An object surface icing detection method based on a flexible thermal sensor

Abstract

The present invention relates to a method for detecting icing on the surface of an object based on a flexible thermal sensor. The core idea is to use the dynamic characteristics of the output signal of the flexible thermal sensor installed flush on the surface of the object to judge the icing condition on the surface of the object by means of the wavelet transform coefficient distribution W<supgt;(n)< / supgt;E(b) of the output signal in a multi-scale and normalized manner. The present invention is applicable to the qualitative detection of icing on the aerodynamic shapes of various complex curved surfaces. At the same time, the discrimination method of whether there is icing or not is very simple, and it is especially suitable for combining with the de-icing unit in the anti-icing system to form a feedback control module, having good prospects for engineering applications.

Description

Technical Field

[0001] The present invention belongs to the field of icing detection and ice prevention and removal, and particularly relates to a method for detecting icing on the surface of an object based on a flexible thermal sensor. Background Art

[0002] Aircraft icing can seriously affect the comprehensive performance of aircraft flight: icing on the fuselage surface often leads to adverse effects such as a decrease in the lift coefficient of the wing surface, early separation, and an increase in flight resistance. Icing in the engine inlet will affect the aerodynamic performance of its inner surface, resulting in airflow disorder and even local separation, and even causing engine blade vibration. If the ice in the inlet breaks off, it may even cause an impact with the engine blade, seriously threatening the safety of the engine and posing a flight safety hazard. Therefore, aircraft icing detection and subsequent ice prevention and removal research are of great significance for improving flight safety.

[0003] For wind power generation, wind turbines in plateau and alpine regions are prone to icing under icing weather conditions, resulting in a change in the aerodynamic shape of the blades, which in turn has an adverse effect on the drag characteristics of the blades, reducing the wind energy conversion efficiency and even damaging the blades, causing economic losses. Therefore, icing detection on wind turbine blades has practical engineering significance for the efficient and energy-saving operation of wind power generation.

[0004] Currently, the main international icing detection sensors mainly include: optical, mechanical, electrical, thermal, etc. The optical detection method determines icing by checking the light resistance or light reflection signal; taking the mechanical method as an example of resonance detection, icing changes the mass or stiffness of the sensitive element, thereby affecting its resonance frequency to achieve detection; the electrical detection methods commonly include piezoelectric and capacitive, which respectively generate electrical signal changes through the force of icing on the sensitive element or the change of the dielectric constant, and then achieve icing detection. However, for these traditional icing detection methods, the overall volume of the detection sensor is relatively large. For applications at flat and thin configurations such as airfoils and the leading edge surface of fan blades: on the one hand, it is difficult to install conformally, and it is easy to cause interference to the flow field to form local turbulence, resulting in inconsistent icing conditions between the surface of the icing sensor and the measured object surface, affecting icing detection; on the other hand, considering the economy of engineering applications, these icing detection means are expensive and are generally only equipped on large aircraft, and have not been widely used on small aircraft, fan blades and other small facilities.

[0005] A method for ice detection based on a flexible thermal sensor proposed by the present invention belongs to thermal measurement. Different from the temperature difference type and thermal resistance type detection means in traditional thermal ice detection, the detection means described in the present invention realizes ice detection by judging the forced convection heat transfer situation of the thermal sensitive unit of the flexible thermal sensor. Although it cannot realize the quantitative measurement of the ice amount, it has the advantages of simple structure, low cost, high sensitivity, high time / space resolution, convenient conformal installation on complex curved surfaces, and qualitative detection of the three-phase medium change during the icing process, meeting the application requirements of ice detection problems in aircraft and wind turbine blade anti-icing projects. Summary of the Invention

[0006] The present invention aims to propose a method for ice detection, specifically an ice detection technology based on a flexible thermal sensor, which can judge the actual ice formation situation on the surface of the object to be measured according to the dynamic characteristics of the output signal of the flexible thermal sensor flush-mounted on the surface of the object to be measured.

[0007] The key feature of the ice detection method described in the present invention is that the main measurement unit is a flexible thermal sensor, and the flexible thermal sensor includes: a thermal sensitive unit 1; a lead unit 2; a PI flexible substrate 3.

[0008] The flexible thermal sensor described in the present invention is characterized in that it is prepared by using microelectromechanical system (MEMS) processing technology, and has the advantages of high spatial resolution measurement with small size and small flow field interference; when driven in a constant temperature mode, the sensor frequency response can reach dozens of kilohertz, and has the advantage of high time resolution measurement.

[0009] The flexible thermal sensor described in the present invention is characterized in that if the fluid medium on the surface of the thermal sensitive unit 1 is pure gas or pure liquid, the working principle of the sensor is forced convection heat transfer, and the working process satisfies the working equation:

[0010]

[0011] Where, E w is the output voltage of the flexible thermal sensor; R w is the working resistance of the thermal sensitive unit 1 of the sensor; λ s is the thermal conductivity of the PI flexible substrate 3 of the sensor; l h is the flow direction working length of the thermal sensitive unit 1 of the sensor; w h is the spanwise width of the thermal sensitive unit 1 of the sensor; A is a constant related to the boundary layer velocity profile and the temperature boundary layer established between the thermal sensitive unit 1 and the surrounding environment; ΔT h-s and ΔT h-f are the temperature differences between the thermal sensitive unit 1 of the sensor and the PI flexible substrate 3 and the flow field respectively; ρ is the density of the fluid medium; c pis the specific heat capacity at constant pressure of the fluid medium; λ f is the thermal conductivity of the fluid medium; μ is the dynamic viscosity of the fluid medium; τ is the fluid wall shear stress at the sensor measurement point.

[0012] For the flexible thermal sensor described in the present invention, it is characterized in that if icing occurs at the measured surface position, the thermal sensing unit 1 of the sensor cannot perform forced convection heat transfer with the flow field. At this time, the working equation of the thermal sensing unit 1 of the flexible thermal sensor should satisfy:

[0013]

[0014] where λ c is the thermal conductivity of ice, and ΔT h-c is the temperature difference between the thermal sensing unit 1 of the sensor and the ice.

[0015] Due to the huge differences in the thermal conductivities between cold air, supercooled water, ice, and the sensor PI flexible substrate 3, and the significant difference in viscosity between cold air and supercooled water, therefore, the gas-liquid phase transition and icing process of the fluid medium flowing through the thermal sensing unit 1 will all cause a series of oscillating and mutant characteristic signals in the output signal of the thermal sensor due to the change in thermal conductivity.

[0016] Furthermore, due to the relationship of λ ice >> λ scw >> λ s >> λ air among cold air, supercooled water, ice, and the sensor PI flexible substrate 3, and the relationship of μ scw >> μ air between cold air and supercooled water, therefore, there is a relationship between the output voltage signals of the thermal sensing unit 1 of the sensor in each state:

[0017] E scw > E air > E ice > E0

[0018] where λ scw , μ scw , E scw are respectively the thermal conductivity, dynamic viscosity of supercooled water, and the output voltage when supercooled water flows through the thermal sensing unit 1; λ air , μ air , E air are respectively the thermal conductivity, dynamic viscosity of cold air, and the output voltage when cold air flows through the thermal sensing unit 1; λ ice , E iceThey are the thermal conductivity of ice and the output voltage when the measured surface and the thermal-sensitive unit 1 are frozen respectively; E0 is the reference voltage of the flexible thermal-sensitive sensor without flow and without freezing under the tested working conditions.

[0019] Combined with the above analysis, the present invention proposes a method for detecting icing on the surface of an object based on a thermal-sensitive sensor, and the specific steps are as follows:

[0020] Step 1: Install the thermal-sensitive sensor flush with the surface of the object to be measured;

[0021] Step 2: Use a constant-temperature drive circuit to drive the thermal-sensitive sensor so that the working temperature T of the thermal-sensitive unit 1 of the sensor h is lower than 0°C and higher than the supercooled water temperature T in the flow field scw , that is, 0 > T h > T scw , to ensure that the thermal-sensitive unit 1 has no influence on the icing of the object surface;

[0022] Step 3: Collect and analyze the output voltage of the thermal-sensitive sensor. The specific analysis method is as follows: When the pulsation intensity of the output signal of the flexible thermal-sensitive sensor is, it indicates that a complex two-phase flow process is occurring on the thermal-sensitive unit 1 (the measured object surface), and the output signal shows a violent fluctuation state. Among them, E r ′ ms is the root mean square value of the pulsation component of the output voltage signal, is the average value of the output voltage signal, and a is the judgment threshold; When a significant negative step characteristic signal appears in the output signal of the sensor, it indicates that icing has started to occur on the thermal-sensitive unit 1 (the measured object surface). The discrimination method is: Through wavelet analysis based on the output signal of the flexible thermal-sensitive sensor, and define the wavelet transform coefficient distribution W (n) E(b) of different scales and normalization to detect the mutation point of the output signal and realize the judgment of the icing start time point.

[0023] Among them, the wavelet transform coefficient distribution W (n) E(b) of different scales and normalization described in the present invention is defined by the following calculation formula:

[0024]

[0025] Among them, C ψ is the wavelet admissibility condition; E(t) is the output time series signal of the sensor; ψ b (n) (t) is the wavelet function at the nth scale; b is the wavelet translation factor.

[0026] Preferably, the Haar wavelet can be used as the mother function to analyze the output signal of the thermal sensor. Under this wavelet mother function, the defined wavelet transform coefficient distribution W of different scales and normalized (n) The maximum point of W (n) E(b) corresponds to the negative step mutation turning point of the sensor output signal; the minimum point of W (n) E(b) corresponds to the positive step mutation turning point of the sensor output signal. Therefore, the moment corresponding to the maximum point of the Haar wavelet normalized wavelet transform coefficient distribution W

[0027] E(b) of the output signal is the starting time point of icing on the thermal element 1 (object surface). (n) Preferably, to accurately determine the position of the mutation point of the sensor output signal, the normalized wavelet transform coefficient distributions W

[0028] Advantageous effects:

[0029] The icing detection method for the object surface described in the present invention combines the advantages of high sensitivity and high spatio-temporal resolution output of the flexible thermal sensor based on MEMS technology, and is applicable to the qualitative detection of icing on various complex curved aerodynamic shapes. At the same time, the discrimination method of whether icing occurs is very simple, which can not only be used for anti-icing prediction, but also be particularly suitable for forming a feedback control module in combination with the de-icing unit in the anti-icing and de-icing system, and has good engineering application prospects. Description of the drawings

[0030] Figure 1 is the structural schematic diagram of the flexible thermal sensor;

[0031] Figure 2 is the cross-sectional view of the A-A section of the flexible thermal sensor;

[0032] Figure 3 is the output voltage signal of the flexible thermal sensor during a certain icing test and its wavelet transform coefficient distribution W (n) E(b) of different scales and normalized;

[0033] Figure 4 is the output voltage signal of the flexible thermal sensor during the time period a-b;

[0034] Figure 5 is the distribution histogram of the output voltage signal of the flexible thermal sensor during the time period a-b;

[0035] Figure 6 is the output voltage signal of the flexible thermal sensor during the time period b-c;

[0036] Figure 7 It is a distribution histogram of the output voltage signal of the flexible thermal sensor during time period b-c;

[0037] Figure 8 It is the output voltage signal of the flexible thermal sensor during time period d-e.

[0038] Marking description:

[0039] Thermal unit 1 of the flexible thermal sensor

[0040] Lead unit 2 of the flexible thermal sensor

[0041] PI flexible substrate 3 of the flexible thermal sensor Specific implementation manner

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0043] An embodiment of the present invention application performs icing detection through a flexible thermal sensor flush-mounted with the surface to be tested. The flexible thermal sensor mainly includes: thermal unit 1; lead unit 2; PI flexible substrate 3.

[0044] The specific implementation steps of this embodiment are as follows:

[0045] Step 1: Flush-mount the thermal sensor with the surface of the object to be measured;

[0046] Step 2: Drive the thermal sensor using a constant temperature drive circuit so that the operating temperature T of the sensor thermal unit 1 h is lower than 0°C and higher than the supercooled water temperature T in the flow field scw , that is, 0 > T h > T scw , to ensure that the thermal unit 1 has no influence on the icing of the object surface;

[0047] Step 3: Collect the output voltage of the thermal sensor;

[0048] Step 4: Analyze the collected voltage signal. In this embodiment, the signal pulsation intensity threshold is set to a = 0.2, and the specific analysis method is as follows:

[0049] The output voltage signal measured by the flexible thermal sensor, as shown in Figure 3 , according to the output signal characteristics of the sensor, it can be judged that during the a-b time period, the output voltage of the sensor (Figure 4 Pulsation intensity of ( The signal shows strong oscillations. The distribution histogram of the output signal ( Figure 5 ) shows a positive skewness distribution, indicating that there are high-frequency and strong high heat transfer rate events occurring on the sensor surface during this period. It corresponds to the cold air containing supercooled water droplets passing through the thermosensitive unit 1 (the surface of the object to be measured). At this time, the fluid medium flowing through the thermosensitive unit 1 is mainly cold air; during the b-c period, the output voltage of the sensor ( Figure 6 ) Pulsation intensity of ( Also shows strong oscillations. The distribution histogram of the output signal ( Figure 7 ) shows a negative skewness distribution, that is, during this period, on the basis of the heat transfer rate event on the sensor surface, local strong low heat transfer rate events occur, indicating that the thermosensitive unit 1 (the surface of the object to be measured) is already in a wet state and a water film has formed on the local surface; during the c-d period, a significant negative step characteristic signal appears in the output signal of the sensor, and through the mutation point detection based on wavelet transform of the present invention, it is shown that icing begins to occur on the thermosensitive unit 1 (the surface of the object to be measured); during the d-e period, the output signal time series of the sensor is stable and no pulsation characteristic signal appears It shows that the thermosensitive unit 1 (the surface of the object to be measured) has been covered by an ice layer. However, due to the local heat release during the development of icing, the constant temperature bridge gradually reduces the driving current to maintain the working temperature stability of the thermosensitive unit 1, resulting in a certain downward trend in the output voltage of the sensor.

[0050] The multi-scale and normalized wavelet transform coefficient distribution W (n) E(b) defined by the present invention is obtained through the 11-scale dyadic wavelet transform with the Haar wavelet function as the mother function. The calculation formula for the multi-scale and normalized wavelet transform coefficient is:

[0051]

[0052] Among them, C ψ Is the wavelet admissibility condition; E(t) is the output time series signal of the sensor; ψ b (n) (t) is the wavelet function at the nth scale; b is the wavelet translation factor.

[0053] The multi-scale and normalized wavelet transform coefficient distribution W (n) E(b) of the Haar wavelet mother function in this embodiment has the characteristics: The maximum point of W (n) E(b) corresponds to the negative step mutation turning point of the sensor output signal; The minimum point of W (n) E(b) corresponds to the positive step mutation turning point of the sensor output signal; Therefore, the multi-scale and normalized Haar wavelet transform coefficient distribution W (n)The time corresponding to the maximum point of E(b) is the starting time point when the thermosensitive unit 1 (the object surface) freezes.

[0054] In this embodiment, to accurately determine the position of the mutation point of the sensor output signal, the distribution W of the normalized wavelet transform coefficients at multiple scales is combined. (n) A comprehensive observation and judgment are made on E(b) to reduce the detection influence caused by output signal frequency aliasing or noise interference, such as Figure 3 shown. According to Figure 3 it can be seen that the time corresponding to the maximum point of W (n) and E(b) at multiple scales is determined as the starting time point of icing.

[0055] The above description is only a preferred embodiment of the present invention and does not limit the present invention. Any modifications, substitutions, and improvements made within the ideas and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for detecting icing on the surface of an object based on a thermal sensor, characterized in that, The specific steps are as follows: Step 1: Install the thermal sensor flush with the surface of the object to be measured; Step 2: Drive the thermosensitive sensor using a constant-temperature drive circuit so that the working temperature T of the thermosensitive unit of the sensor h is lower than 0°C and higher than the supercooled water temperature T in the flow field scw , that is, 0 > T h > T scw , to ensure that the thermosensitive unit has no influence on the icing of the object surface; Step 3: Collect and analyze the output voltage of the thermal sensor. The specific analysis method is as follows: When the pulsation intensity of the output signal of the flexible thermal sensor it indicates that a complex two-phase flow process is occurring on the thermal sensing unit, and the output signal exhibits a violent fluctuation state. Among them, E′ rms is the root mean square value of the pulsation component of the output voltage signal, is the average value of the output voltage signal, and a is the judgment threshold; When the output signal of the sensor generates a significant negative step characteristic signal, it indicates that the ice formation phenomenon begins on the thermosensitive unit, i.e., the surface of the object under test. The discrimination method is as follows: through wavelet analysis based on the output signal of the flexible thermosensitive sensor, and defining the wavelet transform coefficient distribution W of different scales and normalization (n) E(b) is used to detect the mutation point of the output signal to realize the judgment of the ice formation start time point; The distribution W of the multi-scale and normalized wavelet transform coefficients (n) The calculation formula of E(b) is defined as: Among them, C ψ is the wavelet admissibility condition; E(t) is the sensor output timing signal; ψ b (n) (t) is the wavelet function at the n-th scale; b is the wavelet translation factor.

2. A method for detecting icing on the surface of an object based on a thermal sensor according to claim 1, characterized in that, The wavelet analysis based on the output signal of the flexible thermal sensor specifically uses the Haar wavelet as the mother function to analyze the output signal of the thermal sensor, and the distribution W of the Haar wavelet normalized wavelet transform coefficients of the output signal (n) The time corresponding to the maximum point of E(b) is the starting time point when the thermal sensing unit freezes.

Images