The electric vehicle (EV) and electric vehicle charging station (EVCS) have been widely deployed with the devel- opment of large-scale transportation electrifications. However, since charging behaviors of EVs show large uncertainties, the forecasting of EVCS charging power is non-trivial. This paper tackles this issue by proposing a reinforcement learning assisted deep learning framework for the probabilistic EVCS charging power forecasting to capture its uncertainties. Since the EVCS charging power data are not standard time-series data like electricity load, they are first converted to the time-series format. On this basis, one of the most popular deep learning models, the long short-term memory (LSTM) is used and trained to obtain the point forecast of EVCS charging power. To further capture the forecast uncertainty, a Markov decision process (MDP) is employed to model the change of LSTM cell states, which is solved by our proposed adaptive exploration proximal policy optimization (AePPO) algorithm based on reinforcement learning. Finally, experiments are carried out on the real EVCSs charging data from Caltech, and Jet Propulsion Laboratory, USA, respectively. The results and comparative analysis verify the effectiveness and outperformance of our proposed framework.
11 September 2019 in IEEE Transactions on Power Systems
The rapid expansion of wind power has triggered significant wind curtailment because the power system lacks flexibility to deal with the uncertainty and variability of wind power. The operational flexibility of coal-fired power plants is limited by the minimum stable firing rate in the boiler. Steam extraction and thermal energy storage could enable power output adjustment without changing the firing rate in the boiler. Thus, retrofitting existing coal-fired power plants with steam extraction and thermal energy storage is a promising option to accommodate the high penetration of wind power in the power system, especially in coal-dominated power system. This study investigated the operational flexibility of coal-fired power plants retrofitted with steam extraction and thermal energy storage. First, a linear operation model is proposed for retrofitted coal-fired power plants considering new characteristics and technical constraints. Second, a simulation framework based on stochastic unit commitment and rolling economic dispatch is developed to explore the benefits of steam extraction and thermal energy storage in wind-integrated power systems. The simulation results on the modified IEEE 24-bus system demonstrate the effectiveness of retrofitting coal-fired power plants with steam extraction and thermal energy storage for mitigating wind curtailment.
This paper proposes a method of energy storage capacity planning for improving offshore wind power consumption. Firstly, an optimization model of offshore wind power storage capacity planning is established, which takes into account the annual load development demand, the uncertainty of offshore wind power, various types of power sources and line structure. The model aims at the lowest cost of investment, operation and maintenance of the system, and takes lower than a certain abandoned wind level as the strict constraint to obtain two parameters of power capacity and energy capacity of energy storage on the source side. Secondly, taking a coastal power grid as a typical case, the energy storage capacity planning method is verified. Finally, the key factors affecting offshore wind power consumption are summarized, and the sensitivity analysis is carried out from the point of view of the transmission protocol of the transmission lines outside the province and the capacity allocation of the tie lines in the province. This study will be helpful for the planning and operation of the high-proportion of offshore wind energy power systems.
With the grid-connected operation of large-scale wind farms, the contradiction between supply and demand of power systems is becoming more and more prominent. The introduction of multiple types of flexible resources provides a new technical means for improving the supply–demand matching relationship of system flexibility and promoting wind power consumption. In this paper, multi-type flexible resources made up of deep peak regulation of thermal units, demand response, and energy storage were utilized to alleviate the peak regulation pressure caused by large-scale wind power integration. Based on current thermal plant deep peak regulation technology, a three-phase peak regulation cost model of thermal power generation considering the low load fatigue life loss and oil injection cost of the unit was proposed. Additionally, from the perspective of supply–demand balance of power system flexibility, the flexibility margin index of a power system containing source-load-storage flexible resources was put forward to assess the contribution from each flexibility provider to system flexibility. Moreover, an optimal dispatching model of a multi-energy power system with large-scale wind power and multi-flexible resources was constructed, aimed at the lowest total dispatching cost of the whole scheduling period. Finally, the model proposed in this paper was validated by a modified RTS96 system, and the effects of different flexibility resources and wind power capacity on the optimal scheduling results were discussed.
19 September 2018 in IET Renewable Power Generation
Concentrating solar power (CSP) plants produce electricity without any pollutant emission, which is one of the most attractive alternatives to fossil fuels. The thermal energy storage (TES) benefits CSP plants to produce electricity during temporary weather transients and peak‐load demand hours. However, the main drawback of the CSP plants is the high investment costs. To improve the economic viability, it is necessary to design CSP plants appropriately. In this study, an optimal co‐allocation model of solar field and TES for CSP plants considering with the operation demand of power system is developed to deal with the tradeoff between investment costs and benefits. To consider realistic operating constraints account for the interaction between hourly variability, the uncertainties of solar irradiation and wind power are characterised by a set of scenarios. In addition, the large number of scenarios is reduced to relieve the computational burden. The validity of the proposed model is demonstrated by case study results.
Clean and low-carbon energy represented by wind power and photovoltaic power will develop rapidly and will form a new power system with a high proportion of renewable energy. In the context of a low-carbon economy, how to make reasonable planning for power system flexibility resources is crucial for the development of new power systems. In this paper, we establish a multi-flexibility resource planning model for a power system based on a low carbon economy by considering the planning of multi-flexibility resources of “source–load–storage”. First, a ladder-type carbon trading cost accounting model is proposed, and a set of power system flexibility evaluation indexes are proposed. Then, with the objective of minimizing the sum of low carbon operation cost, investment cost, and operation cost of the system, the planning model of multi-flexibility resources is established by considering constraints such as system power balance constraint, investment constraint, and wind power consumption constraint. Finally, the model proposed in this paper is validated by the IEEE-RTS96 system; the results show that: (1) collaborative planning of source–load–storage multi-flexible resources can obtain the best overall system economics, although the investment cost increases by USD 12.6M, the total system cost is reduced by 11.22% due to the reduction in coal generation consumption cost, carbon trading cost, and wind curtailment penalty cost; (2) as the penetration of wind power grows, the demand for energy storage in the power system is gradually increasing; when the installed capacity of wind power grew from 800 MW to 1600 MW, the demand for new thermal power decreased by 53.5% and the demand for new energy storage increased by 200%; (3) the total cost of the planning model considering ladder-type carbon trading decreases by 1.35% compared to the model without carbon trading, and increases by 2.5% compared to the model considering traditional carbon trading, but its carbon emissions decrease by 5.5%.