High-efficiency voltage compatible control method and charging pile module

Abstract

The invention provides a high-efficiency voltage compatible control method and a charging pile module, and relates to the technical field of voltage compatibility, and the method comprises the steps: collecting user habit data and historical data, combining the real-time monitored multi-module state data, and extracting communication protocol data and user demand data from the user habit data. And optimizing the ARIMA model by using the whale algorithm, analyzing the historical distribution instruction, and outputting a control instruction. And extracting distribution influence data through feature engineering, and adjusting a control instruction in combination with communication protocol data and user demand data to generate a CAN instruction. And constructing a main program, a sampling interruption program and a communication interruption program, controlling the operation of the charging pile module according to a CAN instruction, and realizing system initialization, background circulation and fault detection. According to the invention, the control instruction is predicted through the WOA-ARIMA model, the prediction capability and the resource distribution efficiency of the charging pile module are enhanced, the charging demand is predicted more accurately, the mining high-voltage input is satisfied, and a high-frequency and high-efficiency charging modular product can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of voltage compatibility, and particularly to a high-efficiency voltage-compatible control method and a charging pile module. Background Art

[0002] The high-frequency scheme of the traditional 220V / 380V power input charging module is a very mature design scheme and market application. Most design schemes adopt the front-stage BUCK buck conversion and the rear-stage LLC full-bridge conversion scheme. However, there are few high-frequency conversion schemes for the 660V / 1140V power input in mines. In mines, the power frequency (such as thyristor rectification, transformer buck conversion, etc.) conversion is mainly used, which has low efficiency and large volume. A DC-to-DC converter, also known as a DC / DC converter, is an electric energy conversion circuit that can convert a DC power supply into a DC power supply with a different voltage. DC / DC converters can be divided into two types: hard switching and soft switching.

[0003] The switching device of the hard-switching DC / DC converter turns on or off the circuit while bearing voltage or passing current. Therefore, a large switching loss will be generated during the turn-on or turn-off process. For the switching tube of the soft-switching DC / DC converter, during the turn-on or turn-off process, either the voltage applied to it is zero, that is, zero-voltage switching (ZVS), or the current passing through the switching tube is zero, that is, zero-current switching (ZCS). This soft-switching method can significantly reduce the switching loss, enabling the switching frequency to be greatly increased, creating conditions for the miniaturization and modularization of the converter.

[0004] The traditional charging module rectifies and converts the input three-phase 380V circuit into DC at 550V. When designing the power topology, whether it is a traditional IGBT power tube or a silicon carbide power tube, there are very mature devices to choose from. However, once the input voltage level reaches 1140V and the rectification is converted to the DC bus up to 1600V, it is very difficult for the withstand voltage value of the conventional design topology power device to reach. The highest withstand voltage of silicon carbide is 3300V, which not only has a small batch but also is expensive, resulting in too high design costs and being not conducive to mass production.

[0005] The traditional power module is difficult to meet the compatibility with the traditional 380V system and the 660V / 1140V in mines, and it is required to be able to fully power output for different grid inputs. This is a complex design process with difficulties in design topology and cost. At the same time, the traditional control method is difficult to fully allocate the resources of the charging pile module, and the adaptability and accuracy are not high enough. Summary of the Invention

[0006] The present invention provides a high-efficiency voltage-compatible control method and a charging pile module, aiming to solve the defects in the prior art that it is difficult to be compatible with the traditional 380V system and the mining 660V / 1140V system, and the control method is difficult to fully allocate the resources of the charging pile module, and the adaptability and accuracy are not high enough.

[0007] On the one hand, the present invention provides a high-efficiency voltage-compatible control method, including: S1: Collect user habit data and historical data, and perform real-time monitoring on the charging pile module to obtain multi-module status data, and extract communication protocol data and user demand data from the user habit data.

[0008] S2: Analyze the historical data to obtain historical allocation instruction data, use the whale algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model, input the historical allocation instruction data into the WOA-ARIMA model, and output to obtain a control instruction.

[0009] S3: Use the feature engineering method to extract allocation influence data from the multi-module status data, and adjust the control instruction according to the allocation influence data in combination with the communication protocol data and the user demand data to obtain a CAN instruction.

[0010] S4: Construct a main program, a sampling interrupt program, and a communication interrupt program, control the main program according to the CAN instruction, which is divided into two parts: system startup initialization and background loop, and use a fault detection and diagnosis mechanism to detect whether a fault occurs in the main program. If so, enter the fault mode for operation, otherwise continue to execute the background loop.

[0011] According to the high-efficiency voltage-compatible control method provided by the present invention, the steps of obtaining the communication protocol data and the user demand data include: S11: Split the user habit data into user charging records, user device interaction logs, and user setting preferences.

[0012] S12: Extract the message header, frame format, and data field length from the user device interaction log as protocol features, use a machine learning classification model to identify the protocol type to obtain the target type, and calculate the communication times of different protocols per unit time to obtain the interaction frequency. The communication protocol data is composed of the protocol features, the target type, and the interaction frequency.

[0013] S13: Extract explicit demand data from the user setting preferences, identify the typical mode according to the user charging time, divide the user types according to the historical charging power distribution, and analyze the user demand combination according to the user behavior mode. The user demand data is composed of the explicit demand data, the typical mode, the user type, and the user demand combination.

[0014] According to a high-efficiency voltage-compatible control method provided by the present invention, the steps of obtaining historical allocation instruction data include: Collect historical operation data of the charging pile module, remove outliers and missing values from the historical operation data, and convert the historical operation data into a unified format.

[0015] By analyzing the charging instructions in the historical data, identify the charging request instructions and stop charging request instructions initiated by the user, extract the charging process data related to the charging request instructions and stop charging request instructions, and generate historical allocation instructions based on the charging instructions, charging process data, and historical operation data.

[0016] According to a high-efficiency voltage-compatible control method provided by the present invention, the steps of obtaining the WOA-ARIMA model include: S21: Randomly generate a whale population. Each whale individual in the whale population represents a set of possible ARIMA parameters, and optimize the ARIMA parameters by simulating the foraging behavior of whales.

[0017] S22: Use each whale individual to construct an ARIMA model, fit the historical data to the ARIMA model for prediction, calculate the prediction error, and calculate the fitness value using the mean square error method. The formula is expressed as:

[0018] In the formula, is the i-th whale individual is the fitness value of is the mean square error of the ARIMA model corresponding to the i-th whale individual, m is the number of prediction data, yj is the actual observed value, is the predicted value of the ARIMA model, is the position of the i-th whale individual.

[0019] S23: Select the individual with the smallest fitness value among all whale individuals as the current optimal individual, use the whale algorithm to iteratively optimize the current optimal individual. After reaching the preset number of iterations, output the optimal individual, and use the ARIMA parameters corresponding to the optimal individual to construct the WOA-ARIMA model.

[0020] According to a high-efficiency voltage-compatible control method provided by the present invention, the steps of obtaining the optimal individual include: S231: Calculate the control parameter a of the whale population, and calculate the intermediate variable for each whale individual. The formula is expressed as:

[0021]

[0022] Wherein, and are random numbers subject to the uniform distribution U(0,1), and Q and W are intermediate variables.

[0023] S232: Determine whether the absolute value of the intermediate variable Q is greater than 1. If so, the whale individual enters the random exploration stage, moves towards a randomly selected whale individual, and updates the position of the current whale individual to obtain the random whale position. Otherwise, select the method of surrounding the prey or bubble net attack to update the position of the current whale individual to obtain the updated whale position.

[0024] S233: For each updated random whale position and updated whale position, recalculate the fitness value and update the position of the current optimal individual.

[0025] S234: Repeat steps S231 to S233 until the maximum number of iterations is reached, and select the whale individual with the minimum fitness value as the optimal individual.

[0026] According to a high-efficiency voltage-compatible control method provided by the present invention, the formula for obtaining the random whale position is expressed as:

[0027]

[0028] Wherein, is the random whale position, is the position of a randomly selected whale individual, t is the number of iterations, and D is the distance vector.

[0029] According to a high-efficiency voltage-compatible control method provided by the present invention, the steps for obtaining the updated whale position include: Update the position of the current whale individual using the method of surrounding the prey to obtain the updated whale position, and the formula is expressed as:

[0030]

[0031] Wherein, is the updated whale position, is the position of the current optimal individual, is the distance vector two.

[0032] Update the position of the current whale individual using the method of bubble net attack to obtain the updated whale position, and the formula is expressed as:

[0033] Wherein, a is the updated position of the whale, and b is the spiral shape constant. is the distance between and is a random number that follows a uniform distribution U(-1, 1), and e is the natural constant.

[0034] According to a high-efficiency voltage-compatible control method provided by the present invention, the system startup initialization includes: initializing system parameter configuration, Flash allocation, I / O initialization configuration, SPI and CAN communication module configuration, global variable initialization, interrupt vector initialization, PWM module initialization configuration, ADC module configuration, global interrupt vector startup, and peripheral driver software configuration.

[0035] According to a high-efficiency voltage-compatible control method provided by the present invention, the period of executing the background loop is: taking 10 ms as a counting period, taking a 100-us timing interrupt event as a timing period, and the 10-ms counting period is divided into 100 task periods.

[0036] On the other hand, the present invention also provides a high-efficiency voltage-compatible charging pile module, including: Data management system: collecting user habit data and historical data, and performing real-time monitoring on the charging pile module to obtain multi-module status data, and extracting communication protocol data and user demand data from the user habit data.

[0037] Prediction and optimization system: analyzing historical data to obtain historical allocation instruction data, using the whale algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model, inputting the historical allocation instruction data into the WOA-ARIMA model, and outputting to obtain control instructions.

[0038] Control system: using feature engineering methods to extract allocation influence data from multi-module status data, and adjusting control instructions according to the allocation influence data in combination with communication protocol data and user demand data to obtain CAN instructions.

[0039] Execution and monitoring system: constructing a main program, a sampling interrupt program, and a communication interrupt program, controlling the main program according to CAN instructions, which is divided into two parts: system startup initialization and background loop, and using a fault detection and diagnosis mechanism to detect whether a fault occurs in the main program. If so, enter the fault mode for operation, otherwise continue to execute the background loop.

[0040] An efficient voltage-compatible control method and charging pile module provided by the present invention use the whale optimization algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model for predicting control commands, solving the limitations of the traditional ARIMA model in parameter selection. By optimizing parameters with WOA, the prediction accuracy of the model is improved, and at the same time, the prediction ability and resource allocation efficiency of the charging pile module are enhanced. By optimizing the model, the charging demand can be predicted more accurately, reducing resource waste. According to the allocation impact data, combined with communication protocol data and user demand data, the control command is adjusted to generate CAN commands, solving the singularity and inflexibility of the charging pile module in the control command generation stage, improving the adaptability and accuracy of the control command, and better meeting user needs. The function of power conversion is realized through software control, which is more intelligent and convenient. According to different customer needs, it meets the customization requirements, bringing a broader market space for applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0042] Figure 1 is a schematic flowchart of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 2 is a schematic main program flowchart of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 3 is a schematic diagram of the system startup initialization structure of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 4 is a schematic diagram of the background loop structure of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 5 is a schematic flowchart of the sampling interrupt program of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 6 is a schematic flowchart of the communication interrupt program of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 7 is a schematic diagram of the fault detection and diagnosis mechanism structure of an efficient voltage-compatible control method provided by an embodiment of the present invention; Figure 8It is a schematic structural diagram of a control circuit of a high-efficiency voltage-compatible charging pile module provided by an embodiment of the present invention; Figure 9 is Figure 4 a schematic diagram of the counting part process in Specific embodiments

[0043] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0044] Next, in combination with Figures 1 - 8 a high-efficiency voltage-compatible control method and a charging pile module of the present invention will be described.

[0045] As Figure 1 shown, a high-efficiency voltage-compatible control method provided by an embodiment of the present invention includes: S1: Collect user habit data and historical data, and perform real-time monitoring on the charging pile module to obtain multi-module status data, and extract communication protocol data and user demand data from the user habit data.

[0046] In step S1, the steps of obtaining communication protocol data and user demand data include: S11: Split the user habit data into user charging records, user device interaction logs, and user setting preferences. Let the user charging record be A (A 1 , A 2 ,..., A Z ), where A 1 is the charging time, A 2 is the charging duration, A Z is the cost. Let the user device interaction log be B (B 1 , B 2 ,..., B x ), where B 1 is the APP operation, B 2 is the RFID card swipe, B x is the Bluetooth connection. Let the user setting preference be C (C 1 , C 2 ,..., C c ), where C 1 is the upper limit of the charging power, C 2 is the target SOC, C c is the reserved charging period.

[0047] S12: Extract the packet header, frame format, and data field length from the user device interaction log as protocol features, use a machine learning classification model to identify the protocol type to obtain the target type, and calculate the communication times of different protocols per unit time to obtain the interaction frequency. The communication protocol data consists of protocol features, target type, and interaction frequency.

[0048] S13: Extract explicit demand data from user settings preferences, identify typical patterns based on the user's charging time, divide user types according to the historical charging power distribution, analyze the user's behavior pattern to obtain the user demand combination. The user demand data consists of explicit demand data, typical patterns, user types, and user demand combinations. Typical patterns include modes such as weekday commuting charging and weekend long-distance charging, and user types include fast-charging preference types and economic slow-charging types. For example, the judgment combination of the user demand combination is high-power charging plus night charging to judge the user as a taxi driver.

[0049] S2: Analyze the historical data to obtain historical allocation instruction data, use the whale optimization algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model, input the historical allocation instruction data into the WOA-ARIMA model, and output to obtain the control instruction.

[0050] In step S2, the steps to obtain historical allocation instruction data include: Collect the historical operation data of the charging pile module, remove the outliers and missing values in the historical operation data, and convert the historical operation data into a unified format.

[0051] By analyzing the charging instructions in the historical data, identify the charging request instructions and stop charging request instructions initiated by the user, and extract the charging process data related to the charging request instructions and stop charging request instructions. Generate historical allocation instructions according to the charging instructions, charging process data, and historical operation data. The historical allocation instructions include the allocation time of the charging pile, the allocated charging pile ID, and the allocated charging power, etc.

[0052] The steps to obtain the WOA-ARIMA model include: S21: Randomly generate a whale population. Each whale individual in the whale population represents a set of possible ARIMA parameters, and optimize the ARIMA parameters by simulating the foraging behavior of whales.

[0053] S22: Use each whale individual to construct an ARIMA model, fit the historical data to the ARIMA model for prediction, calculate the prediction error, and calculate the fitness value using the mean square error method. The formula is expressed as:

[0054] is the i-th whale individual is the fitness value of is the mean square error of the ARIMA model corresponding to the i-th whale individual. m is the number of prediction data, yj is the actual observed value, is the predicted value of the ARIMA model, is the position of the i-th whale individual.

[0055] S23: Select the individual with the minimum fitness value among all whale individuals as the current optimal individual, and use the whale algorithm to iteratively optimize the current optimal individual. After reaching the preset number of iterations, output the obtained optimal individual, and use the ARIMA parameters corresponding to the optimal individual to construct the WOA-ARIMA model.

[0056] In step S23, the steps to obtain the optimal individual include: S231: Calculate the control parameter a of the whale population, and calculate the intermediate variable for each whale individual. The formula is expressed as:

[0057]

[0058] In the formula, and are random numbers obeying the uniform distribution U(0,1), and Q and W are intermediate variables.

[0059] S232: Judge whether the absolute value of the intermediate variable Q is greater than 1. If so, the whale individual enters the random exploration stage and moves towards a randomly selected whale individual to update the position of the current whale individual to obtain the random whale position. Otherwise, select the method of surrounding the prey or bubble net attack to update the position of the current whale individual to obtain the updated whale position.

[0060] In step S232, the formula for obtaining the random whale position is expressed as:

[0061]

[0062] In the formula, is the random whale position, is the position of a randomly selected whale individual, t is the number of iterations, and D is the distance vector.

[0063] The steps to obtain the updated whale position include: Use the method of surrounding the prey to update the position of the current whale individual to obtain the updated whale position. The formula is expressed as:

[0064]

[0065] In the formula, is the updated position of the whale, is the position of the current optimal individual, is the distance vector two.

[0066] The position of the current whale individual is updated using the bubble net attack method to obtain the updated position of the whale. The formula is expressed as:

[0067] In the formula, is the updated position of the whale, b is the spiral shape constant, is and The distance between them is a random number obeying the uniform distribution U(-1, 1), and e is the natural constant.

[0068] S233: For each updated random position of the whale and the updated position of the whale, recalculate the fitness value and update the position of the current optimal individual.

[0069] S234: Repeat steps S231 to S234 until the maximum number of iterations is reached, and select the whale individual with the minimum fitness value as the optimal individual.

[0070] S3: Use the feature engineering method to extract the allocation influence data from the multi-module status data, and adjust the control instruction according to the allocation influence data in combination with the communication protocol data and the user demand data to obtain the CAN instruction.

[0071] The methods for obtaining the allocation influence data include: removing the outliers and noise data in the multi-module status data, and processing the missing values by filling, interpolation or deletion methods to obtain the complete multi-module data. Extract the time-domain features, frequency-domain features, state-related features, combined features and statistical features from the complete multi-module data. Use the PCA method to reduce the feature dimensions of the time-domain features, frequency-domain features, state-related features, combined features and statistical features, and remove the redundant features to obtain multiple feature factor data. Use the decision tree algorithm to screen the features that have a significant impact on the allocation task from the multiple feature factor data as the allocation influence data.

[0072] S4: Construct the main program, sampling interrupt program and communication interrupt program, control the main program according to the CAN instruction, which is divided into two parts: system startup initialization and background loop, and use the fault detection and diagnosis mechanism to detect whether the main program has a fault. If so, enter the fault mode operation, otherwise continue to execute the background loop.

[0073] Such as Figure 2As shown, the process of the main program includes: after the charging pile module is powered on, the system starts to execute the initialization program, such as Figure 3 As shown, the control process of the main control chip starts as follows: First, the main chip is powered on. After the system completes the initialization of basic peripherals and drivers, it enters the Bootloader. In the Bootloader, it detects whether the control application layer software is normal (judged by the correctness of the check code), and based on this, it determines whether it can directly jump to the application layer software. When the check code is detected to be incorrect, the software will automatically enter the waiting state for flashing. When the check code is detected to be accurate, the software will automatically jump to the application layer software for operation.

[0074] In step S4, the system startup initialization includes: initialization of system parameter configuration, Flash allocation, I / O initialization configuration, SPI and CAN communication module configuration, global variable initialization, interrupt vector initialization, PWM module initialization configuration, ADC module configuration, global interrupt vector startup, and peripheral driver software configuration. The control system starts to receive CAN instructions for output voltage and current setting and switch enabling operations. At this time, the system enters the background loop of the application layer software to implement application layer task scheduling. After the system is powered on, it will perform a delay detection to ensure that all hardware devices have enough time to complete initialization. During the delay detection, the system will perform self-checks on each key component of the charging pile module.

[0075] Such as Figure 4 As shown, the period of executing the background loop is: different tasks in the application layer software are time-shared scheduled in this background loop. With a 10ms as a counting period and a 100us timing interrupt event as the timing period, the 10ms counting period is divided into 100 task cycles.

[0076] The system initializes the CAN communication protocol to ensure that it can send current and temperature data of each channel to the external monitoring system in real time. The system receives and parses CAN instructions from the outside, such as setting voltage / current, start / stop commands, etc., and updates the control parameters. At the same time, the system also receives hardwired commands, which are usually used to directly control the switch state of the charging pile module. According to the parsed instructions, the system will adjust the operating state of the charging pile module, such as setting the output voltage and current, controlling the state of each logic switch, etc.

[0077] Such as Figure 5As shown, the process of the sampling interrupt program is as follows: When the charging pile module receives the power-on instruction, the system starts to execute the initialization program. The system initializes the AD sampling channels to prepare for sampling the key electrical parameters of the charging pile module. The system starts to sample 8 groups of data from the AD channels and filters the collected data. Filtering can remove the noise in the data and improve the accuracy and reliability of the data. The system immediately determines whether the overcurrent and overtemperature protection of the corresponding channels of these 8 groups of data are effective. After completing the sampling and processing of the current 8 groups of data, the system switches to the next group of sampling channels and continues to collect new data. The system enters a loop detection mode and continuously repeats the processes of sampling, filtering, protection judgment, and channel switching. During the loop detection process, the system receives CAN commands from the outside in real time. According to the received commands, the system controls the states of each logic switch to achieve precise control of the charging pile module. When the shutdown instruction is received, the system executes the shutdown process. First, the system stops all charging operations to ensure that the charging pile module is in a safe state. Then, the system saves the current operating state and parameter settings so that it can quickly resume the previous state when starting up next time. Finally, the system executes a series of shutdown operations, including turning off the peripheral power supply, releasing resources, etc., to ensure the safe shutdown of the charging pile module.

[0078] As Figure 6 shown, the process of the communication interrupt program is as follows: When the charging pile module receives the power-on instruction, the system starts to execute the initialization program. The system initializes the CAN communication module to ensure real-time data sending and receiving. The CAN communication module is a key component for the charging pile module to communicate with external systems (such as monitoring centers, user devices). The system sends the current and temperature data of each channel in real time through the CAN communication module. The system enters the control logic state and controls the operation of the charging pile module according to the preset logic and received instructions. The control logic includes: Charging control: Control the start, stop, and adjustment of the charging process according to user requirements and the state of the charging pile module. Fault detection and protection: Real-time detect the operating state of the charging pile module. Once an abnormal situation is found, immediately take protection measures to ensure the safety of the equipment and users. Status feedback: Feedback the operating state of the charging pile module (such as charging status, fault status) to the monitoring system and user devices. The system receives CAN commands from the outside in real time. These commands include but are not limited to the following three types: Charging instruction: Start or stop charging. Parameter adjustment instruction: Adjust parameters such as charging voltage and current. Status query instruction: Query the current operating state of the charging pile module. The system parses the received commands and adjusts the operating state of the charging pile module according to the command content. For example, if the command to start charging is received, the system will start the charging process. If the command to stop charging is received, the system will stop charging. When the shutdown instruction is received, the system executes the shutdown process.

[0079] AsFigure 7 As shown in the figure, the fault detection and diagnosis mechanism adopts the LBC detection and processing mechanism: when a certain fault occurs, the recoverable fault count variable is incremented, and at the same time, the relevant fault is cleared. The sampled values of all recoverable fault LBC periodic count variables are decremented by 1. When the value of the LBC periodic count variable is not greater than 0, it is necessary to determine whether the value of the recoverable fault count variable is greater than the LBC extreme value of the corresponding fault. When the value of the recoverable fault count variable is greater than the LBC extreme value, this fault enters the non-recoverable fault mode. When it is less than or equal to the fault LBC extreme value, it is first necessary to determine whether the recoverable fault counter is greater than 1. When it is greater than 1, the relevant fault needs to enter the recoverable fault mode, otherwise the system operates normally.

[0080] As Figure 8 As shown in the figure, a high-efficiency voltage-compatible control method of the present invention is used to control the entire charging pile module. While considering adding a surge protection circuit to the input, two-stage three-phase EMI filter circuits are added. After being input to the PFC power correction circuit through a six-pulse rectifier circuit, the voltage is stabilized at about DC800V through a BUCK-BOOST circuit (step-up and step-down). The circuit then passes through an LLC full-bridge DCDC conversion circuit to achieve the output required charging voltage. The output voltage is then filtered through LC and EMI filter circuits to output a stable charging voltage to charge the battery pack. The power supply of the auxiliary power supply is taken from the high-voltage input, and a flyback voltage conversion circuit is used to achieve voltage power supply extraction and voltage conversion, and at the same time achieve multiple outputs to meet the requirements of multiple different power supply needs. The high-voltage voltage part all uses a linear optocoupler plus an isolated operational amplifier to achieve isolated sampling. The current sampling part on the high-voltage side all uses a Hall current sampling circuit to achieve complete electrical isolation. The temperature sampling is all carried out on the low-voltage side using an NTC temperature sensor.

[0081] The present invention uses the whale optimization algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model to predict control commands, which solves the limitations of the traditional ARIMA model in parameter selection. By optimizing the parameters with WOA, the prediction accuracy of the model is improved, and at the same time, the prediction ability and resource allocation efficiency of the charging pile module are improved. By optimizing the model, the charging demand can be predicted more accurately, and resource waste can be reduced. According to the allocation impact data, combined with the communication protocol data and user demand data, the control commands are adjusted to generate CAN commands, which solves the singularity and inflexibility of the charging pile module in the control command generation stage, improves the adaptability and accuracy of the control commands, and better meets the user needs. The function of power conversion is realized through software control, which is more intelligent and convenient. According to the different needs of customers, the customization requirements are met, bringing a broader market space to the application.

[0082] Please refer to Figure 9, based on the same general inventive concept, the present invention also protects a high-efficiency voltage-compatible charging pile module, which can adopt the above-mentioned high-efficiency voltage-compatible control method. The charging pile module includes: Data management system: Collect user habit data and historical data, and perform real-time monitoring on the charging pile module to obtain multi-module status data, and extract communication protocol data and user demand data from the user habit data.

[0083] Prediction and optimization system: Analyze the historical data to obtain historical allocation instruction data, use the whale algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model, input the historical allocation instruction data into the WOA-ARIMA model, and output to obtain control instructions.

[0084] Control system: Use feature engineering methods to extract allocation influence data from the multi-module status data, and adjust the control instructions according to the allocation influence data in combination with the communication protocol data and user demand data to obtain CAN instructions.

[0085] Execution and monitoring system: Construct a main program, a sampling interrupt program, and a communication interrupt program, control the main program according to the CAN instructions, which are divided into two parts: system startup initialization and background loop, and use a fault detection and diagnosis mechanism to detect whether a fault occurs in the main program. If so, enter the fault mode for operation, otherwise continue to execute the background loop.

[0086] The present invention collects user habit data and historical data through the data management system, and combines the WOA-ARIMA model in the prediction and optimization system to optimize the allocation strategy of the charging module, enabling it to better adapt to different vehicle models and voltage requirements. The ARIMA model (WOA-ARIMA) is optimized using the whale algorithm to dynamically adjust the output parameters of the charging module, improving the power conversion efficiency. It also uses the control system and the execution and monitoring system to extract allocation influence data in combination with feature engineering methods, and adjusts the control instructions according to these data to generate more accurate CAN instructions, realizing intelligent control. The execution and monitoring system uses a fault detection and diagnosis mechanism to monitor the operating status of the charging pile module in real time. Once a fault is detected, it immediately enters the fault mode for operation to ensure the stability and reliability of the system.

[0087] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A high-efficiency voltage-compatible control method for controlling a charging pile module, characterized in that: include: S1: Collecting user habit data and historical data, and performing real-time monitoring on the charging pile module to obtain multi-module status data, and extracting the user habit data to obtain communication protocol data and user demand data; S2: Analyze the historical data to obtain historical allocation instruction data, use the whale algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model, input the historical allocation instruction data into the WOA-ARIMA model, and output to obtain the control instruction; S3: extracting the multi-module status data using a feature engineering method to obtain allocation impact data, and adjusting the control instruction according to the allocation impact data and in combination with the communication protocol data and the user demand data to obtain a CAN instruction; S4: Construct the main program, sampling interrupt program and communication interrupt program, control the main program according to the CAN instruction, and divide it into two parts: system startup initialization and background cycle. Use the fault detection and diagnosis mechanism to detect whether the main program has a fault. If so, enter the fault mode, otherwise continue to execute the background cycle.

2. A high-efficiency voltage-compatible control method according to claim 1, characterized in that: In step S1, the step of obtaining the communication protocol data and the user demand data includes: S11: Splitting the user habit data into user charging records, user device interaction logs and user setting preferences; S12: extracting a message header, a frame format, and a data field length from the user device interaction log as protocol features, identifying a protocol type using a machine learning classification model to obtain a target type, and calculating the number of communications of different protocols per unit time to obtain an interaction frequency, wherein the protocol features, the target type, and the interaction frequency constitute the communication protocol data; S13: extracting explicit demand data from the user setting preferences, identifying typical patterns according to the user charging time, dividing user types according to the historical charging power distribution, and obtaining user demand combinations according to the user behavior pattern analysis. The user demand data is composed of the explicit demand data, typical patterns, user types and user demand combinations.

3. A high-efficiency voltage-compatible control method according to claim 2, characterized in that: In step S2, the step of obtaining the historical allocation instruction data includes: Collecting historical operation data of the charging pile module, removing abnormal values ​​and missing values ​​in the historical operation data, and converting the historical operation data into a unified format; By analyzing the charging instructions in the historical data, the charging request instructions and the stop charging request instructions initiated by the user are identified, and the charging process data related to the charging request instructions and the stop charging request instructions are extracted, and the historical allocation instructions are generated according to the charging instructions, the charging process data and the historical operation data.

4. The high-efficiency voltage-compatible control method according to claim 1, characterized in that: In step S2, the steps of obtaining the WOA-ARIMA model include: S21: randomly generate a whale population, each whale individual in the whale population represents a set of possible ARIMA parameters, and optimize the ARIMA parameters by simulating the foraging behavior of whales; S22: Use each of the whale individuals to construct an ARIMA model, and fit the historical data to the ARIMA model for prediction, calculate the prediction error, and use the mean square error method to calculate the prediction error to obtain the fitness value, which is expressed as follows: ； In the formula, is the i-th whale individual The fitness value of is the mean square error of the ARIMA model corresponding to the i whale individuals, m is the number of predicted data, yj is the actual observation value, is the predicted value of the ARIMA model, is the position of the i-th individual whale; S23: Select the individual with the smallest fitness value from all the whale individuals as the current optimal individual, use the whale algorithm to iteratively optimize the current optimal individual, and when the preset number of iterations is reached, output the optimal individual, and use the ARIMA parameters corresponding to the optimal individual to construct the WOA-ARIMA model.

5. A high-efficiency voltage-compatible control method according to claim 4, characterized in that: In step S23, the step of obtaining the optimal individual includes: S231: Calculate the control parameter a of the whale population, and calculate the intermediate variable for each individual whale, the formula is expressed as: ； ； In the formula, and is a random number that follows uniform distribution U(0,1), Q and W are intermediate variables; S232: Determine whether the absolute value of the intermediate variable Q is greater than 1. If so, the whale individual enters the random exploration stage, moves to the randomly selected whale individual, updates the current position of the whale individual to obtain the whale random position, otherwise, selects the method of surrounding the prey or attacking with a bubble net to update the current position of the whale individual to obtain the whale updated position; S233: recalculating the fitness value of each updated random position of the whale and the updated position of the whale, and updating the position of the current optimal individual; S234: Repeat steps S231 to S233 until the maximum number of iterations is reached, and select the whale individual with the smallest fitness value as the optimal individual.

6. A high-efficiency voltage-compatible control method according to claim 5, characterized in that: In step S232, the formula for obtaining the random position of the whale is expressed as: ； ； In the formula, is the random position of the whale, is the position of a randomly selected whale individual, t is the number of iterations, and D is the distance vector.

7. The high-efficiency voltage-compatible control method according to claim 5, characterized in that: In step S232, the step of obtaining the updated position of the whale includes: The method of surrounding the prey is used to update the current position of the individual whale to obtain the updated position of the whale, and the formula is expressed as: ； ； In the formula, It is the whale that updates the location. is the current optimal individual position, is the distance vector two; The bubble net attack method is used to update the current position of the individual whale to obtain the updated position of the whale, and the formula is expressed as: ； In the formula, is the whale update position, b is the spiral shape constant, yes and The distance between them is a random number that obeys the uniform distribution U(-1,1), and e is a natural constant.

8. The high-efficiency voltage-compatible control method according to claim 1, characterized in that: In step S4, the system startup initialization includes: initializing system parameter configuration, Flash allocation, I / O initialization configuration, SPI and CAN communication module configuration, global variable initialization, interrupt vector initialization, PWM module initialization configuration, ADC module configuration, global interrupt vector startup and peripheral driver software configuration.

9. The high-efficiency voltage-compatible control method according to claim 1, characterized in that: In step S4, the background cycle is executed in a cycle of: 10 ms as a counting cycle, a 100 us timing interrupt event as a timing cycle, and the 10 ms counting cycle is divided into 100 task cycles.

10. A high-efficiency voltage-compatible charging pile module, which adopts a high-efficiency voltage-compatible control method as claimed in any one of claims 1 to 9, characterized in that: The charging pile module includes: Data management system: collects user habit data and historical data, and monitors the charging pile module in real time to obtain multi-module status data, and extracts the user habit data to obtain communication protocol data and user demand data; Prediction and optimization system: Analyze the historical data to obtain historical allocation instruction data, use the whale algorithm to optimize the autoregressive integrated moving average model to obtain the WOA-ARIMA model, input the historical allocation instruction data into the WOA-ARIMA model, and output the control instruction; Control system: extract the multi-module status data using a feature engineering method to obtain allocation impact data, and adjust the control instruction according to the allocation impact data and in combination with the communication protocol data and user demand data to obtain a CAN instruction; Execution and monitoring system: construct the main program, sampling interrupt program and communication interrupt program, control the main program according to the CAN instruction, and divide it into two parts: system startup initialization and background cycle. Use the fault detection and diagnosis mechanism to detect whether the main program has a fault. If so, enter the fault mode, otherwise continue to execute the background cycle.

Images