Icing Sensor for a Scale Model of a Wind Turbine
Based on fibre-optic icing sensor theory and weak-signal detection theory, the model is designed in detail across four areas: the model blade detector, the signal-processing circuit, the host monitoring software and the communication protocol.
Product Features
- The icing sensor for a scale model of a wind turbine is built to the same scale as a real turbine blade, faithfully reproducing blade icing. It monitors blade aerodynamic profile changes and ice accretion in real time and evaluates the effect on operating efficiency.
Application Scenarios
- Mainly used for blade ice monitoring on wind turbines, in conjunction with wind farm de-icing systems to ensure safe turbine operation.
Technical parameters
| Detection parameters | Icing, ice thickness, wind speed, wind resistance, temperature |
|---|---|
| Measuring Range | Ice 0-0.3mm, 0.3-50mm |
| Resolution | 0.2mm |
| Detection Accuracy | 0.5mm |
| Supply voltage | 3.3V、5V、12V、24V DC |
| Power | <0.4W |
| Sensitivity | 0.4℃ |
| Communication Mode | 4-20mA、RS485 MODBUS |
| Temperature Output | -40℃~85℃ |
| Temperature Compensation Range | -40℃~85℃ |
| Probe Material | SUS304 / PTFE (Teflon) |
| Ambient temperature | -40℃~85℃ |
| Dimensions | Proportionally scaled model |
| Protection Rating | IP67 |
Product Manual
Catalog

Construction site photo
1. Overview
Wind turbine blade icing changes the aerodynamic profile of the blades and reduces the operating efficiency of the wind turbine. If blades are iced, their external shape and aerodynamic performance change, blade drag increases, lift decreases, and the controllability and stability of the whole machine are affected; ultimately the wind-energy conversion efficiency drops, and in severe cases blades can be destroyed and operating accidents occur. Wind turbine blade icing has become an urgent problem to solve for ensuring the safe operation of wind turbines. Based on our original blade icing sensor technology and combined with practical domestic wind-power research, our company developed the wind-generator scale-model icing sensor. The model blades rotate at the same frequency as the wind-generator blades, and the scale model can measure icing value, wind speed, rotational speed, wind resistance, and temperature.
First, the rapid icing-calculation theory, the fiber-optic icing sensor theory, and the weak-signal detection theory are introduced; on that basis, the model is designed in detail in four major aspects: the model blade detector, the signal processing circuit, the host-computer monitoring software, and the communication protocol. The model blade detector and signal-processing circuit realize the conversion from weak icing-state changes at the sensor end to significant voltage-signal changes at the acquisition end; the host monitoring software mainly completes the serial-communication module, the graphic data display module, and the data storage module, processes the icing voltage signals sent from the hardware model, provides a good human-machine interface showing the current icing alarm state, saves real-time data to folders, and completes the analysis of icing signal processing and rainwater interference.
Wind power generation converts wind energy into electricity: wind drives the blades to rotate, and a speed-up gearbox raises the rotational speed to drive the generator. With current windmill technology, power generation can start at a wind speed of about three meters per second (a breeze). It should be noted that icing of wind turbine blades caused by specific climatic conditions poses considerable harm to the safe and efficient operation of wind turbines. Once the blade surface ices, the aerodynamic profile is damaged and the wind-energy conversion efficiency drops sharply; as icing accumulates, the blades are more likely to be destroyed, causing operating accidents. Research and application in the field of wind turbine blade icing-state detection are still at an early stage. By applying existing icing-state analysis methods and the new fiber-optic icing sensor, wind turbine blade icing-state detection can be realized, the blade icing state obtained promptly and accurately, enabling icing early warning and icing detection and eliminating the harm of icing to wind turbines — which is of great significance for the wind power industry.
As a clean renewable energy source, wind energy receives growing attention from countries worldwide. China has abundant wind energy reserves and wide distribution, providing a good foundation for developing the wind power industry. Windmill blades are a key component of wind turbine power generation; icing on the blade surface causes significant changes in mass distribution and airfoil, affects power generation performance, and in severe cases leads to faults, accidents, and even destruction.
At present in China, research and application of icing sensors are still at an early stage, especially in the field of wind turbine blade icing-state detection. Wind turbine blade icing-state detection devices designed domestically mainly use the magnetostrictive principle and can only infer blade icing from the sensor's icing degree; detection effectiveness is greatly affected by electromagnetic interference.
Compared with China, foreign research on wind turbine icing detection started earlier and receives greater attention; research on wind turbine icing models is more thorough, and wind turbine icing detection technology leads the domestic level. The relatively mature icing detection methods mainly include the electrical method, the mechanical method, the optical method, and the waveguide method.
II. Parameters
Detection parameters: icing, icing thickness, wind speed, wind resistance, temperature
Range: ice: 0-0.3, 0.3-50mm
Resolution: 0.2mm
Detection accuracy: 0.5mm
Supply voltage: 3.3V, 5V, 12V, 24VD
Power: <0.4W
Sensitivity: 0.4℃
Communication: 4-20mA, RS485 MODBUS
Temperature output: -40℃~85℃
Temperature compensation range: -40℃~85℃
Probe material: SUS304 / PTFE (Teflon)
Ambient temperature: -40℃~85℃
Dimensions: scale model
Protection rating: IP67
III. Working principle
(1) Optical method
This class of icing detection technology uses the reflection, absorption, and scattering of light by icing to build icing sensors; the fiber-optic icing sensor is the most typical optical sensor. A fiber-optic sensor injects light of a certain wavelength into the ice layer through the transmitting fiber; after reflection and scattering by the ice layer, part of the light returns to the receiving fiber, and the icing thickness can be estimated by detecting the light intensity received.
(2) Waveguide method
This class of icing detection technology uses the absorption and reflection of sound waves or other waves by icing to build icing sensors, including surface acoustic wave type, ultrasonic pulse-echo type, and microwave resonance type.
With the rapid global growth of installed wind power capacity, the safe operation of wind turbines has received much wider attention. However, as the key technology guaranteeing safe turbine operation, wind turbine icing detection technology remains relatively underdeveloped in China. Therefore, whether we can start from fundamental theory and technical-method research and propose research and development that combines rich and profound theoretical content with practical engineering application models is an urgent need for safeguarding China's wind power. The engineering application prospects of this project are clear and broad, and its academic significance and social application value are both substantial.
On the basis of wind turbine icing detection theory, a multi-sensor model dominated by fiber-optic icing sensors is used to detect icing light-intensity signals, which are conditioned by the electronic model; the data acquisition module completes communication between the lower computer and the host computer, finally displaying the icing alarm information on the host interface in real time and controlling the icing alarm output through the relay output module. Finally, laboratory icing and rainwater interference tests yielded the setting basis for the icing alarm decision and the rainwater interference decision.
Research on the key theories of wind turbine icing detection, including the rapid icing-calculation theory, the fiber-optic icing sensor theory, and the weak-signal detection theory.
The wind-generator scale-model icing sensor applies icing-state analysis methods and fiber-optic icing sensors to realize blade icing-state detection, obtaining the blade icing state promptly and accurately for icing early warning and icing detection. The model mainly involves the following three theories: the rapid icing-calculation theory, the fiber-optic icing sensor theory, and the weak-signal detection theory.
1. Rapid icing-calculation theory
To establish the link between model blade detector icing and wind turbine icing, rapid calculation of both model blade detector icing and wind turbine icing is required; we use the following formula to calculate icing thickness:
2. Wall boundary condition
For water-drop phase calculation, unlike the airflow field, a wall-suction boundary condition should be adopted when computing the wall surface — if a water droplet collides with the surface, it is considered to flow out from the collision point.
3. Calculation of water-drop collection rate
The model consists of four major parts: the model blade detector (including the fiber-optic sensor), the photoelectric transceiver module, the signal processing and communication module, and the host computer. The three parts other than the host computer constitute the model hardware.
IV. Signal Processing Circuit Design
The wind-generator scale-model icing sensor comprehensively uses multiple fiber-optic icing sensors of different types to detect blade icing, display the current icing state in real time, and issue alarm signals. The hardware acquires and conditions the front-end sensor data and uploads it to the host computer for further processing. Since the turbine icing detection model safeguards the safe operation of wind turbines, the primary design concerns are stability and accuracy: the model must run stably without failures that interrupt operation, and must have high accuracy so that the alarm starts when the icing thickness at the detection point reaches 2mm; given the complex electromagnetic environment on site at wind farms, the model needs strong anti-interference capability.
The electronic model mainly includes 5 icing signal detection circuits and 1 power conversion circuit. Temperature detection, data acquisition, and 485-bus transmission are handled by the data module. Each icing signal detection circuit comprises a light-source emission circuit, a differential amplifier, a synchronous integrator, a correlator, and a low-pass filter. The square-wave signal generated by the square-wave generator in the light-source emission circuit serves on one hand as the switching signal of the switch circuit — amplified by the transistor to drive the infrared LED — and on the other hand as the reference signal of the synchronous integrator and the correlator; the differential amplifier takes the subtraction of the two received model blade detector signals as its input, effectively reducing background-light influence; the synchronous integrator acts in the signal channel as an automatically frequency-tracking filter, removing part of the noise in the differential amplifier output; the correlator then applies cross-correlation, computing the cross-correlation between the periodic signal under test and a reference signal of the same frequency, so that the information carried by the periodic signal buried in noise can be detected. The temperature model blade detector plays an auxiliary role in icing measurement: on one hand, under certain conditions, ice type is strongly related to temperature; on the other hand, the detection circuit only works and uploads data when temperature conditions are satisfied, and remains idle otherwise.
The development of the wind-generator scale-model icing sensor includes model blade detector (sensor) design, hardware design, and software design. In each part, the model's accuracy, reliability, and real-time performance were comprehensively considered.
V.Icing sensor series selection requirements;
|
No. |
Model |
Specification |
No. |
|
|
|
|
|
|
Item |
|||||
|
1. |
JS-FL |
|
|
||||||||||||
|
2 |
Instrument type |
Ⅰ |
|
|
|
|
|
|
|
Remote transmission |
|||||
|
Ⅱ |
|
|
|
|
|
|
|
Wireless remote transmission |
|||||||
|
Ⅲ |
|
|
|
|
|
|
|
With microwave de-icing |
|||||||
|
3 |
Ambient temperature |
Ⅰ |
-40-40℃
|
||||||||||||
|
4 |
Remote transmission signal |
|
|
A |
|
|
|
|
|
Output 4~20mA DC, two-wire |
|||||
|
|
|
B |
|
|
|
|
|
Output 0~10mA DC, three-wire |
|||||||
|
|
|
C |
|
|
|
|
|
Output 0~4Ω |
|||||||
|
|
|
D |
|
|
|
|
|
HART protocol |
|||||||
|
|
|
E |
|
|
|
|
|
485 communication |
|||||||
|
5 |
Material |
|
|
|
Ⅰ |
|
|
|
|
Carbon steel |
|||||
|
|
|
|
Ⅱ |
|
|
|
|
Stainless steel |
|||||||
|
|
|
|
F |
|
|
|
|
PTFE (Teflon) |
|||||||
|
6 |
Measuring range |
|
|
|
|
L2 |
|
|
|
Range: 0~0.3mm |
|||||
|
|
|
|
|
L2.5 |
|
|
|
Range: 0.3~50mm |
|||||||
|
|
|
|
|
L5 |
|
|
|
Range: custom size |
|||||||
|
7 |
Thread standard |
|
|
|
|
|
DN |
|
|
DN15mm |
|||||
|
|
|
|
|
|
DN |
|
|
DN20~32mm |
|||||||
|
8 |
Medium temperature |
|
|
|
|
|
|
T1 |
|
Normal: -5℃~40℃ |
|||||
|
|
|
|
|
|
|
T2 |
|
High: above 40℃ |
|||||||
|
|
|
|
|
|
|
T3 |
|
Low: below -5℃ |
|||||||
|
9 |
Other |
|
|
|
|
|
|
|
NV |
Supplied with WT-100 all-function remote-control monitoring console |
|||||
VI. Runxinsen Electronics extendable technologies:
1. Connect to the upper-level PLC control model to realize automatic control (Mitsubishi PLC, Siemens PLC, and Schneider PLC recommended).
2. Bluetooth phone display: level height shown on the phone (Bluetooth range 10 m).
3. Supports GPRS remote transmission and GSM remote communication.
4. Supports RS232 and RS485 communication (MODBUS protocol, etc.).
5. Supports DCS configuration display.
VII. Detection performance
The parameters mainly measuring the detection performance of an icing sensor include resolution, sensitivity, temperature coefficient, accuracy, and precision.
Resolution is the minimum icing thickness an icing sensor can perceive.
Sensitivity is the ratio of the change in icing thickness to the change in the icing sensor's output.
The temperature coefficient is the ratio of the change in the icing sensor's output to the change in temperature when there is no icing signal.
Accuracy is the closeness of the center point of a series of data obtained by detecting the same icing thickness with the icing sensor to the actual icing thickness.
Precision is the degree of dispersion of the above series of data points relative to their center point.
VIII. Installation method
Installed on the wind generator platform, the model is fixed in place; when the wind generator model ices up, the model blades can ice up synchronously.
IX. Maintenance and use
The probe should not be covered or blocked by other objects; little maintenance is required — simply check periodically whether any object blocks the probe or whether dust has accumulated excessively and affects accuracy.
X. Ordering information:
1. Model and specification
2. Measurement site
3. Working pressure
4. Working temperature
5. Material requirements
6. Measurement accuracy
7. Output signal requirements
Document Download
风力发电机同比例模型结冰传感器(风力同比例)
Click to download document
Need more information?
For any product inquiries or technical questions, please feel free to contact us at any time.
