Remaining useful life (RUL) prediction for electric vehicle battery (EV battery) is of great significance for the early replacement and regular maintenance of batteries with potential safety hazards in driving conditions. This paper proposes a novel method based on light gradient boosting machine (LightGBM) to predict the RUL of the battery under driving conditions. LightGBM uses the histogram optimization strategy to reduce the number of traversal of the data sample set and improve the robustness of the method; the depth-first splitting (leaf-wise) strategy reduces the risk of overfitting; the gradient-based one-sided sampling strategy (GOSS), reduce the data dimension; use the exclusive feature bundling strategy (EFB) to reduce the feature dimension. However, the LightGBM method has the difficulty of parameter setting. Therefore, this paper uses Hyperopt based on distributed asynchronous algorithm configuration/hyperparameter optimization to optimize its complicated hyperparameters. Subsequently, the method was applied to the prediction of battery RUL under simulated driving conditions. Based on the comparative cases, the results show that this method can guarantee the rapidity, accuracy and robustness of RUL prediction under the condition of low memory usage. © 2020 IEEE.

The high penetration of renewable energy incurs serious power fluctuations in the power system. Large scale electric vehicles (EVs) under the centralized control are able to provide considerable regulation capacity for the power system. In existing modeling methods for centralized EVs, the accurate control results for the regulation services were achieved by modeling each EV individually. However, the computational complexity was a serious problem for modeling large scale EVs. In this paper, a reduced modeling method for centralized EVs is developed by describing a population of EVs with a partial differential equation (PDE). This PDE modeling method uses a finite number of state-of-charge (SOC) intervals to describe the flows of charging EVs, idle EVs and discharging EVs. Considering the three connecting states (CNS) of EVs, this PDE modeling method realizes the power regulation for centralized EVs by changing the CNS of EVs located in different SOC intervals. Comparative simulations validate that the PDE modeling method for large scale centralized EVs achieves high control accuracy and high computational efficiency. © 2020 IEEE.

Wind power generation is one of the important types of distributed generation in microgrid. The random variation of wind speed will affect the stable operation of microgrid. In this paper, the mathematical model of the asynchronous wind turbine is established, and the power characteristics of the wind turbine are analyzed. On this basis, the effect of asynchronous wind turbine on the operation control of microgrid during the wind speed disturbance is analyzed. Furthermore, the microgrid operation control technology based on energy storage system is proposed, including the power circuit and its control strategies. Then a simulation model of microgrid including wind power generation is built by Matlab/Simulink, and the simulation verification under typical operating scenarios such as gust disturbance is carried out. The results show that the energy storage system can effectively buffer the power fluctuation of the asynchronous wind turbine, restrain system voltage change, and improve the stability of microgrid. © 2020 IEEE.

During the actual operation of micro grid, the power fluctuation of tie line caused by the Load power fluctuation of renewable energy or load prediction deviation needs to be controlled in real time to ensure the economy of micro grid operation. This paper aims at a power system model based on the framework of "distribution network - industrial park micro grid - enterprise load user", and studies a power control method of micro grid tie line based on user reward scheme. Firstly, the objective economic function of the micro-grid in the park is established, and then the optimal reward scheme for internal load users of the micro-grid in the park is derived according to two typical load response models. The power fluctuation can be stabilized by encouraging load users to respond to demand, thus the internal load users can benefit while reducing their own economic losses and the day-ahead scheduling results of tie lines can be effectively tracked. Taking a production industrial park as an example, the effectiveness of the proposed optimization model and algorithm is verified by an example analysis. © 2020 IEEE.

This paper investigates the stability of a power system with distributed EV integration embedded by cyber delays in the cyber-physical system. An efficient stability region extraction method based on infinitesimal generator transformation is used to extract all stabilizing values in cyber delay space. We obtain the disrupted time-delay stability region for the power system with a single and double EV aggregators and investigate the interaction among the delay margins and PI controller gains. Verification of theoretical stability region boundaries is shown by time-domain simulations for test points selected from stability regions. Through the case study, it is revealed that with the integration of distributed EVs into the power system, there is no longer a definite time-delay stability margin like the traditional power system, but under some parameter settings, the system stability region is disrupted by several intervals and presents periodic changes. And the cyber delay in EV aggregator with the higher participation ratio has a smaller range of cyber delay region. Results also show that an increase in the integrator gain reduces the stability region, while the range of stability region in time-delay space increases and then decreases as increases, which introduces cyber stability challenges to traditional frequency dynamics. It is expected that the proposed criteria can help to guide the determination of delay requirements for EV aggregators participating in frequency regulation service. © 2020 IEEE.

The uncertainty of multiple sources and loads in the regional integrated energy system (RIES) has a great impact on the planning and operation. This paper proposes a method of multidimensional correlation scenario set generation, which considers the characteristics of multiple sources and loads, time sequence, autocorrelation and cross-correlation. In the planning model, the importance of the energy pipeline is considered, the pipeline risk index is defined, and the multi-objective chance constrained expansion planning of the RIES energy network is studied based on the multidimensional correlation scenario generation method. The case studies show that the multidimensional correlation scenario generation method can effectively generate a mount of approximate realistic scenarios, and the multi-objective stochastic planning scheme of RIES considered the uncertainty, economy and energy pipeline risk. © 2020 IEEE.

District Integrated Energy System can realize the coordination and stepped utilization of multiple energy, improve the integrated utilization efficiency of energy, and take advantages of multiple energy into account. On the other hand, the optimal operation of District Integrated Energy System involves many stakeholders, the investment and operation mode of the system is more complex, and the demand side has different energy demands with different forms and characteristics, which will directly affect the operation economy and environmental protection of the system. This paper mainly studies the multi-objective optimal operation of the District Integrated Energy System considering economy and environmental protection. Firstly, the model of system is constructed, which includes multi-source supply and integrated load. Secondly, a multi-objective optimization problem considering economy and environmental protection is put forward. Finally, the linear weighting method combined with entropy weight method is used to solve the multi-objective optimization problem. The case analysis shows that the system operation can give consideration to economy and environmental protection after multi-objective optimization. © 2020 IEEE.

The integrated energy supplier is able to provide both heat and power to the energy users at the same time. However, the power and heat have different time-scale characterics, which can be made full use of to reduce the energy cost. In this paper, A day-ahead scheduling method for heat-power energy station considering the thermal inertia of heating network is proposed. Firslty, A dynamic model of the heat transmission process considering the thermal inertia of heating network is developed. The virtual energy storage characteristics of the heat transfer fluid in the heating network is analyzed. Then, a day-ahead scheduling model with considering the heat storage capability of the heating network is introduced. The heating network is used as a heat storage system to participate in the day-ahead scheduling process. Numerical studies demonstrate that considering heat storage capability (supplied by the thermal inertia characteristics of the heat network) in the scheduling model is able to reduce the operation cost of the heat-power energy station under the time-of-use electricity price environment. © 2019 IEEE.

Power marketization can revert the property of electricity commodities and improve the efficiency of resource allocation. In this areamaking an effective pricing scheme remains a particular challenge. In this paper, a novel pricing scheme based on evaluation mechanism is proposed, which provide a market-oriented scheme for driving peer-to-peer (P2P) energy trading and optimizing prospective earnings for all households in the microgrid. The evaluation mechanism as a feedback on the historical energy service level of households can make them with better evaluation coefficient have more advantages in optimal quotation point (OQP) and improve the efficiency of the electricity market. Furthermore, the proposed scheme is applied to simulate, and according to the simulation results, the more bidders participate in the auction, the less the prospective earnings are. In addition, when the number of bidders is same, the lower evaluation coefficient, the higher prospective earnings. © 2019 IEEE.

Capacitor voltage balancing is indispensable in normal operation for modular multilevel cascade converter- single star bridge cell (MMCC-SSBC) based static synchronous compensator (STATCOM). This paper proposes a novel individual voltage balancing scheme with dramatically reduced voltage sensors. First, zero-sequence voltage injection is enabled to regulate the cluster voltage based on only one voltage sensor in each phase; Then, through the additional inserted hardware units, individual voltage balancing can be achieved. Compared with conventional software voltage balancing method, the proposed balancing scheme has the advantage of simple implementation, lower transportation delay, fast balancing speed, and dramatically reduced voltage sensors. In addition, only some popular power modules are added to balance the individual capacitor voltage. The effectiveness of the proposed individual voltage balancing scheme is verified by simulation results in MMCC-SSBC based STATCOM. With the proposed scheme, the individual voltage balancing and reactive power compensation can be realized simultaneously.
© 2019 Institution of Engineering and Technology. All rights reserved.