Abstract:Mature electricity market mechanisms are conducive to the development of energy Internet technologies such as virtual power plant, demand response and integrated energysystem. Several countries such as the United States, Denmark, and Germany,have accumulated abundant experience by using electricity market mechanisms to foster the development of energy Internet technologies. The relationship between energy Internet and electricity market is analyzed firstly. In addition, the mechanisms that different energy Internet projects participate in the electricity market are summarized. Finally, based on the experience of these countries, suggestions are proposed to help the development of energy Internet under the circumstance of electricity market in China.

Abstract:In view of the strong nonlinear and strong coupling characteristics of the heat flow equation, a linear optimization method based on multivariate function taylor expansion is proposed. This method transforms the complex nonlinear optimization problem into multiple linear optimization problems that are easy to solve. An iterative method is used to gradually approximate the optimal solution between optimization problems. Taking the minimum total network loss of the electro thermal interconnect system as the optimization goal, the parameter optimization configuration of multiple cogeneration units can be realized at the same time, the network loss is reduced, the system operation cost is reduced, and coupling complementarity and coordinated operation is achieved between different energy sources, which can improve system scheduling, economy and security. The test analysis of the actual system verifies the validity and rationality of the proposed method.

Abstract:Integrated energy system is an effective solution to solve the problems in the current energy field. Gas, electricity,heat, cold and other forms of energy storage play important roles in IES. An integrated energy storage optimization configuration model is proposed, which can be widely applied to multiple IES architectures. The universality of the model is also analyzed. Firstly, a universal model of user side IES is established. Secondly, based onthe universal model of IES, an integrated energy storage optimization configuration model is established to make the configurationmodel applicable to various IES architectures. Aimed at minimizing the annual cost of system life cycle, the model determines the configuration capacity and power of various energy storage. Finally, a case study of IES in a commercial building is analyzed. And the universality of the optimization configuration model in other scenarios is analyzed. The result shows that the proposed model can be applied to IES of multiple architectures to solve the problem of optimal configuration of integrated energy storage in different scenarios.

Abstract:Faced with the optimal operation problem of power system with highpenetration photovoltaics / battery, a coordinated optimization model for power system is established. Firstly, the power output model for photovoltaic system is established, in which the Wasserstein distance is used to calculate the continuous probability distribution of the photovoltaic power output. A variable weighted K means clustering algorithm based on Huffman tree is proposed to cluster the whole photovoltaic scenarios, and the typical photovoltaic power generation scenarios is obtained. Besides that, the simplified kinetic battery model is introduced,which takes into account the depreciation cost of battery. Considering the fuel cost, the startstop cost, the emission cost and the abandoned light, the coordinated optimization model for power systemwith high penetration photovoltaics / battery is established.The model can improve both the capacity of photovoltaic power utilization and the economical and reliable operation of power system in the maximum extent. Finally, a test system is simulated and analyzed by the proposed model, and the results verify the applicability and effectiveness of the model.

Abstract:After the deregulating of the power retail market,the commodity attribute of electric energy is prominent. The priceof electric energy is determined by the quality of electric energy,and the user has the right of free option. For power sale companies,how to reasonably determine the sale price according to the powerquality becomes a burning question. A power quality pricing method based on Stackelberg game is proposed, which contains multi sellers and multiusers. The concepts of internal reference price ofconsumers and“selfselection constraints”are introduced, and the selection model of users is established. On this basis, the user revenue optimal models and power sale company revenue optimal model are established. Stackelberg game is used to find the optimalprice and solve the power quality pricing problem. The case study shows that the proposed power pricing method based on Stackel berg game can reasonably determine power sale price, satisfy thedemand of power quality and revenue of users, and ensure the profit level of power suppliers.

Abstract:In the background of the new power reform, correct and reasonable assessment of the credit risk of power usersand early warning are important factors that restrict the further development of power supply enterprises. In order to carry out creditevaluation of power users and help power supply enterprises to balance the risk level of power trading, based on random forest algorithm, the evaluation principles and indicators of power users arestudied accroding to the evaluation indicators of credit risk in power market. The early warning model of power users evaluates thecredit risk of power users from enterprise market risk, financialrisk, credit risk, operational risk and settlement management risk.It is expected to improve the trading mechanism of electricity market and promote the rational and efficient allocation of resources,help power industry develope healthily under the protection of credit system.

Abstract:Aiming at the calculation requirements of the theoretical line loss level of the station line loss lean management, amethod for estimating the line loss of the theoretical line loss based on the correlation degree variable weight is proposed. Firstly, theinfluencing factors of theoretical line loss are analyzed and a theoretical line loss evaluation index system including the inherentcharacteristics of the station area and operational factors are established. Then, Spearman coefficient, grey correlation degree and cosine quantitative analysis method are used to construct the evaluation index. The evaluation method of the theoretical line loss of thestation area is proposed. Finally, the theoretical line loss level is analyzed by using six different types of stations in Anhui as an example to verify the effectiveness of the proposed method.

Abstract:A load forecasting method based on classified identification deep belief network is proposed to solve the problem ofslow convergence and large prediction error of traditional neural network in load forecasting. In the proposed method, the input historicalload data is normalized firstly. Then, the layered pretraining of thedeep believe machine is implemented by hierarchical unsupervisedgreedy pre training method. Thirdly, the pre training results areused as the initial value of the supervised learning training model.Deep belief network is composed of multilayer restricted boltzmannmachine, and the pre weight of the restricted boltzmann machineand the classification recognition mechanism is trained by means of contrastive divergence. The learning performance of the classified identification deep belief network can be improved by this way. Simulation result shows that the load forecasting algorithm based on the200 times of load training and temperature training, the algorithm has faster convergence speed and higher prediction accuracy than the selforganizing fuzzy neural network and the BP neural network.

Abstract:Currently, economic new normal and frequent extreme weather make influencing factors of power demand more diversified. Forever, integrated forecast methods of such two power demand indicators as power consumption and max power load arenot scientifically formed. Main influencing factors of power demandfrom economic and meteorological dimensions are systematicallyanalyzed and a forecast model system of provincial grid’s annualpower demand is established. The instance of Anhui province demonstrates that the key of improving forecast accuracy is to identify critical influencing factors regardless of basing forecast model on multiple linear regression or intelligent algorithm. Mean forecast error rates of power consumption and max power load remain are both within 2% in this model, becoming an effective method for electricity market analysts to forecast annual power demand.

Abstract:Aiming at the incremental distribution network market at park level under the background of the new round of electricity reform, the problems and demands of load forecasting in power network planning are analyzed, and the description of load forecasting conclusions for incremental distribution network planning is improved, and a load forecasting model considering uncertainties such as distributed energy and occupancy rate is further proposed.Different confidence intervals is set according to the forecasting demand, and the load forecasting of incremental distribution parks under different development conditions is revised. Practical examples show that the method can provide a quantitative idea for precise planning of park incremental distribution network considering uncertainties.

Abstract:In order to solve the problem of distribution network congestion caused by a large number of distributed power sources and flexible load access to the distribution network, a comprehensive solution for distribution network congestion management based on distribution network reconstruction and marketmechanism is proposed. In the proposed scheme, the flexibility of different types of loads and the respective advantages of the two congestion management methods are used to restructure the distribution network into the market mechanism framework to solve the problem of line and flexible load congestion. This comprehensive solution uses the NSGAII algorithm to solve the blocking problem quickly and effectively, and meet the requirements of the minimum energy cost of the distribution system operator side and the minimum electricity cost of the flexible load side. A simulation study is carried out using a standard IEEE 33 node distribution network test system, which verifies the effectiveness and superiority of the proposed comprehensive scheme.

Abstract:An industrial park is characterized by good economic foundation, large energy consumption and industrial agglomeration. It is an“ideal experimental field”for building a flexible,diverse, lowcarbon and efficient integrated energy service system.The structure of integrated energy system in industrial parks is described from three aspects, which are the physical layer, the service layer and the information layer. Then, the integrated energyservice business model suitable for industrial parks is discussedfrom four aspects, which are promoting clean energy consumption,optimizing energy demand, promoting energy efficiency and marketefficiency mining. And from the perspectives of promoting fair market competition and shared development, relevant policy suggestions are put forward. The viewpoints can be referenced for the exploration and establishment of integrated energy service businessmodel in industrial parks.

Abstract:With the continuous growth of demand for national electricity and increasing environmental pollution, demand sideenergy management which designed to improve energy structure,upgrade energy efficiency, reduce power consumption, and decrease operation costs, has begun to be focused on. Large scale shopping malls have great potential of power saving and consumption reduction, due to large number of power devices, long operation hours, high power consumption and so on. An energy optimization operation model is established for the large scale shopping mall by coordinated optimization controlling of central air conditions, PVs and EV charging services. This model takes the minimum operation cost as goal, considering temperature comfort, power balance and other constraints, taking uncertainties of factorssuch as the power of electric vehicles into account. The genetic algorithm with Latin Hypercube sampling is taken to solve the problem. Simulation results of a largescale shopping mall demonstratethat the proposed optimization model could reduce operating costsand bring economic benefits.

Abstract:Within the flourishing of the energy Internet, electric energy storage has become one of the important supportingtechnologies of energy Internet. While focusing on the innovation and development of energy storage physical technology, its business model is also the decisive factor to determine whether the energy storage industry can flourish. Firstly, the relevant policies supporting the development of energy storage are sorted out, and theirimpact on the energy storage industry is discussed. Secondly, thedevelopment status of business models of energy storage at home and abroad are summarized and analyzed, which are in the fields of supporting new energy, microgrid, ancillary services and so on. Finally, it puts forward a potential and feasible business operation model of energy storage based on the reality in China.

Abstract:With the advent of the Internet and the era of bigdata, the value of data is becoming more and more significant, playing a more and more important role in various fields. It also givesbirth to a new research field, that is data privacy security. Powergrid companies collect and store a large amount of power customer data. Solving the problem that how to protect the privacy and security of customer data in the process of using has become particularly urgent. By analyzing the characteristics and application scenarios of power customer data, a construction scheme of data privacy protection system for power customer is proposed, which includes business scenario classification, privacy classification,protectionstrategy and protection method library. An active privacy protection mechanism for power customer data is also built, and the active recommendation of data privacy protection strategies and technology methods for power customers are realized. The practicalcase shows that the data privacy protection system for power customersestablished has good practicability.

Abstract:The shortages of traditional marketing service channels of the power grid are expounded and the reasons for the emergence of new media marketing service channels are also pointed out. Focusing on analyzing the characteristics and advantages of the three mainstream power grid new media marketing service channels, which are Pocket Power, WeChat official account and alipay life, the problems in customer experience, economic benefitsand business control are mentioned. Then, working ideas from three aspects are put forward. Finally, the significance of the development of new media channel services is summarized.

Abstract:The reform of electric power system makes the traditional electric power industry begin to think about how to become a truly customercentered enterprise, the core of which is customer experience management. The basic preparation of customer experience management is introduced. Then, how to build a complete customer experience management system is expounded, as well as the optimization and improvement of customer experience, providing references for the transformation and upgrading of power enterprisesd.

Abstract:How to manage energy scientifically and realize coordination among various energy systems has been a common goal for each country. Therefore, it’s of great significance to study P2G and make it commercialized. The technology and development status of P2G are summarized, the business model is analyzed. Finally,the practical value and the development paths of P2G in Yunnan province are discussed based on the actual situation.