[Stoves] Principal component questions

Wed Aug 23 00:58:08 CDT 2017

A note of warning – principal component analysis is a tricky technique. Famously, Michael Mann used it to develop his “hockey stick” and Steve McIntyre showed how Mann’s misuse would always produce a hockey stick.  Indeed, it could produce one even when fed by random data! Before you plunge in (and it seems so seductive) make sure you have a good mathematician alongside.

From: Stoves [mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Frank Shields
Sent: Wednesday, August 23, 2017 6:59 AM
To: Discussion of biomass cooking stoves
Subject: Re: [Stoves] Principal component questions

Dear Stovers,

Please take a look at this link:

http://setosa.io/ev/principal-component-analysis/

Way over my head but I am thinking something like this could be used to match up biomass to specific stoves. I was thinking of a ’spider graph in the past but this looks better. 

I have a suggested test package, producing many results, that can be done on many different biomass. A specific stove uses the different biomasses tested and the results as to time, fuel used, PM2.5 etc reported. The biomasses that produced acceptable combustion that completes a task people are happy with is listed. We use this method to determine the tests within the package that are important in determining the best fuels to use and the range they must fall into.

I am thinking the vertical column from Alcoholic drinks > to > Sugars we replace with the packages of tests we are doing on the biomass. Like Fixed Carbon, Bulk carbon density, resins, length to width ratio, average length,…… 

Across the top replace England > to > Wales with the fuel used like Fuel A, Fuel B, Fuel …  This using the same stove but different fuels that work. 

The results all need to be in the same units so I suggest establishing a typical range for each test. Resins could go from zero to 10%, Length to width from 1:1 (cube) to 1:50 (long stick). The results found on Fuel A would be reported as a percentage of the typical. If found 2% resin then the reported value is 20 (for 20%). Some might go above 100% but that is ok.

It seems they have a 3 dimension but I don’t see the third values. They have foods vs country so what is the third? 

The purpose is to determine the tests used that make a difference and knowing what they measure and range that produce the best results we can pretest fuels and determine if they will work OR find the stoves that are best for using them. 

Frank 

Frank

Frank Shields

Gabilan Laboratory

Keith Day Company, Inc.

1091 Madison Lane

Salinas, CA  93907

(831) 246-0417 cell

(831) 771-0126 office

fShields at keithdaycompany.com

franke at cruzio.com

_______________________________________________
Stoves mailing list

to Send a Message to the list, use the email address
stoves at lists.bioenergylists.org

to UNSUBSCRIBE or Change your List Settings use the web page
http://lists.bioenergylists.org/mailman/listinfo/stoves_lists.bioenergylists.org

for more Biomass Cooking Stoves,  News and Information see our web site:
http://stoves.bioenergylists.org/

Thanks

Frank

Frank Shields

Gabilan Laboratory

Keith Day Company, Inc.

1091 Madison Lane

Salinas, CA  93907

(831) 246-0417 cell

(831) 771-0126 office

fShields at keithdaycompany.com

franke at cruzio.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20170823/2f61aad0/attachment.html>