Insulation resistance intelligent detection system and method for energy storage converter

Abstract

The invention belongs to the technical field of electrical safety, and particularly discloses an insulation resistance intelligent detection system and method for an energy storage converter, and the system is based on a bridge balance structure, integrates a temperature and humidity sensor and a voltage sensor, and collects environment parameters and bridge unbalance signals in real time. A digital potentiometer and a resistor array are driven through a PID algorithm to dynamically adjust bridge balance, and environmental interference is eliminated in combination with a temperature and humidity compensation model; and carrying out intelligent analysis on the preprocessed multi-dimensional data by utilizing a band-pass filtering and detection model, and outputting a high-precision insulation resistance value. The method achieves full-automatic detection, is high in precision, is quick in response, is suitable for complex working conditions, and remarkably improves the reliability of the insulation performance evaluation of the energy storage converter.

Description

Technical Field

[0001] The present invention belongs to the technical field of electrical safety, and particularly relates to an intelligent insulation impedance detection system and method for an energy storage converter. Background Art

[0002] In the field of safe operation of electrical equipment, insulation impedance detection is a key link in evaluating the insulation performance of an energy storage converter. Traditional detection methods mainly use the bridge method, which judges insulation defects by measuring the resistance value of insulating materials. However, this method has significant limitations: Environmental interference problem: Environmental factors such as temperature, humidity, and electromagnetic interference are likely to cause the destruction of the bridge balance and a decrease in measurement accuracy. For example, temperature changes will change the resistivity of insulating materials, and an increase in humidity may introduce surface leakage current. It is difficult for the traditional bridge method to compensate for these variables in real time, resulting in an error of up to ±10%.

[0003] Insufficient signal processing ability: Traditional methods lack efficient means for extracting features from the noisy bridge output signal and are difficult to distinguish effective signals from interference. Especially in a complex electromagnetic environment, the reliability of the detection results is poor.

[0004] With the development of energy storage converters towards high power, modularization, and non-isolation, the above problems have become increasingly prominent. Although artificial intelligence technology has shown advantages in the field of signal processing, the existing technology has not yet achieved the deep integration of the bridge method and intelligent algorithms, and cannot meet the detection requirements of high precision and real-time. Summary of the Invention

[0005] The purpose of the present invention is to overcome the defects existing in the prior art and provide an intelligent insulation impedance detection system and method for an energy storage converter.

[0006] In the first aspect of the present invention, an intelligent insulation impedance detection system for an energy storage converter is provided, including: Bridge balance structure: Adopting the Wheatstone bridge topology, it consists of a quadrilateral loop formed by four bridge arms, where two diagonals are respectively connected to a DC excitation power supply and a microvolt-level voltage detection port; at least one variable resistance arm and one measurement arm are included in the four bridge arms; Data acquisition module: Used to collect the ambient temperature, humidity, and the bridge output voltage in real time; Control module: Connected to the data acquisition module, calculates the bridge balance deviation based on the ambient temperature, humidity, and the bridge output voltage, and outputs an adjustment instruction; Automatic adjustment circuit: Respectively connected to the control module and the variable resistance arm, responds to the adjustment instruction, and controls the variable resistance arm to adjust the resistance value to achieve dynamic balance of the bridge; The insulation impedance detection module, connected to the data acquisition module, outputs an insulation impedance detection value through a detection model based on the output voltage of the bridge.

[0007] A further solution is that the four bridge arms include: The first standard resistance arm and the second standard resistance arm: both are high-precision metal film resistors, connected in series between the positive pole of the excitation power supply and the common node to form a fixed reference bridge arm; The variable resistance arm: uses a digital potentiometer, connected in series between the common node and the negative pole of the excitation power supply; the digital potentiometer is also connected to the control module through an I²C or SPI communication interface; The measurement arm: connects to the measured circuit of the energy storage converter, is on the same side bridge arm as the variable resistance arm, and is used to connect the insulation impedance to be measured.

[0008] A further solution is that the automatic adjustment circuit includes: A resistor array connected in parallel across the variable resistance arm, and the resistance values of the resistors in the resistor array are distributed according to an exponential or linear law, covering the target adjustment range; Each resistor in the resistor array is in series with a MOSFET switch.

[0009] A further solution is that the data acquisition module includes: A temperature sensor: used to collect the ambient temperature of the bridge balance structure in real time; A humidity sensor: used to collect the ambient humidity in real time; A voltage sensor: a microvolt-level sensor based on a high-impedance differential amplifier, connected to the voltage detection port between the two midpoints of the bridge, and used to measure the bridge unbalance voltage ΔU.

[0010] A further solution is that the control module includes a data acquisition layer, an algorithm processing layer, and a control output layer; The data acquisition layer is respectively connected to the temperature sensor, the humidity sensor, and the voltage sensor, and is used to obtain the ambient temperature, humidity, and the bridge unbalance voltage ΔU; The algorithm processing layer uses the bridge unbalance voltage ΔU as an input variable, compensates ΔU through the ambient temperature and humidity, and calculates the bridge balance adjustment amount through a PDI control algorithm; The control output layer generates an adjustment instruction for the digital potentiometer based on the balance adjustment amount calculated by the PID control algorithm, sends the adjustment instruction to the digital potentiometer through an SPI or I²C communication interface, and adjusts the variable resistance arm through the digital potentiometer to make the bridge output voltage approach zero.

[0011] A further solution is that the algorithm processing layer is also connected to a BMS timing control module, which activates the detection circuit of each module in a time-sharing manner to avoid the parallel effect caused by the simultaneous connection of multiple modules to detect internal resistance.

[0012] A further solution is that the insulation impedance detection module includes a signal preprocessing unit and a detection model; The signal preprocessing unit filters out the low-frequency drift and high-frequency noise of the bridge unbalanced voltage ΔU through a bandpass filter, retains the effective frequency band of the polarization process of the insulating material, performs normalization processing, and converts it into dimensionless data between 0 and 1; The construction process of the detection model is as follows: The insulation resistance detection values ​​of different working conditions in multiple time series are collected by a high-precision insulation resistance meter, and the ambient temperature, humidity and bridge unbalanced voltage ΔU are collected synchronously. The ambient temperature, humidity and bridge unbalanced voltage ΔU data are organized into a three-dimensional tensor format (T, H, ΔU), where T is the ambient temperature and H is the humidity. A large number of insulation resistance detection values ​​and (T, H, ΔU) matching the insulation resistance detection values ​​are obtained for manual expert labeling. After labeling, (T, H, ΔU) is used as the input of the convolutional neural network, and the insulation resistance detection values ​​are used as the output of the convolutional neural network for iterative training, so as to obtain a detection model based on the insulation resistance detection value output by (T, H, ΔU).

[0013] A second aspect of the present invention provides an intelligent detection method for insulation impedance of an energy storage converter, using the above system, comprising the following steps: System initialization: Start the bridge balance structure, adjust the variable resistor arm through the digital potentiometer to make the bridge unbalanced voltage ΔU ≤ 10μV, and record the initial balance parameters; Real-time data acquisition: The data acquisition module synchronously collects ΔU, T, and H to generate a raw signal sequence with a time stamp; Dynamic balance adjustment: The control module calculates the adjustment amount through the PID algorithm according to ΔU, T, and H, and drives the automatic adjustment circuit to switch the digital potentiometer or parallel resistor to make ΔU return to balance; Signal preprocessing: bandpass filter and normalize ΔU, and fuse it with T and H into three-dimensional time series data (T, H, ΔU); Insulation impedance detection: Input the preprocessed three-dimensional data (T, H, ΔU) into the detection model and output the real-time insulation impedance detection value.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention realizes high-precision, high-reliability and full-automatic insulation impedance detection of energy storage converters through the deep integration of bridge dynamic balance control, environmental parameter compensation, intelligent signal processing and multi-module collaborative technology, providing a strong technical guarantee for the safe operation of electrical equipment.

[0015] The present invention collects ambient temperature and humidity in real time and combines the PID control algorithm to dynamically compensate for the unbalanced voltage of the bridge, significantly reducing the measurement error caused by temperature and humidity changes and improving the accuracy of insulation impedance detection. A bandpass filter is used to pre-process the output signal of the bridge, effectively filtering out low-frequency drift and high-frequency noise, retaining the polarization characteristic frequency band of the insulating material, and combining the convolutional neural network model to perform intelligent analysis of multidimensional data (T, H, ΔU), solving the problem of signal distortion in complex electromagnetic environments.

[0016] The automatic adjustment circuit of the present invention is based on a digital potentiometer and a high-speed MOSFET resistor array, has a response time of less than , and cooperates with the PID algorithm to realize fast zeroing of the bridge (ΔU≤10μV), ensuring the stability of the system under dynamic conditions and avoiding the measurement deviation caused by manual adjustment lag in the traditional bridge method.

[0017] The present invention activates each detection circuit in time-sharing manner through the BMS timing control module, avoids mutual interference of internal resistance when multiple modules are connected in parallel, solves the problem of decreased accuracy of multi-channel detection in traditional methods, and is suitable for complex application scenarios of high-power, modular energy storage inverters.

[0018] The present invention utilizes a pre-trained CNN detection model to directly output the insulation resistance value without relying on manual experience judgment, which significantly improves the detection efficiency. At the same time, the system runs automatically throughout the process after initialization, greatly reducing manual intervention and lowering operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The following drawings are only used to illustrate and explain the present invention, and are not used to limit the scope of the present invention, wherein: Figure 1 This is the connection diagram of the insulation impedance intelligent detection system; In the figure: 1. Bridge balance structure; 2. Data acquisition module; 3. Control module; 4. Automatic adjustment circuit; 5. Insulation impedance detection module; 6. BMS timing control module; 7. Standard resistance arm; 8. Variable resistance arm; 9. Measurement arm; 10. Temperature sensor; 11. Humidity sensor; 12. Voltage sensor; 13. Data acquisition layer; 14. Algorithm processing layer; 15. Control output layer; 16. Detection model; 17. Signal preprocessing unit; 18. Resistance array; 19. MOSFET switch. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions, design methods, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0021] Embodiment 1 As Figure 1 shown, this embodiment provides an intelligent insulation impedance detection system for an energy storage converter, including: Bridge balance structure 1: Adopting a Wheatstone bridge topology, it consists of four bridge arms forming a quadrilateral loop, with two diagonals respectively connected to a DC excitation power supply and a microvolt-level voltage detection port; at least one variable resistance arm 8 and one measurement arm 9 are included in the four bridge arms; among them, the four bridge arms include: a first standard resistance arm 7 and a second standard resistance arm 7: both are high-precision metal film resistors, connected in series between the positive pole of the excitation power supply and the common node to form a fixed reference bridge arm; variable resistance arm 8: Using a digital potentiometer, connected in series between the common node and the negative pole of the excitation power supply; the digital potentiometer is also connected to the control module 3 through an I²C or SPI communication interface; measurement arm 9: Connected to the measured circuit of the energy storage converter, on the same side bridge arm as the variable resistance arm 8, for accessing the insulation impedance to be measured; Data acquisition module 2: Used to collect the ambient temperature, humidity, and the output voltage of the bridge in real time; in this embodiment, the data acquisition module 2 includes: temperature sensor 10: A temperature sensor for collecting the ambient temperature of the bridge balance structure 1 in real time; humidity sensor 11: For collecting the ambient humidity in real time; voltage sensor 12: A microvolt-level sensor based on a high-impedance differential amplifier, connected to the voltage detection port between the two midpoints of the bridge, for measuring the bridge unbalance voltage ΔU; Control module 3: Connected to the data acquisition module 2, calculating the bridge balance deviation based on the ambient temperature, humidity, and the output voltage of the bridge, and outputting an adjustment instruction; the control module 3 includes a data acquisition layer 13, an algorithm processing layer 14, and a control output layer 15; the data acquisition layer 13 is respectively connected to the temperature sensor 10, humidity sensor 11, and voltage sensor 12 for obtaining the ambient temperature, humidity, and the bridge unbalance voltage ΔU; the algorithm processing layer 14 uses the bridge unbalance voltage ΔU as an input variable, compensates ΔU through the ambient temperature and humidity, and calculates the bridge balance adjustment amount through a PDI control algorithm; the control output layer 15 generates an adjustment instruction for the digital potentiometer based on the balance adjustment amount calculated by the PID control algorithm, sends the adjustment instruction to the digital potentiometer through an SPI or I²C communication interface, and adjusts the variable resistance arm 8 through the digital potentiometer to make the bridge output voltage approach zero; among them, the process of compensating ΔU through the ambient temperature and humidity is: Compensate the unbalanced voltage ΔU of the bridge based on the temperature characteristics of the resistivity of the insulating material:

[0022] Among them, α is the temperature coefficient of the insulating material to be measured (obtained from the equipment manual, e.g., For epoxy resin, α = -0.02 / °C), and T0 is the initial calibration temperature (e.g., 25°C); Determine the compensation of humidity for the unbalanced voltage ΔU of the bridge by combining the parallel model of surface leakage resistance: Implement humidity detection H. When H > 60%RH, trigger the humidity compensation mechanism; According to the formula Calculate , where k is a reference parameter representing the initial magnitude or reference value of the surface leakage resistance under specific conditions, determining the basic order of magnitude of the leakage resistance, and β is the humidity influence coefficient, reflecting the influence rate and degree of the change in humidity H on the surface leakage resistance . The larger β is, when the humidity H increases, The faster the exponential decay factor decreases with the change of humidity. Both of them are empirical parameters obtained by fitting historical data, used to quantify the influence of humidity on the surface leakage resistance in the insulation impedance, so as to more accurately correct the effect of humidity on the insulation impedance. In this embodiment, k = 100 MΩ, β = 0.05 / RH, indicating that for every 1%RH increase in humidity, The exponential decay factor of will change accordingly, reflecting the dynamic influence of humidity on the leakage resistance; Use the parallel model Calculate the equivalent insulation impedance under the influence of humidity ; The unbalanced voltage of the bridge is related to the impedance of the bridge arm. Use Replace the original Recalculate ΔU, and compare the measured ΔU to correct ΔU through the formula , cancel the interference of the decrease in caused by humidity to ΔU, realize the dynamic compensation of humidity for ΔU, and can more accurately reflect the true state of the insulation impedance in a high-humidity environment, improving the reliability of bridge detection; Automatic adjustment circuit 4: Connect to the control module 3 and the variable resistance arm 8 respectively, respond to the adjustment instruction, and control the variable resistance arm 8 to adjust the resistance value to achieve the dynamic balance of the bridge; The automatic adjustment circuit 4 includes: a resistor array 18 connected in parallel across the variable resistance arm 8, and the resistance values of the resistors in the resistor array 18 are distributed according to an exponential or linear law, covering the target adjustment range; Each resistor in the resistor array 18 is connected in series with a MOSFET switch 19; The insulation impedance detection module 5 is connected to the data acquisition module 2 and outputs an insulation impedance detection value through a detection model 16 based on the bridge output voltage.

[0023] In order to solve the problem of decreased accuracy of multi-channel detection in traditional methods, in this embodiment, the algorithm processing layer 14 is also connected to the BMS timing control module 6, which activates the detection circuit of each module in time-sharing to avoid the parallel effect caused by simultaneous connection of multiple modules to detect internal resistance.

[0024] In the above, the insulation impedance detection module 5 includes a signal preprocessing unit 17 and a detection model 16; The signal preprocessing unit 17 filters out the low-frequency drift and high-frequency noise of the bridge unbalanced voltage ΔU through a bandpass filter, retains the effective frequency band of the polarization process of the insulating material and performs normalization processing to convert it into dimensionless data between 0 and 1; wherein, the construction process of the detection model 16 is: Collect the insulation impedance detection values at multiple time series under different working conditions through a high-precision insulation resistance meter, and synchronously collect the ambient temperature, humidity, and bridge unbalance voltage ΔU. Organize the ambient temperature, humidity, and bridge unbalance voltage ΔU data into a three-dimensional tensor format (T, H, ΔU), where T is the ambient temperature and H is the humidity. Obtain a large number of insulation impedance detection values and the corresponding (T, H, ΔU) for manual expert marking. After marking, use (T, H, ΔU) as the input of the convolutional neural network and the insulation impedance detection value as the output of the convolutional neural network for iterative training to obtain a detection model 16 that outputs the insulation impedance detection value based on (T, H, ΔU). In this embodiment, the detection model 16 uses a convolutional neural network (CNN), and its architecture design is as follows: Input layer: Receive the preprocessed multi-dimensional time series data. The multi-dimensional data (T, H, ΔU) has 3 feature channels, which are the normalized bridge unbalance voltage, real-time temperature, and real-time humidity respectively. Input the bridge signal fluctuations including time series and the temperature and humidity information at the corresponding moments into the network to provide the original data for subsequent feature extraction. The first convolutional layer sets 32 1D convolutional filters, and the kernel size of each filter is 5 (that is, the features of 5 consecutive time points are processed each time). The activation function uses ReLU to extract local time series features through convolutional operations, such as the voltage change rate within 5 ms, the temperature and humidity coupling trend, etc. The ReLU activation function introduces non-linearity, enabling the network to learn complex feature relationships. The max pooling layer has a pooling size of 2, which reduces the dimension of the feature map output by the convolutional layer, retains the maximum value features in the local area (such as the peak value of voltage fluctuations), reduces the data dimension and computational amount, and at the same time suppresses overfitting. The second convolutional layer sets 64 1D convolutional filters, with a kernel size of 3, and the activation function is still ReLU, further extracting more refined time series features, such as the peak value of voltage fluctuations within 3 ms, the temperature and humidity mutation points, etc. The number of filters is increased to 64, enabling the network to learn richer feature representations. The global average pooling layer performs average pooling on the entire feature map. Compress the spatial dimension (time series length) into a global feature vector to avoid overfitting problems caused by too many parameters in the fully connected layer, and at the same time retain the overall statistical information of the feature map. The fully connected layer contains 128 neurons, and the activation function is ReLU. Fuse the local features and global features extracted by the previous layers, and form an abstract representation related to the insulation impedance through non-linear transformation to provide high-level semantic features for the final output. The Dropout layer has a dropout rate set to 0.3. By randomly turning off some neurons, it suppresses network overfitting and improves the generalization ability of the model on unknown data. The output layer has 1 neuron, and the activation function is linear (Linear), which performs a linear transformation on the feature vector output by the fully connected layer and directly outputs the insulation impedance prediction value.

[0025] Embodiment 2 Based on Embodiment 1, this embodiment provides an intelligent insulation impedance detection method for an energy storage converter, including the following steps: System initialization: Start the bridge balance structure and connect a DC excitation power supply to the Wheatstone bridge. The control module sends an initial adjustment instruction to the digital potentiometer through the I²C or SPI communication interface, sets the resistance value of the variable resistance arm to the middle range (for example, if the total resistance value of the digital potentiometer is 100 kΩ, it is initially set to 50 kΩ); monitor the bridge unbalance voltage ΔU at the microvolt-level voltage detection port, and iteratively adjust the resistance value of the digital potentiometer to make ΔU ≤ 10 μV. Record the resistance value of the variable resistance arm at this time, the ambient temperature collected by the temperature sensor, and the ambient humidity collected by the humidity sensor, and store them as initial balance parameters in the register of the control module; Real-time data acquisition: The data acquisition module synchronously drives each sensor to work: The temperature sensor (DS18B20) collects the ambient temperature of the bridge module in real time, with an accuracy of ±0.1 °C; The humidity sensor (SHT30) collects the ambient humidity in real time, with an accuracy of ±2%RH; The voltage sensor based on a high-impedance differential amplifier (input impedance ≥ 10 MΩ) measures the unbalance voltage ΔU at the voltage detection port between the two midpoints of the bridge; Add a time stamp to each set of collected (T, H, ΔU) data to generate a continuous raw signal sequence, and store it in the data buffer for subsequent processing; Dynamic balance adjustment: The algorithm processing layer of the control module performs environmental compensation on the collected ΔU: When H > 60%RH, according to the formula Calculate the surface leakage resistance, and through Correct the influence of the insulation impedance on ΔU; Combine the temperature compensation formula Obtain the error signal e(t) after comprehensive compensation; Use the PID control algorithm to calculate the adjustment amount: Proportional link Quickly respond to the current error; Integral link I Eliminate the steady-state error; Differential link D Predict the change trend of the error; Comprehensively output the adjustment amount ; The control output layer converts the adjustment amount into a digital potentiometer adjustment command or the MOSFET drive signal of the automatic adjustment circuit: If the adjustment amount is to increase the resistance, the resistance value of the digital potentiometer increases at a resolution of 0.1%; If rapid adjustment is required, drive the MOSFET corresponding to the resistance value in the automatic adjustment circuit to conduct, and connect parallel resistors (such as 10 kΩ, 100 kΩ) to the variable resistance arm to make ΔU return to balance within 10 ms (ΔU ≤ 10 μV).

[0026] Signal preprocessing: Perform band-pass filtering and normalization on ΔU, and fuse it with T and H into three-dimensional time-series data (T, H, ΔU); Insulation impedance detection: Input the preprocessed three-dimensional data (T, H, ΔU) into the detection model. The input layer receives the three-dimensional time-series data; The first convolutional layer (32 filters, kernel size 5) extracts local features within 5 ms (such as voltage change rate, temperature and humidity coupling trend); The max pooling layer (pooling size 2) reduces the dimension and retains the peak features; The second convolutional layer (64 filters, kernel size 3) extracts finer features within 3 ms (such as voltage fluctuation peak); The global average pooling layer compresses the feature map into a global vector; The fully connected layer (128 neurons) fuses high-level features; The Dropout layer (dropout rate 0.3) suppresses overfitting; The output layer (1 neuron, linear activation) outputs the real-time insulation impedance detection value.

[0027] Among them, the process of adjusting the variable resistance arm is as follows: Step 1: Receive the adjustment command: The PID algorithm of the control module calculates the required resistance adjustment amount according to the bridge unbalance voltage ΔU and the error signal after environmental compensation; Step 2: Select the adjustment mode: Fine adjustment: Adjust the resistance value through the digital potentiometer, which is suitable for correcting small deviations; Rapid adjustment: When the ΔU deviation is large, enable the resistance array: The control module selects the resistor combination with the closest resistance value in the resistance array according to the magnitude of ΔR, and connects the selected resistor in parallel to the variable resistance arm by turning on the corresponding MOSFET switch to quickly change the total resistance value; Step 3: Execute the resistance value adjustment and MOSFET switch control: If it is necessary to increase the total resistance value, disconnect some parallel resistors (turn off the corresponding MOSFET switch); If it is necessary to decrease the total resistance value, turn on more parallel resistors (turn on the corresponding MOSFET switch); Digital potentiometer adjustment: Send commands through the I²C / SPI interface to adjust its resistance value to the target value.

[0028] Step 4: After adjustment, the voltage sensor re-detects ΔU and feeds it back to the control module. If ΔU still exceeds the threshold (such as > 10 μV), repeat steps 1-3 until the bridge returns to balance.

[0029] Through the above steps, the method realizes high-precision and real-time detection of the insulation impedance of the energy storage converter, effectively compensates for environmental interference, and ensures the reliability of the detection results under complex working conditions.

[0030] The embodiments of the present invention have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to technologies in the market, or to enable other ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An intelligent insulation impedance detection system for an energy storage converter, characterized in that, include: Bridge balance structure: adopts Wheatstone bridge topology, with four bridge arms forming a quadrilateral loop, two diagonals of which are respectively connected to a DC excitation power supply and a microvolt voltage detection port; the four bridge arms include at least one variable resistance arm and one measurement arm; Data acquisition module: used to collect ambient temperature, humidity and bridge output voltage in real time; Control module: connected to the data acquisition module, calculates the bridge balance deviation based on the ambient temperature, humidity and bridge output voltage, and outputs adjustment instructions; Automatic adjustment circuit: connected to the control module and the variable resistor arm respectively, responding to the adjustment instruction, controlling the variable resistor arm to adjust the resistance value to achieve dynamic balance of the bridge; The insulation impedance detection module is connected to the data acquisition module and outputs an insulation impedance detection value through a detection model based on the bridge output voltage.

2. The intelligent insulation impedance detection system for an energy storage converter according to claim 1, wherein The four bridge arms include: The first standard resistance arm and the second standard resistance arm: both are high-precision metal film resistors, connected in series between the positive electrode of the excitation power supply and the common node to form a fixed reference bridge arm; Variable resistor arm: a digital potentiometer is used, which is connected in series between the common node and the negative pole of the excitation power supply; the digital potentiometer is also connected to the control module through an I²C or SPI communication interface; Measuring arm: connected to the measured circuit of the energy storage converter, located on the same side bridge arm as the variable resistor arm, used to access the insulation impedance to be measured.

3. An intelligent insulation impedance detection system for an energy storage converter according to claim 2, characterized in that, The automatic adjustment circuit comprises: A resistor array connected in parallel at both ends of the variable resistor arm, wherein the resistance values ​​of the resistors in the resistor array are respectively exponentially or linearly regular, covering the target adjustment range; Each resistor in the resistor array is connected in series with a MOSFET switch.

4. An intelligent insulation impedance detection system for an energy storage converter according to claim 3, characterized in that, The data acquisition module comprises: Temperature sensor: used to collect the ambient temperature of the bridge in real time; Humidity sensor: used to collect environmental humidity in real time; Voltage sensor: A microvolt sensor based on a high impedance differential amplifier, connected to the voltage detection port between the two midpoints of the bridge, used to measure the unbalanced voltage ΔU of the bridge.

5. An intelligent insulation impedance detection system for an energy storage converter according to claim 4, characterized in that, The control module includes a data acquisition layer, an algorithm processing layer, and a control output layer; The data acquisition layer is connected to the temperature sensor, humidity sensor and voltage sensor respectively, and is used to obtain the ambient temperature, humidity and bridge unbalanced voltage ΔU; The algorithm processing layer uses the bridge unbalance voltage ΔU as an input variable, compensates ΔU by ambient temperature and humidity, and calculates the bridge balance adjustment amount by the PDI control algorithm; The control output layer generates an adjustment instruction for the digital potentiometer based on the balance adjustment amount calculated by the PID control algorithm, sends the adjustment instruction to the digital potentiometer through the SPI or I²C communication interface, and adjusts the variable resistor arm through the digital potentiometer to make the bridge output voltage approach zero.

6. The intelligent insulation impedance detection system for an energy storage converter according to claim 5, characterized in that, The algorithm processing layer is also connected to the BMS timing control module, which activates the detection circuit of each module in a time-sharing manner to avoid the parallel effect caused by the simultaneous connection of multiple modules to detect internal resistance.

7. An intelligent insulation impedance detection system for an energy storage converter according to claim 6, characterized in that, The insulation impedance detection module includes a signal preprocessing unit and a detection model; The signal preprocessing unit filters out the low-frequency drift and high-frequency noise of the bridge unbalance voltage ΔU through a band-pass filter, retains the effective frequency band of the polarization process of the insulating material and performs normalization processing, and converts it into dimensionless data between 0 and 1; The construction process of the detection model is as follows: Collect the insulation impedance detection values at multiple time series under different working conditions through a high-precision insulation resistance meter, and simultaneously collect the ambient temperature, humidity, and bridge unbalance voltage ΔU. Organize the ambient temperature, humidity, and bridge unbalance voltage ΔU data into a three-dimensional tensor format (T, H, ΔU), where T is the ambient temperature and H is the humidity; Obtain a large number of insulation impedance detection values and the corresponding (T, H, ΔU) for manual expert marking. After marking, use (T, H, ΔU) as the input of the convolutional neural network, and use the insulation impedance detection value as the output of the convolutional neural network for iterative training to obtain a detection model that outputs the insulation impedance detection value based on (T, H, ΔU).

8. An intelligent insulation impedance detection method for an energy storage converter, characterized in that, Applying the system according to any one of claims 1-7, comprising the following steps: System initialization: Start the bridge balance structure, adjust the variable resistance arm through a digital potentiometer to make the bridge unbalance voltage ΔU ≤ 10 μV, and record the initial balance parameters; Real-time data acquisition: The data acquisition module synchronously acquires ΔU, T, and H to generate a raw signal sequence with a time stamp; Dynamic balance adjustment: The control module calculates the adjustment amount according to ΔU, T, and H through the PID algorithm, and drives the automatic adjustment circuit to switch the digital potentiometer or shunt resistor to make ΔU return to balance; Signal preprocessing: Perform band-pass filtering and normalization processing on ΔU, and fuse it with T and H into three-dimensional time series data (T, H, ΔU); Insulation impedance detection: Input the preprocessed three-dimensional data (T, H, ΔU) into the detection model to output the real-time insulation impedance detection value.

Images