Abstract

Electric vehicles, including hybrids, will domain the road transport in future cities. Reinforcement learning has shown its capacity in online optimization of energy management strategy for hybrid vehicles. Exploration of new control settings and exploitation of the existing control policy are two key procedures in reinforcement learning but there is a lack of study on how the exploration-to-exploitation (E2E) ratio affects the energy efficiency improvement for hybrid vehicles. This paper introduces two decay functions, ‘Reciprocal function-based decay’ (RBD) and ‘Step-based decay’ (SBD), to generate E2E ratio trajectories for reinforcement learning algorithm which is conventionally based on Exponential decay (EXD) function. By monitoring the improving rate of vehicle energy efficiency in the learning process, the vehicle controlled by Q-learning algorithm based on the SBD function has shown the best compared with the vehicles based on the RBD function and the EXD function. The improving rate can be more than 6.21%. In the HiL testing, the SBD can save 1.52% energy compared to the EXD in real-time control under a predefined driving cycle.

Abstract

Solar District Heating (SDH) systems combined with Seasonal Thermal Energy Storage (STES) are one of the most interesting available strategies to decarbonize the residential sector in Europe, where most of the energy is presently provided by fossil fuels. The present work analyses both technically and economically the retrofitting of an existing district heating based on a biomass boiler into a SDH with STES, comparing two different SDH configurations: a single network, and few separated smaller networks, both serving the same actual heat demand. Dynamic modelling of both systems was carried out using TRNSYS (Transient System Simulation) software. The results of the simulations are then presented and discussed.

Abstract

This paper proposes an integrated framework to achieve a simultaneous real-time reduction of occupants’ thermal dissatisfaction and room HVAC energy consumption. The framework can optimize the HVAC setpoint temperature according to the internal heat gains predicted by a vision-based approach. When there are no occupants found by cameras, this framework will just turn off HVAC systems to reduce energy consumption. When occupants are present, the framework will determine an optimal setpoint temperature to balance occupants’ thermal satisfaction and room HVAC energy consumption. During the simulated four days in the winter in a temperature climate, the utilization of this framework can lead to a reduction of heating energy by up to 36.8% and occupants’ thermal dissatisfaction by up to 5.26%. During another simulated four days in the summer, the cooling energy savings would range from 3.5% to 33.9%, whilst occupants’ thermal dissatisfaction could be decreased by 0.17-2.89%.

Abstract

Recently, the spray heat exchanger became an important technology in a sewage source heat pump (SSHP) to recover thermal energy from oily wastewater. Hence, it’s necessary to study the heat transfer properties and its enhancements. Among all of the influential factors, the arrangement of the tube bundle is a critical factor that influence the liquid film flow of the falling film, and further has a great impact on heat transfer. Therefore, in this paper, a three-dimensional numerical model of the arrangement of triangular tube bundles is established, using a glycerin-water mixed solution to simulate oily wastewater. The flow pattern of the liquid film under different tube row modes and the influence of heat transfer effects are compared, and the temperature inside the liquid film is analyzed. It was found that there is an obvious difference in the falling film flow outside the horizontal tube between the triangular tube bundle arrangement and the in-line tube bundle arrangement. The liquid film flow in the triangular arrangement is more unstable and has a greater impact on the heat transfer of the upper half of the tube. The same phenomenon is observed for the tube bundle below the top row. The local heat transfer coefficient along the circumferential direction of the triangular arrangement is greater than that of the in-line arrangement. Besides, both of the tube bundle arrangements have the phenomenon that the local heat transfer coefficient decreases with the increase in oil content.

Abstract

Microchannel boiling was used for cooling concentrated photovoltaic (CPV) due to its promising heat transfer performance, to maintain the normal operating of CPV at a high efficiency. However, fluctuating and non-uniform heat fluxes of CPVs may lead to deterioration of flow boiling and heat transfer of microchannel, and then reduce the heat dissipation performance of CPV, and lead to damage of solar cells. To improve this key point, a rectangular straight microchannel with deionized (DI) water boiling was numerically simulated using VOF method, four sudden heat flux increase levels were set: 97.96, 199.64, 284.54 and 580 kW/m2. The effects of sudden heat flux increase on the two-phase flow pattern was studied. It was found the higher heat flux increase advanced the generation of water vapor phase and transition. The water vapor volume fraction was instantaneously increased and easily moved to instability under heat flux increase of 580 kW/m2.

Abstract

To explore the influence of circuit layout on the performance of a thermoelectric generator (TEG), this study utilizes Fortran software to establish the engine exhaust of the thermoelectric generation model, and analyze the temperature distribution of the hot and cold ends of the thermoelectric unit, electromotive force, current, and output power distribution along the flow direction, under different circuit layouts. In addition, the power output performance of different circuit layouts, including the full-series-type and multi-stage-series types, are compared. The results indicate that the multi-stage series circuit layout can achieve a higher output power, and the higher the stage number, the higher the power output; however, the increased range becomes significantly smaller. Considering the complexity and connection of the circuit control, the recommended segment number of the multistage series type is three.

Abstract

A microgrid, as a typical low-carbon energy system, has gained great attentions in the last decade. Most studies focused on improving the system performance at the design and control stages while the evaluation of the system performance at the planning stage is usually ignored. Commonly, three key indicators (cost, renewable energy penetration, and reliability) are used to evaluate the system performance. The cost and/or the renewable energy penetration are set as the objectives while the reliability is set as a constraint in design/control optimization. However, due to lacking adequate investigations of system performance at the planning stage, fewer supports are provided in energy policy formulation. This study, therefore, proposed a quantitative approach to evaluate the system performance in terms of these three indicators at the planning stage. By using the proposed model, optimization algorithm, and regression analysis, these three indicators are quantified. A case study is proposed to test and verify this approach. Three different reliability requirements are adopted. In the end, a reasonable range (40%-60%) of renewable energy penetration is recommended. The system performance of the cost-saving (the largest one as 61.5%) is evaluated based on three ratios of reliability requirements and renewable penetration ratios.

Abstract

The time-shifting property is an important parameter to characterize the fluctuation of natural wind at multiple temporal-spatial scales. Accurately grasping the time-shifting characteristics of natural wind can effectively reduce the impact of super-large-scale wind power grid integration on the power system. However, most of the existing methods are based on the ARMA model and time-shifting technology to generate wind speed time series with specified correlation coefficients. It is only the change of time to show that this time-shifting characteristic does exist, but there is no specific quantitative analysis on it. This paper will take the minimum Euclidean distance as the goal, and propose two evaluation indicators, time delay and speed offset, to quantitatively describe this time-shifting characteristics of natural wind associated with multiple time-space scales. First, combined with the empirical Copula function, select the optimal Copula function to describe the joint density distribution of the different wind speed time series, providing an important basis for the calculation of the correlation coefficient. Then analyze the time-shifting characteristics of natural wind based on different time scales (from 10 min to one month) to determine the optimal time analysis scale. Finally, based on the optimal time scale, the change law of the time delay and speed offset between natural wind sequences at different spatial points is studied.

Abstract

Two-phase loop thermosyphon is promising in free cooling of data center, base station, and other industrial buildings with high temperature heat source. The large-scale circulation loop of its natural phase-change cycle is complex with high internal flow resistance in these context and thus is hardly related to the thermodynamic properties of working medium, which has yet to be revealed in detail. In this study, a distributed steady-state model was implemented to verify the heat transfer performance of a two-phase loop thermosyphon with different working mediums under high internal flow resistance for building air-conditioning. Simulative study shows that the behaviors of R744 and R410A are relatively superior, and R744 is outstanding at the extreme high thermal resistance condition. The operating states with different mixed refrigerants are more stable in comparison with pure refrigerants. Eco-friendly mixed refrigerants with low cost and high safety are alternative selections for two-phase loop thermosyphon in building applications.

Abstract

The reform of electricity market in China has made the market participants more active, which has posed a series of challenges for the power system planning with high penetration of renewable energy. This paper proposed a price-driven bi-level model for transmission expansion planning. At the upper level, the investment cost and operation cost of the transmission system is comprehensively considered. While at the lower level, the market clearing model is established, i.e., energy market and reserve market. The local marginal prices are obtained to guide the expansion planning of the upper level. By integrating the market clearing model of the lower level using Karush-Kuhn-Tucker condition, a mixed-integer nonlinear programming problem is formulated, which is further solved by a Heuristic method. In the case study, a modified Garver-6 bus system is utilized to verify the validity of the proposed method.