Performance improvement of conventional and inverted polymer solar cells with hydrophobic fluoropolymer as nonvolatile processing additive

The morphology of the photoactive layer critically affects the performance of the bulk heterojunction polymer solar cells (PSCs). To control the morphology, we introduced a hydrophobic fluoropolymer polyvinylidene fluoride (PVDF) as nonvolatile additive into the P3HT:PCBM active layer. The effect of PVDF on the surface and the bulk morphology were investigated by atomic force microscope and transmission electron microscopy, respectively. Through the repulsive interactions between the hydrophilic PCBM and the hydrophobic PVDF, much more uniform phase separation with good P3HT crystallinity is formed within the active layer, resulting enhanced light harvesting and improved photovoltaic performance in conventional devices. The PCE of the conventional device can improve from 2.40% to 3.07% with PVDF additive. The PVDF distribution within the active layer was investigated by secondary ion mass spectroscopy, confirming a bottom distribution of PVDF. Therefore, inverted device structure was designed, and the PCE can improve from 2.81% to 3.45% with PVDF additive. Our findings suggest that PVDF is a promising nonvolatile processing additive for high performance polymer solar cells.     

Entropy generation and thermodynamic optimization of fully developed laminar convection in a helical coil

The present paper analyses the entropy generation of the fully developed laminar convection in a helical coil with constant wall heat flux and presents the optimal design based on the minimum entropy generation principal. The important design parameters, including Reynolds number (Re), coil-to-tube radius ratio (δ) and nondimensional coil pitch (λ) are varied to investigate their influences on the entropy generation. The results presented in this paper cover Re range of 100–10,000, δ and λ range from 0.01 to 0.3. Compared with Re and δ, the coil pitch λ is found to have minor influence on the entropy generation. For a demonstrated case, the minimum entropy generation occurs in the range bounded by Re from 2271 to 4277 and δ from 0.17 to 0.3, within which the irreversibility of the system is lowest and the system performance would be optimum. The details show that there is an optimal Re for a helical coil with a fixed δ; meanwhile for a helical coil flow with a specified Re, the smaller δ should be selected when the Re is larger than 5000, and the larger δ should be selected when the Re is less than 5000. These results provide worthwhile information for heat exchanger designers to find the optimal helical coil design from the viewpoint of the thermodynamic second law.     

Experimental study of advanced cogeneration system with ammonia–water mixture cycles at bottoming

In this study, an advanced cogeneration system (ACGS) composed of three turbine systems and an ammonia absorption refrigerator is presented. The overall system configurations and some experimental results of the steady state are shown. The effectiveness of the bottoming stage that employs an ammonia–water mixture (AWM) as the working fluid is confirmed by experimental investigation. The experimental investigation shows that the AWM bottoming power-refrigeration cycles contributes to a higher bottoming efficiency, which is about 7.0% in electric power. Otherwise, the efficiency at the middle stage in conventional combined gas and steam turbine power plants is 4.6%. The cogeneration efficiency at the bottoming reached about 26.5% which is the heat and power ratio to the heat input from the heat recovery steam generator.     

Short-term load forecasting using bayesian neural networks learned by Hybrid Monte Carlo algorithm

This paper presents a short term load forecasting model based on Bayesian neural network (shorted as BNN) learned by the Hybrid Monte Carlo (shorted as HMC) algorithm. The weight vector parameter of the Bayesian neural network is a multi-dimensional random variable. In learning process, the Bayesian neural network is considered as a special Hamiltonian dynamical system, and the weights vector as the system position variable. The HMC algorithm is used to learn the weight vector parameter with respect to Normal prior distribution and Cauchy prior distribution, respectively. The Bayesian neural networks learned by Laplace algorithm and HMC algorithm and the artificial neural network (ANN) learned by the BP algorithm were used to forecast the hourly load of 25 days of April (Spring), August (Summer), October (Autumn) and January (Winter), respectively. The roots mean squared error (RMSE) and the mean absolute percent errors (MAPE) were used to measured the forecasting performance. The experimental result shows that the BNNs learned by HMC algorithm have far better performance than the BNN learned by Laplace algorithm and the neural network learned BP algorithm and the BNN learned by HMC has powerful generalizing capability, it can welly solve the overfitting problem.     

基于CVaR的风电机组日前申报出力优化模型研究

在我国,风电机组需提前一天提交次日24小时的申报出力信息,由于来风具有明显的不确定性和随机性特征,风电机组次日实际出力与日前申报出力会存在一定的偏差。为解决以上问题,首先采用K-means聚类法对风电机组实际出力数据进行聚类,并生成样本集;其次,基于风电不确定性样本集构建风险中性风电机组日前申报出力决策优化模型;再次,构建基于条件风险值（CVaR）的日前申报出力优化模型;最后,基于实际数据进行了算例分析。算例分析结果表明,基于CVaR的日前申报出力决策优化模型能有效减少风电机组日前申报出力与次日实际出力的偏差,且使风电机组获得风险价值更好的申报出力,从而实现申报利润的风险可控性。     

燃煤CO2发电系统两级压缩优化

相对于燃煤水蒸汽发电机组,高参数燃煤CO₂发电系统在效率、体积及选材方面都是具有潜在优势的新型发电系统。对燃煤CO₂发电系统中压缩冷却流程进行优化研究,以压缩机总耗功和系统效率为评价指标,采用EES软件模拟分析。首先分析常规燃煤CO₂发电系统中压缩机部分耗功及压缩机出口温度的变化规律,在此基础上依据能量梯级利用原理,提出优化系统。结果表明:压缩机总耗功随着低压压比增大先减小后增大,即低压压缩机存在最佳低压压比,当主蒸汽参数为30MPa/600 ℃时,最佳低压压比为1.66,系统效率为44.1%;低压压缩机出口温度随低压压比增加而增大,而高压压缩机出口温度变化趋势相反;优化压缩冷却流程,得到系统效率提高且随低压压缩机入口温度和压比增大而增加,当低压压缩机入口温度为160 ℃时,系统效率为46.2%,最佳压比为1.325。     

基于随机森林和长短期记忆网络多元负荷预测的综合能源三层规划调度

针对综合能源系统负荷不确定性对规划和调度造成的高成本低效率问题,提出一种基于多元负荷预测的3层规划调度模型,主要包括预测层、规划层和调度层;基于随机森林回归网络和长短期记忆网络构建了多元负荷的长期和短期预测模型;以综合规划调度成本和调度运行成本最小为目标,采用改进粒子群算法和CPLEX求解器获取最优系统综合成本及配置方案;通过不同场景下的规划调度,分析了设备状态与系统成本。通过对比所构建的3层模型与常规双层模型的规划调度结果,证明了3层规划调度模型的经济性与可靠性。     

基于改进自注意力机制生成对抗网络的智能电网GPS欺骗攻击防御方法

为了避免全球定位系统欺骗攻击(GSA)对相量测量装置造成的危害,提出了一种基于改进自注意力机制生成对抗网络(SAGAN)的智能电网GSA防御方法.首先,通过引入深度学习参数,构建了改进网络-物理模型,利用历史数据计算得到当前时刻的量测值.然后,在SAGAN的生成器和判别器网络中分别融入一个时间注意力模块,提出了一种用于实现网络-物理模型的改进SAGAN防御方法.通过训练改进SAGAN,得到一对判别器和生成器,利用判别器检测采集的量测值是否遭受GSA,当检测到攻击时,利用生成器生成的数据替换欺骗数据,从而实现智能电网对GSA的主动防御.最后,基于IEEE 14节点和IEEE 118节点系统进行仿真测试,结果验证了所提方法的可行性和有效性.     

La3+ 掺杂对 CeO2 红外发射性能的影响研究

随着飞行器飞行速度和发动机燃烧室温度的提高, 金属类隐身涂层材料已无法满足环境使用要求。 CeO2 具有较高熔点、 抗氧化、 较低红外发射率等特点, 被认为是红外隐身应用的候选材料, 但其发射率仍无法满足 高温红外隐身需求。 本文通过固相反应法对 CeO2 进行 La3+ 掺杂改性, 研究掺杂量对 CeO2 粉体红外发射性能的 影响规律。 结果表明, 随着掺杂量从 10 mol% 增加到 30 mol%, 粉体在 3~6 μm 波段的发射率不断降低, 其中 Ce0.7La0.3O2 的红外发射率仅为 0.1。