DK181409B1

DK181409B1 - Method to predict and measure precipitation falling onto wind turbine blades.

Info

Publication number: DK181409B1
Application number: DKPA202100167A
Authority: DK
Inventors: Peter Kaad Johansen Oluf; Skov Johansen Martin
Original assignee: Pj Teknik Holding Aps
Priority date: 2021-02-15
Filing date: 2021-02-15
Publication date: 2023-10-23
Also published as: DK202100167A1

Abstract

Wind turbine blades are in their nature fully exposed to precipitation. When the blades rotate the speed of the blades will give rise to wear, where the blades strike precipitation. Especially the front edge of the blades near the tip, where the rotation speed is at the highest, are at high risk of wear and resulting damage. This invention describes a new method for predicting and measuring the precipitation towards the blades, giving the ability to lower the speed of the rotation to lower the wear and damage to the blades. The method is based on measurement of the electrical charge through the blades due to the conditions of the skies and air near the turbine.

Description

DK 181409 B1

Introduction

Wind turbine blades are in their nature fully exposed to precipitation. When the blades rotate the speed of the blades will give rise to wear, where the blades strike precipitation. Especially the front edge of the blades near the tip, where the air speed is at the highest, are at high risk of wear and resulting damage.

With live information on the actual precipitation affecting the blades, it is possible via the turbine control system to lower the speed of the blades to limit the wear and thereby extend the life time and delay blade maintenance service visits.

In order to control the path of a lightning strike discharge most blades make use of conductive lightning receptors placed on the surface of the blade. The receptors are connected to a lightning down conductor inside the blade, which in turn connects to the nacelle and from there the tower down to ground.

As a cost saving measure, the lightning down conductors may be used for the dual purpose of measuring the charge flow originating from the atmospheric conditions (charged particles, electric field potential) surrounding each blade.

Making use of the property that precipitation lead to a distinctly higher electrical charge flow, compared to other atmospheric conditions without precipitation, can be used as a triggering mechanism with a potentially very high reaction speed, i.e. within seconds of a first rain drop / precipitation particle hitting the blades. Exemplifying this, see figure 1.

Areas of use

The invention concerns a method for using electrical charge measurements from wind turbine blades to detect precipitation, in order to prevent or reduce wear of the blades by lowering the rotational speed of the blades.

Prior Art

Several methods exist to measure precipitation on wind turbines as stated in patents WO 2018/091056 and US 2003/0165379.

In WO 2018/091056 a number of methods for measuring rain into turbines stated as:

, DK 181409 B1 - A rain gauge, - An optical rain gauge, - An acoustic rain gauge, - A disdrometer, such as a Disdrometer RD-80 (DISTROMET AG, Zumikon, Switzerland), -An acoustic disdrometer, or - A laser drop-sizing gauge.

In US 2003/0165379 it is mentioned to have sensors to identify particles like rain and hail, but the invention does not describe the sensing methods in any detail.

A system to measure charge flow in wind turbine blades is described in the patent PA 2019 00839.

This patent describe how to measure charge flowing through wind turbine blades and the use of those measurements to detect if the down conduction system is defective. The patent does not describe any use of that system for measurement of precipitation.

In WO 2020/187832 A1 a method is described where measurement of the electrical potential of a wind turbine blade is used as an early severe weather warning system, based on threshold detection.

The method focus on the atmospheric build-up of electrical charge as a precursor to severe weather conditions, and not measurement of precipitation.

US 2012/0287549 A1 describe a lightning protecting device, where the operation relies on measurement of the voltage potential of lightning receptors on the blade. The patent does not involve measurement of precipitation.

In the Journal of Atmospheric Sciences, volume 34, November 1977, a paper titled ‘Characteristics of Raindrop Charge and Associated Electric Field in Different Types of Rain’ is presented. The paper is based on data from a measurement setup targeting detection of the charge of individual raindrops. While the data presented indicate that the average charge of rain drops can be close to 0, it also demonstrates that the charge of individual drops most often lie in the range +0.5-1.0 pC. The paper is in no way related to wind turbines, and the sensing technique is significantly different to the one proposed in the present patent.

Measurement of the charge flow through the wind turbine, or wind turbine blades, with the purpose

; DK 181409 B1 of measuring precipitation is not present in the referenced patents or journal.

The technical problem to be solved

Wear and the resulting damage to wind turbine blades are to be reduced. During periods with precipitation, and where the turbine blades are rotating during normal operation, this can be achieved by reducing the rotational speed. This will in turn reduce the relative velocity of precipitation impacting the surface of the blades, and hereby the wear.

It is of general interest to keep wind turbines operating at their designed optimum blade speed to maximize the power production capabilities. As such, it is important to impose limitations to the blade speed due to precipitation only when necessary. I.e. no 'false alarms’ and no excessive delay when indicating the end of a period of precipitation.

Starting with this baseline, the targeted properties of a precipitation detection method extends to: e Localized, i.e. detection for the individual turbine. e Non-intrusive, i.e. no additional mounting of external equipment, and associated cable routing to the inside of the turbine. e Low cost / low maintenance e No detection of heavy moisture levels, or condensation effects (dew) e Optional indication of the severity of precipitation

The new methodology

Using the electrical charge inherent to precipitation to detect its very presence on wind turbines is a novel method.

The method makes use of the property that the electrical potential at the point of condensation, i.e. clouds often more than a kilometer above ground level, is significantly different from that at ground level. In this context, ground level include the electric potential of the wind turbine structure itself.

The precipitation particle will thus carry a charge, which is equalized when impacting the wind turbine blade. This equalization is detectable via a charge flow, through the blade.

In reference to "Characteristics of Raindrop Charge and Associated Electric Field in Different Types of Rain’, using a sensor that aggregate the absolute change of charge flow over time, will provide a

4 DK 181409 B1 good correlation with the actual rate of precipitation, independent from the average charge of the precipitation. This property may as an example be obtained using a measurement circuit including a component similar to a diode bridge rectifier.

The referenced paper from the Journal of Atmospheric Sciences is indicative of a reasonable correlation between charge flow and the rate of precipitation (e.g. measured in mm/hour). Improved correlation may be obtained based on additional data for seasonal trends and general conditions of the geography in which the technology is deployed. By recording the historic charge levels detected for a specific geographic measurement point (e.g. a turbine), the thresholds defining levels of precipitation may be adapted/scaled. Examples include establishing a median of the measurements over a year, and assuming that there is no precipitation more than half the time, the threshold for no precipitation can be conservatively set to e.g. 3 times the median, based on the knowledge that precipitation usually leads to charge levels at least an order of magnitude greater than the levels seen without precipitation. In a similar fashion, as an example, a 'high’ level of precipitation can be classified based on the highest sustained charge measurement during a minute among a dataset covering a full year, by setting the threshold for 'high' levels of precipitation e.g. 3 times lower than the mentioned sustained charge measurement.

To the extent a qualitative rather than a quantitative measure is needed, the methodology of the patent may be used without local data input. In essence, the method described will be able to answer whether the level of precipitation is none, low, medium or high. The accuracy of these levels may be further refined based on additional information available either offline or online to the sensor, as exemplified in the above paragraphs.

The technical effect

Based on the new precipitation detection methodology, the following is implied: e The detection method is based on a localized measurement of the blades of the wind turbine in question. e It requires no additional external sensing equipment, since the charge flow can be measured on the inside of the turbine blade. e As it will not be directly subjected to the outdoor environment, it may be realized in a form with low maintenance need. With the potential of being added as a pure software functionality on top of a down conductor measurement system, such as that exemplified in

DK 181409 B1

PA 2019 00839, it has the potential of being offered at a comparatively low cost. e Since moisture and condensation around or on the wind turbine blades will carry the electric potential of water which has largely been at the same atmospheric level as that of the wind turbine, it will not be practically measurable. I.e. no or very low risk of detecting dew, 5 condensation or high moisture levels as precipitation. e The severity and ramp up speed of precipitation may be indicated as a relative measure based on the rise time and absolute level of the charge flow, to enable a higher efficiency level in the resulting regulation of the wind turbine blade rotational speeds.

As such, it presents an elegant method of realizing the technical problem to be solved.

Construction example

In terms of hardware, an applicable construction example may be found in PÅ 2019 00839. The new methodology may be implemented based on the description in the present patent.

Description of figures

Figure 1 shows real world precipitation as measured by a wind turbine rain gauge, centered in the figure, presented with a time synchronized overlay of the charge flow measurement of the wind turbine blades (primarily vertical lines, during periods with precipitation). The figure demonstrate the significant difference between periods with and without precipitation, in terms of the detected charge flow level.

Figure 2 illustrate a blade having receptors (1) at locations on the surface of the blade placed from the tip and down the length of the blade towards the root. The receptors are connected via a down- conductor (2). At the root end of the blade, the down-conductor is attached to the hub for further connection towards the tower and down to earth (or sea). The charge measurement system per blade is positioned at the root end of the blade as illustrated by (3).

Claims

Patent claims

Claim I

A method for measuring precipitation that falls on wind turbines, characterized by the measurement and analysis of

3 electric charge flow through down conductors (figure 2, 2), in or on the blades of the windmill,

Requirement 2

The method of claim 1, wherein said analysis includes correlation of historical time series of charge flow measurements to adaptively determine threshold values of charge flow used to indicate the presence or severity of precipitation,

Requirement 3

The method compares to any one of the preceding requirements, in which the presence or severity of precipitation is used to adjust the rotation speed of the wind turbine blades.

Requirement 4

The method according to any one of the previous claims, in which said down conductors are realized with an arbitrary combination of surface conductors or the use of internal conductors, if conductors in this context also include both cables (Figure 2, 2), receptors (Figure 2, 1) , surfaces, and materials that are generally considered to be poor electrical conductors,

Claim 5

The method according to any one of the preceding claims, in which said method is used in parallel with other detection or measurement purposes, such as detection of the functional state of the nedider system itself.