首页> 外国专利> MPPT The Neural Network Algorithm for MPPT Control of Solar Inverter

MPPT The Neural Network Algorithm for MPPT Control of Solar Inverter

机译：MPPT用于逆变器MPPT控制的神经网络算法

页面导航

摘要
著录项
相似文献

摘要

The present invention relates to an algorithm for a photovoltaic power generation controller, in which a technique for tracking a maximum power point (MPPT) maximum power point is used to provide an error backpropagation neural network with fast response characteristics that improves problems in the existing method. This is the control algorithm used. Current current I(k) detected at the present time and previous sampling using the time calculated for each sampling period set by the sensor signal detection unit and the voltage detection unit that detects the current signal generated through the PV module and the output voltage of the solar panel. The two current I(k-1) detected at the time point, the current voltage V(k) detected at the current time point, the previous voltage V(k-1) signal detected at the previous sampling time point, and the current power P(k) and previous Power P(k-1), differential current value (dI) between current current I(k) and previous current I(k-1), and differential voltage between current voltage V(k) and previous voltage V(k-1) Using the value (dV), Error range based on PV minimum power value P(m), PV maximum power value P(M), PV average power P(s), PV maximum efficiency Q(M), PV average efficiency Q(N), TACK TIME It is a radio neural network algorithm. The conventional P&O (Perturbation and Observation) control technique is a problem in the method of finding the maximum power operating point by comparing the previous output power and the current output power, and the solar cell array output voltage momentarily operates at maximum power when the solar radiation amount changes. The problem that the response characteristic of the MPPT control drops momentarily due to the problem that deviates from the following control at the point is improved, MPPT control using a neural network capable of active tracking control that improves the low efficiency problem of low insolation by comparing the conductance of the solar cell output and the incremental conductance like the IncCond (Incremental Conduction) control technique to find the maximum power operating point. Algorithm.

机译：用于光伏发电控制器的算法技术领域本发明涉及一种用于光伏发电控制器的算法，其中用于跟踪最大功率点（MPPT）最大功率点的技术用于提供具有快速响应特性的误差反向传播神经网络，该神经网络改善了现有方法中的问题。。这是使用的控制算法。使用传感器信号检测单元和检测通过PV模块生成的电流信号的电压检测单元设置的每个采样周期计算的时间，在当前时间和先前采样中检测到的当前电流I（k）。太阳能板。在该时间点检测到的两个电流I（k-1），在当前时间点检测到的当前电压V（k），在先前采样时间点检测到的先前电压V（k-1）信号以及电流功率P（k）与先前功率P（k-1），当前电流I（k）与先前电流I（k-1）之间的差分电流值（dI），以及当前电压V（k）与先前电压I（k-1）之间的差分电压电压V（k-1）使用值（dV），基于PV最小功率值P（m），PV最大功率值P（M），PV平均功率P（s），PV最大效率Q（M ），PV平均效率Q（N），TACK TIME这是一种无线电神经网络算法。常规的P＆O（摄动和观察）控制技术是通过比较先前的输出功率和当前的输出功率来找到最大功率工作点的方法中的问题，并且当太阳能电池时，太阳能电池阵列的输出电压会瞬间以最大功率运行。辐射量变化。改善了由于此时偏离跟随控制的问题而导致MPPT控制的响应特性瞬时下降的问题，通过使用具有主动跟踪控制能力的神经网络的MPPT控制，通过比较，改善了低日照的低效率问题太阳能电池输出的电导和增量电导，例如IncCond（增量电导）控制技术，以找到最大功率工作点。算法。

著录项

公开/公告号KR20200094808A

专利类型
公开/公告日2020-08-10

原文格式PDF
申请/专利权人 김홍균;
展开▼

申请/专利号KR20180172614
发明设计人 김홍균;
展开▼

申请日2018-12-28
分类号G06N3/08;G05F1/67;
国家 KR
入库时间 2022-08-21 11:06:16

相似文献

专利
外文文献
中文文献