Abstract:The increasing number of electric vehicles（EVs）makes the uncertainty in the microgrid continue to increase. Aiming at the uncertainty of the arrival time of aggregated EVs and the remaining power at the arrival time, a two-layer model predictive control strategy is established to perform optimal charging and discharging management for EVs connected to the microgrid to minimize the power exchange between the grid and the microgrid. In the optimization of the charging and discharging of the upper-level aggregated electric vehicles, the charging demand of the lower-level individual electric vehicles is considered. According to the urgency of the user’s charging behavior, EVs that have arrived are divided into EVs participating in optimization and EVs without participating in optimization. The simulation results show that the proposed method has better performance in the charge and discharge management of aggregated EVs. In addition, it is closer to the reality and easier to meet the needs of residents.

Abstract:Accurate wind power forecasting is very crucial for the security and stability of power system operation. From a statistical standpoint, a dynamic harmonic regression method is proposed for very-short-term wind power forecasting. Cubic polynomials between wind power and wind speed at different heights are used to construct the regression model. Then, autoregressive integrated moving average model is proposed to model the regression residual in order to fully use historical information of wind power time series. Finally, according to the daily seasonal characteristics of wind power, fourier series is introduced and the final model is established. Results from real-world wind farms show that this method can effectively improve the traditional autoregressive integrated moving average model and regression methods, which can reduceroot mean squared error and improve the prediction accuracy. Compared with two commonly-used existing approaches, persistence and autoregressive integrated moving average model, the proposed model is verified to have higher prediction accuracy, indicating that this method has certain practical application value.

Abstract:The problem of“dual power supply single meter”refers to the use of one energy meter for the terminal load requiring dual power supply, which makes it difficult to determine which transformer area the power consumption of this load is calculated in when the dual power supply is switched. In practice, due to the poor communication of marketing and distribution integration, the dual power supply switching of important loads in the transformer area can not be handled in time, resulting in unqualified line loss. A dual power supply switching sensing method based on power line carrier communication channel impedance is proposed. When dual power switching occurs, the network of the transformer area to which the load metering point belongs changes. The change of input impedance here is sensed through the frequency sweep signal injected by the dual power metering point, so as to realize the automatic identification of dual power switching. Through the active sensing method,the sensing function of dual power switching is increased while retaining the original metering function. By setting a reasonable decision threshold, several groups of confirmatory experiments proved the effectiveness and scientificity of the proposed method.

Abstract:Aiming at the large-scale power outage caused by frequent extreme disasters, a resilience improvement technology of distribution network considering demand response is proposed. Ba-sic concept of distribution network resilience is combed, and two comprehensive resilience indexes, load resilience and DG strength,are put forward to quantitatively evaluate the distribution network resilience. The incentive-based demand response node model and is--land resilience model of distribution network are established. Ac-cording to the characteristics of large fluctuation of DG output in distribution network, a three -stage distribution network resilience improvement strategy based on demand response is proposed. Considering the situation that the distribution network loses the power sup-ply of the main network in the event of extreme natural disasters, an initial island is formed firstly. Then, the demand response technology is used to expand the scope of the island, and finally the dynamic state of the island is updated. Through the improved IEEE 33 node example, the three-stage process of the strategy is described in detail, and two scenarios are set. By comparing the evaluation results of quantitative indicators, the effectiveness and resilience improvement ability of the strategy are verified.

Abstract:The vibration signal of planetary gearbox of wind turbine is a kind of non-linear and non-stationary complex signal.The traditional fault diagnosis method can deal with this kind of signal well in a limited range. The convolution depth belief network is established for planetary gearbox fault diagnosis. In order to prevent the wrong selection of hyper parameters from causing insufficient recognition accuracy, particle swarm optimization algorithm is introduced to optimize the hyper parameters of the network, and the chaos initialization of particles improves the global search ability of particles. Firstly, the original signal is decomposed by VMD to extract the eigenmode function which is relatively concentrated in the impact information as the input data of the network. Then, the training set is used to train, the chaos particle swarm optimization algorithmis used to determine the hyper parameters of the network according to the minimum fitness function, and the layer-by-layer greedy algorithm is used to continuously update the network parameters. Finally, the extracted fault features are classified by a classifier. This method is verified that it can diagnose the fault of planetary gearbox under different conditions.

Abstract:Aiming at the practical problems of low self -sufficient renewable energy utilization, high power purchase cost and in-adequate source -storage -charge coordination in the current energy management of active integrated energy systems, a data-driven real-time dispatching decision model is proposed. The hybrid model is firstly based on a large amount of energy consumption data for its characteristic indicators and time series characteristics, and the K -means algorithm is used for cluster analysis to obtain a representative typical energy consumption pattern. Then the parametric fuzzy inference system combined with the heuristic optimization algorithm is introduced, and the fuzzy logic inference rules are self-optimized based on the representative samples to obtain a real-time control model for the economical scheduling of energy - using systems.In the simulation calculation through the actual data set, the validity and feasibility of the model are verified.

Abstract:In order to improve the consumption rate of clean energy and reduce carbon emissions in the process of building pow-er supply, a flexible AC / DC power supply system topology is presented. The topology realizes the energy interaction of photovoltaic power generation, energy storage and electric vehicle charging and discharging through DC bus, and realizes the energy interaction of AC / DC system through bidirectional AC / DC equipment. In order to ensure the safe, stable and economic operation of the system, amulti time scale scheduling control strategy is proposed. The strategy combines energy storage, photovoltaic power generation, electricvehicle electrical characteristics, power reserve rate and energy re-serve rate in different time scales of millisecond, second to minuteand hour to day to dispatch various power electronic equipment flexibly. The system can maximize the photovoltaic consumption efficiency, reduce power conversion loss, maximize the role of electricvehicles as a flexible resource in energy regulation, and reduce the dependence of flexible DC system on external power supply. The effectiveness of this method is verified by building a MATLAB / Simu-link simulation system.

Abstract:Due to flexible location, short construction cycle and rapid cost reduction, chemical energy storage has been applied to power grid peak load regulation in recent years. However, re-search on factors sensitive to optimal capacity of battery energy storage system suitable for peak load regulation is still insufficient.A model about optimal capacity of battery energy storage for peak load regulation is constructed. The model takes lowest social energy consumption cost as a goal, and planned or forecasted data of allkinds of power source and load in target planning year as boundary condition. The model makes full time production simulation foreach alternative capacity scheme of battery energy storage through-out the year, and solves it by mixed integer linear programming.Based on the work above, the sensitive factors that affect the optimal capacity of battery energy storage are summarized, and empirical research and analysis on the mode and degree of its influence are carried out. The results show that the energy storage factor, including unit investment cost and operating years, as well as demand factor such as peak load supply and low load adjustment, caninfluence the optimal capacity of battery energy storage and the total system cost.

Abstract:In view of the traditional power load forecasting algorithm model of slow training speed and the prediction problem of poor effect, a parallel load forecasting method is proposed based on the deep belief network. Based on parallel computing framework and deep belief network, the method is parallel train the history pow-er load and weather information data, and load values is forcasted through the training model. The experimental results show that the average error between the predicted power load value and the actual value is low and the prediction accuracy is higher than the traditional method. It effectively reduces the elapsed time of consuming training and prediction, and can adapt to the prediction demand in large-scale power data scenarios.

Abstract:The penetration rates of electric vehicles and energy storage in the power system are increasing, and they can participate in the optimal dispatch of the distribution network as the deploy able resources. Firstly, based on the power flow model, energy-storage operation model and EV cluster charging station model, the economic dispatching model of distribution network is established.The goal of day-ahead dispatching is to minimize the network loss and the charging cost of energy storage and charging station. Secondly, a leader -followers distributed solution was designed based on ADMM to give full play to the autonomous coordination of re-sources in the distribution network. Finally, based on a 33-node ex-ample, the simulation results show that the energy storage and charging stations participating in the distribution network dispatching is a two-way benefit for the energy storage operator and distribution network, and the distributed solution based on ADMM is effective.

Abstract:Solid electric heat storage technology converts the off - peak electricity into the heat energy, which is of great significance to the severe electric peak load regulation of the power grid,the utilization of the abandoned wind power and the environmental protection. On the basis of summarizing the research status of solid electric heat storage technology in China and other countries, the research progresses are analyzed in detail from the aspects of system-form, solid thermal storage materials, structure of thermal storage devices, the thermal storage/release performance, operation modes,control strategies and economic analysis. Then the applications oftypical solid electric heat storage devices are given. Finally, in combination with the development requirements of solid electric heat storage technology, the suggestions and prospects could provide references for the future researches and applications.

Abstract:Aiming at the development trend of multi - energy cooperative planning and co-operation of energy systems, a comprehensive energy market trading mechanism including electric energy, thermal energy and hydrogen energy is proposed. Market participants are integrated energy service providers with energy conversion equipment that can supply multiple types of energy and users with flexible demand response capabilities. Based on the market game relationship between integrated energy service providers and users, a Stackelberg game model between them is established. Besides, an upper model of the non-cooperative game of the integrated-energy service providers is established, and the lower model of the evolutionary game between users is established. According to the characteristics of the model, a distributed algorithm for finding the equilibrium solution of the model is proposed. Calculations show that the integrated energy trading mechanism can not only improve energy efficiency, but also promote the flexible response of userload, reduce peak load and enhance user efficiency。

Abstract:The clean heating industry is still in the process of transformation to market - oriented operation, and the investment and operation of electric heating rely on subsidies. Therefore, it isnecessary to study the economic issues of electric heating projects.The economic analysis model and the load aggregation management model of the regenerative electric heating heat source station are built, and the economic benefits of the electric heating heat source station before and after the aggregation management participating in the auxiliary service are compared. The results show that aggregating the electric heating loads with flexible regulation potential to participate in the auxiliary service market can obtain the aggregation income, reduce the electricity cost of heat source stations, enhance the economy of the entire electric heating project.The result has important implications to promote the development of marketization of electric heating.

Abstract:With the continuous deepening of China’s power system reform, significant progress has been made in the construction of the power market. Electricity price is a key influencing factor in the electricity market and each participant conducts electricity transactions based on electricity prices. Therefore, improving the accuracy of electricity price forecasts is very important for every participant in the electricity market. Most of the previous electricity price forecasts used single - layer neural network forecasts,and the accuracy of the forecasts was limited. To this end, according to the accuracy of machine learning in forecasting, the deep be-lief network method is used to predict the day - a - day electricity price. In the calculation example, the real data of the US PJM pow-er market is used for simulation prediction and compared with other neural network prediction models. The results of calculation examples show that the prediction accuracy of the deep belief net-work model is higher. The use of deep belief networks can providean effective method for electricity price forecasting for China’s electricity sales companies.

Abstract:Aiming at the problem of potential local market power in power market, the behavior of the units which exercise market power in the transmission congestion cases is simulated in the day-ahead power market simulation platform. Based on market power evaluation indices, a comprehensive local market power monitoring method is proposed, which can identify the potential local market power in power market effectively. Finally, numerical simulations illustrate that this method can identify the key units and prevent the potential local market power.

Abstract:Aiming at the problems of complex calculation process, low efficiency and high energy consumption of the traditional energy metering multi-dimensional data clustering analysis algorithm, a set of suitable solutions is designed. The solution is based on a big data platform for calculation and storage, and multidimensional analysis of electric energy measurement data is carried out through the chaotic correlation dimension clustering analysis method. This method absorbs the advantages of traditional big data clustering algorithms and chaotic feature extraction, selectethe first minimum value of the interactive information as the best time delay and uses the false nearest neighbor algorithm to select the best embedding dimension u reconstruction phase space. Then,based on the phase space reconstruction, the chaotic correlation dimension characteristics are extracted and the clustering algorithmis combined to cluster the multi-dimensional data of electric energy measurement. Experiments prove that the chaotic correlation dimension clustering analysis method in this study is highly efficient,simple in process and low in energy consumption.

Abstract:The analysis of industry electricity market prosperity is very meaningful to power planning, production, decision-making and demand side management. Based on the big data resources of industry electricity consumption provided by the data collection system of Zhejiang province, the industry electricity market prosperity index model is constructed and the prosperity degree of the industry electricity market is analyzed. Firstly, X - 12-ARIMA（X -12 Auto Regressive Integrated Moving Average）seasonal adjustment model is used to pre-process the data to eliminate seasonal factors. Then, three points forecast - based probing are used to divide the indexes into leading, consistent and lagging indexes respectively. By compiling electricity composite index and diffusion index, the prosperity degree of the industry electricity market is analyzed, and the warning index of industry electricity market is provided correspondingly. Experimental results show that the proposed method is consistent with the actual situations, thus indicating its effectiveness in the prosperity analysis of industry electricity market.

Abstract:Since the launch of the new round of power system reform in China in 2015, the increasingly complicated market settlement relationship and the network -load interaction mode under the market environment are desiring new changes in existing metering system. The general situation of the smart metering system in Texas power market is investigated firstly. By using smart meters and related communication technologies, the system improves the access and utilization efficiency of energy data, and effectively sup-ports the operation of the competitive electricity market in Texas.Then, the development and application status of power metering system in China is summarized, and the problems faced by the metering system under the new power reform situation are sorted out.Finally, suggestions are put forward for the reform of the current measurement system in China to adapt to the complex settlement and network-load interaction in the market environment.