• Energy Research
• Closed Access close
• Restricted close
• engineering and technology close
• CN close
• IT close

Abstract Energy storage systems (ESS) are widely used in active distribution networks (ADN) to smoothen the drastic fluctuation of renewable energy sources (RES). In order to enhance the scalability and flexibility of ESS, a virtual energy storage system (VESS), which is composed of battery energy storage system (BESS), RES as well as flexible loads (FL), is developed in this paper to realize the functionalities of ESS in more cost-effective way in ADN. Aiming at achieving voltage regulation, dynamic pricing strategies based on system voltage condition are designed for VESS. A distributed real-time power management model containing dynamic pricing strategies is proposed to accomplish the voltage regulation and economic power sharing in VESS. Moreover, a set of distributed algorithms, over time-varying unbalanced directed networks, are designed for dynamic pricing strategies and optimal power management model. Furthermore, the convergence property, optimality and system voltage stability are explained by detailed mathematical analysis. Three various case studies which were ran on a real time digital simulator (OPAL-RT OP5600) were designed to validate the effectiveness of the strategy. Finally, simulation results show that the economic power dispatch and voltage regulation are achieved among VESS simultaneously, even in the presence of time-varying directed and unbalanced communication networks.

Abstract To achieve the EU climate and energy objectives, a transition towards a future sustainable energy system is needed. The integration of the huge potential for industrial waste heat recovery into smart energy system represents a main opportunity to accomplish these goals. To successfully implement this strategy, all the several stakeholders' conflicting objectives should be considered. In this paper an evolutionary multi-objective optimization model is developed to perform a sustainability evaluation of an energy system involving an industrial facility as the waste heat source and the neighbourhood as district heating network end users. An Italian case study of heat recovery from a steel casting facility shows how the model allows to properly select the district heating network set of users to fully exploit the available waste energy. Design directions such as the thermal energy storage capacity can be also provided. Moreover, the model enables the analysis of the trade-off between the stakeholders’ different perspectives, allowing to identify possible win-win solutions for both the industrial sector and the citizenship.

Abstract Previous studies have shown that titanium-based glazes containing silicon and calcium tend to crystallize into titanite. In this work, ZnO was found to be an effective additive to produce rutile but not titanite. Images of the micromorphology showed that the glaze doped with 8 wt% ZnO was UV–VIS–NIR of acicular rutile and granular silica. The refractive index of crystalline rutile is higher than that of crystalline titanite, which leads to higher reflectance in the near-infrared range. However, excessive rutile counteracted this improvement in the UV–VIS–NIR spectrum reflectance of solar energy because it decreased the visible spectrum reflectance. As a result, the best ratio of added ZnO was confirmed to be 2 wt%.

Abstract In this study, crude cellulose derived from cornstalk, after bleaching, was used as raw material for the synthesis of sodium carboxymethyl cellulose (CMC) by reacting with the cellulose with NaOH and chloroacetic acid at 75 °C for 1.5 h. Effects of alkali dosage, concentration of chloroacetic acid on the physical and chemical properties of the CMC products were investigated. It was revealed that the reactants alkali reagent/chloroacetic acid/cellulose at the molar ratio of 4.6:2.8:1and 4:2.5:1, or at the molar ratio of NaOH/ClCH 2 COOH ≈1.6–1.64, resulted in CMC products of relatively high water solubility. The viscosity-average molecular weight M v of these two CMC products obtained at molar ratios of 4.0:2.5:1 and 4.6:2.8:1 is in the range of 1.94 × 10 4 –2.48 × 10 4 g mol −1 , and the average DS of the two products are 0.57 and 0.85, respectively. As the solute concentration is above 2 wt%, the viscosity of the CMC-water solution exhibits nonlinear (exponential) increasing with increasing the solute concentration (typical of non-Newton fluids).

The microbial electrolysis desalination and chemical-production cell (MEDCC) is a device to desalinate seawater, and produce acid and alkali. The objective of this study was to enhance the desalination and chemical-production performance of the MEDCC using two types of stack structure. Experiments were conducted with different membrane spacings, numbers of desalination chambers and applied voltages. Results showed that the stack construction in the MEDCC enhanced the desalination and chemical-production rates. The maximal desalination rate of 0.58 ± 0.02 mmol/h, which was 43% higher than that in the MEDCC, was achieved in the four-desalination-chamber MEDCC with the AEM-CEM stack structure and the membrane spacing of 1.5mm. The maximal acid- and alkali-production rates of 0.079 ± 0.006 and 0.13 ± 0.02 mmol/h, which were 46% and 8% higher than that in the MEDCC, respectively, were achieved in the two-desalination-chamber MEDCC with the BPM-AEM-CEM stack structure and the membrane spacing of 3mm.

Experimental investigation into the effects of different pilot amounts of dimethyl ether (DME) on the performance and emission of a single-cylinder directinjection DME engine is conducted. The results show that a DME engine can operate at a wider range of speeds and loads at quasi-homogenous charge compression ignition (QHCCI) mode. The brake thermal efficiency increases while the exhaust temperature decreases. NOx emission decreases by about 30%–50% although there is a slight increase in HC and CO emissions. NOx, HC and CO emissions increase with an increase in the amount of DME pilot. QHCCI is a good way to increase thermal efficiency and decrease NOx emission.

Abstract Predicting and optimizing radiative thermal properties have been acknowledged as an efficient way to improve thermal insulation performance of fibrous materials with high porosity. Based on experimental investigation of infrared spectral of ultrafine fibrous insulations with diameters of 520–650 nm, a method of calculating radiative thermal properties was presented by combining Rosseland equation, Mie scattering theory, Beer’s law and Subtractive Kramers–Kronig (SKK) relation. To ensure the calculation correct the uniqueness analysis was performed for Poly(vinylidene fluoride) (PVDF) fibers, which indicated the valid fiber diameter was less than 1.06 μm. The calculated thermal radiative conductivities by using the method agreed well with the measured data. The effect of fiber diameter on the thermal properties of the fibrous insulations was also investigated to minimize the radiative thermal conductivity. The results indicated that the minimized radiative thermal conductivities by regulating fiber diameters could be approximately 25% smaller than those for experimental fiber diameters. The method of predicting and minimizing radiative thermal conductivities of fibrous insulations demonstrated in this paper could be of great advantage to thermal engineering applications aiming to reducing heat loss and saving energy.

Reactive power consumption of the modern power system is a very important problem that to solve well. Due to the line commutated current source converter-based (LCC) HVDC system is a multivariable and strong coupling nonlinear system, it is difficult to model reactive power consumption of receiving end of LCC-HVDC by traditional analysis method. This paper proposes the reactive power consumption model of receiving end of LCC-HVDC on account of the radial basis function (RBF) neural network. The model with the turn-off angle for input, with reactive power loss on the inverter side of LCC-HVDC for output, using 500kV, 1000MW LCC-HVDC system PSCAD simulation test data training RBF neural network and testing network generalization ability. The results indicate that the reactive power loss model on account of the RBF neural network has much faster convergence speed and higher convergence precision.