Charging pile power self-adaptive adjusting system and method based on temperature feedback

Abstract

The invention belongs to the technical field of new energy automobile charging equipment, and relates to a charging pile power adaptive adjustment system and method based on temperature feedback, and the method comprises the steps: carrying out the data fusion of a plurality of collected temperature values through employing a Kalman filtering algorithm, and obtaining a target temperature fusion value representing the overall thermal state of a charging pile; the target temperature fusion value is compared with a preset temperature safety threshold value, and the output power of the charging pile is dynamically adjusted according to a comparison result; the future temperature trend is predicted through a charging pile internal temperature prediction model and an air cooling / liquid cooling cooperative temperature adjustment mechanism, and a system can take actions in advance before the temperature does not reach a dangerous threshold but is determined to exceed the standard, so that the passive situation of traditional thermal management is thoroughly changed, real preventive control is realized, and the safety of the system is improved. The output power can be actively reduced when it is detected that the temperature in the pile rises, and overheating damage is prevented; and when the temperature returns to normal, the original power is automatically recovered, so that the reliability and the service life of the charging pile are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of new energy vehicle charging equipment, and relates to a charging pile power self-adaptive adjustment system and method based on temperature feedback. BACKGROUND

[0002] At present, the global electric vehicle charging facility market is in a high-speed expansion stage. The mainstream charging pile manufacturers and research institutions generally focus on the technical competition on the extreme charging speed and extensive equipment compatibility. This trend directly leads to the rapid popularization of high-power density charging piles, and the output power of the charging piles is constantly breaking through new highs, rapidly moving from dozens of kilowatts to 300 kilowatts or even higher. However, this competition orientation of "power-only theory" and "compatibility-only theory" leads to the fact that the industry resources are excessively tilted to the frontiers such as power electronic topology optimization and wide voltage range output control, and a basic subject related to the long-term reliable operation of the equipment, i.e. the thermal management guarantee mechanism under the changing real working conditions, especially the long-time high-load operation, is largely ignored. Many existing designs only meet the basic temperature rise test in the standard laboratory environment, and the built-in heat dissipation scheme (such as a fan with fixed speed or a simple temperature threshold control) is often a passive and lagging coping strategy, lacking the intelligent ability of accurate perception, prediction and active regulation of the internal thermal state. This imbalance in design concept makes a large number of operating charging piles inherently carry the gene of overheating risk in complex application scenarios, which poses a hidden danger to the long-term stable operation of the charging piles.

[0003] Therefore, most of the charging piles in the current market are mainly designed to improve the charging speed and compatibility, and less consideration is given to how to ensure the stability and safety of the equipment under long-time high-load operation. Especially in high-temperature environments, the key components in the charging pile, such as power modules and main control boards, will generate a large amount of heat. If this heat cannot be effectively dissipated in time, it will lead to a decrease in device performance, shorten the service life, and increase the maintenance cost. In addition, although some existing temperature control measures such as installing a heat dissipation fan can alleviate the heating problem to a certain extent, they cannot fundamentally solve the overheating risk caused by high environmental temperature or poor heat dissipation. SUMMARY

[0004] In view of the problems in the prior art, the application aims to provide a charging pile power self-adaptive adjustment system and method based on temperature sensor feedback, which can actively reduce the output power when detecting the temperature rise in the pile to prevent overheating damage, and automatically restore the original power when the temperature returns to normal, thereby improving the reliability and service life of the charging pile.

[0005] The application is realized by the following technical solutions: A charging pile power self-adaptive adjustment method based on temperature feedback, comprising: Real-time collection of temperature values of multiple key heating areas inside the charging pile; Data fusion of the collected multiple temperature values is performed using a Kalman filtering algorithm to obtain a target temperature fusion value representing the overall thermal state of the charging pile; The target temperature fusion value is compared with a preset temperature safety threshold, and the output power of the charging pile is dynamically adjusted according to the comparison result; When the target temperature fusion value exceeds the preset temperature safety threshold, the output power is reduced; when the target temperature fusion value falls below the preset temperature safety threshold, the output power is restored; When the target temperature fusion value does not exceed the preset temperature safety threshold, but the charging pile internal temperature prediction model predicts that the internal temperature of the charging pile will exceed the preset temperature safety threshold in the future, the operation of reducing the output power is performed in advance; at the same time, the charging pile is physically cooled by combining the air cooling / liquid cooling cooperative temperature regulation mechanism, so that the charging pile operates within the safe temperature range.

[0006] Preferably, the collected multiple temperature values are data fused using a Kalman filtering algorithm, specifically: Initialization step: set the initial state estimate value of the temperature and the initial covariance matrix; Prediction step: based on the heat transfer model of the charging pile, the temperature state at the current time is predicted according to the temperature parameter at the last time, to obtain the predicted state value and the predicted covariance; Update step: obtain the current temperature measurement value, calculate the Kalman gain, and update the predicted state value and the predicted covariance using the Kalman gain to obtain the state estimate value of the multiple different key heating areas at the current time, take the average value as the target temperature fusion value.

[0007] Preferably, the key heating areas include the power module, the main control board and the DC input / output side of the charging pile.

[0008] Preferably, the output power of the charging pile is dynamically adjusted, specifically including: Set multiple temperature safety threshold intervals, each interval corresponding to a different power adjustment strategy; When the target temperature fusion value enters a higher temperature safety threshold interval, the output power is reduced according to the strategy corresponding to the interval, and the reduction amplitude is related to the degree of temperature exceeding the threshold; When the target temperature fusion value falls to a lower temperature safety threshold interval, the output power is gradually increased according to the strategy corresponding to the interval.

[0009] Preferably, the charging pile internal temperature prediction model is constructed as follows: Collecting historical temperature data, power output data and external environmental factor data of the charging pile; After preprocessing the collected data, key features related to the internal temperature of the charging pile are extracted; Through feature engineering method, the key features are converted into feature vectors reflecting the temperature change law of the charging pile; Machine learning algorithm is used to train the extracted feature vectors to build an internal temperature prediction model of the charging pile.

[0010] Preferably, based on the internal temperature prediction model of the charging pile, the temperature change trend of the charging pile in the future period of time is predicted through the machine learning model according to the historical operation data and external environmental data of the charging pile.

[0011] Preferably, the air cooling / liquid cooling collaborative temperature regulation mechanism is as follows: According to the target temperature fusion value, the air cooling system and / or the liquid cooling system inside the charging pile are controlled to adjust the cooling intensity, the air cooling system adjusts the speed of the cooling fan to cool the inside of the charging pile, and the liquid cooling system adjusts the flow of the cooling liquid to cool the inside of the charging pile.

[0012] A charging pile power self-adaptive adjustment system based on temperature feedback, comprising: temperature acquisition module, data fusion module, central control module, power adjustment module, prediction module and cooling control module; The temperature acquisition module includes a plurality of temperature sensors arranged in the key heating areas inside the charging pile, respectively, for monitoring the working temperature inside the charging pile and collecting the temperature values of each area in real time; The data fusion module is used to fuse the collected multiple temperature values through Kalman filtering algorithm to obtain a target temperature fusion value representing the overall thermal state of the charging pile; The central control module is connected with the data fusion module, configured with a preset temperature safety threshold, and compares the target temperature fusion value with the preset temperature safety threshold; The power adjustment module dynamically adjusts the output power of the charging pile according to the instructions from the central control module and the comparison results; The prediction module is connected with the central control module, and predicts the temperature change trend of the charging pile in the future period of time through the machine learning model according to the historical operation data and external environmental data of the charging pile; The cooling control module is connected with the central control module and is configured to control at least one of the air cooling system and the liquid cooling system to perform collaborative temperature regulation and physically cool the charging pile according to the instructions from the central control module.

[0013] Preferably, the air cooling system comprises a plurality of adjustable speed cooling fans installed inside the charging pile, and the speed of the fans is controlled by a central control module; The liquid cooling system comprises a circulating pump, a radiator, a cooling liquid pipeline and a temperature sensor; the cooling liquid flows through the key heat generating areas inside the charging pile under the driving of the circulating pump, absorbs heat and then dissipates to the air through the radiator; the central control module controls the speed of the circulating pump and the flow of the cooling liquid according to the temperature value collected by the temperature sensor, so as to effectively control the temperature inside the charging pile.

[0014] A charging pile comprising the temperature feedback based charging pile power adaptive adjustment system.

[0015] Compared with the prior art, the present application has the following beneficial technical effects: The present application aims to provide a temperature sensor feedback based charging pile power adaptive adjustment system and method, which can actively reduce the output power when detecting the temperature rise inside the pile to prevent overheating damage, and automatically restore the original power when the temperature returns to normal, thereby improving the reliability and service life of the charging pile. By using the Kalman filtering algorithm, the system can effectively filter out the measurement noise and instantaneous interference of a single sensor to obtain a target fusion temperature value that can more truly and stably reflect the overall thermal state of the charging pile, providing a highly reliable data cornerstone for subsequent decision-making, fundamentally avoiding false actions caused by data distortion and significantly improving the reliability of the system. By predicting the future temperature trend through the charging pile internal temperature prediction model, the system can take action in advance before the temperature reaches the dangerous threshold but is determined to exceed the standard, which completely changes the passive situation of traditional thermal management and realizes real preventive control. Through real-time monitoring and immediate response mechanism, the occurrence of hardware damage or other safety accidents caused by high temperature is effectively avoided; the material aging phenomenon caused by long-term high temperature environment is reduced, which helps to maintain the best performance state of the charging pile. Even in extreme weather conditions, the charging pile can also operate stably, improving the user's satisfaction. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0017] Figure 1 The system block diagram of the temperature feedback based charging pile power adaptive adjustment system of embodiment 1; Figure 2A temperature feedback charging pile power adaptive adjustment system system control flowchart for example 1; Figure 3 A data fusion module flowchart for example 1; Figure 4 A central processing module flowchart for example 1; DETAILED DESCRIPTION In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0018] The present application relates to the technical field of new energy vehicle charging equipment, and particularly relates to an intelligent control method and system based on a temperature sensor for real-time monitoring of internal temperature of a charging pile and automatic adjustment of output power according to temperature changes. With the rapid growth of the electric vehicle market, efficient and reliable charging infrastructure has become one of the key demands. However, traditional charging piles are prone to overheating problems under high load conditions, which not only affects the charging efficiency, but also can cause hardware failure and even safety accidents. Therefore, it is particularly important to develop a system that can effectively manage the internal temperature of the charging pile and dynamically adjust the working state accordingly. In contrast, the present application can significantly improve this situation by real-time monitoring of the internal temperature of the charging pile and dynamically adjusting the output power according to the actual situation. This method not only prevents safety hazards caused by excessive temperature, but also ensures that the charging pile is always in the best working state, thereby improving the reliability and user experience of the overall system.

[0019] The present application provides a charging pile power adaptive adjustment method based on temperature feedback, comprising: S1, real-time acquisition of temperature values of multiple key heating areas inside the charging pile; The key heating areas include the power module, the main control board and the DC input side of the charging pile.

[0020] S2, data fusion of the collected multiple temperature values is performed using a Kalman filtering algorithm to obtain a target temperature fusion value representing the overall thermal state of the charging pile; The data fusion of the collected multiple temperature values is performed using a Kalman filtering algorithm, specifically: Initialization step: set the initial state estimate value of temperature and the initial covariance matrix; Prediction step: based on the heat transfer model of the charging pile, the temperature state at the current time is predicted according to the temperature parameter at the last time, and the predicted state value and the predicted covariance are obtained; Update step: obtain the current temperature measurement value, calculate the Kalman gain, and update the predicted state value and the predicted covariance by using the Kalman gain, so as to obtain the state estimate value of the current time of the plurality of different key heat generating areas, and take the average value as the target temperature fusion value.

[0021] The heat transfer model of the charging pile is essentially a mathematical abstraction of the internal thermal dynamics process of the charging pile, which is established in accordance with the law of conservation of energy, that is, the change amount of the temperature in the charging pile per unit time is equal to the difference between the heat generated and the heat dissipated to the environment. The specific expression is:

[0022] Among them, C is the heat capacity (J / °C) of the key heat generating components of the charging pile, which represents the heat required to raise one degree, T is the estimated temperature (°C) inside the charging pile, t is the time (s); P loss is the total power loss, which is a function of the output power P output , , η is the efficiency, h is the comprehensive heat transfer coefficient, A is the effective heat dissipation area; T ambient is the ambient temperature (°C); Application of the model in the Kalman filter prediction stage

[0023] Among them, T k∣k-1 is the predicted temperature at the current time k ; T k-1∣k-1 is the optimal estimated temperature at the last time k -1; P loss,k-1 is the power loss at the last time; T ambient,k-1 is the ambient temperature at the last time; S3, compare the target temperature fusion value with a preset temperature safety threshold, and dynamically adjust the output power of the charging pile according to the comparison result; When the target temperature fusion value exceeds the preset temperature safety threshold, the output power is reduced; and when the target temperature fusion value falls below the preset temperature safety threshold, the output power is restored. The output power of the charging pile is dynamically adjusted, and the method specifically comprises the following steps: A plurality of temperature safety threshold intervals are set, and each interval corresponds to a different power adjustment strategy. When the target temperature fusion value enters a higher temperature safety threshold interval, the output power is reduced according to the strategy corresponding to the interval, and the reduction amplitude is related to the degree of temperature exceeding the threshold. When the target temperature fusion value falls to a lower temperature safety threshold interval, the output power is gradually increased according to the strategy corresponding to the interval.

[0024] Based on the charging pile internal temperature prediction model, the temperature change trend of the charging pile in the future period of time is predicted according to the historical operation data and external environment data of the charging pile through the machine learning model. When the target temperature fusion value has not yet exceeded the preset temperature safety threshold, but the charging pile internal temperature prediction model predicts that the internal temperature of the charging pile will exceed the preset temperature safety threshold in the future period of time, the operation of reducing the output power is performed in advance; and the physical cooling of the charging pile is performed in combination with the air cooling / liquid cooling cooperative temperature adjustment mechanism, so that the charging pile operates within the safe temperature range, and the power adaptive adjustment of the charging pile is realized.

[0025] The charging pile internal temperature prediction model, and the specific construction process is as follows: The historical temperature data, power output data and external environmental factor data of the charging pile are collected. After the collected data is preprocessed, the key features related to the internal temperature of the charging pile are extracted. Through the feature engineering method, the key features are converted into feature vectors reflecting the temperature change law of the charging pile. The extracted feature vectors are trained by using a machine learning algorithm to construct a charging pile internal temperature prediction model.

[0026] The machine learning model is a long short-term memory network model or a support vector regression model.

[0027] The air cooling / liquid cooling cooperative temperature adjustment mechanism specifically comprises: According to the target temperature fusion value, the air cooling system and / or the liquid cooling system inside the charging pile are controlled to adjust the cooling intensity, the air cooling system adjusts the speed of the cooling fan to cool the inside of the charging pile, and the liquid cooling system adjusts the flow of the cooling liquid to cool the inside of the charging pile.

[0028] The application also provides a charging pile power adaptive adjustment system based on temperature feedback, which comprises: temperature acquisition module, data fusion module, central control module, power adjustment module, prediction module and cooling control module; The temperature acquisition module comprises a plurality of temperature sensors arranged in key heating areas inside the charging pile respectively, for monitoring the working temperature inside the charging pile and collecting temperature values of each area in real time. The data fusion module is configured to perform data fusion on the collected temperature values through Kalman filtering algorithm to obtain a target temperature fusion value representing the overall thermal state of the charging pile. The central control module is connected with the data fusion module and is configured with a preset temperature safety threshold, and compares the target temperature fusion value with the preset temperature safety threshold. The power adjustment module dynamically adjusts the output power of the charging pile according to the instructions from the central control module and the comparison result. The prediction module is connected with the central control module and predicts the temperature change trend of the charging pile in a future period of time through a machine learning model according to historical operation data and external environment data of the charging pile. The cooling control module is connected with the central control module and is configured to control at least one of the air cooling system and the liquid cooling system to cooperate in temperature adjustment according to the instructions from the central control module, so as to physically cool the charging pile.

[0029] The air cooling system comprises a plurality of adjustable speed cooling fans installed inside the charging pile, and the speed of the fans is controlled by the central control module.

[0030] The liquid cooling system comprises a circulating pump, a radiator, a cooling liquid pipeline and a temperature sensor; the cooling liquid flows through the key heating areas inside the charging pile under the driving of the circulating pump, absorbs heat and then dissipates to the air through the radiator; the central control module controls the speed of the circulating pump and the flow of the cooling liquid according to the temperature value collected by the temperature sensor, so as to effectively control the temperature inside the charging pile.

[0031] By adopting the Kalman filtering algorithm, the system can effectively filter out the measurement noise and instantaneous interference of a single sensor, obtain a target fusion temperature value that can more truly and stably reflect the overall thermal state of the charging pile, and provide a highly reliable data cornerstone for subsequent decision-making, thereby fundamentally avoiding misoperation caused by distorted data and significantly improving the reliability of the system. By predicting the future temperature trend through the charging pile internal temperature prediction model, the system can take action in advance before the temperature reaches the dangerous threshold but is determined to exceed the standard, which completely changes the passive situation of traditional thermal management and realizes true preventive control. Through real-time monitoring and immediate response mechanism, the system can effectively avoid the occurrence of hardware damage or other safety accidents caused by high temperature; reduce the material aging phenomenon caused by long-term high-temperature environment, and help to maintain the best performance state of the charging pile. Even in extreme weather conditions, the charging pile can still operate stably, improving the satisfaction of users.

[0032] Embodiment 1 (1) Temperature acquisition module: contains multiple temperature sensors, which are arranged in the key heating areas (such as power modules, main control boards, etc.) inside the charging pile to comprehensively monitor the working temperature inside the charging pile. The collected temperature values are , ,…, .

[0033] (2) Data fusion module: responsible for comprehensive judgment and fusion calculation of the temperatures collected by multiple temperature sensors to obtain the final temperature value .

[0034] Fusion algorithm selection: the Kalman filtering algorithm is adopted for data fusion in the present application. The Kalman filtering algorithm is a kind of efficient self-recursion filter, which can estimate the state of the system from a series of noisy measurements, is suitable for multi-sensor data fusion scenarios, can effectively reduce the influence of measurement noise, and improve the accuracy of fusion results. The specific fusion process is as follows: S1, initialization: Initial state estimate value : assuming that the initial state estimate value is 35℃ Initial covariance matrix : assuming that the measurement error standard deviation of each temperature sensor is 0.5℃, the initial covariance matrix is = 0.25 (i.e. 0.52).

[0035] S2, prediction stage: State prediction : according to the heat transfer model of the charging pile, assuming that the temperature at the last time is 35℃, the power output is 50kW, the environmental temperature is 30℃, and the heat increment is 2℃, the temperature at the current time is predicted to be: = 35 + 2 = 37°C.

[0036] Covariance prediction : Assuming process noise covariance Q = 0.1 (representing uncertainty in system model), then: = + Q = 0.25 + 0.1 = 0.35 S3, Update stage: Measurement : Assuming the measurements of the three temperature sensors are 36°C, 37°C, and 38°C respectively. Calculate the average of the measurements as the measurement at the current time: = =

[0037] Kalman gain : Assuming measurement noise covariance R = 0.25 (representing a standard deviation of 0.5°C in sensor measurement error), then: = = = ≈ 0.583 State update : = + ( - ) = 37 + 0.583 × (37 - 37) = 37°C Covariance update : = (1 - ) = (1 - 0.583) × 0.35 ≈ 0.148 S4, Combine measurements from multiple sensors: First sensor update: = = = ≈ 0.583 = + ( - ) = 37 + 0.583 × (36 - 37) ≈ 36.417°C Second sensor update: = = = ≈0.583 = + ( - )=37+0.583×(37-37)≈37℃ Third sensor update: = = = ≈0.583 = + ( - )=37+0.583×(38-37)≈37.583℃ S5, final fusion value: Take the average of the updated values of the three sensors to get the target temperature fusion value: = = ≈37.2℃ (4) Control logic steps as follows: Temperature collection: Use high-precision temperature sensors to continuously monitor the temperature values of key positions inside the charging pile. For example, assume that three temperature sensors are installed on the power module, main control board and DC input side of the charging pile, respectively, and the temperature values collected are =36℃, =37℃ and =38℃.

[0038] Data fusion: The collected temperature values are fused and calculated by the fusion algorithm. After the initialization, prediction and update stages, the target temperature fusion value =37.2℃.

[0039] Threshold setting: The user sets a series of temperature safety thresholds in advance, such as normal working temperature range 25℃~55℃, warning temperature 60℃, emergency shutdown temperature 85℃, etc.

[0040] Temperature comparison: Compare the real-time collected temperature data with the set temperature safety threshold. For example, the current target temperature fusion value =37.2℃, is within the normal working temperature range.

[0041] Power adjustment: Once it is found that the temperature of a certain area exceeds the safety limit, the corresponding power reduction strategy is immediately started. The specific power reduction amplitude can be flexibly adjusted according to the actual temperature deviation. For example, when the temperature exceeds 60℃, reduce the output power by 10% for every 5℃ rise, and the minimum is 40% of the rated power. When the temperature falls below 60℃, gradually increase the output power to 100% in stages.

[0042] Recording and reporting: During the entire process, all operations and their corresponding temperature changes are recorded in detail and uploaded to the cloud server regularly for technical personnel to refer to and analyze. For example, the record shows that the temperature reaches 65℃ at a certain time, the system automatically reduces the output power from 60kW to 54kW, and uploads this operation record and temperature change value to the cloud server.

[0043] Combining the internal temperature prediction model of the charging pile and the air cooling / liquid cooling collaborative temperature regulation mechanism Internal temperature prediction model of charging pile: Data collection and preprocessing: The cloud server collects historical temperature data, power output data, and external environmental factors (such as weather temperature, humidity, wind speed, etc.) data of the charging pile. The collected data is cleaned, normalized, and other preprocessing operations are performed to remove noise and outliers, so as to facilitate subsequent analysis and modeling.

[0044] Feature extraction: Extract key features related to the internal temperature of the charging pile from the preprocessed data, such as temperature change rate, power and temperature correlation, external environmental temperature and charging pile internal temperature lag relationship, etc. Through feature engineering methods, the original key feature data is converted into a feature vector that can effectively reflect the temperature change law of the charging pile.

[0045] Model training: Use machine learning algorithms (such as long short-term memory network model LSTM, support vector regression SVR model, etc.) to train the extracted feature vectors to build an internal temperature prediction model of the charging pile. The LSTM algorithm can well handle the long-term dependence relationship in time series data and is suitable for predicting the internal temperature change trend of the charging pile; the SVR algorithm has good generalization ability and prediction accuracy and performs well in handling small sample data. According to the actual running data of the charging pile and the prediction requirements, select the appropriate machine learning algorithm for model training.

[0046] Model application: The trained prediction model can receive real-time current operating state data and external environmental factor data of the charging pile, and predict the possible temperature peaks inside the charging pile in the future period of time. According to the prediction result, the output power of the charging pile is adjusted in advance to avoid affecting the charging efficiency and equipment safety due to high temperature. For example, if it is predicted that the internal temperature of the charging pile may exceed the preset temperature safety threshold in high temperature weather, the output power can be reduced in advance, and the physical cooling is performed by combining the air cooling / liquid cooling cooperative cooling mechanism to ensure that the charging pile operates within the safe temperature range.

[0047] Air cooling / liquid cooling cooperative cooling mechanism: Air cooling system: Install multiple adjustable speed cooling fans inside the charging pile. According to the real-time monitoring value of the internal temperature of the charging pile, the speed of the fan is controlled by the central control module. When the temperature exceeds a certain threshold, the fan speed is increased to increase air flow and speed up heat dissipation. The speed of the fan can be adjusted in multiple stages according to the temperature to achieve precise air cooling control.

[0048] Liquid cooling system: For some high-power and high-heat charging piles, a liquid cooling system can also be equipped. The liquid cooling system is composed of a circulating pump, a radiator, a cooling liquid pipeline and a temperature sensor. The cooling liquid flows through the key heat generating areas inside the charging pile under the drive of the circulating pump, absorbs heat and then dissipates to the air through the radiator. The central control module controls the speed of the circulating pump and the flow of the cooling liquid according to the temperature value collected by the temperature sensor, to effectively control the internal temperature of the charging pile. In extreme high temperature environment, the liquid cooling system can work with the air cooling system to jointly reduce the internal temperature of the charging pile. When the temperature exceeds the adjustment range of the air cooling system, start the liquid cooling system, through the synergistic effect of the two cooling methods, to ensure that the internal temperature of the charging pile always remains within the safe range.

[0049] Figure 1 System block diagram of the temperature feedback charging pile power adaptive adjustment system of embodiment 1. The system is composed of multiple temperature sensors, a central processor, a display and a cloud server.

[0050] Figure 2 System control flow chart of the temperature feedback charging pile power adaptive adjustment system of embodiment 1. The temperature acquisition module obtains the temperature values collected by the temperature sensors in the key areas inside the charging pile , ,…, , the data fusion module fuses and calculates , ,…, outputs , the central processing module compares the user set temperature safety threshold with Compare and adjust the output power, and record and report the process data to the cloud server.

[0051] Figure 3 The data fusion module flowchart for Example 1. Mainly, the temperature values collected by the temperature sensor of the temperature acquisition module are fused and calculated by the Kalman filtering algorithm, and the output .

[0052] Figure 4 The central processing module flowchart for Example 1. Mainly, the T_final output by the data fusion module is compared with the user-set temperature safety threshold value, and if the temperature exceeds the threshold value, the power is adjusted; otherwise, it is further determined whether the power needs to be restored according to whether the temperature returns to the safe range. The operation record and temperature change value are recorded and reported to the cloud server.

[0053] A charging pile power self-adaptive adjustment system based on temperature feedback, 1. Hardware arrangement: In addition to the temperature sensors already arranged in the power module, main control board and DC output side inside the charging pile, 4 groups of high-precision digital temperature sensors (±0.1℃) are additionally arranged at the inlet / outlet air ports, liquid cooling plate surface and environmental temperature probe. All sensors communicate with the ARM main control board through CAN_FD bus with a sampling period of 1s.

[0054] 2. Online deployment of charging pile internal temperature prediction model 2.1 Model offline training Data source: The cloud server pulls the full amount of operation logs (temperature, power, environmental temperature and humidity, wind speed) and fault records every day, which are stored in the time series database InfluxDB after Kafka stream cleaning.

[0055] Feature engineering: sliding window 30min, extracting 36-dimensional features such as temperature gradient, power-temperature lag 5min correlation, environmental dew point, and air cooling duty cycle.

[0056] Algorithm: adopt double-layer Stacking structure, the first layer is LSTM (128 hidden units) + SVR (RBF kernel), and the second layer is LightGBM regressor, the target is the future 10min power module temperature peak value.

[0057] Training frequency: once a week, using the past 30 days data, after 5-fold CV, if MAE<0.8℃, update the model. The model file (≈4MB) is encrypted by AES-128 algorithm and sent to the local TF card through MQTT communication.

[0058] 2.2 Model online inference Inference framework: ARM mainboard runs TensorFlow Lite Micro (TFLM) inference engine, occupying RAM <2MB, and inference time is 8ms.

[0059] Triggering logic: The model input vector is called every 30s, and if the predicted value T_pred≥ warning threshold 60℃, the power reduction is triggered 5min in advance, with a reduction ΔP=k·(T_pred-60), k=2% / ℃.

[0060] 3. Air cooling / liquid cooling collaborative temperature regulation mechanism 3.1 Air cooling subsystem, fan: 4 120mm PWM fans (0-100% duty cycle adjustable), partition control (2 for inlet and outlet).

[0061] Control strategy: - Normal zone (T<50℃): duty cycle 30%, noise≤45dB(A).

[0062] - Warning zone (50℃≤T<60℃): duty cycle linearly increases to 70%.

[0063] - Dangerous zone (T≥60℃): duty cycle 100%, and simultaneously requests liquid cooling intervention.

[0064] 3.2 Liquid cooling subsystem Composition: 25% ethylene glycol aqueous solution, 12V DC brushless pump (0-6L / min), aluminum micro-channel liquid cooling plate attached to power module, external radiator with 2 auxiliary fans.

[0065] Control strategy: - When T≥60℃ and air cooling has been running at full speed for 30s without temperature reduction, the main control board issues a command to start the liquid cooling pump through CAN communication, with an initial flow rate of 3L / min, increasing by 0.5L / min for every 1℃ increase, up to 6L / min.

[0066] - When T<55℃, first reduce the pump speed to 2L / min, and if there is no temperature rise for 120s, completely shut down to prevent frequent start-stop.

[0067] 3.3 Collaborative logic Priority: advance action predicted by the charging pile internal temperature prediction model > real-time threshold > user manual instruction.

[0068] Arbitration: if the prediction of the charging pile internal temperature prediction model and the real-time threshold are triggered at the same time, the more conservative power-temperature compromise curve is adopted to ensure that the fan and liquid cooling pump do not produce control conflicts in any case.

[0069] 4. Operation flow (1) Temperature sensor 1s collects once → Kalman filter fusion → T_final.

[0070] (2) T_final is synchronized into AI inference module to obtain T_pred_10min.

[0071] (3) If T_pred_10min≥60℃, immediately reduce power according to ΔP=k·(T_pred-60), and increase the duty ratio of air cooling.

[0072] (4) If T_final is still ≥60℃ after 30s, start the liquid cooling pump and enter the cooperative cooling mode.

[0073] (5) When T_final<55℃, the liquid cooling pump first slows down → stop the pump; the air cooling duty ratio slows down → 30%; the power is restored to 100% according to a slope of 5% / min.

[0074] (6) The power, temperature, fan speed, pump flow, and AI prediction value of the whole process are packaged into JSON, and uploaded to the cloud every 60s through TLS1.3, which is used for the next round of model iteration.

[0075] It should be understood that the terms "comprising" and "including" as used in the specification and the appended claims indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0076] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description herein is for describing the specific embodiments only and is not intended to be limiting. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0077] The above description is only the preferred embodiment of the present application, and is not intended to limit the present application in any form. Any person skilled in the art can easily implement the present application according to the drawings and the above description. However, any equivalent changes, modifications and evolutions made by those skilled in the art within the scope of the technical solutions of the present application, using the technical content disclosed above, are equivalent embodiments of the present application. At the same time, any equivalent changes, modifications and evolutions made by those skilled in the art according to the essence of the present application to the above embodiments are still within the protection scope of the technical solutions of the present application.

Claims

1. A temperature feedback based power self-adaptive adjustment method for charging piles, characterized in that, The application relates to a charging pile temperature safety control method and device. Real-time acquisition of temperature values of multiple key heating areas inside a charging pile; Data fusion of the acquired multiple temperature values by using a Kalman filtering algorithm to obtain a target temperature fusion value representing the overall thermal state of the charging pile; Comparison of the target temperature fusion value with a preset temperature safety threshold value, and dynamic adjustment of the output power of the charging pile according to the comparison result; When the target temperature fusion value exceeds the preset temperature safety threshold value, the output power is reduced; when the target temperature fusion value falls below the preset temperature safety threshold value, the output power is restored. When the target temperature fusion value does not exceed the preset temperature safety threshold value, but the internal temperature of the charging pile is predicted to exceed the preset temperature safety threshold value in a future period of time through an internal temperature prediction model, the operation of reducing the output power is performed in advance; meanwhile, the charging pile is physically cooled by combining a forced air cooling / liquid cooling collaborative temperature regulation mechanism, so that the charging pile operates within a safe temperature range.

2. The method of claim 1, wherein, Data fusion of the acquired multiple temperature values by using a Kalman filtering algorithm, specifically including: Initialization step: setting an initial state estimation value of temperature and an initial covariance matrix; Prediction step: predicting the temperature state at the current moment based on a heat transfer model of the charging pile according to the temperature parameters at the last moment to obtain a predicted state value and a predicted covariance; Update step: obtaining the current temperature measurement value, calculating the Kalman gain, and updating the predicted state value and the predicted covariance by using the Kalman gain to obtain the state estimation value of the multiple different key heating areas at the current moment, and taking the average value as the target temperature fusion value. 3.The temperature feedback based power self-adaptive adjustment method of a charging pile according to claim 1, characterized in that, The key heating areas include a power module, a main control board and a direct current input / output side of the charging pile.

4. The method of claim 1, wherein, Dynamic adjustment of the output power of the charging pile, specifically including: Setting multiple temperature safety threshold value intervals, each interval corresponding to a different power adjustment strategy; When the target temperature fusion value enters a higher temperature safety threshold value interval, the output power is reduced according to the strategy corresponding to the interval, and the reduction amplitude is related to the degree of temperature exceeding the threshold value; When the target temperature fusion value falls to a lower temperature safety threshold value interval, the output power is gradually increased according to the strategy corresponding to the interval.

5. The method of claim 1, wherein, The internal temperature prediction model of the charging pile, specifically including the following construction process: Acquisition of historical temperature data, power output data and external environmental factor data of the charging pile; After preprocessing the acquired data, key features related to the internal temperature of the charging pile are extracted; Through a feature engineering method, the key features are converted into a feature vector reflecting the temperature change law of the charging pile; A machine learning algorithm is used to train the extracted feature vector to construct an internal temperature prediction model of the charging pile.

6. The method of claim 1, wherein, Based on the internal temperature prediction model of the charging pile, the temperature change trend of the charging pile in a future period of time is predicted through a machine learning model according to the historical operation data and external environmental data of the charging pile.

7. The method of claim 1, wherein, The forced air cooling / liquid cooling collaborative temperature regulation mechanism, specifically including: According to the target temperature fusion value, the air cooling system and / or the liquid cooling system inside the charging pile are controlled to adjust the cooling intensity, the air cooling system cools the inside of the charging pile by adjusting the rotating speed of the cooling fan, and the liquid cooling system cools the inside of the charging pile by adjusting the flow of the cooling liquid.

8. A temperature feedback-based charging pile power self-adaptive adjustment system, characterized in that, Comprise: a temperature acquisition module, a data fusion module, a central control module, a power adjustment module, a prediction module and a cooling control module; The temperature acquisition module comprises a plurality of temperature sensors arranged in the key heating areas inside the charging pile, respectively, for monitoring the working temperature inside the charging pile and collecting the temperature values of each area in real time; The data fusion module is used for data fusion of the collected multiple temperature values through Kalman filtering algorithm to obtain a target temperature fusion value representing the overall thermal state of the charging pile; The central control module is connected with the data fusion module, is configured with a preset temperature safety threshold, and compares the target temperature fusion value with the preset temperature safety threshold; The power adjustment module dynamically adjusts the output power of the charging pile according to the instructions issued by the central control module and the comparison result; The prediction module is connected with the central control module, and predicts the temperature change trend of the charging pile in a future period of time according to the historical operation data and external environment data of the charging pile through a machine learning model; The cooling control module is connected with the central control module, and is configured to control at least one of the air cooling system and the liquid cooling system to cooperate in temperature adjustment according to the instructions of the central control module, so as to physically cool the charging pile.

9. The temperature feedback based charging pile power self-adaptive adjustment system according to claim 8, characterized in that, The air cooling system comprises a plurality of adjustable rotating speed cooling fans installed inside the charging pile, and the rotating speed of the fan is controlled by the central control module; The liquid cooling system comprises a circulating pump, a radiator, a cooling liquid pipeline and a temperature sensor; the cooling liquid flows through the key heating areas inside the charging pile under the driving of the circulating pump, absorbs heat and is dissipated to the air through the radiator, and the central control module controls the rotating speed of the circulating pump and the flow of the cooling liquid according to the temperature value collected by the temperature sensor, so as to effectively control the temperature inside the charging pile.

10. A charging post, characterized in that, Comprise the charging pile power adaptive adjustment system based on temperature feedback as claimed in claim 8 or 9.

Images