Ice melting and deicing method and system based on induction type magnetic thermal sound

Abstract

The invention relates to the technical field of induction type magnetic thermoacoustic, and discloses an ice melting and deicing method and system based on induction type magnetic thermoacoustic. The method comprises the following steps: acquiring ice surface state signal data in real time, and constructing a multi-dimensional signal processing system to output a standardized ice melting signal; establishing a parameter database based on a magneto-thermo-acoustic conversion technology, and positioning target parameters by using a feature matching algorithm to generate a control label; a historical ice melting case information base is constructed, a multi-factor weight distribution model is combined, weighting is carried out through a decision tree algorithm, ice melting matching degree scores are calculated, and a priority sequence is generated; solving an optimal ice melting scheme by adopting an improved genetic algorithm; and outputting a result through the interactive verification platform, and tracking and feeding back. According to the method, the ice melting process can be accurately regulated and controlled, multiple factors are comprehensively considered, the ice melting efficiency is effectively improved, the energy consumption is reduced, intelligent and accurate ice melting is achieved, and the method is suitable for ice melting and deicing work in multiple scenes.

Description

Technical Field

[0001] The present invention relates to the field of inductive magnetothermoacoustic technology, and in particular to an ice melting and deicing method and system based on inductive magnetothermoacoustic technology. Background Art

[0002] In the fields of electricity, transportation, etc., ice cover poses a serious threat to equipment operation and traffic safety. Research on ice melting and de-icing technology has always been an industry focus. Traditional ice melting and de-icing methods have many shortcomings. On the one hand, at the signal acquisition and processing level, most technologies rely on a single type of sensor to obtain ice surface signals. The collected data dimension is single and cannot fully reflect the physical characteristics of the ice surface. In addition, the signal preprocessing process is simple, and only basic filtering and denoising are performed. It is difficult to effectively remove interference clutter in complex environments, resulting in low accuracy and reliability of subsequent ice melting signals.

[0003] In terms of parameter matching and control, existing technologies lack a systematic parameter database and are unable to comprehensively collect and manage standard parameters of common application scenarios, equipment types, and ice layer materials based on magnetothermal acoustic conversion technology. The correlation between parameters has not been fully explored, making it difficult to quickly and accurately locate target parameters and generate reasonable control labels when ice melting needs are detected, affecting the timeliness and accuracy of ice melting operations.

[0004] In terms of the use of historical data, traditional methods have not built a historical ice melting case information database, and are unable to dynamically record ice melting records in different scenarios, materials and ice thicknesses. There is a lack of summary and utilization of historical experience. When faced with new ice melting tasks, optimization based on historical data cannot be carried out, resulting in a lack of reference for the formulation of ice melting plans, making it difficult to adapt to complex and changing actual conditions.

[0005] In the priority allocation link, the existing technology has not established a scientific multi-factor weight allocation model, and is unable to comprehensively consider factors such as scene type, material properties, ice thickness, and environmental temperature and humidity to reasonably rank the priorities of ice melting tasks. This can easily lead to unreasonable resource allocation, resulting in high-importance and high-complexity ice melting tasks not being processed in a timely manner, affecting the overall ice melting efficiency.

[0006] In terms of solution optimization, traditional ice-melting methods lack algorithm optimization capabilities when solving ice-melting solutions, and are unable to efficiently optimize with the goals of maximizing ice-melting efficiency, minimizing energy consumption thresholds, and shortening action time. This results in high energy consumption, long action time, and low efficiency in the ice-melting solution, making it difficult to meet the requirements of energy saving and efficiency in actual applications.

[0007] The existing ice-melting system lacks an effective feedback tracking mechanism and is unable to continuously track the actual ice-melting effect and optimize it, resulting in poor adaptability and reliability of the system and difficulty in maintaining good ice-melting performance under different environments and working conditions. Summary of the Invention

[0008] The object of the present invention is to provide an ice melting and deicing method and system based on induction magneto-thermoacoustic technology to solve the problems raised in the above-mentioned background technology.

[0009] To achieve the above objectives, the present invention provides an ice melting and deicing method based on induction magnetothermoacoustic technology, the method comprising: Collect ice surface status signal data in real time, build a multi-dimensional signal processing system, and output standardized ice melting signals through the signal preprocessing module; A parameter database is established based on magnetothermal acoustic conversion technology. When ice melting needs are detected, the target parameters are located through feature matching algorithms and control tags are generated. Build a historical ice melting case information database to dynamically record ice melting records in different scenarios, materials, and ice thickness; A multi-factor weight distribution model was established. By combining scene type, material properties, ice thickness, and ambient temperature and humidity factors, a decision tree algorithm was used to assign weights to each indicator. A matching model was constructed using a weighted summation method to calculate the ice melting matching score, thereby generating an ice melting priority sequence. An improved genetic algorithm is used to solve the optimal ice melting scheme, with the optimization objectives set in the direction of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time. The final ice-melting results are output through the interactive verification platform, and the actual ice-melting effect feedback is continuously tracked.

[0010] Preferably, the real-time collection of ice surface status signal data, the construction of a multi-dimensional signal processing system, and the output of a standardized ice melting signal through a signal preprocessing module are specifically as follows: Based on the inductive sensor to obtain the signal data of the ice surface area, the signal acquisition terminal receives the original signal from the sensor and performs time domain filtering, frequency domain denoising, and edge enhancement preprocessing to remove environmental interference clutter; According to the physical properties of the ice surface and the magnetothermoacoustic response law, the parameters closely related to ice melting characteristics are selected as candidate features based on the feature selection algorithm; The candidate features are divided into magnetic permeability features, thermal diffusion features, and acoustic impedance features. A multi-level feature system is constructed, with the overall ice melting matching degree as the top-level feature. The top-level feature is decomposed downward into several first-level features, and each first-level feature is further subdivided into several second-level features. Based on the multi-level feature system, a neural network model is used to complete the signal preprocessing module training; Correlate the ice surface position information with the ice melting characteristics of the pre-processed signal to establish a position-feature mapping relationship; Signal analysis software was selected as the magnetothermoacoustic analysis platform, and areas were marked on the interface according to the actual position of the ice surface. Different brightness levels were used for visualization based on the differences in features. Areas with large feature differences were highlighted, and areas with small feature differences were low-brightness.

[0011] Preferably, the parameter database is established based on the magnetothermoacoustic conversion technology, specifically: Collect standard parameter data for common application scenarios, equipment types, and ice layer materials, including basic parameters such as magnetic saturation value, thermal conductivity value, and sound velocity value; Obtaining physical property information of parameters, including medium density, interface bonding strength, and surface roughness parameters; Parameters are abstracted into data nodes, and the magneto-thermo-acoustic correlations between parameters are abstracted into data associations, thus constructing a node-association data structure. Establish database architecture, including data table definition, field settings and data relationship establishment, storage parameter information, attribute information and association relationships; Import the preprocessed data into the database of the magnetothermoacoustic analysis software to establish a parameter database.

[0012] Preferably, the method of locating target parameters and generating control tags by using a feature matching algorithm is as follows: Based on the parameter database, each parameter node is regarded as a sample point in the data set, and the magnetothermoacoustic correlation between parameters is regarded as the similarity between samples. The signal characteristics of the area to be de-ice-melted are used as query points. The characteristic distance to each sample point is calculated using a dynamic programming algorithm. When the sample point with the highest matching degree is found, the characteristic information of the sample point is recorded. Analyze the calculated characteristic distance and combine it with ice thickness and material characteristics to obtain the matching range of the target parameters; Extract the magnetic saturation value, thermal conductivity value, and sound velocity value parameters of the target parameters from the parameter database to generate control parameters; The control parameters of the target parameters, scene type, and ice layer material are combined into a label and visualized on the magnetothermoacoustic analysis software interface.

[0013] Preferably, the historical ice melting case information database is constructed to dynamically record ice melting records of different scenes, materials, and ice thicknesses, specifically: Obtain the main dimensions for building the information database, including ice melting records of different scene types, ice layer materials, and ice layer thickness; At the same time, specific information fields are planned for each dimension, including application scenarios, equipment types, ice layer materials, standard parameters, actual control parameters, and user operation feedback; By connecting the de-icing equipment management system with the user feedback platform, the matching results, energy consumption feedback, and operation effect information of historical de-icing cases are continuously updated; Build a historical ice melting case information database based on main dimensions and data fields.

[0014] Preferably, the multi-factor weight distribution model is established by combining scene type, material properties, ice thickness, and ambient temperature and humidity factors, using a decision tree algorithm to assign weights to each indicator, and using a weighted summation method to construct a matching model to calculate the ice melting matching score, thereby generating an ice melting priority sequence, specifically: The importance of ice melting is divided into different levels based on the scenario type, standard parameters, and ice thickness factors. High-frequency scenarios or scenarios with clear standard parameters are assigned a high importance level, while low-frequency scenarios or scenarios without clear parameters are assigned a low importance level. The ice layer material is divided into metal surface ice, concrete surface ice, and glass surface ice. The complexity level of ice melting on metal surface is high, while the complexity level of ice melting on glass surface is low. The ambient temperature and humidity are divided into standard temperature and humidity, high temperature and high humidity environment, and low temperature and low humidity environment; Based on the decision tree algorithm, weights are assigned to ice melting importance level, complexity level, and environmental temperature and humidity factors; The matching model is constructed using a weighted summation method. Each indicator value is multiplied by the corresponding weight and then added together to obtain the matching score of each ice melting task. According to the matching scores calculated by the matching model, all ice-melting tasks are sorted. Tasks with higher scores have higher matching priorities, thus generating an ice-melting priority sequence.

[0015] Preferably, the improved genetic algorithm is used to solve the optimal ice melting solution, and the optimization objectives are set in the direction of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time, specifically: Randomly generate a set of candidate solutions as the initial population; Calculate the fitness value of each candidate solution according to the optimization goal; Select the optimal solution as the parent individual according to the fitness value, and update the crossover probability and mutation probability of other individuals; By simulating the genetic evolution and information transmission behavior of the population, the candidate solution is continuously updated iteratively until the termination condition is met; A set of optimal solutions is selected from the final population as the optimal ice melting solution.

[0016] Preferably, the ice melting matching score is specifically: The matching score is determined by the weight of the importance level of ice melting, the weight of the complexity level, the weight of the ambient temperature and humidity, and the weight of the time factor. The weight of the importance level of ice melting corresponds to the influence of the scene type and standard parameters, the weight of the complexity level corresponds to the influence of the ice layer material, the weight of the ambient temperature and humidity corresponds to the influence of the temperature and humidity conditions, and the weight of the time factor corresponds to the urgency of the ice melting demand. Each weight is multiplied by the corresponding level score and then added up to obtain the final matching score.

[0017] Preferably, the optimization goal is specifically: The optimization objectives include maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time. The ice melting efficiency objective is achieved by calculating the magnetothermoacoustic matching degree between the candidate solution and the target parameters. The energy consumption threshold objective is achieved by limiting the energy consumption difference range between the candidate solution and the standard parameters. The action time objective is achieved by reducing the number of algorithm iterations and calculation time. The final optimization objective is the comprehensive optimal of the three sub-objectives.

[0018] Preferably, the present invention further includes an induction-type magnetothermoacoustic ice melting and deicing system for implementing the above-mentioned induction-type magnetothermoacoustic ice melting and deicing method, the system comprising the following modules: Signal acquisition and preprocessing module, used to collect ice surface status signal data in real time, build a multi-dimensional signal processing system, and output standardized ice melting signals; The control tag generation module establishes a parameter database based on magnetothermal acoustic conversion technology. When ice melting needs are detected, it locates the target parameters through a feature matching algorithm and generates a control tag. The historical case information database module is used to dynamically record ice melting records of different scenes, materials, and ice thicknesses; The priority generation module establishes a multi-factor weight distribution model, combining scene type, material properties, ice thickness, and ambient temperature and humidity factors. It uses a decision tree algorithm to assign weights to each indicator, and uses a weighted summation method to build a matching model to calculate the ice melting matching score, thereby generating an ice melting priority sequence. The optimal solution solution module uses an improved genetic algorithm to solve the optimal ice melting solution, with the optimization goals of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time; The feedback tracking module outputs the final ice melting results through the interactive verification platform and continuously tracks the actual ice melting effect feedback.

[0019] Compared with the prior art, the present invention has the following beneficial effects: Through inductive sensors, ice surface status signals are collected in real time, and a multi-dimensional signal processing system is constructed to pre-process the signals through time domain filtering, frequency domain denoising, edge enhancement, etc. At the same time, relevant parameters are screened based on the feature selection algorithm, and a multi-level feature system is constructed. The training is completed using a neural network model. It can comprehensively and accurately obtain information such as the magnetic permeability characteristics, thermal diffusion characteristics, and acoustic impedance characteristics of the ice surface, associate the ice surface position information with the ice melting characteristics and visualize them, making the analysis of the ice surface status more accurate and intuitive, and providing a reliable signal basis for subsequent ice melting operations.

[0020] Based on the magnetothermal acoustic conversion technology, a parameter database is established to collect standard parameters and physical property information of common scenarios, equipment types, and ice materials, and to construct a node-association data structure, which can systematically manage parameter information and association relationships. When the need for ice melting is detected, the feature matching algorithm is used to calculate the feature distance with a dynamic programming algorithm. The target parameter matching range is determined in combination with the ice thickness and material characteristics, and the control parameters are extracted and labels are generated, thereby achieving rapid and accurate positioning of the target parameters and improving the pertinence and timeliness of ice melting operations.

[0021] Build a historical ice-melting case information database to dynamically record ice-melting records in different scenarios, materials, and ice thicknesses. By connecting with the ice-melting equipment management system and the user feedback platform data, continuously update matching results, energy consumption feedback, operation effects and other information. This can make full use of historical experience and provide reference for new ice-melting tasks, making the formulation of ice-melting plans more scientific and reasonable, and continuously optimizing ice-melting effects.

[0022] A multi-factor weight allocation model is established, combining scene type, material properties, ice thickness, and environmental temperature and humidity factors. The decision tree algorithm is used to assign weights to each indicator. The ice melting matching score is calculated through weighted summation and a priority sequence is generated. This model can comprehensively consider multiple factors, reasonably allocate resources, and ensure that ice melting tasks with high importance and complexity are given priority, thereby improving the overall efficiency and resource utilization of ice melting operations.

[0023] An improved genetic algorithm is used to solve the optimal ice-melting solution, with the optimization goals of maximizing ice-melting efficiency, minimizing energy consumption threshold and shortening action time. By randomly generating the initial population, calculating the fitness value, selecting parent individuals and updating the crossover mutation probability, the candidate solution is continuously iterated and updated. It is possible to find the comprehensive optimal ice-melting solution in a relatively short time, significantly improving ice-melting efficiency, reducing energy consumption and shortening action time.

[0024] By outputting the final ice-melting results through the interactive verification platform and continuously tracking the actual ice-melting effect feedback, the ice-melting plan can be adjusted and optimized according to the actual situation, making the system more adaptable and reliable, and maintaining good ice-melting performance in different environments and working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a working principle diagram of the induction-type magnetic thermoacoustic ice melting and de-icing method according to the present invention; Figure 2 Flowchart constructed for the signal preprocessing module; Figure 3 Flowchart established for parameter database; Figure 4 Flowchart for feature matching and control label generation. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0027] See also Figure 1-Figure 4 The present invention provides an ice melting and deicing method based on induction magnetic thermoacoustic, and the specific implementation steps are as follows: Collect ice surface status signal data in real time, build a multi-dimensional signal processing system, and output standardized ice melting signals through the signal preprocessing module.

[0028] A parameter database is established based on magnetothermal acoustic conversion technology. When ice melting needs are detected, the target parameters are located and control tags are generated through feature matching algorithms.

[0029] Build a historical ice melting case information database to dynamically record ice melting records in different scenarios, materials, and ice thicknesses.

[0030] A multi-factor weight distribution model is established. By combining scene type, material properties, ice thickness, and ambient temperature and humidity factors, a decision tree algorithm is used to assign weights to each indicator. A matching model is constructed using a weighted summation method to calculate the ice melting matching score, thereby generating an ice melting priority sequence.

[0031] An improved genetic algorithm is used to solve the optimal ice melting scheme, and the optimization objectives are set in the direction of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time.

[0032] The final ice-melting results are output through the interactive verification platform, and the actual ice-melting effect feedback is continuously tracked.

[0033] Example 1: In the process of collecting ice surface status signal data in real time, constructing a multi-dimensional signal processing system, and outputting a standardized ice-melting signal through a signal preprocessing module, an inductive sensor is first used to acquire signal data from the ice surface area. The inductive sensor here can be selected based on the actual application scenario. For example, in the ice-melting scenario of power transmission lines, an inductive coil sensor arranged around the line can be used, and in the ice-melting scenario of large structures such as bridges, an embedded magneto-thermo-acoustic composite sensor can be used. After receiving the raw signal from the sensor, the signal acquisition terminal performs time-domain filtering on it. By setting a specific time window and filter coefficient, time-domain noise caused by environmental vibration, electromagnetic interference, etc. is removed. Frequency-domain denoising is then performed. The time-domain signal is converted to the frequency domain using a Fourier transform, and high-frequency interference signals and low-frequency environmental noise are filtered out using a bandpass filter. Edge enhancement preprocessing is then performed. Methods such as differential operators are used to highlight edge features in the signal, thereby effectively removing environmental interference clutter.

[0034] Based on the physical properties of the ice surface, such as its density and crystal structure, as well as its magnetothermoacoustic response (i.e., the thermal and acoustic effects produced by ice in a magnetic field), a feature selection algorithm is used to select parameters closely related to ice melting from the preprocessed signal as candidate features. Feature selection algorithms, such as the Relief-F algorithm, determine candidate features by calculating the correlation between each parameter and the ice melting effect. Candidate features are categorized into three types: magnetic permeability, thermal diffusion, and acoustic impedance. The magnetic permeability reflects the magnetization of the ice surface in a magnetic field and can be determined by measuring its relative magnetic permeability. The thermal diffusion reflects the thermal conductivity of the ice surface and can be determined through thermal conduction experiments to determine the thermal diffusion coefficient. The acoustic impedance reflects the propagation and reflection characteristics of the ice surface to sound waves and can be measured using acoustic wave detection equipment.

[0035] When constructing a multi-level feature system, the overall ice melting matching degree is used as the top-level feature, which is then decomposed into several first-level features, such as the magnetothermal acoustic response intensity and feature stability. Each first-level feature is further subdivided into several second-level features. For example, the magnetothermal acoustic response intensity can be subdivided into magnetic field intensity response features, temperature change features, etc., and feature stability can be subdivided into feature consistency at different time points and feature repeatability under different environmental conditions. Based on the multi-level feature system, a neural network model is used to complete the training of the signal preprocessing module. The neural network model can use a convolutional neural network (CNN) or a recurrent neural network (RNN). By inputting a large amount of ice surface signal data and the corresponding ice melting effect labels, the model is trained and the model weights and biases are adjusted to enable it to accurately preprocess the input signal.

[0036] The location information of the ice surface is correlated with the ice melting characteristics of the preprocessed signal to establish a location-feature mapping. Specifically, each sensor collecting signals is assigned a unique location identifier, such as the tower number and line segment on a transmission line, or the specific location coordinates on a bridge. The signal characteristics collected by the sensor are then associated with the location identifier to form a location-feature mapping table. Signal analysis software is used as the magnetothermoacoustic analysis platform. This software can be based on commonly available professional signal analysis software such as MATLAB and LabVIEW. Areas are marked on the software interface according to the actual location of the ice surface. For example, the locations of each section of the transmission line are marked on a 3D model, and the locations of different areas are marked on a plan view of a bridge. Different brightness levels are used for visualization based on feature differences: areas with large feature differences are highlighted, while areas with small feature differences are dimmed. For example, if the magnetic permeability characteristics of a certain area differ significantly from those of other areas, this area will be highlighted on the interface to provide operators with an intuitive understanding of the distribution of ice surface features.

[0037] Throughout the implementation process, the signal acquisition terminal must possess high-speed data acquisition and real-time processing capabilities to ensure timely acquisition and preprocessing of ice surface status signals. The placement of inductive sensors must consider ice surface coverage and signal acquisition uniformity to avoid signal acquisition blind spots. Training the neural network model requires a large amount of sample data, encompassing ice surface signals from diverse scenarios, materials, and ice thicknesses, to ensure model generalization and accuracy. Establishing position-feature mapping relationships requires accurate recording of sensor location information and corresponding signal features to avoid mapping errors. The selection of signal analysis software must consider its functionality, operational ease, and visualization capabilities, meeting the requirements of magnetothermoacoustic analysis. Through these steps, real-time acquisition of ice surface status signal data is achieved, a multi-dimensional signal processing system is constructed, and standardized ice-melting signals are output through the signal preprocessing module, providing reliable signal support for subsequent ice-melting and de-icing operations.

[0038] Example 2: When establishing a parameter database based on magnetothermoacoustic conversion technology, standard parameter data for common application scenarios, equipment types, and ice layer materials are first collected. Common application scenarios include transmission lines, bridges, communication base stations, etc., and equipment types include different models of transmission line towers, bridge steel structures, base station antennas, etc. The ice layer materials include pure ice, mixed ice crystals, snow-covered ice layers, etc. The collected standard parameter data include basic parameters such as magnetic saturation value, thermal conductivity value, and sound velocity value. For example, the magnetic saturation value of pure ice is approximately within a certain range, the value of thermal conductivity value at a specific temperature, and the performance of sound velocity value at different ice layer densities. The physical property information of the parameters is obtained, where the medium density is calculated by measuring the mass and volume of the ice layer sample, the interface bonding strength is determined by tensile test or shear test, and the surface roughness parameters are measured using a surface profilometer.

[0039] Parameters are abstracted into data nodes, each containing information such as the parameter's name, value, and unit. The magneto-thermoacoustic relationships between parameters are abstracted into data associations. For example, changes in the magnetic saturation value affect changes in the thermal conductivity value; this influence is a data association. A node-association data structure is constructed, with data nodes as vertices and data associations as edges, forming a network structure that clearly represents the relationships between parameters. A database architecture is established, including the definition of data tables, such as parameter tables, attribute tables, and association tables. Field settings are determined based on actual needs. For example, a parameter table contains fields such as parameter ID, parameter name, parameter value, and unit. Data relationships are established through foreign key associations to store parameter information, attribute information, and associations.

[0040] The preprocessed data is imported into the magnetothermoacoustic analysis software database. During the import process, the data needs to be verified to ensure its accuracy and completeness, and to avoid missing or incorrect data, thereby establishing a parameter database. When locating the target parameters and generating control labels through the feature matching algorithm, based on the established parameter database, each parameter node is regarded as a sample point in the data set. Each sample point contains multiple characteristic attributes, and the magnetothermoacoustic correlation between parameters is regarded as the similarity between samples. That is, the higher the degree of correlation between two sample points, the greater their similarity.

[0041] Using the signal characteristics of the area to be melted as the query point, a dynamic programming algorithm is used to calculate the characteristic distance to each sample point. The dynamic programming algorithm decomposes complex problems into sub-problems and gradually solves the optimal solution. When calculating the characteristic distance, distance metrics such as Euclidean distance and Manhattan distance can be used. When the sample point with the highest matching degree is traversed, the characteristic information of the sample point is recorded, including the parameter value corresponding to the sample point, the application scenario, the ice layer material, etc. The calculated characteristic distance is analyzed and combined with the ice layer thickness and material property information to determine the matching range of the target parameter. For example, when the characteristic distance is less than a certain threshold, the sample point is considered to be within the matching range of the target parameter.

[0042] The magnetic saturation value, thermal conductivity value, and sound velocity value parameters of the target parameters are extracted from the parameter database to generate control parameters. The control parameters need to be adjusted according to the actual ice melting requirements. For example, appropriate corrections may be required for different ambient temperatures. The control parameters of the target parameters, the scenario type, and the ice layer material are combined into a label. The label format can be designed according to the requirements of the magnetothermoacoustic analysis software, such as using JSON or XML format, and visually displayed on the magnetothermoacoustic analysis software interface. On the software interface, labels can be displayed in the form of lists, charts, etc., making it easier for operators to view and understand.

[0043] When establishing a parameter database, it is necessary to ensure that the collected data is representative and reliable. For data on special scenarios or rare ice materials, experiments or actual measurements are necessary to avoid using outdated or inaccurate data. When constructing a node-association data structure, it is necessary to fully consider the various associations between parameters to ensure that the data structure accurately reflects the physical laws of magnetothermal acoustic conversion technology. When using a dynamic programming algorithm to calculate feature distances, it is necessary to reasonably set the algorithm's parameters, such as the distance measurement method and threshold, to improve the accuracy and efficiency of feature matching. When generating control labels, it is necessary to ensure that the label information is complete and accurate and can be correctly identified and processed by the magnetothermal acoustic analysis software. Through the above steps, a parameter database based on magnetothermal acoustic conversion technology is established, and the target parameters are located through a feature matching algorithm to generate control labels, providing accurate parameter support and control basis for subsequent ice melting and de-icing work.

[0044] Example 3: Build a historical ice-melt case database to dynamically record ice melt records for different scenarios, materials, and ice thicknesses. First, obtain the key dimensions for building the database. These dimensions include ice melt records for different scenario types, ice materials, and ice thicknesses. Scenario types can be categorized into transmission line scenarios, bridge scenarios, building roof scenarios, and communications equipment scenarios. Each scenario type can be further subdivided. For example, transmission line scenarios can be categorized by voltage level into high voltage and ultra-high voltage. Ice materials can be categorized into pure ice, mixed ice (e.g., a mixture of ice and snow), salt-coated ice, and ice containing impurities. Ice thickness can be divided into different intervals based on the actual measurement range, such as 0-10mm, 10-20mm, and 20-30mm.

[0045] At the same time, specific information fields are planned for each dimension, including application scenarios, equipment types, ice layer materials, standard parameters, actual control parameters, and user operation feedback. Application scenarios must record the specific usage environment in detail, such as the specific section of a transmission line or the name and location of a bridge. Equipment types must be specified to specific equipment models, such as a certain type of transmission tower or a certain type of bridge support. In addition to recording the general category of ice layer materials, the specific characteristics of the material, such as impurity content and density, must also be recorded. Standard parameters include the corresponding magnetic saturation value, thermal conductivity value, and sound velocity value for the scenario and material. Actual control parameters are the magnetothermoacoustic control parameters used in the actual ice melting process, such as the applied magnetic field strength, heating power, and sound wave frequency. User operation feedback includes the operator's subjective evaluation of the ice melting effect and any problems encountered during operation.

[0046] By connecting the de-icing equipment management system with the user feedback platform, the matching results, energy consumption feedback, and operation effect information of historical de-icing cases are continuously updated. The de-icing equipment management system stores the equipment's operating data, including the time of each de-icing operation, the parameters used, the equipment's operating status, etc.; the user feedback platform collects feedback information from operators and maintenance personnel. During docking, the data in the two systems are synchronized and integrated through the data interface to ensure that the data in the historical de-icing case information library can be updated in a timely manner. For example, when a de-icing operation is completed, the de-icing equipment management system will automatically upload the operation data of the operation to the information library, and the feedback information submitted by the operator on the user feedback platform will also be synchronized to the information library.

[0047] Build a historical ice melting case database based on key dimensions and data fields. This database can be stored in a relational database such as MySQL or Oracle. Create multiple data tables within the database, corresponding to different dimensions and information fields. By establishing relationships between tables, efficient data storage and querying can be achieved. For example, create a scenario table, a material table, a thickness table, and a case table. The case table is linked to the scenario, material, and thickness tables via foreign keys to store specific information about each case.

[0048] A multi-factor weighting model was established. By combining scenario type, material properties, ice thickness, and ambient temperature and humidity, a decision tree algorithm was used to assign weights to each indicator. A weighted summation method was used to construct a matching model and calculate the ice melting match score, thereby generating an ice melting priority sequence. Ice melting importance levels were assigned based on scenario type, standard parameters, and ice thickness. High-priority scenarios or requirements with clear standard parameters were assigned a high importance level. For example, in the transmission line scenario, the need for ice melting on high-voltage lines, due to its significant impact on power supply and the relatively clear standard parameters, were assigned a high importance level. Low-priority scenarios or requirements without clear parameters were assigned a low importance level. For example, the need for ice melting on building roofs in remote areas, which is less frequently used and has unclear standard parameters, were assigned a low importance level.

[0049] Ice layers are categorized into metal surface ice, concrete surface ice, and glass surface ice. Metal surface ice has a higher level of complexity in melting because metal has good thermal conductivity and the ice layer has a stronger bond with the metal surface. Therefore, more factors need to be considered when melting the ice, such as avoiding damage to the metal surface. Glass surface ice has a lower level of complexity because the glass surface is relatively smooth, resulting in a weaker bond between the ice layer and the glass, making melting easier. Ambient temperature and humidity are categorized into standard temperature and humidity, high temperature and high humidity, and low temperature and low humidity. Standard temperature and humidity refers to a normal range of temperature and humidity. High temperature and high humidity may affect the performance of ice melting equipment, while low temperature and low humidity may cause the ice layer to become harder, increasing the difficulty of melting.

[0050] Based on the decision tree algorithm, weights are assigned to the importance level of ice melting, complexity level, and environmental temperature and humidity factors. The decision tree algorithm builds a tree structure by learning from historical data to classify and predict new data. When constructing the decision tree, the ice melting effect is used as the target variable, and the importance level of ice melting, complexity level, environmental temperature and humidity factors are used as feature variables. The information gain of each feature variable is calculated to determine its importance in the decision tree, thereby obtaining the weight of each indicator. Suppose the weight of the importance level of ice melting is , the weight of the complexity level is , the weight of the environmental temperature and humidity factors is , the time factor weight is , each weight satisfies ,in Reflects the impact of scene type and standard parameters on ice melting tasks, Reflect the influence of ice material, Indicates the effect of ambient temperature and humidity conditions, The urgency of the ice melting needs.

[0051] The matching model is constructed by weighted summation. Each indicator value is multiplied by the corresponding weight and then added together to obtain the matching score of each ice melting task. Assume that the ice melting importance level score is , the complexity level score is , the environmental temperature and humidity factor score is , the time factor score is , then the matching score The calculation formula is:

[0052] in, Assign values ​​based on the importance of ice melting. A high importance level can be assigned a higher score, such as 80-100 points, and a low importance level can be assigned a lower score, such as 0-40 points. The value is assigned based on the complexity level of the ice layer material. The metal surface has a high complexity level of ice, which can be assigned a value of 60-100 points, while the glass surface has a low complexity level of ice, which can be assigned a value of 0-40 points. According to the classification and assignment of ambient temperature and humidity, the standard temperature and humidity environment is assigned a value of 40-60 points, and the high temperature and high humidity or low temperature and low humidity environment is assigned a value of 0-40 points or 60-100 points according to the specific situation; Assign points based on the urgency of the ice melting demand, with urgent needs assigned a score of 80-100 points and non-urgent needs assigned a score of 0-40 points.

[0053] All ice-melting tasks are sorted based on the matching scores calculated by the matching model. Tasks with higher scores are assigned higher matching priorities, thus generating an ice-melting priority sequence. During the sorting process, common sorting algorithms such as bubble sort and quick sort can be used to ensure the accuracy and rationality of the priority sequence.

[0054] When constructing a historical ice-melting case information database, it is necessary to ensure the integrity and accuracy of the data, and to supplement and correct missing data in a timely manner to avoid data quality issues affecting subsequent analysis and decision-making. When dividing the importance and complexity levels of ice melting, it is necessary to fully consider the needs and characteristics of the actual application scenarios to ensure the rationality of the level division. When using the decision tree algorithm to assign weights, it is necessary to use enough historical data for training to improve the accuracy and reliability of the weights. When calculating the matching score, it is necessary to ensure that the assignment standards of the scores of each indicator are consistent to avoid deviations caused by human factors. Through the above steps, the construction of a historical ice-melting case information database and the establishment of a multi-factor weight distribution model were realized, providing a scientific basis and method for the prioritization of ice-melting tasks.

[0055] Example 4: During the implementation of the multi-factor weight distribution model, the ice layer material must be classified and its corresponding complexity level determined. For example, when dealing with ice on a metal surface, due to the excellent thermal conductivity of metal itself, the ice layer often bonds tightly to the metal surface. During the ice melting process, not only must the melting efficiency of the ice layer be considered, but also damage to the metal substrate caused by magneto-thermo-acoustic effects must be avoided. For example, ice formed on the steel surface of a transmission line tower. In this case, the ice melting complexity level for ice on a metal surface is set to high. However, the situation is different for ice on a concrete surface. Concrete has a large surface porosity, and the ice layer's bond to the concrete surface is relatively weak. However, concrete structures may have defects such as cracks. During ice melting, it is necessary to prevent moisture penetration and structural damage. The complexity level of ice on a glass surface is between that of metal and glass. For example, ice on glass surfaces, such as the glass panels of bridge guardrails or glass windows of buildings, has a smooth surface and low adhesion strength to the glass. During ice melting, the focus is on ice melting, without excessive consideration of substrate damage. Therefore, the complexity level of ice on glass surfaces is classified as low.

[0056] The classification of ambient temperature and humidity also needs to be combined with specific scenarios. The standard temperature and humidity environment can be set between 0°C and 10°C and between 40% and 60% relative humidity. Under these conditions, the physical properties of the ice layer are relatively stable, and the ice melting operation is less affected by environmental factors. In high temperature and high humidity environments, such as rainy areas in summer, the temperature may reach above 25°C and the relative humidity exceeds 80%. At this time, the ice layer may be in a semi-melted state, and water evaporation and equipment heat dissipation must be considered during the ice melting process. In low temperature and low humidity environments, such as the extremely cold northern regions, where the temperature is below -10°C and the relative humidity is below 30%, the ice layer will become harder, and the transfer efficiency of magneto-thermal-acoustic energy may be affected. Both of these environments are non-standard temperature and humidity environments and need to be reflected in the weight distribution.

[0057] The decision tree algorithm uses data from a database of historical ice-melting cases to assign weights to each factor. For example, when dealing with ice-melting tasks on a high-voltage section of a transmission line, this scenario is frequently used and has clear standard parameters, so the importance of ice-melting is determined to be high. The ice layer is made of metal, which has a high complexity level. The ambient temperature and humidity are within the standard range, so the corresponding weight distribution tends to favor the importance and complexity levels. The decision tree algorithm determines the weight ratio of each factor by learning from a large number of similar cases. For example, in a particular training run, the weight of the importance of ice-melting is 0.4, the weight of the complexity level is 0.3, the weight of the ambient temperature and humidity is 0.2, and the weight of the time factor is 0.1. These weights are continuously updated and adjusted based on new case data.

[0058] When building a matching model to calculate the ice melting matching score, we use a specific ice melting task as an example. Assume that an ice melting task occurs on a bridge on a main road in a city. This bridge is a key location at a transportation hub, and the scenario type is a bridge. Due to the frequent traffic, it is a high-frequency scenario. Furthermore, the bridge management department has established clear standard ice melting parameters, so the ice melting importance score is 90. The ice layer is made of concrete surface ice. Considering the safety of the bridge structure, the ice melting complexity score is 60. The ambient temperature and humidity at the time were low and low, at -15°C and 20% relative humidity. In these conditions, the ice layer is hard, making ice melting more difficult. Therefore, the ambient temperature and humidity factor score is 30. Furthermore, due to the urgent need for ice melting due to an important transportation event the next day, the time factor score is 90.

[0059] According to the matching model, each indicator value is multiplied by its corresponding weight and then added together to calculate the matching score. If the weights for each factor are 0.4 for ice melting importance, 0.3 for complexity, 0.2 for ambient temperature and humidity, and 0.1 for time, the matching score for this task is: 90 × 0.4 + 60 × 0.3 + 30 × 0.2 + 90 × 0.1 = 36 + 18 + 6 + 9 = 69. Another ice melting task occurs at a communication base station in a mountainous area. The scenario type is communication equipment, a low-frequency use scenario with unclear standard parameters. The ice melting importance score is 40; the ice layer is made of glass surface ice, with a complexity score of 20; the ambient temperature and humidity are within the standard range, with a score of 50; the ice melting requirement is routine maintenance, with a time score of 30. Assuming the same weight distribution as in the previous example, the matching score is: 40×0.4+20×0.3+50×0.2+30×0.1=16+6+10+3=35 points.

[0060] The calculated matching scores are used to rank ice melting tasks. The bridge ice melting task, with a score of 69, is higher than the communication base station's score of 35. Therefore, the bridge ice melting task has a higher priority and should be implemented first. In practice, multiple ice melting tasks may exist simultaneously. A matching score must be calculated and ranked for each task to generate a priority sequence. For example, if there are three tasks simultaneously: Task A (matching score 75), Task B (matching score 55), and Task C (matching score 80), the priority sequence is Task C, Task A, and Task B, and ice melting operations will be performed in this order.

[0061] When calculating the ice melting match score, the relationship between each weight and the corresponding level score needs to be dynamically adjusted based on actual conditions. For example, in cold winter regions, where low temperatures and low humidity are common, the weight of the ambient temperature and humidity factors may be appropriately increased to more accurately reflect the environmental impact of the ice melting task. The weight of the time factor corresponds to the urgency of the ice melting need. For example, if sudden disaster weather causes rapid ice accumulation, the time factor weight will be significantly increased, moving such tasks forward in the priority sequence.

[0062] Throughout the implementation process, it is necessary to continuously collect new ice-melting case data, continuously enrich the historical ice-melting case information database, provide more training samples for the decision tree algorithm, and make the weight distribution more accurate. At the same time, the matching model should be regularly evaluated and optimized to ensure that it can reasonably calculate the ice-melting matching score in different scenarios and environments and generate an accurate priority sequence. For example, historical cases should be sorted out every quarter and the weights of the decision tree model should be updated to adapt to the environmental differences brought about by seasonal changes and the needs of newly emerging ice-melting scenarios. In this way, the dynamic optimization of the multi-factor weight distribution model and the accurate calculation of the matching score can be achieved, providing a scientific and reasonable basis for the priority sorting of ice-melting tasks and ensuring the efficient implementation of ice-melting and de-icing work.

[0063] Example 5: When using an improved genetic algorithm to solve the optimal ice melting solution, it is first necessary to randomly generate a group of candidate solutions as the initial population. The structural design of the candidate solution needs to match the parameter system of the ice melting solution. For example, each candidate solution can be represented as a vector containing magnetothermoacoustic control parameters, including parameters such as magnetic field intensity, heating power, acoustic wave frequency, and action time. The size of the initial population is determined according to the complexity of the problem and is usually set to 50 to 200 candidate solutions to ensure the diversity of the population. When generating the initial population, the value range of each parameter is determined based on the standard parameters in the parameter database and actual ice melting experience. For example, the value range of the magnetic field intensity can be set to 100mT to 500mT, and the value range of the heating power can be set to 500W to 2000W to ensure that the initial candidate solution is within a reasonable range.

[0064] The fitness value of each candidate solution is calculated based on the optimization objectives, which include maximizing ice-melting efficiency, minimizing the energy consumption threshold, and shortening the action time. The ice-melting efficiency target is achieved by calculating the magneto-thermo-acoustic matching degree between the candidate solution and the target parameters. The higher the magneto-thermo-acoustic matching degree, the greater the ice-melting efficiency potential of the candidate solution. The energy consumption threshold target is achieved by limiting the energy consumption difference range between the candidate solution and the standard parameters. The smaller the energy consumption difference, the more it meets the energy-saving requirements. The action time target is achieved by reducing the number of algorithm iterations and calculation time, and is also related to the action time parameter set in the candidate solution. The design of the fitness function requires comprehensive consideration of these three sub-goals. For example, the ice-melting efficiency matching degree can be used as a positive correlation factor, and the energy consumption difference and action time as negative correlation factors. The fitness value of each candidate solution is obtained through weighted summation.

[0065] Based on the fitness value, the optimal solution is selected as the parent individual, and the crossover and mutation probabilities of the remaining individuals are updated. Candidate solutions with higher fitness values ​​have a greater probability of being selected as the parent. Common selection methods include roulette wheel selection and tournament selection. For example, in roulette wheel selection, the probability of each candidate solution being selected is proportional to its fitness value. The higher the fitness value, the greater its proportion in the roulette wheel, and the higher its probability of selection. After the parent individual is selected, the crossover and mutation probabilities of the remaining individuals are dynamically adjusted based on the parent's fitness value. Generally, individuals with higher fitness values ​​have lower crossover and mutation probabilities to preserve excellent genes, while individuals with lower fitness values ​​have higher crossover and mutation probabilities to increase population diversity and explore new solution spaces.

[0066] By simulating the genetic evolution and information transfer behavior of a population, candidate solutions are continuously iterated and updated until the termination condition is met. The genetic evolution process includes crossover and mutation operations. Crossover involves exchanging the genes of two parent individuals to generate new offspring individuals. The location and method of crossover are determined by the characteristics of the parameters. For example, arithmetic crossover can be used for continuous parameters, and single-point or multi-point crossover can be used for discrete parameters. Mutation randomly changes the genes of individuals to prevent the population from falling into a local optimal solution. The probability and magnitude of mutation are adjusted according to the parameter value range and the optimization process. During the iterative process, the optimal solution and average fitness value of each generation are recorded. The iteration process is terminated when the number of iterations reaches the preset maximum value or when the optimal solution no longer changes significantly over several consecutive generations.

[0067] A set of optimal solutions is selected from the final population as the optimal ice-melting solution. The optimal solution in the final population is the candidate solution with the highest fitness value after multiple generations of evolution. This solution comprehensively considers factors such as ice-melting efficiency, energy consumption, and action time, and is expected to achieve good results in actual ice-melting operations. For example, the optimal ice-melting solution might have a magnetic field strength of 300mT, a heating power of 1200W, an acoustic frequency of 20kHz, and an action time of 30 minutes. This solution strikes a good balance between ice-melting efficiency, energy consumption, and action time.

[0068] To achieve the specific ice-melting efficiency target within the optimization objective, the magnetothermoacoustic parameters in the candidate solution must be compared with the target parameters to calculate the degree of match between the two. The target parameters are derived from a parameter database and are determined based on factors such as the material, thickness, and scenario type of the ice layer in the area to be melted. For example, for ice layers on metal surfaces, the target magnetic saturation value is within a specific range, and the thermal conductivity and speed of sound values ​​also have corresponding standards. The degree of match between the candidate solution parameters and the target parameters is calculated to determine the degree of ice-melting efficiency. The smaller the deviation, the higher the degree of match, and the better the ice-melting efficiency target is achieved.

[0069] Achieving the energy consumption threshold target requires calculating the theoretical energy consumption of the candidate solution during the ice melting process and comparing it with the energy consumption under standard parameters, limiting the difference between the two to a certain range. The energy consumption under standard parameters is obtained from data statistics in a historical ice melting case database. For example, for ice layers of certain scenarios and materials, the standard energy consumption is 1.5 kW·h per square meter of ice. The theoretical energy consumption of the candidate solution is calculated based on its parameters and the characteristics of the ice layer. For example, the energy consumption is calculated by multiplying the heating power by the action time. This energy consumption is then adjusted based on the area and thickness of the ice layer to ensure that the energy consumption of the candidate solution does not exceed a certain percentage of the standard energy consumption, such as 120%.

[0070] The action time goal is achieved by optimizing the algorithm's iterative process to reduce computational time, for example by employing more efficient crossover and mutation operations to lower the algorithm's time complexity. Furthermore, the action time parameters in the candidate solutions are controlled to minimize the action time while ensuring ice melting efficiency. For example, by adjusting the magnetothermoacoustic parameters, the action time can be shortened from 40 minutes to 30 minutes while ensuring complete ice melting, thus achieving the action time goal.

[0071] When implementing an improved genetic algorithm, it's important to properly set the algorithm's initial parameters, such as the initial population size, crossover probability, mutation probability, and number of iterations. These parameters can affect the algorithm's convergence speed and solution quality. For example, a too small initial population size can lead to insufficient diversity and a tendency to fall into a local optimum. A too high crossover probability can destroy optimal gene combinations, while a too low crossover probability can affect the population's evolutionary speed. Therefore, in practical applications, the algorithm parameters need to be debugged and optimized based on the specific ice melting problem and historical data to achieve optimal solutions.

[0072] The final ice-melting results are output through the interactive verification platform, and the actual ice-melting effect feedback is continuously tracked. The interactive verification platform provides a human-computer interaction interface where operators can view the specific parameters and expected effects of the optimal ice-melting solution. At the same time, the platform will output the solution to the ice-melting equipment to guide the operation of the equipment. After the ice-melting operation is completed, the platform will collect the equipment's operating data and the operator's feedback information, including actual ice-melting time, energy consumption, ice melting conditions, etc., and compare and analyze this feedback information with the data in the historical ice-melting case information library to evaluate the actual effect of the optimal ice-melting solution. If it is found that the solution has deficiencies in certain aspects, such as excessive energy consumption or too long ice-melting time, this feedback information will be added as new case data to the historical case library to provide a reference for solving the optimal ice-melting solution next time, thereby achieving continuous optimization of the ice-melting solution.

[0073] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0074] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The ice melting and deicing method based on induction magnetic thermoacoustic is characterized by: The following steps are involved: Collect ice surface status signal data in real time, build a multi-dimensional signal processing system, and output standardized ice melting signals through the signal preprocessing module; A parameter database is established based on magnetothermal acoustic conversion technology. When ice melting needs are detected, the target parameters are located through feature matching algorithms and control tags are generated. Build a historical ice melting case information database to dynamically record ice melting records in different scenarios, materials, and ice thickness; A multi-factor weight distribution model was established. By combining scene type, material properties, ice thickness, and ambient temperature and humidity factors, a decision tree algorithm was used to assign weights to each indicator. A matching model was constructed using a weighted summation method to calculate the ice melting matching score, thereby generating an ice melting priority sequence. An improved genetic algorithm is used to solve the optimal ice melting scheme, with the optimization objectives set in the direction of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time. The final ice-melting results are output through the interactive verification platform, and the actual ice-melting effect feedback is continuously tracked.

2. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 1 is characterized in that: The real-time collection of ice surface status signal data, the construction of a multi-dimensional signal processing system, and the output of a standardized ice melting signal through a signal preprocessing module are specifically as follows: Based on the inductive sensor to obtain the signal data of the ice surface area, the signal acquisition terminal receives the original signal from the sensor and performs time domain filtering, frequency domain denoising, and edge enhancement preprocessing to remove environmental interference clutter; According to the physical properties of the ice surface and the magnetothermoacoustic response law, the parameters closely related to ice melting characteristics are selected as candidate features based on the feature selection algorithm; The candidate features are divided into magnetic permeability features, thermal diffusion features, and acoustic impedance features. A multi-level feature system is constructed, with the overall ice melting matching degree as the top-level feature. The top-level feature is decomposed downward into several first-level features, and each first-level feature is further subdivided into several second-level features. Based on the multi-level feature system, a neural network model is used to complete the signal preprocessing module training; Correlate the ice surface position information with the ice melting characteristics of the pre-processed signal to establish a position-feature mapping relationship; Signal analysis software was selected as the magnetothermoacoustic analysis platform, and areas were marked on the interface according to the actual position of the ice surface. Different brightness levels were used for visualization based on the differences in features. Areas with large feature differences were highlighted, and areas with small feature differences were low-brightness.

3. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 2, characterized in that: The parameter database is established based on the magnetothermoacoustic conversion technology, specifically: Collect standard parameter data for common application scenarios, equipment types, and ice layer materials, including basic parameters such as magnetic saturation value, thermal conductivity value, and sound velocity value; Obtaining physical property information of parameters, including medium density, interface bonding strength, and surface roughness parameters; Parameters are abstracted into data nodes, and the magneto-thermo-acoustic correlations between parameters are abstracted into data associations, thus constructing a node-association data structure. Establish database architecture, including data table definition, field settings and data relationship establishment, storage parameter information, attribute information and association relationships; Import the preprocessed data into the database of the magnetothermoacoustic analysis software to establish a parameter database.

4. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 3 is characterized in that: The target parameters are located and control labels are generated by the feature matching algorithm, specifically: Based on the parameter database, each parameter node is regarded as a sample point in the data set, and the magnetothermoacoustic correlation between parameters is regarded as the similarity between samples. The signal characteristics of the area to be de-ice-melted are used as query points. The characteristic distance to each sample point is calculated using a dynamic programming algorithm. When the sample point with the highest matching degree is found, the characteristic information of the sample point is recorded. Analyze the calculated characteristic distance and combine it with ice thickness and material characteristics to obtain the matching range of the target parameters; Extract the magnetic saturation value, thermal conductivity value, and sound velocity value parameters of the target parameters from the parameter database to generate control parameters; The control parameters of the target parameters, scene type, and ice layer material are combined into a label and visualized on the magnetothermoacoustic analysis software interface.

5. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 4 is characterized in that: The historical ice melting case information database is constructed to dynamically record ice melting records of different scenes, materials, and ice thicknesses, specifically: Obtain the main dimensions for building the information database, including ice melting records of different scene types, ice layer materials, and ice layer thickness; At the same time, specific information fields are planned for each dimension, including application scenarios, equipment types, ice layer materials, standard parameters, actual control parameters, and user operation feedback; By connecting the de-icing equipment management system with the user feedback platform, the matching results, energy consumption feedback, and operation effect information of historical de-icing cases are continuously updated; Build a historical ice melting case information database based on main dimensions and data fields.

6. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 5, characterized in that: The multi-factor weight distribution model is established by combining scene type, material properties, ice thickness, and ambient temperature and humidity factors. A decision tree algorithm is used to assign weights to each indicator. A matching model is constructed using a weighted summation method to calculate the ice melting matching score, thereby generating an ice melting priority sequence, specifically: The importance of ice melting is divided into different levels based on the scenario type, standard parameters, and ice thickness factors. High-frequency scenarios or scenarios with clear standard parameters are assigned a high importance level, while low-frequency scenarios or scenarios without clear parameters are assigned a low importance level. The ice layer material is divided into metal surface ice, concrete surface ice, and glass surface ice. The complexity level of ice melting on metal surface is high, while the complexity level of ice melting on glass surface is low. The ambient temperature and humidity are divided into standard temperature and humidity, high temperature and high humidity environment, and low temperature and low humidity environment; Based on the decision tree algorithm, weights are assigned to ice melting importance level, complexity level, and environmental temperature and humidity factors; The matching model is constructed using a weighted summation method. Each indicator value is multiplied by the corresponding weight and then added together to obtain the matching score of each ice melting task. According to the matching scores calculated by the matching model, all ice-melting tasks are sorted. Tasks with higher scores have higher matching priorities, thus generating an ice-melting priority sequence.

7. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 6, characterized in that: The improved genetic algorithm is used to solve the optimal ice melting solution, and the optimization objectives are set in the direction of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time. Specifically, Randomly generate a set of candidate solutions as the initial population; Calculate the fitness value of each candidate solution according to the optimization goal; Select the optimal solution as the parent individual according to the fitness value, and update the crossover probability and mutation probability of other individuals; By simulating the genetic evolution and information transmission behavior of the population, the candidate solution is continuously updated iteratively until the termination condition is met; A set of optimal solutions is selected from the final population as the optimal ice melting solution.

8. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 7, characterized in that: The ice melting matching score is specifically: The matching score is determined by the weight of the importance level of ice melting, the weight of the complexity level, the weight of the ambient temperature and humidity, and the weight of the time factor. The weight of the importance level of ice melting corresponds to the influence of the scene type and standard parameters, the weight of the complexity level corresponds to the influence of the ice layer material, the weight of the ambient temperature and humidity corresponds to the influence of the temperature and humidity conditions, and the weight of the time factor corresponds to the urgency of the ice melting demand. Each weight is multiplied by the corresponding level score and then added up to obtain the final matching score.

9. The ice melting and deicing method based on induction magnetic thermoacoustic according to claim 8, characterized in that: The optimization objectives are specifically: The optimization objectives include maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time. The ice melting efficiency objective is achieved by calculating the magnetothermoacoustic matching degree between the candidate solution and the target parameters. The energy consumption threshold objective is achieved by limiting the energy consumption difference range between the candidate solution and the standard parameters. The action time objective is achieved by reducing the number of algorithm iterations and calculation time. The final optimization objective is the comprehensive optimal of the three sub-objectives.

10. An induction-type magnetothermoacoustic ice melting and deicing system, used to implement the induction-type magnetothermoacoustic ice melting and deicing method according to any one of claims 1 to 9, characterized in that: Includes the following modules: Signal acquisition and preprocessing module, used to collect ice surface status signal data in real time, build a multi-dimensional signal processing system, and output standardized ice melting signals; The control tag generation module establishes a parameter database based on magnetothermal acoustic conversion technology. When ice melting needs are detected, it locates the target parameters through a feature matching algorithm and generates a control tag. The historical case information database module is used to dynamically record ice melting records of different scenes, materials, and ice thicknesses; The priority generation module establishes a multi-factor weight distribution model, combining scene type, material properties, ice thickness, and ambient temperature and humidity factors. It uses a decision tree algorithm to assign weights to each indicator, and uses a weighted summation method to build a matching model to calculate the ice melting matching score, thereby generating an ice melting priority sequence. The optimal solution solution module uses an improved genetic algorithm to solve the optimal ice melting solution, with the optimization goals of maximizing ice melting efficiency, minimizing energy consumption threshold and shortening action time; The feedback tracking module outputs the final ice melting results through the interactive verification platform and continuously tracks the actual ice melting effect feedback.

Images