可再生能源的电流感应

By Alex Latham, Systems Engineer, and
Shaun Milano, Strategic Marketing Manager
亚博棋牌游戏Allegro微系统有限责任公司

In grid-tied photovoltaic systems, designers must focus on minimizing cost per watt to deliver the best possible return on investment

随着对全球能源需求的越来越担心和环境意识的整体扩张，电力电子应用的设计者处于恒定压力，以提高效率。亚博尊贵会员智能电网的出现，插入式混合动力电动车（PHEV）和全电池电动车（BEV）以及电网绑定光伏（PV）和其他网格捆绑的可再生能源系统，需要开发高 -效率功率逆变器。通常，更高的效率与更高的应用成本和降低的性能相关。

技术的真正进步能够提高效率，同时保持或提高成本和性能。例如，随着微处理器成本的不断下降，更先进的控制算法可以廉价地嵌入到那些以前简单、低效、性能差的模拟控制成为常态的应用中。亚博尊贵会员

With this trend continuing, the limiting factor for performance and cost is shifting to the numerous sensors required for advanced control algorithms.

对于用于电流的电流光伏系统中使用的电流传感器，设计旨在最大限度地减少每种瓦特的成本，以便在太阳能中提供最佳可能的投资回报（图1）。

图1。整个并网系统都需要电流传感器来控制变流器和逆变器，优化太阳能电池板的功率提取，并进行安全故障检测。

光伏系统

对于并网光伏系统，太阳能电池板的能量转换通常分两个阶段进行。首先，DC/DC转换器用于将来自面板或阵列的电压转换为接近电网电压的电压，以及最大化从面板提取的功率。然后，使用逆变器将直流电转换为交流电并与电网同步。由于每个转换系统通常由一个制造商整体设计，因此这些阶段或系统中的任何一个都没有标准或广泛认可的拓扑结构；每个阶段的设计和优化都是为了与其他阶段一起最有效地工作，以获得最高的整体效率。无论系统设计如何，电流传感器在控制不同阶段的转换中都起着重要的作用，直接影响着系统的效率和有效性。

Conversion and MPPT

在任何光电转换系统,最大的战俘er point tracking (MPPT) is an important aspect, enabling maximum energy extraction from an array. The efficiency of MPPT algorithms is greatly impacted by the sensors used for feedback. The basis of MPPT is that at any given time the I-V curve of each solar panel indicates the optimum operating point, or maximum power point (MPP, see figure 2). However, it is also evident that the I-V curve and, subsequently, the MPP will change with both irradiance and temperature. It is the MPPT algorithm’s job to constantly search for the MPP.

有许多不同的MPPT算法；然而，大多数算法通过测量功率（在DC/DC转换器级的输入或输出处）并改变转换器的工作点沿I-V曲线移动来找到MPP。这些算法基本上都执行某种形式的梯度上升，估计电流（或电压）与功率曲线的斜率，并用它来确定如何改变工作点。如果坡度为正，则增加工作点；如果坡度为负，则减少工作点。

One of the most common algorithms is called perturb and observe. Here, the power from the solar panel or array is measured, the operating point is changed, and the power is measured again. If the power goes up, the operating point is moved in the same direction; if the power goes down, the operating point is moved in the opposite direction. This results in tracking of the MPP, though it also results in steady-state oscillation around the MPP. To minimize the steady-state oscillation the size of the steps can be reduced. However, the update rate must be increased to maintain the tracking speed.

最终，跟踪效率的限制因素将是功率测量中的噪声。这是因为将步长减小到功率测量的噪声级会混淆算法并导致围绕MPP的振荡。所有的“爬山”算法，包括扰动和观测，以及增量电导，或纹波相关控制等，都将从根本上受到功率测量传感器中噪声的限制。

图2. The graph above shows how I-V curves change with both irradiance and temperature. Subsequently, the maximum power point (MPP) will change, and the tracking algorithms must determine the new MPP.

对于大多数MPPT应用，电流亚博尊贵会员传感器所需的带宽为数百赫兹，这意味着通常会对传感器的输出进行滤波或平均，以降低噪声并提高跟踪效率。然而，如果使用纹波相关控制，则传感器是必要的，纹波相关控制使用DC/DC转换器的固有输入纹波来测量电压与功率曲线的斜率。

The inverter stage

In a photovoltaic system, the DC/ DC converter stage is typically followed by an inverter stage, which ties the output of the system to the grid (see figure 1). This signal must be compliant with the utility system’s requirements, including proper synchronization to the grid and low total harmonic distortion (THD). Current sensors are needed in the systems’ control loops to ensure proper connection to the grid. The sensors must not only accurately measure both AC and DC currents, they must have high dynamic performance: very fast response times are needed to react quickly to any change in the grid, shutting off or disconnecting the system in the case of a short circuit (that is, ground fault) or loss of the grid connection (anti-islanding). The high output bandwidth is also needed to measure high-frequency AC currents and harmonics at different points in the system.

For inverters without a transformer or with a high-frequency transformer, sensors must exhibit low offset drift with temperature (high accuracy) so as to control the DC component in the AC current fed to the grid. Regulations differ from country to country for this electrical DC current limit, but generally it is small: on the order of tens to hundreds of milliamperes.

In a typical inverter configuration, there are a number of locations where a current sensor might be used (see figure 3). While one could use an expensive, high-precision current sensor, it is also possible to achieve high accuracy by using one or more economical current sensor ICs in conjunction with the microprocessor already necessary for the system. The on-board intelligence can be used to calibrate out offset error. For example, there will be periods of time where sensors 2 and 3 will have no current flowing through them. One can sample the output of the sensor ICs during those periods and use a moving average filter to generate an internal zero amperage output reference for that sensor IC, dramatically decreasing the offset drift, limited only by the noise and size of the averaging filter.

图3。在典型逆变器配置图中可以看到多个电流传感器。为简单起见，未显示滤波器组件；但是，它们通常放置在变压器的任一侧（如果使用），设计师通常使用变压器的漏感作为滤波器的一部分。

For example, using aACS710-6BB.在20 kHz（具有适当的过滤以避免抗锯齿）的采样，平均2,000个样本（在流过传感器IC的零电流的许多不同持续时间），可以实现1.5 mA的分辨率，零电流偏移可以校准。当该校准的传感器IC与其他传感器IC串联串联时，可以确定两者之间的关系，校准传感器4将在温度下提供非常低的偏移漂移，使其用于感测DC输入到网格的电流的组件。最终，根据所选择的特定拓扑和控制方案，存在许多类似的技术来校准电流传感器，这是根据所选择的特定拓扑和控制方案，可以降低整体系统成本并提供显着的空间节省。亚博尊贵会员

本文档中包含的信息不构成关于本文件主题的Allegro向客户提供的任何代表性，保证，保证，担保或诱导。所提供的信息并不能保证基于此信息的过程是可靠的，或者Allegro探讨了所有可能的失败模式。客户有责任对最终产品进行充分的资格测试，以确保它是可靠的，满足所有设计要求。