An electrostatic spray head assembly that prevents the atomizing system from becoming charged
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIANGSU HUAIQING AGRICULTURAL CO LTD
- Filing Date
- 2026-04-10
- Publication Date
- 2026-06-09
Smart Images
Figure CN122164571A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electrostatic spraying technology, and more specifically to an electrostatic nozzle assembly that prevents the atomization system from becoming charged. Background Technology
[0002] The core of electrostatic spraying technology is to atomize liquid and then apply an electric charge to the droplets, causing the charged droplets to be oriented and adsorbed under the action of an electric field, thereby improving the adhesion effect and atomization utilization rate. It is widely used in agricultural plant protection, industrial spraying, disinfection and sterilization and other fields.
[0003] In existing electrostatic spraying technology, high-voltage static electricity is typically applied directly to the atomizing nozzle to charge the droplets. However, since the sprayed liquid is often conductive, the high-voltage charge can be conducted back through the liquid to the entire liquid supply system, causing various components of the atomizing system itself to become charged. This traditional structure has several drawbacks, including: 1. High insulation protection requirements and high system cost: Since high voltage charge can be conducted to the entire liquid supply system through the liquid, in order to ensure safe use, it is necessary to carry out comprehensive insulation design for nozzles, liquid pipelines, connectors, water tanks and other potentially electrified components, and select high-performance insulation materials, along with corresponding protective measures. This not only increases the complexity of system design, but also significantly increases material and manufacturing costs, thus limiting its promotion in low- and medium-cost application scenarios.
[0004] 2. Charge accumulation and partial discharge are prone to occur, affecting lifespan and reliability: After being conducted along the liquid path, the charge is prone to accumulate at uneven, sharp, and turning parts of the pipeline, forming a concentrated electric field and causing partial discharge, which corrodes and damages water circuit components, may lead to pipeline breakdown and leakage, increase maintenance frequency, and reduce system stability and service life.
[0005] 3. Dispersed electrostatic effects and poor droplet charging: In traditional structures, the charge output by the electrostatic generator not only acts on the sprayed mist but is also dispersed to multiple parts such as the nozzle, pipeline, and water tank, making it difficult to concentrate on charging the droplets and resulting in a decrease in the effective charging ratio. This further leads to insufficient droplet charge and poor charging uniformity, affecting the droplet adsorption effect on the target area and the atomization utilization rate, making it difficult to meet the application requirements of high-efficiency spraying operations.
[0006] In summary, existing atomizing nozzle structures easily allow charge to diffuse along the liquid path into the liquid supply system, resulting in problems such as high insulation protection requirements, significant charge accumulation and partial discharge risks, and insufficient droplet charging efficiency. Therefore, it is necessary to provide a new electrostatic nozzle solution to overcome these shortcomings. Summary of the Invention
[0007] To achieve the above-mentioned objectives, this invention provides an electrostatic nozzle assembly that prevents the atomization system from becoming charged, the specific technical solution of which is as follows: An electrostatic nozzle assembly for preventing the atomization system from becoming charged, comprising: An insulated nozzle is used to atomize liquid and spray it outward as a mist. A discharge body is disposed on the path of the mist stream ejected from the insulating nozzle and spaced apart from the insulating nozzle, for applying static electricity to the mist stream; A grounding component is disposed at the end of the insulating nozzle away from the discharge body and is electrically connected to the liquid circuit system. It is used to conduct the returned charge to the ground to prevent high-voltage charge from being conducted to the liquid circuit system.
[0008] The discharge body includes a discharge mesh area, and the mist flow is sprayed outward after passing through the discharge mesh area.
[0009] Furthermore, the cross-sectional area defined by the outer contour of the discharge mesh area is set in a positively proportional relationship with the cross-sectional area of the mist flow in the plane where the discharge body is located.
[0010] Furthermore, the discharge body also includes a plurality of discharge needles disposed around the periphery of the discharge mesh area, the plurality of discharge needles being distributed at intervals along a circular trajectory surrounding the center of the discharge mesh area.
[0011] Furthermore, the plurality of discharge needles have clearance openings in the area below the discharge mesh area to allow mist flow to pass through.
[0012] Furthermore, the angle range corresponding to the avoidance notch in the annular distribution direction of the plurality of discharge needles is 120° to 180°.
[0013] Furthermore, the outer ends of the discharge needles located below the horizontal plane at the center of the discharge mesh area are all bent upwards.
[0014] Furthermore, the insulating nozzle is a nozzle made of insulating material, or at least a nozzle with an insulating coating on its outer surface.
[0015] In addition, the electrostatic nozzle assembly also includes a distance adjustment component. The discharge body is installed on the mist path ejected by the insulating nozzle through the distance adjustment component to adjust the relative distance between the discharge body and the insulating nozzle.
[0016] The beneficial effects of this application are as follows: By setting the nozzle as an insulated nozzle and placing the discharge body in the path of the mist ejected from the insulated nozzle at a distance from it, the discharge body applies electrostatics to the mist while reducing the possibility of high-voltage charges being directly conducted to the nozzle body. Simultaneously, by incorporating a grounding component that contacts the liquid flowing through the insulated nozzle, the charges transmitted back from the discharge body through the mist can be promptly guided to the ground, making it difficult for high-voltage charges to be further conducted to the liquid circuit system connected to the insulated nozzle. This reduces the risk of electrification in downstream components such as water pipelines, water tanks, and water pumps, lowers the difficulty of insulation protection for the atomization system, and helps reduce the cost of insulation materials and manufacturing. Attached Figure Description
[0017] Figure 1 This is a perspective view of the electrostatic nozzle assembly in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the discharge body in Embodiment 2 of this application; Figure 3 This is a schematic diagram of the structure of the discharge body in Embodiment 3 of this application. Detailed Implementation
[0018] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. All equivalent substitutions, improvements, and modifications made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
[0019] This invention provides a radar-based underground bamboo shoot detection device that can address the problems of concealed burial locations of bamboo shoots in underground soil environments, weak target echoes, complex background clutter, and low efficiency of manual search. It enables rapid detection, feature enhancement, intelligent identification, and result visualization of underground bamboo shoot targets, thereby improving the accuracy and efficiency of bamboo shoot detection.
[0020] In one embodiment, the underground winter bamboo shoot detection device based on radar technology of the present invention includes a radar detection module, a data processing unit, an image enhancement processing module, a model training module, a result display module, and a positioning module.
[0021] The system comprises the following components: a radar detection module for transmitting electromagnetic waves into the underground medium and receiving echo signals reflected from underground targets and the soil environment; a data processing unit connected to the radar detection module for preprocessing, feature extraction, and image conversion of the echo signals; an image enhancement module connected to the data processing unit for enhancing image data or feature maps; a model training module for constructing and training a bamboo shoot recognition model based on sample data and using the trained model to identify underground targets; a result display module for outputting the bamboo shoot detection results and graphically displaying the location, depth, and / or contour information of the bamboo shoots; and a positioning module for recording the spatial location information of the radar detection path and associating it with the radar echo data to achieve spatial positioning of underground bamboo shoot targets.
[0022] In practical applications, the device can be integrated into portable detection terminals, backpack detection equipment, hand-pushed detection platforms, vehicle-mounted platforms, or unmanned mobile platforms. Preferably, the device has a portable or semi-portable structure to meet the mobile detection needs in mountainous bamboo forest environments.
[0023] In one embodiment, the radar detection module is a ground-penetrating radar module. This ground-penetrating radar module includes a transmitting antenna, a receiving antenna, a control circuit, and a data acquisition circuit. The transmitting antenna is used to transmit detection electromagnetic waves within a preset frequency range into the ground; the receiving antenna is used to receive echo signals reflected from underground bamboo shoot targets, soil layer structures, rocks, roots, and other underground media interfaces; the control circuit is used to control the transmission timing, power, pulse repetition frequency, and sampling triggering process of the detection electromagnetic waves; and the data acquisition circuit is used to perform analog-to-digital conversion, sampling buffering, and data transmission on the received echo signals.
[0024] In one optional implementation, the ground-penetrating radar module employs a high-frequency antenna suitable for detecting shallow underground targets, balancing detection depth and resolution. Considering that bamboo shoots are typically buried at a certain depth below the surface, antenna components with different center frequencies can be selected based on different soil types, moisture content, and target size. For scenarios where the target is shallow and higher resolution is required, a higher center frequency can be used; for scenarios with significant soil attenuation or higher detection depth requirements, a lower center frequency can be used.
[0025] In one embodiment, the radar detection module moves along a preset path to scan and continuously acquire echo data from multiple detection locations, thereby forming underground profile data. The detection path can be a straight line, a curve, a grid-like path, or a regional coverage path. Data from multiple detection paths can be further stitched together to form a regional underground distribution map, facilitating the location and statistical analysis of bamboo shoot targets.
[0026] The data processing unit is used to perform signal preprocessing, feature extraction, image conversion, and depth estimation on radar echo signals.
[0027] In one embodiment, the data processing unit performs at least one preprocessing operation on the echo signal, including denoising, time-zero correction, gain compensation, background removal, filtering, and time-depth conversion. Among these, denoising can be used to reduce random interference caused by system noise, environmental noise, or sampling errors; time-zero correction can be used to correct the offset of the electromagnetic wave incident start time, so that each echo can be compared under a unified reference time axis; gain compensation can be used to compensate for the energy attenuation of electromagnetic waves with increasing depth during underground propagation, thereby enhancing the visibility of weak targets in deep underground areas; background removal can be used to reduce the influence of stable background echoes and fixed structure echoes on target identification; filtering can use bandpass filtering, mean filtering, median filtering, wavelet filtering, or other suitable methods to improve the signal-to-noise ratio of the target signal.
[0028] In a preferred embodiment, the original echo signal is first subjected to time-zero correction, then bandpass filtering and background removal are performed, and then gain compensation is performed based on propagation attenuation to obtain preprocessed data that is more suitable for subsequent feature extraction and image conversion.
[0029] In one embodiment, the data processing unit extracts time-domain features, frequency-domain features, and / or texture features based on the preprocessed echo signal.
[0030] The time-domain features may include echo amplitude, energy, envelope, peak position, waveform width, hyperbolic morphological parameters, or target echo duration, etc. The frequency domain characteristics may include spectral distribution, dominant frequency, frequency band energy, frequency centroid, or spectral variation trend, etc. The texture features may include image texture information obtained based on gray-level co-occurrence matrix, local binary pattern, edge gradient distribution, or local structural changes.
[0031] By extracting the above features, the difference between underground bamboo shoots and other underground reflectors such as soil background, rocks, roots, or cavities can be enhanced, providing an input basis for subsequent image enhancement and intelligent recognition.
[0032] In one embodiment, the data processing unit converts the echo signal and / or preprocessed data into image data or feature maps for target recognition. The image data may be one or more of the following: A-scan waveform, B-scan radar profile, C-scan planar image, time-depth map, amplitude map, energy map, spectrum map, or multi-channel fused feature map.
[0033] Preferably, echo data from multiple consecutive detection locations are arranged in spatial order to form a B-scan profile, reflecting the continuous distribution characteristics of underground targets along the scanning direction. For winter bamboo shoot targets, they typically appear in the profile as locally enhanced echoes, concentrated morphological regions, or regions with specific geometric features. Visual representation facilitates subsequent image enhancement and intelligent recognition.
[0034] In one embodiment, the data processing unit is further configured to estimate the depth information of the underground bamboo shoot target based on the radar echo propagation time and the electromagnetic parameters of the underground medium.
[0035] Specifically, depth can be calculated based on the two-way propagation time of the target echo and the electromagnetic wave propagation velocity in the subsurface medium. The propagation velocity can be estimated based on the soil dielectric constant, empirical models, field calibration results, or measurements from a known reference body. For soils with different moisture contents and compaction degrees, corresponding electromagnetic parameter correction relationships can be established to improve the accuracy of depth estimation.
[0036] In one implementation, the data processing unit can adaptively update the electromagnetic parameters of the subsurface medium by combining multiple repeated measurements, comparisons of adjacent profiles, or known calibration points, thereby reducing the impact of soil environment changes on depth estimation results.
[0037] The image enhancement processing module is used to further enhance the image data or feature map output by the data processing unit, so as to improve the expressive ability of the target-related features of winter bamboo shoots and suppress complex soil background and non-target interference.
[0038] In one embodiment, the image enhancement processing module includes an image preprocessing unit, a background suppression unit, a color space feature enhancement unit, and a multidimensional attention enhancement unit.
[0039] 1. Image Preprocessing Unit The image preprocessing unit is used to adaptively scale, normalize, convert color spaces, and / or augment the input image data or feature maps.
[0040] Among them, adaptive scaling can be used to unify the input size to adapt to the subsequent recognition model; normalization can be used to reduce the difference in data amplitude range between different collection batches, different soil areas and different equipment states; color space conversion can be used to enhance the contrast of certain target features in different representation channels; data augmentation can include rotation, mirroring, cropping, noise perturbation, brightness adjustment, contrast adjustment, random occlusion or channel perturbation, etc., to improve the model's adaptability and generalization ability to underground bamboo shoot targets in different scenarios.
[0041] The background suppression unit is used to generate a target foreground mask using a staged mask generation and repair method, and extract the target region based on the foreground mask, so as to reduce the influence of soil background, clutter interference and non-target regions on winter bamboo shoot identification.
[0042] In one embodiment, the background suppression unit can first generate an initial mask based on the image grayscale distribution, energy concentration, edge changes, or local saliency information; then perform morphological processing, hole filling, edge repair, connection region optimization, or false target removal on the initial mask to obtain a more accurate foreground mask; finally, use the foreground mask to extract or weighted enhance the target area, so that the area where the underground winter bamboo shoots may be located is highlighted, while large areas of soil background and irrelevant areas are suppressed.
[0043] In one alternative implementation, the background suppression process can be achieved by combining threshold segmentation with region inpainting, or by learning-based foreground segmentation; the present invention does not limit this to either approach.
[0044] The color space feature enhancement unit is used to convert the image data to a preset color representation space and introduce attention mechanisms for different channels to enhance the color-related feature expression capability of underground bamboo shoots.
[0045] In a preferred embodiment, the preset color representation space is the YES color space. The color space feature enhancement unit converts the input image to the YES color space and introduces attention mechanisms on the Y, E, and S channels respectively to improve the distinguishability between the underground bamboo shoot target and the background.
[0046] The Y channel can be used to characterize brightness or intensity-related information, while the E and S channels can be used to characterize different color or structural differences. By weighting each channel separately, key responses related to the bamboo shoot target can be highlighted, while background noise and non-target responses can be reduced. The attention weights for each channel can be determined based on channel statistical characteristics, local response intensity, global contextual information, or training results.
[0047] It should be noted that although the preferred embodiment uses the YES color space, other representation spaces that are beneficial to enhancing the difference between the underground winter bamboo shoot target and the background can also be used in other embodiments. As long as the target features can be enhanced, they can fall within the protection scope of this invention. The multidimensional attention enhancement unit is used to perform multidimensional attention modeling and fusion of feature maps to improve the model's ability to perceive key features of underground winter bamboo shoots.
[0048] In a preferred embodiment, the multidimensional attention enhancement unit is a ternary attention enhancement unit, used to perform attention modeling and fusion of channel dimension, spatial dimension and cross-dimensional features on the feature map.
[0049] Among them, channel-dimensional attention is used to evaluate the importance of different feature channels for the bamboo shoot recognition task and enhance effective channels; spatial-dimensional attention is used to determine the spatial regions in the image that are more likely to correspond to the bamboo shoot target; and cross-dimensional feature attention is used to establish the coupling relationship between channel information and spatial information, thereby obtaining a more comprehensive target representation. Through the three-dimensional attention mechanism, attention can be paid to "what features to look at", "where to look", and "how different dimensions cooperate", thereby improving the recognition performance of underground bamboo shoot targets in complex backgrounds.
[0050] In one implementation, the multidimensional attention enhancement unit can be set at the input end of the recognition model, the intermediate layer of the backbone feature extraction network, or the pre-output fusion layer to achieve flexible deployment according to the actual model architecture.
[0051] The model training module is used to build and train a winter bamboo shoot recognition model based on sample data, so as to identify underground winter bamboo shoot targets according to the feature data and / or the enhanced image features.
[0052] In one embodiment, the bamboo shoot identification model is a machine learning model and / or a deep learning model. The machine learning model may include models such as support vector machines, random forests, decision trees, K-nearest neighbors, and Naive Bayes; the deep learning model may include convolutional neural networks, residual neural networks, object detection networks, semantic segmentation networks, attention enhancement networks, or multi-branch fusion networks.
[0053] In one embodiment, the model training module is trained based on a pre-built underground bamboo shoot sample database. The sample database includes target bamboo shoot samples and non-target bamboo shoot samples. The non-target bamboo shoot samples may include stones, roots, cavities, soil interfaces, and other underground reflectors easily confused with bamboo shoots. Sample data may originate from field-collected data, manually labeled data, augmented data under different environmental conditions, and simulation-generated data.
[0054] In a preferred embodiment, the model training process includes the following steps: 1) Collect and label underground radar echo data; 2) Preprocess the sample data and perform image enhancement; 3) Construct training, validation, and test sets; 4) Select a suitable recognition model structure and set the loss function, optimizer, and training parameters; 5) Iteratively train the model and adjust the hyperparameters based on the validation results; 6) Output the trained bamboo shoot recognition model.
[0055] After training, the recognition model receives the feature data and / or the enhanced image features, and outputs the recognition result that the underground target belongs to the bamboo shoot target and the corresponding confidence level. The recognition result can be a binary classification result of target presence / absence, or it can be a target category, candidate box position, pixel-level segmentation result, or target contour information.
[0056] In one implementation, the model training module can be deployed on a local computing terminal, an edge computing device, or a remote server. For resource-constrained field detection devices, training can be completed on the server side, and the trained model can be loaded onto the field device for inference and recognition.
[0057] The result display module is used to output the detection results of underground winter bamboo shoots based on the identification results, and to display the location and / or depth information of the underground winter bamboo shoots in a graphical manner.
[0058] In one embodiment, the result display module is used to generate at least one of an underground target distribution map, profile map, depth map, or location map based on the identification results, and to indicate the burial location, burial depth, and / or target outline of the winter bamboo shoot target.
[0059] For example, the relative position and depth of the bamboo shoot target under the scanning path can be marked in the cross-sectional view; the actual spatial position of the bamboo shoot target can be marked in the planar positioning view in combination with the coordinate information provided by the positioning module; the burial depth of different targets can be displayed in the depth map through color gradient or numerical labels; and multiple bamboo shoot targets can be statistically displayed and regionally summarized in the target distribution map.
[0060] In one embodiment, the results display module may also output auxiliary information such as target confidence level, target number, detection time, detection path, soil parameter estimation results, and suggested excavation area to facilitate on-site judgment and subsequent operations by the operator.
[0061] The positioning module is used to record the spatial location information of the radar detection path and associate it with the radar echo data to achieve spatial positioning of underground bamboo shoot targets.
[0062] In one embodiment, the positioning module may employ a satellite positioning unit, an inertial measurement unit, an odometer, a visual positioning unit, or a combination thereof to acquire the motion trajectory and spatial coordinate information of the detection device. For forest areas, mountainous areas, or areas with limited satellite signals, a combination of inertial navigation and path estimation can be used for compensated positioning.
[0063] In one embodiment, the location information output by the positioning module is time-synchronized or indexed with radar echo data at each detection time, thereby mapping the identified underground bamboo shoot target to its actual ground coordinates. Operators can then quickly locate the bamboo shoot target area, improving mining efficiency and reducing surface disturbance caused by indiscriminate excavation.
[0064] In one embodiment, the operation of the device of the present invention includes the following steps: S1, activate the radar detection module and scan the target area along the preset detection path to obtain underground echo data; S2, the data processing unit preprocesses the echo data, including denoising, time-zero correction, gain compensation, background removal and / or filtering, and extracts time-domain features, frequency-domain features and / or texture features; S3, the data processing unit converts the preprocessed echo data into image data or feature map, and estimates the target depth by combining the electromagnetic parameters of the underground medium; S4, the image enhancement processing module enhances image data or feature maps, including image preprocessing, background suppression, color space feature enhancement, and multidimensional attention enhancement; S5, the model training module calls the trained winter bamboo shoot recognition model to recognize the enhanced image features and / or feature data, and outputs the recognition result of whether the underground target is a winter bamboo shoot target and the corresponding confidence level; S6, the result display module, in conjunction with the positioning module, outputs the location, depth, outline, and distribution information of the bamboo shoot target.
[0065] Compared with existing methods that rely on manual experience to determine the location of underground winter bamboo shoots by judging surface uplifts, soil cracks, or bamboo rhizomes, this invention has at least the following advantages: Using radar detection methods to achieve non-destructive underground detection, it is possible to obtain response information of underground targets without excavation; By preprocessing signals and extracting features, the signal-to-noise ratio and distinguishability of underground winter bamboo shoot target echoes are improved. By using background suppression, color space feature enhancement, and multidimensional attention enhancement in the image enhancement processing module, complex soil backgrounds and clutter interference are effectively reduced. Intelligent identification of underground winter bamboo shoots can be achieved by using machine learning models and / or deep learning models, which can improve the accuracy and automation level of identification. By combining the results display module and the positioning module, the burial location and depth information of underground bamboo shoot targets can be presented intuitively, which is convenient for on-site operations. This invention is applicable to different soil environments, different bamboo forest topography, and different winter bamboo shoot burial depths, and has good application and promotion value.
[0066] It should be noted that the present invention is not limited to the embodiments described above. For those skilled in the art, various modifications and improvements can be made to the structure, connection relationships, algorithm flow, and parameter settings of each module without departing from the concept of the present invention.
[0067] For example, the background suppression unit, color space feature enhancement unit, and multi-dimensional attention enhancement unit in the image enhancement processing module can be arranged in series or in parallel and then feature fusion is performed; the model training module can use a single model or multiple models cascaded or integrated for decision-making; the result display module can be set on the local terminal or the results can be sent to a remote display terminal via wireless communication.
[0068] Furthermore, in some embodiments, the radar detection module and the positioning module can be integrated into one unit, and the data processing unit, image enhancement processing module, and model training module can be integrated on the same processor platform or distributed across multiple processor platforms to work collaboratively. As long as the technical solution falls within the substantive content and protection scope of this invention, it should be considered within the protection scope of this invention.
Claims
1. An electrostatic nozzle assembly for preventing the atomization system from becoming charged, characterized in that, include: Insulated nozzle (1) is used to atomize liquid and spray it outward as a mist; A discharge body (2) is disposed on the path of the mist flow ejected by the insulating nozzle (1) and spaced apart from the insulating nozzle (1) for applying static electricity to the mist flow; The grounding component (3) is disposed on the end of the insulating nozzle (1) away from the discharge body (2) and is electrically connected to the liquid circuit system to conduct the return charge to the ground so as to prevent the high voltage charge from being conducted to the liquid circuit system.
2. The electrostatic nozzle assembly according to claim 1, characterized in that: The discharge body (2) includes a discharge mesh area (21), and the mist flows outward after passing through the discharge mesh area (21).
3. The electrostatic nozzle assembly according to claim 2, characterized in that: The cross-sectional area defined by the outer contour of the discharge mesh area (21) is set in a positively proportional relationship with the cross-sectional area of the mist flow in the plane where the discharge body (2) is located.
4. The electrostatic nozzle assembly according to claim 2, characterized in that: The discharge body (2) also includes a plurality of discharge needles (22) disposed around the discharge mesh area (21), and the plurality of discharge needles (22) are distributed at intervals along a ring trajectory around the center of the discharge mesh area (21).
5. The electrostatic nozzle assembly according to claim 4, characterized in that: Multiple discharge needles (22) have clearance openings in the area below the discharge mesh area (21) to allow mist flow to pass through.
6. The electrostatic nozzle assembly according to claim 5, characterized in that: The angle range of the avoidance notch corresponding to the annular distribution direction of the plurality of discharge needles (22) is 120° to 180°.
7. The electrostatic nozzle assembly according to claim 4, characterized in that: The outer ends of the discharge needles (22) located below the center horizontal plane of the discharge mesh area (21) are all bent upwards.
8. The electrostatic nozzle assembly according to claim 1, characterized in that: The insulating nozzle (1) is a nozzle made of insulating material, or a nozzle with an insulating coating on its outer surface.
9. The electrostatic nozzle assembly according to claim 1, characterized in that: It also includes a distance adjustment component (4), wherein the discharge body (2) is installed on the mist path ejected by the insulating nozzle (1) through the distance adjustment component (4) to adjust the relative distance between the discharge body (2) and the insulating nozzle (1).