An intelligent control method and system for charging piles based on the Internet of Things

Abstract

The application relates to the technical field of intelligent control, and discloses a charging pile intelligent control method and system based on the Internet of Things, which comprises the following steps: collecting real-time data of an electric vehicle battery through a charging pile sensor and performing smoothing processing to obtain a battery state sequence; calculating a current polarization voltage according to the sequence; determining a lithium precipitation critical voltage threshold based on the battery temperature and the state of charge; if the current polarization voltage reaches or exceeds the threshold, predicting the future change trend of the polarization voltage by using a neural network model and performing risk assessment; dynamically adjusting a charging strategy according to the assessment result, calculating a new maximum allowable current; updating charging pile control parameters according to the new maximum allowable current to generate an optimized charging curve; and finally monitoring the battery response and determining a final charging strategy according to the polarization voltage stability. The method can realize real-time and accurate prevention and control of the lithium precipitation risk in the fast charging process, and significantly improves the charging safety and the battery life.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent control method and system for charging piles based on the Internet of Things. Background Technology

[0002] In the rapidly developing modern new energy vehicle industry, the manufacturing of electric vehicle charging infrastructure plays a key pillar role in supporting the industry's sustainable growth. In particular, the intelligent management of charging piles, which is directly related to energy efficiency and battery safety, is becoming increasingly important because it not only affects user experience but also involves grid stability and extended battery life.

[0003] In one existing technology, a local self-organizing network is typically used to collect multiple data points such as current, voltage, and temperature in real time through a low-power module embedded in the charging pile. A smart gateway on-site performs protocol conversion, and the data is then transmitted back to a cloud platform for aggregation. Charging control is then performed using a pre-defined fixed charging curve or simple feedback adjustment. However, in environments with fluctuating battery states, traditional charging control methods can easily lead to a disconnect between the charging process and the actual internal reactions of the battery. During lithium battery charging, a polarization voltage is generated inside the battery, reflecting its immediate response to the charging current. If this polarization voltage continuously approaches or exceeds the lithium plating threshold, it can cause safety hazards, leading to internal short circuits or increased risk of thermal runaway.

[0004] Therefore, existing technologies that rely on fixed charging curves or simple feedback mechanisms cannot accurately estimate and track polarization voltage in real time. This makes it impossible to accurately assess the risk of lithium plating and adjust the charging strategy dynamically in a timely manner, which has a significant negative impact on the safety of fast charging and battery life. Summary of the Invention

[0005] This invention provides an intelligent control method and system for charging piles based on the Internet of Things, which addresses the key issue of how to dynamically determine the maximum allowable current by acquiring battery data in real time and accurately calculating the deviation between polarization voltage and lithium plating threshold, thus ensuring the safety of fast charging.

[0006] Firstly, in order to solve the above-mentioned technical problems, the present invention provides an intelligent control method for charging piles based on the Internet of Things, comprising:

[0007] Real-time data of electric vehicle batteries is collected by sensors in charging piles, and the real-time data is smoothed to obtain a smooth battery state sequence.

[0008] Based on the battery state sequence, the current polarization voltage is calculated using a second-order RC equivalent circuit model;

[0009] The lithium plating boundary voltage is obtained based on the temperature and the state of charge, and the difference is calculated with a preset voltage safety margin to obtain the critical voltage threshold.

[0010] If the current polarization voltage is greater than or equal to the critical voltage threshold, the trend of polarization voltage change is predicted using a preset neural network model, and the trend of change is risk-mapped to obtain a risk assessment result.

[0011] Based on the risk assessment results, the corresponding polarization voltage change trend value is extracted. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, the current charging control strategy is adjusted and the maximum allowable current is calculated.

[0012] The control parameters of the charging pile are updated according to the maximum allowable current, a new charging control point is calculated and connected to obtain the optimized charging curve.

[0013] Based on the optimized charging curve, the battery response is monitored to obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence. The fluctuation variance is compared with a preset stability threshold, and the final charging strategy control and adjustment are determined based on the comparison result.

[0014] In one optional implementation, the step of collecting real-time data of the electric vehicle battery through the charging pile sensor and smoothing the real-time data to obtain a smooth battery state sequence includes:

[0015] Real-time data including battery terminal voltage, loop current, state of charge, and temperature are collected by the charging pile's sensors.

[0016] The real-time data is smoothed using a Kalman filter algorithm to obtain a smoothed battery state sequence.

[0017] In one optional implementation, the step of calculating the current polarization voltage using a second-order RC equivalent circuit model based on the battery state sequence includes:

[0018] The parameters of the second-order RC equivalent circuit model are obtained by performing parameter identification on the battery state sequence.

[0019] The static open-circuit voltage value and ohmic voltage drop component are determined based on the battery state sequence.

[0020] The total polarization voltage residual vector is obtained by subtracting the static open-circuit voltage value and the ohmic voltage drop component from the battery terminal voltage observation value collected by the charging pile sensor.

[0021] Based on the model parameters, a state-space equation is constructed, and the total polarization voltage residual vector is substituted into the state-space equation to analyze the electrochemical polarization voltage component and the concentration polarization voltage component.

[0022] The current polarization voltage is determined by linearly superimposing the electrochemical polarization voltage component and the concentration polarization voltage component.

[0023] In one optional implementation, the step of obtaining the lithium plating boundary voltage based on the temperature and the state of charge and subtracting it from a preset voltage safety margin to obtain a critical voltage threshold includes:

[0024] The lithium plating boundary voltage is obtained by retrieving a preset lithium plating boundary voltage mapping table based on the temperature and the state of charge.

[0025] The critical voltage threshold is obtained by subtracting a preset voltage safety margin from the lithium plating boundary voltage.

[0026] In one optional implementation, predicting the trend of polarization voltage using a preset neural network model includes:

[0027] Historical polarization voltage time series data are extracted and standardized to obtain an input feature vector containing voltage difference information;

[0028] The input feature vector is fed into a pre-trained long short-term memory network model for time-series extrapolation to obtain the polarization voltage prediction value within the future time window;

[0029] Calculate the first-order difference of the predicted polarization voltage to determine the trend of polarization voltage variation that characterizes the direction of voltage evolution.

[0030] In one optional implementation, adjusting the current charging control strategy and calculating the maximum allowable current includes:

[0031] The target current decay is calculated based on the real-time charging current value and the preset current lower limit.

[0032] The corrected maximum allowable current is obtained by subtracting the target current decay from the real-time charging current value.

[0033] In one optional implementation, the step of updating the control parameters of the charging pile based on the maximum allowable current, calculating and connecting a new charging control point, and obtaining an optimized charging curve includes:

[0034] The maximum allowable current is encapsulated into a control command frame and sent to the charging pile to adjust the output voltage and current, thereby generating actual output current data.

[0035] The optimal voltage reference value is calculated based on the actual output current data, the current state of charge of the battery, and the temperature.

[0036] A charging control point is constructed based on the optimal voltage reference value and the actual output current data. Multiple charging control points are connected to form an optimized charging curve.

[0037] In one optional implementation, the step of monitoring the battery response based on the optimized charging curve, obtaining a polarization voltage sequence and calculating the fluctuation variance of the polarization voltage sequence, comparing the fluctuation variance with a preset stability threshold, and determining the final charging strategy control and adjustment based on the comparison result includes:

[0038] The real-time terminal voltage value of each battery cell and the corresponding estimated open-circuit voltage data are collected based on the optimized charging curve.

[0039] The polarization voltage sequence is obtained by subtracting the open-circuit voltage estimation data from the real-time terminal voltage value, and the fluctuation variance of the polarization voltage sequence is calculated.

[0040] If the fluctuation variance is less than a preset stability threshold, then maintain the control command for the current state;

[0041] If the fluctuation variance is not less than the stability threshold, a preset conservative charging instruction is executed, and then the process is traced back to the step of collecting real-time data of the electric vehicle battery through the charging pile sensor, smoothing the real-time data to obtain a smooth battery state sequence, and then a new round of charging strategy adjustment is carried out.

[0042] Secondly, the present invention provides an intelligent control system for charging piles based on the Internet of Things, comprising:

[0043] The data acquisition and processing module is used to acquire real-time data of electric vehicle batteries through charging pile sensors, and to smooth the real-time data to obtain a smooth battery state sequence.

[0044] The polarization voltage calculation module is used to calculate the current polarization voltage based on the battery state sequence using a second-order RC equivalent circuit model.

[0045] The critical voltage determination module is used to obtain the lithium plating boundary voltage based on the temperature and the state of charge, and then subtract it from a preset voltage safety margin to obtain the critical voltage threshold.

[0046] The risk prediction module is used to predict the trend of polarization voltage change using a preset neural network model if the current polarization voltage is greater than or equal to the critical voltage threshold, and to perform risk mapping on the trend of change to obtain a risk assessment result.

[0047] The strategy adjustment module is used to extract the corresponding polarization voltage change trend value based on the risk assessment result. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, the current charging control strategy is adjusted and the maximum allowable current is calculated.

[0048] The control update module is used to update the control parameters of the charging pile according to the maximum allowable current, calculate and connect the new charging control point, and obtain the optimized charging curve.

[0049] The monitoring and adjustment module is used to monitor the battery response based on the optimized charging curve, obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence, compare the fluctuation variance with a preset stability threshold, and determine the final charging strategy control and adjustment based on the comparison result.

[0050] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the IoT-based intelligent control method for charging piles as described above.

[0051] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the IoT-based intelligent control method for charging piles described above.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] (1) The present invention uses the Kalman filter algorithm to smooth the raw data collected by the sensor and combines it with the second-order RC equivalent circuit model to accurately calculate the real-time polarization voltage, which effectively overcomes the estimation deviation caused by the dynamic change of battery parameters and realizes accurate perception of the internal state of the battery.

[0054] (2) This invention introduces a trend prediction model based on neural networks (such as long short-term memory networks). When the polarization voltage is close to the safety threshold, it can proactively assess its future trend and lithium plating risk, realizing the transformation from passive threshold judgment to proactive risk prediction, and significantly improving the timeliness of risk prevention and control.

[0055] (3) The present invention establishes a complete closed-loop control and dynamic optimization mechanism, dynamically adjusts the maximum allowable charging current according to the real-time risk assessment results, and generates an optimized charging curve. Under the premise of ensuring battery safety, it maximizes the fast charging efficiency and achieves an effective balance between safety and charging speed, thereby extending the battery life. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the intelligent control method for charging piles based on the Internet of Things provided in the first embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of the structure of the Internet of Things-based intelligent control system for charging piles provided in the second embodiment of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Reference Figure 1 The first embodiment of the present invention provides an intelligent control method for charging piles based on the Internet of Things, including the following steps:

[0060] S101, real-time data of electric vehicle battery is collected through charging pile sensors, and the real-time data is smoothed to obtain a smooth battery state sequence.

[0061] S102, calculate the current polarization voltage using a second-order RC equivalent circuit model based on the battery state sequence;

[0062] S103, obtain the lithium plating boundary voltage based on the temperature and the state of charge, and subtract it from the preset voltage safety margin to obtain the critical voltage threshold.

[0063] S104, if the current polarization voltage is greater than or equal to the critical voltage threshold, then the change trend of the polarization voltage is predicted using a preset neural network model, and the change trend is risk-mapped to obtain a risk assessment result.

[0064] S105, Based on the risk assessment results, extract the corresponding polarization voltage change trend value. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, adjust the current charging control strategy and calculate the maximum allowable current.

[0065] S106, Update the control parameters of the charging pile according to the maximum allowable current, calculate the new charging control point and connect it to obtain the optimized charging curve;

[0066] S107, Based on the optimized charging curve, monitor the battery response, obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence, compare the fluctuation variance with a preset stability threshold, and determine the final charging strategy control and adjustment based on the comparison result.

[0067] In step S101, real-time data of the electric vehicle battery is collected through the charging pile sensor, and the real-time data is smoothed to obtain a smoothed battery state sequence, including:

[0068] Real-time data including battery terminal voltage, loop current, state of charge, and temperature are collected by the charging pile's sensors;

[0069] The real-time data is smoothed using a Kalman filter algorithm to obtain a smoothed battery state sequence.

[0070] First, real-time data during the charging process of electric vehicle batteries is synchronously collected at a fixed sampling frequency (e.g., 10 Hz) by the sensor array integrated in the charging pile (including voltage sensors, current sensors, and temperature sensors). This includes battery terminal voltage (unit: V), loop current (unit: A), and battery temperature (unit: °C). All sensor data are appended with millisecond-level timestamps to ensure that multi-dimensional data are strictly aligned on the time axis. The state of charge is estimated in real time using the ampere-hour integration method. Specifically, starting from the battery's fully charged state, the real-time current measured by the high-precision current sensor is integrated over time to obtain the cumulative charge or discharge amount. This is then divided by the battery's rated capacity, and the estimated state of charge value is obtained by combining the initial state of charge with coulombic efficiency correction.

[0071] Secondly, to address the random noise and transient anomalies that may exist in the real-time data due to electromagnetic interference, circuit noise, or fluctuations in the sensor itself, a Kalman filter algorithm is used to smooth the raw data. Specifically, the battery voltage, current, and temperature are defined as system state variables. Within each sampling period, an iterative "prediction-update" operation is performed: first, the state at the current moment is predicted based on the state estimate from the previous moment and the system dynamic model (e.g., a constant-state model or a simple battery equivalent circuit model); then, the predicted value is corrected using the actual observations collected at the current moment (i.e., sensor readings), thus outputting the optimal estimate of the state at that moment. This process is iteratively repeated, ultimately smoothing the raw data to obtain a smooth battery state sequence. For example, the raw voltage observation may fluctuate randomly between 3.60V and 3.65V; after Kalman filtering, the output sequence presents a smooth trajectory stable around 3.62V, more clearly reflecting the true state of the battery voltage.

[0072] In step S102, the current polarization voltage is calculated using a second-order RC equivalent circuit model based on the battery state sequence, including: parameter identification of the battery state sequence to obtain the model parameters of the second-order RC equivalent circuit model;

[0073] The static open-circuit voltage value and ohmic voltage drop component are determined based on the battery state sequence.

[0074] The total polarization voltage residual vector is obtained by subtracting the static open-circuit voltage value and the ohmic voltage drop component from the battery terminal voltage observation value collected by the charging pile sensor.

[0075] Based on the model parameters, a state-space equation is constructed, and the total polarization voltage residual vector is substituted into the state-space equation to analyze the electrochemical polarization voltage component and the concentration polarization voltage component.

[0076] The current polarization voltage is determined by linearly superimposing the electrochemical polarization voltage component and the concentration polarization voltage component.

[0077] First, parameter identification is performed on the battery state sequence to obtain the model parameters of the second-order RC equivalent circuit model. It should be noted that the second-order RC equivalent circuit model is a physical circuit model that represents the complex electrochemical processes inside the battery as equivalent to a circuit consisting of an ohmic internal resistance R0, an electrochemical polarization resistance R1, an electrochemical polarization capacitance C1, a concentration polarization resistance R2, and a concentration polarization capacitance C2, where R1 and C1 are connected in parallel, and R2 and C2 are connected in parallel. The model parameters are determined by least squares fitting. Specifically, the voltage-current relationship equation is first derived based on the second-order RC equivalent circuit model. Then, with the objective function being the minimum sum of squared errors between the model output voltage and the actual voltage in the battery state sequence, an optimization problem is constructed for the above five model parameters. This optimization problem is then solved iteratively to obtain the optimal solution that minimizes the objective function. The convergence criterion is that the 2-norm of the difference between the parameter vectors of two adjacent iterations is less than a preset threshold. This threshold can be set according to the battery type and service life, with a basic convergence threshold of 1×10⁻⁻⁻⁶. 4 The lifespan of new batteries (≤1 year) can be reduced to 5×10⁻ 5 Older batteries (used for more than 3 years) can be upgraded to 5×10⁻ 4 For example, after fitting the state sequence of a ternary lithium battery using the least squares method, the model parameters obtained are R0=0.02Ω, R1=0.015Ω, C1=800F, R2=0.03Ω, and C2=2000F.

[0078] Next, the static open-circuit voltage and ohmic voltage drop component are determined based on the battery state sequence. The static open-circuit voltage value can be obtained by looking up a pre-established OCV-SOC table, which establishes the correspondence between the battery's state of charge (SOC) and open-circuit voltage (OCV) (OCV-SOC curve). The corresponding static open-circuit voltage value is obtained by looking up the table based on the SOC data in the battery state data. Subsequently, based on the ohmic internal resistance (R0) in the second-order RC equivalent circuit model and the current data in the battery state sequence, the ohmic voltage drop component is calculated according to Ohm's law. During the calculation, attention is paid to the influence of the current direction on the sign of the voltage drop (the current is positive during charging, and the ohmic voltage drop is in the same direction as the current; the current is negative during discharging, and the voltage drop is in the opposite direction). For example, if the static open-circuit voltage value is 3.65V, the current charging current is 10A, and the ohmic internal resistance is R0 = 0.02Ω, the calculated ohmic voltage drop component is 0.02Ω × 10A = 0.2V.

[0079] Then, by subtracting the static open-circuit voltage value and the ohmic voltage drop component from the battery terminal voltage observation value collected by the charging pile sensor, the total polarization voltage residual vector can be obtained. Specifically, by subtracting the static open-circuit voltage value and the ohmic voltage drop component sequence from the terminal voltage sequence in the battery state sequence, a one-dimensional vector with the same length as the battery state sequence can be obtained, which is the total polarization voltage residual vector.

[0080] Subsequently, based on the model parameters of the previously determined second-order RC equivalent circuit model, the state-space equations are constructed. Specifically, the state variables are defined as the electrochemical polarization voltage U1(k) (where k represents the current time) and the concentration polarization voltage U2(k), and the state vector is x(k) = [U1(k), U2(k)]^T, where T represents the transpose. The state equation is defined as x(k+1) = Ax(k) + BI(k), where x(k+1) is the state vector at time k+1, and I(k) is the input current at time k (positive for charging, negative for discharging). A is the state transition matrix, with the following specific form:

[0081]

[0082] The input matrix has the following specific form:

[0083]

[0084] Where Δt is the system sampling time interval, and R1, C1, R2, and C2 are the parameters in the previously determined second-order RC equivalent circuit model.

[0085] Subsequently, the observation equation y(k) = Cx(k) + v(k) is established, where y(k) is the observed value at time k, i.e., the value of the corresponding time in the total polarization voltage residual, C is the observation matrix [1, 1], indicating that the two state components are linearly superimposed, and v(k) is the observation noise, which is set to white noise with a mean of 0. Based on the constructed state-space equation, Kalman filtering or a Luneburger observer is used to perform recursive calculations starting from the initial state. Combining the total polarization voltage residual vector as the observed value, the time series of the electrochemical polarization voltage component and the concentration polarization voltage component are estimated in real time. Finally, the electrochemical polarization voltage component sequence and the concentration polarization voltage component sequence can be obtained, and their dimensions are the same as the battery state sequence. The two components at the current time are extracted from the sequence and linearly superimposed to obtain the current polarization voltage value.

[0086] In step S103, the lithium plating boundary voltage is obtained based on the temperature and the state of charge, and the difference is calculated with a preset voltage safety margin to obtain the critical voltage threshold, including:

[0087] The lithium plating boundary voltage is obtained by retrieving a preset lithium plating boundary voltage mapping table based on the temperature and the state of charge.

[0088] The critical voltage threshold is obtained by subtracting a preset voltage safety margin from the lithium plating boundary voltage.

[0089] First, it should be noted that the lithium plating boundary voltage mapping table is pre-calibrated through numerous battery charge-discharge experiments. Indexed by battery temperature and state of charge (SOC), it stores the boundary voltage values ​​corresponding to the onset of lithium plating under different operating conditions. For example, one way to construct this table is as follows: at different constant temperatures (e.g., 0°C, 10°C, 25°C, 40°C) and different initial SOCs, the battery is charged with a series of constant currents. Simultaneously, the lithium plating initiation point is detected online using high-precision differential voltage analysis (DVA) or capacity differential curve (dQ / dV) methods. The battery terminal voltage corresponding to this point is recorded, which is the lithium plating boundary voltage under that temperature-SOC combination. Integrating all the obtained lithium plating boundary voltages yields the lithium plating boundary voltage mapping table. For instance, taking a certain ternary lithium battery as an example, by consulting its corresponding lithium plating boundary voltage mapping table, at a temperature of 25°C and a SOC of 80%, the corresponding lithium plating boundary voltage is found to be 4.15V.

[0090] Next, the preset voltage safety margin value is read. This voltage safety margin is a pre-set constant, set based on statistical analysis of the amplitude range of model estimation errors and measurement noise under various operating conditions, and comprehensively considering the uncertainties caused by battery aging, to cover deviations in most cases. For example, analysis of a large amount of experimental data revealed that the polarization voltage estimation error is typically within ±0.02V, the peak-to-peak value of sensor noise does not exceed 0.01V, and the threshold drift caused by aging is within 0.01V. Therefore, the voltage safety margin can be set to 0.05V to ensure sufficient safety buffer under various uncertainties and avoid misjudgments due to instantaneous errors.

[0091] Finally, the critical voltage threshold obtained by subtracting the safety margin value from the lithium plating boundary voltage value obtained from the query is the critical voltage threshold used for subsequent comparison and risk assessment.

[0092] In step S104, if the current polarization voltage is greater than or equal to the critical voltage threshold, the changing trend of the polarization voltage is predicted using a preset neural network model, and the changing trend is risk-mapped to obtain a risk assessment result.

[0093] Among these, the prediction of polarization voltage variation trends using a pre-defined neural network model includes:

[0094] Historical polarization voltage time series data are extracted and standardized to obtain an input feature vector containing voltage difference information;

[0095] The input feature vector is fed into a pre-trained long short-term memory network model for time-series extrapolation to obtain the polarization voltage prediction value within the future time window;

[0096] Calculate the first-order difference of the predicted polarization voltage to determine the trend of polarization voltage variation, which characterizes the direction of voltage evolution.

[0097] First, historical polarization voltage time-series data within the most recent time window (e.g., the past 30 seconds) is extracted. Next, the extracted historical polarization voltage time-series data is standardized to obtain an input feature vector containing voltage difference information. This standardization is achieved by subtracting the sequence mean and dividing by the sequence standard deviation. The construction of the input feature vector involves using the standardized time-series data itself, along with the calculated voltage differences (first-order differences) between adjacent time points, as features.

[0098] Subsequently, the constructed input feature vector is fed into a pre-trained Long Short-Term Memory (LSTM) network model for time-series extrapolation, obtaining a sequence of polarization voltage predictions for the next time window (e.g., the next 10 seconds). It should be noted that the LSTM model is pre-trained, and its training process is as follows: During the actual operation of the charging station, a large amount of historical charging data is collected, including time-series evolution data of polarization voltage under different batteries, temperatures, and states of charge, forming a training dataset. A network structure is constructed containing one LSTM layer (128 hidden units), one fully connected layer, and one output layer. The model input is the polarization voltage feature vector of the past 30 seconds, and the output is the polarization voltage prediction for the next 10 seconds. The loss function uses mean squared error (MSE). Training is performed using the Adam optimizer with an initial learning rate of 0.001, employing a learning rate decay strategy during training. The model is trained on the training set for 300 epochs, and its performance is evaluated on an independent validation set. Training stops when the validation set loss no longer decreases significantly, and the model weights are preserved. Hyperparameters such as the number of hidden units and the learning rate are optimized and determined on the validation set through grid search.

[0099] Then, the first difference (i.e. the difference between adjacent predicted values) of the polarization voltage prediction sequence is calculated to determine the polarization voltage change trend value that characterizes the direction of voltage evolution.

[0100] Finally, based on the polarization voltage change trend value, a preset risk level mapping table is retrieved, and the potential risk assessment result for the current battery state is output. The risk level mapping table defines the risk level corresponding to different trend ranges; for example, a trend value less than 0.01 V / s is low risk, 0.01 to 0.03 V / s is medium risk, and greater than 0.03 V / s is high risk. This mapping relationship is established based on the statistical distribution of the polarization voltage rise rate and the probability of eventual risk through retrospective analysis of massive historical safe charging data and charging data showing lithium plating risks. For example, statistics show that when the polarization voltage rise rate is consistently below 0.01 V / s, the probability of risk occurring within the next minute is less than 0.1%; when the rate is between 0.01 and 0.03 V / s, the risk probability rises to 1% to 5%; and when the rate is above 0.03 V / s, the risk probability rises sharply to over 10%. Therefore, the risk level mapping relationship is set based on this statistical probability result.

[0101] It is worth noting that if the current polarization voltage is less than the critical voltage threshold, the current charging control is determined to be in a safe state. At this time, the process will backtrack to step S101 and continue to monitor the real-time battery status.

[0102] In step S105, based on the risk assessment results, the corresponding polarization voltage change trend value is extracted. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, the current charging control strategy is adjusted, and the maximum allowable current is calculated.

[0103] This includes adjusting the current charging control strategy and calculating the maximum allowable current, including:

[0104] The target current decay is calculated based on the real-time charging current value and the preset current lower limit.

[0105] The corrected maximum allowable current is obtained by subtracting the target current decay from the real-time charging current value.

[0106] First, the risk assessment result output in step S104 is received, and the key polarization voltage change trend values ​​are analyzed and extracted.

[0107] Next, it should be noted that the preset trend tolerance threshold represents the upper limit of the acceptable rate of risk escalation for the system. Its value can be set based on battery safety specifications and statistical analysis of extensive experimental data. Specifically, the setting is based on: combining the maximum safe charging rate recommendations provided by battery manufacturers, and through statistical analysis of the natural fluctuations and growth rates of polarization voltage under numerous safe charging scenarios, selecting a critical value that covers most safe operating conditions and can provide timely warnings of abnormally rapid growth. For example, analysis shows that in 95% of safe charging processes, the polarization voltage rise rate is below 0.015 V / s; therefore, 0.015 V / s is set as the preset trend tolerance threshold.

[0108] Next, the extracted trend value is compared with the trend tolerance threshold. If the trend value does not exceed the trend tolerance threshold, the current charging strategy is maintained and monitoring continues. If the trend value exceeds the trend tolerance threshold, it indicates that the risk is accumulating rapidly, and immediate intervention measures must be taken to reduce the risk, triggering an adjustment to the current charging control strategy.

[0109] Finally, after the adjustment is triggered, the target current decay is calculated based on the real-time charging current value and the preset lower current limit. The lower current limit is a preset safe current value used to ensure basic charging needs. It is set based on the battery's chemical characteristics and thermal management limits, determining an absolutely safe current value that will not cause thermal runaway or severe lithium plating even under extreme conditions. For example, for a certain type of battery, experiments using constant current charging to the lithium plating boundary showed that no significant lithium plating was observed at currents below 0.5C (equivalent to 30A for a 60Ah battery capacity). Therefore, the lower current limit can be preset to the current value corresponding to the 0.5C rate of this battery (30A). The calculation rule for the target current decay is: set proportionally according to the degree of exceeding the threshold. For example, let's define an attenuation coefficient k, which is calculated by subtracting the trend tolerance threshold from the current trend value and then dividing by the difference between the high-risk threshold and the trend tolerance threshold (here, the high-risk threshold is the threshold used to determine the high-risk level in step S104 for risk level mapping). Then, the target current attenuation is calculated by subtracting the current lower limit from the real-time charging current and then multiplying by the attenuation coefficient k. For example, if the real-time current is 80A, the current lower limit is 30A, the current trend is 0.02 V / s, the tolerance threshold is 0.015 V / s, and the high-risk threshold is 0.03 V / s, then k = (0.02-0.015) / (0.03-0.015) ≈ 0.33, and the target current attenuation is 0.33 × (80-30) ≈ 16.5A.

[0110] The target current attenuation is subtracted from the real-time charging current value to obtain the corrected maximum allowable current (80A - 16.5A = 63.5A), and this value is written into the current limit register.

[0111] In step S106, the control parameters of the charging pile are updated according to the maximum allowable current, a new charging control point is calculated and connected to obtain an optimized charging curve, including:

[0112] The maximum allowable current is encapsulated into a control command frame and sent to the charging pile to adjust the output voltage and current, thereby generating actual output current data.

[0113] The optimal voltage reference value is calculated based on the actual output current data, the current state of charge of the battery, and the temperature.

[0114] A charging control point is constructed based on the optimal voltage reference value and the actual output current data. Multiple charging control points are connected to form an optimized charging curve.

[0115] First, the corrected maximum allowable current value stored in the current limit register is obtained, and this value is encapsulated into a control command frame that conforms to the charging pile communication protocol (such as CAN bus protocol or custom TCP / UDP protocol) and sent to the main control unit of the charging pile.

[0116] Then, the main control unit of the charging pile parses the control command frame, extracts the maximum allowable current value, and adjusts the output voltage and current through its internal power electronic module (such as a DC-DC converter) to make the actual output current approach but not exceed the maximum allowable current, thereby generating actual output current data.

[0117] Next, the actual output current data is transmitted back to the battery management system or cloud control platform. Based on this data, and combined with the battery's current state of charge and temperature data, the system calculates the optimal voltage reference value. Specifically, the formula for calculating the optimal voltage reference value is:

[0118]

[0119] in, The calculated optimal voltage reference value, To determine based on the state of charge and temperature The static open-circuit voltage is obtained by looking up a table. The specific method for obtaining this voltage has already been given in step S102 and will not be explained again here. This is the actual output current data. The internal resistance is the ohmic resistance obtained from the second-order RC equivalent circuit model in step S102.

[0120] Subsequently, based on the optimal voltage reference value and the actual output current data, a new charging control point (i.e., a "current-voltage" pair) is constructed. As the charging process progresses, a series of such charging control points are dynamically generated according to the above steps, and they are connected in time sequence to form a new, optimized charging curve to guide the subsequent charging process.

[0121] S107, Based on the optimized charging curve, monitor the battery response to obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence. Compare the fluctuation variance with a preset stability threshold, and determine the final charging strategy control and adjustment based on the comparison result, including:

[0122] The real-time terminal voltage value of each battery cell and the corresponding estimated open-circuit voltage data are collected based on the optimized charging curve.

[0123] The polarization voltage sequence is obtained by subtracting the open-circuit voltage estimation data from the real-time terminal voltage value, and the fluctuation variance of the polarization voltage sequence is calculated.

[0124] If the fluctuation variance is less than a preset stability threshold, then maintain the control command for the current state;

[0125] If the fluctuation variance is not less than the stability threshold, a preset conservative charging command is executed. Then, the process reverts to the step of collecting real-time data of the electric vehicle battery through the charging pile sensor, smoothing the real-time data to obtain a smooth battery state sequence, and performs a new round of charging strategy adjustment.

[0126] First, charging is performed according to the optimized charging curve generated in step S106, and the battery response is continuously monitored during this process. Specifically, the real-time terminal voltage value of each battery cell is collected, and based on the current state of charge and temperature data, the corresponding open-circuit voltage estimation data is obtained through the table lookup method in step S102.

[0127] Then, the polarization voltage sequence during the monitoring period is obtained by subtracting the open-circuit voltage estimation data from the real-time terminal voltage value.

[0128] Next, the fluctuation variance of the polarization voltage sequence is calculated to quantify the stability of the polarization state.

[0129] Subsequently, the calculated fluctuation variance is compared with a preset stability threshold. This stability threshold is an empirical value (e.g., 1×10⁻⁶) set by analyzing the fluctuation characteristics of polarization voltage under a large number of safe and stable charging conditions. -4 V² is used to determine whether the battery's response to the current charging strategy is stable. The threshold is set based on the variance distribution of polarization voltage fluctuations under various battery conditions and health states when using a safe charging strategy. The high percentile of the distribution (such as the 95th percentile) is taken as the stability threshold to ensure that the fluctuations under most safe operating conditions are less than this value.

[0130] Finally, based on the comparison results, the final charging strategy control and adjustment are determined: if the fluctuation variance is less than the preset stability threshold, a control command to maintain the current state is generated, indicating that the safety of the current charging process is guaranteed, and charging continues according to the optimized charging curve; if the fluctuation variance is not less than the stability threshold, it indicates that the battery response is unstable and the polarization state fluctuates greatly. At this time, the system will execute a preset conservative charging command (for example, immediately reducing the charging current to an extremely low safety value, such as 5A corresponding to 0.1C) to ensure absolute safety. Then, the control flow is traced back to step S101, and a new round of data acquisition, state estimation, and strategy calculation is restarted, thereby realizing closed-loop adaptive optimization.

[0131] In summary, this invention discloses an intelligent control method for charging piles based on the Internet of Things. Through data acquisition and smoothing, accurate calculation of polarization voltage, dynamic threshold determination, neural network risk trend prediction, dynamic adjustment of charging strategy based on prediction, optimized generation of charging curves, and closed-loop monitoring and verification, it achieves real-time, accurate, and forward-looking prevention and control of lithium plating risk during the fast charging process of electric vehicles. Under the premise of ensuring the core safety of the battery, it effectively improves charging efficiency and extends battery cycle life.

[0132] Reference Figure 2 The second embodiment of the present invention provides an intelligent control system for charging piles based on the Internet of Things, comprising:

[0133] The data acquisition and processing module is used to acquire real-time data of the electric vehicle battery through the charging pile sensor, and to smooth the real-time data to obtain a smooth battery state sequence.

[0134] The polarization voltage calculation module is used to calculate the current polarization voltage based on the battery state sequence using a second-order RC equivalent circuit model.

[0135] The critical voltage determination module is used to obtain the lithium plating boundary voltage based on the temperature and the state of charge, and then subtract it from a preset voltage safety margin to obtain the critical voltage threshold.

[0136] The risk prediction module is used to predict the trend of polarization voltage change using a preset neural network model if the current polarization voltage is greater than or equal to the critical voltage threshold, and to perform risk mapping on the trend of change to obtain a risk assessment result.

[0137] The strategy adjustment module is used to extract the corresponding polarization voltage change trend value based on the risk assessment result. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, the current charging control strategy is adjusted and the maximum allowable current is calculated.

[0138] The control update module is used to update the control parameters of the charging pile according to the maximum allowable current, calculate and connect the new charging control point, and obtain the optimized charging curve.

[0139] The monitoring and adjustment module is used to monitor the battery response based on the optimized charging curve, obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence, compare the fluctuation variance with a preset stability threshold, and determine the final charging strategy control and adjustment based on the comparison result.

[0140] It should be noted that the IoT-based intelligent control system for charging piles provided in this embodiment of the invention is used to execute all the process steps of the IoT-based intelligent control method for charging piles in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0141] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a smart control program for charging piles. When the processor executes the computer program, it implements the steps described in the various method embodiments above, for example... Figure 1 The step S101 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments, such as the data acquisition and processing module.

[0142] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0143] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0144] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0145] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0146] If the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0147] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0148] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A smart control method for charging piles based on the Internet of Things, characterized in that, include: Real-time data of electric vehicle batteries is collected by sensors in charging piles, and the real-time data is smoothed to obtain a smooth battery state sequence; wherein the real-time data includes battery terminal voltage, loop current, state of charge and temperature. Based on the battery state sequence, the current polarization voltage is calculated using a second-order RC equivalent circuit model; The lithium plating boundary voltage is obtained based on the temperature and the state of charge, and the difference is calculated with a preset voltage safety margin to obtain the critical voltage threshold. If the current polarization voltage is greater than or equal to the critical voltage threshold, the trend of polarization voltage change is predicted using a preset neural network model, and the trend of change is risk-mapped to obtain a risk assessment result. Based on the risk assessment results, the corresponding polarization voltage change trend value is extracted. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, the current charging control strategy is adjusted and the maximum allowable current is calculated. The control parameters of the charging pile are updated according to the maximum allowable current, a new charging control point is calculated and connected to obtain the optimized charging curve. Charge according to the optimized charging curve, monitor the battery response based on the optimized charging curve, obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence, and compare the fluctuation variance with a preset stability threshold. If the fluctuation variance is less than the preset stability threshold, the control command for maintaining the current state is maintained; if the fluctuation variance is not less than the stability threshold, the preset conservative charging command is executed, and then the process is traced back to the step of collecting real-time data of the electric vehicle battery through the charging pile sensor, smoothing the real-time data to obtain a smooth battery state sequence, and a new round of charging strategy adjustment is carried out.

2. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The process of collecting real-time data from the electric vehicle battery via charging pile sensors and smoothing the real-time data to obtain a smoothed battery state sequence includes: Real-time data including battery terminal voltage, loop current, state of charge, and temperature are collected by the charging pile's sensors. The real-time data is smoothed using a Kalman filter algorithm to obtain a smoothed battery state sequence.

3. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The step of calculating the current polarization voltage using a second-order RC equivalent circuit model based on the battery state sequence includes: The parameters of the second-order RC equivalent circuit model are obtained by performing parameter identification on the battery state sequence. The static open-circuit voltage value and ohmic voltage drop component are determined based on the battery state sequence. The total polarization voltage residual vector is obtained by subtracting the static open-circuit voltage value and the ohmic voltage drop component from the battery terminal voltage observation value collected by the charging pile sensor. Based on the model parameters, a state-space equation is constructed, and the total polarization voltage residual vector is substituted into the state-space equation to analyze the electrochemical polarization voltage component and the concentration polarization voltage component. The current polarization voltage is determined by linearly superimposing the electrochemical polarization voltage component and the concentration polarization voltage component.

4. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The step of obtaining the lithium plating boundary voltage based on the temperature and the state of charge, and subtracting it from a preset voltage safety margin to obtain the critical voltage threshold includes: The lithium plating boundary voltage is obtained by retrieving a preset lithium plating boundary voltage mapping table based on the temperature and the state of charge. The critical voltage threshold is obtained by subtracting a preset voltage safety margin from the lithium plating boundary voltage.

5. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The method of predicting the trend of polarization voltage using a preset neural network model includes: Historical polarization voltage time series data are extracted and standardized to obtain an input feature vector containing voltage difference information; The input feature vector is fed into a pre-trained long short-term memory network model for time-series extrapolation to obtain the polarization voltage prediction value within the future time window; Calculate the first-order difference of the predicted polarization voltage to determine the trend of polarization voltage variation that characterizes the direction of voltage evolution.

6. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The adjustment of the current charging control strategy and the calculation of the maximum allowable current include: The target current decay is calculated based on the real-time charging current value and the preset current lower limit. The corrected maximum allowable current is obtained by subtracting the target current decay from the real-time charging current value.

7. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The process of updating the control parameters of the charging pile based on the maximum allowable current, calculating and connecting a new charging control point, and obtaining an optimized charging curve includes: The maximum allowable current is encapsulated into a control command frame and sent to the charging pile to adjust the output voltage and current, thereby generating actual output current data. The optimal voltage reference value is calculated based on the actual output current data, the current state of charge of the battery, and the temperature. A charging control point is constructed based on the optimal voltage reference value and the actual output current data. Multiple charging control points are connected to form an optimized charging curve.

8. The intelligent control method for charging piles based on the Internet of Things according to claim 1, characterized in that, The process of monitoring battery response based on the optimized charging curve, obtaining a polarization voltage sequence, and calculating the variance of the fluctuation of the polarization voltage sequence includes: The real-time terminal voltage value of each battery cell and the corresponding estimated open-circuit voltage data are collected based on the optimized charging curve. The polarization voltage sequence is obtained by subtracting the open-circuit voltage estimation data from the real-time terminal voltage value, and the fluctuation variance of the polarization voltage sequence is calculated.

9. An intelligent control system for charging piles based on the Internet of Things, characterized in that, include: The data acquisition and processing module is used to acquire real-time data of the electric vehicle battery through the charging pile sensor, and to smooth the real-time data to obtain a smooth battery state sequence; wherein the real-time data includes battery terminal voltage, loop current, state of charge and temperature. The polarization voltage calculation module is used to calculate the current polarization voltage based on the battery state sequence using a second-order RC equivalent circuit model. The critical voltage determination module is used to obtain the lithium plating boundary voltage based on the temperature and the state of charge, and then subtract it from a preset voltage safety margin to obtain the critical voltage threshold. The risk prediction module is used to predict the trend of polarization voltage change using a preset neural network model if the current polarization voltage is greater than or equal to the critical voltage threshold, and to perform risk mapping on the trend of change to obtain a risk assessment result. The strategy adjustment module is used to extract the corresponding polarization voltage change trend value based on the risk assessment result. If the polarization voltage change trend value exceeds the preset trend tolerance threshold, the current charging control strategy is adjusted and the maximum allowable current is calculated. The control update module is used to update the control parameters of the charging pile according to the maximum allowable current, calculate and connect the new charging control point, and obtain the optimized charging curve. The monitoring and adjustment module is used to charge according to the optimized charging curve, monitor the battery response based on the optimized charging curve, obtain the polarization voltage sequence and calculate the fluctuation variance of the polarization voltage sequence, and compare the fluctuation variance with a preset stability threshold. If the fluctuation variance is less than the preset stability threshold, the control command for maintaining the current state is maintained; if the fluctuation variance is not less than the stability threshold, the preset conservative charging command is executed, and then the process is traced back to the step of collecting real-time data of the electric vehicle battery through the charging pile sensor, smoothing the real-time data to obtain a smooth battery state sequence, and a new round of charging strategy adjustment is carried out.

Images