An internationally applicable charging pile and intelligent management system

Abstract

This invention provides an internationally compatible charging pile and intelligent management system, relating to the field of electric vehicle charging pile technology. It includes: a multi-standard compatible adapter module for integrating adapters of various standards to support charging of electric vehicles of different standards at the charging pile; and an adapter identification and information transmission module for automatically identifying the type of adapter connected to the charging pile using an adapter matching algorithm, and transmitting the identified adapter type to an automatic output parameter adjustment module. This invention, by integrating multi-standard compatible adapters, intelligent identification and output adjustment, remote monitoring and management, and user-friendly interaction, achieves efficient, convenient, and safe charging services, meeting the charging needs of electric vehicles of different standards.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging pile technology, and in particular to an internationally universal charging pile and intelligent management system. Background Technology

[0002] Traditional charging stations mostly support only a single charging standard, while the electric vehicle market has a variety of different charging standards.

[0003] When users drive electric vehicles that do not meet local standards to other countries or regions, they often find it difficult to find suitable charging stations. This standard incompatibility issue causes great inconvenience to users and limits the cross-regional use of electric vehicles.

[0004] In addition, the traditional management methods for charging stations are relatively outdated. The process of borrowing and returning adapters is cumbersome and often requires manual intervention, which not only leads to low management efficiency but also increases management costs.

[0005] For example, at some public charging stations, users need to go to the service counter to borrow adapters and return them to a designated location after charging. This management method not only wastes users' time but also creates an additional workload for charging station managers. Furthermore, the lack of intelligent methods for monitoring and maintaining the adapters' status leads to a shortened lifespan and further increases management costs. Summary of the Invention

[0006] The technical problem to be solved by this invention is to provide an internationally universal charging pile and intelligent management system, which solves the problem of standard incompatibility in the electric vehicle charging pile market.

[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0008] Firstly, an internationally universal charging station includes:

[0009] Multi-standard compatible adapter module, used to integrate adapters for multiple standards to support electric vehicles of different standards to charge at charging stations;

[0010] The connector identification and information transmission module is used to automatically identify the type of adapter connected to the charging pile through the adapter matching algorithm. If the adapter type is identified, the adapter type is transmitted to the automatic output parameter adjustment module.

[0011] The automatic output parameter adjustment module is used to automatically detect and intelligently adjust the output current and voltage according to the different standard conversion connectors connected, using a flower pollination algorithm to adapt to the charging needs of different electric vehicles.

[0012] The intelligent management module is used to intelligently identify and manage the status of connectors, and to enable self-service borrowing and returning of adapters by monitoring the use, location and damage of connectors in real time, so as to obtain maintenance reports;

[0013] The remote monitoring module is used to remotely monitor, diagnose faults, and perform data statistical analysis of charging piles via a remote server.

[0014] The user interaction module is used to view charging status, cost information, and operation guides for borrowing and returning adapters. It also supports users in initiating charging and making payments.

[0015] Furthermore, through a connector matching algorithm, the type of connector used to access the charging station is automatically identified. If the connector type is identified, it is transmitted to the automatic output parameter adjustment module, including:

[0016] The charging port of the charging pile continuously monitors the port status through sensors. When the sensor detects that the adapter is connected, the charging pile’s built-in camera captures an image of the adapter and extracts key features from the image, including shape, size and texture, to form a feature vector.

[0017] The feature vector is compared with the standard features in the feature database, the similarity between each standard feature and the feature vector is calculated, and the matching joint type, i.e. the matching item, is determined based on the similarity.

[0018] The matching items are transmitted to the automatic output parameter adjustment module via the internal communication bus of the charging pile.

[0019] Furthermore, based on the different standard adapters used, the system automatically detects and intelligently adjusts the output current and voltage using a flower pollination algorithm to adapt to the charging needs of different electric vehicles, including:

[0020] When an electric vehicle is connected to a charging station, the charging station's sensors detect the type of adapter connector used to identify the electric vehicle's charging standard.

[0021] The charging standard is analyzed to obtain the current and voltage range and charging speed requirements under the charging standard, and the parameters of the flower pollination algorithm are initialized, including the population size, number of iterations and fitness function.

[0022] The final current and voltage combination is searched using a flower pollination algorithm, where each flower represents a current and voltage combination;

[0023] For each flower, the fitness function evaluates the performance of the corresponding flower based on the electric vehicle's charging requirements, including charging speed, battery protection, and energy efficiency, in order to obtain the fitness function evaluation result.

[0024] Based on the fitness function evaluation results, an iterative search is performed, each flower is continuously adjusted, and the fitness function evaluation and iterative search process is repeated until the preset number of iterations is reached. The final combination of current and voltage is determined based on the fitness value of each flower.

[0025] Based on the final current and voltage combination, during the charging process, the charging pile continuously monitors the output of current and voltage and the battery status of the electric vehicle, including charge, temperature and internal resistance. If any abnormality occurs, including overheating, overcurrent, or battery failure, the charging pile will immediately take corresponding protective measures, including reducing output, cutting off power or issuing an alarm.

[0026] When an electric vehicle finishes charging or the user requests to disconnect via the charging station interface, the charging station stops outputting current and voltage, thus completing the charging process to accommodate the charging needs of different electric vehicles.

[0027] Furthermore, the system intelligently identifies and manages connector status, and through real-time monitoring of connector usage, location, and damage, enables self-service borrowing and returning of adapters to generate maintenance reports, including:

[0028] Each connector is equipped with a unique identifier and its basic information is initialized, including type, specifications, production date, and initial condition.

[0029] Various sensors are deployed in the application scenarios of the connector, including counters, positioners, and damage detection sensors, and the usage data, location data, and damage data of the connector are collected in real time through the sensors;

[0030] The usage data, location data, and damage data of the joints are trained and optimized using a pre-defined random forest model to obtain joint status identification results and generate predictive maintenance suggestions.

[0031] When a user selects a connector to borrow through the interactive interface, the user's identity is verified, the connector status is updated to "borrowing", and the borrowing time and user information are recorded; when the user returns the connector, the user's identity is verified again, the connector status is updated to "available", and the return time and connector usage are recorded.

[0032] Based on real-time and historical data, the status of the joints is analyzed regularly to generate maintenance reports. The reports include statistics on the frequency of joint use, location distribution maps, damage analysis, and predictive maintenance recommendations.

[0033] Furthermore, a pre-defined random forest model is used to train and optimize the usage, location, and damage data of the joints to obtain joint status identification results and generate predictive maintenance suggestions, including:

[0034] Collect usage data, location data, and damage data of the connectors, and divide the collected data into training set and test set;

[0035] Based on the training and test sets, construct the basic framework of the random forest model and set the parameters of the random forest model, including the number of decision trees, maximum depth, and minimum number of sample splits.

[0036] The random forest model is initially trained using the training set to obtain preliminary training results. Based on these results, the random forest model is then optimized and adjusted, including increasing or decreasing the number of decision trees and adjusting the feature selection strategy, to obtain an optimized random forest model.

[0037] The optimized random forest model is used to predict the test set data. Each decision tree produces a prediction result, and the prediction results of all decision trees are combined to obtain the final joint state identification label.

[0038] Based on the final connector status identification tag, charging pile operation data, and historical maintenance records, the connector status is analyzed to obtain predictive maintenance recommendations, including recommendations for connector repair, replacement, and inspection, as well as corresponding maintenance times and priorities.

[0039] Furthermore, remote monitoring, fault diagnosis, and data statistical analysis of charging piles can be achieved through a remote server, including:

[0040] Based on the charging pile's operational data, potential faults are predicted and diagnosed using a fault diagnosis model on a remote server, and fault alarm information is generated.

[0041] The remote server collects, organizes, and analyzes the charging pile's operational data, including charging volume, charging duration, connector status, and ambient temperature, generating various reports and charts, including charging volume statistics reports and failure rate analysis charts. It also provides customized data statistical analysis services, including charging volume statistics for specific time periods and failure rate analysis for specific failure types.

[0042] Furthermore, based on the charging pile's operational data, potential faults are predicted and diagnosed using a fault diagnosis model on a remote server, and fault alarm information is generated, including:

[0043] The charging pile's operating data, including current, voltage, power, temperature, and charging time, is transmitted to a remote server.

[0044] The fault diagnosis model on the remote server predicts and diagnoses the operating data of the charging pile to obtain the predicted status of the charging pile, so as to determine whether there is any abnormality or fault in the charging pile.

[0045] Based on the predicted status of the charging pile, if an abnormality or fault is detected in the charging pile, the type of abnormality or fault is determined, including electrical fault, mechanical fault, and communication fault.

[0046] The fault diagnosis model analyzes the causes of the faults, including battery aging, poor connector contact, and poor heat dissipation. Based on the fault type and cause, the model generates fault alarm information, including the fault type, fault cause, time of fault occurrence, charging station identifier, and suggested solutions.

[0047] Secondly, an internationally recognized intelligent management system for charging piles includes:

[0048] The charging station integrates multiple standard adapters to support various electric vehicle charging standards;

[0049] When users borrow adapters, they can use online payment to enable self-service borrowing, and the system can intelligently identify and record the borrowing status of the adapters.

[0050] Users connect their electric vehicles to charging stations and select the corresponding charging standard. The system automatically detects the charging needs of the electric vehicles and obtains the detection results.

[0051] Based on the test results, the output parameters of the charging pile, including current and voltage, are intelligently adjusted to adapt to the selected charging specifications.

[0052] During the charging process, the charging station continuously monitors the charging status, including key parameters such as current, voltage, and charging time, and handles any abnormal situations.

[0053] Once charging is complete, the charging station records charging data, including charging time, charging amount, electricity consumption, and generates a corresponding charging report.

[0054] The above-described solution of the present invention has at least the following beneficial effects:

[0055] By integrating adapters for multiple standards, this charging station supports various electric vehicle charging standards, including Chinese, European, American, Japanese, and Tesla standards. This means that regardless of the standard of the electric vehicle a user drives, it can be charged at this station. This improves charging convenience and reduces the inconvenience users face due to incompatible charging standards.

[0056] The connector identification and information transmission module automatically identifies the type of adapter connector being connected and transmits this information to the automatic output parameter adjustment module, ensuring the safety and efficiency of the charging process. Simultaneously, the intelligent management module intelligently identifies and manages the status of the connectors, enabling self-service borrowing and returning of adapter connectors by monitoring their usage, location, and damage in real time. This not only improves the utilization rate of adapter connectors but also reduces management costs and provides users with a convenient charging experience. The generation of maintenance reports also helps to promptly identify and resolve potential problems, ensuring the long-term stable operation of the charging station.

[0057] The automatic output parameter adjustment module utilizes a flower pollination algorithm to automatically detect and intelligently adjust the output current and voltage based on the different standard adapters connected, adapting to the charging needs of various electric vehicles. This function ensures the safety and efficiency of the charging process, preventing equipment damage or safety accidents that may result from current or voltage mismatch.

[0058] The remote monitoring module enables remote monitoring, fault diagnosis, and data statistical analysis of charging piles via a remote server. This allows operators to monitor the operational status of charging piles in real time, promptly identify and handle faults, and improve operational efficiency. Simultaneously, the data statistical analysis function helps operators understand charging demand, optimize resource allocation, and provide strong support for future charging infrastructure construction.

[0059] The user interaction module provides a user-friendly interface, allowing users to easily view charging status, cost information, and instructions on borrowing and returning adapters. Furthermore, this module supports users initiating charging and making payments, enabling self-service charging. This not only enhances the user experience but also helps improve the utilization rate and service quality of charging stations. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of an internationally universal charging pile provided by an embodiment of the present invention.

[0061] Figure 2 This is a flowchart illustrating an internationally universal intelligent management system for charging piles, provided by an embodiment of the present invention. Detailed Implementation

[0062] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0063] like Figure 1As shown, an embodiment of the present invention proposes an internationally universal charging pile, comprising:

[0064] Multi-standard compatible adapter module 11 is used to integrate adapters for multiple standards to support electric vehicles of different standards to charge at charging stations;

[0065] The connector identification and information transmission module 12 is used to automatically identify the type of adapter connected to the charging pile through the adapter matching algorithm. If the adapter type is identified, the adapter type is transmitted to the automatic output parameter adjustment module.

[0066] The automatic output parameter adjustment module 13 is used to automatically detect and intelligently adjust the output current and voltage according to the different standard conversion connectors connected, using the flower pollination algorithm, to adapt to the charging needs of different electric vehicles.

[0067] The intelligent management module 14 is used to intelligently identify and manage the status of the connectors, and to realize the self-service borrowing and returning of the adapters by monitoring the use, location and damage of the connectors in real time, so as to obtain maintenance reports.

[0068] The remote monitoring module 15 is used to remotely monitor, diagnose faults, and perform data statistical analysis of charging piles via a remote server.

[0069] User interaction module 16 is used to view charging status, fee information, and operation guides for borrowing and returning adapters, and supports users to start charging and make payments.

[0070] In this embodiment of the invention, the multi-standard compatible adapter module 11 improves the universality of the charging pile, enabling electric vehicles of different standards to be charged on the same charging pile, which is convenient for electric vehicle users and reduces the construction and maintenance costs of the charging pile, because there is no need to set up a separate charging pile for each electric vehicle standard.

[0071] The connector identification and information transmission module 12 enables automatic identification of the adapter connector, improves the intelligence level of the charging process, and quickly and accurately transmits connector type information.

[0072] The automatic output parameter adjustment module 13 uses a flower pollination algorithm to automatically detect and intelligently adjust the output current and voltage, ensuring the safety and efficiency of the charging process, adapting to the charging needs of different electric vehicles, and improving the compatibility of charging piles and user satisfaction.

[0073] The intelligent management module 14 enables intelligent identification and management of the adapters, improving their efficiency and management level. By monitoring the adapter status in real time, damage or abnormalities can be detected and addressed promptly, ensuring smooth charging. It also supports self-service borrowing and returning of adapters, facilitating user convenience and reducing the workload of management personnel.

[0074] The remote monitoring module 15 enables remote monitoring of the charging piles, allowing users to understand their operating status and working conditions anytime, anywhere. It supports fault diagnosis and data statistical analysis, providing strong support for the maintenance and optimization of the charging piles.

[0075] User interaction module 16 provides an intuitive and easy-to-use user interface, allowing users to view charging status, cost information, and instructions on borrowing and returning adapters. It supports user initiation of charging and payment, improving the convenience and user experience of the charging process.

[0076] In a preferred embodiment of the present invention, integrating multiple standard adapters to support electric vehicles of different standards to charge at charging stations may include:

[0077] First, research was conducted on mainstream electric vehicle charging standards in the market, such as China's GB / T, Europe's Type 2, North America's SAE J1772, and Japan's CHAdeMO. Based on the research results, the types and quantities of adapters that need to be integrated were determined. The charging interface of the charging station was designed to ensure a stable and safe connection to various adapters. Corresponding adapters were designed for each charging standard, ensuring a tight and safe connection to the charging station's interface. The adapters contained the necessary circuitry and components to achieve energy conversion and transmission. According to the requirements of each charging standard, the internal electrical connections and conversion circuits of the adapters were designed to ensure accurate energy conversion and meet the charging needs of electric vehicles. Necessary protection circuits, such as overcurrent and overvoltage protection, were incorporated into the adapters to ensure safety during the charging process. Considering that the adapters may be exposed to harsh environments, their waterproof, dustproof, and corrosion-resistant capabilities were enhanced.

[0078] Establish a unified standard for adapter interfaces to ensure seamless connection to charging station interfaces. The interface design should be compatible and scalable, considering future interface upgrades or changes. Design each adapter as an independent module for easy installation, maintenance, and replacement. Integrate an adapter identification module within the charging station to automatically recognize the type of adapter connected. Based on the identification result, the charging station can automatically adjust its output parameters to adapt to the charging needs of different electric vehicles. Establish an intelligent management system to monitor and manage adapter usage in real time, including usage duration and damage status.

[0079] In a preferred embodiment of the present invention, the adapter type of the charging pile is automatically identified by an adapter matching algorithm. If the adapter type is identified, the adapter type is transmitted to the automatic output parameter adjustment module, which may include:

[0080] The charging port of the charging pile continuously monitors the port status through sensors. When the sensor detects that the adapter is connected, the charging pile’s built-in camera captures an image of the adapter and extracts key features from the image, including shape, size and texture, to form a feature vector.

[0081] The feature vector is compared with the standard features in the feature database, the similarity between each standard feature and the feature vector is calculated, and the matching joint type, i.e. the matching item, is determined based on the similarity.

[0082] The matching items are transmitted to the automatic output parameter adjustment module via the internal communication bus of the charging pile.

[0083] In this embodiment of the invention, the charging pile is equipped with various sensors that can monitor the physical state of the charging interface in real time, such as whether an object is inserted, the depth of insertion, and whether the contact is good. The sensors transmit the monitored data to the charging pile's control unit in the form of electrical signals. After receiving the signals from the sensors, the control unit determines whether a converter connector is connected. Once a connector is detected, the control unit triggers the charging pile's built-in camera to acquire an image. The camera captures an image of the connector's appearance and transmits the image data to the image processing module. After receiving the image data from the camera, the image processing module uses image processing algorithms to analyze the image and extract key features, such as the connector's shape, size, and surface texture. These features are encoded into feature vectors.

[0084] The charging station internally stores a feature database containing feature vectors for various standard adapters. The image processing module compares the extracted feature vectors with the standard features in the database one by one. During the comparison, the image processing module calculates the similarity between each standard feature and the feature vector. The calculated similarity value is used to determine which standard feature matches the current feature vector.

[0085] The image processing module determines the standard feature that best matches the current feature vector based on the calculated similarity value. The connector type corresponding to this standard feature is the matching item. Once the matching item is determined, the image processing module transmits the matching item information to the automatic output parameter adjustment module via the internal communication bus of the charging pile. The communication bus can be a CAN bus, Ethernet, etc., used for data transmission between various modules within the charging pile.

[0086] Assume the charging station supports two charging standards: Chinese GB / T and European Type 2. Sensors and cameras are installed at the charging interface of the charging station. When an electric vehicle supporting the GB / T standard connects to the charging station, the sensor detects the connector insertion, and the camera begins capturing images of the connector. The image processing module extracts the connector's feature vectors, including its unique shape, size, and texture. These feature vectors are compared with standard features in a feature database. The module finds the highest similarity to the GB / T standard feature vectors, thus determining the GB / T standard as the match. This match is transmitted via a communication bus to an automatic output parameter adjustment module, which automatically adjusts the charging current and voltage according to the GB / T standard.

[0087] After automatically recognizing the adapter type, the charging station can quickly adjust its output parameters, reducing user waiting time and improving charging efficiency. Support for automatic adapter recognition across multiple charging standards allows the charging station to be compatible with more types of electric vehicles, expanding the range of charging services. By accurately matching the adapter type, the charging station ensures that its output parameters match the charging needs of the electric vehicle, avoiding safety hazards such as overcharging and overcurrent. The automatic recognition and adjustment function reduces the need for manual intervention, lowering maintenance costs and workload. Users do not need to manually select the charging standard; the charging station automatically recognizes and adjusts, improving user convenience and satisfaction.

[0088] In a preferred embodiment of the present invention, the formula for calculating the similarity between each standard feature and the feature vector is as follows:

[0089]

[0090] Where S represents the similarity between two feature vectors; a, b, and h represent adjustment parameters; n represents the dimension of the feature vector; i represents the index variable; v i This represents the component of the eigenvector v in the i-th dimension; v s,i Represents the standard eigenvector v s The component in the i-th dimension; y represents the power parameter of the logarithmic transformation term; e represents the base of the natural logarithm; f represents the coefficient parameter of the exponential transformation term; θ represents a non-zero constant; z represents the power parameter of the fractional transformation term.

[0091] In this embodiment of the invention, the feature vector v to be compared and the standard feature vector v are determined. s Both vectors are n-dimensional, where n is the dimension of the feature vector.

[0092] For each dimension i (from 1 to n), compute the feature vector v and the standard feature vector vi. s The square of the difference along this dimension, i.e. (v i -v s,i) 2 Sum the squares of the differences in all dimensions to obtain the square of the Euclidean distance. Taking the square root of this sum yields the Euclidean distance. To perform a logarithmic transformation on the Euclidean distance, first add 1 to avoid the case where the logarithm is undefined.

[0093] Raising the logarithmic result to the power y is Multiply the result by the adjustment parameter a to obtain the result. Euclidean distance Multiply by the coefficient parameter f, and calculate the exponent of the product. Multiply the result by the adjustment parameter b.

[0094] Euclidean distance Divide by itself and add a non-zero constant θ to obtain Promoting the result to the power z yields... Multiply the result by the adjustment parameter h to obtain the result. Add the results of the three transformations above to obtain the denominator. Take the reciprocal of the denominator to obtain the similarity S.

[0095] By combining logarithmic, exponential, and fractional transformations, this formula comprehensively captures various similarity features between feature vectors, including nonlinear relationships and subtle differences. Adjusting the parameters a, b, h, power parameters y, z, and coefficient parameters f, θ provides high flexibility. The sum of a, b, and h equals 1, where a = 0.3, b = 0.3, and h = 0.4; the power parameters y = 2, z = 2, f = 2, and θ = 0.1. By adjusting these parameters, the formula can adapt to different data distributions and feature vector comparison tasks, working effectively for both high-dimensional sparse vectors and low-dimensional dense vectors. The fractional transformation term provides a smooth transition, allowing similarity to gradually change between near and far distances, avoiding abrupt changes in the similarity curve.

[0096] In a preferred embodiment of the present invention, based on the different standard adapters used, a flower pollination algorithm is employed to automatically detect and intelligently adjust the output current and voltage to adapt to the charging needs of different electric vehicles. This may include:

[0097] When an electric vehicle is connected to a charging station, the charging station's sensors detect the type of adapter connector used to identify the electric vehicle's charging standard.

[0098] The charging standard is analyzed to obtain the current and voltage range and charging speed requirements under the charging standard, and the parameters of the flower pollination algorithm are initialized, including the population size, number of iterations and fitness function.

[0099] The final current and voltage combination is searched using a flower pollination algorithm, where each flower represents a current and voltage combination;

[0100] For each flower, the fitness function evaluates the performance of the corresponding flower based on the electric vehicle's charging requirements, including charging speed, battery protection, and energy efficiency, in order to obtain the fitness function evaluation result.

[0101] Based on the fitness function evaluation results, an iterative search is performed, each flower is continuously adjusted, and the fitness function evaluation and iterative search process is repeated until the preset number of iterations is reached. The final combination of current and voltage is determined based on the fitness value of each flower.

[0102] Based on the final current and voltage combination, during the charging process, the charging pile continuously monitors the output of current and voltage and the battery status of the electric vehicle, including charge, temperature and internal resistance. If any abnormality occurs, including overheating, overcurrent, or battery failure, the charging pile will immediately take corresponding protective measures, including reducing output, cutting off power or issuing an alarm.

[0103] When an electric vehicle finishes charging or the user requests to disconnect via the charging station interface, the charging station stops outputting current and voltage, thus completing the charging process to accommodate the charging needs of different electric vehicles.

[0104] In this embodiment of the invention, when the charging pile is started, all sensors are initialized, including sensors for detecting the type of adapter connector. The charging interface is continuously monitored to detect whether the electric vehicle is connected. Once a connection is detected, the sensors read the physical or electronic identifier of the adapter connector, identify the adapter connector type through a built-in database or communication protocol, and thus determine the charging standard of the electric vehicle. Based on the identified adapter connector type, the corresponding charging standard, including current and voltage ranges, charging speed requirements, etc., is retrieved from the database or standard protocol. Parameters of the pollination algorithm are set, including population size (i.e., the number of candidate current and voltage combinations), number of iterations (the number of loops in the search process), and fitness function (an indicator used to evaluate the performance of current and voltage combinations). An initial population is randomly generated or generated based on historical data, with each individual representing a current and voltage combination. For each individual, its performance is evaluated using the fitness function, considering factors such as charging speed, battery protection, and energy efficiency. Based on the fitness evaluation results, the pollination algorithm (such as global and local pollination strategies) is applied to adjust the individuals in the population, generating new current and voltage combinations. The fitness evaluation and iterative search are performed repeatedly until the preset number of iterations is reached. Based on the fitness value of each individual, the population is sorted, and the individual with the highest fitness is selected as the final combination of current and voltage.

[0105] During charging, the system continuously monitors current and voltage output, as well as the electric vehicle's battery status (charge level, temperature, internal resistance). Thresholds are set to detect abnormalities such as overheating, overcurrent, or battery malfunction. Upon detection of an anomaly, immediate protective measures are taken, such as reducing output, cutting off power, or issuing an alarm, to ensure charging safety. The system monitors battery level to determine if charging completion conditions have been met and responds to user disconnection requests via the charging station interface. When charging completion conditions are met or a user request is received, current and voltage output are stopped, completing the charging process.

[0106] Assume an electric vehicle supporting the CCS1 standard is connected to a charging station. The charging station's sensors detect the connection and identify the adapter as CCS1 type. The charging station retrieves the charging parameters for the CCS1 standard from its database, finding that it supports a maximum current of 125A, a voltage range of 200V-500V, and fast charging. The flower pollination algorithm is set to a population size of 50 and 100 iterations, with the fitness function considering charging speed, battery temperature, and internal resistance changes. The algorithm begins its iterative search, evaluating the impact of each current and voltage combination on battery charging speed and safety in each iteration, adjusting the combination to optimize the fitness. After 100 iterations, the current and voltage combination with the highest fitness (e.g., 100A, 400V) is selected as the final output. During charging, the charging station continuously monitors the output and battery status, automatically reducing the output to protect the battery when the battery temperature approaches the safety limit. When the battery level reaches 90%, the charging station stops outputting power, completing the charging process.

[0107] By automatically searching for the optimal current and voltage combination using a flower pollination algorithm, the charging process is made intelligent and automated, improving charging efficiency and safety. It can identify and adapt to different electric vehicle charging standards, ensuring broad compatibility and adaptability of charging stations. Optimizing the current and voltage combination reduces energy waste and improves energy efficiency during charging. Real-time monitoring and anomaly detection mechanisms ensure the safety of the charging process, effectively preventing safety risks such as overheating, overcurrent, and battery failure. It quickly responds to user requests, providing convenient charging services and enhancing user experience and satisfaction. By collecting and analyzing charging data, algorithm parameters and charging strategies can be continuously optimized to further improve charging efficiency and safety.

[0108] In a preferred embodiment of the present invention, the formula for calculating the fitness function is:

[0109]

[0110] w1, w2, w3, w4, and w5 represent weighting coefficients, for example, w1 = 0.2, w2 = 0.2, w3 = 0.1, w4 = 0.3, and w5 = 0.2 respectively; α, β, γ, δ, and ∈ represent exponential parameters, for example, α = 2, β = 3, γ = 2, δ = 4, and ∈ = 2 respectively; V a Indicates the current charging voltage; V max Indicates the maximum voltage under the charging standard; T b Indicates the current battery temperature; T o Indicates the target operating temperature of the battery; T r Indicates the acceptable temperature range for the battery; I o Indicates the current output current; I max Indicates the maximum current under the charging standard; Ec E represents the energy consumed. t R represents the total energy capacity of the battery. b R represents the current internal resistance of the battery. max This indicates the maximum acceptable internal resistance of the battery.

[0111] In this embodiment of the invention, weighting coefficients w1, w2, w3, w4, w5 and exponential parameters α, β, γ, δ, ∈ are initialized based on the charging standard and the characteristics of the electric vehicle. These parameters need to be fine-tuned based on actual testing and experience. The maximum voltage V under the charging standard is set. max Maximum current I max The target operating temperature T of the battery o Acceptable temperature range T of the battery r Total energy capacity of the battery E t And the maximum acceptable value R of the battery internal resistance max During the charging process, the charging station continuously collects the current charging voltage V. a Current output current I o Current battery temperature T b Energy consumed E c and the current internal resistance R of the battery b Using the collected real-time data and initialized parameters, the fitness function F is calculated:

[0112] Voltage adaptability section Assess the utilization of the current voltage relative to the maximum voltage.

[0113] Temperature adaptability section Assess how close the battery temperature is to the target temperature, and consider the acceptable temperature range.

[0114] Current adaptability section Assess the utilization of the current current relative to the maximum current and use it as a negative term to limit excessive current.

[0115] Energy efficiency fitness section Assess the percentage of energy consumed relative to the total energy and treat it as a negative item to encourage efficient charging.

[0116] Internal resistance fitness section Assess the extent to which the battery's current internal resistance is relative to its acceptable maximum value.

[0117] The fitness values ​​of each component are weighted and summed using weighted coefficients w1, w2, w3, w4, and w5 to obtain the final fitness function value F. In each iteration of the flower pollination algorithm, the calculated fitness function value F is used to evaluate the performance of each current and voltage combination. Based on the fitness evaluation results, the current and voltage combinations are adjusted to search for a better solution. Based on the optimal current and voltage combination found, the output of the charging station is adjusted to meet the charging needs of the electric vehicle. By comprehensively considering multiple factors such as voltage, current, temperature, energy efficiency, and internal resistance, the fitness function guides the algorithm to search for a better current and voltage combination, thereby improving charging efficiency. The temperature and internal resistance components in the fitness function ensure that the battery maintains a suitable temperature and internal resistance range during charging, which helps extend battery life and maintain battery performance.

[0118] By identifying the charging standards of different electric vehicles and adjusting charging parameters accordingly, charging stations can be compatible with multiple electric vehicles, improving the utilization rate of charging facilities. The current and temperature components in the fitness function limit excessive current and temperature, preventing safety accidents during charging and ensuring charging safety. By considering energy efficiency factors, the fitness function guides the algorithm to search for more energy-efficient charging solutions, reducing energy waste and improving energy utilization efficiency. Through intelligent adjustment of charging parameters, charging stations can provide faster and safer charging services, enhancing user experience and satisfaction. Furthermore, compatibility with multiple electric vehicle charging standards also contributes to improving the convenience and ease of use of charging facilities.

[0119] In a preferred embodiment of the present invention, intelligent identification and management of connector status, and real-time monitoring of connector usage, location, and damage, enables self-service borrowing and returning of adapter connectors to obtain maintenance reports, which may include:

[0120] Each connector is equipped with a unique identifier and its basic information is initialized, including type, specifications, production date, and initial condition.

[0121] Various sensors are deployed in the application scenarios of the connector, including counters, positioners, and damage detection sensors, and the usage data, location data, and damage data of the connector are collected in real time through the sensors;

[0122] The usage data, location data, and damage data of the joints are trained and optimized using a pre-defined random forest model to obtain joint status identification results and generate predictive maintenance suggestions.

[0123] When a user selects a connector to borrow through the interactive interface, the user's identity is verified, the connector status is updated to "borrowing", and the borrowing time and user information are recorded; when the user returns the connector, the user's identity is verified again, the connector status is updated to "available", and the return time and connector usage are recorded.

[0124] Based on real-time and historical data, the status of the joints is analyzed regularly to generate maintenance reports. The reports include statistics on the frequency of joint use, location distribution maps, damage analysis, and predictive maintenance recommendations.

[0125] In this embodiment of the invention, a database table is designed to store the unique identifier (such as a QR code, RFID tag, etc.) and basic information (type, specifications, production date, initial status, etc.) of the connector. When a connector is produced or put into storage, a unique identifier is assigned to each connector, and its basic information is entered into the database. Based on the requirements of the connector's usage scenario, an appropriate sensor type (such as a counter, locator, damage detection sensor, etc.) is selected. Sensors are installed on the connector or related equipment, and it is ensured that the sensors are functioning properly to collect usage data, location data, and damage data of the connector in real time.

[0126] Develop a sensor data acquisition program to periodically or in real-time read data from sensors. Send the acquired data to the data processing center for storage and analysis. Collect historical data, including joint usage, location, and damage data, along with corresponding joint status labels. Use this data to train a random forest model, adjust model parameters to optimize performance, and deploy the trained model to the data processing center for real-time joint status prediction. When new sensor data is received, use the trained random forest model to make predictions and obtain joint status identification results. Based on the joint status identification results and preset rules, generate predictive maintenance suggestions (such as regular inspections and component replacement).

[0127] On the production line, each connector is labeled with a QR code containing its unique identifier and basic information. Usage counters, positioners, and damage detection sensors are installed at key locations along the line. Counters record the number of times a connector is used, positioners track its location in real time, and damage detection sensors monitor its physical condition. A sensor data acquisition program periodically reads data from the sensors and sends it to a data processing center. A random forest model is trained using historical data and its performance is continuously optimized based on new data. When new sensor data is received, the model predicts the connector's condition (e.g., normal, damaged, impending damage) and generates maintenance recommendations (e.g., immediate replacement, periodic inspection).

[0128] Automated and intelligent connector management reduces the need for manual intervention and improves management efficiency. Predictive maintenance recommendations enable timely detection and handling of connector problems, avoiding production interruptions and additional maintenance costs caused by connector failures. Real-time monitoring of connector status allows for the timely detection and handling of potential safety hazards, ensuring the safe operation of the production line. Analyzing connector usage frequency and location distribution allows for optimization of connector layout and usage strategies, improving connector utilization. Self-service borrowing and returning functions enable users to conveniently and quickly obtain and return connectors, enhancing the user experience. By collecting and analyzing connector data, strong data support can be provided for management decisions, helping management make more informed choices.

[0129] In another preferred embodiment of the present invention, a preset random forest model is used to train and optimize the usage data, location data, and damage data of the joints to obtain joint status identification results and generate predictive maintenance suggestions, which may include:

[0130] Collect usage data, location data, and damage data of the connectors, and divide the collected data into training set and test set;

[0131] Based on the training and test sets, construct the basic framework of the random forest model and set the parameters of the random forest model, including the number of decision trees, maximum depth, and minimum number of sample splits.

[0132] The random forest model is initially trained using the training set to obtain preliminary training results. Based on these results, the random forest model is then optimized and adjusted, including increasing or decreasing the number of decision trees and adjusting the feature selection strategy, to obtain an optimized random forest model.

[0133] The optimized random forest model is used to predict the test set data. Each decision tree produces a prediction result, and the prediction results of all decision trees are combined to obtain the final joint state identification label.

[0134] Based on the final connector status identification tag, charging pile operation data, and historical maintenance records, the connector status is analyzed to obtain predictive maintenance recommendations, including recommendations for connector repair, replacement, and inspection, as well as corresponding maintenance times and priorities.

[0135] In this embodiment of the invention, the charging pile's sensors collect usage data (such as charging times, charging duration, charging current, etc.), location data (such as the specific location of the connector on the charging pile), and damage data (such as the time and type of connector damage). The collected data is cleaned to remove outliers and duplicates, ensuring accuracy and consistency. The cleaned data is divided into training and testing sets according to a certain ratio. The training set is used for model training and parameter optimization, while the testing set is used for model validation and performance evaluation. An empty random forest model framework is initialized, defining a set of decision trees within it to store multiple decision trees generated during training. Basic parameters of the random forest model are set; these parameters affect the model's training process and final performance. For example, the number of decision trees determines the ensemble size of the model, the maximum depth controls the tree complexity, and the minimum number of sample splits prevents overfitting.

[0136] The random forest model is initially trained using the training set data. During training, each decision tree splits and grows based on input features (connector usage data, location data, and damage data) and labels (connector status). The performance of the initially trained model is evaluated using the test set data, calculating metrics such as accuracy, recall, and F1 score. Based on the initial training results and performance evaluation metrics, the random forest model is optimized and adjusted. For example, if the model's accuracy is low, consider increasing the number of decision trees or adjusting the feature selection strategy; if the model's complexity is too high, consider reducing the maximum depth of the decision trees or increasing the minimum number of sample splits.

[0137] An optimized random forest model is used to predict the test set data. Each decision tree independently produces a prediction result (i.e., a joint state label) based on the input features. The final joint state identification label is obtained by combining the prediction results of all decision trees and using methods such as weighted averaging. The weighted averaging method refers to averaging the prediction results according to the weights of each decision tree.

[0138] By combining the final connector status identification tag, charging pile operational data (such as current, voltage, charging time, etc.), and historical maintenance records, a thorough analysis of the connector status is conducted. For example, if the connector status is identified as damaged, further analysis of the type, cause, and possible consequences of the damage is required. Based on the analysis results and the actual situation, predictive maintenance recommendations are generated. For example, if the connector is severely damaged, immediate replacement should be recommended; if the connector shows minor damage or signs of aging, regular inspection and maintenance can be recommended; if the connector status is normal, routine inspections can be recommended according to the schedule. Furthermore, the appropriate maintenance time and priority need to be determined based on the urgency and importance of the maintenance.

[0139] Assume the charging station is equipped with various sensors to collect real-time data on connector usage (e.g., number of charging sessions, charging duration), location (e.g., connector position within the charging station), and damage (e.g., whether the connector is damaged, and the extent of damage). Using the collected data as a training set, the parameters of a random forest model are set (e.g., number of decision trees = 100, maximum depth = 10, minimum number of splits = 2) for initial training. Based on the initial training results, the model showed low accuracy in identifying slightly damaged connectors. Therefore, the number of decision trees was increased to 200, the feature selection strategy was adjusted, redundant features were removed, and the model was retrained. The optimized model was used to predict the connector status on the test set data, yielding connector status identification labels. Compared with the true labels on the test set, the model's accuracy improved to 95%. Based on the connector status identification labels, charging station operation data, and historical maintenance records, predictive maintenance recommendations are generated. For example, for connectors identified as impending damage, it is recommended to check and maintain them before the next charging session; for already damaged connectors, it is recommended to replace them immediately.

[0140] By optimizing the parameters and feature selection strategy of the random forest model, the accuracy of connector status identification has been improved, reducing false positives and false negatives. Predictive maintenance suggestions allow for timely detection and handling of connector problems, preventing charging interruptions and additional maintenance costs due to connector failure. Accurate connector status identification ensures the normal operation of charging stations, reduces charging delays and waiting times caused by connector issues, and improves charging efficiency. Timely detection and handling of damaged connectors can prevent safety accidents caused by connector failures, protecting the safety of users and charging stations. Analyzing connector status identification tags, charging station operation data, and historical maintenance records provides strong data support for management decisions, helping management make more informed choices.

[0141] In a preferred embodiment of the present invention, remote monitoring, fault diagnosis, and data statistical analysis of charging piles are achieved through a remote server, which may include:

[0142] Based on the charging pile's operational data, potential faults are predicted and diagnosed using a fault diagnosis model on a remote server, and fault alarm information is generated.

[0143] The remote server collects, organizes, and analyzes the charging pile's operational data, including charging volume, charging duration, connector status, and ambient temperature, generating various reports and charts, including charging volume statistics reports and failure rate analysis charts. It also provides customized data statistical analysis services, including charging volume statistics for specific time periods and failure rate analysis for specific failure types.

[0144] In this embodiment of the invention, the charging pile transmits operational data (such as charging amount, charging duration, connector status, ambient temperature, etc.) to a remote server in real time via a built-in communication module (such as 4G / 5G, Wi-Fi, etc.). The remote server is equipped with a data receiving interface to ensure stable and efficient reception of data sent by the charging pile. A fault diagnosis model is deployed on the remote server, and historical fault data is used to train the model, optimize model parameters, and improve the model's predictive and diagnostic capabilities. When operational data from the charging pile is received, the data is input into the fault diagnosis model, which predicts and diagnoses potential faults and generates fault alarm information.

[0145] Based on the output of the fault diagnosis model, the system determines whether there are potential faults in the charging piles. If a fault is found, a fault alarm is generated, including detailed information such as fault type, fault level, and fault location. The alarm information is promptly sent to relevant personnel via SMS, email, and app push notifications for timely handling. The remote server cleans and preprocesses the received charging pile operation data, removing outliers and noise to ensure data quality. The cleaned data is then organized according to specific formats and rules and stored in a database. Data analysis tools or algorithms are used to analyze the organized data and extract valuable information. Based on the data analysis results, various reports and charts are generated, such as charging volume statistics reports and fault rate analysis charts. These reports and charts should clearly and intuitively display the charging pile's operation and fault status for easy understanding and analysis by relevant personnel.

[0146] The remote server provides a customized data statistical analysis service interface, allowing users to select specific time periods, fault types, and other parameters for queries. Based on the user-selected parameters, the server filters and analyzes the data in the database, generating corresponding reports and charts. These reports and charts are then returned to the user via an API interface or a web page.

[0147] The charging piles within the charging station transmit their operational data in real-time to a remote server via a 4G network. A fault diagnosis model based on a random forest is deployed on the remote server, trained using historical fault data. Upon receiving operational data from the charging piles, the model predicts and diagnoses potential faults. If the fault diagnosis model detects a potential fault (such as battery overheating or damaged charging interface), it generates a fault alarm and sends it to the charging station's management personnel via SMS. The remote server cleans, organizes, and analyzes the received charging pile operational data, extracting valuable information. Based on the data analysis results, it generates charging volume statistics reports and fault rate analysis charts. The reports display the daily changes in charging volume at the charging station, while the charts show the fault rate distribution for different types of faults. Charging station management personnel can query specific time periods and fault types through the remote monitoring system's web page. For example, they can select to query battery overheating faults from the past month, and the server will generate corresponding reports and charts based on the selection.

[0148] Remote monitoring and fault diagnosis models enable timely detection of potential charging pile faults and generate fault alarms, improving fault response speed and reducing charging interruption time caused by faults. Data analysis and customized services provide a more accurate understanding of the charging pile's operational status and fault conditions, offering targeted maintenance suggestions to maintenance personnel and reducing maintenance costs. Accurately identifying charging pile faults ensures normal operation, reduces charging delays and waiting times caused by faults, and improves charging efficiency. Timely detection and handling of potential charging pile faults can prevent safety accidents caused by faults, protecting the safety of users and charging stations. Data analysis and report generation provide strong data support for charging station management decisions, helping management make more informed decisions, such as optimizing charging pile layout and improving charging efficiency.

[0149] In another preferred embodiment of the present invention, predicting and diagnosing potential faults using a fault diagnosis model on a remote server and generating fault alarm information may include:

[0150] The charging pile's operating data, including current, voltage, power, temperature, and charging time, is transmitted to a remote server.

[0151] The fault diagnosis model on the remote server predicts and diagnoses the operating data of the charging pile to obtain the predicted status of the charging pile, so as to determine whether there is any abnormality or fault in the charging pile.

[0152] Based on the predicted status of the charging pile, if an abnormality or fault is detected in the charging pile, the type of abnormality or fault is determined, including electrical fault, mechanical fault, and communication fault.

[0153] The fault diagnosis model analyzes the causes of the faults, including battery aging, poor connector contact, and poor heat dissipation. Based on the fault type and cause, the model generates fault alarm information, including the fault type, fault cause, time of fault occurrence, charging station identifier, and suggested solutions.

[0154] In this embodiment of the invention, the charging pile's built-in sensors collect real-time operational data such as current, voltage, power, temperature, and charging time. The charging pile packages the collected operational data according to a predefined protocol and format via its built-in communication module (e.g., 4G / 5G, Wi-Fi). The packaged data is transmitted to a remote server via a network. The server is configured with a data receiving interface to ensure stable and efficient reception of data sent by the charging pile. A fault diagnosis model is deployed on the remote server. The fault diagnosis model is pre-trained using a large amount of historical charging pile operational data and corresponding fault labels to optimize model parameters and improve the model's prediction and diagnostic capabilities. When the server receives operational data from the charging pile, it inputs the data into the fault diagnosis model, which then predicts and diagnoses the charging pile's state, obtaining a predicted state result.

[0155] The fault diagnosis model determines whether the charging pile has any abnormalities or faults based on the prediction results. If the prediction results indicate that the charging pile is abnormal or faulty, it proceeds to the next step of analyzing the fault type and cause. The fault diagnosis model further analyzes the type of abnormality or fault based on the charging pile's operating data and prediction results. Fault types include electrical faults (such as overcurrent, overvoltage, short circuit, etc.), mechanical faults (such as connector damage, wear of transmission components, etc.), and communication faults (such as data loss, communication interruption, etc.). Combining the charging pile's operating data and fault type, the fault diagnosis model analyzes the specific cause of the fault. Electrical faults may be caused by battery aging, circuit short circuits, etc.; mechanical faults may be caused by poor connector contact, wear of transmission components, etc.; communication faults may be caused by network instability, equipment failure, etc. Based on the fault type and cause, the fault diagnosis model generates fault alarm information, including the fault type, fault cause, time of fault occurrence, charging pile identifier (such as serial number, location information, etc.), and suggested solutions (such as replacing the battery, checking the connector, restarting the equipment, etc.). Fault alarm information is promptly sent to relevant personnel via SMS, email, APP push, etc., for timely handling.

[0156] The charging piles within the charging station transmit real-time operational data, including current, voltage, power, temperature, and charging time, to a remote server via a 4G network. A neural network-based fault diagnosis model is deployed on the remote server. Upon receiving operational data from a charging pile, the model predicts and diagnoses its status, detecting an abnormally high current in one pile. Based on the prediction, the fault diagnosis model determines that the charging pile has an electrical fault, further analyzing it to identify an overcurrent fault. Combining the charging pile's operational data, the model analyzes that the cause of the fault may be battery aging leading to increased internal resistance, resulting in the abnormally high current. The fault diagnosis model generates a fault alarm message, including the fault type (overcurrent fault), cause (battery aging), time of occurrence, charging pile identifier, and suggested solutions (battery replacement). The fault alarm message is sent to the charging station's management personnel via SMS, who promptly dispatch to the site to address the fault.

[0157] By using a fault diagnosis model on a remote server to predict and diagnose the operational data of charging piles, abnormal or faulty states of charging piles can be accurately identified, improving the accuracy of fault detection. The fault diagnosis model can promptly detect abnormalities or faults in charging piles and generate fault alarm information to send to relevant personnel, shortening fault response time and reducing charging interruption time caused by faults. Through the fault type, cause, and suggested solutions provided in the fault alarm information, maintenance personnel can more accurately troubleshoot and repair faults, reducing maintenance costs. Timely detection and handling of charging pile faults can prevent safety accidents caused by faults, ensuring the safety of users and charging stations. By analyzing the fault and operational data of charging piles, charging station managers can understand the usage and fault status of charging piles, providing data support for optimizing charging station operations. For example, the layout of charging piles can be adjusted based on fault data to improve charging efficiency.

[0158] like Figure 2 As shown, embodiments of the present invention also provide an intelligent management system for internationally universal charging piles, comprising:

[0159] The charging station integrates multiple standard adapters to support various electric vehicle charging standards;

[0160] When users borrow adapters, they can use online payment to enable self-service borrowing, and the system can intelligently identify and record the borrowing status of the adapters.

[0161] Users connect their electric vehicles to charging stations and select the corresponding charging standard. The system automatically detects the charging needs of the electric vehicles and obtains the detection results.

[0162] Based on the test results, the output parameters of the charging pile, including current and voltage, are intelligently adjusted to adapt to the selected charging specifications.

[0163] During the charging process, the charging station continuously monitors the charging status, including key parameters such as current, voltage, and charging time, and handles any abnormal situations.

[0164] Once charging is complete, the charging station records charging data, including charging time, charging amount, electricity consumption, and generates a corresponding charging report.

[0165] It should be noted that this method is the same as the method of the internationally universal charging pile mentioned above. All the implementation methods in the above internationally universal charging pile embodiments are applicable to this embodiment and can achieve the same technical effect.

[0166] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An internationally universal charging pile, characterized in that, include: Multi-standard compatible adapter module, used to integrate adapters for multiple standards to support electric vehicles of different standards to charge at charging stations; The connector identification and information transmission module is used to automatically identify the type of adapter connected to the charging pile through the adapter matching algorithm. If the adapter type is identified, the adapter type is transmitted to the automatic output parameter adjustment module. An automatic output parameter adjustment module is used to automatically detect and intelligently adjust the output current and voltage based on different standard adapters, using a pollination algorithm to adapt to the charging needs of different electric vehicles, including: For each flower, the fitness function evaluates the performance of the corresponding flower based on the charging requirements of the electric vehicle, including charging speed, battery protection, and energy efficiency, to obtain the fitness function evaluation result. Based on the fitness function evaluation result, an iterative search is performed to continuously adjust each flower, and the fitness function evaluation and iterative search process is repeated until the preset number of iterations is reached. The final combination of current and voltage is determined based on the fitness value of each flower. The intelligent management module is used to intelligently identify and manage the status of connectors, and to enable self-service borrowing and returning of adapters by monitoring the use, location and damage of connectors in real time, so as to obtain maintenance reports; The remote monitoring module is used to remotely monitor, diagnose faults, and perform data statistical analysis of charging piles via a remote server. The user interaction module is used to view charging status, cost information, and operation guides for borrowing and returning adapters. It also supports users in initiating charging and making payments.

2. The internationally universal charging pile according to claim 1, characterized in that, The adapter matching algorithm automatically identifies the adapter type of the charging station. If the adapter type is identified, it is transmitted to the automatic output parameter adjustment module, including: The charging port of the charging pile continuously monitors the port status through sensors. When the sensor detects that the adapter is connected, the charging pile’s built-in camera captures an image of the adapter and extracts key features from the image, including shape, size and texture, to form a feature vector. The feature vector is compared with the standard features in the feature database, the similarity between each standard feature and the feature vector is calculated, and the matching joint type, i.e. the matching item, is determined based on the similarity. The matching items are transmitted to the automatic output parameter adjustment module via the internal communication bus of the charging pile.

3. The internationally universal charging pile according to claim 2, characterized in that, Based on the different standard adapters used, the system automatically detects and intelligently adjusts the output current and voltage using a pollination algorithm to adapt to the charging needs of different electric vehicles, including: When an electric vehicle is connected to a charging station, the charging station's sensors detect the type of adapter connector used to identify the electric vehicle's charging standard. The charging standard is analyzed to obtain the current and voltage range and charging speed requirements under the charging standard, and the parameters of the flower pollination algorithm are initialized, including the population size, number of iterations and fitness function. The final current and voltage combination is searched using a flower pollination algorithm, where each flower represents a current and voltage combination.

4. The internationally universal charging pile according to claim 3, characterized in that, Based on the different standard adapters used, the system automatically detects and intelligently adjusts the output current and voltage using a pollination algorithm to adapt to the charging needs of different electric vehicles, including: Based on the final current and voltage combination, during the charging process, the charging pile continuously monitors the output of current and voltage and the battery status of the electric vehicle, including charge, temperature and internal resistance. If any abnormality occurs, including overheating, overcurrent, or battery failure, the charging pile will immediately take corresponding protective measures, including reducing output, cutting off power or issuing an alarm. When an electric vehicle finishes charging or the user requests to disconnect via the charging station interface, the charging station stops outputting current and voltage, thus completing the charging process to accommodate the charging needs of different electric vehicles.

5. The internationally universal charging pile according to claim 4, characterized in that, The system intelligently identifies and manages connector status, and enables self-service borrowing and returning of adapters by monitoring connector usage, location, and damage in real time, generating maintenance reports, including: Each connector is equipped with a unique identifier and its basic information is initialized, including type, specifications, production date, and initial condition. Various sensors are deployed in the application scenarios of the connector, including counters, positioners, and damage detection sensors, and the usage data, location data, and damage data of the connector are collected in real time through the sensors; The usage data, location data, and damage data of the joints are trained and optimized using a pre-defined random forest model to obtain joint status identification results and generate predictive maintenance suggestions. When a user selects a connector to borrow through the interactive interface, the user's identity is verified, the connector status is updated to "borrowing", and the borrowing time and user information are recorded; when the user returns the connector, the user's identity is verified again, the connector status is updated to "available", and the return time and connector usage are recorded. Based on real-time and historical data, the status of the joints is analyzed regularly to generate maintenance reports. The reports include statistics on the frequency of joint use, location distribution maps, damage analysis, and predictive maintenance recommendations.

6. The internationally universal charging pile according to claim 5, characterized in that, The system trains and optimizes the joint's usage, location, and damage data using a pre-defined random forest model to obtain joint status identification results and generate predictive maintenance suggestions, including: Collect usage data, location data, and damage data of the connectors, and divide the collected data into training set and test set; Based on the training and test sets, construct the basic framework of the random forest model and set the parameters of the random forest model, including the number of decision trees, maximum depth, and minimum number of sample splits. The random forest model is initially trained using the training set to obtain preliminary training results. Based on these results, the random forest model is then optimized and adjusted, including increasing or decreasing the number of decision trees and adjusting the feature selection strategy, to obtain an optimized random forest model. The optimized random forest model is used to predict the test set data. Each decision tree produces a prediction result, and the prediction results of all decision trees are combined to obtain the final joint state identification label. Based on the final connector status identification tag, charging pile operation data, and historical maintenance records, the connector status is analyzed to obtain predictive maintenance recommendations, including recommendations for connector repair, replacement, and inspection, as well as corresponding maintenance times and priorities.

7. The internationally universal charging pile according to claim 6, characterized in that, Remote monitoring, fault diagnosis, and data statistical analysis of charging piles are achieved through a remote server, including: Based on the charging pile's operational data, potential faults are predicted and diagnosed using a fault diagnosis model on a remote server, and fault alarm information is generated. The remote server collects, organizes, and analyzes the charging pile's operational data, including charging volume, charging duration, connector status, and ambient temperature, generating various reports and charts, including charging volume statistics reports and failure rate analysis charts. It also provides customized data statistical analysis services, including charging volume statistics for specific time periods and failure rate analysis for specific failure types.

8. The internationally universal charging pile according to claim 7, characterized in that, Based on the charging pile's operational data, potential faults are predicted and diagnosed using a fault diagnosis model on a remote server, and fault alarm information is generated, including: The charging pile's operating data, including current, voltage, power, temperature, and charging time, is transmitted to a remote server. The fault diagnosis model on the remote server predicts and diagnoses the operating data of the charging pile to obtain the predicted status of the charging pile, so as to determine whether there is any abnormality or fault in the charging pile. Based on the predicted status of the charging pile, if an abnormality or fault is detected in the charging pile, the type of abnormality or fault is determined, including electrical fault, mechanical fault, and communication fault. The fault diagnosis model analyzes the causes of the faults, including battery aging, poor connector contact, and poor heat dissipation. Based on the fault type and cause, the model generates fault alarm information, including the fault type, fault cause, time of fault occurrence, charging station identifier, and suggested solutions.

9. An intelligent management system for an internationally compatible charging pile, comprising an internationally compatible charging pile as described in any one of claims 1 to 8, characterized in that, The system implements the following functions: The charging station integrates multiple standard adapters to support various electric vehicle charging standards; When users borrow adapters, they can use online payment to enable self-service borrowing, and the system can intelligently identify and record the borrowing status of the adapters. Users connect their electric vehicles to charging stations and select the corresponding charging standard. The system automatically detects the charging needs of the electric vehicles and obtains the detection results. Based on the test results, the output parameters of the charging pile, including current and voltage, are intelligently adjusted to adapt to the selected charging specifications. During the charging process, the charging station continuously monitors the charging status, including key parameters such as current, voltage, and charging time, and handles any abnormal situations. Once charging is complete, the charging station records charging data, including charging time, charging amount, electricity consumption, and generates a corresponding charging report.

Images