the system are obtained with the goal of the lowest operating cost. In intra-day scheduling, a two-layer rolling model is proposed to obtain the adjusted output of the port energy equipment, which considers the prediction error of shore power load and renewable energy as well as the different response speeds of cooling, heating and power. The calculation results show that the collaborative optimization scheduling of refrigerated containers and the container IPES can effectively reduce the port operation cost and carbon emissions. The two-stage day-ahead and intra-day rolling optimization scheduling can improve the economy and stability of the system.

With the construction of a diversified electricity market, the factors affecting electricity prices are gradually increasing, and the market environment has undergone more drastic changes. In order to improve the accuracy of short-term electricity price prediction, an improved Transformer-particle swarm optimization (PSO) short-term electricity price prediction method considering multiple factors affecting electricity prices is proposed. First, based on the consideration of historical electricity prices and historical loads, the relevant factors of electricity price formation are further analyzed. The autocorrelation function is used to analyze the multi-cycle characteristics of electricity price and adjust input sequence, which overcomes the problem of limited prediction accuracy caused by using historical data only and adjusting the input sequence by experience. Then, by combining long short-term memory (LSTM), self-attention mechanism, multi-layer attention mechanism, and adopting a multi-input structure, an improved Transformer model is established to further enhance the ability of the LSTM model to capture long short-term dependencies between different time step information, to overcome the information utilization bottleneck of LSTM, and to adapt to complex multiple sequence inputs including historical electricity prices and various electricity price causes. In addition, the PSO intelligent algorithm is utilized to search for the optimal learning rate of the model at different learning stages, overcoming the limitations of manually adjusting the learning rate. Finally, the PJM market electricity price is used for example analysis. The results show that the proposed short-term electricity price prediction model can be applied to the market environment where electricity prices are affected by various factors and drastic changes, and effectively improve the accuracy of short-term electricity price prediction.

Fault diagnosis in power distribution networks is crucial for fault location, enhancement of fault processing efficiency, and reduction of power outage losses. Currently, the impact of switch operations and other interferences is seldomly considered in fault diagnosis algorithm designing and testing, which may lead to frequent mal-function and poor performance in practical applications. In this paper, a detailed analysis and modeling of the transient process of switch operation in distribution networks is proposed with the combination of the Mayr and the Helmer models. The transient waveform of the on-site operation process is compared and analyzed with the simulation waveforms generated in PSCAD. Based on the accuracy verification of the model, typical fault scenarios in distribution networks, including switch operation processes, are constructed for fault diagnosis algorithm tests. Compared to the traditional model, the model proposed can simulate and generate disturbance data close to the on-site switch operation process for reliability testing of fault diagnosis algorithms. Finally, several suggestions for optimizing the fault diagnosis algorithm and testing process are proposed through result analysis.

Aimed at the coupling problem of the combined wind-storage system participating in different call time scale scenarios in electricity markets, an optimal dispatching method of the combined wind-storage system oriented to the application of multi-time scale scenarios in electricity markets is proposed to guide the combined wind-storage system to suppress short-term wind power fluctuation, and participate in the electric energy market and the reserve ancillary service market, so as to realize the collaborative optimization among different call time scale scenarios application and maximize the economic benefits of the combined wind-storage system. First, considering the profit mechanism of different scenarios, the objective function is established with the objective of maximizing the economic benefits of multiple scenarios of the combined wind-storage system. Then, the constraints of the combined wind-storage system participating in various application scenarios and multi call time scale coupling constraints are established. Finally, the numerical simulation verifies that the proposed method can improve the comprehensive operation profit of the combined wind-storage system in the day-ahead electric energy market and the reserve ancillary service market while ensuring that the wind power fluctuation does not exceed the limit.

To improve the current automated guided vehicle (AGV) charging strategy at automated terminals, which is not fully coordinated with the distributed power supply, a joint optimization method of AGV logistics scheduling and orderly charging is proposed. First, the synergetic relationship between AGV logistics scheduling and charging scheduling is analyzed, and a joint optimization framework is built. Then, a method to calculate the distance traveled by AGVs while considering the segregation requirements of trucks inside and outside the terminal is proposed. Afterwards, for the AGV charging module, the judgment conditions of AGV charging status and the pile selection method are defined. Furthermore, to minimize the cost of purchasing electricity at the terminal, a joint optimization model of logistics scheduling and orderly charging is constructed by considering time-of-use tariff, distributed power feed-in tariff, power balance constraint, state of charge constraint at the termination moment, upper and lower bound constraints of decision variables, and logistics scheduling constraint. Finally, a fast solution method based on improved particle swarm optimization algorithm is proposed, of which the effectiveness and economic efficiency are verified by an actual case of a terminal.

In modern ship degaussing systems, degaussing windings are mainly distributed based on the shape of ship bulkhead, which is difficult to ensure the degaussing effect of magnetic induction intensity of unit winding of each degaussing winding. In order to solve this problem, this paper introduces a tilted correlation screening in high-dimensional variable filter, which splits and recombines the original coils, and re-divides the original degaussing sections, so as to improve the degaussing efficiency of each coil. Aiming at the problem of sparse parameter vectors and multiple collinearity in the calculation of degaussing current after winding restructuring, this paper proposes a slant correlation screening and partial ridge regression algorithm. Through simulation, when the threshold is 0.73 and 0.91, the algorithm reduces 10.08% and 17.59% respectively compared with the least square method, while the residual root mean square error decreases by 10.45% and 12.17%. The simulation results show that the degaussing effect is significantly improved after the algorithm is adopted.

Aimed at the problem of low efficiency of ship pipeline design, an optimization method of pipeline layout is proposed. An optimization mathematical model is established by comprehensively considering the engineering background of safety, economy, coordination and operability, and the defects of ant colony optimization algorithm in dealing with mixed pipeline layout conditions are improved. A spatial state transition strategy for optimizing feasible solution search, a pheromone diffusion mechanism for improving pheromone inspiration effect and accelerating algorithm convergence are proposed, and a multi-ant colony co-evolution mechanism is designed for mixed pipeline layout conditions. Based on the secondary development technology, the application of this method in the third-party design software is realized, and verified by a nuclear primary pipeline layout project. The results show that the pheromone Gaussian diffusion multi ant colony optimization (PG-MACO) algorithm has a better performance and layout effect than the traditional ant colony algorithm. The routing efficiency is improved by 58.38%, the convergence algebra is shortened by 43.24%, the pipeline length is shortened by 33.88%, and the number of pipeline bends is reduced by 41.67%, which verifies the effectiveness and engineering practicability of the proposed method.

An accurate and timely monitoring for the inertia support capability of the point of interconnection of aggregated sources to the grid in a low-inertia new power system is crucial for the safety, stability, and economic operation of the system. In order to explain the basic idea of the online point-to-grid inertia monitoring method, the definition of inertia of power system based on the swing equation and existing online monitoring methods are analyzed. Then, in order to improve the accuracy of the existing online inertia monitoring method, an equivalent inertia constant identification method based on the regression method is developed. Combining the proposed inertia constant identification method with the online inertia monitoring method, a systematic method for online monitoring of the inertia support capacity of point-to-grid in new power system is developed based on synchronous phasor measurement units. Finally, the simulation analysis of a modified New England 10-machine 39-bus system proves the accuracy and the feasibility of the developed real-time inertia monitoring method for the new power system.

The quantitative evaluation of the uncertainty in distributed photovoltaic power is significant for the safe and stable operation of distribution network. Considering the significant differences in power characteristics of different output fluctuation patterns, in order to obtain a prediction model suitable for different fluctuation patterns and to perform a refined assessment of power uncertainty, this paper proposes a method for pattern recognition and ultra-short-term probabilistic forecasting of power fluctuating in aggregated distributed photovoltaic clusters. First, the satellite cloud images and photovoltaic power data are integrated, and the pattern recognition model of fluctuation is constructed via the feature extraction of power fluctuation, realizing the mining of fluctuation rules. On this basis, the difference in predictability of different fluctuation patterns and the correlation between fluctuation patterns and prediction errors are considered via classification modeling, so that the width of prediction interval can better adapt to the characteristics of prediction error distribution. Thus, refined consideration of power uncertainty of different fluctuation patterns is realized to improve the precision of probabilistic prediction, provide more references for power grid dispatching, and weaken the influence of the strong volatility in distributed photovoltaic power on the power system.

Although the adaptive bistable wave energy generation device solves the problem that the bistable system may be difficult to cross the barrier when the amplitude of the incident wave is small, its efficiency can still be improved. Previous studies have proved that the change of the parameters of the device will have a great impact on its performance, and the optimal device parameters are closely related to the spectral peak frequency at a given time. Therefore, in the control study of the device, a control scheme is designed and the device parameters are adjusted accordingly in order to improve efficiency assuming that the peak frequency within a period of time is predictable. In this study, three control parameters are selected, and the optimal device parameter library with different spectral peak frequencies is determined by simulation calculation. The control module is then added to the simulation program to control the parameters by interpolation. The results show that the device with variable parameter control improves energy capture efficiency.

In order to predict the crashworthiness of.pngfened plate structures of ships under eccentric vertical quasi-static ballast of wedge, a new simplified analytical method is proposed in this paper. The plastic deformation area of the rectangular plate is divided into eight asymmetrical plates and the aggregate is divided into two asymmetrical parts. An linear mode is adopted in the deformation of rectangular plate and aggregate. From the perspective of the internal dynamics of ship collision, and based on the rigid-plastic theory, a simplified analytical method is deduced between the deformation damage resistance and lateral deformation of rectangular plate and aggregate under the eccentric vertical action of wedge impact, considering tensile and bending effects of the membrane. The proposed method is verified by the simulation results of nonlinear finite element software Abaqus. The results show that the proposed method has a high prediction accuracy, and can be used in the hull structure design stage to quickly predict the crashworthiness of the ship side structure.

The rapid development of electric vehicles (EVs) and unmanned aerial vehicles (UAVs) provides new ways for personnel search and material distribution during emergency periods. This paper proposes an EV-UAV joint rescue system, in which the UAVs use the EVs as charging and maintenance base stations to provide various services for the objects to be rescued, and the EVs can use distributed generations to obtain diversified electricity supply, which improves the adaptability and endurance level of the system in emergencies. The coordinated scheduling model of the EV-UAV system is established in the mixed-integer linear programming (MILP) formulation, which considers factors including electricity consumption, electricity replenishment, loading capacity, distribution route, and distribution time window of the EVs and the UAVs. Case studies verify the validity of the model proposed, compare the EV-UAV and ground vehicle (GV)-UAV rescue systems, and illustrate the technical characteristics and application potential of the EV-UAV system in emergency assistance.

Agricultural microgrids offer a promising solution for energy supply in remote rural areas in a low-cost manner. In this paper, under uncertain conditions of renewable energy output and electricity load demand, a robust optimal scheduling model combined with the isolated agricultural microgrid and irrigation system containing a pumped hydro storage (PHS) power station is proposed, considering the factors that the wind-landscape pumped storage integrated agricultural microgrid can satisfy the uncertain fluctuations of power load demand and water load demand. By utilizing the abundant water resources in rural areas and the advantages of landscape drainage and storage compensation, the total cost of the system is minimized while the absorption of renewable energy is increased. Considering distributed generation, power load demand and water load demand, turbine flow, and irrigation flow, the proposed model is characterized by diversity, multi-constraint, and discontinuity. A gravitational whale optimization algorithm (GWOA) is proposed to solve the model. The simulation results of an agricultural microgrid show that the GWOA can obtain a more competitive solution than the CPLEX solver and other newly developed algorithms do. In addition, the impact of the change of water load demand caused by precipitation uncertainty on the operating cost of the irrigation system and the necessity of using PHS power station are explored.

With the increasing penetration of new energy sources and the development of new power systems, grid-following converter (GFL) plays a crucial role in maintaining the stability of power systems. However, existing transient stability analyses of GFLs assume that the direct current (DC) side behaves as a constant-voltage source, neglecting the effects of DC-bus voltage control. This paper aims to investigate the transient instability mechanism of GFL considering DC-bus voltage control. First, a transient synchronous stability model considering DC voltage control is established, followed by an analysis of the transient synchronous stability of GFL under DC-bus voltage control. The findings indicate that DC voltage control increases the active current reference value and decreases the equivalent damping of the GFL, which in turn reduces its transient synchronous stability of GFL. By increasing the proportional coefficient or reducing the integral coefficient of DC-bus voltage control, transient synchronous stability can be appropriately improved. Finally, the theoretical analysis is validated through MATLAB/Simulink simulations.

To meet the construction demand of clean energy demonstration bases, a gain allocation strategy for the joint optimization operation of wind-solar-pumped storage-hydrogen multi-stakeholder energy system based on the cooperative game theory is proposed. In order to take into consideration the security of system operation, evaluation indicators for the complementarity of on-grid output are constructed. The stakeholders of wind, solar, pumped storage, and power-to-hydrogen cooperate through the internal electricity transaction to construct a joint scheduling model with the optimization goal of maximizing the operation benefits. Then, the minimum cost remaining saving (MCRS) method in the cooperative game theory is applied to allocate the synergistic benefits based on the scheduling results. The simulation results of a 12-stakeholder wind-solar-pumped storage-hydrogen clean energy demonstration base show that each stakeholder can derive positive gains through joint operation, and the reservoir capacity of pumped storage station, on-grid price and operation security demand will affect the cooperative synergistic benefits of the system.

How to make full use of existing information to improve the accuracy of fault diagnosis in distribution networks, and provide accurate research and judgement for emergency repair of distribution networks, is an urgent problem to be solved. To address the problem of the single source of fault diagnosis information in existing distribution networks, a fault diagnosis model of distribution network is proposed which integrates the medium and low voltage information of the distribution networks and the outgoing current information of the substation. The model first applies the existing overcurrent diagnosis method to the problem of large-scale distribution network, and adopts hierarchical reduction of the size of the distribution networks to improve the location speed of fault section. Then, in view of the accuracy of overcurrent alarm information, an auxiliary fault judgment method for distribution networks based on switch relay protection sequence of events (SOE) data and substation outgoing load sag data is proposed. Finally, the steps for fault diagnosis in distribution networks of multi-directional information and data fusion in practical engineering are summarized, which provides reference for fault diagnosis of dispatchers. Engineering practice proves that the method proposed in this paper can effectively diagnose faults and is very adaptable to large-scale distribution networks. The auxiliary diagnosis model combining switch operation SOE and telemetering voltage information can compensate for the accuracy requirements of the overcurrent diagnosis model for remote communication information, which is complementary to each other and has a good engineering value.

The energy management of the electricity-heating integrated energy system (IES) is related to the economic benefits and multi-energy complementary capabilities of a park, but it faces the challenges of the randomness of renewable energy and the uncertainty of load. First, in this paper, a mathematical model of the energy management problem for the electricity-heating IES is conducted, and each energy supply subsystem is empowered as an agent. Based on the deep deterministic policy gradient (DDPG) algorithm, a system energy management model is established that comprehensively considers the real-time energy load of the subsystem, the time-of-use pricing, and the output of each equipment. Then, the federated learning technology is used to interact with the gradient parameters of the energy management model of the three subsystems during the training process to synergistically optimize the training effect of the model, which can protect the data privacy of each subsystem while breaking the data barriers. Finally, an example analysis verifies that the proposed federated-DDPG energy management model can effectively improve the economic benefits of the park-level IES.

Under the low-carbon development goal, energy storage allocation is the key measure to ensure the safe and economic operation of low-carbon parks, and to reduce carbon emissions. To solve the problems of inaccurate carbon emission calculation and insufficient utilization of equivalent energy storage resources in low-carbon parks, this paper proposes a dynamic emission factor calculation method based on the carbon emission flow theory, which realizes the accurate measurement of indirect carbon emissions from park electricity consumption. Then, taking into account the available equivalent energy storage resources in the park, it proposes an energy storage capacity optimization allocation model considering the equivalent energy storage characteristics of thermal system, and uses the big M method to equivalently transform the nonlinear constraints in the model. Finally, it conducts simulation analysis based on a case system to verify the correctness and effectiveness of the proposed model.

A droop-free distributed energy storage control strategy based on the diffusion algorithm is proposed to address the inability of droop control to simultaneously achieve bus voltage stability and accurate power distribution among energy storage systems when the line impedance is mismatched and bus voltage is inconsistent. First, the diffusion algorithm is applied to the distributed estimation of direct current (DC) microgrid to obtain the global average, and the difference between the voltage rating and the average voltage is used as a compensation term to restore the bus voltage deviation. Then, in order to achieve the accurate distribution of power between energy storage systems with different rated capacities and different states of charge (SOC), a standard power of energy storage is designed and the difference between the standard power and the average standard power is used as a compensation term for the equilibrium of SOC between energy storage systems. Finally, a model based on RT-LAB is built to verify the effectiveness of the designed control strategy in four different operating modes. The experimental results show that the proposed control strategy can achieve the bus voltage recovery and the accurate distribution of energy storage power in the isolated DC microgrid.

In order to study the difference of dynamic response between large diameter variable section single pile and pile group foundations in seismic subsidence soil with different thicknesses under ground motion, based on the Xiang’an Bridge of Xiamen Second East Channel, a shaking table model test was conducted to study the difference of dynamic response of soil subsidence, horizontal displacement of pile top, acceleration of pile body and bending moment between single pile and pile group foundation when the thickness of seismic subsidence soil layer is 30, 40, and 50 cm. The results show that with the increase of the thickness of seismic subsidence layer, the seismic subsidence, horizontal displacement of pile top, acceleration and bending moment of single pile and pile group foundations increase gradually, and the acceleration and bending moment change abruptly at variable section. In the seismic subsidence layer with the same thickness, the seismic subsidence of soil around pile group foundation is larger than that of single pile, but the acceleration of pile group foundation, horizontal displacement of pile top and bending moment of pile body are smaller than those of single pile. It is suggested that in the design of pile foundation in seismic subsidence site, the difference of dynamic response between variable section single pile and pile group should be mainly considered to ensure the seismic performance of pile foundation.

It is a significant technical challenge to exploit deep-sea polymetallic nodules with high efficiency and low disturbance. The mechanical behavior of seabed nodule collecting is very complicated, which is a multi-physical coupling process involving three-dimensional turbulent flow, discrete particle movement, and fine particle soil failure. In this paper, three main deep-sea hydraulic nodule collecting methods, i.e., the suck-up-based method, the Coandă-effect-based method, and the double-jet hydraulic method, are investigated by numerical simulation on the performance of nodule collecting and environmental disturbance. The realizable K-Epsilon two-layer model and the discrete element method are used to simulate the turbulent flow of the liquid phase and nodule particles in the solid phase respectively. The effect of collection flow q_m and drag velocity v on collection rate η, turbulent kinetic energy k, and volume fraction φ of the seawater-sediment mixture in the collecting flow field is analyzed. The flow velocity, pressure, and nodule distribution are explored. The results indicate that, at the same q_m and v, the double-jet hydraulic model will achieve the largest η, while the suck-up-based model will achieve the least η. The double-jet hydraulic model has the most significant disturbance to the near-bottom flow field and the most obvious sediment spreading phenomenon. In contrast, the suck-up-based model and the Coandăeffect-based model have less environmental disturbance, which is more conducive to the requirements of environmental protection. The Coandă-effect-based model shows minor sensitivity to q_m and v and a good balance between high nodule collecting efficiency and low environmental disturbance. This paper will provide a scie.pngic basis for revealing the nodule collecting mechanism and designing and developing a nodule collecting device.

The uncertainty of wind power output and the difficulty of power storage restrict the development of new energy. As a high-quality secondary energy, hydrogen energy is green and pollution-free and has a high energy density. In order to cope with the volatility and randomness of new energy output, this paper proposes a multi-time scale probabilistic production simulation method for wind-solar hydrogen integrated energy system considering hydrogen storage. First, thermal energy recovery is considered in the hydrogen storage system model, and the wind-solar hydrogen integrated energy system model including electrothermal hydrogen multiple energy storage is constructed. Then, multi-time scale probabilistic production simulation is conducted for the wind-solar hydrogen integrated energy system, and the system maintenance arrangement and hydrogen storage seasonal distribution scheme are obtained through medium and long-term production simulation. The simulation results are taken as the boundary for short-term production simulation to achieve the cooperation of electrothermal hydrogen multiple energy storage and to smooth the random fluctuation of wind and solar power. Finally, an IEEE-RTS79 node example is given to verify that the proposed method can improve the reliability, flexibility and low-carbon feature of system operation.

The horizontal vibration model of partially buried pile group under vertical load is established based on Biot saturated porous media theory, considering the flow characteristics of liquefied soil. Considering the pile-soil coupling and continuous displacement conditions, the pile interaction factor solutions and the horizontal dynamic pile impedance solutions in liquefied soil under complex conditions are obtained using the variable separation method and the operator decomposition method. Parameter studies are conducted on liquefied soil characteristics and vertical loadings. The results show that at the same vibration frequency, the horizontal dynamic.pngfness of pile group decreases with the increase of surface liquefied soil thickness. When the liquefaction thickness is large, the dynamic.pngfness decreases significantly with the increase of frequency, and a negative.pngfness appears. Vertical pile top loading can reduce the dynamic.pngfness of pile group in liquefaction soils. The weakening effect becomes more obvious with the increase in the thickness of liquefied soil.

The constant power load (CPL) in a DC microgrid can reduce the effective damping of the system, resulting in high frequency voltage oscillations on the DC bus, which threatens the safe and stable operation of the system. To address this issue, this paper proposes a DC-bus voltage oscillation suppressor based on an active capacitor and its control method. The oscillation suppressor is connected in parallel to the DC bus, enabling direct interaction with the DC bus. The energy storage capacitor in the oscillator suppressor effectively stores the transient energy generated by voltage oscillations, thereby reducing the amplitude of voltage oscillation and improving the voltage stability of the bus. The voltage of the power supply in the oscillation suppressor adapts to the voltage of the DC bus, allowing for stable operation in the face of load changes in the system. The design offers advantages such as plug-and-play functionality, strong applicability, and flexible control. In addition, by analyzing the operating mode and mechanism of the oscillation suppressor, a small signal model is established, and the influence of controller parameters on the stability and dynamic performance of the suppressor is analyzed, based on which the controller parameter optimization scheme is proposed. Finally, the effectiveness of the oscillatory suppressor is validated through the experimental results.

To address the challenges of complex detection background and poor detection performance for small targets, a transmission line channel security detection algorithm based on the fusion of window self-attention network and the YOLOv5 model is proposed. First, the Swin Transformer (S-T) is employed to optimize the backbone network, expanding the perception field of the model and enhancing its ability to extract effective information. Then, the adaptive spatial feature fusion (ASFF) module is improved to enhance the feature fusion ability of the model. Finally, considering the mismatch between the real frame and the predicted frame, the structural similarity intersection over union (SIoU) is introduced to optimize the boundary errors and improve the generalization ability of the model. The experimental results show that the model proposed achieves a multi-target intrusion detection accuracy of 90.2%, and with significant improvements in the detection of small targets. This approach better meets the requirements of foreign object detection in transmission line channels compared to other object detection algorithms.

The continuous increase in renewables penetration poses a severe challenge to the frequency control of interconnected power grid. Since the conventional automatic generation control (AGC) strategy does not consider the power flow constraints of the network, the traditional approach is to make tentative generator power adjustments based on expert knowledge and experience, which is time consuming. The optimal power flow-based AGC optimization model has a long solution time and convergence issues due to its non-convexity and large size. Deep reinforcement learning has the advantage of “offline training and online end-to-end strategy formation”, which yet cannot ensure the security of artificial intelligence (AI) in power grid applications. A coordinated optimal control method is proposed for active power and frequency control based on safe deep reinforcement learning. First, the method models the frequency control problem as a constrained Markov decision process, and an agent is designed by considering various safety constraints. Then, the agent is trained using the example of East China Power Grid through continuous interactions with the grid. Finally, the effect of the agent and the conventional AGC strategy is compared. The results show that the proposed approach can quickly generate control strategies under various operating conditions, and can assist dispatchers to make decisions online.

The cavity expansion theory provides a theoretical basis for the displacement prediction of the installed piles and the analysis of pressuremeter test results. However, there is a lack of systematic analysis of the influences of soil structure in the existing solutions. In this paper, based on the Cemented-CASM considering the cementation effect, an elastoplastic solution for the cylindrical cavity expansion under undrained condition was presented by deriving the elastic-plastic.pngfness matrix reflecting the soil stress-strain relationship in combination with the hardening laws and larger-strain theory. To verify the validation of the present solution, it was compared with the solution in Modified Cam Clay soils when ignoring the soil structure and compared with the numerical results of pressuremeter tests when the soil structure was considered. Parametric analyses were conducted to illustrate the effects of soil structure and destructuration on the cavity expansion process. The results show that the effective stresses and pore water pressure caused by the cavity expansion in structured soils are larger than those in reconstituted soils, while the plastic range around the cavity becomes smaller due to the soil structure. With the damage of the structure, the soil exhibits a strain softening behavior, while the radial stress increases and then decreases as the distance from the cavity wall decreases. When the soil structure is completely destroyed, the effective stresses around the cavity are consistent with those in the reconstituted soils. As the soil structure strengthens, destructuration become faster under the same variation in cavity expansion pressure or cavity radius, and the strain softening behavior becomes more pronounced.

With the rapid development of commercial complexes in recent years, energy consumption and carbon emissions of buildings are growing continuously. In this context, the low-carbon planning of commercial complexes is studied including shopping, restaurants, offices, and accommodation. In addition, a low-carbon planning model for building considering ladder carbon reward and punishment with the introduction of the time-sharing carbon measurement model of superior network power purchase and an integrated demand response (IDR) considering the differentiated predicted mean vote (PMV) of each functional area of the building is established. First, the time-sharing carbon measurement model of superior network power purchase is introduced to evaluate the equivalent carbon emissions of the building. Then, a ladder carbon reward and punishment model is built to measure the carbon emissions of the building. Based on the characteristics of the commercial complex planned in this paper, the IDR considering load time shifting and energy use reduction, end-use energy equipment substitution, and energy use type conversion at the superior energy end is constructed. Afterwards, a low-carbon planning model for building is established to determine the equipment type and capacity considering the PMV for each functional area of the building. Especially, the objective function is to optimize the total annual planning cost of the building by taking into account carbon reward and punishment costs of the building. A distributionally robust optimization model based on Kullback-Leibler divergence is proposed to cope with the output volatility of the connected distributed photovoltaic and photovoltaic curtain wall. Finally, the effects of the ladder carbon reward and punishment mechanism of the time-sharing carbon measurement model of superior network power purchase, the IDR of the differentiated PMV of each functional area of the building, and the volatility of the connected photovoltaic output on the building planning are analyzed to verify the effectiveness of the planning model proposed for energy saving and emission reduction in the building.

The multicollinearity of measurement data leads to the low accuracy of the data-driven methods for estimating voltage-power sensitivity in distribution networks. In this paper, a data-driven method embedded with topological information is proposed to address the problem. First, the voltage-power sensitivity matrix is decomposed into principal and secondary components, where the principal component is closely related to the distribution network topology and the secondary component is the error between the principal component and the actual value. Then, the principal and secondary components are estimated sequentially in two stages, and their data-driven estimation models based on quadratic programming are established, respectively. The key of the model in the first stage is the constraint based on the distribution network topology information, and the key of the model in the second stage is the constraint that the ratio of the secondary component to the principal component is tiny. Finally, the accuracy and efficiency of the proposed method is validated in the IEEE 33-bus system with a set of measurement data, and comparisons are made with ordinary least square regression, ridge regression, and LASSO regression. The simulation results show that the accuracy of the proposed method is significantly improved by orders of magnitude.

Short-term load forecasting is generally applied in power system real-time dispatching and day-ahead generation planning, which is of great significance for power system economic dispatching and safe operation of the system. Many researches on short-term load forecasting using smart models have been conducted at home and abroad. However, how to obtain the optimal structure and parameters accurately and quickly poses a challenge to short-term load forecasting, because the prediction performance of smart forecasting methods is more easily affected by the structure and parameters of the method, and the personality difference of the prediction object itself makes it difficult for the parameters to be reused. Aiming at this problem, a tree-structured Parzen estimator (TPE)-based boosting short-term load forecasting method is proposed. The results show that the proposed method can achieve rapid optimization of structure and parameters, which is verified in the application in short-term load forecasting of a southern province in China to improve the prediction accuracy.