Comparison between dynamic programming and genetic algorithm for hydro unit economic load dispatch

The hydro unit economic load dispatch （ELD） is of great importance in energy conservation and emission reduction. Dynamic programming （DP） and genetic algorithm （GA） are two representative algorithms for solving ELD problems. The goal of this study was to examine the performance of DP and GA while they were applied to ELD. We established numerical experiments to conduct performance comparisons between DP and GA with two given schemes. The schemes included comparing the CPU time of the algorithms when they had the same solution quality, and comparing the solution quality when they had the same CPU time. The numerical experiments were applied to the Three Gorges Reservoir in China, which is equipped with 26 hydro generation units. We found the relation between the performance of algorithms and the number of units through experiments. Results show that GA is adept at searching for optimal solutions in low-dimensional cases. In some cases, such as with a number of units of less than 10, GA＇s performance is superior to that of a coarse-grid DP. However, GA loses its superiority in high-dimensional cases. DP is powerful in obtaining stable and high-quality solutions. Its performance can be maintained even while searching over a large solution space. Nevertheless, due to its exhaustive enumerating nature, it costs excess time in low-dimensional cases.

Modeling and Solution of Economic Dispatch Problem for GTCC Units

Economic dispatch problem lies at the kernel among different issues in GTCC units’ operation, which is about minimizing the fuel consumption for a period of operation so as to accomplish optimal load dispatch among units. This paper has analyzed the load dispatch model of gas turbine combined-cycle (GTCC) units and utilizes a quantum genetic algorithm to optimize the solution of the model. The performance of gas turbine combined-cycle units varies with many factors and this directly leads to variation of model parameters. To solve the dispatch problem, variable constraints are adopted to correct the parameters influenced by ambient conditions. In the simulation, comparison of dispatch models for GTCC units considering and not considering the influence of ambient conditions indicates that it is necessary to adopt variable constraints for the dispatch model of GTCC units. To optimize the solution of the model, a Quantum Genetic Algorithm is used considering its advantages in searching performance. QGA combines the quantum theory with evolutionary theory of genetic algorithm. It is a new kind of intelligence algorithm which has been successfully employed in optimization problems. Utilizing quantum code, quantum gate and so on, QGA shows flexibility, high convergent rate, and global optimal capacity and so on. Simulations were performed by building up models and optimizing the solutions of the models by QGA. QGA shows better effect than equal micro incremental method used in the previous literature. The operational economy is proved by the results obtained by QGA. It can be concluded that QGA is quite effective in optimizing economic dispatch problem of GTCC units.

Optimal Load Dispatch of Gas Turbine Power Generation Units based on Multiple Population Genetic Algorithm

In this paper, a multiple population genetic algorithm (MPGA) is proposed to solve the problem of optimal load dispatch of gas turbine generation units. By introducing multiple populations on the basis of Standard Genetic Algorithm (SGA), connecting each population through immigrant operator and preserving the best individuals of every generation through elite strategy, MPGA can enhance the efficiency in obtaining the global optimal solution. In this paper, MPGA is applied to optimize the load dispatch of 3×390MW gas turbine units. The results of MPGA calculation are compared with that of equal micro incremental method and AGC instruction. MPGA shows the best performance of optimization under different load conditions. The amount of saved gas consumption in the calculation is up to 2337.45m3N/h, which indicates that the load dispatch optimization of gas turbine units via MPGA approach can be effective.

基于火电灵活性改造的可再生能源消纳系统成本优化

随着风电等可再生能源的大规模发展,带来了消纳难等新挑战,通过火电灵活性改造可以增强电力系统灵活调节能力、提升可再生能源消纳空间,同时需要考虑火电灵活性改造的经济性和对可再生能源消纳的影响。因此,综合考虑火电机组常规运行成本及灵活性改造后增加的各项成本,结合碳排放成本和弃风成本,建立以总成本最小为目标的火电灵活性改造前后优化调度模型,采用对比差值法计算基于火电灵活性改造的可再生能源消纳系统成本。通过算例,将所建模型应用于对比改造前后、不同碳排放强度、是否考虑弃风等不同方案下的机组运行状况及消纳系统成本大小,并拟合得到可再生能源渗透率和消纳系统成本的关系曲线,验证了模型的有效性。

计及电转气-风-火-核-碳捕集的多源联合系统优化调度

在风-火-核-碳捕集多源联合系统中,核电机组在参与电网调峰时存在调峰深度选择不精确的问题。此外,该多源联合系统还存在风电消纳不足的问题。为此,构建了一种计及线性化核电机组调峰深度模型的电转气(power to gas,P2G)-风-火-核-碳捕集多源联合系统,并对该系统进行了日前优化调度。首先,基于核电机组负荷跟踪模式,通过引入连续变量,提高了调峰深度选择的准确性;然后,分析了碳捕集电厂-P2G联合运行模式及需求响应资源对促进风电消纳的积极作用;最后,以系统综合运行成本最低为目标函数,同时考虑碳交易机制,在Matlab平台搭建仿真模型,验证了所构建多源联合系统的有效性。结果表明,相较于核电机组采用固定调峰档位的多源联合系统,所构建的多源联合系统能够在保证核电机组安全稳定运行的同时,实现风电完全消纳,系统碳排放量与综合运行成本分别下降了13.74%与6.27%,提高了系统运行的低碳性与经济性。

计及改进阶梯型碳交易和热电联产机组灵活输出的园区综合能源系统低碳调度

为了进一步降低园区综合能源系统(park-level integrated energy system,PIES)碳排放量,优化热电联产(combined heat and power,CHP)机组出力的灵活性,提出一种考虑改进阶梯型碳交易和CHP热电灵活输出的PIES低碳经济调度策略。首先,将遗传算法与模糊控制相结合,设计一种遗传模糊碳交易参数优化器,从而对现有阶梯型碳交易机制进行改进,实现该机制参数的自适应变化;其次,在传统CHP中加入卡琳娜(Kalina)循环与电锅炉(electricboiler,EB),构造CHP热电灵活输出模型,以同时满足电、热负荷的不同需求;然后,提出一种柔性指标——电、热输出占比率,进而计算出电、热输出占比率区间,以衡量CHP运行灵活性;最后,将改进阶梯型碳交易机制和CHP热电灵活输出模型协同优化,以系统运行成本和碳交易成本之和最小为目标,构建PIES低碳经济优化模型。算例分析表明,所提策略可有效降低经济成本和碳排放量,同时还可扩展CHP灵活输出调节范围,能够为PIES低碳经济调度提供参考。

基于CHP机组碳排放分析的综合能源系统低碳调度优化方法

针对现有研究对于热电联产(CHP)机组碳排放计算不符合实际物理特性的问题,提出了基于CHP碳排放分析的综合能源系统(IES)低碳调度优化方法。从CHP机组内部结构入手,分析CHP机组内部各部分之间能量的传递过程,绘制CHP机组功率与碳排流向图,列写各个部分的功率平衡方程,得到CHP机组排放废气的等效热功率;计及碳排放量与废气排放速率的关系,结合废气排放的热力学方程,得到CHP机组废气排放速率与输入气流量及输出功率的关系;提出基于CHP机组碳排放分析的IES低碳调度优化模型,以总运行成本最小为目标,求解IES低碳调度优化结果;通过算例分析不同CHP机组运行方式下的优化调度结果,并与现有方法进行对比,验证了所提模型在降低碳排放方面的有效性。

沿海区域综合能源生产单元随机优化调度

为实现沿海区域的海上风电场、海上采气平台和陆上热电联供燃气电厂等多种能源生产子单元的协同化运行,考虑可再生能源出力和氢负荷的随机波动,提出沿海区域综合能源生产单元(coastal integrated energy production units,CIEPU)随机优化调度模型。采用参数化代价函数近似(parametric cost function approximation,PCFA)的动态规划算法求解随机优化调度模型。通过一种基于梯度下降的求解方法--Adadelta法,获得策略函数的一阶信息,并计算梯度平方的指数衰减平均值,以更新策略函数的迭代步长;对随机优化调度模型进行策略参数逼近,从而得到近似最优的策略参数,并逐一时段求解出CIEPU的最优调度计划。最后,以某个CIEPU为例,分析计算结果表明,所提出方法获得的优化调度方案可以提高CIEPU运行的经济性并降低碳排放量,验证了所提方法的准确性和高效性。

计及不可调度DG的孤岛串级交流微电网功率控制方法

针对由可调度和不可调度分布式电源DG(distributed generation)组成的串级孤岛微电网MGs(microgrids)功率控制机理复杂、电压和频率调节难度大等问题,提出了一种新型分散主从功率分配控制方法,以实现含二类DG串级孤岛微电网的电压、频率调节以及功率分配。该方法分别设计了主DG和从DG的控制方法,其中,主DG负责调节微电网的频率和电压,从属DG的MPPT可在一定母线电压下运行,其可最大限度地利用不可调度DG的输出功率,在轻负荷条件下可采用自动功率削减,并有效保障系统的安全稳定性。最后,通过仿真和实验验证了所提方法的有效性。

水电站机组负荷分配中动态规划与遗传算法性能比较

以三峡水库26台机组为研究背景,以计算精度、计算时间和算法稳定性为评价指标,采用数值试验方法,开展了动态规划和遗传算法在机组负荷分配中的性能比较研究。设计了计算精度相同比较计算时间和计算时间相同比较计算精度的具有可比性的数值实验方案,得到了动态规划和遗传算法的性能随机组数变化的关系。结果表明,在机组数小于10台时,遗传算法的性能优于动态规划;当机组数大于10台时,遗传算法性能低于动态规划法。

计及碳捕集的含新能源电网低碳调度策略

在碳达峰、碳中和背景下,碳捕集机组可有效降低电力系统碳排放。但是在含新能源的电网中,碳捕集机组频繁参与系统调峰会降低系统的经济性。为此,文中首先在分析灵活运行碳捕集机组运行原理和碳交易工作机制的基础上,提出引入抽水蓄能机组辅助碳捕集机组参与系统调峰,以促进风电消纳,使碳捕集机组更专注捕碳工作并降低系统碳排放量;然后针对风电并网后的不确定性,引入模糊理论,将系统功率约束中风电和负荷用模糊参数表示,使约束转变为基于可信性的模糊机会约束,运用清晰等价类将模糊机会约束清晰化;接着以系统净收益最高为目标函数,综合考虑机组上网收益、抽蓄收益、碳交易收益、运行成本以及系统安全性约束等因素,构建碳捕集-抽蓄联合运行模型;最后通过CPLEX求解模型。仿真结果表明,引入抽蓄机组后系统净收益提高7.62%,碳排放降低7.01%。引入抽水蓄能机组能够兼顾系统的经济性和环保性。

策略路由技术在厂站远动机通信中的应用

在电力调度数据网双平面建设过渡期某些复杂场景下,要求厂站远动机通过数据网双平面独立响应只有1个IP地址的主站前置机通信连接。远动机应用静态路由技术则只能选择双平面之一作为通信网络,当所选择的平面发生网络故障时,远动机与主站通信中断且无法切换到备用平面。为此提出在远动机通信中应用策略路由技术,通过分析策略路由技术的各种可用方法,根据远动机通信报文特点,选择源地址匹配方法来解决该问题。试验结果表明,策略路由技术能在任一平面网络故障时,自动从备用平面响应同一主站前置机的通信连接,保障了电力调度数据网通信网络冗余的安全性。

MicroGrid Designer:user-friendly design,operation and control assist tools for resilient microgrid and autonomous community

During this decade,many countries have experienced natural and accidental disasters,such as typhoons,floods,earthquakes,and nuclear plant accidents,causing catastrophic damage to infrastructures.Since the end of 2019,all countries of the world are struggling with the COVID-19 and pursuing countermeasures,including inoculation of vaccine,and changes in our lifestyle and social structures.All these experiences have made the residents in the affected regions keenly aware of the need for new infrastructures that are resilient and autonomous,so that vital lifelines are secured during calamities.A paradigm shift has been taking place toward reorganizing the energy social service management in many countries,including Japan,by effective use of sustainable energy and new supply schemes.However,such new power sources and supply schemes would affect the power grid through intermittency of power output and the deterioration of power quality and service.Therefore,new social infrastructures and novel management systems to supply energy and social service will be required.In this paper,user-friendly design,operation and control assist tools for resilient microgrids and autonomous communities are proposed and applied to the standard microgrid to verify its effectiveness and performance.

Optimization of unit commitment based on genetic algorithms

How to solve unit commitment and load dispatch of power system by genetic algorithms is discussed in this paper. A combination encoding scheme of binary encoding and floating number encoding and corresponding genetic operators are developed. Meanwhile a contract mapping genetic algorithm is used to enhance traditional GA’s convergence. The result of a practical example shows that this algorithm is effective.

风险量化的高比例新能源电力系统概率调度

新能源大规模并网给电力系统的运行带来重大技术挑战,传统确定性调度模式缺乏对不确定性条件下的安全风险量化管控,而鲁棒优化调度过于保守。基于随机优化的概率调控技术具有风险可控、统计最优性等特点,但是存在概念与内涵不清、概率建模困难、复杂机会约束模型求解低效等问题,概率调度技术的实际应用面临困难。为此,提出了风险量化的概率调度定义与技术内涵,阐释了处理新能源发电不确定性所遵循的基本原理,然后重点介绍了如下概率调度的关键技术:1)新能源概率建模的数学基础;2)概率备用决策与机组组合;3)考虑弃电与有序用电的概率日前调度计划;4)随机鲁棒自适应实时调度。最后,简要分析了所提方法在省级电网的应用效果,并对后续的研究进行了展望。

考虑广义储能与碳捕集设备联合调峰的电力系统低碳经济调度

在运用碳捕集设备提高深度调峰机组负荷低谷时段出力的同时,采用广义储能提高负荷高峰时段的碳捕集水平,是实现碳达峰、碳中和的重要途径之一。提出考虑广义储能与火电深度调峰的低碳经济调度模型,通过引入碳捕集设备与广义储能解决机组深度调峰损耗大及负荷高峰时段碳排放量高的问题。将深度调峰机组改造为碳捕集机组,提高其深度调峰时段的出力水平;将由价格型需求响应与储能装置构成的广义储能用于调度模型中,提高负荷高峰时段的碳捕集水平;以系统总运行成本最优为目标,制定各主体出力方案。仿真结果表明,所提模型能够缓解机组深度调峰压力,提高碳捕集机组的捕碳水平,兼顾系统的经济效益与低碳性能。

基于深度强化学习的新型电力系统调度优化方法综述

随着新能源并网规模不断扩大,能源形式更加灵活多变,电力系统调度运行面临新的挑战。随着系统复杂度和不确定性增加,传统基于物理模型的优化方法难以建立精确的模型进行实时快速求解,而深度强化学习(DRL)可以从历史经验中自适应地学习调度策略并实时决策,避免了复杂的建模过程,以数据驱动的方式应对更高的不确定性和复杂度。文中首先介绍了新型电力系统调度运行问题;然后,介绍了DRL原理及其分类算法;接着,分析了各类DRL算法在求解新型电力系统调度决策问题时的优势与劣势;最后,对需进一步研究的方向进行了展望。

基于改进遗传算法的虚拟发电厂经济调度优化方法

常规的虚拟发电厂经济调度优化方法主要针对发电调度的变量进行分析,往往出现局部最优解的情况,影响调度优化效果。因此,文章设计了基于改进遗传算法的虚拟发电厂经济调度优化方法。本研究提取虚拟发电厂机组发电并网特征,对各类发电机组的发电情况进行分析;利用改进遗传算法,构建虚拟发电厂调度优化模型,以经济与技术为约束条件,找出不同机组的最优调度优化策略,进而实现虚拟发电厂经济调度的优化;采用对比实验的方式,验证了该优化方法的调度效果更佳,能够应用于实际生活中。

Puzzle:面向深度学习集成芯片的可扩展框架

芯粒集成逐渐成为不同场景下敏捷定制深度学习芯片的高可扩展性的解决方案,芯片设计者可以通过集成设计、验证完成的第三方芯粒来降低芯片开发周期和成本,提高芯片设计的灵活性和芯片良率.在传统的芯片设计和商业模式中,编译器等专用软件工具链是芯片解决方案的组成部分,并在芯片性能和开发中发挥重要作用.然而,当使用第三方芯粒进行芯片敏捷定制时,第三方芯粒所提供的专用工具链无法预知整个芯片的资源,因此无法解决敏捷定制的深度学习芯片的任务部署问题,而为敏捷定制的芯片设计全新的工具链需要大量的时间成本,失去了芯片敏捷定制的优势.因此,提出一种面向深度学习集成芯片的可扩展框架(scalable framework for integrated deep learning chips)--Puzzle,它包含从处理任务输入到运行时管理芯片资源的完整流程,并自适应地生成高效的任务调度和资源分配方案,降低冗余访存和芯粒间通信开销.实验结果表明,该可扩展框架为深度学习集成芯片生成的任务部署方案可自适应于不同的工作负载和硬件资源配置,与现有方法相比平均降低27.5%的工作负载运行延迟.

基于旗鱼优化算法的电力系统动态经济调度方法研究

电力系统动态经济调度(DED)是一种复杂的多目标优化问题,旨在使发电机组的总燃料成本和污染排放最小化。然而,由于约束条件的复杂性,一般的数学方法难以解决这个问题。因此,本文采用了旗鱼优化算法(SFO)来解决考虑风电情况下的动态经济调度问题。为了验证SFO算法的性能,本文使用了5台火电机组的测试系统,并在MATLAB软件中对模型进行了仿真和测试。实验结果表明,SFO算法能够有效降低电力系统的燃料成本和污染排放,从而提高电力行业的经济效益。