这样简单的算法是当地经验,出于同样的原因,所以根据你的位置可以有非常糟糕的性能。在另一方面,一些人认为正确校准它们可以(超过90%)[1]准确。它们中的大多数都是位于Zambretti算法或它的某些变体。该算法产生的最初预测设备中实现公司(Negretti和西班牙舞)[2]XX世纪早期。[![在这里输入图像描述][3]][3]它认为的绝对值的压力,压力的趋势、季节和风向(尽管风向和季节影响一个小的输出)。公式的算法描述和翻译(本文)[1],但是在总结它的工作原理计算Z美元数量列表预测如下:1。$ P $在测量压力,温度在摄氏温度T和美元海拔米h美元计算大气压力降低到海平面P_0美元。有很多公式,一个常见的一个是:$ P_0 = P \离开(1 - \压裂{0.0065 h} {T + 0.0065 h + 273.15} \右)^{-5.257}2美元。计算趋势的压力,如果压力下降是* * * *计算预测数Z = 130美元\压裂{P_0}{81}——如果美元稳定压力* * * *计算预测数Z = 147美元\压裂{5 P_0}{376}——如果美元上涨压力* * * *计算预测数Z = 179美元\压裂{2 P_0}{129} 3美元。 Adjust $Z$ for wind direction: - For Northerly winds adjust $Z=Z+1$ - For Southerly winds adjust $Z=Z-2$ 4. Adjust $Z$ for the season: - If Winter adjust $Z=Z-1$ - If Summer adjust $Z=Z+1$ 5. Look up for the forecast from the following table: [![enter image description here][4]][4] This was made for the UK, and many people have developed modified version to work in other areas. You will find many implementations online with downloadable code, as [this one in javascript][5], or [this one in C for Arduino UNO][6], an explanation of this Arduino UNO implementation can be found [here][7]. [1]: https://web.archive.org/web/20110610213848/http://www.meteormetrics.com/zambretti.htm [2]: https:// [3]: https://i.stack.imgur.com/jfAuY.png [4]: https://i.stack.imgur.com/WCUeP.jpg [5]: http://www.beteljuice.co.uk/zambretti/forecast.html [6]: https://github.com/fandonov/weatherstation [7]: https://cdn.hackaday.io/files/274581174763072/english_description.pdf
Baidu
map