With the further development of China’s electricity market’s reform, the marketization brings price risk to the market participants. Moreover, under China’s Carbon Neutrality strategy, the electricity market participants will face not only the traditional risk, but also the carbon risk. As an important tool for risk management, electricity futures are widely used in many mature electricity markets and have attracted attention from the market participants in China. To quantitatively analyze the impact of introducing electricity futures on the spot market, this paper proposes a bi-level electricity simulation model. The first level models the transaction behavior of electricity futures, and the second level models the optimal bidding strategy of generation companies under the market with power futures. Numerical experiments simulate a six-unit power market environment with two different contract signing methods. The simulation results show that the introduction of electricity futures can help the spot market to balance the price in both peak and valley period, thus helping the electricity market in both scenarios.
This paper constructs a master-slave game model with embedded non-cooperation and cooperation game to explore the impact of coal-fired power unit decision-making on the economic efficiency and carbon emissions from a system perspective under different proportion of new energy penetration. The model takes into account the technical characteristics of coal-fired units to accurately account for the total emissions of coal-fired units in different operating conditions. The results show that, when the new energy penetration rate is low, the cooperation strategy among coal-fired power units has good performance in both economic and environmental benefits. With the increase of new energy penetration, the cooperation strategy among coal-fired power units is still helpful for carbon reduction, but the overall economy will decrease. When the level of new energy penetration is high, the non-cooperative strategy can enhance both economic and carbon reduction effects. Some valuable policy implications are proposed according to the research findings.
With the goal of carbon peaking in 2030, the carbon peak of thermal power industry in the energy industry has attracted much attention. In this paper, firstly, according to the extension of Kaya’s constant equation, the main factors affecting carbon emissions are analyzed and obtained: population, economy, industrial structure, energy consumption intensity and energy consumption structure. Secondly, linear regression, RBF neural network, ARIMA and BP neural network models are established according to data from 2000 to 2018, to get the optimal prediction models by comparison. Finally, on the basis of the optimal model, carbon emissions from 2021 to 2050 are predicted under four different development scenarios: baseline development, industrial optimization, technological breakthrough, and low-carbon development, and then carbon peak situation is analyzed on this basis. The results show that it has the earliest peak time and lowest peak value in the low-carbon development scenario, which is the preferred mode to achieve peak carbon emissions in China’s thermal power industry, and provides a reference for promoting the thermal power industry to achieve lower peak carbon emissions as soon as possible.
Through the analysis of the modular multilevel converter (MMC) with the traditional d-q axes decoupling controller, this paper reveals the coupling effect between the DC voltage and the sub-module capacitor voltage, which also results in the equivalent capacitance on the DC port and the deviation of the capacitor voltage. Through considering the sum of the upper and lower arm voltages as the control variable, the DC current is introduced as the third inner-loop state variable, and a three-axis decoupling controller with three inner-loop state variables is proposed. Applying the introduced DC current state variable, the control of the DC voltage can be decoupled from the sub-module capacitor voltage. The direct control of the sub-module capacitor voltage and the fast and flexible control of the DC voltage/current can be realized simultaneously. On the basis of the proposed inner-loop three-axis decoupling controller, various outer-loop controllers are designed for different operating modes of MMC, which can flexibly realize the control of DC voltage, active power, reactive power, and capacitor voltages. The feasibility and effectiveness of the proposed three-axis decoupled controller have been verified by using simulation results for different operating modes.
Since the level number of modular multi-level converters (MMC) used in medium voltage DC systems is small, it is difficult to operate with lower operating losses and higher voltage quality. Thus, a level multiplication MMC (LM-MMC) topology and level multiplication modulation (LMM) strategy is designed. By introducing B-submodules with capability of parallel output non-integer level, the output level of LM-MMC can be multiplicated. Secondly, the operation mode of submodule is rotated according to the sorting result, which can effectively balance the capacitor voltages. In addition, when the LMM strategy is applied to the LM-MMC topology, the sum of output level in the upper and lower arms is constant, and the circulating current of the converter is lower. Finally, according to the simulation and physical experiment results, the operating loss of LMM strategy applied to the LM-MMC topology is lower, and the output waveform quality is higher.
As the cooling, heating and power system for traditional buildings ignores the energy storage characteristics and does not consider the participation of energy storage in decision-making, it cannot adjust the energy storage pressure, thus resulting in energy waste. Therefore, a hybrid load-following operation strategy considering energy storage characteristics is proposed in this paper. At the same time, in order to reduce the amount of model calculation and accurately select typical day of the load, a K-means clustering combined with CRITIC weighting is proposed. The system is optimized from three aspects: economy, energy and environment. The optimization results show that the K-means clustering algorithm with CRITIC weighting has higher contour coefficient and better clustering effect. At the same time, the comprehensive performance of the operation strategy proposed in this paper reaches 29.66%, which is much higher than the electric load following strategy and heat load following strategy. Compared with the traditional operation strategy, the proposed operation strategy has the best comprehensive performance, which verifies the feasibility and superiority of the proposed operation strategy.
Rural integrated energy system can effectively improve energy efficiency and economy while meeting the diversified energy demand of rural users through the coordination and complementarity of various energy sources. Firstly, considering adapting to typical rural scenarios, a three-layer cooperative self-regulation framework for hierarchical cooperative operation optimization of rural integrated energy is proposed. Then, according to the typical equipment of the rural integrated energy system, a combined optimization dispatching model of the three-layer rural integrated energy system, including source, storage and load, and the corresponding optimization dispatching process, are established. In the dispatching model, residual heat of biogas generator unit and air-source heat pump cooperate to provide heating for users in winter. In summer, the residual heat of biogas generator set is recovered, and the rural users are cooled by the combination of lithium bromide refrigerator and air-source heat pump. Finally, an example is given to analyze the multi-layer collaborative optimization method for rural integrated energy system. The results show that the optimization method can improve the economy of rural residents’ energy consumption, and the effectiveness of the proposed method is verified.
In the case of planned power outage in the distribution network, the home microgrid based on cloud energy storage enters the island operation state. Considering the limited available energy storage capacity of cloud energy-storage users and the uncertainty of renewable energy output, the strategy of emergency energy management is proposed to reduce user’s power outage loss and user dissatisfaction. Firstly, the cloud energy storage is divided into power storage and transaction storage. The capacity of the power storage can ensure the power supply of user’s important loads during the planned power outage; Secondly, the emergency energy management model is established with the goal of reducing user power outage loss and user dissatisfaction, and the weight factors is introduced to measure the importance of the two. Then the genetic algorithm is used to solve the model, and the optimal solution is the optimal scheduling scheme for user’s emergency energy. Finally, it is verified through experimental simulation that the proposed optimization strategy can effectively reduce user power outage loss and user dissatisfaction.
In order to solve the problems of high cost and untimely communication caused by the lack of mutual trust among various subjects in the process of demand response, on the basis of combing the general technical principles of block chain, this paper presents the overall architecture, transaction flow, data chain design, intelligent contract design, network building scheme and consensus mechanism of the blockchain demand response scheme, and carries out benefit analysis from multiple dimensions. It provides a train of thought for not only making use of the advantages of centralized management, but also providing non-centralized deposit certificate and supervision function. The simulation builds the corresponding alliance chain network on the demand composed of seven nodes, verifies the feasibility of the architecture and model, and gives an example of interface design. By building the benefit function of demand response blockchain application and taking the demand response scenario in Henan as an example, this paper demonstrates that the application of block chain technology to demand response is economic and feasible.
Aiming at the large number of power equipment defect texts accumulated in the daily operation of distribution network, deeply mining the characteristics of defect texts and learning the power equipment situation is an important direction for the refinement development of intelligent distribution network. However, there are few research results in this field in China, and there is lack of analysis of key challenges and targeted solutions. Therefore, this paper comprehensively evaluates the main technologies of text mining of power equipment defects in China, and analyzes the difficulties faced. Taking the power equipment defect text as the research object, this paper firstly introduces the key technologies of text mining of power equipment defects from error identification and quality improvement, and then expounds other text mining technologies for power equipment defects in intelligent distribution network from the aspects of automatic classification of defect severity level, defect detail extraction and automatic health state evaluation. Furthermore, combined with the development trend of intelligent distribution network in the future, the future of text mining technologies for power equipment defects is prospected, in order to provide reference for lean operation and maintenance of intelligent distribution network.
To prevent the security problem of cascading failures caused by source faults, this paper proposes a preventative control strategy of cascading failure considering security and the economy. Firstly, according to the action characteristics of the relay protection, a mathematical form is given to discriminate the cascading failure. Secondly, the security and economy of the system are evaluated in terms of power grid security margin and operating cost of power generation, respectively. The initial failure is selected by the branch vulnerability assessment method to construct a preventive control model of cascading failure for different initial failures actions. The model is a two-layer optimization mathematical model. The inner model is solved by the particle swarm optimization algorithm to get the power grid security margin. The outer model is solved by the improved multi-objective particle swarm optimization algorithm to minimize generation cost and maximize power grid security margin. Finally, the calculated Pareto set is evaluated using fuzzy set theory to determine the optimal output strategy of generators. Simulation studies are conducted with the IEEE 39-bus system and IEEE 118-bus system to verify the feasibility of the proposed method in this paper.
After the Yu-E back-to-back flexible direct current system was put into operation, the asynchronous interconnection of the Southwest and Central China power grids was realized, and an LCC/VSC hybrid infeed power system including the Southwest power grid and Three Gorges generators, VSC and LCC such as Longzheng HVDC was formed. Aiming at the system stability requirements after AC or DC failures under such a typical structure, this paper considers giving full play to the fast adjustable characteristics of VSC and LCC, and proposes a coordinated control system involving AC lines, generator sets, VSC and LCC. The core is the coordinated control strategy of inertial support of VSC/LCC, which can realize the inertia supporting and auxiliary frequency modulation of the entire hybrid system. An electromagnetic transient simulation system including three-region AC power grids and detailed models of VSC and LCC is established, and typical AC and DC faults that may cause instability have also been analyzed and simulated. Under the action of coordinated control strategies, the system can quickly recover to new stability operation status, so as to meet the needs of dynamic power coordinated control of AC/DC power grids in the near area.
Power load profile clustering usually classifies load profiles according to shape difference and numerical difference of the curves. In this paper, an ensemble clustering algorithm based on granular computing and dual-scale similarity is proposed. The K-means algorithm, which takes Euclidean distance and Pearson correlation coefficient as similarity measures, is used to generate base clustering. Then the part of base clustering is selected to participate in the ensemble algorithm through granular computing. Finally, the similarity matrix is generated and the hierarchical clustering is used to obtain the final clustering. The result of the experiment shows that, the proposed algorithm can overcome the limitation that the traditional load profile clustering can only measure the load similarity from the value or shape, and significantly improves the quality of power load profile clustering.
In this paper, a group of integrated of and coordinated control strategy for new energy DC collection system is proposed, which is divided into two parts: power control and voltage control. Power control includes planned power control within the group and power control of connecting lines between groups. The planned power control within the group takes into account the time-of-use price of power grid and energy storage, which can improve the economy of system operation. The connection-wire power control based on DC group control error (DGCE) is adopted for the power control of the connecting line between clusters, which not only ensures the power interchange benefit between clusters, but also realizes the independence of clustering operation as far as possible. The voltage control adopts voltage secondary control combined with droop control based on distributed consistency algorithm to make the DC collection system run at a better voltage level. Finally, MATLAB/Simulink simulation software is used to verify the operation state of DC collection system in unit working day. The results show that the economy and voltage level of the system are improved after the control strategy is adopted, and the power flow is more reasonable.
Under the background of new power systems dominated by renewable energy, it is necessary to grasp the grid-connection performance of renewable energy station in real time. Firstly, grid-connection performance evaluation of renewable energy station under distributed framework is proposed in this paper, and the application platform architecture under this framework is designed by introducing edge computing idea. Then, the index system for grid-connection performance evaluation of renewable power station is established and the calculation methods for key parameters are given. Finally, an application example of frequency response capability evaluation is presented. The practice shows that the PLL method has certain advantages in accuracy and convergence for frequency parameter estimation. In addition, because of the advantages in data accuracy, remote communication and computing efficiency, the idea of distributed evaluation can provide a solution for panoramic monitoring and control of renewable energy stations.
