Battery internal state real-time measurement system based on complex domain signal processing

Abstract

The application relates to the technical field of lithium battery state monitoring, and provides a battery internal state real-time measurement system based on complex domain signal processing, which mainly comprises an upper computer and a hardware measurement end. An excitation generating unit of the hardware measurement end applies at least one frequency AC excitation signal to a measured battery, a sampling unit collects response signals of magnetic field signals and voltage signals of the battery under the action of the excitation signal in parallel, a complex domain signal processing unit is connected with the sampling unit, real and imaginary parts of each signal are synchronously acquired by real-time orthogonal demodulation of the response signals, and a communication interface unit is used for uploading complex characteristic quantities to the upper computer. The system and method provided by the application can improve battery measurement to complex domain measurement, capture phase information, and realize accurate identification of early lithium precipitation and other microscopic failures. Meanwhile, the real-time and interactivity of monitoring are considered, and the system and method are widely applied to production detection and battery full life cycle safety evaluation and the like.

Description

Technical Field

[0001] This invention relates to the field of lithium battery state monitoring technology, and in particular to a real-time measurement system for the internal state of a battery based on complex domain signal processing. Background Technology

[0002] Lithium-ion batteries, with their advantages of high energy density, long cycle life, and no memory effect, have become core power supply devices for consumer electronics, new energy vehicles, and energy storage systems. During charge-discharge cycles, a series of complex electrochemical reactions occur within the battery, including lithium-ion insertion / extraction, migration, and charge transfer, accompanied by various side reactions and microstructural evolution. Under conditions such as fast charging, low temperatures, and overcharging, lithium ions cannot effectively insert into the negative electrode lattice, easily leading to reduction deposition on the negative electrode surface, forming lithium plating, which then continues to grow into dendritic lithium dendrites. These abnormal microstructural changes not only consume active lithium and disrupt the stability of the solid electrolyte interface film, directly causing reversible capacity decay and shortened cycle life; but also, lithium dendrites piercing the separator can trigger internal short circuits, initiating chain reactions such as electrolyte decomposition and exothermic electrode materials, becoming the root cause of battery thermal runaway and threatening operational safety. Accurate measurement of these early failure characteristics is of great significance for battery research and development, production quality control, and service safety assessment.

[0003] CN121613341A discloses an online state estimation method and system for lithium battery digital twins, comprising: training a TP-FNO model, mapping operating parameters and spatiotemporal variables to the lithium-ion solid phase concentration distribution within the positive and negative electrode active particles; embedding the model into a nonlinear state-space equation, combining the posterior estimate of the system state at the previous moment with the current input to obtain the prior estimate of the system state and the prior covariance; calculating the predicted voltage value through the observation equation, and comparing it with the real-time measurement value to obtain the observation residual; employing a state estimation algorithm, using the observation residual to update the prior estimate of the system state and the prior covariance to obtain the posterior estimate of the system state and the posterior covariance; feeding back the posterior estimation result, constructing the model input features for the next moment, and realizing closed-loop iteration.

[0004] CN121385644A discloses a wideband online impedance measurement system and method for lithium batteries without the need for additional circuitry, relating to the field of lithium battery state monitoring technology. The system includes a power DC-DC converter, a drive circuit, a sampling module, and a digital controller. The digital controller includes a converter control module and an impedance calculation and output module. The output of the drive circuit is connected to the control terminal of the switching device of the power DC-DC converter. The output of the power DC-DC converter is connected to the input of the sampling module. The output of the sampling module is connected to both the input of the converter control module and the input of the impedance calculation and output module. The output of the converter control module is connected to the input of the drive circuit.

[0005] Existing battery state monitoring and anomaly diagnosis methods primarily rely on changes in the amplitude of electrical parameters such as voltage and current. The effective information obtained is limited in scope and data volume. These methods can only reflect changes in the overall macroscopic electrical characteristics of the battery, failing to accurately characterize early feature signals corresponding to microstructural evolution such as lithium plating and dendrite growth within the battery. They also struggle to effectively distinguish between fluctuations in normal operating conditions and abnormal signals indicating potential faults, leading to misjudgments or missed diagnoses. Consequently, they cannot meet the practical needs for accurate identification and safety warnings of early battery faults. Summary of the Invention

[0006] Long-term practical experience has shown that existing battery state detection methods typically only focus on amplitude changes in signals such as voltage and current, while ignoring the significant impact of internal battery processes such as polarization, diffusion, and lithium plating on the signal phase. Phase information contains key characteristics of early battery failure; traditional detection methods, by discarding phase information, struggle to effectively detect and warn of failure in its early stages. Furthermore, measuring the magnetic field distribution on the battery surface requires multi-sensor collaboration, but existing measurement systems suffer from significant deficiencies in multi-channel synchronous sampling, with nanosecond-level clock skew between channels. For measurement scenarios such as magnetic field gradients that rely on high-precision phase difference calculations, this clock deviation directly causes phase calculation distortion or even complete failure, making it impossible to obtain reliable internal state characterization information.

[0007] In view of this, the present invention provides a real-time battery internal state measurement system based on complex domain signal processing, comprising, The host computer is connected to the hardware measurement terminal through a communication interface to provide a human-machine interface, receive and display complex feature quantities and / or raw sampled data from the hardware measurement terminal in real time, and store the complex feature quantities and / or raw sampled data in a local storage medium. Hardware measurement terminals, including, An excitation generating unit is used to apply an AC excitation signal of at least one frequency to the battery under test; The sampling unit is used to acquire, in parallel, at least 128 response signals generated by the battery under the action of the excitation signal with a time synchronization accuracy of less than or equal to 10 picoseconds. The response signals include magnetic field signals and voltage signals. A complex domain signal processing unit, connected to the sampling unit, is used to perform real-time quadrature demodulation on the response signal to synchronously acquire the real and imaginary components of each signal, and to calculate complex feature quantities characterizing the internal state of the battery based on the real and imaginary components. The communication interface unit is used to upload the complex feature quantity to the host computer and receive configuration parameters from the host computer.

[0008] Preferably, the complex characteristic quantity includes the complex impedance characteristic value, λ=α+βi Where α represents the battery's resistance characteristic and β represents the battery's reactance characteristic; the complex domain signal processing unit monitors the battery status by tracking the movement trajectory of λ in the complex plane in real time.

[0009] Preferably, the complex feature quantity includes the phase winding index Φwinding, and the complex domain signal processing unit calculates the phase winding index by performing path integration on the phase data of at least four magnetic field sensors that are spatially distributed in a grid pattern.

[0010] Preferably, the complex characteristic quantity further includes complex numbers. , in, To enable real-time calibration based on the proportional relationship between voltage V and phase change rate θ, This is for real-time calibration based on the Joule heat generation rate QJoule and the topological entropy growth rate dStopo / dt.

[0011] Preferably, the hardware measurement terminal further includes a feature trigger, which is directly connected to the complex domain signal processing unit and is used to output a hardware pulse signal within 1 nanosecond when the complex feature quantity exceeds a preset threshold; the preset threshold includes the phase winding index Φwinding changing from 0 to a non-zero value, or The growth rate exceeded the preset threshold.

[0012] Preferably, the sampling unit includes: a clock source with jitter less than 50 femtoseconds; and a clock distribution network connected to the clock source; The system has 128 analog-to-digital converters, each with a resolution of at least 18 bits and a sampling rate of at least 15 Mbps, and all converters are controlled by the same trigger signal.

[0013] Preferably, the complex domain signal processing unit is implemented based on a field-programmable gate array (FPGA), and internally includes at least a first logic region and a second logic region interconnected by long lines; The first logic region is used to process response signals from the inactive region of the battery and to establish and update a dynamic background noise model in real time. The second logic region is used to receive the response signal from the battery active region, perform orthogonal projection using the background noise model to subtract background noise, and then calculate the complex feature quantity based on the signal after noise subtraction.

[0014] This invention also discloses a method for real-time measurement of the internal state of a battery based on complex domain signal processing, as described above, the method comprising: Step S1: The host computer sets various parameters of the hardware measurement front end, including excitation signal parameters, sampling channel parameters, and trigger thresholds; the hardware measurement end applies an AC excitation signal to the battery under test according to the configuration parameters; Step S2: The hardware measurement end acquires at least 128 magnetic field signals and voltage signals generated by the response of the battery under test in parallel with a time synchronization accuracy of less than or equal to 10 picoseconds; Step S3: The hardware measurement end performs real-time quadrature demodulation on each acquired signal through hard-wired logic circuits, and synchronously extracts the real and imaginary parts of the signal; Step S4: Based on the real and imaginary parts, the hardware measurement end calculates in real-time complex characteristic quantities used to characterize the internal state of the battery, including complex impedance characteristic value λ and phase winding index Φwinding, complex... At least one of them; Step S5, when the complex feature quantity exceeds a preset threshold, the hardware-level feature trigger outputs a hardware pulse signal within 1 nanosecond; the hardware measurement terminal uploads the complex feature quantity to the host computer for data processing through the communication interface; Step S6: The host computer displays the complex feature quantities in real time, stores the complex feature quantities and / or raw sampling data according to the user configuration, and outputs the abnormal points of the tested battery.

[0015] Preferably, the method includes, In step S4, the phase winding index Φwinding is performed by hardwired logic within one clock cycle.

[0016] The present invention also discloses a machine-readable storage medium storing instructions for causing a machine to execute the method of the real-time measurement system for battery internal state based on complex domain signal processing as described in any of the preceding claims.

[0017] This invention provides a real-time battery internal state measurement system based on complex domain signal processing, mainly composed of a host computer and a hardware measurement terminal. The host computer is connected to the hardware measurement terminal via a communication interface. The excitation generation unit of the hardware measurement terminal applies an AC excitation signal of at least one frequency to the battery under test. The sampling unit collects the response signals of the magnetic field signal and voltage signal generated by the battery under the action of the excitation signal in parallel. The complex domain signal processing unit is connected to the sampling unit and performs real-time orthogonal demodulation on the response signal to synchronously acquire the real and imaginary components of each signal, and calculates the complex characteristic quantities characterizing the internal state of the battery accordingly. The communication interface unit is used to upload the complex characteristic quantities to the host computer and receive configuration parameters from the host computer. The system and method provided by this invention can upgrade battery measurement from traditional scalar measurement to complex domain measurement. By capturing neglected phase information, it can achieve accurate identification of micro-failures such as early lithium plating. Experimental data shows that it can provide early warning when voltage and temperature are normal. At the same time, through software and hardware co-design, it takes into account both real-time performance and interactivity, and can be widely used in battery research and development, production testing, and battery life cycle safety assessment. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of a real-time battery internal state measurement system based on complex domain signal processing, according to one embodiment of the present invention. Detailed Implementation

[0019] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0020] To address the issue that current battery state detection methods typically focus only on amplitude changes in signals such as voltage and current, neglecting the significant impact of internal processes like polarization, diffusion, and lithium plating on signal phase, this invention proposes a real-time battery internal state measurement system based on complex domain signal processing. Phase information contains key characteristics of early battery failure, and traditional detection methods, by discarding this phase information, struggle to effectively detect and warn of early-stage failures. Furthermore, measuring the magnetic field distribution on the battery surface requires multi-sensor collaboration, but existing measurement systems suffer from significant deficiencies in multi-channel synchronous sampling, exhibiting nanosecond-level clock skew between channels. For measurement scenarios such as magnetic field gradients that rely on high-precision phase difference calculations, this clock skew directly causes phase calculation distortion or even complete failure, hindering the acquisition of reliable internal state characterization information. Figure 1As shown, the real-time battery internal state measurement system based on complex domain signal processing includes: The host computer is connected to the hardware measurement terminal through a communication interface to provide a human-machine interface, receive and display complex feature quantities and / or raw sampled data from the hardware measurement terminal in real time, and store the complex feature quantities and / or raw sampled data in a local storage medium. Hardware measurement terminals, including, An excitation generating unit is used to apply an AC excitation signal of at least one frequency to the battery under test; The sampling unit is used to acquire, in parallel, at least 128 response signals generated by the battery under the action of the excitation signal with a time synchronization accuracy of less than or equal to 10 picoseconds. The response signals include magnetic field signals and voltage signals. A complex domain signal processing unit, connected to the sampling unit, is used to perform real-time quadrature demodulation on the response signal to synchronously acquire the real and imaginary components of each signal, and to calculate complex feature quantities characterizing the internal state of the battery based on the real and imaginary components. The communication interface unit is used to upload the complex feature quantity to the host computer and receive configuration parameters from the host computer.

[0021] This invention provides a real-time battery internal state measurement system based on complex domain signal processing, mainly composed of a host computer and a hardware measurement terminal. The host computer is connected to the hardware measurement terminal via a communication interface. The excitation generation unit of the hardware measurement terminal applies an AC excitation signal of at least one frequency to the battery under test. The sampling unit collects the response signals of the magnetic field and voltage signals generated by the battery under the excitation signal in parallel. The complex domain signal processing unit is connected to the sampling unit and performs real-time orthogonal demodulation on the response signals to synchronously acquire the real and imaginary components of each signal, and calculates the complex characteristic quantities characterizing the internal state of the battery accordingly. The communication interface unit is used to upload the complex characteristic quantities to the host computer and receive configuration parameters from the host computer. The system provided by this invention can upgrade battery measurement from traditional scalar measurement to complex domain measurement. By capturing neglected phase information, it can accurately identify microscopic failures such as early lithium plating. Experimental data shows that it can provide early warnings when voltage and temperature are normal. At the same time, through software and hardware co-design, it balances real-time performance and interactivity, and can be widely used in battery research and development, production testing, and battery life cycle safety assessment.

[0022] To better measure magnetic field and voltage signals and achieve parallel, high-speed, and high-precision acquisition of multi-point physical field signals from the battery, the sensors support connection to a Tunnel Magnetoresistance (TMR) sensor array for measuring the weak magnetic field distribution on the battery surface. For example, four TMR magnetic field sensors are mounted on the surface of each 18650 cell to measure the weak magnetic field changes generated during the cell's charging and discharging process. Voltage sensors are used to measure voltage changes or directly read voltage change signals from the Battery Management System (BMS). The sampling unit includes a jitter clock distribution network and 128 independent 18-bit analog-to-digital converters (ADCs). The jitter clock distribution network contains a dedicated phase-locked loop (PLL) circuit to provide a unified reference clock for all sampling channels. The output jitter of this clock source is strictly suppressed to below 50 femtoseconds (fs). It includes 128 independent 18-bit analog-to-digital converters (ADCs), each with a sampling rate of no less than 15 megasamples per second (MSPS). All 128 channels respond to the same trigger signal, and the timestamp deviation of their sampling time is strictly controlled within 10 picoseconds (ps). This ensures the accuracy of subsequent phase difference calculations between multiple signals.

[0023] The communication interface unit is used to realize data exchange and command transmission between the hardware measurement front end and the host computer, including a high-speed data interface and a control interface. For example, USB 3.0, GigE Vision, or PCIe are used to upload measurement data and feature values ​​to the host computer in real time. The control interface is used to receive configuration parameters from the host computer, including the frequency, amplitude, and sweep mode of the excitation signal, as well as the selection of the sampling channel and trigger threshold.

[0024] The excitation unit, used to generate the electrical signal to excite the battery, includes a Direct Digital Synthesizer (DDS) and a power amplification and injection circuit. The DDS can generate sinusoidal excitation signals with adjustable amplitudes from 1 MHz to 100 kHz. The frequency, amplitude, and sweep mode are configurable via host computer software. For example, an FPGA-controlled DDS generates a 1 MHz-10 kHz swept sine wave, which is injected into the positive and negative terminals of the battery pack through a linear power amplifier. The power amplification and injection circuit then amplifies the signal generated by the DDS before injecting it into the electrodes of the battery under test.

[0025] Implemented using hard-wired logic within a Field-Programmable Gate Array (FPGA), this system processes massive amounts of data collected by the sampling unit in real time. It is configured to perform quadrature demodulation on each acquired voltage, current, or magnetic field signal in real time, synchronously outputting the real and imaginary parts of the signal, i.e., the in-phase component I and the quadrature component Q, thus completely preserving the amplitude and phase information of the signal. Based on the electrochemical model of the battery, matrix operations are performed on the I / Q data in the complex domain to extract complex impedance characteristic values ​​in real time. In a more preferred embodiment of this invention, the complex characteristic values ​​include complex impedance characteristic values. λ=α+βi Where α represents the battery's resistance characteristic, and β represents its reactance characteristic; the complex domain signal processing unit monitors the battery state by tracking the movement trajectory of λ in the complex plane in real time. The real part α reflects the battery's ohmic polarization and charge transfer impedance, while the imaginary part β reflects the battery's diffusion impedance and inductive reactance characteristics.

[0026] For example, during the internal resistance testing phase before the battery leaves the factory, the sampling unit collects 128 magnetic field signals in parallel and establishes an environmental background noise model through the first logic region. By tracing the trajectory of the complex impedance characteristic value λ on the complex plane, the system finds that the trajectory curvature of the abnormal cell deviates significantly from the standard envelope. Even if the DC internal resistance α of the cell is qualified, the uneven polarization distribution reflected by the reactance characteristic β indicates that it poses a medium- to long-term safety risk.

[0027] To analyze the phase distribution of multi-point magnetic field signals, the phase differences between adjacent sensor nodes are accumulated to monitor whether entanglement or vortexing occurs in the spatial distribution of the phase field. In a more preferred embodiment, the complex characteristic quantity includes a phase entanglement index Φwinding. The complex domain signal processing unit calculates the phase entanglement index by performing path integration on the phase data of at least four spatially grid-distributed magnetic field sensors. More preferably, when the 128-channel analog-to-digital converter is driven by the same 50fs jitter clock source, the sampled data stream enters the field-programmable gate array (FPGA). Phase path integration is completed within one clock cycle through hard-wired logic, thereby ensuring zero-latency capture of lithium plating singularities.

[0028] Among them, link variables for, , The phase of adjacent sensor nodes.

[0029] Assuming there are 4 sensor nodes, calculate the Wilson ring W. Extracting local Berry curvature , The phase winding index Φwinding is calculated as follows: When the phase winding index Φwinding jumps from 0 to a non-zero integer value, it indicates that a local topological defect has appeared inside the battery, namely a lithium plating singularity or dendrite.

[0030] To provide early warning of lithium plating under high-rate fast charging conditions, in this preferred embodiment, the excitation generation unit applies a 127Hz sinusoidal perturbation signal to the battery under test. When the charging rate reaches 3C, the complex domain signal processing unit observes that the phase winding index Φwinding begins to fluctuate slightly from 0, indicating that a local circulating current or lithium plating singularity may have occurred inside the battery. At this time, the voltage and temperature sensors have not yet detected the anomaly, but due to the complex domain signal processing unit… imaginary part Based on the rate of increase of topological entropy, which exceeds 15% / s, a hardware-level feature trigger is activated, and the system can successfully determine that the negative electrode surface has formed a local manifold topological lock, i.e., early lithium plating budding.

[0031] To better identify the phase change pattern, in a more preferred embodiment of the present invention, the complex characteristic quantity further includes a complex number. , in, To enable real-time calibration based on the proportional relationship between voltage V and phase change rate θ, This is for real-time calibration based on the Joule heat generation rate QJoule and the topological entropy growth rate dStopo / dt.

[0032] when When exponential growth occurs, it indicates that the system has entered a strongly dissipative non-Hermitian mode and is about to experience thermal runaway. T is the temperature. When the growth rate exceeds a preset threshold, the trigger can also be activated. This design is directly related to the early warning of non-Hermitian sensitivity outbreaks and thermal runaway, enabling the system to capture weak early signals that are physically amplified near the EP point.

[0033] To achieve real-time hardware-level triggering of battery abnormal states without the need for a host computer, the response time is compressed to within 1 nanosecond, significantly improving the real-time performance of fault detection. Furthermore, the triggering conditions can be flexibly configured through a programmable threshold register to adapt to different operating conditions and battery types, improving system versatility and detection accuracy. In a more preferred embodiment of the invention, the hardware measurement terminal further includes a feature trigger, which is directly connected to the complex domain signal processing unit. This feature trigger is used to output a hardware pulse signal within 1 nanosecond when the complex feature quantity exceeds a preset threshold. The feature trigger, directly connected to the complex domain signal processing unit, includes a high-speed comparator and a programmable threshold register. The preset threshold includes the phase winding index Φwinding transitioning from 0 to a non-zero value, or... The growth rate exceeded a preset threshold. For example, the feature trigger captured this transition and output a high-level pulse through the I / O interface within 1 ns. Simultaneously, the host computer software received the trigger signal, displayed a red alarm window on the interface, and automatically saved the original data for 10 seconds before and after the alarm time. After 360 minutes, the cell experienced thermal runaway, verifying the accuracy of the measurement. Users can then use the offline analysis function of the host computer software to replay the original data at the alarm time and conduct in-depth analysis of the lithium plating process.

[0034] To eliminate inter-channel clock skew at its source, ensure accurate and reliable phase difference calculation, and meet the high-precision measurement requirements such as magnetic field gradients, this invention significantly improves the acquisition accuracy and dynamic range of weak signals, such as weak magnetic fields on battery surfaces and low-amplitude response signals, effectively distinguishing minute feature changes corresponding to early faults. In a more preferred embodiment, the sampling unit includes: a clock source with jitter less than 50 femtoseconds; a clock distribution network connected to the clock source; and 128 analog-to-digital converters (ADCs), each with a resolution of at least 18 bits and a sampling rate of at least 15 Mbps, all controlled by the same trigger signal.

[0035] To meet the high-speed processing requirements of real-time quadrature demodulation and noise suppression, and to achieve parallel and independent processing of active region signals and inactive region background signals, stable data interaction and consistent time delay between regions are ensured, guaranteeing strict synchronization between the noise model and signal processing. In a more preferred embodiment of the present invention, the complex domain signal processing unit is implemented based on a field-programmable gate array (FPGA), and internally includes at least a first logic region and a second logic region interconnected by long lines. The first logic region is used to process response signals from the inactive region of the battery and to establish and update a dynamic background noise model in real time. The second logic region is used to receive the response signal from the battery active region, perform orthogonal projection using the background noise model to subtract background noise, and then calculate the complex feature quantity based on the signal after noise subtraction.

[0036] Among them, the long-line interconnect is used for ultra-low latency data exchange between two SLRs to ensure the real-time performance of background model subtraction, with an interconnect latency of less than 500ps.

[0037] The present invention also provides a method for a real-time measurement system of battery internal state based on complex domain signal processing as described above, the method comprising, Step S1: The host computer sets various parameters of the hardware measurement front end, including excitation signal parameters, sampling channel parameters, and trigger thresholds; the hardware measurement end applies an AC excitation signal to the battery under test according to the configuration parameters; Step S2: The hardware measurement end acquires at least 128 magnetic field signals and voltage signals generated by the response of the battery under test in parallel with a time synchronization accuracy of less than or equal to 10 picoseconds; Step S3: The hardware measurement end performs real-time quadrature demodulation on each acquired signal through hard-wired logic circuits, and synchronously extracts the real and imaginary parts of the signal; Step S4: Based on the real and imaginary parts, the hardware measurement end calculates in real-time complex characteristic quantities used to characterize the internal state of the battery, including complex impedance characteristic value λ and phase winding index Φwinding, complex... At least one of them; Step S5, when the complex feature quantity exceeds a preset threshold, the hardware-level feature trigger outputs a hardware pulse signal within 1 nanosecond; the hardware measurement terminal uploads the complex feature quantity to the host computer for data processing through the communication interface; Step S6: The host computer displays the complex feature quantities in real time, stores the complex feature quantities and / or raw sampling data according to the user configuration, and outputs the abnormal points of the tested battery.

[0038] This invention provides a method for real-time measurement of battery internal state based on complex domain signal processing. Steps S1-S6 enable high-speed completion of all processes from excitation, synchronous sampling, real-time demodulation, complex feature calculation to hardware-level triggering by the hardware measurement end, eliminating reliance on software and host computers and reducing system latency. It directly judges based on complex features such as complex impedance λ, phase winding index Φwinding, and complex number κ, which is more sensitive to early, weak anomalies than traditional methods that only use amplitude, and can identify potential problems such as lithium plating and dendrite formation at the nascent stage of a fault. It achieves sub-microsecond anomaly alarms and protection triggering, far faster than software processing, significantly improving battery safety early warning and fault response capabilities.

[0039] To completely avoid delays caused by software calculations or multi-cycle operations and ensure real-time output of phase-related characteristics, providing an immediate and reliable basis for subsequent nanosecond-level hardware triggering, in a more preferred embodiment of the present invention, the method includes, in step S4, the phase winding index Φwinding being completed by hard-wired logic within one clock cycle.

[0040] The present invention also discloses a machine-readable storage medium storing instructions for causing a machine to perform the method of the real-time measurement system for battery internal state based on complex domain signal processing as described above.

[0041] In the above embodiments, the descriptions of each embodiment have different focuses. Parts not described in detail in a certain embodiment can be referred to in the relevant descriptions of other embodiments. The above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A real-time measurement system for the internal state of a battery based on complex domain signal processing, characterized in that, The real-time battery internal state measurement system based on complex domain signal processing includes: The host computer is connected to the hardware measurement terminal through a communication interface to provide a human-machine interface, receive and display complex feature quantities and / or raw sampled data from the hardware measurement terminal in real time, and store the complex feature quantities and / or raw sampled data in a local storage medium. Hardware measurement terminals, including, An excitation generating unit is used to apply an AC excitation signal of at least one frequency to the battery under test; The sampling unit is used to acquire, in parallel, at least 128 response signals generated by the battery under the action of the excitation signal with a time synchronization accuracy of less than or equal to 10 picoseconds. The response signals include magnetic field signals and voltage signals. A complex domain signal processing unit, connected to the sampling unit, is used to perform real-time quadrature demodulation on the response signal to synchronously acquire the real and imaginary components of each signal, and to calculate complex feature quantities characterizing the internal state of the battery based on the real and imaginary components. The communication interface unit is used to upload the complex feature quantity to the host computer and receive configuration parameters from the host computer.

2. The real-time battery internal state measurement system based on complex domain signal processing according to claim 1, characterized in that, The complex characteristic quantities include complex impedance characteristic values, λ=α+βi Where α represents the battery's resistance characteristic and β represents the battery's reactance characteristic; the complex domain signal processing unit monitors the battery status by tracking the movement trajectory of λ in the complex plane in real time.

3. The real-time battery internal state measurement system based on complex domain signal processing according to claim 1, characterized in that, The complex feature quantity includes the phase winding index Φwinding, which is calculated by the complex domain signal processing unit through path integration of phase data from at least four magnetic field sensors that are spatially distributed in a grid pattern.

4. The real-time battery internal state measurement system based on complex domain signal processing according to claim 3, characterized in that, The complex characteristic quantities also include complex numbers. , in, To enable real-time calibration based on the proportional relationship between voltage V and phase change rate θ, This is for real-time calibration based on the Joule heat generation rate QJoule and the topological entropy growth rate dStopo / dt.

5. The real-time battery internal state measurement system based on complex domain signal processing according to claim 4, characterized in that, The hardware measurement terminal also includes a feature trigger, which is directly connected to the complex domain signal processing unit and is used to output a hardware pulse signal within 1 nanosecond when the complex feature quantity exceeds a preset threshold. The preset threshold includes the phase winding index Φwinding changing from 0 to a non-zero value, or The growth rate exceeded the preset threshold.

6. The real-time battery internal state measurement system based on complex domain signal processing according to claim 5, characterized in that, The sampling unit includes: a clock source with jitter less than 50 femtoseconds; a clock distribution network connected to the clock source; and 128 analog-to-digital converters, each with a resolution of not less than 18 bits and a sampling rate of not less than 15 megabits per second, and all analog-to-digital converters are controlled by the same trigger signal.

7. The real-time battery internal state measurement system based on complex domain signal processing according to any one of claims 1-6, characterized in that, The complex domain signal processing unit is implemented based on a field-programmable gate array and contains at least a first logic region and a second logic region interconnected by long lines. The first logic region is used to process response signals from the inactive region of the battery and to establish and update a dynamic background noise model in real time. The second logic region is used to receive the response signal from the battery active region, perform orthogonal projection using the background noise model to subtract background noise, and then calculate the complex feature quantity based on the signal after noise subtraction.

8. A method for a real-time measurement system of battery internal state based on complex domain signal processing as described in any one of claims 1-7, characterized in that, The method includes, Step S1: The host computer sets various parameters of the hardware measurement front end, including excitation signal parameters, sampling channel parameters, and trigger thresholds; the hardware measurement end applies an AC excitation signal to the battery under test according to the configuration parameters; Step S2: The hardware measurement end acquires at least 128 magnetic field signals and voltage signals generated by the response of the battery under test in parallel with a time synchronization accuracy of less than or equal to 10 picoseconds; Step S3: The hardware measurement end performs real-time quadrature demodulation on each acquired signal through hard-wired logic circuits, and synchronously extracts the real and imaginary parts of the signal; Step S4: Based on the real and imaginary parts, the hardware measurement end calculates in real-time complex characteristic quantities used to characterize the internal state of the battery, including complex impedance characteristic value λ and phase winding index Φwinding, complex... At least one of them; Step S5, when the complex feature quantity exceeds a preset threshold, the hardware-level feature trigger outputs a hardware pulse signal within 1 nanosecond; the hardware measurement terminal uploads the complex feature quantity to the host computer for data processing through the communication interface; Step S6: The host computer displays the complex feature quantities in real time, stores the complex feature quantities and / or raw sampling data according to the user configuration, and outputs the abnormal points of the tested battery.

9. The method according to claim 8, characterized in that, The method includes, In step S4, the phase winding index Φwinding is performed by hardwired logic within one clock cycle.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to perform the method of the real-time measurement system for battery internal state based on complex domain signal processing as described in any one of claims 8 or 9.

Images