基于负荷曲线的水电站群短期运行效益研究

以金沙江中游梯级水电站群为例,结合电网需求,建立了基于负荷曲线运行的梯级水电站群短期联合发电优化调度模型,并采用逐次逼近法与逐步优化法的耦合算法对模型进行求解,比较分析了基于该模型的短期运行发电量效益。结果表明,梯级联合运行更符合电网调度需求,且在周调节基础上进一步日调节,综合效益增幅显著。     

南津渡引水式径流电站日优化运行调度研究

针对南津渡径流式电站水位变化较大的特点。分别应用水力学和总径流线性响应模型方法建立了前池水位和尾水水位的计算方案。提出南津渡水电站的日优化运行模型。根据不同的来水情况，分别采取离散微分动态规划，有后效性动态规划逐次逼近法求解，充分发挥两种算法的快速性，取得了较好的运行效果。与现行调度方法比较表明，优化调度和经济运行可显著提高水电站的发电效益。     

光水混合发电系统优化策略研究

为了提高独立能源发电的可靠性与运行效率,结合光水发电系统的混合发电模式,搭建出光水混合发电系统的优化模型,并基于改进型粒子群优化算法(CPSO)对混合光水发电系统进行仿真计算,比较了混合发电与单一发电系统中传输功率及效益。结果表明,通过对光水混合发电系统中水电优化,解决了光伏发电不稳定问题,更好地利用了太阳能与水电资源优化互补的特性,实现了效益最大化。     

金沙江下游梯级与三峡梯级水库联合发电调度

针对金沙江下游梯级水库的相继建成将会改变长江流域梯级水库综合调度格局问题,构建了金沙江下游梯级水库与三峡梯级水库联合发电调度模型,并采用逐次优化算法进行了求解.结果表明,采用联合优化调度后整个梯级发电效益明显,与原设计方案相比梯级平均发电量可增加6.29;与各梯级单独优化运行相比,梯级发电量可增加1.48%.     

水库长期优化调度发电效益最大模型探讨

在水库长期优化调度问题中，一般以发电量最大为优化调度模型。考虑到整个系统的影响．综合水电站水库自身约束和系统调峰要求．建立以系统发电效益最大为目标的优化调度模型．并通过实例计算比较这两种模型，结果表明：发电效益最大模型较发电量最大模型更为符合实际。     

基于SA-NSGA-Ⅱ算法的水库多目标优化调度研究

针对带精英策略的快速非支配排序遗传算法NSGA-II求解多目标优化问题时存在局部搜索能力较差的缺陷,提出一种基于逐次逼近方法改进的快速非支配排序遗传算法(SA-NSGA-Ⅱ),该算法通过不断调整搜索空间来减小不可行域和支配解占比,从而增强局部搜索能力,快速逼近Pareto真实前沿。以平均出力最大和下游河道适宜生态流量改变度最小为目标,建立小浪底水库多目标优化调度模型,分别采用SA-NSGA-Ⅱ和NSGA-Ⅱ求解模型并比较优化效果。结果表明,在两种算法求解模型生成的混合Pareto前沿中,SA-NSGA-Ⅱ、NSGA-Ⅱ所生成的Pareto前沿点分别占比80.45%、19.55%;SA-NSGA-Ⅱ、NSGA-Ⅱ生成的Pareto前沿中,分别有86.90%、52.67%的点处于非支配地位,且SA-NSGA-Ⅱ较NSGA-Ⅱ的算法运算时间减少了15.32%。因此,在相同初始条件下,SA-NSGA-Ⅱ的优化效果优于NSGA-Ⅱ,验证了SA-NSGA-Ⅱ在水库多目标优化调度中的适用性。     

含混合式抽水蓄能电站的梯级水电站优化调度

为提高梯级水电站调度运行效益、优化水库水位变化过程、减少弃水,考虑湖北省某流域梯级水电站规划新增混合抽水蓄能电站,将抽水蓄能机组作为水电站发电备用机组或利用其抽水工况实现梯级水电站年发电量的最大化。针对不同来水频率下服从皮尔逊Ⅲ型分布的天然入库,提出逐步优化算法(POA)与水位廊道耦合方法,求解含混合抽水蓄能的梯级水电站调度模型,能有效保证在多约束条件下构建发电量最大的中长期调度规则。算例结果表明,新增混合抽水蓄能电站后不仅可增加本级水电站发电量,还可提高抽水发电转换效率、减少弃水,为梯级水电站规划新增抽水蓄能电站提供参考。     

考虑收益风险的梯级水电站群中长期合约电量最优组合模型

针对梯级水电站群参与不同时间尺度的合约电量交易时天然来水和交易电价的不确定性因素导致不同市场份额难以确定、水资源利用不充分及不同时间尺度电量分配不合理导致收益存在较大风险问题，提出了考虑收益风险的梯级水电站群中长期合约电量最优组合模型，该模型采用给定置信度下整个梯级发电收益风险最小为目标，其求解首先由多年来水数据生成随机月尺度来水过程，然后根据梯级水电站占市场比重不同模拟出三种电价函数，采用逐步优化和逐次逼近混合算法对电站出力和合约电量组合循环迭代求解。最后以西南地区某流域梯级水电站为例进行验证。结果表明，所提出的优化调度模型能够合理制定年度和月度双边合约电量申报策略，有效降低收益风险，为梯级水电站群参与不同时间尺度的双边交易市场提供可靠依据。     

基于改进多目标蝙蝠算法的洛河流域水库优化调度

为了发挥水资源综合利用的最大效益,实现水库发电和供水效益双赢的优化调度需求,以水库发电量最大和水库下游用水区域总缺水量最小为目标建立了多目标优化调度模型。同时,提出改进多目标蝙蝠算法(IMOBA)用于求解多目标优化调度模型,该算法引入外部精英蝙蝠集EBS()来存储算法寻优过程中的非劣解,设计一种基于混沌原理的脉冲发射率生成机制来增强算法的搜索能力,并加入变异操作以提高算法的种群多样性。将改进多目标蝙蝠算法应用在洛河流域故县水库,得到丰、平、枯三种典型年下的故县水库调度方案集。结果表明,IMOBA算法合理有效且优于多目标蝙蝠算法(MOBA),能较好地反映洛河流域故县水库的发电和供水要求。同时验证了多目标优化调度模型的可行性,为流域水库多目标优化调度方案的制定提供了新途径。     

考虑反调节水库作用的木里河梯级水电站水库中长期优化调度研究

木里河是雅砻江中游最大支流,目前已形成两库四级梯级水电站群,充分利用梯级末端立洲水库的反调节能力,将有效提升流域整体发电效益。为分析梯级水电最优运行方式,以流域梯级枯期最小出力最大和梯级年发电量最大为目标函数,建立了木里河流域两库四级的优化调度模型,并采用逐渐优化算法(POA)求解,最大化利用末端水库反调节能力。结果表明,通过立洲水库的反调节作用,木里河流域梯级枯水期的可靠出力得到有效改善,发电效益得到有效提升。     

Bacterial Foraging Algorithm based solar PV parameter estimation

The abundance and non-polluting nature of solar energy has aroused the interest of many researchers. This worldwide attention of photovoltaic panels has led to the need of generating accurate model for solar photovoltaic (PV) module before proceeding to the installation part. However, accurate modeling of solar PV characteristics is difficult; since the manufacturer’s datasheet provides only four values such as V_mp, I_mp, V_oc, and I_sc. Further, for accurate modeling precise estimation of model parameters at different environmental conditions are very essential. On the other hand, optimization technique is a very powerful tool to obtain solutions to complex non-linear problems. Hence, in this paper, Bacterial Foraging Algorithm is proposed to model the solar PV characteristics accurately. A new equation has been evolved to determine the values of V_oc, V_mp accurately; since these values decides the closeness of the simulated characteristics. Model parameters are extracted for three different types of solar PV panels. A systematic evaluation and performance comparison of Bacterial Foraging Algorithm with other optimization techniques such as Genetic Algorithm and Artificial Immune System has been done and the best computational technique is derived based on performance criteria such as accuracy, consistency, speed of convergence and absolute error. Extensive computations are carried out for the proposed method, as well as for Genetic Algorithm and Artificial Immune System to substantiate the findings.     

基于煤耗特性的火电厂动态成本模型研究

在"厂网分开,竞价上网"的电力体制下,电能成本的分析对发电企业的生产运营已显得尤为重要。重点研究在"煤电联动"机制还没有完全实现、机组负荷不稳定的条件下,电能成本如何变化。在分析了电能成本构成的基础上,运用定性、定量相结合的方法给出了机组成本、利润函数模型,并以实际数据分析了机组负荷率、煤耗率及燃煤价格对成本利润的影响。最后以某300MW电厂为例,给出了机组实时成本利润曲线。对火电企业的运营有一定指导意义。     

Maximize the operating profit of a SWRO-PRO integrated process for optimal water production and energy recovery

Pressure retarded osmosis (PRO) is a promising technology to reduce the specific energy consumption and the operating expenditure of a seawater reverse osmosis (SWRO) plant. In this study, a simple analytical PRO model is developed to predict the PRO performance as the dilution of draw solutions occurs. The model can predict the PRO performance with a high accuracy without carrying out complicated integrations and experiments. The operating profit of SWRO-PRO is also studied by calculating the profit generated for every m³ of seawater entering the process because maximizing the operating profit is the uttermost objective of the SWRO-PRO process. Based on the PRO analytical model, the operating profit and the dynamics of the SWRO-PRO process, a strategy has been proposed to maximize the operating profit of the SWRO-PRO process while maintaining the highest power density of the PRO membranes. This study proves that integration of SWRO with PRO can (1) push the SWRO to a higher recovery and maintain its high profitability, (2) effectively reduce the specific energy consumption of desalination by up to 35% and (3) increase the operating profit up to 100%.     

激光无线输能系统光伏特性及最大功率跟踪研究

针对激光无线输能系统接收端激光光伏阵列由于激光强度不均匀和不稳定而导致其输出P-U特性曲线呈现多峰、波动性强、稳定性差的问题,首先结合砷化镓激光光伏电池的输出特性,建立激光光伏电池阵列模型,分析其多峰产生的原因。然后利用捕猎搜索算法动态调节粒子群算法中粒子的搜索空间,改进粒子的控制机制,提高了算法的收敛速度、跟踪精度和稳定性,且可有效避免算法陷入局部最优解。最后基于交错并联Boost电路,通过Matlab/Simulink仿真验证了该算法在激光无线输能系统输出最大功率跟踪方面应用的有效性和可行性。     

多水库优化调度的研究

本文针对红水河上四个梯级水电站建立了确定型,四库优化调度的数子模型,并用动态规划求得初始解的基础上应用逐次优化方法求解该问题。本文中对逐次优化法的收敛进行了证明,编制了计算软件,并给出计算结果。     

A new method for optimal parameters identification of a PEMFC using an improved version of Monarch Butterfly Optimization Algorithm

In this paper, a circuit-based model of proton exchange membrane fuel cell (PEMFC) is developed for optimal selection of the model parameters. The optimization is based on using an improved version of Monarch Butterfly Optimization (IMBO) algorithm for minimizing the Integral Time Absolute Error between the measured output voltage and the output voltage of the achieved model. For validation of the proposed method, two different case studies including 6 kW NedSstack PS6 and 2 kW Nexa FC PEMFC stacks have been employed and the results have been compared with the experimental data and some well-known metaheuristics including Chaotic Grasshopper Optimization Algorithm (CGOA), Grass Fibrous Root Optimization Algorithm (GRA), and basic Monarch Butterfly Optimization (MBO) to indicate the superiority of the proposed method against the compared methods. Final results show a satisfying agreement between the proposed IMBO and the experimental data.     

An evaluation of the life cycle cost of rapeseed oil as a straight vegetable oil fuel to replace petroleum diesel in agriculture

The use of straight vegetable oil (SVO) as biofuel has been recognized as a valid substitute of diesel fuel in the agricultural sector under specific circumstances. Its direct use reduces most of the chemical processes involved when converting it into biodiesel, thus lowering harmful emissions. This study presents the economic analysis of a self-supply farming model that uses rapeseed as its fuel base. This model addresses agricultural environmental concerns and can even minimize dependence on the fluctuating costs of diesel fuel. The use of SVO in agriculture can help reduce farmers’ vulnerability to fossil fuel prices. The economic evaluation of the model proposed in this study shows clear economic benefits of introducing rapeseed to the traditional crop rotation of wheat and barley. The key factors analyzed in this model are diesel fuel price, diesel fuel grants and crop aids. The current situation in Spain favors the use of diesel fuel in agriculture rather than rapeseed SVO due to an 8% profit difference. However, results show that changes in key factors slightly affect the profit margin, calculating a difference of only 3.7% for particular factor combinations. Combined environmental-friendly agriculture supporting policies are necessary to cover this slight profit difference to promote this biofuel.     

二倍体多目标遗传算法在水资源优化中的应用

针对遗传算法存在早熟、收敛性和解分布性差的问题,在孟德尔多目标简单遗传算法的基础上提出一种求解多目标的二倍体遗传算法,引入变异算子和聚类方法使求得的Pareto解集具有更好的分布性.通过多目标测试函数和二倍体多目标进化算法在流域水资源调度配置中应用,测试结果较好.     

汽轮机叶片枞树型叶根轮缘优化研究

叶片是汽轮机中重要的零部件之一,其可靠性直接关系到整个机组的安全运行。因此,通过改进叶片结构来弥补材料强度方面的不足,减少叶片事故,进而提高整个机组的安全性是非常必要的。对汽轮机叶片枞树型叶根轮缘完成了多变量优化分析工作,基于APDL编程语言对枞树型叶根轮缘进行了多参数建模,通过求解7个特征变量的优化值,以达到叶根轮缘最大、等效应力最小的目的;并对具体的叶片分别采用零阶算法结合一阶算法、智能优化算法(遗传算法、粒子群算法、模拟退火算法)及模式搜索等方法进行了优化。分析结果表明,在综合考虑精度及优化时间的情况下,模式搜索算法是解决本问题的最佳方法。研究结果可以为汽轮机叶片叶根轮缘部分的设计提供理论支持,并在一定程度上提高透平机组的运行可靠性。     

Kinetics of methane decomposition to COx-free hydrogen and carbon nanofiber over Ni–Cu/MgO catalyst

Kinetic modeling of methane decomposition to CO_x-free hydrogen and carbon nanofiber has been carried out in the temperature range 550–650 °C over Ni–Cu/MgO catalyst from CH₄–H₂ mixtures at atmospheric pressure. Assuming the different mechanisms of the reaction, several kinetic models were derived based on Langmuir–Hinshelwood type. The optimum value of kinetic parameters has been obtained by Genetic Algorithm and statistical analysis has been used for the model discrimination. The suggested kinetic model relates to the mechanism when the dissociative adsorption of methane molecule is the rate-determining stage and the estimated activation energy is 50.4 kJ/mol in agreement with the literature. The catalyst deactivation was found to be dependent on the time, reaction temperature, and partial pressures of methane and hydrogen. Inspection of the behavior of the catalyst activity in relation to time, led to a model of second order for catalyst deactivation.