"The company said in statement that
it believe that “this was an isolated incident as turbine malfunctions
are rare." The
company said it is working to determine “the cause of the turbine
failure and assessing other turbines on site as a quality-control
measure.” ------------------------------------------------------------------ There's two statistical generalizations here:
The first, is the term 'Isolated incident' which should be interpreted as 'they have only seen it once here'.
The second is "malfunctions are rare"... which means, out of a group of X units, they have seen failures Y times.
To identify "malfunctions" as rare is clearly a misrepresentation, as function means it fully operational in all aspects. Anyone who drives through a wind farm on a windy day and sees one of ten machines not spinning, is witness to 10% observable malfunction rate... AND... although they're turning, they may not actually be GENERATING... so from a visual observation standpoint, the malfunction rate could be substantially higher than 10%. An excellent example was the hundreds of wind turbines located on the eastern ridge of the Tehachapi basin from the mid '80's through about four years ago... around 60-70 units originally built in the 25-50kw range spun vigorously all those years, but never produced, because the gearboxes had failed. To keep the impression of function, they simply removed the reduction gears from the units and allowed the sails and pitch control governors to continue operating. The third aspect of this, is the advertised reasoning for 'assessing other turbines on site as a quality-control measure'. This is a misstatement.
Quality control, is when you define and affect a method for assessing the success of a production process, and in doing so, you determine the failure rate of parts, in the pursuit of having a low failure rate with respect to reasonable production cost. Failures have NOTHING to do with Quality Control... other than the statistical RATE of failures is relatively predictable.
Failures have to do with Quality Assurance. Quality Assurance is the process of methodically REMOVING FAILURES as EARLY in all processes as possible. Here's an in-scheme explanation:
A supplier provides a thousand bolts to a manufacturer, with a Quality Control target of under 3 failures per 100 bolts... That means, in a 1000-bolt order, they'll have equal to, or less than 30 bad bolts. The supplier maintains their contractual obligation, therefore, by providing an extra SIX bolts for every 100 they send out (3 for the expected failures, and 3 more just for good measure). To QUALIFY the shipment, the supplier takes a SAMPLE of every-so-many bolts, and subjects it to a variety of tests to prove that they're being made 'right'. If they get a bad one, they go back through their manufacturing process and CORRECT it. They go through, pull out all the bad ones, and restart the process.
Then the batch goes out the door.
At the reciever, the bolts are counted, then sent into their system for use.
IF they have a QA tier in place, those bolts are checked with a variety of tests to assure that EVERY BAD BOLT is found. One of the first tests, is to WEIGH each bolt. Then they run it through a machine that aligns it just right, then drop it onto a special angled plate, where it bounces off and lands on a conveyor. There's a microphone array on the plate, and they LISTEN to the sound that bolt makes when it hits the plate... between these two tests (which can happen really fast) they can find LOTS of bad bolt problems, and reject the suspect bolts, so they never get to the assembly line.
Sending guys out to look at a wind turbine structure AFTER it's been built, ain't Quality Control...
it's damage control.
Statistics are used to suggest 'possibility and probability' of something happening. The true definition of 'Possibility and Probability' - If something happens ONCE... it is POSSIBLE. - If something happens TWICE- It is PROBABLE.
They fracture, burn, melt, fall, explode, crumple. 90% of a windmill's problem, is it's elevation. Things that are far off the ground are subject to some pretty incredible circumstances- temperature, deflection, vibration, ice... and finally, gravity.
Only two things you can depend on: Gravity, and Triangles. Gravity and Triangles will ALWAYS prevail.
------------- Ten Amendments, Ten Commandments, and one Golden Rule solve most every problem. Citrus hand-cleaner with Pumice does the rest.
|