A method and system for detecting zero value high voltage impulse of an insulator

Abstract

The application relates to a kind of insulator zero value high voltage impact detection method and system, belong to insulator detection technical field, comprising: the detection signal of insulator is carried out noise reduction processing;The current voltage feature is obtained and the defect quantization index is constructed to the signal after noise reduction is carried out feature extraction;Using convolutional neural network extracts insulator image feature and outputs the confidence of insulator defect;According to the temperature distribution image of insulator, the temperature feature of insulator is extracted and temperature defect identification index is constructed;According to defect quantization index, the confidence of insulator defect and temperature defect identification index, output final defect judgment result.The application carries out professional noise reduction and feature extraction to detection signal, and is supplemented with the cross verification of image and temperature information, even if single channel exists noise or shielding, can be compensated by other channels, this multi-source information fusion method reduces the uncertainty of single judgment index, makes insulator defect judgment more stable and reliable.

Description

Technical Field

[0001] This invention relates to the field of insulator testing technology, specifically to a method and system for zero-value high-voltage impulse testing of insulators. Background Technology

[0002] Currently, insulators are the most important insulating and supporting components in transmission lines and substations, and their operating status directly affects the safety and stability of the power grid. During long-term operation, insulators are susceptible to defects such as internal cracks, aging of the adhesive layer, damage to the sheds, increased creepage marks, and zero-value insulators due to the combined effects of environmental pollution, ultraviolet aging, wind and rain, temperature changes, and electrical stresses from lightning strikes and electric arcs. Among these, zero-value insulators, due to their severely deteriorated or completely lost electrical insulation performance, are often difficult to identify visually, but significantly reduce the effective creepage distance and withstand voltage of the entire insulator string, posing a significant hidden danger for flashover and tripping accidents. Therefore, effective detection of the zero-value state of insulators is of great engineering significance.

[0003] Existing methods for detecting insulator defects and zero values ​​mainly rely on manual inspection. Manual inspection usually depends on power maintenance personnel to visually inspect the surface of the insulator using telescopes, high-definition cameras, etc. This type of method is relatively sensitive to obvious mechanical damage and missing skirts, but it is weak in identifying non-visual defects such as internal aging and zero values. Moreover, the test results are overly dependent on personnel experience, are highly subjective, and are difficult to standardize and quantify. Summary of the Invention

[0004] To address the aforementioned problems, the purpose of this invention is to provide a method and system for detecting zero-value high-voltage impulse insulators.

[0005] A method for zero-value high-voltage impulse testing of insulators includes: Step 1: Obtain the detection signal of the insulator; the detection signal includes current signal and voltage signal; Step 2: Perform noise reduction processing on the detection signal to obtain the noise-reduced signal; Step 3: Extract current and voltage features from the denoised signal and construct defect quantification indicators; Step 4: Obtain insulator images, use convolutional neural networks to extract insulator image features and output the confidence level of insulator defects; Step 6: Extract the temperature features of the insulator from the temperature distribution image and construct temperature defect identification indicators; Step 7: Output the final defect judgment result based on the defect quantification index, the confidence level of insulator defects, and the temperature defect identification index.

[0006] Preferably, step 2: performing noise reduction processing on the detection signal to obtain a noise-reduced signal includes: Step 2.1: Perform empirical mode decomposition on the detected signal to obtain the IMF components; Step 2.2: Perform wavelet packet transform on the IMF components to obtain the wavelet packet decomposition coefficients; Step 2.3: Construct the coefficient matrix using wavelet packet decomposition coefficients; Step 2.4: Perform singular value decomposition on the coefficient matrix to obtain the denoised signal.

[0007] Preferably, in step 2.2, wavelet packet transform is performed on the IMF components using wavelet basis functions to obtain wavelet packet decomposition coefficients; wherein, the formula for calculating the wavelet packet decomposition coefficients is:

[0008] in, Denotes the decomposition coefficients of the k-th wavelet packet at the j-th level. This represents the i-th IMF component. Describe the wavelet basis functions. Indicates time.

[0009] Preferably, step 2.4: performing singular value decomposition on the coefficient matrix to obtain the denoised signal includes: Step 2.4.1: Perform singular value decomposition on the coefficient matrix, retain singular values ​​greater than the threshold, and obtain a new diagonal matrix; Step 2.4.2: Reconstruct the denoised wavelet packet coefficient matrix based on the new diagonal matrix; Step 2.4.3: Perform inverse wavelet packet transform on the denoised wavelet packet coefficient matrix to obtain the denoised IMF components; Step 2.4.4: Sum the denoised IMF components to obtain the denoised signal.

[0010] Preferably, in step 3, the current and voltage characteristics include peak value, RMS value, kurtosis, waveform factor, impulse energy, spectral centroid, and total harmonic distortion; wherein, the defect quantification index is calculated using the following formula:

[0011] in, Indicates the quantitative indicators of defects. Indicates peak current. Indicates the reference peak current. Indicates kurtosis, Indicates reference kurtosis. Represents the total harmonic distortion rate. Indicates the reference total harmonic distortion rate. Indicates impact energy. Indicates the reference impact energy. Indicates peak voltage. Indicates the reference peak voltage. Indicates the weight of the first feature. Indicates the weight of the second feature. Indicates the weight of the third feature. This represents the weight of the fourth feature. Indicates the weight of the fifth feature. .

[0012] Preferably, in step 6, the temperature characteristics include: maximum temperature, average temperature, temperature standard deviation, and hot spot area ratio; wherein, the hot spot area ratio is calculated using the following formula:

[0013] in, Indicates the area ratio of hotspot areas. This indicates the number of pixels above the temperature threshold. Indicates the length of the temperature distribution image. This indicates the width of the temperature distribution image.

[0014] Preferably, in step 6, the formula for calculating the temperature defect identification index is:

[0015] in, This indicates an index for identifying temperature defects. Indicates the highest temperature. Indicates the highest reference temperature. Indicates reference temperature. Indicates the standard deviation of the reference temperature. Indicates the area ratio of hotspot areas. , Indicates the weight.

[0016] Preferably, step 7: outputting the final defect judgment result based on the defect quantification index, the confidence level of the insulator defect, and the temperature defect identification index, including: The final defect judgment result is output using a weighted voting method based on defect quantification indicators, confidence level of insulator defects, and temperature defect identification indicators.

[0017] The present invention also provides an insulator zero-value high-voltage impulse detection system, comprising: A signal acquisition module is used to acquire the detection signals of the insulator; the detection signals include current signals and voltage signals. The signal noise reduction module is used to process the detection signal to obtain a noise-reduced signal; The signal feature extraction module is used to extract features from the denoised signal to obtain current and voltage features and construct defect quantification indicators. The image defect recognition module is used to acquire insulator images, extract insulator image features using a convolutional neural network, and output the confidence level of insulator defects. The temperature defect identification module is used to extract the temperature characteristics of the insulator based on the temperature distribution image of the insulator and to construct temperature defect identification indicators. The signal acquisition module is used to output the final defect judgment result based on the defect quantification index, the confidence level of the insulator defect, and the temperature defect identification index.

[0018] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps in the above-described method for zero-value high-voltage impulse detection of insulators.

[0019] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: This invention relates to a method for zero-value high-voltage impulse detection of insulators. Compared with the prior art, this invention performs professional noise reduction and feature extraction on the detection signal, and is supplemented by cross-verification of image and temperature information. Even if there is noise or obstruction in a single channel, it can be compensated by other channels. This multi-source information fusion method reduces the uncertainty of a single judgment indicator, making defect judgment more stable and reliable.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the detection rod in the insulator zero-value high-voltage impulse detection device provided by the present invention; Figure 2 This is a schematic diagram of the main unit in the insulator zero-value high-voltage impulse detection device provided by the present invention; Figure 3 This is a schematic diagram of the insulator zero-value high-voltage impulse detection device provided by the present invention; Figure 4 Circuit diagram of the insulator zero-value high-voltage impulse detection device provided by the present invention; Figure 5 The present invention provides a flowchart of a method for zero-value high-voltage impulse detection of insulators. Detailed Implementation

[0023] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0025] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0026] Please see Figure 1-5 A method for zero-value high-voltage impulse testing of insulators, comprising: Step 1: Obtain the detection signal of the insulator; the detection signal includes current signal and voltage signal; like Figure 1-4 As shown, the present invention obtains the detection signal of the insulator through the insulator zero-value high-voltage impulse detection device. The insulator zero-value high-voltage impulse detection device includes: magnet 1, high-voltage rod 2, fixing frame 3, high-voltage line 4, high-voltage pole 5 and main unit (built-in power supply, key board, core board and high-voltage transformer).

[0027] The host computer controls the high-voltage transformer via the core board to apply a zero-value high-voltage impulse voltage with a peak value of 60 kV to the insulator under test to complete the impulse test of the insulator. During the impulse test, the high-voltage rod is conductive and can swing freely under the influence of a magnet to better collect the current and voltage signals of the insulator, which are then input to the host computer for processing. Furthermore, the high-voltage rod 5 in this invention is retractable, facilitating operation in different environments.

[0028] Step 2: Perform noise reduction processing on the detection signal to obtain the noise-reduced signal; Step 2 includes: Step 2.1: Perform empirical mode decomposition on the detected signal to obtain the IMF components; For the original detection signal (x(t)), its EMD (Empirical Mode Decomposition) process can be described as follows: 1. Find all local maxima and local minima of (x(t)).

[0029] 2. Obtain the upper envelope by fitting local maxima points using cubic spline curves. The lower envelope is obtained by fitting local minimum points. .

[0030] 3. Calculate the average value of the upper and lower envelopes. .

[0031] 4. Subtract this average value from the original signal to obtain the intermediate signal. .

[0032] 5. Judgment Is it an intrinsic mode function (IMF)? If not, then... As the new original signal, repeat steps 1-4 to obtain... until If the IMF criteria are met, it is denoted as the first IMF component. .

[0033] 6. Separate from the original signal The remaining signal is obtained. .

[0034] 7. Treating this as a new original signal, repeat steps 1-6 to obtain... until the remaining signal It becomes a monotonic function or a constant.

[0035] 8. The original signal can be represented as the sum of all IMF components and the final residual signal:

[0036] Step 2.2: Perform wavelet packet transform on the IMF components to obtain the wavelet packet decomposition coefficients; 1. Select wavelet basis functions And the number of decomposition layers (N).

[0037] 2. The basis functions of the (j)th layer and (k)th wavelet subspace are: Quantity wavelet packet decomposition coefficients for:

[0038] in yes .

[0039] Step 2.3: Construct the coefficient matrix using wavelet packet decomposition coefficients; Step 2.4: Perform singular value decomposition on the coefficient matrix to obtain the denoised signal.

[0040] The singular value decomposition denoising process is as follows: 1. Wavelet packet coefficients of the (i)th IMF component at layer (j). Constructing the coefficient matrix .

[0041] 2. For the matrix Perform singular value decomposition:

[0042] in, and It is an orthogonal matrix. It is a diagonal matrix, and its diagonal elements It is a singular value.

[0043] 3. Determine the singular value threshold (Methods based on signal-to-noise ratio or energy percentage can be used).

[0044] 4. Threshold the singular values, retaining those greater than the threshold to obtain a new diagonal matrix. ,in:

[0045] 5. Reconstruct the denoised wavelet packet coefficient matrix:

[0046] 6. For the reconstructed wavelet packet coefficients Perform inverse wavelet packet transform to obtain the denoised IMF components. .

[0047] 7. Sum all the denoised IMF components to obtain the signal after EMD-WPT-SVD three-stage denoising:

[0048] Step 3: Extract current and voltage features from the denoised signal and construct defect quantification indicators; In step 3, the current and voltage characteristics include: 1. Peak Value:

[0049] 2. RMS Value:

[0050] Where (T) is the signal period or sampling duration.

[0051] 3. Kurtosis:

[0052] in , respectively, are the discretized current and voltage sample values, and (N) is the number of sampling points.

[0053] 4. Form Factor:

[0054] in, , This is the rectified average value.

[0055] 5. Impact Energy:

[0056] 6. Spectral Centroid: For current signals Performing a Fourier transform yields (I(f)), whose power spectral density is .

[0057]

[0058] in, This is the cutoff frequency. Voltage signal. The calculations are similar.

[0059] 7. Total Harmonic Distortion (THD):

[0060] in, These are the effective values ​​of the (h)th harmonic of the current and voltage signals, respectively. This is the effective value of the fundamental frequency.

[0061] By combining all current and voltage characteristics, a defect quantification index (Q) is constructed:

[0062] in, These are the reference peak current, reference kurtosis, reference total harmonic distortion, reference impulse energy, and reference peak voltage of an intact insulator under the same 60kV impulse. For each feature weight, satisfying .

[0063] Step 4: Obtain insulator images, use convolutional neural networks to extract insulator image features and output the confidence level of insulator defects.

[0064] In this embodiment of the invention, ultraviolet or visible light images of insulators under 60kV high-voltage electric impulse are used. Where H is the height, W is the width, and C is the number of channels, for defect identification.

[0065] Taking a typical CNN structure as an example, it includes convolutional layers, pooling layers, and fully connected layers.

[0066] 1. Convolutional Layer: For the (l)th convolutional layer, the input feature map is ,use Each size is convolution kernel and bias Output feature map :

[0067] in, For activation functions (such as ReLU). , , For the number of fillers, The step size.

[0068] 2. Pooling Layer: Max pooling is used, and the input feature map is applied to the (l)th pooling layer. Output feature map .

[0069] 3. Generate candidate defect regions using methods such as selective search. ( ).

[0070] 4. Divide each candidate region Mapping this onto the feature map of the last convolutional layer yields a fixed-size feature vector. .

[0071] 5. For the eigenvectors Perform classification (defect / non-defect, and defect type) and bounding box regression:

[0072]

[0073] in, For class probability distribution, For bounding box regression parameters, These are learnable parameters.

[0074] 6. Confidence level of image recognition output defects :

[0075] in, This represents the maximum probability of belonging to the defect category across all candidate regions.

[0076] Step 6: Extract the temperature features of the insulator from the temperature distribution image and construct temperature defect identification indicators; In step 6, an image of the temperature distribution of the insulator under a 60kV high-voltage electric impulse is acquired based on infrared thermal imaging technology. Temperature feature extraction and defect identification are performed. Among them, temperature characteristics include: 1. Maximum Temperature:

[0077] 2. Average Temperature:

[0078] 3. Temperature Standard Deviation:

[0079] 4. Hotspot Area Ratio: 5. Set temperature threshold (Above the sum of ambient temperature and normal temperature rise), statistical temperature is higher than number of pixels :

[0080] Constructing temperature defect identification indicators :

[0081] in, These represent the maximum reference temperature and the standard deviation of the reference temperature for intact insulators under the same 60kV impact, respectively.

[0082] Step 7: Output the final defect judgment result based on the defect quantification index, the confidence level of insulator defects, and the temperature defect identification index.

[0083] In step 7, the final defect judgment result is output by combining electrical signals, image, and temperature characteristics using a weighted voting method. (0: No defects, 1: Defective):

[0084] in, The probability of defects in the image model is (0~1). The temperature model defect probability (0~1), weights Decision threshold Determined through optimization using training data.

[0085] This invention performs professional noise reduction and feature extraction on the detection signal, and supplements it with cross-validation of image and temperature information. Even if there is noise or occlusion in a single channel, it can be compensated by other channels. This multi-source information fusion method reduces the uncertainty of a single judgment indicator and makes defect judgment more stable and reliable.

[0086] The present invention also provides a computer-readable storage medium storing a computer program thereon, characterized in that, when the computer program is executed by a processor, it implements the steps in the above-described method for zero-value high-voltage impulse detection of insulators. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the above-described method for zero-value high-voltage impulse detection of insulators, and will not be elaborated here.

[0087] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting zero-value high-voltage impulse insulators, characterized in that, include: Step 1: Obtain the detection signal of the insulator; the detection signal includes current signal and voltage signal; Step 2: Perform noise reduction processing on the detection signal to obtain the noise-reduced signal; Step 2: Denoising the detection signal to obtain a denoised signal, including: Step 2.1: Perform empirical mode decomposition on the detected signal to obtain the IMF components; Step 2.2: Perform wavelet packet transform on the IMF components to obtain the wavelet packet decomposition coefficients; In step 2.2, wavelet packet transform is performed on the IMF components using wavelet basis functions to obtain wavelet packet decomposition coefficients; the formula for calculating the wavelet packet decomposition coefficients is as follows: in, Denotes the decomposition coefficients of the k-th wavelet packet at the j-th level. This represents the i-th IMF component. Describe the wavelet basis functions. Indicates time; Step 2.3: Construct the coefficient matrix using wavelet packet decomposition coefficients; Step 2.4: Perform singular value decomposition on the coefficient matrix to obtain the denoised signal; Step 2.4: Performing singular value decomposition on the coefficient matrix to obtain the denoised signal includes: Step 2.4.1: Perform singular value decomposition on the coefficient matrix, retain singular values ​​greater than the threshold, and obtain a new diagonal matrix; Step 2.4.2: Reconstruct the denoised wavelet packet coefficient matrix based on the new diagonal matrix; Step 2.4.3: Perform inverse wavelet packet transform on the denoised wavelet packet coefficient matrix to obtain the denoised IMF components; Step 2.4.4: Sum the denoised IMF components to obtain the denoised signal; Step 3: Extract current and voltage features from the denoised signal and construct defect quantification indicators; In step 3, the current and voltage characteristics include peak value, RMS value, kurtosis, waveform factor, impulse energy, spectral centroid, and total harmonic distortion; wherein, the defect quantification index is calculated using the following formula: in, Indicates the quantitative indicators of defects. Indicates peak current. Indicates the reference peak current. Indicates kurtosis, Indicates reference kurtosis. Represents the total harmonic distortion rate. Indicates the reference total harmonic distortion rate. Indicates impact energy. Indicates the reference impact energy. Indicates peak voltage. Indicates the reference peak voltage. Indicates the weight of the first feature. Indicates the weight of the second feature. Indicates the weight of the third feature. This represents the weight of the fourth feature. Indicates the weight of the fifth feature. ; Step 4: Obtain insulator images, use convolutional neural networks to extract insulator image features and output the confidence level of insulator defects; Step 5: Extract the temperature features of the insulator from the temperature distribution image and construct temperature defect identification indicators; Step 6: Output the final defect judgment result based on the defect quantification index, the confidence level of insulator defects, and the temperature defect identification index.

2. The method for zero-value high-voltage impulse detection of insulators according to claim 1, characterized in that, In step 5, the temperature characteristics include: maximum temperature, average temperature, temperature standard deviation, and hot spot area ratio; wherein, the hot spot area ratio is calculated using the following formula: in, Indicates the area ratio of hotspot areas. This indicates the number of pixels above the temperature threshold. Indicates the length of the temperature distribution image. This indicates the width of the temperature distribution image.

3. The method for zero-value high-voltage impulse detection of insulators according to claim 2, characterized in that, In step 5, the formula for calculating the temperature defect identification index is: in, This indicates an index for identifying temperature defects. Indicates the highest temperature. Indicates the highest reference temperature. Indicates reference temperature. Indicates the standard deviation of the reference temperature. Indicates the area ratio of hotspot areas. , Indicates the weight.

4. The method for zero-value high-voltage impulse detection of insulators according to claim 3, characterized in that, Step 6: Output the final defect judgment result based on the defect quantification index, the confidence level of the insulator defect, and the temperature defect identification index, including: The final defect judgment result is output using a weighted voting method based on defect quantification indicators, confidence level of insulator defects, and temperature defect identification indicators.

5. A zero-value high-voltage impulse detection system for insulators, characterized in that, include: A signal acquisition module is used to acquire the detection signals of the insulator; the detection signals include current signals and voltage signals. The signal noise reduction module is used to process the detection signal to obtain a noise-reduced signal; The noise reduction process, which involves denoising the detected signal to obtain the denoised signal, includes: Step 2.1: Perform empirical mode decomposition on the detected signal to obtain the IMF components; Step 2.2: Perform wavelet packet transform on the IMF components to obtain the wavelet packet decomposition coefficients; In step 2.2, wavelet packet transform is performed on the IMF components using wavelet basis functions to obtain wavelet packet decomposition coefficients; the formula for calculating the wavelet packet decomposition coefficients is as follows: in, Denotes the decomposition coefficients of the k-th wavelet packet at the j-th level. This represents the i-th IMF component. Describe the wavelet basis functions. Indicates time; Step 2.3: Construct the coefficient matrix using wavelet packet decomposition coefficients; Step 2.4: Perform singular value decomposition on the coefficient matrix to obtain the denoised signal; Step 2.4: Performing singular value decomposition on the coefficient matrix to obtain the denoised signal includes: Step 2.4.1: Perform singular value decomposition on the coefficient matrix, retain singular values ​​greater than the threshold, and obtain a new diagonal matrix; Step 2.4.2: Reconstruct the denoised wavelet packet coefficient matrix based on the new diagonal matrix; Step 2.4.3: Perform inverse wavelet packet transform on the denoised wavelet packet coefficient matrix to obtain the denoised IMF components; Step 2.4.4: Sum the denoised IMF components to obtain the denoised signal; The signal feature extraction module is used to extract current and voltage features from the denoised signal and construct defect quantification indicators. The current and voltage features include peak value, RMS value, kurtosis, waveform factor, impulse energy, spectral centroid, and total harmonic distortion. The defect quantification indicator is calculated using the following formula: in, Indicates the quantitative indicators of defects. Indicates peak current. Indicates the reference peak current. Indicates kurtosis, Indicates reference kurtosis. Represents the total harmonic distortion rate. Indicates the reference total harmonic distortion rate. Indicates impact energy. Indicates the reference impact energy. Indicates peak voltage. Indicates the reference peak voltage. Indicates the weight of the first feature. Indicates the weight of the second feature. Indicates the weight of the third feature. This represents the weight of the fourth feature. Indicates the weight of the fifth feature. ; The image defect recognition module is used to acquire insulator images, extract insulator image features using a convolutional neural network, and output the confidence level of insulator defects. The temperature defect identification module is used to extract the temperature characteristics of the insulator based on the temperature distribution image of the insulator and to construct temperature defect identification indicators. The signal acquisition module is used to output the final defect judgment result based on the defect quantification index, the confidence level of the insulator defect, and the temperature defect identification index.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps in the zero-value high-voltage impulse detection method for insulators as described in any one of claims 1-4.

Images