Time for a reality check

When I read the two articles in the January 2013 issue of EIBI on measurement and verification (M&V), I was torn between whether to laugh or weep in despair.

I recently retired after 30 years working in energy management, I now have the time to ponder the significance of the new edition of the EVO M&V Protocol and reflect on Prof Fisk’s inaugural speech as president of CIBSE last May.

In his speech Prof Fisk said “It would be a very English tragedy to have got a lead on Europe …  only to fritter it away because of some creaky statutes”. I have an alert for Prof Fisk and CIBSE; what is beginning to unfold here is worse than tragedy, it is a scandal.

Five things I note in the EVO Protocol are: it is called a “Protocol”, this is an American initiative that in method and technique is roughly where best practice in the UK was at the beginning of May 1980; this 2012 edition is the first that makes any mention of monitoring and targeting (M&T). Vilnis Vesma, representing his company Degree Days Direct, was the only UK representative involved in the 2012 edition; and while the first edition carried a disclaimer regarding its legal status, this one doesn’t.

We must be clear: this is about interpreting data and application of the result; M&T is not a subset of M&V but both M&T and M&V are overlapping application subsets of a repertoire that spans both. Let’s call it MMV&T.

My fear is that all the effort to develop and promote techniques for MMV&T in the UK – that has cost the Government programme more than £20m in today’s money – will go to waste while UK companies think something called an “International Protocol” carries more weight than a CIBSE Memorandum. I have other concerns too, I don’t have the space here to even scratch the surface but I have written them up in a 50,000 word, evidence-based diatribe that I will be inviting people to respond to before I publish it, but I can give a hint.

Last year CIBSE launched the 3rd edition of its Guide F, described by Prof Fisk at its launch as “fully revised.”. The part relevant to MMV&T is chapters 15-19. Figure 19-14 is a CUSUM graph.  Compare this with the same graph 2nd in the edition of Guide F, or a copy of FEB7.  See the revision?  Someone has edited the dates to make this 26 year-old graph in Guide F appear still relevant. I have a personal interest here because this graph comes from my 1986 book Preparing the Company Energy Plan that CIBSE have never properly acknowledged, while I am the researcher dismissed in the preface to TM41 as having “demonstrated his commitment to an alternative view of the science”!

CUSUM is not mentioned in the Protocol because, while parametric CUSUM has been in everyday use in energy management in the UK since the late 1980s, it is hardly known in energy management outside the UK or in any other field of human endeavour. Really! Try finding this combination of regression and CUSUM in a statistics textbook at any level or the research literature. It is only known to energy managers because of my own efforts to promote it through three books, 11 other guides including GPGs 112, 125, 165, GIL50, FEB7 and many articles, all now out of print. 

Knowing this, now compare the references where the new edition of Guide F states “TM41 sets out a number of forms the weather dependency graph can take”. The 2nd edition made exactly the same statement but offered instead GIL50, GPG 125 and GPG 112.  “Revision” here was no more than changing the reference without checking what it provides.

Now compare Figure 5.4 in TM41 the worked example in GPG310 (2002). See it? Exactly the same data with two entirely different and incompatible interpretations! In GPG310 Prof Day, who wrote one of the articles in the Jan issue of EiBi, thought “£4,694 of savings were made” in TM41 he makes no mention of savings. This second interpretation is from his 2003 paper in BSER&T where he offered it as “an alternative approach to the interpretation of performance lines set out in [GPGs] 165 and 310”. Is he right? No. I have evidence from the site, it was a faulty pipe layout.  In an ESCO contract using Day’s interpretation the hospital would have paid out for the apparent savings but still have the faulty pipe layout and be carrying the cost of the waste.

Look at the example of CUSUM in TM41. The conclusion is that fitting a timeswitch to a reverse cycle heat pump has saved £1,140 (22 per cent). Using the degree days Day used with the same data at the aM&T conference in March 2004, these savings reduce to £616 (12 per cent). Prof Day is right in his article “it is essential to have a … transparent process for measurement and verification”! (my italics). His presentation does not appear to be transparent.

Am I just nit-picking? No.TM41 is like this all the way through. Has anyone yet had a sensible answer for the preheat time using the method on page 54?  I’ve been through Day’s dissertation, I am confident he never has either. The UK still leads the world in MMV&T but nothing now in print reflects this and in buildings this lead is not about to be “frittered away” – CIBSE is squandering it.