Storage battery automatic polishing evaluation system and method

Abstract

The invention relates to the technical field of storage battery maintenance, in particular to an automatic storage battery polishing evaluation system which comprises a control unit, the control unit is used for controlling the polishing state, and the control unit comprises a shell, an internal circuit and a human-computer interaction interface; the polishing unit is used for executing polishing operation on the storage battery post, and the polishing unit comprises an execution module and a joint module; the monitoring unit is used for monitoring the grinding state; and the Kelvin clip type measuring pen is used for measuring electrical parameters of the polished storage battery, and the core controller, the control unit, the polishing unit and the monitoring unit are in communication connection with the core controller. The method has the beneficial effects that objective quantification of the oxidation degree is achieved through machine vision and an algorithm, and personalized polishing parameters are dynamically generated for precise self-adaptive machining.

Description

Technical Field

[0001] This invention relates to the field of battery maintenance technology, and in particular to an automatic battery polishing and evaluation system and method. Background Technology

[0002] With the rapid development of industries such as power, communications, data centers, and new energy, batteries, as critical backup power sources and energy storage units, require reliable performance and timely maintenance. As key interfaces for energy transmission, battery terminals are prone to oxide film formation on their surfaces during long-term use. This oxide layer significantly increases contact resistance, leading to overheating at connection points, inaccurate voltage measurements, deterioration of battery pack consistency, and even safety accidents.

[0003] The treatment and performance testing of battery terminal oxidation are hampered by several key issues. Battery terminal grinding and maintenance rely entirely on manual experience to judge the degree of oxidation, and setting grinding parameters by feel leads to unstable process quality, uncontrollable material loss, low efficiency, and the inability to achieve adaptive and precise processing. Furthermore, traditional methods lack the ability to conduct online precision testing of the surface geometry after grinding and to simultaneously conduct in-depth diagnostics of the battery's internal electrical performance, resulting in maintenance operations remaining at the level of extensive surface treatment. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an automatic grinding and evaluation system for a storage battery, comprising a control unit for controlling the grinding state; A polishing unit is used to perform polishing operations on the battery terminals. The polishing unit includes an execution module and a joint module. The monitoring unit is used to monitor the grinding status; Kelvin clip-on measuring pen, used to measure the electrical parameters of a polished battery; The core controller, control unit, grinding unit, and monitoring unit are connected to the core controller for communication.

[0005] As a preferred embodiment of the automatic battery polishing and evaluation system of the present invention, the joint module includes three sets of rotary joints and one set of pitch joints, and the three sets of rotary joints are hinged in sequence.

[0006] As a preferred embodiment of the automatic battery polishing and evaluation system of the present invention, the execution module includes a motor and a polishing drill bit; the motor is connected to a pitch joint, the output shaft of the motor is connected to the polishing drill bit, and silicon carbide abrasive particles are attached to the surface of the polishing drill bit.

[0007] As a preferred embodiment of the automatic battery polishing evaluation system of the present invention, the monitoring unit includes a camera for acquiring images of the polishing area and a pressure sensor for acquiring polishing pressure.

[0008] In a second aspect, the present invention provides an automatic grinding evaluation method for storage batteries, including calculating grinding control parameters based on a grinding control parameter generation model; The grinding drill bit is controlled to grind the pole post according to the grinding control parameters. After grinding is completed, the grinding quality is evaluated, a grinding quality evaluation result is generated, and the grinding control parameter generation model is updated based on the grinding quality evaluation result.

[0009] As a preferred embodiment of the automatic battery polishing and evaluation method of the present invention, the polishing and evaluation method further includes: When the target polishing time reaches the preset standard, polishing is stopped and the polishing quality is evaluated.

[0010] As a preferred embodiment of the automatic grinding and evaluation method for batteries of the present invention, the grinding device is controlled by PID to control the grinding force, and the grinding speed and time are adaptively adjusted according to the degree of oxidation.

[0011] As a preferred embodiment of the automatic battery polishing evaluation method of the present invention, the evaluation of polishing quality includes: Calculate the thickness of the residual oxide layer on the surface of the electrode post after polishing; Evaluate the surface roughness of the polished pole piece; The grinding quality is evaluated based on the thickness of the residual oxide layer and the roughness of the ground electrode surface, and a grinding quality evaluation result is generated.

[0012] As a preferred embodiment of the automatic battery polishing evaluation method of the present invention, the calculation of polishing control parameters based on the polishing control parameter generation model includes: By acquiring images of the electrode posts using a camera, identifying the positions of the electrode posts, and quantitatively evaluating the degree of surface oxidation, an oxidation degree evaluation value is obtained. Based on the oxidation degree assessment value, the polishing control parameters are calculated.

[0013] In a preferred embodiment of the automatic battery grinding evaluation method of the present invention, the roughness of the surface of the ground terminal post is determined by using the principle of laser triangulation and statistical parameters of height variation. Compared with the prior art, the beneficial effects of the present invention are as follows: By using machine vision and algorithms, the degree of oxidation is objectively quantified, and personalized polishing parameters are dynamically generated for precise adaptive processing. At the same time, high-precision sensing technology is integrated to evaluate the polishing geometry and material loss online, and advanced electrochemical impedance spectroscopy analysis and temperature compensation technology are combined to conduct in-depth diagnosis of the battery's internal health status. Finally, through data-driven online learning and multi-objective optimization algorithms, the system can continuously optimize process parameters from historical operations, completely transforming terminal maintenance from a rough operation relying on manual experience to a standardized, traceable intelligent process, significantly improving processing accuracy, efficiency and consistency, and achieving a leap from surface treatment to performance assurance. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of an automatic battery polishing and evaluation system.

[0016] Figure 2 The method and process for automatic grinding and evaluation of storage batteries. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0018] Example 1, referring to Figure 1-2 This is the first embodiment of the present invention, which provides an automatic battery polishing and evaluation system, comprising: Control unit 1 is used to control the polishing state. Control unit 1 includes a housing, internal circuitry, and human-machine interface. Grinding unit 2 is used to perform grinding operations on the battery terminals. Grinding unit 2 includes execution module 21 and joint module 22. The joint assembly includes three sets of rotary joints 222 and one set of pitch joints 221. The three sets of rotary joints 222 are hinged in sequence. One end of one set of rotary joints 222 is hinged to the outer casing. The pitch joint 221 is hinged to the set of rotary joints 222 that is away from the outer casing.

[0019] The execution module 21 includes a motor 212 and a grinding drill bit 211; the motor 212 is connected to the pitch joint 221, the output shaft of the motor 212 is connected to the grinding drill bit 211, and silicon carbide abrasive particles are attached to the surface of the grinding drill bit 211.

[0020] The monitoring unit is used to monitor the grinding status. The monitoring unit includes a camera for acquiring images of the grinding area and a pressure sensor for acquiring grinding pressure. Kelvin clip-on measuring pen, used to measure the electrical parameters of a polished battery; The core controller, control unit 1, grinding unit 2, monitoring unit, and Kelvin clip-on measuring pen are connected to the core controller for communication.

[0021] It should be noted that a ring-shaped LED light can be installed on the outer side of the grinding drill bit 211. The ring-shaped LED light surrounds the grinding drill bit 211 and has an illumination intensity of ≥500 lux. The LED light is encapsulated by a light-transmitting insulating cover. The ring-shaped LED light can improve the brightness during shooting and ensure the uniformity of illumination.

[0022] The miniature camera can be a 1 / 4-inch CMOS miniature camera, fixed to the side of the top of the robot arm with the lens facing the grinding area, and connected to the control unit 1 through a high-definition transmission line; the core controller is an STM32H743 microcontroller.

[0023] It should be noted that the automatic battery polishing and evaluation system also includes an 18650 lithium battery pack for power supply, equipped with a power management module, supporting 5V / 2A output to power each functional unit. A single full charge can provide continuous operation for ≥8 hours and standby time for ≥72 hours. The terminal screen displays the remaining power in real time, and triggers a buzzer when the power is low (≤10%).

[0024] Example 2, refer to Figure 1-2 As an embodiment of the present invention, based on the above embodiment, an automatic grinding and evaluation method for storage batteries is provided, including steps S100-S400: S100. Calculate grinding control parameters based on the grinding control parameter generation model; S200, control the grinding drill bit 211 to grind the pole post according to the grinding control parameters, evaluate the grinding quality after grinding is completed, generate the grinding quality evaluation result, and update the grinding control parameter generation model based on the grinding quality evaluation result. S300 performs a quality assessment on the polished battery and outputs the battery quality assessment results.

[0025] It should be noted that the core problems of battery terminal grinding and maintenance are that it relies entirely on manual experience to judge the degree of oxidation, sets grinding parameters by feel, resulting in unstable process quality, uncontrollable material loss, low efficiency, and inability to achieve adaptive and precise processing. In addition, traditional methods lack the ability to conduct online precision detection of the surface geometry quality after grinding and the ability to simultaneously conduct in-depth diagnosis of the internal electrical performance of the battery, which results in maintenance work remaining at the level of rough surface treatment.

[0026] Therefore, to address the aforementioned issues, steps S100-S300 integrate a multi-joint robotic arm, a grinding execution module, a vision and pressure monitoring unit, and a precision electrical measurement module, all under the unified control of a core controller. This achieves fully automated closed-loop operation from terminal positioning and oxidation degree identification to adaptive precision grinding and post-grinding electrical performance testing. It effectively replaces the traditional extensive mode of manual operation relying on visual inspection and touch, significantly improving the positioning accuracy, process consistency, and overall efficiency of grinding operations, and laying the hardware foundation for standardized and intelligent battery maintenance.

[0027] Example 3, referring to Figure 1-2 As an embodiment of the present invention, based on the above embodiment, an automatic grinding evaluation method for storage batteries is provided. In this embodiment, step S100, based on the grinding control parameter generation model, calculates the grinding control parameters, including the following steps A1-A7: A1: The image of the pole to be polished is acquired by a camera, the pixel of the pole center in the image coordinate system is identified, and the image is processed to obtain a standardized image. The image processing includes at least noise suppression, adaptive brightness and contrast correction, and adaptive histogram equalization. The image data after noise suppression and adaptive brightness and contrast correction satisfy the following formula: ; In the formula, These are the preprocessed image pixel values. These are the original image pixel values; This is the contrast adjustment factor. For brightness adjustment parameters, This is the noise suppression coefficient. This represents the noise component.

[0028] Understandably, the contrast adjustment coefficient α and the brightness adjustment parameter β are dynamically calculated based on the global statistical features of the input image, including the average gray value and gray-level distribution range, to achieve automatic normalization of image brightness and adaptive stretching of contrast. Noise suppression is achieved using Gaussian filtering, and the noise component... It characterizes the random interference present in the image, and its statistical properties conform to a Gaussian distribution.

[0029] In one alternative implementation, adaptive histogram equalization can employ a method based on local cumulative distribution function remapping, the transformation of which can be expressed as the formula: ; In the formula, For the enhanced image, CDF is the cumulative distribution function. and These are the minimum and maximum values ​​of the cumulative distribution function, respectively.

[0030] In another alternative implementation, adaptive histogram equalization can also employ a contrast-limited adaptive histogram equalization method to enhance contrast while suppressing excessive noise amplification.

[0031] It should be noted that a coordinate system calibration procedure will be performed when the system is used for the first time or during regular maintenance to establish a unified visual-mechanical spatial mapping relationship. This unified visual-mechanical spatial mapping relationship can be established by using a fixed corner point of the battery rack as the origin to create a world coordinate system.

[0032] A2: Extract texture, color, and shape features from the polar surface of the standardized image to construct a comprehensive feature vector. ; In the formula, F is the comprehensive feature vector. For texture feature vectors, For color feature vectors, This is the shape feature vector.

[0033] It should be noted that texture features can be extracted using the gray-level co-occurrence matrix method. The formula for calculating the elements of the gray-level co-occurrence matrix of the image is as follows: ; In the formula, For elements of the gray-level co-occurrence matrix, Here, is the image grayscale value, d is the pixel distance, and is the orientation angle. Let M be the Dirac function, and M and N be the number of rows and columns of the image, respectively.

[0034] A3: The oxidation degree assessment value is calculated based on the comprehensive feature vector. The oxidation degree assessment value is calculated using a fuzzy logic method. ; In the formula, This is an assessment value for the degree of oxidation. Let be the weight coefficient of the i-th feature. Let be the membership function of the i-th feature, and n be the total number of features.

[0035] In one alternative implementation, the weight coefficients of the features can be set based on an expert assignment method using domain knowledge, with the initial weights directly assigned by battery maintenance experts based on their experience.

[0036] In another alternative implementation, the weighting coefficients of the features can also be calculated based on historical sample data using statistical analysis methods.

[0037] A4: The grinding force is controlled by a closed-loop PID controller, and its control equation is as follows: ; In the formula, The grinding force output by motor 212. Sharpening your skills towards your goal For actual polishing force; To collect data via a pressure sensor. , , These are the proportional, integral, and derivative control parameters, respectively. A5: Calculate the target grinding speed based on the oxidation degree assessment value. The calculation formula is as follows: ; In the formula, Adjust the grinding speed to the target. This refers to the actual grinding speed. and For adjustment coefficients, Contact area; In an optional implementation, the actual grinding speed in step B2 can be acquired by a magneto-electric encoder located at the rear end of the output shaft of motor 212. In another alternative implementation, the actual grinding speed in step B2 can also be estimated using back electromotive force (EMF). When the DC motor 212 rotates, its windings generate a back EMF proportional to the rotational speed. By measuring the voltage and current across the motor 212, the back EMF can be estimated, thereby calculating the rotational speed.

[0038] In one alternative implementation, the contact area in step B2 can be estimated based on the principle of contact mechanics according to a preset drill bit geometry model and the actual grinding force. In another optional implementation, the real-time load current of the contact area grinding motor 212 in step B2 is used as a feedback variable related to the contact area to participate in the adaptive adjustment calculation of the grinding speed.

[0039] A6: Calculate the target polishing time based on the oxidation degree assessment value. The calculation formula is as follows: ; In the formula, To make the most of the time, Time for basic polishing This is an assessment value for the degree of oxidation. The hardness of the battery terminals is obtained by querying a pre-stored database using the battery model. The reference hardness is the process hardness value of the reference material used during system calibration, and it is a pre-stored constant.

[0040] It should be noted that the basic polishing time and reference hardness are preset parameters based on the actual polishing situation; The basic grinding time can be determined in the following way: During the system development phase, using standard grinding tools and under optimized basic process parameters, repeatedly grind standard samples with reference oxidation degree and reference material hardness, and take the statistical average of the time required to grind until the surface quality meets the standard as the basic grinding time. The reference hardness can be the process hardness value of the standard test pole material used when the system is calibrated for the basic grinding time.

[0041] A7: The calculated target grinding force, target grinding time and target grinding speed are output to the core controller. The core controller sets the target grinding force, target grinding time and target grinding speed as control grinding parameters to perform real-time grinding of the pole piece.

[0042] Preferably, step S100, through adaptive image preprocessing, multi-dimensional feature fusion extraction, and fuzzy logic comprehensive evaluation, intelligently maps the oxidation degree assessment value into precise grinding force, speed, and time control parameters, realizing the transformation of the grinding process from experience-based setting to adaptive decision-making. The system employs a dynamic calculation formula based on the oxidation degree, combined with closed-loop PID force control and multi-sensor feedback, ensuring differentiated and high-precision automated grinding of poles with different oxidation states and material hardness. This effectively balances oxide layer removal effect, material loss control, and grinding efficiency, significantly improving the consistency, adaptability, and intelligence level of the process.

[0043] In this embodiment of the application, step 2300 involves controlling the grinding drill bit 211 to grind the electrode post according to the grinding control parameters. After grinding is completed, the grinding quality is evaluated, a grinding quality evaluation result is generated, and the grinding control parameter generation model is updated based on the grinding quality evaluation result, including the following steps B1-B5: B1: Using the principle of optical interference, the thickness of the residual oxide layer on the surface of the polished pole post is determined by analyzing the reflection spectrum. Specifically: ; In the formula, Reflection intensity, Let r1 be the incident light intensity and r1 be the incident light intensity. The interface reflectance coefficient, For phase difference, Wavelength; It should be noted that, and The calculations are based on the theoretical calculations of the material's optical constants. Specifically, the system pre-stores a database of the optical constants of common oxide film materials at different wavelengths, as well as the optical constants of the electrode substrate materials. For the battery model under test, the system calls up the corresponding optical constants according to its electrode material type and typical oxidation products, and directly calculates r1 and r2 using the Fresnel formula mentioned above. .

[0044] The relationship between phase difference and residual oxide layer thickness is as follows: ; In the formula, n is the refractive index of the thin film, and d is the thickness of the residual oxide layer. The angle of incidence is denoted as .

[0045] It should be noted that the polishing is complete when the target polishing time ends.

[0046] B2: Using the laser triangulation principle, the surface roughness of the polished pole piece is evaluated through statistical parameters of height change. Specifically: ; In the formula, The roughness is the arithmetic mean, and L is the measurement length. For height function, This represents the average height.

[0047] Understandably, the average height is the arithmetic mean of the contour data points; the measurement length is the evaluation length for calculating contour statistics, and the height contour data of the selected local area is obtained through visual or laser sensors.

[0048] B3: Evaluate the sanding quality and generate sanding quality evaluation results, specifically: ; In the formula, For the polishing quality score, β1, β2, and β3 are weighting coefficients. The thickness of the residual oxide layer, For material loss, This represents the arithmetic mean roughness.

[0049] In one alternative implementation, the material loss in step C3 can be calculated using the volumetric method. The system calculates the volume of the removed material by measuring the dimensional changes before and after grinding, and then calculates the mass loss by combining this with the known density of the pole material.

[0050] In another optional implementation, the material loss in step C3 can also be obtained by process monitoring integration. The system monitors the pressure of the grinding drill bit 211 in real time through a pressure sensor, and combines it with precisely controlled grinding time, rotation speed and other parameters. Based on the removal rate-pressure-speed model calibrated in advance through experiments, the cumulative removal amount is calculated in real time by integration.

[0051] It should be noted that β1, β2, and β3 were assigned initial weights directly by experts in the field of battery maintenance based on process standards and safety specifications.

[0052] B4: With the goal of maximizing the long-term expected polishing quality score Q, continuously refine the polishing control parameters. The specific steps are as follows: B41: Each complete set of polishing control parameters and its corresponding quality score is stored as a data sample in the system's experience playback buffer. The system uses an online learning method to continuously optimize the polishing parameters. ; In the formula, For the updated parameters, For the current parameter, For learning rate, for The objective function, This is the gradient operator.

[0053] B42: Experience replay mechanism is used to stabilize the learning process: ; In the formula, Here, N is the replay loss function, and N is the number of replay samples. For sample weights, For the target value, These are predicted values.

[0054] It should be noted that N is the number of samples randomly drawn from the experience replay buffer each time the model is updated. This parameter affects the stability and efficiency of learning, and is usually preset to a fixed value based on the buffer size and computing resources.

[0055] This is used to adjust the importance of different samples in the loss calculation. Its settings can follow this strategy: all samples have equal weight, i.e. = 1.

[0056] B5: Then, substitute the corrected grinding control parameters into the grinding control parameter generation model, update the parameters in the formulas involved in the model, and generate a new grinding control parameter generation model.

[0057] It should be noted that the evaluation and optimization steps are seamlessly integrated into the closed-loop control process of the system, forming an intelligent closed loop of execution, evaluation, learning, optimization, and re-execution. Each polishing operation is not only a production task, but also a learning opportunity to provide data for the system optimization model and make it smarter.

[0058] Preferably, step S200 uses integrated optical interferometry to measure oxide film thickness in real time, laser triangulation to assess surface roughness, and combines volumetric or process integration methods to accurately obtain material loss, thus constructing a multi-parameter fusion automated grinding quality assessment system. This system can not only achieve high-precision, non-contact online quality inspection, but also continuously optimize grinding process parameters through feedback data, thereby ensuring effective removal of the oxide layer while accurately controlling material loss, and comprehensively improving the consistency, efficiency, and intelligent level of battery terminal grinding.

[0059] In this embodiment of the application, step S300 involves a quality assessment of the polished battery and outputting the battery quality assessment result, including the following steps C1-C4: C1: Clamp the Kelvin clip-on measuring pen onto the polished and cleaned electrode surface, and inject a specific frequency AC test current through the current electrode. Simultaneously, the response voltage is measured using a Kelvin clip-on measuring pen. The measurement system employs a four-wire Kelvin method to eliminate the influence of lead resistance and contact resistance, thereby obtaining an accurate DC internal resistance value. This value serves as a preliminary assessment of the battery's ohmic internal resistance. The resistance measurement formula is as follows: ; In the formula, The internal resistance of the battery. To measure voltage, For testing current.

[0060] C2: To obtain more comprehensive information about the battery's internal state, the system further employs a multi-frequency AC impedance spectroscopy method. This method injects a series of small AC current signals with different frequencies ω into the battery and measures its voltage response, thereby obtaining the complex impedance Z(ω). ; In the formula, It is a complex impedance; Ω is the ohmic resistance, which characterizes the pure resistance of the electrolyte, electrode materials, and connections. and Polarization resistor, and It is a time constant. ω is the angular frequency.

[0061] By fitting the impedance spectrum, the system can be decoupled and output the aforementioned set of parameters with clear electrochemical and physical significance. , , , , ...}.

[0062] C3: The electrochemical parameters of the battery are significantly affected by temperature. To ensure the comparability of measurement results under different environments, the system performs temperature compensation for all resistive parameters. The compensation algorithm is as follows: + In the formula, This is the resistance value after temperature compensation. The actual measured resistance value, i.e. The temperature coefficient of the battery material. The actual temperature is obtained through measurement by a temperature sensor. The reference temperature is a standard temperature set by those skilled in the art, typically 25°C.

[0063] C4: Outputs the battery's internal resistance, temperature-compensated resistance, and complex impedance as analytical results that can be used to assess battery quality, facilitating operators' analysis of the battery's health status.

[0064] Preferably, step S300 integrates four-wire Kelvin method, multi-frequency AC impedance spectroscopy analysis, and temperature compensation algorithm to achieve accurate and standardized measurement of battery internal resistance from basic ohmic values ​​to in-depth electrochemical parameters. This system not only eliminates contact resistance interference but also obtains core parameters reflecting the battery's internal aging state through impedance spectroscopy analysis. After temperature normalization, it ultimately outputs a multi-level, highly comparable battery electrical performance diagnostic dataset. This provides operators with comprehensive analytical support far exceeding traditional single internal resistance values, enabling precise assessment of battery health, differentiation between internal faults and surface problems, and quantification of the electrical effects of polishing processes.

[0065] The following is an optional implementation of the oxidation degree recognition method. To significantly improve the accuracy and adaptability of oxidation degree recognition, step S100 also employs a deep learning-based method, specifically implemented through a pre-trained convolutional neural network model: The input to the convolutional neural network is the polar cylindrical surface image region after preprocessing in step S100. The mathematical expression of the convolutional neural network structure is as follows: ; In the formula, For network output, Let be the weight matrix of the i-th layer. Let f be the bias vector of the i-th layer, f be the activation function, and x be the input feature.

[0066] Preferably, the weight matrix in the convolutional neural network model With bias vector It is obtained through supervised training of the network, with the goal of making the network output y as close as possible to the real labeled value.

[0067] The loss function design takes into account classification accuracy and regression error: ; In the formula, For the total loss function, For classifying losses, To regress the loss, λ1, λ2, and λ3 are regularization terms, and λ1, λ2, and λ3 are weighting coefficients.

[0068] It should be noted that the weight coefficients are initially determined by expert experience within a given range, and then the final values ​​are determined through automated hyperparameter search. The weight coefficients determined through this process ensure that the convolutional neural network model achieves balanced and excellent performance in both oxidation degree classification and thickness regression tasks; λ1, λ2, and λ3 can be 0.5, 0.4, and 0.1, respectively. In summary, machine vision and algorithms are used to objectively quantify the degree of oxidation and dynamically generate personalized polishing parameters for precise adaptive processing. Simultaneously, high-precision sensing technology is integrated to evaluate polishing geometry and material loss online, and advanced electrochemical impedance spectroscopy analysis and temperature compensation technology are combined to perform in-depth diagnosis of the battery's internal health status. Finally, through data-driven online learning and multi-objective optimization algorithms, the system can continuously optimize process parameters from historical operations, completely transforming terminal maintenance from a crude operation relying on manual experience into a standardized, traceable intelligent process. This significantly improves processing accuracy, efficiency, and consistency, achieving a leap from surface treatment to performance assurance.

[0069] The following is another optional implementation method for adaptive optimization of grinding process parameters. To overcome the contradictions between multiple performance objectives during the grinding process and to achieve adaptive global optimization of process parameters, a multi-objective optimization algorithm can be used as the core, specifically: The optimization process addresses the conflicting objectives of grinding: high quality, high speed, and low material loss. Since a single parameter cannot simultaneously optimize all three, intelligent trade-offs are necessary. This leads to a multi-objective optimization model. The optimization variables in this model are combinations of grinding process parameters; the optimization objectives are maximizing the quality score, minimizing processing time, and minimizing material loss.

[0070] A multi-objective algorithm is used to solve this model. The algorithm works by randomly generating multiple sets of parameters as the initial population. Using the system's prediction model, it evaluates the three objective values ​​corresponding to each set of parameters. The algorithm then selects and sorts solutions according to the "Pareto optimality" principle—that is, retaining those solutions that achieve the best balance among the three objectives. For example, solution A has the highest quality but takes the longest time, while solution B is the fastest but has slightly higher resource consumption; both are feasible optimal balance points. Through simulating the evolutionary process of selection, crossover, and mutation, new parameter combinations are continuously generated and selected, causing the population to evolve towards a better set of balance points. The algorithm ultimately outputs a set of optimal balance solutions, i.e., the Pareto solution set, rather than a single answer. The specific algorithm formula is as follows: ; In the formula, For a multi-objective function vector, Let i be the objective function. Let m be the vector of decision variables, and m be the number of objective functions.

[0071] The Pareto optimal solution set is obtained through non-dominated sorting: ; In the formula, Let S be the Pareto optimal solution and S be the feasible solution set.

[0072] Furthermore, this study constructed a complete simulation model in the MATLAB / Simulink environment to verify the feasibility and effectiveness of the system. The simulation parameters were set based on actual engineering requirements, including battery specifications, oxidation range, and measurement accuracy requirements.

[0073] Experimental data show a comparison of the system's measurement accuracy under different oxidation levels: Table 1: Comparison of measurement accuracy under different oxidation levels Table 2: Evaluation Results of Polishing Effect Simulation results show that the system significantly improves measurement accuracy, processing efficiency, and automation. The measurement error is reduced from ±5.5% of the traditional method to ±1.3%, the processing time is shortened from 25 minutes to 8 minutes, the manual intervention rate is reduced from 100% to 15%, and the robustness and adaptability of the system are also significantly improved, enabling it to meet the processing needs of battery terminals of different types and conditions.

[0074] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0075] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An automatic grinding and evaluation system for storage batteries, characterized in that: include, Control unit (1), the control unit (1) is used to control the grinding state; A polishing unit (2) is used to perform polishing operations on the battery terminals. The polishing unit (2) includes an execution module (21) and a joint module (22). Monitoring unit, the monitoring unit is used to monitor the polishing status; A Kelvin clip-on measuring pen, used to measure the electrical parameters of a polished battery; The core controller, the control unit (1), the grinding unit (2), and the monitoring unit are communicatively connected to the core controller.

2. The automatic grinding and evaluation system for storage batteries as described in claim 1, characterized in that, The joint module (22) includes three sets of rotary joints (222) and one set of pitch joints (221), and the three sets of rotary joints (222) are hinged in sequence.

3. The automatic battery polishing and evaluation system as described in claim 2, characterized in that, The execution module (21) includes a motor (212) and a grinding drill bit (211); the motor (212) is connected to the pitch joint (221), the output shaft of the motor (212) is connected to the grinding drill bit (211), and silicon carbide abrasive particles are attached to the surface of the grinding drill bit (211).

4. The automatic grinding and evaluation system for storage batteries as described in claim 1, characterized in that, The monitoring unit includes a camera for acquiring images of the polishing area and a pressure sensor for acquiring the polishing pressure.

5. A method for evaluating the polishing of a storage battery, using the automatic storage battery polishing evaluation system as described in any one of claims 1-4, characterized in that, include, Based on the grinding control parameter generation model, the grinding control parameters are calculated; The grinding drill bit (211) grinds the pole post according to the grinding control parameters. After grinding is completed, the grinding quality is evaluated, a grinding quality evaluation result is generated, and the grinding control parameter generation model is updated based on the grinding quality evaluation result. The quality of the polished battery is evaluated, and the battery quality evaluation results are output.

6. The method for evaluating the polishing of a storage battery as described in claim 5, characterized in that, The polishing evaluation method also includes: When the target polishing time reaches the preset standard, polishing is stopped and the polishing quality is evaluated.

7. The method for evaluating the polishing of a storage battery as described in claim 5, characterized in that, The controlled grinding device uses PID control to control the grinding force and adaptively adjusts the grinding speed and time according to the degree of oxidation.

8. The method for evaluating the polishing of a storage battery as described in claim 7, characterized in that, The assessment of polishing quality includes: Calculate the thickness of the residual oxide layer on the surface of the electrode post after polishing; Evaluate the surface roughness of the polished pole piece; The grinding quality is evaluated based on the thickness of the residual oxide layer and the roughness of the polished pole surface, and a grinding quality evaluation result is generated.

9. The method for evaluating the polishing of a storage battery as described in claim 5, characterized in that, The model based on grinding control parameters calculates the grinding control parameters as follows: By acquiring images of the electrode posts using a camera, identifying the positions of the electrode posts, and quantitatively evaluating the degree of surface oxidation, an oxidation degree evaluation value is obtained. Based on the oxidation degree assessment value, the polishing control parameters are calculated.

10. The method for evaluating the polishing of a storage battery as described in claim 8, characterized in that, The roughness of the polished pole surface is determined using the principle of laser triangulation and statistical parameters of height variation.

Images