Search

CN-122026156-A - Global communication intelligent converter based on multi-mode AI and control method thereof

CN122026156ACN 122026156 ACN122026156 ACN 122026156ACN-122026156-A

Abstract

The application relates to the technical field of embedded intelligent control, and provides a global intelligent converter based on multi-mode AI and a control method thereof. The method is applied to the global intelligent converter, the converter comprises a physical plug conversion module, a power conversion module, an embedded processing module, a global communication module and a multi-mode sensing and interaction module, the physical plug conversion module is controlled to be matched with socket standards of a target area, the power conversion module is controlled to complete the adaptive conversion of preset wide input voltage, safe power supply is provided for an external load and the global intelligent converter, the global communication module is controlled to search and register a mobile communication network accessed to the target area, the embedded processing module drives the multi-mode sensing and interaction module to collect environment and user interaction data, local processing of voice recognition, image recognition and semantic understanding is completed, and corresponding multi-language voice interaction, visual recognition service and satellite positioning navigation service are executed according to received trigger interaction instructions.

Inventors

  • WANG GUIMING
  • XU ZEBIN

Assignees

  • 珠海市钛芯动力科技股份有限公司

Dates

Publication Date
20260512
Application Date
20260413

Claims (10)

  1. 1. The global intelligent conversion method based on the multi-mode AI is applied to a global intelligent converter and is characterized by comprising a physical plug conversion module, a power conversion module, an embedded processing module, a global communication module and a multi-mode sensing and interaction module, wherein the method comprises the following steps: The physical plug conversion module is controlled to adapt to socket standards of a target area, the power conversion module is controlled to complete the adaptive conversion of preset wide-range input voltage, and safe power supply is provided for an external load and the global intelligent converter; The global communication module is controlled to search and register a mobile communication network accessed to a target area, and the embedded processing module drives the multi-mode sensing and interaction module to acquire environment and user interaction data so as to complete local processing of voice recognition, image recognition and semantic understanding; and executing one or more of corresponding multilingual voice interaction, visual recognition service and satellite positioning navigation service according to the received trigger interaction instruction, and monitoring and controlling the power supply state of the global intelligent converter in real time.
  2. 2. The method of claim 1, wherein the controlling the physical plug conversion module to adapt to the socket standard of the target area comprises: The method comprises the steps of collecting appearance image data of a target socket through a multi-mode sensing and interaction module, inputting the collected image data into a socket standard identification model pre-trained in an embedded processing module, and identifying and obtaining socket types and electrical parameter standards corresponding to a target area; according to the recognized socket standard, a plug driving control instruction is generated and sent to a driving unit of the physical plug conversion module, and the telescopic pin assembly is controlled to adjust the pitch, the form and the extension length of pins, so that the mechanical adaptation with the target socket standard is completed; And collecting pin access feedback signals of the physical plug conversion module, and locking the current state of the pin assembly after the mechanical adaptation is confirmed.
  3. 3. The method of claim 1, wherein the controlling the power conversion module to perform the adaptive conversion of the preset wide input voltage to provide the external load and the global intelligent converter with the safe power comprises: The sampling unit of the power conversion module is used for collecting alternating voltage parameters and frequency parameters of an input side in real time, inputting the collected alternating voltage parameters and frequency parameters into a power adaptation control algorithm of the embedded processing module, and matching to obtain rectification, voltage regulation and voltage stabilization control parameters corresponding to the input parameters; generating a pulse width modulation driving control signal according to the control parameters obtained by matching, and sending the pulse width modulation driving control signal to a power conversion unit of a power conversion module to finish conversion from a wide input voltage to a preset rated output voltage and frequency; and collecting voltage, current and power parameters of an output side, generating a turn-off control instruction when the voltage, current and power parameters are detected to exceed a preset safety threshold, cutting off power supply of a corresponding output loop, and outputting an alarm prompt through a multi-mode sensing and interaction module.
  4. 4. The method of claim 1, wherein the controlling the global communication module to search for and register for a mobile communication network accessing the target area comprises: Acquiring current satellite positioning data through an embedded processing module, matching to obtain current target region information, and calling prestored mobile communication network system, frequency band and operator priority configuration parameters corresponding to the region; according to the acquired mobile communication network system, frequency band and operator priority configuration parameters, controlling a radio frequency unit of a global communication module to sequentially scan mobile communication network signals of corresponding frequency bands, and acquiring signal strength, network delay and tariff related parameters of each available network; inputting the acquired signal strength, network delay and tariff related parameters into a pre-trained network optimization decision model, outputting an optimal access network, controlling a global communication module to finish registration and access of the optimal access network, and establishing a data transmission link between an embedded processing module and the Internet.
  5. 5. The method according to claim 1, wherein the local processing for performing speech recognition, image recognition and semantic understanding by driving the multi-modal sensing and interaction module to collect environment and user interaction data via the embedded processing module comprises: The audio acquisition unit, the image acquisition unit and the touch control interaction unit of the multi-mode sensing and interaction module are driven by the embedded processing module to respectively acquire voice interaction data, environment image data and touch control instruction data of a user; The method comprises the steps of carrying out noise reduction, endpoint detection and feature extraction on collected voice data, inputting extracted voice features into a locally deployed multilingual voice recognition model to finish conversion from voice content to text information, carrying out pretreatment and feature extraction on the collected image data, inputting the collected image data into a locally deployed lightweight image recognition model to finish recognition of a target object, characters and scenes; and inputting the converted text information and the recognized image semantic information into a locally deployed multilingual large-language semantic understanding model to complete recognition and semantic analysis of user intention, and outputting a corresponding structured interaction instruction.
  6. 6. The method of claim 1, wherein the performing one or more of a corresponding multilingual voice interaction, visual recognition service, and satellite positioning navigation service according to the received trigger interaction instruction comprises: After a trigger interaction instruction of a user is obtained through semantic analysis, a corresponding functional unit is called according to a service type corresponding to the instruction; When the service type is multilingual voice interaction, converting source language text information input by a user into target language text through a preset multilingual machine translation model, converting the target language text into voice data of corresponding languages through a voice synthesis model, outputting the voice data through an audio playing unit of a multimodal perception and interaction module, and simultaneously completing bidirectional real-time dialogue interaction; When the service type is visual identification service, the image acquisition unit is driven to acquire a target image, identification of target characters and objects is completed through the image identification model, language conversion is completed through the translation model, and converted contents are synchronously output with the audio playing unit through the display unit; When the service type is satellite positioning navigation service, current satellite positioning data is acquired, an optimal navigation path is generated through a path planning algorithm by combining destination information input by a user, and real-time voice navigation and picture guidance are completed through a multi-mode sensing and interaction module.
  7. 7. The method of claim 1, wherein the real-time monitoring and controlling of the power supply status of the universal intelligent converter comprises: the method comprises the steps that voltage, current, power and temperature parameters of an input side and an output side of a power conversion module and access state parameters of a physical plug conversion module are collected in real time through an embedded processing module, and the embedded processing module is used for constructing a real-time power supply state data set; Inputting the real-time power supply state data set corresponding to the real-time power supply state data set into a pre-trained power supply abnormality detection model, and identifying to obtain a risk level corresponding to a preset abnormal state type, wherein the preset abnormal state type at least comprises one or more of overvoltage, overcurrent, overtemperature, short circuit and electric leakage; When the low risk abnormality is identified according to the risk level, an emergency shutdown instruction is generated, all output loops of the power conversion module are cut off, pin assemblies of the physical plug conversion module are locked, emergency alarm information is output through the multi-mode sensing and interaction module, and corresponding abnormal data is uploaded to a preset cloud management platform through the global communication module.
  8. 8. The method according to claim 1, wherein the method further comprises: Continuously acquiring interactive behavior data, equipment operation data and environment data and scene data of a region where the equipment is located of a user through an embedded processing module, and constructing a user-specific cross-border travel behavior data set; Inputting the constructed cross-border travel behavior data set into a user habit adaptation model based on federal learning training, finishing iterative updating of the cross-border travel behavior data set, and learning to obtain language preference, electricity utilization habit, interaction habit and travel scene preference of the user; When the user is detected to enter a new target area, a full-flow adaptation control instruction is generated based on the trained user habit adaptation model, automatic adaptation of a physical plug and automatic matching of power supply parameters, automatic optimal access of a communication network of the target area and automatic switching of interactive languages are sequentially completed, and local navigation, consumption and safety prompt information matched with the travel preference of the user is pushed.
  9. 9. The method according to claim 1, wherein the method further comprises: establishing a close range communication link and a remote data synchronous link with a user portable intelligent terminal through a global communication module, and acquiring the electric quantity state, the network state and the position data of the user intelligent terminal in real time; when the environment data collected by the multi-mode sensing and interaction module is combined through the embedded processing module, an emergency travel scene with weak network signals and/or insufficient terminal electric quantity of a user is identified, an emergency guarantee mode is started; In an emergency guarantee mode, quick charging and power supply is provided for a user intelligent terminal through a power supply conversion module, network sharing hot spot service is provided for the user intelligent terminal through a global communication module, surrounding environment data are continuously collected through a multi-mode fusion sensing algorithm, emergency risk avoidance path planning is completed by combining satellite positioning data, real-time emergency guidance is provided for a user through multi-language voice interaction, and real-time position and state information of the user are synchronized to a preset cloud account of an emergency contact.
  10. 10. The global intelligent converter based on the multi-mode AI is characterized by comprising a physical plug conversion module, a power conversion module, an embedded processing module, a global communication module and a multi-mode sensing and interaction module, wherein the global intelligent converter is used for realizing the method according to any one of claims 1-9; The intelligent power supply system comprises a physical plug conversion module, a power supply conversion module, a global communication module, an embedded processing module, a multi-mode sensing and interaction module, a user interaction module, a voice recognition and semantic understanding local processing module, a control power supply conversion module, a global communication module, a power supply control module, a control power supply control module and a control module, wherein the physical plug conversion module is controlled to adapt to socket standards of a target area, the control power supply conversion module is used for completing adaptation conversion of preset wide input voltage, providing safe power supply for an external load and a global intelligent converter, the global communication module is controlled to search and register a mobile communication network accessed to the target area, the embedded processing module is used for driving the multi-mode sensing and interaction module to collect environment and user interaction data so as to complete voice recognition, image recognition and semantic understanding local processing, and one or more of corresponding multi-language voice interaction, visual recognition service and satellite positioning navigation service are executed according to received triggering interaction instructions, and power supply states of the global intelligent converter are monitored and controlled in real time.

Description

Global communication intelligent converter based on multi-mode AI and control method thereof Technical Field The application relates to the technical field of embedded intelligent control, in particular to a global intelligent converter based on multi-mode AI and a control method thereof. Background With the penetration of global economy integration, the cross-border travel activities such as cross-border travel, international business transactions, involved engineering operations and the like continue to increase in scale, and the demands of users on cross-border travel matched equipment are rapidly upgraded from the traditional single power supply adaptation function to the integrated, intelligent and full-scene coverage direction. The global communication conversion equipment is used as a necessary basic tool for cross-border travel, and the technical upgrading and the functional optimization of the global communication conversion equipment become the core direction of industry attention. The prior general global communication conversion equipment has the core that only the physical plug conversion component and the basic power supply adaptation circuit are integrated, so that the basic power utilization adaptation problem caused by the incompatibility of socket mechanical standards and power grid voltage standards in different countries and regions can be solved, and at least the following core technical defects exist: 1. The mechanical adaptation requires users to inquire the socket standard of a target area in advance and manually adjust the pin structure, is complex in operation, is easy to cause adaptation errors due to unfamiliar users to the standard, even causes potential safety hazards such as poor contact and short circuit, and cannot realize automatic identification and adaptation; 2. The power supply management capability is weak, the basic broad-width voltage input adaptation is realized only by adopting a voltage regulating circuit with fixed topology, the real-time monitoring of the input and output all-link electric parameters is lacking, the passive fusing protection can be realized only by relying on fuses for abnormal working conditions such as overvoltage, overcurrent and overtemperature, the active early warning and graded management and control cannot be realized, and the obvious short plates exist for the electric safety protection; 3. The system has extremely single function, can only solve the basic power consumption requirement, can not solve the core pain points of language communication obstacle, difficult access of an overseas network, inconvenient navigation of strange environment and the like which are commonly existed in cross-border travel, and a user needs to additionally carry a plurality of terminals such as a carry-on WiFi, a translator, navigation equipment and the like, thereby greatly increasing travel burden and equipment management cost. Accordingly, a need exists for a method that addresses at least one of the problems described above. Disclosure of Invention The application provides a global communication intelligent converter based on a multi-mode AI and a control method thereof, which aim to solve the problems that the prior conventional global communication conversion equipment only integrates a physical plug conversion component and a basic power supply adaptation circuit, and only solves the basic power utilization adaptation problem caused by the incompatibility of socket mechanical standards and power grid voltage standards in different countries and regions. In a first aspect, an embodiment of the present application provides a global intelligent conversion method based on a multi-mode AI, which is applied to a global intelligent converter, where the global intelligent converter includes a physical plug conversion module, a power conversion module, an embedded processing module, a global communication module, and a multi-mode sensing and interaction module, and the method includes: The physical plug conversion module is controlled to adapt to socket standards of a target area, the power conversion module is controlled to complete the adaptive conversion of preset wide-range input voltage, and safe power supply is provided for an external load and the global intelligent converter; The global communication module is controlled to search and register a mobile communication network accessed to a target area, and the embedded processing module drives the multi-mode sensing and interaction module to acquire environment and user interaction data so as to complete local processing of voice recognition, image recognition and semantic understanding; and executing one or more of corresponding multilingual voice interaction, visual recognition service and satellite positioning navigation service according to the received trigger interaction instruction, and monitoring and controlling the power supply state of the global intelligent converter in real time. In some e