Icing detection device, method, computer equipment, storage medium and program product

Abstract

The present application relates to the technical field of icing detection, and discloses an icing detection device, method, computer equipment, storage medium and program product, the icing detection device comprises a complex impedance measurement unit and a data processing unit, the complex impedance measurement unit is connected with the data processing unit, wherein the complex impedance measurement unit comprises a plurality of electrodes with preset shapes, the electrodes are arranged according to a preset rule, and at least two electrodes are selected as a group of electrode pairs. The icing detection method comprises: collecting complex impedance data of a plurality of frequency bands at a to-be-detected position by using a plurality of electrode pairs, and performing feature processing on the complex impedance data of each frequency band to obtain complex impedance feature values; collecting temperature information at the to-be-detected position by using a temperature measurement unit; temperature compensation is performed on the complex impedance feature values by using the temperature information, and the icing condition of the to-be-detected position is detected to obtain icing detection information. The present application can solve the problems of poor applicability and low accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of icing detection, and in particular to an icing detection device, method, computer device, storage medium and program product. BACKGROUND

[0002] Icing is a common natural phenomenon, and icing can cause serious harm to normal production and life. Icing causes wind turbine generators to be unable to generate electricity normally, and icing of power transmission and transformation equipment can cause the power grid to be unable to operate safely and stably, and icing of airfoils can cause airplanes to be unable to fly safely, etc.

[0003] Current icing detection methods mainly include mechanical method / weight method, image method, optical fiber sensing method, etc. The mechanical method / weight method calculates the icing thickness by measuring the strain or weight change of the measured object after icing. The image method obtains icing images through a camera or a drone and calculates the icing thickness by using image recognition technology. The optical fiber sensing method calculates the icing thickness by obtaining the stress change of the measured object through a grating sensor or a distributed Brillouin scattering optical fiber sensor.

[0004] For some scenarios such as wind turbine blades and airplane wings, the appearance contour surface has high requirements. Currently, the icing monitoring sensor has poor applicability and low accuracy due to difficulty in arrangement. SUMMARY

[0005] Therefore, the present application provides an icing detection device, method and related device to solve the problem of poor applicability and low accuracy.

[0006] In a first aspect, the present application provides an icing detection device, which comprises a complex impedance measurement unit and a data processing unit. The complex impedance measurement unit is connected with the data processing unit. The complex impedance measurement unit comprises a plurality of electrodes with preset shapes. The electrodes are arranged according to a preset rule. At least two electrodes are selected as a group of electrode pairs. The complex impedance measurement unit is used to collect complex impedance data of a plurality of frequency bands at a to-be-detected position by using a plurality of groups of electrode pairs. The data processing unit is used to process the complex impedance data and determine icing detection information at the to-be-detected position based on the complex impedance data.

[0007] In this implementation, the icing detection device of the present application has a complex impedance measurement unit with a plurality of groups of electrode pairs. A plurality of groups of complex impedance data are collected by using a plurality of groups of electrode pairs. The icing detection information is determined based on the data processing unit. Compared with the double-electrode scheme, the present application expands the data source by electrode number and electrode layout design, improves the measurement effect, and improves the accuracy of icing detection.

[0008] In an optional implementation, the icing detection device further comprises a temperature measurement unit, wherein the temperature measurement unit is configured to collect temperature information at the to-be-detected position; and the data processing unit is configured to take the complex impedance data and the temperature information as input data of an icing detection model, process the complex impedance data and the temperature information, and output icing detection information of the to-be-detected position, the icing detection model being a cascaded machine learning model.

[0009] In an optional implementation, the electrodes include excitation electrodes and measurement electrodes, each group of electrode pairs includes at least one excitation electrode and at least one measurement electrode, and the electrode pairs have different characteristic parameters, including an inter-electrode distance, an electrode width, and an electrode area.

[0010] In a second aspect, the present application provides an icing detection method applied to the icing detection device described above, the icing detection method comprising: collecting, by a plurality of groups of electrode pairs, complex impedance data of a plurality of frequency bands at a to-be-detected position, and performing feature processing on the complex impedance data of each frequency band to obtain complex impedance feature values; collecting, by a temperature measurement unit, temperature information at the to-be-detected position; and performing temperature compensation on the complex impedance feature values by using the temperature information, detecting icing conditions of the to-be-detected position, and obtaining icing detection information.

[0011] In this implementation, the icing detection method of the present application, on the basis of the icing detection device, collects a plurality of groups of complex impedance data by using a complex impedance measurement unit of a plurality of groups of electrode pairs, collects temperature information, performs icing detection, and obtains icing detection information. The present application expands the data source by electrode quantity and electrode layout design, performs data compensation on the complex impedance feature values by using the temperature information, improves the measurement effect, and improves the accuracy of icing detection.

[0012] In an optional implementation, the temperature compensation on the complex impedance feature values by using the temperature information, the detection of the icing conditions of the to-be-detected position, and the obtaining of the icing detection information comprise: inputting the complex impedance feature values and the temperature information into an icing detection model, detecting the icing conditions of the to-be-detected position, and outputting the icing detection information; and / or calculating the icing detection information of the to-be-detected position by using the complex impedance feature values and a preset ice thickness fitting curve.

[0013] In an optional implementation, the icing detection model includes an ice type classifier, and the temperature compensation on the complex impedance feature values by using the temperature information, the detection of the icing conditions of the to-be-detected position, and the obtaining of the icing detection information comprise: inputting the complex impedance feature values and the temperature information into the ice type classifier; calculating a probability that the icing type of the to-be-detected position is a preset icing type, and outputting the preset icing type with the largest probability as the icing type of the to-be-detected position; wherein the preset icing type includes clear ice, hoar frost, mixed ice, water, pollutants, and dry.

[0014] In an optional implementation, the icing detection model further comprises an ice thickness regressor, and after outputting the preset icing type with the maximum probability as the icing type of the to-be-detected position, the method further comprises: acquiring the ice thickness regressor corresponding to the icing type; inputting the complex impedance feature value and the temperature information into the ice thickness regressor, fusing ice thickness results of different machine learning algorithms, and calculating the ice thickness value of the to-be-detected position.

[0015] In this implementation, the application uses a cascade machine learning algorithm to judge the icing state, and uses an ice type classifier and an ice thickness regressor to respectively judge the ice type and the ice thickness, so that the icing measurement has robustness, stability and accuracy in data calculation through algorithm design and application.

[0016] In an optional implementation, the feature processing of the complex impedance data of each frequency band to obtain the complex impedance feature value comprises: performing feature processing on the complex impedance data by using an RC parallel model to obtain a first complex impedance feature value, the first complex impedance feature value comprising an equivalent capacitance value and a loss tangent value; and calculating a second complex impedance feature value based on the first complex impedance feature value, the second complex impedance feature value comprising a low-frequency capacitance value, a high-frequency capacitance value, a relaxation time, a high-low frequency capacitance difference value, a high-low frequency capacitance ratio value, a loss peak number, a first loss peak peak value, a second loss peak peak value, and a loss peak peak value ratio.

[0017] In an optional implementation, the complex impedance feature value comprises a capacitance value, and the calculation of the icing detection information of the to-be-detected position by using the complex impedance feature value and a preset capacitance-ice thickness fitting curve comprises: constructing a plurality of capacitance-ice thickness fitting curves, the plurality of capacitance-ice thickness fitting curves being ice thickness detection curves of a plurality of electrode pairs that are pre-established; and determining the ice thickness value of the to-be-detected position by using the equivalent capacitance value and the capacitance-ice thickness fitting curve.

[0018] In an optional implementation, the determination of the ice thickness value of the to-be-detected position by using the equivalent capacitance value and the capacitance-ice thickness fitting curve comprises: sorting the electrode pairs according to the electrode spacing from small to large, and sequentially determining the corresponding thickness information by using the electrode pairs and the capacitance-ice thickness fitting curves of the electrode pairs; determining whether the thickness information is less than a preset threshold value of the electrode pair, and when the thickness information is less than the preset threshold value of the electrode pair, taking the thickness information as the ice thickness value of the to-be-detected position; otherwise, calculating the thickness information by using the next electrode pair until the ice thickness value of the to-be-detected position is obtained.

[0019] In this implementation, the application uses a multi-stage capacitance-ice thickness fitting curve query method to further reduce the complexity of ice thickness measurement while ensuring the signal-to-noise ratio of ice thickness measurement.

[0020] In a third aspect, the present application provides a computer device, comprising a memory and a processor, which are connected with each other in communication, and the memory stores computer instructions, and the processor executes the ice detection method of the first aspect or any of the corresponding embodiments thereof by executing the computer instructions.

[0021] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions for making a computer execute the ice detection method of the first aspect or any of the corresponding embodiments thereof.

[0022] In a fifth aspect, the present application provides a computer program product, which comprises computer instructions for making a computer execute the ice detection method of the first aspect or any of the corresponding embodiments thereof. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0024] Figure 1 is a schematic diagram of an ice detection system according to an embodiment of the present application;

[0025] Figure 2 is a schematic diagram of a combination of electrode pairs of a complex impedance measurement unit according to an embodiment of the present application;

[0026] Figure 3 is a schematic diagram of a specific complex impedance measurement unit according to an embodiment of the present application;

[0027] Figure 4 is a flowchart of an ice detection method according to an embodiment of the present application;

[0028] Figure 5 is a flowchart of another ice detection method according to an embodiment of the present application;

[0029] Figure 6 is a schematic diagram of an ice detection model according to an embodiment of the present application;

[0030] Figure 7 is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0032] The current ice thickness measurement method realizes ice thickness measurement by measuring the change of vibration frequency or impedance of an iced object, and has low ice thickness measurement precision. For a scene with high requirements on the contour surface, the applicability is poor and the accuracy is low. Therefore, according to the embodiments of the present application, an ice detection system is provided, which can improve the accuracy of ice detection. As shown in Figure 1 Figure 1 is a schematic diagram of an ice detection system according to an embodiment of the present application. In actual use, the ice detection system is placed on a device to be detected, as shown in Figure 1 The ice detection system includes a substrate 1, an ice detection device 2, a data transmission component 3 and a power supply component 4.

[0033] The substrate 1 carries the ice detection device 2, the data transmission component 3 and the power supply component 4. The ice detection device 2 is used to detect the icing condition of the to-be-detected position and send the detection data to the data transmission component 3. The data transmission component 3 sends the detection data to a monitoring center to realize ice monitoring and alarm. The power supply component 4 is used to power the ice detection system.

[0034] In an implementation manner, the substrate can be selected as a flexible substrate, so that the device perfectly fits the irregular curved surface shape, and the design has higher fitting degree.

[0035] In an implementation manner, the ice detection device includes a complex impedance measurement unit and a data processing unit, and the complex impedance measurement unit is connected with the data processing unit.

[0036] In an implementation manner, the electrode of the ice detection device is copper skin, which can perfectly fit the irregular curved surface shape, and the design has higher fitting degree.

[0037] The complex impedance measurement unit includes a plurality of electrodes with preset shapes, the electrodes are arranged according to preset rules, at least two electrodes are selected as a group of electrode pairs, and the complex impedance measurement unit is used to collect complex impedance data of a plurality of frequency bands at the to-be-detected position by using a plurality of groups of electrode pairs.

[0038] Specifically, please refer to Figure 2 , Figure 2 is a schematic diagram of a complex impedance measurement unit electrode pair combination according to an embodiment of the present application. As shown in Figure 2 ​As shown, the complex impedance measurement unit distributes multiple electrodes on the substrate, and the multiple electrodes are arranged according to multiple preset rules, and different independent electrodes are combined by using a relay to obtain an electrode pair.

[0039] Exemplarily, E135 is three excitation electrodes, and E24 is two measurement electrodes, or E135 is a measurement electrode, and E24 is an excitation electrode.

[0040] In the arrayed electrodes, when the electrode pairs are combined, a detection penetration depth range coverage principle is followed, that is, in all combinations, the spacing between the excitation electrodes and the measurement electrodes should contain the possible maximum value and the minimum value.

[0041] In an implementation, at least one of the electrode pairs has a different characteristic parameter, which has different response characteristics to the ice layer. The characteristic parameter includes an electrode distance, an electrode width, and an electrode area, etc.

[0042] Specifically, the electrodes include excitation electrodes and measurement electrodes, the length and width of each electrode are different, the distance between adjacent electrodes is different, and each electrode pair includes at least one excitation electrode and at least one measurement electrode. Each electrode pair has a different detection depth, and when the ice thickness changes, the complex impedance data measured by each electrode pair at different frequency points changes accordingly.

[0043] In a specific implementation, the length of each independent electrode in the arrayed electrodes is the same, one excitation electrode is shared on the left side, and the width and spacing of the four measurement electrodes are increased in turn, and the specific parameters can be determined according to simulation optimization.

[0044] Exemplarily, please refer to Figure 3 , Figure 3 is a schematic diagram of a specific complex impedance measurement unit according to an embodiment of the application. As shown in Figure 3 , the lengths of the five independent electrodes are the same, the leftmost electrode is an excitation electrode, and the other four electrodes are measurement electrodes, and the four measurement electrodes share one excitation electrode. Among them, the width and spacing of the four measurement electrodes are increased in turn, and the specific parameters can be determined according to simulation optimization. In this specific implementation, the width ratio of the four measurement electrodes is 1:2:3:4, the spacing ratio between the five electrodes is 2:3:4:5, and the width of the excitation electrode is the average of the widths of all the measurement electrodes.

[0045] The data processing unit is configured to process the complex impedance data and determine ice detection information at the to-be-detected position based on the complex impedance data.

[0046] In an implementation, the ice detection device further includes a temperature measurement unit.

[0047] The temperature measuring unit is configured to collect temperature information at the to-be-detected position.

[0048] The data processing unit is configured to process the complex impedance data and determine icing detection information at the to-be-detected position based on the complex impedance data.

[0049] The icing detection model is a cascaded machine learning model.

[0050] In an implementation, the complex impedance data is processed by using the icing detection model after pre-training.

[0051] Specifically, the data processing unit is configured to process the complex impedance data and the temperature information as input data of the icing detection model, and output the icing detection information at the to-be-detected position. The icing detection model is a cascaded machine learning model.

[0052] In this implementation, the icing detection device has a complex impedance measuring unit with multiple electrode pairs. The complex impedance data of multiple groups is collected by using the multiple electrode pairs, and the icing detection information is determined based on the data processing unit. Compared with the double-electrode scheme, the number of electrodes and the electrode layout design are expanded to expand the data source, improve the measurement effect, and improve the accuracy of icing detection.

[0053] According to the embodiment of the present application, a kind of icing detection method embodiment is provided, it needs to be explained, the steps shown in the flowchart of the drawing can be executed in computer system such as a group of computer executable instructions, and, although logical order is shown in flowchart, in some cases, the steps shown or described can be executed in different order from here.

[0054] In the present embodiment, an icing detection method is provided, which can be used in the icing detection device described above, Figure 4 is a flowchart of a kind of icing detection method according to the embodiment of the present application, it needs to be noted that, if there is substantially the same result, the present embodiment is not limited to Figure 4 The order of the flow shown in the figure. As Figure 4 Shown, the flow includes the following steps:

[0055] Step S401, using multiple electrode pairs respectively in to-be-detected position on the complex impedance data of multiple frequency bands are collected, and the complex impedance data of each frequency band is processed to obtain complex impedance feature value.

[0056] The complex impedance data at the to-be-detected position is collected by using the complex impedance measuring unit on the icing detection device.

[0057] Specifically, one electrode pair of the complex impedance measurement unit is selected to switch a plurality of frequencies, and the complex impedance data of a plurality of frequency bands at the to-be-detected position is collected. The frequency can be a preset frequency. The electrode pair is switched, a plurality of frequencies are switched by using the next electrode pair, and the complex impedance data of a plurality of frequency bands at the to-be-detected position is collected again. In this way, the complex impedance data of a plurality of frequency bands of each electrode pair is obtained.

[0058] Further, the complex impedance data of a plurality of frequency bands of each electrode pair is processed by using data processing methods such as data cleaning and data conversion.

[0059] In an implementation manner, the complex impedance data is converted by using a network model, so as to facilitate subsequent detection accuracy.

[0060] In step S402, the temperature information at the to-be-detected position is collected by using the temperature measurement unit.

[0061] In step S403, the complex impedance characteristic value is temperature-compensated by using the temperature information, the icing condition at the to-be-detected position is detected, and icing detection information is obtained.

[0062] The complex impedance characteristic value is taken as original data measured by each group of electrode pairs, the complex impedance characteristic value measured by each group of electrode pairs is temperature-compensated by using the collected temperature information of the ice layer, in a case where the complex impedance information and the temperature information at the to-be-detected position are determined, the complex impedance information and the temperature information of a plurality of preset icing types are compared, and the icing detection information at the to-be-detected position is determined.

[0063] The icing detection information includes an icing type. The icing type includes clear ice, frost ice, mixed ice, water, pollutants, and dryness.

[0064] The icing detection method provided in this embodiment, on the basis of the icing detection device, collects a plurality of groups of complex impedance data by using a plurality of groups of electrode pairs of the complex impedance measurement unit, collects temperature information, performs icing detection, and obtains icing detection information. The number of electrodes and the electrode layout design are expanded to expand the data source, the temperature information is supplemented to the complex impedance characteristic value, the measurement effect is improved, and the accuracy of the icing detection is improved.

[0065] The icing detection method of the present application can be applied to icing monitoring scenes such as wind turbine generators, photovoltaic panels, aircraft, roads, and bridges, and has a wide application prospect. In this embodiment, an icing detection method is provided, which can be used in the above-mentioned icing detection device, Figure 5 is a flowchart of another icing detection method according to an embodiment of the present application. It should be noted that the flow order shown in this embodiment is not limited if there is substantially the same result. Figure 5 As shown in Figure 5 , the flow includes the following steps:

[0066] Step S501, using multiple sets of electrode pairs to respectively collect complex impedance data of multiple frequency bands at the to-be-detected position, and respectively performing feature processing on the complex impedance data of each frequency band to obtain complex impedance feature values.

[0067] Specifically, the above step S501 includes:

[0068] Step S5011, using multiple sets of electrode pairs to respectively collect complex impedance data of multiple frequency bands at the to-be-detected position.

[0069] The complex impedance data at the to-be-detected position is collected by using a complex impedance measurement unit on the icing detection device.

[0070] Specifically, one electrode pair of the complex impedance measurement unit is selected to switch multiple frequencies, and the complex impedance data of multiple frequency bands at the to-be-detected position is collected. The frequency can be a preset frequency. The electrode pair is switched, and the next electrode pair is used to switch multiple frequencies, and the complex impedance data of multiple frequency bands at the to-be-detected position is collected again. In this way, the complex impedance data of multiple frequency bands of each electrode pair is obtained.

[0071] Step S5012, performing feature processing on the complex impedance data by using an RC parallel model to calculate first complex impedance feature values.

[0072] The first complex impedance feature values include an equivalent capacitance value and a loss tangent value.

[0073] Specifically, the RC parallel model is used to perform feature processing on the complex impedance data to obtain an equivalent capacitance C P and a loss tangent tanδ.

[0074] Step S5013, calculating second complex impedance feature values based on the first complex impedance feature values.

[0075] The second complex impedance feature values include a low-frequency capacitance value, a high-frequency capacitance value, a relaxation time, a high-low frequency capacitance difference value, a high-low frequency capacitance ratio, a loss peak number, a first loss peak peak value, a second loss peak peak value, and a loss peak peak value ratio.

[0076] Specifically, calculating the second complex impedance feature values based on the first complex impedance feature values includes:

[0077] The relaxation time τ is calculated as:

[0078]

[0079] The low-frequency capacitance value C low is calculated as:

[0080] C 1ow =C f , f = 1 kHz.

[0081] wherein the high-frequency capacitance value C high is calculated as:

[0082] C high = C f f = 100 kHz.

[0083] wherein the high-low frequency capacitance difference value ΔC P is calculated as:

[0084] ΔC P = C low -C high .

[0085] wherein the high-low frequency capacitance ratio value R C is calculated as:

[0086]

[0087] wherein the peak-seeking algorithm is used to detect the loss peaks and determine the number N of loss peaks, and the first loss peak value P1 and the second loss peak value P2 are obtained. If the number of loss peaks in the measurement frequency range is 1, then P2 = 0, and if the number of loss peaks in the measurement frequency range is greater than 2, then it is determined that the contaminant interference state exists.

[0088] wherein the loss peak value ratio R P is calculated as:

[0089]

[0090] In step S502, the temperature measurement unit is used to collect temperature information at the to-be-detected position.

[0091] In step S503, the icing condition of the to-be-detected position is detected based on the complex impedance characteristic value and the temperature information, and icing detection information is obtained.

[0092] In one implementation, the complex impedance characteristic value and the temperature information are input into an icing detection model to detect the icing condition of the to-be-detected position, and output icing detection information.

[0093] The icing detection model includes an ice type classifier and an ice thickness regressor. Please refer to Figure 6 , Figure 6 is a schematic diagram of an icing detection model according to an embodiment of the present application. First, the surface state is recognized and the ice type is classified by the temperature and the complex impedance spectrum characteristic value and the ice type classifier, and the icing type is obtained. Then, the ice thickness range or the ice thickness is calculated by the ice type, the temperature, the complex impedance spectrum characteristic value measured by each group of electrodes, and the ice thickness regressor in the icing state.

[0094] In one implementation, the icing detection model is a cascaded machine learning model.

[0095] Specifically, the step S504 includes:

[0096] In step S5041, an ice type classifier is used to obtain the ice type of the to-be-detected position.

[0097] The complex impedance characteristic value and the temperature information are input into the ice type classifier, a probability that the ice type of the to-be-detected position is a preset ice type is calculated, and the preset ice type with the maximum probability is output as the ice type of the to-be-detected position.

[0098] The ice type classifier algorithm for ice type and state recognition can be selected according to the computing resources and index requirements.

[0099] In an implementation manner, to realize lightweight design of the ice thickness measurement model, the ice type classifier is a random forest model, the input feature data is the temperature and the complex impedance spectrum characteristic value measured by each group of electrodes, and the output is different ice types and surface states such as clear ice, frost ice, mixed ice, water, pollutants and dryness.

[0100] In step S5042, an ice thickness regressor is used to obtain the ice thickness value of the to-be-detected position.

[0101] One or more ice thickness regressors based on different machine learning algorithms are configured for each ice type. The ice thickness regressor corresponding to the ice type is obtained, the complex impedance characteristic value and the temperature information are input into the ice thickness regressor, the ice thickness results of different machine learning algorithms are fused, and the ice thickness value of the to-be-detected position is calculated.

[0102] The ice thickness regressor algorithm for ice thickness measurement can be selected according to the computing resources and index requirements.

[0103] In an implementation manner, to realize lightweight design of the ice thickness measurement model, the ice thickness regressor is a decision tree model, the input feature data is the temperature and the complex impedance spectrum characteristic value measured by each group of electrodes, and the output is the calculated ice thickness value range.

[0104] For example, the ice thickness range is divided into: less than 1 mm, 1-5 mm, 5-10 mm, 10-20 mm, and greater than 20 mm.

[0105] In an implementation manner, the ice thickness results of different machine learning algorithms are weighted and averaged to calculate the ice thickness value of the to-be-detected position.

[0106] In another implementation manner, the complex impedance characteristic value and the temperature information are input into an ice detection model, the ice condition of the to-be-detected position is detected, and ice detection information is output; and the complex impedance characteristic value and a preset capacitance ice thickness fitting curve are used to calculate the ice detection information of the to-be-detected position.

[0107] Specifically, the step S504 includes:

[0108] Step S5041, obtaining the ice type of the to-be-detected position by using the ice type classifier.

[0109] The ice detection model is the ice type classifier. The ice type of the to-be-detected position is obtained by using the ice type classifier according to the method of step S5041. Details are not described herein.

[0110] Step S5042,

[0111] The ice thickness value of the to-be-detected position is obtained by using the ice thickness regressor or is calculated by using the complex impedance characteristic value and the preset capacitance ice thickness fitting curve.

[0112] Specifically, step S5042 includes:

[0113] Step a1, constructing a plurality of capacitance ice thickness fitting curves.

[0114] The plurality of capacitance ice thickness fitting curves are ice thickness detection curves of a plurality of electrode pairs established in advance. That is, for each electrode pair, a capacitance ice thickness fitting curve is determined.

[0115] Step a2, determining the ice thickness value of the to-be-detected position by using the equivalent capacitance value and the capacitance ice thickness fitting curve.

[0116] Each electrode pair has a corresponding thickness detection threshold value. The electrode pair suitable for the current ice surface thickness detection is determined by using the method of the present application, and the ice thickness value is obtained based on the corresponding capacitance ice thickness fitting curve.

[0117] The capacitance value at a certain high signal-to-noise ratio frequency measured by the electrode pair with the shortest electrode spacing is preferentially used for thickness calculation. The electrode pairs are sorted according to the electrode spacing from small to large. The corresponding thickness information is determined by using the electrode pairs and the capacitance ice thickness fitting curves of the electrode pairs in turn. It is judged whether the thickness information is less than the preset threshold value of the electrode pair. When the thickness information is less than the preset threshold value of the electrode pair, the thickness information is taken as the ice thickness value of the to-be-detected position. Otherwise, the next electrode pair is used to calculate the thickness information, until the ice thickness value of the to-be-detected position is obtained.

[0118] Specifically, a first equivalent capacitance value corresponding to the complex impedance data collected by a first electrode pair with the minimum electrode spacing is obtained, a first capacitance-ice thickness fitting curve of the first electrode pair is obtained, and a first thickness information corresponding to the first equivalent capacitance value is calculated in the first capacitance-ice thickness fitting curve. When the first thickness information is less than a first preset threshold of the first electrode pair, it indicates that the current electrode pair is suitable for detecting the ice surface, and the first thickness information is taken as the ice thickness value of the position to be detected. Otherwise, a second equivalent capacitance value corresponding to the complex impedance data collected by a second electrode pair is obtained, a second capacitance-ice thickness fitting curve of the second electrode pair is obtained, and a second thickness information corresponding to the second equivalent capacitance value is calculated in the second capacitance-ice thickness fitting curve. Details are not repeated hereinafter.

[0119] On the basis of the ice detection device, the complex impedance measurement unit of the plurality of electrode pairs is used to collect a plurality of complex impedance data, and the temperature information is collected, the ice detection is carried out, and the ice detection information is obtained. The application expands the data source through the number of electrodes and the electrode layout design, and the temperature information supplements the data of the complex impedance characteristic value, improves the measurement effect, and improves the accuracy of the ice detection. Specifically, the application applies a cascade machine learning algorithm to judge the ice state, and uses an ice type classifier and an ice thickness regressor to judge the ice type and the ice thickness respectively. Through the design and application of the algorithm in the ice measurement, the ice measurement has the robustness, stability and accuracy of data calculation. The multi-level capacitance-ice thickness fitting curve query mode further reduces the complexity of the ice thickness measurement, while ensuring the signal-to-noise ratio of the ice thickness measurement.

[0120] The embodiment of the application also provides a computer device. Please refer to Figure 7 , Figure 7 is a structural schematic diagram of a computer device provided by an optional embodiment of the application, as shown in Figure 7 , the computer device comprises one or more processors 10, a memory 20, and an interface for connecting various components, including a high-speed interface and a low-speed interface. Various components are communicatively connected to each other by different buses, and can be installed on a common mainboard or in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in the memory or memory to display graphical information on a GUI on an external input / output device, such as a display device coupled to the interface. In some optional embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memories, if necessary. Similarly, multiple computer devices can be connected, each device providing part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 7 In the embodiment, the processor 10 is taken as an example.

[0121] The processor 10 can be a central processing unit, a network processing unit, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a generic array logic, or any combination thereof.

[0122] The memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform the methods illustrated in the above embodiments.

[0123] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system and application programs required by at least one function. The data storage area can store data created according to the use of the computer device, and the like. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely from the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0124] The memory 20 can include a volatile memory, such as a random access memory, and can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid state disk. The memory 20 can further include a combination of the above-mentioned types of memories.

[0125] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 can be connected by a bus or other means, Figure 7 For example, the connection by the bus is taken as an example.

[0126] The input device 30 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, and the like. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), and the like. The display device includes, but is not limited to, a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0127] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded to a local storage medium through network, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0128] Part of the present application can be applied as a computer program product, for example, computer program instructions, when executed by a computer, the operation of the computer can invoke or provide the method and / or technical solutions according to the present application. Those skilled in the art should understand that the form of computer program instructions in computer readable medium includes but is not limited to source file, executable file, installation package file, etc. Correspondingly, the way of computer program instructions executed by computer includes but is not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer readable medium can be any available computer readable storage medium or communication medium accessible to the computer.

[0129] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. An icing detection method, applied to an icing detection device, characterized in that, The icing detection device includes a complex impedance measurement unit, a temperature measurement unit, and a data processing unit. The complex impedance measurement unit includes multiple electrodes of preset shapes, arranged according to a preset rule, with at least two electrodes selected as a pair. The method includes: Multiple sets of electrodes are used to collect complex impedance data of multiple frequency bands at the location to be detected, and feature processing is performed on the complex impedance data of each frequency band to obtain complex impedance feature values, the complex impedance feature values ​​including equivalent capacitance values; Temperature information is collected at the location to be detected using a temperature measurement unit; Temperature compensation is performed on the complex impedance characteristic value using the temperature information, and the icing condition at the detection location is detected to obtain icing detection information, including: The complex impedance characteristic value and the temperature information are input into the icing detection model to detect the icing condition at the location to be detected and output icing detection information. The icing detection model includes an ice type classifier and an ice thickness regressor, and includes: inputting the complex impedance characteristic value and the temperature information into the ice type classifier; calculating the probability that the icing type at the location to be detected is a preset icing type, and outputting the preset icing type with the highest probability as the icing type at the location to be detected; wherein, the preset icing type includes clear ice, frost ice, mixed ice, water, contaminant, and dry. Obtain the ice thickness regressor corresponding to the ice type; input the complex impedance characteristic value and the temperature information into the ice thickness regressor, fuse the ice thickness results from different machine learning algorithms, and calculate the ice thickness value at the location to be detected; or Multiple capacitance-ice-thickness fitting curves are constructed, and these multiple capacitance-ice-thickness fitting curves are pre-established ice-thickness detection curves for multiple electrode pairs. The electrode pairs are sorted from smallest to largest according to the electrode spacing, and the corresponding thickness information is determined by using the equivalent capacitance value of the electrode pair and the capacitance-ice thickness fitting curve of the electrode pair in turn. Determine whether the thickness information is less than a preset threshold of the electrode pair. If the thickness information is less than the preset threshold of the electrode pair, use the thickness information as the ice thickness value of the location to be detected. The preset threshold is the thickness detection threshold. Otherwise, the thickness information is calculated using the next electrode pair until the ice thickness value at the location to be detected is obtained; The step of performing feature processing on the complex impedance data for each frequency band to obtain complex impedance characteristic values ​​includes: The complex impedance data is processed using an RC parallel model to calculate a first complex impedance characteristic value, which includes the equivalent capacitance value and the loss tangent value. The second complex impedance characteristic value is calculated based on the first complex impedance characteristic value. The second complex impedance characteristic value includes low-frequency capacitance value, high-frequency capacitance value, relaxation time, high-low frequency capacitance difference, high-low frequency capacitance ratio, number of loss peaks, first loss peak-to-peak value, second loss peak-to-peak value, and loss peak-to-peak ratio.

2. The icing detection method according to claim 1, characterized in that, The step of using the temperature information to perform temperature compensation on the complex impedance characteristic value, and detecting the icing condition at the location to be detected, to obtain icing detection information includes: Using the complex impedance characteristic value and the preset capacitance-ice thickness fitting curve, the icing detection information of the location to be detected is calculated.

3. The icing detection method according to claim 1, characterized in that, The step of determining the corresponding thickness information by sequentially using the electrode pair and the capacitance-ice thickness fitting curve of the electrode pair includes: Obtain the first equivalent capacitance value corresponding to the complex impedance data collected by the first electrode pair with the smallest electrode spacing; Obtain the ice thickness fitting curve of the first capacitance of the first electrode pair; Calculate the first thickness information corresponding to the first equivalent capacitance value from the first capacitance ice thickness fitting curve.

4. A computer device, characterized in that, include: The system includes a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the icing detection method according to any one of claims 1 to 3.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the icing detection method according to any one of claims 1 to 3.

6. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the icing detection method according to any one of claims 1 to 3.

Images