多逆变器型微电网协调控制及优化调度方法研究
文献类型：学位论文
| 作者 | 李忠文
|
| 学位类别 | 博士 |
| 答辩日期 | 2016-11-28 |
| 授予单位 | 中国科学院沈阳自动化研究所 |
| 导师 | 于海斌 ; 臧传治 |
| 关键词 | 微电网 分布式分层控制 下垂控制 多智能体系统 逆变器接口电源 |
| 其他题名 | Research on Cooperative Control and Optimal Dispatch of Multi-inverter Microgrids |
| 学位专业 | 控制理论与控制工程 |
| 中文摘要 | 在接入多个逆变器接口分布式电源的微电网中，微电网通过相应的逆变器实现微电网控制。本文根据控制任务特点和其对时间要求的不同，在三个层面展开研究。第一层为微电网本地控制层，其控制目标是对电源的输出电压、电流和功率进行本地控制，使其电气量能够快速响应上层控制器给定的设定值，该层的时间尺度为毫秒级（对应第二章和第三章的研究内容）。第二层为微电网电压和频率稳定控制层，其控制目标是监测和估计微电网系统状态，并通过设定微电网本地控制层的设定值，使微电网系统补偿其电压和频率的偏移量，该层的时间尺度为秒级（对应第四章和第五章的研究内容）。第三层为微电网优化调度层，其目标是优化用于微电网电压和频率稳定控制的分布式电源的机组组合和输出功率以及储能设备的充放电策略，使微电网的运行成本最小化，该层的时间尺度为分钟或小时级（对应第六章的研究内容）。在微电网本地控制层，主要研究逆变器接口电源的本地控制策略。逆变器接口分布式电源的控制模式包括有功无功（Active-Reactive Power, PQ）控制模式和电压/频率（Voltage/Frequency, V/F）控制模式。处于PQ运行模式的分布式电源只能运行在并网状态。通常PQ控制模式通过同步旋转框架下的比例积分控制（Proportional plus Integral，PI）或静态框架下的比例谐振（Proportional plus Resonant，PR）控制实施。然而，上述控制策略响应速度慢并且对参数不确定性和负荷扰动鲁棒性差。针对上述限制条件，本文提出了基于H2/H∞的PQ控制型逆变器矢量控制策略，并采用粒子群优化算法搜寻最优H2/H∞控制器参数，以代替传统试凑法。仿真及实验结果表明了本文提出的控制算法比传统PI控制算法具有更好的暂态和稳态性能。微电网中V/F控制模式下的分布式电源是保证微电网孤岛运行的关键，为微电网提供电压参考。然而传统基于PI级联控制模式的V/F控制方法在参数波动时恶化，难以满足要求。本文针对具有输出LC滤波和负荷的V/F控制型逆变器系统，提出了一种新型的级联控制策略。该控制策略在电流控制内环采用滑模控制器，在电压控制外环采用混合H2/H∞控制器，具有固定开关频率，总谐波失真低，鲁棒性强，响应速度快等优点。仿真及硬件实验表明本文提出的控制策略比传统基于PI的级联控制策略具有更好的暂态和稳态性能。在微电网电压和频率稳定控制层，主要研究系统电压和频率偏移的补偿策略。传统下垂控制（Droop Control）方法可避免分布式电源间的环流，得到了广泛应用。然而，传统下垂控制方法具有稳态频率和电压偏差的缺点。针对此问题，本文提出了基于多智能体系统（Multi-agent System，MAS）的分布式自动发电控（Automatic Generation Control，AGC）和自动电压控制（Automatic Voltage Control，AVC）策略，该控制策略实时调整各微电网本地控制层的设定值，实现系统的无偏运行，既可以运行在孤岛模式也可以运行在并网模式，支持两种运行模式间平滑切换。通过PSCAD软件仿真验证了本文提出的控制策略的有效性。然而，为了进一步提高微电网的经济效率，需要将大时间尺度微电网优化调度层与小时间尺度的电压和频率稳定控制层紧密融合。为此，本文进一步提出了支持经济性的分布式自动发电控制算法（Economic AGC，EAGC）。该算法可在补偿微电网系统电压和频率的偏移量的同时，最优化各微电网本地控制层的设定值。此外，EAGC算法分布式实施，易于实现，可扩展性好，适用于微电网各种运行模式，如：孤岛模式、并网模式和暂态模式。为了提高针对智能体故障和通信链路故障的鲁棒性，智能体系统之间的通信拓扑满足 规则。通过具有多个仿真情景的仿真测试验证了本文提出的算法的有效性。在微电网优化调度层，考虑微电网中新能源发电、负荷的随机性以及电力价格的变动性，提出一种用于优化分布式电源的机组组合和输出功率以及储能设备的充放电决策的混合两阶段随机规划和滚动优化策略（Stochastic Programming and Receding Horizon Control，SPRHC）。通过该优化策略，可优化用于微电网系统电压和频率稳定控制的分布式电源的机组组合。该策略综合了两阶段随机规划和滚动优化的优点，提高了微电网运行的经济性。随机规划（Stochastic Programming，SP）策略对各种随机性进行数学建模，以最小化系统运行成本的期望为优化目标，建立最优随机优化数学模型，得到期望意义下最优的调度计划。而滚动优化（Receding Horizon Control，RHC）策略，通过反馈机制即时引入实时更新的预测信息，进一步降低不确定性对优化运行的影响。仿真表明本文提出的算法对于孤岛模式和并网模式的微电网都具有更好的效果。 |
| 英文摘要 | In the microgrid that contains certain inverter-interfaced Distributed Generators (DGs), the control of the microgrid is fulfilled through the control of according converters. In this paper, the studies of the microgrid are divided into three control levels, according to the differences of control objectives and time-scales between each control level. The primary level is the local control level of the microgrid, the control objective is to locally control the output voltage, current and powers of an inverter-interfaced DGs, and impel the DGs to quickly track the settings given by the uper level controllers, the time-scale of this level is in millisecond (corresponding to the research contents of the second and the third chapters). The second control level is the voltage and frequency stability control level of the microgrid, the control objective is to monitor and estimate the state of a microgrid, and through regulating the reference values that is send to the local control level, the offset of voltage and frequency of the microgrid can be compensated, the time-scale of this level is in second (corresponding to the research contents of the forth and the fifth chapters). The tertiary control level is the economic dispatch level of the microgrid, the objective is to reduce the operation cost of a microgrid through optimizing unit commitments and active power distribution of the DGs and the charging and discharging profiles of storage system, the time-scale of this level is in minutes or hours (corresponding to the research contents of the sixth chapter). In the local control level of the microgrid, the local control strategies of inverter-interfaced DGs are mainly studied. The control mode of inverter-interfaced DGs contains the PQ control mode and V/F control mode. The DGs controlled in PQ mode can only operate in the grid-connected mode. Traditionally, PQ control mode are implemented by using PI controllers in the rotating synchronous frame or PR controllers in the stationary frame. However, those mechanisms show slow response time for tracking the changing reference commands and are not robust against uncertainties and volatilities in the grid. In this dissertation, in order to deal with those drawbacks, a mixed H2/H∞ optimal control based vector control strategy is implementted for a PQ controlled converter, and the particle swarm optimization algorithm is applied to search a mixed H2/H∞ optimal controller, instead of using a guess-and-check method. Simulation results show that the proposed control method has much more improved transient and steady performance than the traditional PI control methods. In a microgrid, the V/F controlled DG is an important component for an islanded microgrid, it builds up a reference voltage for the microgrid. However, conventional PI-based nested-loop control method for V/F control mode has a deteriorative performance under parameter variations, and is difficult to satisfy the performance requirement. In this dissertation, a novel nested-loop control strategy is proposed for control of a V/F controlled inverter system containing an LC output filter and loads. It utilizes a sliding mode control in the inner current loop and a mixed H2/H∞ optimal control in the outer voltage loop, which provides the advantages of constant switching frequency, low Total Harmonic Distortion (THD), robustness and fast transient response. The simulation and hardware experiments presented in this paper demonstrate the proposed controller’s improved transient and steady-state performance over conventional PI-based nested-loop control strategy. In the voltage and frequency stability control level of the microgrid system, the compensation strategy for offset of the voltage and frequency is mainly studied. The droop-control method is popular since it avoids circulating currents among DGs. However, traditional droop control methods have the drawbacks of steady-state frequency and voltage deviation. In order to deal with this problem, this dissertation proposes a MAS based distributed AGC and AVC strategy, which can adjust the setting values of the primary control level in real-time, and thus restore the deviation of the system, and it can operate in both the island mode and the grid-connected mode and ensure smooth switching between these two modes. Simulation studies in PSCAD demonstrate the effectiveness of the proposed control method. However, in order to improve the economical efficiency of a microgrid, the gap between large time-scale microgrid economic dispatch level and small time-scale voltage and frequency stability control levle need to be decreased. Thus, this dissertation proposes a distributed EAGC algorithm. This proposed algorithm can not only compensate the offset of microgrid voltage and frequency, but also optimal the setting values for the DGs’ local control level at the same time. Furthermore, EAGC algorithm is distributedly implemented, easy to fulfil, scalable, and suitable for all possible operation modes of a microgrid, such as: islanded mode, grid-connected mode, and transition mode. To improve the robustness against agent loss or communication link fault, the topology of agents’ communication network is designed to satisfy the N-1 rule. The effectiveness of the proposed algorithm is demonstrated through the simulation tests which involve plenty of case studies. In the microgrid energy management level, considering the randomness of power output of RESs and load demand, and the fluctuation of electricity prices, a combined Stochastic Programming and Receding Horizon Control (SPRHC) strategy is proposed for optimizing unit commitments and active power distribution of the DGs and the charging and discharging profiles of storage system. With the proposed strategy, the unit commitments can be optimized for the DGs that responsible for the voltage and frequency stability control. The proposed strategy combines the advantages of two-stage Stochastic Programming (SP) and Receding Horizon Control (RHC) strategy, and improves the economics for the operation of a microgrid. With a SP strategy, the uncertainties of a microgrid are modeled and an optimal stochastic programming model is built with the objection of minimizing the expected operation cost of the microgrid, and derive an optimal scheduling plan in the sense of expectation. With a RHC strategy, the uncertainty within the microgrid can be further compensated through a feedback mechanism with the lately updated forecast information. The results of numerical experiments explicitly demonstrate the superiority of the proposed strategy for both island and grid-connected operating modes of a microgrid. |
| 语种 | 中文 |
| 产权排序 | 1 |
| 页码 | 135页 |
| 源URL | [http://ir.sia.cn/handle/173321/19448]  |
| 专题 | 沈阳自动化研究所_工业控制网络与系统研究室 |
| 推荐引用方式 GB/T 7714 | 李忠文. 多逆变器型微电网协调控制及优化调度方法研究[D]. 中国科学院沈阳自动化研究所. 2016. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。