Control method of intelligent charging pile and related equipment

Abstract

The invention belongs to the technical field of charging equipment, and discloses a control method of an intelligent charging pile and related equipment. The control method of the intelligent charging pile comprises the steps of receiving and fusing a user instruction based on multi-modal interaction, analyzing a charging demand based on the user instruction, obtaining a real-time parameter according to the charging demand, inputting the real-time parameter into a constructed LSTM load prediction model, outputting a prediction result, and controlling the charging pile based on the prediction result. A control instruction is obtained, the prediction result and the control instruction are dynamically calculated through a multi-objective optimization function, the optimal charging output power is obtained, and charging is executed based on the optimal charging output power; according to the invention, the perception and interaction depth of the system can be obviously improved, comprehensive consideration and prospective regulation and control of the power grid state, the user demand and the energy are realized, and the charging strategy is converted from static preset to dynamic real-time optimization.

Description

Technical Field

[0001] This invention relates to the field of charging equipment technology, and specifically to a control method and related equipment for an intelligent charging pile. Background Technology

[0002] Charging piles are the core equipment for providing electric vehicles with electrical energy. They are often referred to as "gas stations for electric vehicles" and are widely distributed in residential areas, shopping malls, highway service areas, and other locations. Their core function is to convert AC power from the power grid into DC power that is compatible with the electric vehicle's battery, enabling safe and efficient charging.

[0003] Charging piles are energy replenishment stations for new energy vehicles. They convert grid power into electrical energy that can be stored in the vehicle's power battery through a dedicated interface to meet the vehicle's driving range requirements. According to the installation scenario, charging piles can be divided into home wall-mounted charging piles and public floor-mounted charging piles; according to the charging speed, they include slow charging piles (AC piles) and fast charging piles (DC piles) to meet daily energy replenishment and emergency fast charging needs.

[0004] However, current mainstream charging piles still have significant functional limitations. They focus only on basic charging needs and lack in-depth user interaction and energy coordination capabilities. Their charging strategies are rigid and difficult to adapt to personalized needs. Furthermore, the charging strategies lack dynamic adjustment capabilities, resulting in poor user charging experience, inefficient energy utilization, inability to balance grid load and distributed energy, damage to batteries, or delays in travel. This not only increases the cost of recharging for car owners but also restricts the coordinated development of charging piles and the new energy vehicle industry.

[0005] Therefore, there is an urgent need for a technology that can solve the problems of lacking deep user interaction and energy coordination capabilities, as well as the lack of dynamic adjustment capabilities in charging strategies. Summary of the Invention

[0006] The purpose of this invention is to provide a control method and related equipment for intelligent charging piles to overcome the problems existing in the prior art. This invention can significantly improve the perception and interaction depth of the system, and also realize the comprehensive consideration and forward-looking control of power grid status, user needs and energy, so that the charging strategy can be transformed from static preset to dynamic real-time optimization.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a control method for an intelligent charging pile, comprising the following steps: Step 1: Receive and integrate user commands based on multimodal interaction; Step 2: Analyze the charging demand based on the user's instructions and obtain real-time parameters according to the charging demand; Step 3: Input the real-time parameters into the constructed LSTM load prediction model and output the prediction results; Step 4: Control the charging pile based on the prediction results to obtain control commands; Step 5: The prediction results and control commands are dynamically calculated using a multi-objective optimization function to obtain the optimal charging output power; Step 6: Perform charging based on the optimal charging output power.

[0008] In some embodiments, receiving and fusing user instructions based on multimodal interaction specifically includes: Step 1.1: Enhance the language instructions from the speech recognition unit and then extract the key intents; Step 1.2: Arbitrate the conflict between the touch commands and setting commands from the touch display screen and the mobile APP respectively to obtain the operation commands; Step 1.3: Merge the key intents and operation instructions to obtain user instructions.

[0009] In some embodiments, the real-time parameters include grid load, photovoltaic energy storage, and vehicle status information.

[0010] In some embodiments, the construction of the LSTM load forecasting model includes: Step 3.1: Obtain historical parameters, including historical grid load, historical photovoltaic energy storage, and historical vehicle status information; Step 3.2: The historical parameters are cleaned, denoised, and normalized sequentially to obtain the supervised learning dataset; Step 3.3: Construct the LSTM load prediction model, which includes an input layer, one or more LSTM hidden layers, a Dropout layer, and an output layer connected in sequence. Step 3.4: Divide the supervised learning dataset into a training set and a validation set. Use the training set to train the LSTM load prediction model and use the validation set to fine-tune the hyperparameters of the LSTM load prediction model to minimize the loss function between the predicted and actual values. This will result in the constructed LSTM load prediction model, which can be used to predict the charging load in real time for specific future periods.

[0011] In some embodiments, controlling the charging pile based on the prediction results to obtain control commands specifically includes: When the forecast indicates that photovoltaic power supply is sufficient, switch to green energy. When the forecast indicates a low grid load, fast charging is initiated. When the forecast indicates a high grid load, the power-limited charging mode is activated. When the forecast indicates peak power demand, it is recommended to charge during off-peak hours.

[0012] In some embodiments, the formula of the multi-objective optimization function includes:

[0013] In the formula, Represents a multi-objective optimization function; Indicates the cost of the power grid; Indicates the weight of power grid costs; Indicates the user's waiting time; Indicates the weight of user waiting time; Indicates carbon emissions; This indicates the carbon emission weight.

[0014] In some embodiments, during the charging process, the temperature of the charging cable is monitored in real time, a charging report is generated based on the temperature of the charging cable, and the charging report is fed back through the user terminal.

[0015] Secondly, the present invention provides a control system for an intelligent charging pile, comprising: The multimodal interaction module is used to receive and integrate user commands based on multimodal interaction. The environmental perception module is used to parse charging needs based on user commands and obtain real-time parameters according to the charging needs; The prediction result output module is used to output the real-time parameters to the constructed LSTM load prediction model and output the prediction results. The charging pile control module is used to control the charging pile based on the prediction results and obtain control commands. The dynamic energy optimization module is used to dynamically calculate the optimal charging output power by combining the prediction results and control commands through a multi-objective optimization function. The charging module is used to perform charging based on the optimal charging output power.

[0016] Thirdly, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0018] The above technical solution has the following advantages or beneficial effects: Firstly, this invention provides a control method for intelligent charging piles. By deeply integrating user intent and environmental data through multimodal interaction technology, it significantly improves the perception and interaction depth of the system. By combining LSTM load forecasting with multi-objective dynamic optimization, it achieves comprehensive consideration and forward-looking regulation of grid status, user demand, and clean energy. This method transforms the charging strategy from static preset to dynamic real-time optimization, effectively solving the technical problems of insufficient energy coordination capability and rigid strategy. Ultimately, it improves charging efficiency by more than 20%, reduces operating costs by 30%, and reduces user operation steps by 50%.

[0019] In some embodiments, the multimodal instruction fusion method of the present invention effectively overcomes the reliability problem of voice interaction in noisy environments through voice enhancement and key intent extraction technologies, ensuring the accurate capture of core user instructions; by introducing a cross-channel conflict arbitration mechanism, it systematically solves the coordination problem of multi-source operation instructions such as touch and APP, avoiding misoperation caused by instruction conflicts; the method intelligently fuses ambiguous user language intent with precise device operation instructions to generate semantically clear and executable user instructions, making the human-computer interaction experience more natural, smooth and accurate, and achieving the beneficial effect of reducing user operation steps by 50%.

[0020] In some embodiments, the present invention introduces grid load, photovoltaic energy storage and vehicle status information as core real-time parameters to construct an integrated collaborative sensing network of "vehicle-charging pile-grid-storage". This method breaks through the limitation of traditional charging piles that only focus on the status of the charging pile end, and realizes global and accurate perception of the energy supply side, grid carrying capacity and user vehicle demand. It provides a complete data foundation for subsequent intelligent decision-making and dynamic power adjustment, and fundamentally improves the collaborative optimization capability of the system.

[0021] In some embodiments, the LSTM load forecasting model construction method of the present invention constructs a high-precision supervised learning dataset by integrating multi-source heterogeneous data such as historical grid load, photovoltaic energy storage, and vehicle status, providing a comprehensive information foundation for prediction. By adopting an LSTM network structure including a Dropout layer, model overfitting is effectively avoided, ensuring prediction generalization ability. This model can accurately learn the nonlinear variation law of charging load under complex spatiotemporal environments, and achieve advanced prediction of load for specific future periods. This technology upgrades charging management from passive response to proactive forward control, providing key decision-making basis for subsequent multi-objective dynamic optimization, and fundamentally improving the energy coordination and scheduling capabilities of charging piles.

[0022] In some embodiments, the present invention achieves precise linkage between charging strategy and energy status by directly mapping load forecasting results to forward-looking control commands. This method enables charging piles to proactively switch to green energy, guide grid charging during off-peak hours, and smooth peak loads, transforming load forecasting from a "predictive" capability into a "regulatory" action, significantly improving the system's absorption rate of renewable energy and the grid's proactive support capability.

[0023] In some embodiments, the present invention constructs a multi-objective optimization function for grid cost, user waiting time and carbon emissions, and introduces configurable weight coefficients to achieve a precise balance between economic efficiency, user experience and environmental benefits in the charging strategy. By quantifying complex operational objectives into solvable mathematical problems, the system can calculate the global optimal solution from conflicting objectives, thereby significantly improving overall social benefits while reducing operating costs by 30%.

[0024] In some embodiments, the present invention constructs a "monitoring-feedback" safety closed loop by monitoring cable temperature in real time during charging and generating a visual report. This mechanism transforms the hidden physical state into perceptible data, enabling users to intuitively grasp the charging safety status and receive early warnings in the early stages of abnormal temperature rise. This upgrades traditional passive protection to proactive prevention and transparent management, greatly improving the safety of the charging process and user trust.

[0025] Secondly, this invention provides a control system for intelligent charging piles. This intelligent charging pile control system constructs a complete intelligent closed loop of "perception-decision-execution" through the coordinated operation of five modules: multimodal interaction, environmental perception, load prediction, dynamic optimization, and safe execution. The system achieves fully automated operation from accurate understanding of user intent, multi-source data fusion perception, and forward-looking load prediction to multi-objective dynamic optimization decision-making. It not only fundamentally solves the problem of rigid charging strategies but also upgrades charging piles into edge computing nodes integrating intelligent interaction, energy coordination, and proactive safety, significantly improving the overall reliability, economy, and intelligence level of the system.

[0026] Thirdly, the present invention provides a computer device that, through a processor executing a specific computer program, can efficiently implement the steps of the method of the present invention. When performing data processing tasks, the computer device can accurately perform numerical calculations and logical judgments, avoiding errors caused by human factors. At the same time, since the computer program has high stability and reliability, it can ensure the accuracy and consistency of the data processing results.

[0027] Fourthly, the present invention provides a computer-readable storage medium. By programming the steps of the method of the present invention into a computer program and storing it on the computer-readable storage medium, users can easily load these programs onto any compatible computer device and execute them without rewriting or converting the code, which greatly improves the convenience and flexibility of program execution. Attached Figure Description

[0028] Figure 1 This is a schematic flowchart of a control method for a smart charging pile according to some embodiments of this specification; Figure 2 This is a schematic diagram of the control method framework for an intelligent charging pile according to some embodiments of this specification; Figure 3 This is a schematic diagram of the control system framework of a smart charging pile according to some embodiments of this specification; Figure 4 This is a schematic diagram of the structure of a computer device according to some embodiments of this specification. Detailed Implementation

[0029] The present invention will be further described in detail below with reference to specific embodiments. These descriptions are for explanation purposes only and are not intended to limit the scope of the invention. To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention. It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Example: This invention can significantly improve the perception and interaction depth of the system, and also realize the comprehensive consideration and forward-looking control of the power grid status, user needs and energy, so that the charging strategy can be transformed from static preset to dynamic real-time optimization.

[0031] This invention provides a control method for intelligent charging piles. Figure 1 This is a schematic flowchart illustrating a control method for a smart charging pile according to some embodiments of this specification. Figure 2 This is a schematic diagram of the control method for a smart charging pile according to some embodiments of this specification. The control method for the smart charging pile includes the following steps: Step 1: Receive and integrate user commands based on multimodal interaction, specifically including: Step 1.1, Language instruction priority processing: Enhance the language instructions from the speech recognition unit, and then extract the key intents; Step 1.2, Conflict arbitration between touch operation and APP settings: Conflict arbitration is performed on touch commands and setting commands from the touch screen and the mobile APP respectively to obtain operation commands; Step 1.3: Merge the key intents and operation instructions to obtain user instructions.

[0032] In some embodiments, the emergency stop cross-modal response mechanism is as follows: when an emergency stop command is received, the cross-modal response mechanism is activated, ignoring the priority of other commands, and immediately executing the stop charging operation.

[0033] Step 2: Analyze the charging demand based on the user's instructions and obtain real-time parameters according to the charging demand; In some embodiments, the real-time parameters include grid load, photovoltaic energy storage, and vehicle status information.

[0034] In some embodiments, the vehicle status information may include battery status information, charging capability status information, vehicle identity and configuration status information, and safety status information; wherein, the battery status information includes battery state of charge information, battery health status information, battery voltage / current status information, battery temperature status information, and battery internal resistance status information; the charging capability status information includes maximum allowable charging voltage / current / power status information, vehicle charging curve status information, and expected charging target status information; the vehicle identity and configuration status information includes vehicle identification number, battery type, and battery capacity; and the safety status information includes insulation resistance status information, contactor status information, and fault codes.

[0035] In some embodiments, battery state of charge (SOH) information, i.e., remaining capacity, is usually expressed as a percentage and is the most direct factor determining charging demand and duration; battery health status information reflects the overall health and capacity degradation of the battery; batteries with low SOH may require a more conservative charging strategy; battery voltage / current status information is the current real-time voltage and current values, which are key parameters for charging closed-loop control; battery temperature status information refers to the temperature of multiple temperature measurement points within the battery pack and is one of the most important safety indicators; excessively high temperatures will trigger power reduction or charging cessation; battery internal resistance status information reflects the internal characteristics of the battery; increased internal resistance will affect charging efficiency and heat generation.

[0036] In some embodiments, the maximum allowable charging voltage / current / power status information refers to the upper limit of charging that the vehicle battery can withstand, and the output of the charging pile must never exceed this limit; the vehicle charging curve status information refers to the ideal voltage-current curve recommended for the vehicle, and the charging pile will try to follow this curve to achieve optimal charging; and the desired charging target status information refers to the target set by the user.

[0037] In some embodiments, the vehicle identification number is used for authentication, personalization settings, and billing; the battery type is such as ternary lithium battery, lithium iron phosphate battery, etc., and the optimal charging strategy may differ for different types of batteries; the battery capacity is the total charge.

[0038] In some embodiments, insulation resistance status information refers to the insulation condition between the high-voltage system and the vehicle chassis; excessively low insulation resistance is a serious safety hazard. Contactor status information refers to whether the main positive and main negative contactors in the battery pack are properly engaged to ensure the correct connection of the charging circuit. Fault codes refer to current or historical fault information reported by the vehicle BMS.

[0039] Step 3: Input the real-time parameters into the constructed LSTM load prediction model and output the prediction results.

[0040] In some embodiments, the construction of the LSTM load forecasting model includes: Step 3.1: Obtain historical parameters, including historical grid load, historical photovoltaic energy storage, and historical vehicle status information; Step 3.2: The historical parameters are cleaned, denoised, and normalized sequentially to obtain the supervised learning dataset; Step 3.3: Construct a Long Short-Term Memory (LSTM) load prediction model, which includes an input layer, one or more LSTM hidden layers, a Dropout layer, and an output layer connected in sequence. Step 3.4: Divide the supervised learning dataset into a training set and a validation set. Use the training set to train the LSTM load prediction model and use the validation set to fine-tune the hyperparameters of the LSTM load prediction model to minimize the loss function between the predicted and actual values. This will result in the constructed LSTM load prediction model, which can be used to predict the charging load in real time for specific future periods.

[0041] In some embodiments, the loss function used is the mean squared error loss function, and the Adam optimizer is used to optimize the model parameters.

[0042] In some embodiments, the system further includes an online model update step: the dynamic energy management unit continuously collects new charging loads and related data; periodically or when the prediction error exceeds a set threshold, it uses the newly collected data to incrementally train or retrain the deployed LSTM load prediction model to update the model parameters.

[0043] In some embodiments, historical vehicle status information may include historical battery status information, historical charging capability status information, historical vehicle identity and configuration status information, and historical safety status information; wherein, historical battery status information includes historical battery state of charge information, historical battery health status information, historical battery voltage / current status information, historical battery temperature status information, and battery internal resistance status information; historical charging capability status information includes historical maximum allowable charging voltage / current / power status information, historical vehicle charging curve status information, and historical expected charging target status information; historical vehicle identity and configuration status information includes historical vehicle identification number, historical battery type, and historical battery capacity; historical safety status information includes historical insulation resistance status information, historical contactor status information, and historical fault codes.

[0044] Step 4: Control the charging pile based on the prediction results to obtain control commands, specifically including: When the forecast indicates that photovoltaic power supply is sufficient, switch to green energy. When the forecast indicates a low grid load, fast charging is initiated. When the forecast indicates a high grid load, the power-limited charging mode is activated. When the forecast indicates peak power demand, it is recommended to charge during off-peak hours.

[0045] Step 5: The prediction results and control commands are dynamically calculated using a multi-objective optimization function to obtain the optimal charging output power.

[0046] In some embodiments, the formula of the multi-objective optimization function includes:

[0047] In the formula, Represents a multi-objective optimization function; Indicates the cost of the power grid; Indicates the weight of power grid costs; Indicates the user's waiting time; Indicates the weight of user waiting time; Indicates carbon emissions; This indicates the carbon emission weight.

[0048] In some embodiments, the fuzzy control algorithm for charging power of the present invention can cope with the nonlinear characteristics of the charging process, the blockchain-based energy transaction records ensure the immutability and traceability of distributed energy transaction data such as photovoltaics, and the energy storage and charging strategy in the event of a sudden power outage automatically switches to the energy storage system to continue to power the vehicle, effectively ensuring the reliability of the charging process and the continuity of the user experience.

[0049] Step 6: Perform charging based on the optimal charging output power.

[0050] In some embodiments, during the charging process, the temperature of the charging cable is monitored in real time, a charging report is generated based on the temperature of the charging cable, and the charging report is fed back through the user terminal.

[0051] In some embodiments, before charging begins, the user is authenticated and authorized for automated payment via a biometric authentication module; during or after charging, an electronic invoice is pushed to the user via the multimodal interaction interface.

[0052] In some embodiments, extended human-computer interaction functions are also included: the display screen provides charging progress, cost estimate, and recommendations for nearby services (such as restaurants and parking lots); the voice assistant supports multilingual operation guidance; and the mobile APP allows remote reservation of charging time slots and obtaining electronic invoices.

[0053] This invention provides a control method for smart charging piles, applicable to emergency charging scenarios during heavy rain. Background: During the evening rush hour, a sudden downpour causes a power grid fluctuation warning. Users drive their electric vehicles to this smart charging pile hoping to quickly replenish their battery. The method includes the following steps: Step 1: Receive and integrate user commands based on multimodal interaction, specifically including: Voice command (high priority): When the user approaches the charging station, due to noise interference from wind and rain (low signal-to-noise ratio), they can use voice commands to say: Charge as much as you can now. Touch command: At the same time, the user can manually drag the charging level setting slider to 80% on the charging station's touch screen; Environmental data: The system's environmental perception module receives a power grid fluctuation warning, indicating that the current power grid stability is poor and there is a risk of tripping.

[0054] Language command priority processing: First, ambiguous voice commands are enhanced and noise reduced. The core keyword "charge now" is extracted by the NLP unit, which clarifies the user's urgent intention to "charge immediately, regardless of battery level".

[0055] Arbitration of conflicts between touch operation and APP settings: The system recognizes that the "80%" setting on the touch screen is a specific battery target. At this time, there is a potential conflict between the voice's intention to "ignore battery" and the "specific target" of the touch operation.

[0056] A compromise solution is generated: The multimodal instruction fusion processor combines user intent (immediate charging, desired 80% charge) and environmental constraints (unstable power grid) to generate a compromise user instruction: execute the immediate charging operation, but the maximum charging limit must be subject to power grid security constraints.

[0057] Step 2: Parse the charging demand based on the user's instructions and obtain real-time parameters according to the charging demand, specifically including: Analysis of charging needs: The system identified the core needs as "immediate startup and rapid battery replenishment"; Real-time parameter acquisition: Vehicle status information: After the charging gun is connected, the system obtains the user's vehicle data through the BMS: the current SOC (state of charge) is 30%, the battery temperature is 35℃, and the maximum allowable charging power is 60kW; Grid load: The grid monitoring unit reports that the current regional grid load has reached 92%, which is under heavy load; Photovoltaic / energy storage status: Due to the heavy rain in the evening, photovoltaic power generation is zero; the energy storage system has 40% remaining power.

[0058] Step 3: Input the real-time parameters into the constructed LSTM load forecasting model and output the forecast results, specifically including: Model prediction: The system inputs the current real-time parameters (high grid load, no photovoltaic, vehicle charging demand, etc.) into the pre-built LSTM load prediction model.

[0059] Output prediction results: The model predicts that the power grid load in the area will continue to exceed the safety threshold (95%) within the next 30 minutes. If a large number of charging piles are operating at full capacity, it is very likely to cause a local power outage. Therefore, the prediction result is: the power grid is in an extremely vulnerable state and load limiting is necessary.

[0060] Step 4: Control the charging pile based on the prediction results and obtain control commands, specifically including: activating the power-limited charging mode.

[0061] Step 5 involves dynamically calculating the prediction results and control commands using a multi-objective optimization function to obtain the optimal charging output power. This includes: Multi-objective dynamic optimization calculation: The system inputs all the above information into the multi-objective optimization function for dynamic calculation.

[0062] Power Grid Costs and Security (Extremely high weight): To avoid the enormous costs and social impact of a power grid collapse, at this time... (Grid cost / safety weight) is dynamically adjusted to the highest value by the system; user waiting time While users desire fast charging, this objective carries less weight in the face of grid security; carbon emissions In the current emergency scenario, this objective has the lowest weight.

[0063] Calculation and solution: After the optimizer solves the problem, it finds that the optimal charging output power that can be allocated to the charging pile is 30 kW (only half of the vehicle's maximum allowable power) under the premise of ensuring that the power grid does not collapse. Based on this, it is calculated that charging to 65% capacity is a balance point that can meet the user's basic needs and comply with the power grid safety threshold under the current vehicle status.

[0064] Step 6: Perform charging based on the optimal charging output power.

[0065] Charging and Confirmation: The system does not execute immediately. Instead, it pushes a confirmation message to the user's phone via the app: Due to grid limitations, to ensure stable power supply, it is recommended to charge to 65% immediately (approximately 40 minutes). Do you wish to continue? The suggested charge level combines the user's touch settings (80%), emergency intent (charge now), and system safety boundary (65%).

[0066] User confirmation and charging start: The user clicks "Confirm," and the system immediately begins charging the vehicle at an optimized power of 30 kW.

[0067] Safety monitoring and feedback: Throughout the charging process, the system monitors the charging cable temperature, battery temperature, and insulation resistance in real time to ensure absolute safety in heavy rain. It then generates a charging report, which is displayed to the user in real time via the app, showing information such as current power, estimated full charge time, amount charged, and cost.

[0068] Charging complete: When the vehicle's SOC reaches 65%, the charging station automatically stops charging, and the system pushes a "Charging complete" notification and electronic invoice via the app. The entire charging process is safe and efficient, meeting users' emergency needs in extreme weather conditions while perfectly fulfilling its responsibilities as a "smart node" of the power grid for peak shaving, valley filling, and safety assurance.

[0069] This invention provides a control system for an intelligent charging pile. Figure 3This is a schematic diagram of a control system framework for a smart charging pile according to some embodiments of this specification, including: The multimodal interaction module is used to receive and integrate user commands based on multimodal interaction. The environmental perception module is used to parse charging needs based on user commands and obtain real-time parameters according to the charging needs; The prediction result output module is used to input real-time parameters into the constructed LSTM load prediction model and output the prediction results; The charging pile control module is used to control the charging pile based on the prediction results and obtain control commands. The dynamic energy optimization module is used to dynamically calculate the optimal charging output power by combining the prediction results and control commands through a multi-objective optimization function. The charging module is used to perform charging based on the optimal charging output power.

[0070] In some embodiments, multimodal interaction is performed through a multimodal interaction module, which includes configurations such as a speech recognition / NLP processing unit, a touch screen interactive interface, a mobile APP remote control, and a biometric (fingerprint / face) authentication module.

[0071] In some embodiments, the dynamic energy optimization module includes the design of a charging pile core controller, i.e., a multimodal instruction fusion processor, through a user habit learning algorithm and a dynamic pricing strategy generator.

[0072] In some embodiments, real-time feedback is provided through a grid load monitoring unit, a distributed energy coordinator (photovoltaic / energy storage), and a charging load prediction model.

[0073] In some embodiments, a safety protection module is also included, comprising a real-time temperature monitoring unit for charging cables, an automatic power-off unit for abnormal high temperatures, a data encryption communication module, a leakage protection unit, an IP65 waterproof structure, a smoke sensor and an automatic fire extinguishing device, a charging gun insertion / removal status detection unit, and a fault self-diagnosis system.

[0074] In some embodiments, this system will acquire real-time parameters and upload them to the cloud management platform.

[0075] In some embodiments, the system also includes a modular expansion design, with hardware interfaces supporting 5G communication and edge computing module upgrades; the detachable structure facilitates future maintenance.

[0076] In some embodiments, the present invention controls the output power of the charging system in real time based on the power distribution side of the charging system, which can improve charging efficiency by more than 20% (through dynamic power adjustment), reduce operating costs by 30% (photovoltaic + energy storage peak shaving and valley filling), reduce user operation steps by 50% (automated identity verification and payment), and reduce the safety accident rate by 90% (multiple active protection mechanisms).

[0077] See Figure 4 In one embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to realize a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used for the operation of a control method for a smart charging pile.

[0078] In one embodiment of the present invention, a computer-readable storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system; and the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the control method for the smart charging pile in the embodiment.

[0079] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A control method of an intelligent charging pile, characterized in that, The method comprises the following steps: receiving and fusing user instructions based on multi-modal interaction; analyzing charging demand based on user instructions, and obtaining real-time parameters according to the charging demand; inputting the real-time parameters into a constructed LSTM load prediction model to output a prediction result; controlling the charging pile based on the prediction result to obtain a control instruction; dynamically calculating the prediction result and the control instruction through a multi-objective optimization function to obtain an optimal charging output power; performing charging based on the optimal charging output power. 2.The control method of claim 1, wherein, The receiving and fusing user instructions based on multi-modal interaction specifically comprises: enhancing language instructions from a voice recognition unit and then extracting key intentions; conflict arbitration of touch instructions and setting instructions from a touch display screen and a mobile terminal APP respectively to obtain operation instructions; fusing the key intentions and the operation instructions to obtain user instructions.

3. The control method of the intelligent charging pile according to claim 1, characterized in that, The real-time parameters include power grid load, photovoltaic energy storage and vehicle state information.

4. The control method of claim 1, wherein, The construction of the LSTM load prediction model comprises: obtaining historical parameters, including historical power grid load, historical photovoltaic energy storage and historical vehicle state information; cleaning, denoising and normalizing the historical parameters in sequence to obtain a supervised learning data set; constructing an LSTM load prediction model, including an input layer, one or more LSTM hidden layers, a Dropout layer and an output layer connected in sequence; dividing the supervised learning data set into a training set and a validation set, training the LSTM load prediction model using the training set, and optimizing the hyperparameters of the LSTM load prediction model through the validation set to minimize the loss function between the predicted value and the true value, thereby obtaining the constructed LSTM load prediction model for real-time prediction of charging load in a specific period in the future.

5. The control method of the intelligent charging pile according to claim 1, characterized in that, The controlling the charging pile based on the prediction result to obtain a control instruction specifically comprises: switching to green energy when the prediction result is that photovoltaic power supply is sufficient; starting fast charging when the prediction result is that the power grid is low load; starting a power-limited charging mode when the prediction result is that the power grid is high load; suggesting peak-shifting charging when the prediction result is a peak electricity period.

6. The control method of the intelligent charging pile according to claim 1, characterized in that, The formula of the multi-objective optimization function comprises: wherein represents a multi-objective optimization function; represents a grid cost; represents a grid cost weight; represents a user wait time; represents a user wait time weight; represents carbon emissions; represents a carbon emissions weight.

7. The control method of the intelligent charging pile according to claim 1, characterized in that, During the charging process, the temperature of the charging cable is monitored in real time, a charging report is generated according to the temperature of the charging cable, and the charging report is fed back through the user terminal.

8. A control system of an intelligent charging pile, characterized in that, The control method of the intelligent charging pile according to any one of claims 1-7, comprising: a multi-modal interaction module for receiving and fusing user instructions based on multi-modal interaction; an environment perception module for analyzing charging demand based on user instructions, and obtaining real-time parameters according to the charging demand; a prediction result output module for inputting the real-time parameters into a constructed LSTM load prediction model to output a prediction result; a charging pile control module for controlling the charging pile based on the prediction result to obtain a control instruction; a dynamic energy optimization module for dynamically calculating the prediction result and the control instruction through a multi-objective optimization function to obtain an optimal charging output power; an execution charging module for performing charging based on the optimal charging output power.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The computer program is executed by the processor to implement the steps of the control method of the intelligent charging pile according to any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the steps of the control method of the intelligent charging pile according to any one of claims 1-7.

Images