Automobile inflator pump power supply control method and system and automobile inflator pump

Abstract

The invention relates to the technical field of automobile inflator pump control, and discloses an automobile inflator pump power supply control method and system and an automobile inflator pump. According to the method, multi-dimensional power supply data of the automobile inflator pump are collected, dynamic feature vectors are extracted, and a global association feature matrix is generated by using a convolutional neural network. And constructing a power supply mode knowledge base, and generating a power supply control strategy sequence and a resource allocation scheme in combination with a genetic algorithm. An improved particle swarm optimization algorithm is adopted to optimize a strategy, abnormity is detected and the strategy is reconstructed based on a fuzzy logic system, and a knowledge graph is constructed to verify the physical compatibility of the strategy. The power supply state can be accurately analyzed, the optimization strategy can be efficiently generated, abnormal conditions can be rapidly handled, the operation stability and the inflation efficiency of the automobile inflation pump are improved, the service life of a battery is prolonged, and energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive accessory air pump control, and particularly to a power control method, system and automotive air pump for an automotive air pump. Background Art

[0002] In the daily use of automobiles, the stability of tire pressure is crucial for driving safety and vehicle performance. As a key device for maintaining tire pressure, the performance of an automotive air pump directly affects the user experience and the safe driving of the vehicle. However, there are many problems in the power control of current automotive air pumps, which seriously restrict their further development and application.

[0003] The power control method of early automotive air pumps was relatively simple, usually only based on basic voltage and current detection for control, and could not comprehensively consider the complex working conditions when the air pump was operating. For example, under different environmental temperature and humidity conditions, the performance of internal electronic components of the air pump would change, thus affecting the power stability and the working efficiency of the air pump. However, traditional control methods often ignored these factors, resulting in the air pump being prone to failure and having a shortened service life in high-temperature and high-humidity environments.

[0004] With the development of technology, although some air pumps have started to pay attention to more operating parameters, there are still deficiencies in data processing and control strategy formulation. When existing air pumps collect multi-dimensional power data, they cannot effectively extract valuable feature information, resulting in inaccurate judgment of the power state. For example, only collecting voltage and current data, but not being able to deeply analyze the power ripple spectrum characteristics, load fluctuation detection characteristics, etc., making it difficult to quickly make a correct response in the face of abnormal power fluctuations, affecting the normal operation of the air pump.

[0005] In addition, the current power control strategy lacks effective utilization of historical data. When encountering similar power usage situations or failures, it cannot quickly draw on past experience for processing, but instead repeatedly tries different control methods, which not only wastes time but may also cause more serious problems due to untimely processing. For example, when the battery health status affects the inflation efficiency, without referring to historical optimization strategies, only blindly adjusting the power output cannot achieve the best matching control of the battery and the air pump, further increasing energy consumption.

[0006] Moreover, when the existing power control system of automotive air pumps faces complex hardware topologies, there are problems with unreasonable control strategies. The energy transfer constraints and hardware resource conflict relationships between various power regulation operations are not fully considered, resulting in the inability of components to work together during actual operation and reducing the overall performance of the air pump. For example, motor speed regulation and battery charging operations may interfere with each other due to resource conflicts, affecting the working efficiency and stability of the air pump. Summary of the Invention

[0007] The purpose of the present invention is to provide a power control method, system, and automotive air pump for an automotive air pump to solve the problems raised in the above background technology.

[0008] To achieve the above purpose, the present invention provides the following technical solution: A power control method for an automotive air pump, the method comprising:

[0009] Collect multi-dimensional power data of the automotive air pump;

[0010] Extract signal features from the multi-dimensional power data to generate dynamic feature vectors for each dimension;

[0011] Map the dynamic feature vectors to a joint analysis space based on a pre-trained convolutional neural network model to generate a global correlation feature matrix;

[0012] Construct a power mode knowledge base, which stores historical power usage patterns, fault cases, optimization strategies, and corresponding execution efficiency data;

[0013] Generate a power control strategy sequence and resource allocation plan for the current air pump through a genetic algorithm based on the global correlation feature matrix and the historical patterns in the knowledge base.

[0014] Preferably, the multi-dimensional power data includes voltage, current, ambient temperature, humidity, inflation rate, battery health status, user preset pressure value, motor speed, and radiator working status;

[0015] The dynamic feature vectors include power ripple spectrum features, load fluctuation detection features, temperature gradient change features, dynamic distribution features of battery internal resistance, and motor torque response delay features.

[0016] Preferably, the generating a power control strategy sequence and resource allocation plan for the current air pump through a genetic algorithm includes:

[0017] Generate an initial candidate control strategy set according to the matching degree between the global correlation feature matrix and the fault patterns in the knowledge base;

[0018] Construct a policy execution dependency graph based on the hardware topology structure of the inflator pump. The nodes in the policy execution dependency graph represent power regulation operations, and the edges represent the energy transfer constraints or hardware resource conflict relationships between operations;

[0019] Combined with real-time power data, iteratively optimize the initial candidate control policy set through an improved particle swarm optimization algorithm, and output the final control sequence and resource allocation result.

[0020] Preferably, the convolutional neural network model includes:

[0021] Use a multi-scale convolutional layer to extract power signal features in different frequency bands;

[0022] Weightedly fuse the features of each frequency band through a channel attention mechanism to generate the global correlation feature matrix.

[0023] Preferably, the improved particle swarm optimization algorithm includes:

[0024] Introduce an adaptive inertia weight adjustment strategy, and the weight is dynamically updated according to the real-time error of inflation efficiency and the battery energy consumption rate;

[0025] Adopt an elite retention mechanism to screen the non-dominated solution set in the iterative process, and enhance the global search ability through a Cauchy mutation operator.

[0026] Preferably, the signal feature extraction adopts a wavelet transform and time-frequency joint analysis method, specifically including:

[0027] Decompose the power signal through a Morlet wavelet basis function to extract the energy entropy features at each scale;

[0028] Perform singular value decomposition on the time-frequency matrix to generate a dynamic feature vector representing the signal mutation and steady-state components.

[0029] Preferably, the method further includes:

[0030] Detect abnormal patterns in real-time power data based on a fuzzy logic system;

[0031] When an abnormality is detected, trigger the dynamic reconstruction of the control policy sequence, and update the resource allocation plan according to the reconstruction result.

[0032] Preferably, the dynamic reconstruction uses a mixed-integer linear programming solver, specifically including:

[0033] Define the hard constraints of the power operation sequence constraint, the battery charge and discharge cycle limit, and the radiator start and stop time window;

[0034] Generate a feasible solution set through a column generation algorithm, and screen the comprehensive optimal solution based on the entropy weight TOPSIS method.

[0035] Preferably, the method further includes:

[0036] Construct a knowledge graph of the power supply component relationship, where the nodes in the graph represent the electrical components or fault types of the inflator pump, and the edges represent the energy coupling relationship or fault conduction path between the components;

[0037] In the policy sequence generation stage, verify the physical compatibility between the policy and the knowledge graph through the subgraph isomorphism detection algorithm.

[0038] Preferably, the present invention further includes a power supply control system for an automotive inflator pump, and the system includes:

[0039] Data acquisition module: used to acquire multi-dimensional power supply data of the automotive inflator pump;

[0040] Feature extraction module: extract signal features from the multi-dimensional power supply data to generate dynamic feature vectors for each dimension;

[0041] Feature correlation module: map the dynamic feature vectors to the joint analysis space based on a pre-trained convolutional neural network model to generate a global correlation feature matrix;

[0042] Knowledge base construction module: construct a power supply mode knowledge base, which stores historical power supply usage patterns, fault cases, optimization strategies, and corresponding execution efficiency data;

[0043] Policy generation module: generate a power supply control policy sequence and resource allocation plan for the current inflator pump through a genetic algorithm according to the global correlation feature matrix and the historical patterns in the knowledge base.

[0044] An automotive inflator pump includes an inflator pump housing main body, an air pump assembly, and a mobile power source. An assembly groove for the mobile power source to be movably arranged therein is provided on the inflator pump housing main body. An electrical interface terminal electrically connected to the air pump assembly is provided in the assembly groove. The mobile power source is electrically connected to the air pump assembly through the electrical interface terminal. A spring piece is provided in the assembly groove. One end of the spring piece is fixed to the bottom of the assembly groove, and the other end is in contact with the mobile power source. A control button is provided on the surface of the inflator pump housing main body, and the bottom end of the control button is connected to a button bracket in contact with the spring piece.

[0045] One side of the air pump assembly is connected to a fan assembly. An air inlet corresponding to the fan assembly is provided on the side wall of the inflator pump housing main body. An air outlet communicating with the air pump assembly is provided on the other side of the inflator pump housing main body.

[0046] A lighting lamp is embedded on the surface of the inflator pump housing main body, and an openable and closable sealing baffle is provided at the opening of the assembly groove.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] At the data acquisition and analysis level, by collecting multi-dimensional power data of the automotive air pump, including information such as voltage, current, ambient temperature, humidity, inflation rate, battery health status, user preset pressure value, motor speed, and radiator working status, it provides a rich data basis for comprehensively understanding the working status of the air pump. The wavelet transform and time-frequency joint analysis method are used to extract signal features from these data, generating dynamic feature vectors in each dimension such as power ripple spectrum features and load fluctuation detection features, which can more accurately grasp the real-time state changes of the power supply. This makes the detection of power supply abnormalities by the system more sensitive. Compared with the traditional method that only relies on basic voltage and current detection, it can detect potential problems in advance, greatly improving the reliability of the air pump operation and reducing the probability of failures caused by power supply problems.

[0049] Based on the pre-trained convolutional neural network model, the dynamic feature vectors are mapped to the joint analysis space to generate a global correlation feature matrix, further exploring the deep connections between data. This matrix comprehensively reflects the global correlation of each dimension feature, providing strong support for formulating precise control strategies subsequently. In this way, the system can more accurately predict the change trend of the power supply state, make adjustments in advance, ensure that the air pump always operates in the best working state, and improve the inflation efficiency.

[0050] In terms of generating control strategies, the constructed power supply mode knowledge base stores historical power supply usage patterns, fault cases, optimization strategies, and corresponding execution efficiency data. Combining the global correlation feature matrix with the historical patterns in the knowledge base, the genetic algorithm is used to generate the power supply control strategy sequence and resource allocation plan for the current air pump. This process fully draws on past experiences, avoids blind attempts, and greatly shortens the time for formulating control strategies. Moreover, the generated strategies can be dynamically adjusted according to the real-time state of the air pump, realizing the refined management of the power supply. For example, when the battery health status is poor, it can automatically adjust the charging and discharging strategies, ensuring the normal operation of the air pump while extending the battery life and reducing energy consumption.

[0051] The application of the improved particle swarm optimization algorithm further optimizes the generation process of the control strategy by introducing an adaptive inertia weight adjustment strategy, an elite retention mechanism, and a Cauchy mutation operator. The adaptive inertia weight is dynamically updated according to the real-time error of the inflation efficiency and the battery energy consumption rate, enabling the algorithm to flexibly adjust the search strategy according to the actual situation when searching for the optimal solution, thereby improving the convergence speed and solution quality of the algorithm. The elite retention mechanism ensures that excellent solutions are not lost during the iteration process, and the Cauchy mutation operator enhances the global search ability to prevent the algorithm from falling into local optimal solutions. This makes the generated control strategy and resource allocation scheme more reasonable, effectively enhancing the overall performance of the inflator pump.

[0052] In addition, an abnormal pattern detection is performed on the real-time power supply data based on a fuzzy logic system, and when an abnormality is detected, a dynamic reconstruction of the control strategy sequence is triggered through a mixed-integer linear programming solver while updating the resource allocation scheme. This abnormal handling mechanism can quickly respond to power supply anomalies, minimize the impact of the anomalies on the operation of the inflator pump, and ensure the stable operation of the inflator pump. The constructed knowledge graph of the power component relationships and the verification of the physical compatibility of the strategy through the subgraph isomorphism detection algorithm during the strategy sequence generation stage ensure that the control strategy conforms to the hardware structure and physical principles of the inflator pump, avoiding hardware damage or performance degradation caused by unreasonable strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is the working principle diagram of the power control method for the vehicle inflator pump described in the present invention;

[0054] Figure 2 is a schematic diagram of real-time power supply data anomaly detection and control strategy reconstruction;

[0055] Figure 3 is a schematic diagram of dynamic reconstruction solution and optimization;

[0056] Figure 4 is a schematic diagram of the construction of the power component relationship knowledge graph and strategy verification;

[0057] Figure 5 is a schematic diagram of the structure of the vehicle inflator pump;

[0058] Figure 6 is a schematic diagram of the internal structure of the vehicle charging pump.

[0059] In the figure: 1. Main body of the inflator pump housing; 2. Air pump assembly; 3. Mobile power supply; 4. Sealing baffle; 5. Control button; 6. Assembly groove; 7. Spring piece; 8. Fan assembly; 9. Air inlet; 10. Air outlet; 11. Lighting lamp; 12. Electrical interface terminal. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0061] Please refer to Figures 1-6 , the present invention provides a power control method for an automotive air pump, specifically including the following steps:

[0062] Using a variety of sensors and detection devices, various power-related data during the operation of the automotive air pump are obtained in real time. These data involve multiple dimensions and cover various state information of the power supply during the operation of the air pump, providing basic data support for subsequent precise analysis and control.

[0063] Adopting specific signal processing techniques, representative features are extracted from the multi-dimensional power data collected, and these features are integrated to generate dynamic feature vectors for each dimension to more effectively reflect the changing trend of the power supply state.

[0064] Using a convolutional neural network model that has been trained with a large amount of data to process the generated dynamic feature vectors, mapping them to a joint analysis space, and generating a feature matrix in this space that can reflect the global correlation relationship between the features of each dimension, thereby mining the deep connections in the data.

[0065] Establish a power mode knowledge base for storing historical power usage patterns, past fault cases, optimization strategies for different situations, and the corresponding execution efficiency data of these strategies. This knowledge base is an important reference basis for generating power control strategies in the future.

[0066] Compare and analyze the generated global correlation feature matrix with the historical patterns in the knowledge base, and with the help of the genetic algorithm, combined with the actual situation of the current air pump, generate a set of power control strategy sequences suitable for the current working conditions and the corresponding resource allocation plan to achieve optimized control of the power supply of the automotive air pump and improve the working efficiency and performance of the air pump.

[0067] The following further illustrates the present invention in conjunction with Embodiments 1 to 6:

[0068] Embodiment 1:

[0069] In actual operation, the collection of multi-dimensional power supply data is crucial. Among them, voltage data can be collected through a high-precision voltage sensor, which can monitor the voltage values ​​of the air pump power input and output in real time, providing a basis for judging the stability of the power supply. Current data is obtained using a current transformer to accurately measure the current in the circuit in order to understand the power consumption of the air pump when it is working.

[0070] The ambient temperature and humidity are collected with the help of temperature and humidity sensors. These sensors are installed in appropriate locations of the air pump and can sense the changes in ambient temperature and humidity in real time. This is because ambient temperature and humidity will affect the performance and power supply stability of the air pump. For example, in a high temperature and high humidity environment, the performance of electronic components may decline, causing power supply abnormalities.

[0071] The inflation rate is a key indicator to measure the efficiency of the air pump. By installing a flow sensor on the inflation pipe, the volume of gas filled per unit time can be accurately measured to obtain the inflation rate. The battery health status is monitored by the battery management system (BMS). The BMS detects the battery voltage, current, temperature, and charge and discharge times, and uses a specific algorithm to estimate the remaining capacity, internal resistance and other key information of the battery to evaluate the health of the battery.

[0072] The user preset pressure value is the inflation target value set by the user according to actual needs, which can be input through the operation panel on the air pump or a mobile device connected to it. The motor speed is obtained through the speed sensor installed on the motor shaft. The change in speed reflects the working state and load condition of the motor. The working state of the radiator is determined by detecting the fan speed, temperature and other information of the radiator to determine whether the radiator is working normally and whether it can dissipate heat effectively.

[0073] Signal feature extraction is performed on the collected data. Taking the power ripple spectrum feature as an example, a specific signal processing algorithm is used to analyze the voltage signal to obtain the frequency component and amplitude distribution of the power ripple. The load fluctuation detection feature monitors and analyzes the changes in the current signal to determine whether the load is stable. When the load fluctuates, relevant feature information is captured in a timely manner.

[0074] The extraction of temperature gradient change characteristics is to perform time series analysis on the temperature data collected by the temperature and humidity sensor, calculate the rate of change of temperature at adjacent moments, and obtain the temperature gradient. The dynamic distribution characteristics of the battery internal resistance use the data provided by the battery management system and combine relevant algorithms to analyze the change law of the battery internal resistance under different charging and discharging conditions. The motor torque response delay characteristics measure the delay time of the motor torque response by comparing the motor control signal and the actual speed change, thereby generating the motor torque response delay characteristics.

[0075] By comprehensively collecting multi-dimensional power supply data and extracting targeted features, dynamic feature vectors for each dimension are generated, providing rich and valuable information for formulating subsequent power supply control strategies. These feature vectors can more accurately reflect the actual working state of the inflator power supply, contributing to the precise control of the power supply and improving the overall performance and reliability of the inflator.

[0076] Embodiment 2:

[0077] In the process of generating the power supply control strategy sequence and resource allocation plan for the current inflator, the genetic algorithm first generates an initial candidate control strategy set according to the matching degree between the global correlation feature matrix and the fault modes in the knowledge base. Each feature in the global correlation feature matrix is compared one by one with the fault mode features stored in the knowledge base to calculate the matching degree. For example, when a feature vector in the global correlation feature matrix has a high similarity with a feature vector under a certain fault mode in multiple dimensions, it is considered that the global correlation feature matrix has a high matching degree with this fault mode.

[0078] Sort the fault modes according to the matching degree from high to low, and select the historical control strategies corresponding to several fault modes with higher matching degrees as the initial candidate control strategy set. These initial candidate strategies are based on the strategies adopted in similar situations that occurred in the past and have certain reference value.

[0079] Next, a strategy execution dependency graph is constructed based on the inflator hardware topology. In the hardware system of the inflator, there are energy transfer constraints and hardware resource conflict relationships between different power regulation operations. For example, the speed regulation operation of the motor may affect the discharge current of the battery, thereby affecting the energy consumption of the battery, which reflects the energy transfer constraint; while some regulation operations may need to occupy the same hardware resource at the same time, such as multiple control signals may need to be output through the pins of the same microcontroller, which generates hardware resource conflicts.

[0080] In the strategy execution dependency graph, each power regulation operation is used as a node, and edges are used to represent the energy transfer constraints or hardware resource conflict relationships between operations. In this way, the mutual relationships between different operations can be clearly shown, providing an intuitive basis for optimizing the strategy in the future.

[0081] Combined with real-time power supply data, the candidate strategy set is iteratively optimized through an improved particle swarm optimization algorithm. The improved particle swarm optimization algorithm introduces an adaptive inertia weight adjustment strategy, and this weight is dynamically updated according to the real-time error of the inflation efficiency and the battery energy consumption rate. Suppose the real-time error of the inflation efficiency is , representing the difference between the current actual inflation efficiency and the target inflation efficiency; the battery energy consumption rate is , representing the proportion of the energy consumed by the battery per unit time in the total energy. The inertia weight The update formula for is and are the maximum and minimum values of the inertia weight respectively, is the maximum number of iterations, is the current number of iterations. In this way, when the inflation efficiency error is large and the battery energy consumption rate is high, the inertia weight will be adjusted accordingly, making the algorithm more inclined to global search to find a better strategy; on the contrary, it will pay more attention to local search to refine the current better solution.

[0082] At the same time, an elite retention mechanism is adopted to screen the non-dominated solution set in the iterative process. In each iteration, the non-dominated solutions in the current population are saved. These non-dominated solutions represent strategies that perform well in different aspects in the current iteration. Then, the Cauchy mutation operator is used to enhance the global search ability. The Cauchy mutation operator will randomly generate a mutation value according to the Cauchy distribution based on the current solution to perturb the solution, thereby increasing the possibility of the algorithm jumping out of the local optimal solution and enabling the algorithm to search in a wider solution space.

[0083] After multiple iterations of optimization, the control sequence and resource allocation results that conform to the actual situation of the current air pump are finally output, realizing the optimal control of the power supply and improving the working efficiency and performance of the air pump.

[0084] Embodiment 3:

[0085] This embodiment focuses on the application of the convolutional neural network model in the power control of automotive air pumps. The convolutional neural network model aims to more effectively process dynamic feature vectors and generate a global correlation feature matrix.

[0086] The multi-scale convolutional layer of the model is an important part to achieve this goal. The multi-scale convolutional layer uses convolutional kernels of different sizes to perform convolutional operations on the input dynamic feature vectors. For example, a small-sized convolutional kernel (such as ) is set to extract local detail features and capture relatively subtle changes in the power supply signal; large-sized convolutional kernels (such as or ) are used to obtain more extensive context information and grasp the overall trend of the signal. Through this multi-scale convolution method, the power supply signal features in different frequency bands can be extracted.

[0087] After extracting the features of different frequency bands, the channel attention mechanism is used to weight and fuse these features. The channel attention mechanism adaptively adjusts the features of different channels by calculating the importance weights of each channel. Specifically, first, a global average pooling operation is performed on the feature map in the spatial dimension to compress the features of each channel into a single value, obtaining the channel descriptor. Then, a neural network containing a fully connected layer processes the channel descriptor to generate the weight value for each channel. This weight value reflects the importance of the features of this channel in the overall features. Finally, the features of each channel are multiplied by the corresponding weight value to achieve the weighted fusion of the features of each frequency band, thereby generating the global correlation feature matrix.

[0088] Taking a simple example to illustrate, assume that the input dynamic feature vector passes through a multi-scale convolutional layer and obtains three feature maps of different frequency bands , , . After global average pooling, the channel descriptors , , are obtained respectively. Through the processing of the neural network, the corresponding weight values , , are obtained. Then the calculation method of the finally fused feature map is , and this is the generated global correlation feature matrix.

[0089] Through the collaborative work of the multi-scale convolutional layer and the channel attention mechanism, the convolutional neural network model can fully exploit the information in the dynamic feature vector, generate a more representative global correlation feature matrix, provide strong support for generating the power control strategy and resource allocation scheme based on this matrix subsequently, and improve the accuracy and effectiveness of the power control of the automotive air pump.

[0090] Example 4:

[0091] In the improved particle swarm optimization algorithm, the adaptive inertia weight adjustment strategy, the elite retention mechanism, and the Cauchy mutation operator cooperate with each other to effectively improve the algorithm performance.

[0092] The adaptive inertia weight adjustment strategy dynamically updates the inertia weight according to the real-time error of the inflation efficiency and the battery energy consumption rate. The real-time error of the inflation efficiency reflects the gap between the actual inflation efficiency of the current air pump and the expected target. When the actual inflation efficiency is lower than the target value, the error is positive, and the larger the error, the more serious the deviation of the current inflation efficiency from the ideal state. The battery energy consumption rate reflects the consumption speed of the battery energy during the inflation process. If the battery energy is consumed too quickly, it will affect the endurance and overall performance of the air pump.

[0093] In actual calculations, assume the real-time error of the charging efficiency is calculated through the difference between the current actual charging volume and the target charging volume within a unit time , that is . The battery energy consumption rate is calculated by the ratio of the energy consumed by the current battery to the initial total energy of the battery within a unit time , that is .

[0094] According to the formula , at the beginning of the algorithm , the inertia weight is close to . At this time, the algorithm has strong global search ability, and particles can explore the solution space within a large range to find possible optimal solution regions. As the iteration progresses, if the charging efficiency error is large and the battery energy consumption rate is high has a large value, the inertia weight will rapidly decrease, making the particles more inclined to global search to jump out of the current possible local optimal solution. On the contrary, if the charging efficiency error is small and the battery energy consumption rate is low, the inertia weight will relatively stably stay at a higher value, and the algorithm pays more attention to local search to finely adjust the current better solution.

[0095] In the iterative process, the elite retention mechanism screens the population generated in each iteration. The non-dominated solutions, that is, those solutions that are not dominated by other solutions in terms of the objective function (such as comprehensive consideration of multiple objectives like charging efficiency, battery energy consumption, etc.), are retained. These non-dominated solutions represent better strategies in different aspects in the current iteration. For example, a certain solution may perform well in terms of charging efficiency, while another solution is better in terms of battery energy consumption, and they are both retained.

[0096] The Cauchy mutation operator adds the ability for the algorithm to jump out of the local optimal solution. In each iteration, for some particles, a mutation value is generated according to the Cauchy distribution. The Cauchy distribution has a heavy tail. Compared with the traditional Gaussian mutation, it enables the particles to search at places far from the current position. Assume the current position of the particle is , the mutation value is , and the position after mutation is . Through this mutation operation, the algorithm can explore a wider solution space, avoid falling into the local optimal solution, and increase the probability of finding the global optimal solution.

[0097] By comprehensively applying these three mechanisms, the improved particle swarm optimization algorithm can more efficiently search the solution space when generating power control strategies and resource allocation schemes, improve the convergence speed and solution quality of the algorithm, and provide better policy support for the power control of automotive air pumps.

[0098] Example 5:

[0099] In the process of automotive air pump power control, signal feature extraction and abnormal pattern detection and processing are important links to ensure the stable operation of the power supply.

[0100] First, the signal feature extraction adopts the wavelet transform and time-frequency joint analysis method. The power supply signal is decomposed by the Morlet wavelet basis function to extract the energy entropy features at each scale. The Morlet wavelet basis function has good time-frequency localization characteristics and can perform a detailed analysis of the power supply signal at different time and frequency scales. Assuming the power supply signal is , using the Morlet wavelet basis function to perform wavelet transform on it, obtaining the wavelet coefficients , where is the scale parameter, is the translation parameter. By calculating the energy entropy of the wavelet coefficients at different scales , here represents different translation positions, and the energy entropy reflects the complexity and uncertainty of the signal at this scale, thus extracting the energy entropy features at each scale.

[0101] Next, perform singular value decomposition on the time-frequency matrix to generate a dynamic feature vector representing the signal mutation and steady-state components. The time-frequency matrix is composed of the wavelet coefficients obtained by wavelet transform. Let the time-frequency matrix be , and perform singular value decomposition on it , where and are unitary matrices, is a diagonal matrix, and the elements on its diagonal are singular values. The magnitude and distribution of the singular values reflect different characteristics of the signal. Larger singular values correspond to the main components of the signal, and smaller singular values correspond to the secondary components or noise of the signal. By selecting appropriate singular values and combining them into a dynamic feature vector, the signal mutation and steady-state components can be effectively characterized.

[0102] During the real-time monitoring process, an anomaly pattern detection is performed on the real-time power supply data based on a fuzzy logic system. The fuzzy logic system processes uncertain information by defining fuzzy sets and fuzzy rules. For example, for voltage data, fuzzy sets such as "normal voltage", "slightly low voltage", "too low voltage" are defined, along with corresponding membership functions to describe the degree to which the voltage data belongs to each fuzzy set. Based on the fuzzy information of multiple power supply data dimensions (such as voltage, current, temperature, etc.), a series of fuzzy rules are formulated. For example, when the voltage belongs to "too low voltage" and the current belongs to "too large current", it is judged that an abnormal situation may exist.

[0103] When an anomaly is detected, a dynamic reconfiguration of the control strategy sequence is triggered, and the resource allocation scheme is updated according to the reconfiguration result. The dynamic reconfiguration uses a mixed-integer linear programming solver. First, the power supply operation sequence constraints are defined to ensure that each power supply regulation operation is executed in the correct order, avoiding system failures caused by improper operation sequences. For example, before performing a charging operation on the battery, it is necessary to ensure that the charging circuit is correctly connected and the relevant protection mechanisms are activated.

[0104] At the same time, the battery charge-discharge cycle count limit is considered to extend the battery's service life. The number of charge-discharge cycles of the battery is limited, and overcharging and over-discharging will accelerate battery aging. Therefore, when formulating the control strategy, unnecessary charge-discharge operations should be avoided frequently, so that the number of charge-discharge cycles of the battery is within a reasonable range.

[0105] In addition, the hard constraints of the radiator start-stop time window also need to be considered. The role of the radiator is to dissipate heat for the air pump to ensure that it operates within an appropriate temperature range. In some cases, such as when the air pump operates at high load for a long time, the radiator needs to be started in a timely manner; while when the temperature is low or the air pump is in a low-load state, the radiator can be appropriately turned off to save energy.

[0106] A feasible solution set is generated through a column generation algorithm. The column generation algorithm expands the solution space by gradually generating new columns (corresponding to different power supply control strategy options) to find solutions that satisfy various constraint conditions. Then, the comprehensive optimal solution is selected based on the entropy weight TOPSIS method. The entropy weight TOPSIS method determines the weights according to the information entropy of each index, and then ranks the feasible solutions by calculating the distances between each solution and the ideal solution and the negative ideal solution, and selects the comprehensive optimal solution. In this way, when a power supply anomaly is detected, the control strategy and the resource allocation scheme can be quickly adjusted to ensure the stable operation of the automotive air pump.

[0107] Example 6:

[0108] In the entire process of automotive air pump power supply control, constructing a knowledge graph of power supply component relationships and verifying the physical compatibility between the strategy and the knowledge graph during the strategy sequence generation stage play a crucial role in ensuring the stable and efficient operation of the air pump power supply.

[0109] When constructing the knowledge graph of the power supply component relationship, various electrical components in the air pump and possible fault types are set as nodes. In terms of electrical components, like the battery, which is a key power supply component of the air pump, stores and outputs electrical energy, and is a core node in the knowledge graph. It has many attributes, such as battery capacity, voltage, internal resistance, etc., and these attributes directly affect the working duration and performance of the air pump. The motor is also important. It converts electrical energy into mechanical energy to drive the air pump to work. Attributes such as the rotational speed, torque, and power of the motor are also key information that the knowledge graph needs to record. In addition, the controller is responsible for regulating the operation of the entire power supply system, and the radiator ensures that the system works at an appropriate temperature. They all exist as independent nodes in the knowledge graph.

[0110] And the fault types are also included in the node category. For example, the battery overcharge fault, which is a situation where the battery charge exceeds the safety limit due to improper control during the charging process; the motor overheat fault, which is usually caused by long-term high-load operation or poor heat dissipation. Each fault type node is associated with the relevant electrical component node, and this association reflects the causal relationship of the fault occurrence.

[0111] The edges between nodes are used to represent the energy coupling relationship between components or the fault conduction path. In terms of the energy coupling relationship, there is a close energy transfer connection between the battery and the motor. The electrical energy output by the battery is transmitted to the motor through the circuit to drive the motor to operate. In the knowledge graph, a directed edge is drawn from the battery node to the motor node, and the attributes of the edge can be marked with information such as output voltage and current magnitude to clearly present the energy coupling situation between the two. Another example is that when the motor has an overheat fault, the excessive temperature may affect the nearby controller through the conduction of metal components or heat radiation. In the knowledge graph, there will be an edge from the motor overheat fault node to the controller fault node, and the attributes of this edge will record information such as the possibility of fault conduction, time delay, and the possible controller fault types caused, thus constructing a complete knowledge graph of the power supply component relationship.

[0112] In the strategy sequence generation stage, the subgraph isomorphism detection algorithm is used to verify the physical compatibility between the strategy and the knowledge graph. When generating a power supply control strategy sequence, it is transformed into a subgraph form. The nodes in the subgraph correspond to the electrical component operations or fault response measures involved in the strategy, and the edges represent the logical relationships between these operations or measures. Then, the subgraph is compared with the pre-constructed knowledge graph of the power supply component relationship.

[0113] For example, a control strategy is to start the charging operation when the battery power is lower than a certain threshold, and at the same time adjust the motor speed to reduce energy consumption. In the sub-graph corresponding to this strategy, there will be corresponding operation marks on the battery node and the motor node, and there are edges indicating the sequence and mutual influence of these two operations. Through the sub-graph isomorphism detection algorithm, check whether an isomorphic part of this sub-graph can be found in the knowledge graph. If an isomorphic part can be found, it means that the strategy is reasonable and feasible at the physical level because it conforms to the actual energy coupling relationship and fault conduction logic among the components of the inflator pump; conversely, if no isomorphic part is found, it means that the strategy may lead to physical conflicts or does not conform to the actual working principle, and the strategy needs to be adjusted or re-formulated.

[0114] By constructing the knowledge graph of the power supply component relationship and performing sub-graph isomorphism detection, potential problems can be discovered in a timely manner during the strategy formulation stage, ensuring that the generated power supply control strategy sequence not only meets the control requirements but also conforms to the physical characteristics of the inflator pump, thereby improving the accuracy and reliability of the power supply control of the automotive inflator pump and ensuring the stable operation of the inflator pump.

[0115] An automotive inflator pump includes an inflator pump housing main body 1, an air pump assembly 2, and a mobile power supply 3. An assembly groove 6 for the mobile power supply 3 to be movably arranged therein is formed on the inflator pump housing main body 1. An electrical interface terminal 12 electrically connected to the air pump assembly 2 is provided in the assembly groove 6. The mobile power supply 3 is electrically connected to the air pump assembly 2 through the electrical interface terminal 12. A spring piece 7 is provided in the assembly groove 6. One end of the spring piece 7 is fixed to the bottom of the assembly groove 6, and the other end contacts the mobile power supply 3. A control button 5 is provided on the surface of the inflator pump housing main body 1, and the bottom end of the control button 5 is connected to a button bracket that contacts the spring piece 7.

[0116] It should be noted that the button bracket in this embodiment further includes a linkage rod. The two ends of the linkage rod respectively contact the spring piece 7 and the control button 5. During use, the pressing action of the control button 5 is transmitted to the spring piece 7 through the linkage rod. After the free end of the spring piece 7 receives the thrust of the linkage rod, it uses its own elastic deformation to push upward. When the control button 5 is released, the elastic restoring force of the spring piece 7 makes its free end return to the initial position, completing a complete linkage operation;

[0117] In addition, the control button 5 and the spring piece 7 are physically connected through the linkage rod. This connection method is simple and stable, which can ensure the sensitivity and reliability of the button operation. The elastic design of the spring piece 7 enables it to still maintain good performance after multiple operations, reducing the risk of failure caused by mechanical fatigue.

[0118] Moreover, the mobile power supply 3 is electrically connected to the air pump assembly 2 through the electrical interface terminals 12. Specifically, the electrical interface terminals 12 are installed inside the main body of the air pump housing 1 near the installation position of the mobile power supply 3. The electrical interface terminals 12 are composed of multiple metal contact pieces, and these contact pieces respectively correspond to the positive and negative poles and the control signal lines of the mobile power supply 3;

[0119] The contact pieces of the electrical interface terminals 12 have a certain elasticity, providing a stable contact pressure when the mobile power supply 3 is inserted to ensure the reliability of the electrical connection. The side or bottom of the mobile power supply 3 is provided with metal contact points that match the electrical interface terminals 12, and the positions and shapes of these contact points correspond one by one to the contact pieces of the electrical interface terminals 12;

[0120] In this way, when the mobile power supply 3 is inserted into the inside of the air pump housing main body 1, its metal contact points will contact the contact pieces of the electrical interface terminals 12. Due to the elastic design of the electrical interface terminals 12, the contact pieces will automatically clamp the contact points of the mobile power supply 3 to ensure a stable electrical connection between the two;

[0121] Disassembly process: When it is necessary to disassemble the mobile power supply 3, press the control button 5, and use the linkage rod to drive the elastic piece 7 to push out the mobile power supply 3. Due to the elasticity of the electrical interface terminals 12, the contact pieces will automatically loosen, and the mobile power supply 3 can be taken out smoothly, and at the same time the electrical connection is automatically disconnected;

[0122] Thus, the detachable design of the mobile power supply 3 enables users to quickly replace the mobile power supply with insufficient power without waiting for charging, improving the usage efficiency of the device. Users can choose mobile power supplies with different capacities according to actual needs to adapt to different usage scenarios;

[0123] Moreover, the mobile power supply 3 is installed inside the main body of the air pump housing 1 without occupying extra space, making the overall structure of the device more compact and suitable for use in limited spaces. The electrical interface terminals 12 integrate the functions of the mobile power supply 3 and the air pump assembly 2, reducing the use of external cables and improving the overall aesthetics and reliability of the device.

[0124] One side of the air pump assembly 2 is connected to a fan assembly 8. An air inlet 9 corresponding to the fan assembly 8 is provided on the side wall of the main body of the air pump housing 1, and an air outlet 10 communicating with the air pump assembly 2 is provided on the other side of the main body of the air pump housing 1. During use, the fan assembly 8 ventilates the inside of the pump body through the air inlet 9 and the air outlet 10 to continuously cool down.

[0125] In addition, a lighting lamp 11 is embedded on the surface of the main body of the air pump housing 1. The lighting lamp 11 provides temporary lighting to cope with the use in places with poor lighting at night or in underground garages. A switchable sealing baffle 4 is provided at the opening of the assembly groove 6 for hermetically sealing the assembly groove 6 to provide sealing protection.

[0126] It should be noted that, in this document, relational terms such as first and second are only used 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 "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0127] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A power control method for an automotive air pump, characterized in that, Including: Collecting multi-dimensional power data of an automotive air pump; Performing signal feature extraction on the multi-dimensional power data to generate dynamic feature vectors for each dimension; Mapping the dynamic feature vectors to a joint analysis space based on a pre-trained convolutional neural network model to generate a global correlation feature matrix; Constructing a power mode knowledge base, which stores historical power usage patterns, fault cases, optimization strategies, and corresponding execution efficiency data; Generating a power control strategy sequence and a resource allocation plan for the current air pump through a genetic algorithm according to the global correlation feature matrix and the historical patterns in the knowledge base.

2. The power control method according to claim 1, wherein The multi-dimensional power data includes voltage, current, ambient temperature, humidity, inflation rate, battery health status, user preset pressure value, motor speed, and radiator working status; The dynamic feature vectors include power ripple spectrum features, load fluctuation detection features, temperature gradient change features, dynamic distribution features of battery internal resistance, and motor torque response delay features.

3. The power control method according to claim 1, characterized in that The generating of the power control strategy sequence and the resource allocation plan for the current air pump through a genetic algorithm includes: Generating an initial candidate control strategy set according to the matching degree between the global correlation feature matrix and the fault patterns in the knowledge base; Constructing a strategy execution dependency graph based on the hardware topology of the air pump, where the nodes in the strategy execution dependency graph represent power regulation operations, and the edges represent energy transfer constraints or hardware resource conflict relationships between operations; Combining real-time power data, iteratively optimizing the initial candidate control strategy set through an improved particle swarm optimization algorithm, and outputting the final control sequence and resource allocation result.

4. The power control method according to claim 3, wherein The convolutional neural network model includes: Using multi-scale convolutional layers to extract power signal features in different frequency bands; Weightedly fusing the features of each frequency band through a channel attention mechanism to generate the global correlation feature matrix.

5. The power control method according to claim 3, wherein The improved particle swarm optimization algorithm includes: Introducing an adaptive inertia weight adjustment strategy, where the weight is dynamically updated according to the real-time error of inflation efficiency and the battery energy consumption rate; Adopting an elite retention mechanism to screen the non-dominated solution set in the iterative process, and enhancing the global search ability through a Cauchy mutation operator.

6. The power control method according to claim 1, wherein The signal feature extraction adopts a wavelet transform and time-frequency joint analysis method, specifically including: Decomposing the power signal through a Morlet wavelet basis function to extract the energy entropy features at each scale; Performing singular value decomposition on the time-frequency matrix to generate dynamic feature vectors representing signal mutation and steady-state components; The method further includes: Performing abnormal mode detection on real-time power data based on a fuzzy logic system; When an abnormality is detected, triggering dynamic reconstruction of the control strategy sequence and updating the resource allocation plan according to the reconstruction result; The dynamic reconstruction adopts a mixed integer linear programming solver, specifically including: Defining hard constraints such as power operation sequence constraints, battery charge and discharge cycle times limit, and radiator start-stop time window; Generating a feasible solution set through a column generation algorithm and screening the comprehensive optimal solution based on the entropy weight TOPSIS method; The method further includes: Constructing a knowledge graph of power component relationships, where the nodes in the graph represent air pump electrical components or fault types, and the edges represent energy coupling relationships or fault conduction paths between components; In the policy sequence generation stage, the physical compatibility between the policy and the knowledge graph is verified through a subgraph isomorphism detection algorithm.

7. A power control system for an automotive air pump, characterized in that, The power control method according to any one of claims 1-6 is adopted, including: A data acquisition module: used for acquiring multi-dimensional power data of an automotive air pump; A feature extraction module: extracting signal features from the multi-dimensional power data to generate dynamic feature vectors for each dimension; A feature association module: mapping the dynamic feature vectors to a joint analysis space based on a pre-trained convolutional neural network model to generate a global association feature matrix; A knowledge base construction module: constructing a power mode knowledge base, where the knowledge base stores historical power usage patterns, fault cases, optimization strategies, and corresponding execution efficiency data; A policy generation module: generating a power control policy sequence and a resource allocation plan for the current air pump through a genetic algorithm according to the global association feature matrix and the historical patterns in the knowledge base.

8. An automotive air pump, characterized in that, The power control system according to claim 7 is adopted, including an air pump housing main body (1), an air pump assembly (2), and a mobile power source (3), characterized in that: an assembly groove (6) for movably arranging the mobile power source (3) therein is formed on the air pump housing main body (1), an electrical interface terminal (12) electrically connected to the air pump assembly (2) is arranged in the assembly groove (6), the mobile power source (3) is electrically connected to the air pump assembly (2) through the electrical interface terminal (12), a spring piece (7) is arranged in the assembly groove (6), one end of the spring piece (7) is fixed to the bottom of the assembly groove (6), and the other end is in contact with the mobile power source (3), a control button (5) is arranged on the surface of the air pump housing main body (1), and the bottom end of the control button (5) is connected to a button bracket in contact with the spring piece (7).

9. The automotive air pump according to claim 8, wherein One side of the air pump assembly (2) is connected with a fan assembly (8), an air inlet (9) corresponding to the fan assembly (8) is formed on the side wall of the air pump housing main body (1), and an air outlet (10) communicated with the air pump assembly (2) is arranged on the other side of the air pump housing main body (1).

10. The automotive air pump according to claim 8, characterized in that, A lighting lamp (11) is embedded on the surface of the air pump housing main body (1), and an openable and closable sealing baffle (4) is arranged at the opening of the assembly groove (6).

Images