Proceedings of 2021 International Top-Level Forum on Engineering Science and Technology Development Strategy

Inhaltsverzeichnis

1. Stochastic Optimal Strategy of Direct Electric Quantity Purchase by Large Consumers in Medium and Long-Term Transactions

   Huang Qingqing, Jiang Cenxi, Chen Jinhui, Zhang Zhe, Fu Rong

   Abstract

   Affected by the variations of consumer demands, large direct-purchase electricity consumers have large deviations in electricity and electricity price forecasts, which will increase the deviation penalty of electricity and power purchase costs. Considering the uncertainty of consumer demand, the optimization decomposition method is studied to compute the monthly power electricity from the consumer’s annual bilateral transaction power electricity. So the minimum expected cost of the monthly electricity purchase and the deviation penalty is expressed as the comprehensive objective function. Thus, the consumer’s annual bilateral transaction electricity and monthly centralized bidding electricity purchase are figured out with a stochastic optimal strategy. Combining the data of actual consumer electricity purchases, it verifies the economics and effectiveness of the optimization decomposition method for direct electricity quantity purchase by large consumers in medium and long-term transactions.

2. An Active Power Control Strategy Adapted to Renewable Energy Absorption Responsibility Weight Under Spot Market Environment

   Aimei Liu, Li Chang, Chunxiang Yang, Jiping Wu, Rongzhang Cao

   Abstract

   In China, the responsibility weight system for renewable energy absorption is being promoted simultaneously with the construction of the electricity spot market. Under this environment, this paper synergistically integrates an active power control method of renewable energy considering absorption responsibility weight and guaranteeing accommodation mechanism. An active power distribution strategy is triggered when renewable energy abandonment occurs, which aims to promote the priority execution of guaranteed accommodation schedule and the spot market trading power. Secondary active power allocation strategy based on prediction bias confidence is introduced into the strategy. The proposed control strategy is verified based on the actual operation effect of the power grid and achieves a good effect.

3. Morphological Clustering Algorithm of Daily Output Curve of Wind Farm Based on Multi-scale and Entropy Weight Method

   Qinran Zhu, Xiaoyan Bian, Peijie Lou, Zhexuan Tang

   Abstract

   Due to the random fluctuation of daily output curve of wind farm, the traditional methods for morphological clustering are difficult to apply, and can not accurately reflect the characteristics of fluctuation. To solve this problem, this paper proposes a morphological clustering algorithm based on multi-scale similarity and entropy weight method. Firstly, the multi-scale difference measure can be obtained by four difference distances and quantile characteristics assignment. Secondly, the entropy weight method is applied to determine the weights of each scale, and obtains the difference measure matrix. The morphological distance is used instead of the traditional Euclidean distance for clustering. Finally, the feasibility of the proposed algorithm is verified by taking the wind farm output data of 100 days in a certain area as an example.

4. Converter Recovery Strategy to Reduce the Commutation Failure Risk of Capacitor Commuted HVDC Transmission System

   Yue Pan, Wei Li, Yu Wang, Jie Lei, Ling Zhu, Qianwei Liu

   Abstract

   Commutation failure is an important factor affecting the grid-connected safety of LCC DC systems. The Capacitor Commutated Converter (CCC) is used to improve the ability of DC to resist commutation failure, but it also introduces new problems. In this paper, firstly, the recovery characteristics of the CCC converter in the AC asymmetric fault scenario is analyzed, based on which, a “capacitor bypass-parallel grounding” fault recovery strategy is proposed. This strategy can reduce the risk of commutation failure during AC fault period and accelerate the speed of the power recovery after the fault cleaning, so as to reduce the impact of the fault on the system. Base on Cigre_Benchmark electromagnetic transient model, CCC HVDC system has been verified in simulation experiments. The proposed strategy can effectively reduce the number of commutation failures, the duration of commutation failures and the loss of active power.

5. Study on Protection Schemes of Variable Speed Pumped Storage Units

   Wang Kai, Li Jianchun, Li Huazhong, Xu Huabin, Wang Guang

   Abstract

   Compared with constant speed pumped storage unit, the variable speed unit has many advantages of operation characteristics, such as larger range faster adjustment of active power regulation, higher capacity of leading phase operation and higher comprehensive efficiency. It has become the preferred construction type of pumped storage power stations in the world. It is very important of comprehensive and reliable protection configuration for generator unit. First, based on the electrical differences between variable speed pumped storage unit and constant speed pumped storage unit, this paper analyzed the applicability of conventional protection principle for constant speed pumped storage unit in variable speed pumped-storage unit. Then, some special protection configurations were studied, such as differential main protection, internal fault protection of rotor winding and startup process protection. In addition, this paper proposed typical protection configuration of 300 MW variable speed pumped storage unit. The protection device based on this scheme has passed the dynamic model test. Finally, the dynamic experiment is carried out to validate the new criterion. The experimental results show that the methods can deal with unique faults model in rotor winding proposed in this paper. The principle will be applied in the new variable speed pumped storage unit of China.

6. A Coordinated Control Strategy for Stable Operation and Mode Smooth Switching of AC/DC Hybrid Distribution Network with DAB

   Yu Gao, Shuo Cheng, Yan Chen, Baichen Zeng, Qi Wang

   Abstract

   Aiming at the problems of DC voltage fluctuation and load power imbalance during the operation mode switching of multi-terminal AC/DC hybrid distribution network with dual active bridge (DAB), a coordinated control strategy of AC and DC double-side coordination is proposed. Firstly, the mechanism of DC voltage fluctuation and the imbalance of load power in the AC/DC distribution mode switching process is analyzed. Then multi-terminal AC/DC hybrid distribution network’s structure with dual active bridge is proposed. Meanwhile, the traditional phase shift control for dual active bridge’s control angle θ should not be too large. Otherwise, it will generate a great backflow power problem. A dual control method of control angle θ and duty cycle D is proposed. This method can not only take advantage of the energy bidirectional control of the dual active bridge, but also good for Multi-terminal flexible AC/DC hybrid distribution, it can form a certain buffer due to the impact caused by mode switching, which makes the grid operation mode switch smoother. Finally, a coordinated control strategy applied to the AC/DC dual-side distribution network is proposed. The power buffering capability of the control strategy enables the two sides of the AC and DC to cooperate with each other to make the AC/DC hybrid distribution network operate stably. The effectiveness of the control method is verified by the PSCAD/EMTDC simulation platform.

7. Multi-energy Flow Calculation Method for Integrated Energy Systems Under Different Operation Modes

   Lin Tingting, Wu Guilian, Zheng Jieyun

   Abstract

   The alternating iteration method is widely used for calculating multi-energy flow of integrated energy systems. Due to the interaction devices of power, thermal and gas and diversity of energy supply conditions, there will exist many situations in alternating iteration. To solve the above problems, this paper first analyzes the influence of devices interaction and diversity of energy supply on flow calculation in integrated energy system, then four operation modes are summarized to solve devices interaction and diversity of energy supply, finally the alternating iteration flow calculation under each mode is proposed. The correctness and validity of the proposed method is verified by a case of district integrated energy system.

8. A Comprehensive Energy Efficiency Evaluation Method for Multi-energy System Based on Exergy Analysis

   Zhao Yongkai, Ren Shuai, Zhao Xianhao, Chen Xiuxin, Zhang Yuanbo, Li Yingji

   Abstract

   Existing evaluation methods do not take into account the quality of energy, which may sometimes overestimate the energy efficiency and lead to the loss of energy quality. In this paper, a comprehensive energy efficiency (CEE) evaluation method for multi-energy systems is established based on exergy analysis. Firstly, the subsystems are defined as the equipment in the system based on the different energy conversion types such as “electricity to cold/heat” and “gas to cold/heat”. Then, by introducing the energy quality coefficient (EQC), energy supply and demand of different quality in the subsystem are converted into a unified energy level, so that a variety of energy conversion efficiency methods are established. Furthermore, the CEE of the system can be obtained by summarizing the energy conversion efficiencies and energy weight coefficient. Through case analysis, this method can effectively figure out the link of energy quality waste, and offer suggestions for the reasonable design and operation of multi-energy system.

9. Research on Key Technologies of Line Differential Protection Based on 5G Communication

   Xiaojiang Zheng, Gang Yao, Tonghua Wu, Xianliang Teng, Feng Hong, Weiwei Wang

   Abstract

   With the advanced 5G communication technology and the wireless communication environment of 5G communication network with high bandwidth, low delay and high reliability, line differential protection based on 5G communication has become the development trend of MV (Medium Voltage) and LV (Low Voltage) distribution network protection. This paper provides an overall architecture scheme of line differential protection based on 5G communication to further explain the characteristics of 5G network, and aiming at the problem of delay jitter existing in 5G network, presents a sampling point sequence number synchronization method based on absolute time-marker of external clock to analyze data encapsulation, interaction and traffic consumption of bilateral differential protection, and the corresponding technical solutions, and at last, it proposes a processing mechanism of protection device in dealing with clock loss and data anomaly, which provides a complete technical route for line differential protection based on 5G communication, facilitates the large-scale application of longitudinal differential protection in the closed-loop operation of MV and LV power network, and ensures that the requirements of quickness and selectivity can be met in the traditional non-unit protection of MV and LV power network, so as to improve the reliability level of power network.

10. Comparison of the Effect of Regularization Techniques and Lookback Window Length on Deep Learning Models in Short Term Load Forecasting

    Aysegul Kahraman, Peng Hou, Guangya Yang, Zhile Yang

    Abstract

    Management of electric power balance requires accurate forecasting of load and generation, especially in the context of renewable energy adoption. In this context, forecasting electric load requires more attention to decrease the uncertainties in the system operation. There have been many studies under this context, however, the effect of the lookback window for both deep learning and regularization techniques has not been fully investigated in the literature. In this study, we developed a comparative study based on 4 typical deep learning techniques, namely MLP, 1D-CNN, LSTM, and a hybrid model that is a combination of 1D-CNN and LSTM to forecast the electrical load. The effect of both regularization methods and lookback window length has been investigated in detail and found that they improved the forecasting performance based on the complexity and features of the networks. The methods are evaluated in terms of 4 different metrics namely MSE, MAE, MAPE and, R². The results show LSTM outperformed the other methods in general, and the increase of lookback length improved its performance with the average MAPE less than 2%.