Short circuit protection prediction method for new energy battery with rubber shell

Abstract

The invention provides a short-circuit protection prediction method for a new energy battery with a rubber shell, and the method comprises the steps: carrying out the characterization analysis of the surface and internal microtopography of the rubber shell in a residual stress concentration region through a microscope observation mode, obtaining the morphological characteristics of microcrack formation, and combining with the mechanical performance parameters of a material, including fracture toughness and fracture strain, the mechanical property formed by the microcracks is analyzed; if it is judged that the insulativity of the constructed three-dimensional solid model of the new energy battery rubber shell is reduced, the injection molding process parameters including the injection molding temperature, the dwell time and the cooling rate are optimized, and an optimal process parameter combination is obtained through orthogonal experiment design; a rubber shell injection molding quality reliability evaluation system is constructed, residual stress, microcrack density, mechanical properties and electrical properties are integrated, a fuzzy comprehensive evaluation method is adopted, the quality of rubber shells is quantitatively evaluated, and defective rubber shells are screened and improved.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a short-circuit protection prediction method for a new energy battery with a plastic shell. Background Art

[0002] During the injection molding process of new energy battery with plastic shell, residual stress will be generated inside the plastic shell due to improper setting of injection molding process parameters, such as too high injection pressure and too short cooling time. These residual stresses will gradually release during the use of the plastic shell, causing the size of the plastic shell to change and reducing the mechanical strength of the plastic shell. At the same time, the presence of residual stress will also cause the generation and expansion of microcracks in the microstructure of the plastic shell. Microcracks will continue to accumulate during the battery charge and discharge cycle, eventually leading to cracking of the plastic shell, causing electrolyte leakage and causing safety problems. The presence of microcracks will also affect the insulation performance of the plastic shell, increase the risk of leakage and short circuit, and weaken the short-circuit protection capability of the battery pack. Therefore, how to deeply analyze the generation mechanism of residual stress in the injection molding process, optimize the process parameters, minimize residual stress, and improve the mechanical strength and electrical safety of the plastic shell is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a short circuit protection prediction method for a new energy battery with a plastic shell, which mainly includes:

[0004] Based on the constructed three-dimensional solid model of the new energy battery shell, the finite element simulation analysis method is used to input the mechanical properties parameters of the material, the geometric dimensions of the parts and the injection molding process parameter data, simulate the injection molding process, and obtain the residual stress distribution cloud map inside the shell. The residual stress concentration area is determined through the stress distribution cloud map;

[0005] The surface and internal microscopic morphology of the rubber shell in the residual stress concentration area is characterized and analyzed by microscope observation to obtain the morphological characteristics of microcrack formation, and the mechanical properties of the microcrack formation are analyzed by combining the material mechanical properties parameters, including fracture toughness and fracture strain;

[0006] Through the mechanical property test of the constructed three-dimensional solid model of the new energy battery shell, the residual stress level is determined according to the yield strength of the material, the influence of the residual stress level on the mechanical properties of the shell is analyzed, and the residual stress control threshold is preset. When the residual stress exceeds the residual stress control threshold, it is determined that the mechanical properties of the shell do not meet the requirements;

[0007] Conducting a dielectric constant and volume resistivity electrical performance test method on the constructed three-dimensional solid model of the new energy battery gel shell, evaluating the electrical insulation performance of the constructed three-dimensional solid model of the new energy battery gel shell, and determining the microcrack density control range. If the microcrack density exceeds the range, it is determined that the insulation performance of the constructed three-dimensional solid model of the new energy battery gel shell is reduced;

[0008] If it is determined that the insulation of the constructed three-dimensional solid model of the new energy battery gel shell is reduced, the injection molding process parameters are optimized, including injection temperature, holding time, and cooling rate, and the optimal process parameter combination is obtained through orthogonal experimental design;

[0009] Based on the optimal process parameter combination, the injection molded plastic shell is produced, the injection molded plastic shell is annealed to promote the release of residual stress, low-frequency vibration is applied to the plastic shell to induce residual stress redistribution, reduce stress concentration, inhibit microcrack propagation, and annealing and vibration aging treatment are performed to improve the residual stress state of the plastic shell;

[0010] After annealing and vibration aging treatment, the surface of the rubber shell is scanned by infrared thermal imaging method to identify the local heating area caused by residual stress. Combined with the stress distribution cloud map in the three-dimensional solid model of the rubber shell, the influence of micro-crack extension in the hot spot area on the insulation performance of the battery tab is predicted, the risk of battery short circuit in the hot spot area is evaluated, and the optimization suggestions for the anti-short circuit design are put forward for the rubber shell in the high-risk area;

[0011] A reliability evaluation system for the injection molding quality of plastic shells was established. The residual stress, microcrack density, mechanical properties, and electrical properties were comprehensively considered, and a fuzzy comprehensive evaluation method was used to quantitatively evaluate the quality of plastic shells, so that defective plastic shells could be screened and improved.

[0012] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0013] The present invention discloses a short-circuit protection prediction method for a new energy battery with a plastic shell. The method first analyzes the residual stress distribution in the plastic shell injection molding process through finite element simulation, and determines the microcrack formation characteristics in combination with microscopic observation. Then, mechanical and electrical performance tests are carried out, and a correlation model between residual stress, microcrack density and plastic shell performance is established, and a control threshold is set. When the performance does not meet the requirements, the residual stress state is improved by optimizing the injection molding process parameters, annealing treatment and vibration aging. Infrared thermal imaging technology is used to identify hot spots, evaluate short-circuit risks, and propose optimization suggestions for anti-short-circuit design. Finally, a plastic shell quality reliability evaluation system is constructed, and a fuzzy comprehensive evaluation method is used to quantitatively evaluate and screen the plastic shell quality. The present invention effectively suppresses the extension of microcracks, improves the mechanical properties and electrical insulation properties of the plastic shell, and significantly improves the overall quality reliability of the new energy battery plastic shell by systematically analyzing and controlling residual stress. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flow chart of a short-circuit protection prediction method for a new energy battery with a plastic shell.

[0015] Figure 2 It is a schematic diagram of a short-circuit protection prediction method for a new energy battery with a plastic shell according to the present invention.

[0016] Figure 3 It is another schematic diagram of a short-circuit protection prediction method for a new energy battery with a plastic shell according to the present invention. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0018] like Figure 1 -3. In this embodiment, a short circuit protection prediction method for a new energy battery with a plastic shell may specifically include:

[0019] S101. Based on the constructed three-dimensional solid model of the new energy battery plastic shell, the finite element simulation analysis method is adopted, the mechanical property parameters of the material, the geometric dimensions of the parts and the injection molding process parameter data are input, the injection molding process is simulated, and the residual stress distribution cloud map inside the plastic shell is obtained. The residual stress concentration area is determined through the stress distribution cloud map.

[0020] The size data of the shell structure in the three-dimensional solid model are collected, and the shell elastic deformation curve data are generated according to the calibration measurement data through calibration measurement of the material elastic coefficient and Poisson's ratio; the injection melt temperature, injection pressure and injection rate data are collected according to the shell elastic deformation curve data, load boundary conditions are applied to the injection station sampling points, and stress calculation input data are obtained by multivariate linear regression; the finite element meshing tool is called according to the stress calculation input data, the three-dimensional solid model is meshed, and the unit node stress numerical distribution data set is obtained based on Hooke's law; stress cloud map data are constructed according to the stress numerical distribution data set, the principal stress value is calculated by using the vonMises stress criterion, the residual stress cloud map data is obtained by calibrating the stress peak point coordinates, and the residual stress concentration area is determined by the stress distribution cloud map.

[0021] Exemplarily, the size data of the plastic shell structure is collected from the three-dimensional solid model, the elastic coefficient and Poisson's ratio of the material are calibrated and measured, and the plastic shell measurement position points are sampled according to the material test standard, and the plastic shell elastic deformation curve data is generated by least squares fitting. According to the plastic shell elastic deformation curve data, the injection melt temperature, injection pressure, and injection rate data recorded by the process control unit are collected, and the load boundary conditions are applied to the injection molding station sampling points. The stress calculation input data is generated by multivariate linear regression for the process parameter data. According to the stress calculation input data, the finite element meshing tool is called in the control unit to mesh the three-dimensional solid model, and the unit node stress is calculated based on Hooke's law to generate a stress numerical distribution data set. In the injection molding simulation station, the stress cloud map data is constructed according to the stress numerical distribution data set, the principal stress is calculated using the vonMises stress criterion, the stress peak point coordinates are calibrated, and the residual stress cloud map data is generated. In the three-dimensional solid model collection, the size data of the plastic shell structure includes key features such as wall thickness, fillet radius, and rib distribution. The measurement points are usually collected at the battery plastic shell connection, the load-bearing body, and the edge weak area. When calibrating the elastic coefficient of the material, the elastic modulus of the polymer material was measured at 23°C, and the Poisson's ratio was 2000MPa and 0.35, respectively. These basic mechanical parameters constitute the basic data for subsequent injection molding simulation. The injection molding process involves three main control parameters: melt temperature, mold pressure, and injection rate. In the injection molding station, the temperature sensor records the melt temperature of 220°C, the pressure sensor collects the injection pressure of 15MPa, and the injection rate is controlled at 25cm 3 / s. The sampling points of the workstations are arranged at the cavity feed port, gate, and runner end. Boundary constraints are imposed at these locations and load data are recorded. In the finite element simulation of the injection molding process, hexahedral units are used for mesh division, and the mesh size is set to 2mm. The area with a curvature greater than 60 degrees is locally encrypted, and the mesh size after encryption is 0.5mm. By calculating the stress of the grid nodes, the stress distribution value of the shell wall is obtained. The stress calculation is based on the linear elastic constitutive equation, considering the influence of the temperature field on the material properties. In the injection molding process simulation stage, the stress cloud map is represented by a multi-color ribbon, the red area represents the high stress area, and the blue area represents the low stress area. The stress cloud map data shows that stress concentration occurs at the corners of the shell and the root of the reinforcement. The maximum stress value reaches 35MPa, and the coverage area of ​​the stress concentration area accounts for 5% of the surface area. According to the vonMises stress criterion, the main stress direction is at an angle of 45 degrees to the direction of the reinforcement, and the stress peak points are mainly distributed in the transition area of ​​the shell corner. During the injection molding process, the material shrinks and deforms from the melt to the solidification process, with a shrinkage rate of 1.5%, which causes residual stress in the gel shell. The residual stress distribution cloud map shows that there are large stress concentrations in the area where the gel shell wall thickness changes, the gate position, and the rib connection. Through the stress distribution law, high-risk areas with stress exceeding 25MPa are identified. These areas are prone to cracking or deformation during subsequent use. The injection molding simulation results show that the internal stress distribution of the battery gel shell is closely related to the material flow behavior. During the cavity filling process, the melt temperature gradient and pressure distribution unevenness have a significant effect on the residual stress distribution. When the injection rate is increased from 20cm 3 / s increased to 30cm 3 / s, the residual stress peak value increases by 20%, and the stress concentration area expands by 1.5 times. When the mold temperature is controlled at 80℃, the residual stress distribution of the plastic shell molded part is more uniform, and the stress peak value is reduced by 15%.

[0022] S102. Use microscope observation to characterize and analyze the surface and internal microscopic morphology of the rubber shell in the residual stress concentration area, obtain the morphological characteristics of microcrack formation, and combine the material mechanical property parameters, including fracture toughness and fracture strain, to analyze the mechanical properties of microcrack formation.

[0023] The surface morphology image of the rubber shell is collected by a microscopic scanning device, and the microcrack boundary contour curve data is obtained by using a Sobel edge extractor according to the morphology image; a strain sampling point array is arranged within the gauge range for the microcrack boundary contour curve data, and the corresponding relationship between the fracture strain value and the microcrack boundary position is recorded according to the strain sampling point array to obtain a strain data field distribution diagram; the microcrack morphology characteristics are segmentedly extracted by using an edge detection operator according to the strain data field distribution diagram, and the crack depth value, crack elongation rate, and crack orientation angle are calculated according to the microcrack morphology characteristics to obtain a crack morphology characteristic parameter matrix; the crack tip opening displacement value, stress intensity factor, and fracture toughness are calculated according to the crack morphology characteristic parameter matrix in combination with the generalized fracture criterion, and the normal stress component and the shear stress component are calibrated in the micro-region fracture unit through the opening displacement value to obtain the crack extension mechanical parameter field.

[0024] Exemplarily, the surface morphology image of the rubber shell is collected in a preset area according to the microscopic scanning device, the microcrack boundary is depicted by the Sobel edge extractor, and the microcrack contour points are fitted by cubic spline interpolation to generate crack boundary contour curve data. According to the material tensile test standard, the fracture area of ​​the rubber shell is precisely loaded, and 128x128 dot matrix strain sampling points are arranged within the gauge range, and the corresponding relationship between the fracture strain value and the microcrack boundary position is recorded to generate a strain data field distribution map. In the strain data field distribution map, the edge detection operator is used to extract the microcrack morphological features in sections, and the crack depth value, crack elongation, and crack orientation angle are calculated from them to construct a crack morphological characteristic parameter matrix. According to the crack morphological characteristic parameter matrix, the crack tip opening displacement value, stress intensity factor, and fracture toughness are calculated in combination with the generalized fracture criterion, and the normal stress component and shear stress component are calibrated in the micro-region fracture unit to generate a crack extension mechanical parameter field. In microscopic scanning, the microcrack morphology was observed by a 1000x microscope, the scanning area was set to 10mm×10mm, the scanning step was 0.1μm, and the image resolution reached 2048×2048 pixels. When Sobel edge extraction was used to depict the microcrack boundary, the grayscale threshold was set to 128, and the detected crack width ranged from 0.5μm to 5μm, and the length distribution ranged from 50μm to 500μm. When cubic spline interpolation was used to fit the crack profile, the curvature change point was selected as the feature point, and the interpolation node spacing was 0.2μm. During the tensile loading of the material, the loading rate was controlled at 0.5mm / min, and 128×128 dot matrix strain sampling points were arranged within a 10mm gauge length, with a point spacing of 0.078mm. The fracture strain value reached 15% in the tensile fracture area, the strain was concentrated at the tip of the microcrack, and the local strain value exceeded 25%. The strain data field distribution diagram shows that a butterfly-shaped plastic deformation zone is formed around the microcrack, and the size of the plastic zone is about twice the length of the crack. In the analysis of the microcrack morphology, the edge detection operator uses a 5×5 pixel window, and the crack depth value is distributed between 15% and 45% of the material thickness. The crack elongation rate shows an exponential growth characteristic, increasing from the initial 0.1μm / s to 1.0μm / s. The angle between the crack orientation angle and the main stress direction is mostly distributed between 85 degrees and 95 degrees, showing a quasi-vertical expansion trend. In the micro-region fracture unit analysis, the crack tip opening displacement value is parabolic with the increase of the distance from the crack tip, and the maximum opening value is 2.5μm. The stress intensity factor calculation adopts the displacement extrapolation method, and the critical stress intensity factor value is 2.5MPa·m^0.5. The fracture toughness measurement value is 15kJ / m 3, the maximum value of the positive stress component appears at 0.2mm in front of the crack tip, which is 350MPa, and the peak value of the shear stress component is 175MPa. Under the action of cyclic load, the microcrack extension of the battery shell material shows fatigue characteristics, and the crack extension rate and the amplitude of the stress intensity factor satisfy the power function relationship. When the cyclic load frequency is 10Hz, the crack extension rate is 0.01μm / cycle. The microcrack extension path is affected by the microstructure of the material and deflects at the grain boundary with a deflection angle between 15 and 35 degrees. During the formation of microcracks, the fracture surface morphology shows typical cleavage fracture characteristics, and the cleavage step height is between 0.5μm and 2μm. The fracture surface roughness Ra value is 0.8μm, and obvious fatigue bands are observed in the crack initiation area, with a band spacing of 0.2μm. The secondary cracks branch from the main crack surface, with a bifurcation angle between 30 and 60 degrees, forming a network crack morphology.

[0025] S103. Perform mechanical property tests on the constructed three-dimensional solid model of the new energy battery shell, determine the residual stress level according to the yield strength of the material, analyze the influence of the residual stress level on the mechanical properties of the shell, preset the residual stress control threshold, and when the residual stress exceeds the residual stress control threshold, it is determined that the mechanical properties of the shell do not meet the requirements.

[0026] Orthogonal loading points are arranged on the surface of the rubber shell according to the three-dimensional solid load loader, and the displacement field data of the rubber shell surface is collected by a digital image correlation optical strain meter, and a uniaxial tensile load is applied to the loaded area to obtain the rubber shell strain distribution data; the tensile strain value, the compressive strain value and the shear strain value of the measuring point are extracted from the rubber shell strain distribution data, and the linear segment value of the material is calibrated by a tensile stress-strain curve calculator to obtain the yield point stress-strain curve; for the yield point stress-strain curve, the shear modulus and the bulk modulus of the rubber shell material are solved by a Lame parameter calculator, and the residual stress value and the stress concentration factor are calculated to obtain the residual stress distribution data; according to the residual stress distribution data, a numerical comparator is used to compare the residual stress value with a preset stress threshold, and the residual stress peak point position and the stress level are calibrated to obtain the mechanical property judgment result.

[0027] Exemplarily, 8 orthogonal loading points are arranged on the surface of the shell according to the three-dimensional solid load loader, and the displacement field data of the shell surface is collected by the digital image correlation optical strain meter, and a uniaxial tensile load is applied to the loading area to generate the shell strain distribution data. The tensile strain value, compressive strain value, and shear strain value of each measuring point are extracted from the shell strain distribution data, and the linear segment value of the material is calibrated by the tensile stress strain curve calculator to generate the yield point stress strain curve. On the basis of the yield point stress strain curve, the shear modulus and bulk modulus of the shell material are solved by the Lame parameter calculator, and the residual stress value and stress concentration factor are calculated to generate the residual stress distribution data. According to the residual stress distribution data, the residual stress value is compared with the preset stress threshold by a numerical comparator, and the residual stress peak point position and stress level are calibrated to generate the mechanical property judgment result. In the shell mechanical property test, the load loader evenly arranges 8 orthogonal loading points on the shell surface, and the loading point spacing is 20mm, forming a complete strain measurement network. When the digital image correlation optical strain meter collects the surface displacement field, the resolution reaches 0.1μm, and the measurement area covers a range of 100mm×100mm. The uniaxial tensile load increases from 0 to 5000N, with a sampling interval of 100N, and the strain field evolution process is recorded. In the strain distribution measurement, the tensile strain value increases from 0 to 0.2% in the linear segment, the compressive strain value reaches 0.1% in the transverse direction, and the maximum shear strain is 0.15%. The stress-strain relationship of the linear segment of the material shows good linear characteristics, and the slope corresponds to an elastic modulus of 2200MPa. When the strain exceeds 0.2%, the stress-strain curve begins to deviate from the straight line, indicating that the material enters the yield stage. Material parameter calculation shows that the shear modulus of the shell material is 850MPa and the bulk modulus is 2350MPa. The residual stress distribution on the shell surface shows uneven characteristics, and stress concentration occurs at the corners and the root of the stiffener, and the stress concentration factor reaches 2.3. The maximum residual stress occurs at the connection between the stiffener and the matrix, reaching 42MPa. The residual stress distribution law shows that there is obvious stress concentration in the area where the thickness of the shell changes suddenly, at the connection of the reinforcement ribs, and at the corner transition area. When the preset stress threshold is set to 35MPa, the residual stress in the local area of ​​the shell exceeds the standard. The stress exceeding area is mainly distributed at the root of the reinforcement rib, accounting for 3.5% of the total area. The position of the residual stress peak point corresponds to the geometric discontinuity of the structure. After the battery shell is injection molded, the residual stress level inside the material directly affects its mechanical properties. The process parameters have a significant effect on the residual stress distribution. When the injection pressure increases from 40MPa to 60MPa, the average residual stress increases by 25%. The mold temperature also has an important influence on the residual stress. When the mold temperature increases from 60℃ to 80℃, the residual stress decreases by 20%. The mechanical properties test results show that when the residual stress exceeds 35% of the yield strength, plastic deformation begins to occur in the local area of ​​the material.Under cyclic loads, high residual stress areas are prone to stress concentration, leading to fatigue crack initiation. When the residual stress reaches 45% of the yield strength, the fatigue life of the material is reduced by 50%. Under temperature cycling conditions, the residual stress and thermal stress are superimposed, exacerbating the damage evolution process of the material. In the temperature cycle from -40°C to 60°C, the residual stress changes with temperature by ±15MPa.

[0028] S104. Conducting electrical performance testing methods of dielectric constant and volume resistivity on the constructed three-dimensional solid model of the new energy battery shell to evaluate the electrical insulation performance of the constructed three-dimensional solid model of the new energy battery shell and determine the control range of microcrack density. If the microcrack density exceeds the range, it is determined that the insulation performance of the constructed three-dimensional solid model of the new energy battery shell is reduced.

[0029] According to the three-dimensional entity test platform, test electrodes are arranged on the surface of the rubber shell, the surface dielectric constant value is collected by a capacitance meter, and the volume resistivity value is recorded by a high-voltage DC resistance meter to obtain basic electrical performance data; the dielectric constant distribution value and the volume resistivity distribution value are extracted from the basic electrical performance data, and the insulation strength of the material is scanned and measured by a voltage-current characteristic curve instrument to obtain insulation performance evaluation data; for the insulation performance evaluation data, a regional growth segmenter is used to perform block processing on the microcrack image, and the number of cracks per unit area and the crack surface area are calculated by a standard convolutional neural network to generate microcrack density feature data; according to the microcrack density feature data, the microcrack feature value and the insulation performance parameter are mapped by a correlation matrix calculator to determine the calibration result of the insulation strength weakening area.

[0030] Exemplarily, 8 test electrodes are arranged on the surface of the plastic shell according to the three-dimensional entity test platform, and an AC voltage is applied to the test point at a frequency of 1 kHz through a capacitance meter, and the surface dielectric constant value is collected synchronously. The volume resistivity value of each point is recorded at 1 kilovolt using a high-voltage DC resistance meter to generate basic electrical performance data. The dielectric constant distribution value, dielectric loss angle, and volume resistivity distribution value are extracted from the basic electrical performance data, and the material insulation strength is scanned and measured by a voltage and current characteristic curve instrument to generate insulation performance evaluation data. Based on the insulation performance evaluation data, the microcrack image is processed in blocks using a regional growth segmenter, and the number of cracks per unit area, the crack surface area, and the crack size spectrum are calculated by a standard convolutional neural network to generate microcrack density feature data. According to the microcrack density feature data, the microcrack feature value is mapped to the insulation performance parameter through an association matrix calculator, the insulation strength weakening area is calibrated, and the insulation strength determination result is generated. In the electrical performance test, the test electrode adopts an 8-point arrangement method, the electrode diameter is 10mm, and the surface is silver-plated to improve conductivity. When a 1kHz AC voltage is applied, the voltage amplitude is 100V. Data is collected through a capacitance meter, and the measured dielectric constant is distributed between 2.8 and 3.2. During the high-voltage DC test, a 1kV voltage is applied, and the volume resistivity measurement value reaches 10^14Ω·cm, indicating that the material has good insulation properties. During the dielectric performance evaluation process, the dielectric constant distribution shows anisotropic characteristics, and the amplitude of variation in the thickness direction of the shell is ±0.2. The measured value of the dielectric loss angle is 0.02, reflecting that the energy loss of the material is small under the alternating electric field. The volume resistivity distribution diagram shows that the resistivity value in the microcrack area drops to 10^12Ω·cm. The step voltage method is used for insulation strength measurement, and the breakdown voltage reaches 25kV / mm. In the microcrack feature extraction, the regional growth segmenter sets the growth threshold to 8 pixels, and the crack boundary grayscale gradient is greater than 50. The standard convolutional neural network identifies the number of cracks per unit area as 5 / mm 3 , the crack surface area accounts for 2.5%. Crack size spectrum analysis shows that the length distribution is between 20μm and 200μm, and the width ranges from 2μm to 10μm. In the insulation performance evaluation, the correlation matrix calculation shows that the microcrack density is negatively correlated with the dielectric constant, with a correlation coefficient of -0.85. When the crack density exceeds 3 / mm 3When the temperature rises to 80℃, the dielectric constant increases by 0.5, and the dielectric loss angle increases to 0.035. Microcracks are prone to form conductive channels in high temperature and high humidity environments, accelerating the degradation of insulation performance. The insulation strength test results show that crack orientation has a significant effect on the breakdown path. Cracks perpendicular to the electric field have little effect on the insulation strength, while cracks parallel to the electric field are prone to breakdown. The electric field enhancement effect at the end of the crack reduces the local breakdown voltage by 30%. Under the action of the alternating electric field, local discharge at the microcracks accelerates the degradation of insulation performance, and the local temperature rises to 15℃. When the microcrack network forms a conductive path, the leakage current density increases significantly. Under standard voltage, the leakage current density in the healthy area is 0.1μA / cm 3 , while the crack-intensive area increased to 1.5μA / cm 3 During long-term operation, the corrosion effect caused by partial discharge further enlarges the crack size, forming a cyclic effect of deteriorating insulation performance.

[0031] The dielectric constant and volume resistivity of the plastic shell material are measured using a dielectric constant tester and a high resistance meter. Based on the electrical insulation performance standard, it is determined whether the plastic shell material meets the insulation requirements. The number and distribution of internal microcracks are counted, and the correlation between the microcrack density and insulation performance is analyzed to obtain the control range and threshold of the microcrack density of the plastic shell material.

[0032] An AC voltage and a DC voltage are applied to the surface of the rubber shell by using a testing instrument to obtain the capacitance value and the resistance value, and the comprehensive electrical characteristic data is obtained according to the breakdown voltage scan; the dielectric constant value and the dielectric loss angle are extracted according to the comprehensive electrical characteristic data, and the insulation performance characteristic data are obtained by comparing them with the preset insulation standard value through a numerical comparator; the regional growth algorithm is run on the insulation performance characteristic data to extract the number of cracks per unit area and the crack extension length, and the insulation strength characteristic matrix is ​​constructed through a convolutional neural network; the correlation value between the crack density parameter and the insulation strength parameter is calculated according to the insulation strength characteristic matrix, and if the correlation value exceeds the preset standard threshold, the microcrack control interval data is determined.

[0033] Exemplarily, according to the standard measurement specification, the test electrodes are arranged in an 8-point array on the surface of the plastic shell, and the capacitance value is collected by applying a 1 kHz AC voltage through a dielectric constant tester, and a 1 kilovolt DC voltage is simultaneously applied by a high resistance meter to record the resistance value, and the breakdown voltage is step-scanned to generate comprehensive electrical characteristic data. The dielectric constant value, dielectric loss angle, volume resistivity value, and breakdown field strength value are extracted from the comprehensive electrical characteristic data, and compared with the preset insulation standard value through a numerical comparator, and the insulation strength attenuation curve is collected to generate insulation performance characteristic data. Based on the insulation performance characteristic data, the microcrack morphology is divided into zones and counted using a regional grower, and the number of cracks per unit area, the crack extension length, and the crack penetration depth are extracted from them, and the insulation strength characteristic matrix is ​​constructed through a convolutional neural network. According to the insulation strength characteristic matrix, the correlation coefficient calculator is used to calculate the correlation between the crack density parameter and the insulation strength parameter, and the crack density control upper limit and lower limit are calibrated based on the electrical insulation standard to generate microcrack control interval data.

[0034] , ρ represents the Pearson correlation coefficient between the crack density parameter and the insulation strength parameter, Xi represents the crack density of the ith observation, Yi represents the insulation strength of the ith observation, and Represent the sample means of crack density and insulation strength respectively. In the electrical performance test, the test electrodes are arranged in an 8-point array with an electrode spacing of 25mm, covering the main test area of ​​the rubber shell. When a 1kHz AC voltage is applied, the voltage amplitude is 100V, and the capacitance value measured by the dielectric constant tester is between 100pF and 120pF. The high resistance meter applies a 1kV DC voltage, and the measured resistance value reaches 10^14Ω, and the breakdown voltage scanning range increases from 1kV to 30kV. The insulation performance parameters show that the dielectric constant of the rubber shell material is 2.8, the dielectric loss angle is 0.02, and the volume resistivity is 10^14Ω·cm. The breakdown field strength test adopts the step voltage method, the voltage increment is 1kV / 30s, and the breakdown field strength is 25kV / mm. The insulation strength decay curve shows nonlinear characteristics, and the decay rate is accelerated in the microcrack area. In the microcrack morphology analysis, the regional growth algorithm sets the grayscale threshold to 128 and the growth step to 2 pixels. The number of cracks per unit area is 5 / mm 3 The crack extension length is distributed between 50μm and 500μm, and the penetration depth reaches 25% of the material thickness. The feature matrix extracted by the convolutional neural network contains 16×16 feature points, each of which records the crack morphological parameters. The correlation analysis between insulation strength and crack density shows that the two show a significant negative correlation, with a correlation coefficient of -0.85. When the crack density exceeds 3 / mm 3 When the local insulation strength drops by 30%, the crack density is controlled at 4 / mm based on the electrical insulation standard. 3, the lower limit is 0.5 / mm 3 . In practical applications, the electrical properties of the plastic shell material are significantly affected by environmental conditions. When the humidity increases from 45% to 85%, the volume resistivity decreases by 2 orders of magnitude. When the temperature rises to 80°C, the dielectric constant increases by 0.3 and the dielectric loss angle increases to 0.035. Microcracks are more likely to form conductive paths in high temperature and high humidity environments. Insulation breakdown tests show that the presence of microcracks changes the electric field distribution. At the crack tip, the local electric field strength increases by 2.5 times, which promotes the formation of electrical trees. When the crack direction is consistent with the electric field direction, the breakdown voltage is reduced by 40%. Under an alternating electric field, local discharge at the crack accelerates material aging. The effect of microcrack density on leakage current is significant. Under standard voltage, the leakage current density of the complete area is 0.1μA / cm 3 , while the crack area rises to 1.2μA / cm 3 As the operating time increases, the material erosion caused by partial discharge further expands the crack size, forming a vicious cycle of insulation degradation. The decay rate of electrical insulation performance is exponentially related to the crack growth rate.

[0035] S105. If it is determined that the insulation of the constructed three-dimensional solid model of the new energy battery gel shell is reduced, the injection molding process parameters are optimized, including injection temperature, holding time, and cooling rate, and the optimal process parameter combination is obtained through orthogonal experimental design.

[0036] According to the injection molding specification, a process parameter calculator is used to arrange the injection melt temperature range, holding time range, and cooling rate range in combination, and the material viscosity value and crystallinity value corresponding to each group of parameters are calibrated through the physical property parameter database to obtain an injection molding process parameter matrix; each group of process configuration parameters are extracted from the injection molding process parameter matrix, and the cavity pressure sensing point, temperature sensing point, and flow front sensing point are recorded in real time in the injection molding controller to obtain an injection molding process parameter set; for the injection molding process parameter set, a gradient boosting tree is used to extract features from parameter curve data, and the insulation contribution value and parameter coupling index of a single parameter are calculated to generate a process parameter influencing factor set; according to the process parameter influencing factor set, a parameter response surface is constructed through a central composite designer, and the parameter space is traversed and searched under the constraint of optimizing insulation performance, and the optimal value of the melt temperature, the optimal value of the holding time, and the optimal value of the cooling rate are calibrated.

[0037] Exemplarily, the injection molding parameter interval is set in a three-factor four-level orthogonal table according to the injection molding specification, and the injection molding melt temperature interval, the holding time interval, and the cooling rate interval are combined and arranged by a process parameter calculator. The material viscosity value and crystallinity value corresponding to each group of parameters are calibrated through the physical property parameter database to generate an injection molding process parameter matrix. Each group of process configuration parameters is extracted from the injection molding process parameter matrix, and the cavity pressure sensing point, temperature sensing point, and flow front sensing point are recorded in real time in the injection molding controller. The parameter curve data and insulation performance data in the plastic shell molding process are collected to generate an injection molding process parameter set. Based on the injection molding process parameter set, the gradient boosting tree is used to extract the characteristics of the parameter curve data, and the insulation contribution value, parameter coupling effect index, and physical property parameter balance point of a single parameter are calculated to generate a process parameter influencing factor set. According to the process parameter influencing factor set, a parameter response surface is constructed through a central composite designer, and the parameter space is traversed and searched under the constraint of optimal insulation performance, and the optimal value of the melt temperature, the optimal value of the holding time, and the optimal value of the cooling rate are calibrated to generate a process parameter collaborative configuration scheme. In the optimization of injection molding process parameters, the three-factor four-level orthogonal design includes three key parameters: melt temperature, holding time, and cooling rate. The melt temperature range is set from 220℃ to 260℃, and a level is taken every 10℃. The holding time is from 5s to 20s, and the interval is 5s. The cooling rate is between 10℃ / s and 40℃ / s, and the level value is set at an interval of 10℃ / s. The material viscosity is 280Pa·s at a temperature of 220℃, and it drops to 180Pa·s when the temperature rises to 260℃. During the parameter collection process, the cavity pressure sensing points are arranged at the gate, the end of the runner, and the corner, and the sampling frequency is 100Hz. The pressure in the cavity rises from 0MPa at the beginning of injection to 40MPa in the holding stage, and drops to 0MPa at the end of the molding cycle. The temperature sensing point records show that the local cooling rate difference reaches 15℃ / s during the process of the melt temperature dropping from 260℃ to 80℃. The filling time recorded by the flow front sensor is 2.5s. The analysis of the influence of process parameters shows that the melt temperature has the most significant effect on the crystallinity of the material, with a contribution of 45%. The interaction between the holding time and the cooling rate is obvious, and the coupling index is 0.72. When the melt temperature is lower than 230°C, the material viscosity is too high, resulting in insufficient filling. When the temperature is higher than 250°C, the thermal degradation of the material is aggravated, and the insulation performance decreases by 15%. The response surface optimization results show that under the optimization constraint of insulation performance, the optimal melt temperature is 240°C, the material viscosity is 220Pa·s at this temperature, and the crystallinity reaches 32%. The optimal holding time is 15s, and the holding pressure is maintained at 35MPa. The optimal cooling rate is 25°C / s, and the total duration of the molding cycle is controlled within 45s. The insulation performance of the plastic shell is closely related to the molding process, and the crystallinity of the material is the key bridge connecting the process parameters and the insulation performance. When the crystallinity is lower than 25%, an amorphous region is formed inside the material, and the insulation resistance is reduced by 2 orders of magnitude.When the crystallinity is too high and exceeds 40%, the internal stress of the material is concentrated and microcracks are easily formed. During the injection molding process, the control of the holding time directly affects the residual stress level inside the product. When the holding time is extended from 5s to 20s, the residual stress is reduced by 35%. The uniformity of the holding pressure plays a key role in the deformation of the product. When the unevenness of the pressure distribution exceeds 20%, the warpage deformation of the product increases by 0.5mm. The cooling rate has a significant effect on the internal structure of the material. When the rate is lower than 15℃ / s, the crystallization is sufficient but the cycle is extended, and the production efficiency is reduced by 30%. When the rate exceeds 35℃ / s, the inner and outer layers are cooled unevenly, shrinkage stress is generated, and the density of microcracks increases by 2 times. Uniform cooling plays a key role in reducing defects and improving insulation performance.

[0038] S106. Produce injection molded plastic shells based on the optimal process parameter combination, anneal the injection molded plastic shells to promote residual stress release, apply low-frequency vibration to the plastic shells to induce residual stress redistribution, reduce stress concentration, inhibit microcrack growth, and perform annealing and vibration aging treatments to improve the residual stress state of the plastic shells.

[0039] According to the process control parameters, a temperature sensor array is arranged on the surface of the rubber shell, and a temperature collector is used to obtain temperature field distribution and crystallinity change data to obtain an annealing temperature change curve; the heating rate and insulation time parameters are extracted through the annealing temperature change curve, and a stress relaxation calculator is used to numerically solve the displacement field inside the rubber shell to obtain a stress relaxation field distribution diagram; according to the stress relaxation field distribution diagram, an acoustic exciter is used to apply a frequency scanning signal to the rubber shell, and the frequency scanning signal is processed by a neural network fitter to obtain stress redistribution data; according to the stress redistribution data, a vibration sensor array is arranged in the stress concentration area of ​​the rubber shell, and the vibration sensor array acquisition signal is processed by a phase delay calculator to obtain a microcrack growth rate and a stress intensity factor.

[0040] Exemplarily, a 16-point temperature sensor array is arranged on the surface of the plastic shell according to the process control parameters, and the temperature field distribution and the change of material crystallinity are recorded by a temperature collector. The overall temperature field is calibrated by an infrared thermal imager, and the crystallization transition point data of the plastic shell is marked within the standard annealing cycle to generate an annealing temperature change curve. The heating rate, the insulation time, and the crystallinity change value are extracted from the annealing temperature change curve, and the internal displacement field of the material is numerically solved by a stress relaxation calculator. The stress gradient change value is recorded during the annealing process to generate a stress relaxation field distribution map. On the basis of the stress relaxation field distribution map, an acoustic exciter is used to apply a frequency scan to the plastic shell, and the displacement response and strain distribution are recorded by frequency band from low frequency to high frequency. The neural network fitter is used to construct the stress field evolution law and generate stress redistribution data. According to the stress redistribution data, a vibration sensor array is arranged in the stress concentration area of ​​the plastic shell, and the stress wave propagation law is extracted by a phase delay calculator, and the microcrack growth rate and stress intensity factor are calibrated to generate stress control optimization data. During the annealing process, the temperature sensor array was arranged in a 4×4 matrix on the surface of the plastic shell, and the spacing between the sensing points was 25mm. The annealing temperature was increased from room temperature to 80℃, and the heating rate was controlled at 2℃ / min. The crystallinity of the material increased from 28% to 35% with the increase of temperature, and thermal imaging showed that the uniformity of the temperature field was within the range of ±2℃. The standard annealing cycle was set to 4 hours, including 1 hour of heating, 2 hours of insulation, and 1 hour of cooling. The stress relaxation process showed that the internal displacement field of the material showed obvious time dependence during the 80℃ insulation stage. The displacement field calculation results showed that the residual stress decreased by 45% after 2 hours of insulation. The stress gradient value decreased from the initial 5MPa / mm to 1MPa / mm, and the stress field distribution tended to be uniform. The reorganization of the internal structure of the material further increased the crystallinity to 38%. In the acoustic excitation test, the frequency was scanned from 10Hz to 200Hz, and the amplitude was set to 0.5mm. In the low frequency band below 50Hz, the displacement response is in phase with the excitation signal, and the strain distribution is uniform. When the frequency rises to 150Hz, a resonance peak appears, and the displacement amplification factor reaches 2.5. The stress evolution law fitted by the neural network shows that the vibration process redistributes stress and reduces the local stress concentration by 35%. The vibration sensor array is arranged in the stress concentration area, and the sensor spacing is 10mm. Phase delay calculation shows that the propagation speed of stress waves in the material is 1200m / s. The expansion rate of microcracks under vibration is reduced by 80%, and the stress intensity factor is reduced from 1.2MPa·m^0.5 to 0.6MPa·m^0.5. The synergistic effect of annealing and vibration aging is reflected in many aspects. The annealing process promotes the movement of molecular chains and reduces the overall stress level. Vibration loading breaks the local stress concentration through the propagation of stress waves. The increase in crystallinity enhances the overall performance of the material and reduces the tendency of microcracks to expand. The uniformity of temperature field distribution has a significant effect on the stress release effect. When the temperature field inhomogeneity exceeds 5℃, new thermal stress is generated, which in turn aggravates the stress concentration.When the heating rate is too fast and exceeds 5℃ / min, the temperature gradient between the surface and the inside of the material increases, forming a new stress field. The choice of vibration frequency is closely related to the natural frequency of the material. When vibration is applied near the natural frequency, resonance occurs inside the material, promoting stress redistribution. When the amplitude is too large and exceeds 1mm, fatigue damage to the material will occur. When the frequency is too high and exceeds 180Hz, the damping of the material increases, and it is difficult for the vibration energy to be transmitted to the inside. The microcrack suppression effect is directly related to the evolution of the stress field. After stress redistribution, the stress intensity at the crack tip is reduced, which inhibits crack propagation. The internal structure of the material is reorganized, the fracture toughness is improved, and the ability to resist crack propagation is enhanced. After annealing and vibration aging treatment, the service life of the material is significantly extended.

[0041] S107. After annealing and vibration aging treatment, the surface of the rubber shell is scanned by infrared thermal imaging method to identify the local heating area caused by residual stress. Combined with the stress distribution cloud map in the three-dimensional solid model of the rubber shell, the influence of microcrack extension in the hot spot area on the insulation performance of the battery tab is predicted, the risk of battery short circuit in the hot spot area is evaluated, and the optimization suggestions for the anti-short circuit design are put forward for the rubber shell in the high-risk area.

[0042] According to the scanning data obtained by the thermal imaging acquisition device, the surface temperature of the rubber shell is calibrated by an infrared detector, and the infrared detector uses an area integrator to calculate the hot zone coverage to obtain the hot zone distribution characteristic data; the temperature gradient field and the thermal stress field are extracted from the hot zone distribution characteristic data, and the stress distribution cloud map is partitioned and matched by the regional growing algorithm to obtain a thermal stress coupling distribution map; for the thermal stress coupling distribution map, a multi-layer perceptron is used to perform feature recognition on the microcrack growth path, and the distance between the tab area and the crack tip is calculated in the spatial coordinate system to obtain the microcrack evolution prediction data; according to the microcrack evolution prediction data, a resistance threshold detection dot matrix is ​​arranged near the tab insulation layer, and the short circuit risk area is calibrated by the insulation resistance attenuation curve to obtain a structural optimization layout map.

[0043] Exemplarily, the surface of the plastic shell is scanned according to the 0.1 mm scanning resolution set by the thermal imaging acquisition device, and the surface temperature is calibrated by an infrared detector. If the local temperature exceeds the set multiple of the reference temperature value, the hot zone coverage is calculated by the area integrator to generate the hot zone distribution characteristic data. The temperature gradient field, thermal stress field, and hot zone density value are extracted from the hot zone distribution characteristic data, and the stress distribution cloud map is partitioned and matched by the regional growth algorithm. The temperature field mutation position is calibrated in the residual stress high value area to generate a thermal stress coupling distribution map. On the basis of the thermal stress coupling distribution map, the multi-layer perceptron is used to identify the characteristics of the microcrack growth path, and the distance between the ear area and the crack tip, the insulation layer thickness change rate, and the crack propagation speed are calculated in the spatial coordinate system to generate microcrack evolution prediction data. According to the microcrack evolution prediction data, the resistance threshold detection dot matrix is ​​arranged near the ear insulation layer, the short circuit risk area is quantitatively calibrated by the insulation resistance attenuation curve, the thickening position of the insulation protection layer and the buffer layer arrangement position are marked, and the structural optimization layout diagram is generated. In thermal imaging testing, the surface of the plastic shell is scanned with a scanning resolution of 0.1mm, and the temperature measurement accuracy reaches 0.1℃. The reference temperature is set at 25℃. When the local temperature exceeds the reference temperature by 2℃, it is determined to be an abnormal hot spot. The calculation of the hot zone area shows that the coverage area of ​​a single hot spot is within 0.5mm 3 Up to 2mm 3 The hot spot density in the high stress area reaches 5 / cm 3 . The temperature gradient field analysis shows that the temperature around the hot spot decreases by 0.2℃ every 0.5mm, forming an obvious temperature gradient. The calculation results of the thermal stress field show that 0.5MPa thermal stress is generated for every 1℃ increase in temperature. The overlap between the stress distribution identified by the regional growth algorithm and the hot spot distribution reaches 85%, and the temperature of the residual stress high value area exceeds that of the surrounding area by 1.5℃. In the prediction of the microcrack growth path, the multi-layer perceptron input layer contains 16 stress field feature points and 16 temperature field feature points. The minimum distance between the crack and the lug area is 2mm, and the thickness of the insulation layer is thinned by 15% at the crack tip. The crack propagation rate is positively correlated with the temperature field gradient, and the propagation rate in the hot spot area reaches 0.01mm / hour. The resistance threshold detection adopts an 8×8 dot matrix arrangement with a detection point spacing of 1mm. The warning is triggered when the insulation resistance drops from the initial value of 10^14Ω to 10^12Ω. The insulation protection layer is thickened by 0.5mm in the high-risk area, and the buffer layer is arranged within 2mm around the lug. After structural optimization, the temperature gradient in the hot spot area is reduced by 30%. The coupling effect of residual stress and temperature field is particularly evident in the lug area. When the residual stress exceeds 20MPa, the local temperature rise reaches 3°C. Under the action of thermal stress, microcracks preferentially expand along the direction of maximum temperature gradient, and the risk of forming through cracks increases significantly. The insulation performance degradation process has a cumulative effect. The initial microcracks cause local heating, and thermal stress accelerates crack expansion, forming a positive feedback. When the crack density exceeds 3 / mm3 When the temperature rise rate in the area is accelerated, the temperature rises by 0.1°C per hour. The vibration and thermal cycle in the hot spot area accelerate the damage of the insulation layer. In the anti-short circuit design, the thickening of the material and the use of the buffer layer are effective. The stress in the thickened area is reduced by 40%, and the hot spot temperature drops back to the reference temperature. The stress dispersion effect of the buffer layer reduces the crack growth rate by 75%. In the temperature cycle test, the insulation performance of the optimized structure remains stable after 500 cycles from -40°C to 80°C. The key to structural optimization lies in the balance between stress dispersion and heat diffusion. If the thickened area is too large, it will affect the volumetric efficiency of the battery pack, and if it is too small, the stress concentration effect will still exist. The thickness of the buffer layer gradually decreases away from the ear, forming a gradient structure to avoid sudden stress changes.

[0044] S108. Construct a quality reliability evaluation system for plastic shell injection molding. Comprehensively consider residual stress, microcrack density, mechanical properties, and electrical properties. Use a fuzzy comprehensive evaluation method to quantitatively evaluate the quality of plastic shells and screen and improve plastic shells with defects.

[0045] The detection data of stress distribution value, crack density value, mechanical strength value and insulation resistance value are obtained on the surface of the rubber shell through a detection data collector, and the detection data collector is arranged with stress sampling points, crack density observation points, mechanical performance test points and insulation performance test points; according to the stress distribution value, crack density value, mechanical strength value and insulation resistance value, a range normalizer is used to perform interval mapping on the collected values ​​to obtain a standardized evaluation parameter matrix; for the standardized evaluation parameter matrix, a principal component analyzer is used to calculate the correlation of characteristic indicators, and a hierarchical structure method is used to quantify the weights of performance parameters to obtain a comprehensive performance evaluation matrix; for the comprehensive performance evaluation matrix, a fuzzy relationship matrix is ​​constructed using triangular fuzzy numbers, and various parameters are weighted and integrated through an indicator synthesis operator, and a quality evaluation result set and a quality improvement parameter set are generated by comparing with a preset quality level threshold.

[0046] Exemplarily, according to the detection data collector, 16 stress sampling points, 8 crack density observation points, 12 mechanical performance test points, and 10 insulation performance test points are arranged on the surface of the rubber shell, and the collected values ​​are interval mapped by the range normalizer, and a standardized evaluation parameter matrix is ​​generated based on the performance index standard library. From the standardized evaluation parameter matrix, the stress distribution value, crack density value, mechanical strength value, and insulation resistance value of each point are extracted, and the correlation calculation of the characteristic index is performed using the principal component analyzer, and the performance parameters are weighted quantified in combination with the hierarchical structure method to generate a comprehensive performance evaluation matrix. On the basis of the comprehensive performance evaluation matrix, a fuzzy relationship matrix is ​​constructed using triangular fuzzy numbers, and each parameter is weighted and integrated through an index synthesis operator. The rubber shell is graded and evaluated against the preset quality grade threshold to generate a quality evaluation result set. According to the quality evaluation result set, a defect morphology identifier is used to calibrate the improvement parameters in the low-value area of ​​the rubber shell quality rating, including the wall thickness optimization value, material strengthening value, and structural reinforcement value. The improvement action points are marked in the spatial coordinate system to generate a quality improvement parameter set. In the multi-point performance test, stress sampling points are evenly arranged along the circumference of the rubber shell with a spacing of 25mm. The stress value is 15MPa in the normal area and 35MPa in the abnormal area. The crack density observation points are arranged in the stress concentration area, and the number of cracks per unit area ranges from 0.5 / mm 3 Up to 5 lines / mm 3 The tensile strength recorded at the mechanical performance test point is 45MPa, and the elongation at break is 150%. The volume resistivity measured at the insulation performance test point is 10^14Ω·cm. The standardization process uses the range method to map each indicator to the range of 0 to 1. The stress index is normalized based on the design allowable stress of 20MPa. The crack density is based on the safety threshold of 1 / mm 3 The mechanical properties are based on the standard value of 45MPa, and the insulation properties are based on 10^13Ω·cm. The principal component analysis results show that the correlation coefficient between residual stress and crack density is 0.85, and the weights of the two on quality are 0.35 and 0.3 respectively. The correlation coefficient between mechanical properties and insulation properties is 0.6, and the weights are 0.2 and 0.15 respectively. For every 5MPa increase in residual stress, the crack density increases by 0.5 / mm 3. Fuzzy evaluation uses triangular fuzzy numbers to characterize the index membership, and sets four levels: excellent, good, qualified, and unqualified. When the comprehensive score is lower than 0.6, it is judged as unqualified. The weighted average operator is used for weight integration, taking into account the interaction between indicators. The excellent quality range is 0.85 to 1, and the good range is 0.7 to 0.85. The defect improvement parameters contain multiple dimensions. The wall thickness optimization value is determined based on the stress distribution, and the thickness is increased by 0.5mm in the high stress area. The material strengthening value is achieved by improving the crystallinity, which increases from 30% to 35%. The structural reinforcement value considers the arrangement of local reinforcing ribs, and the rib height is 0.8mm. Multi-parameter collaborative optimization shows that there is a coupling effect between the indicators. The increase in wall thickness increases the difficulty of injection molding and extends the molding cycle by 5%. Material strengthening improves the mechanical properties, but excessive crystallization reduces toughness. Although structural reinforcement reduces stress, it may introduce new stress concentration. In the process of quality improvement, a balance needs to be struck between the improvement parameters. When the wall thickness increase exceeds 0.8mm, it will lead to an increase in internal stress. When the crystallinity exceeds 38%, the material becomes brittle and the impact resistance decreases. Too high a density of reinforcement ribs will affect the fluidity of the material and form welding defects. After the evaluation system is established, the quality control benchmark is formed through the accumulation of batch data. The stress level of the excellent plastic shell is stable below 15MPa, and the crack density is less than 0.8 / mm 3 , tensile strength exceeds 50MPa, and insulation resistance remains above 10^14Ω·cm. These data provide a quantitative basis for process optimization.

[0047] The above description is merely a preferred embodiment of one or more embodiments of the present specification and is not intended to limit one or more embodiments of the present specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present specification shall be included in the scope of protection of one or more embodiments of the present specification.

Claims

1. A short circuit protection prediction method for a new energy battery with a plastic shell, characterized in that: The method comprises: Based on the constructed three-dimensional solid model of the new energy battery shell, the finite element simulation analysis method is used to input the mechanical properties parameters of the material, the geometric dimensions of the parts and the injection molding process parameter data, simulate the injection molding process, and obtain the residual stress distribution cloud map inside the shell. The residual stress concentration area is determined through the stress distribution cloud map; The surface and internal microscopic morphology of the rubber shell in the residual stress concentration area is characterized and analyzed by microscope observation to obtain the morphological characteristics of microcrack formation, and the mechanical properties of the microcrack formation are analyzed by combining the material mechanical properties parameters, including fracture toughness and fracture strain; Through the mechanical property test of the constructed three-dimensional solid model of the new energy battery shell, the residual stress level is determined according to the yield strength of the material, the influence of the residual stress level on the mechanical properties of the shell is analyzed, and the residual stress control threshold is preset. When the residual stress exceeds the residual stress control threshold, it is determined that the mechanical properties of the shell do not meet the requirements; Conducting a dielectric constant and volume resistivity electrical performance test method on the constructed three-dimensional solid model of the new energy battery gel shell, evaluating the electrical insulation performance of the constructed three-dimensional solid model of the new energy battery gel shell, and determining the microcrack density control range. If the microcrack density exceeds the range, it is determined that the insulation performance of the constructed three-dimensional solid model of the new energy battery gel shell is reduced; If it is determined that the insulation of the constructed three-dimensional solid model of the new energy battery gel shell is reduced, the injection molding process parameters are optimized, including injection temperature, holding time, and cooling rate, and the optimal process parameter combination is obtained through orthogonal experimental design; Based on the optimal process parameter combination, the injection molded plastic shell is produced, the injection molded plastic shell is annealed to promote the release of residual stress, low-frequency vibration is applied to the plastic shell to induce residual stress redistribution, reduce stress concentration, inhibit microcrack propagation, and annealing and vibration aging treatment are performed to improve the residual stress state of the plastic shell; After annealing and vibration aging treatment, the surface of the rubber shell is scanned by infrared thermal imaging method to identify the local heating area caused by residual stress. Combined with the stress distribution cloud map in the three-dimensional solid model of the rubber shell, the influence of micro-crack extension in the hot spot area on the insulation performance of the battery tab is predicted, the risk of battery short circuit in the hot spot area is evaluated, and the optimization suggestions for the anti-short circuit design are put forward for the rubber shell in the high-risk area; A reliability evaluation system for the injection molding quality of plastic shells was established. The residual stress, microcrack density, mechanical properties, and electrical properties were comprehensively considered, and a fuzzy comprehensive evaluation method was used to quantitatively evaluate the quality of plastic shells, so that defective plastic shells could be screened and improved.

2. The method according to claim 1, characterized in that The three-dimensional solid model of the new energy battery plastic shell constructed is based on the finite element simulation analysis method, and the mechanical property parameters, part geometric dimensions and injection molding process parameter data of the material are input to simulate the injection molding process, and the residual stress distribution cloud map inside the plastic shell is obtained. The residual stress concentration area is determined through the stress distribution cloud map, including: Collecting the size data of the rubber shell structure in the three-dimensional solid model, and generating the rubber shell elastic deformation curve data according to the calibration measurement data through calibration measurement of the material elastic coefficient and Poisson's ratio; According to the elastic deformation curve data of the plastic shell, the injection melt temperature, injection pressure and injection rate data are collected, load boundary conditions are applied to the sampling points of the injection station, and multiple linear regression is used to obtain stress calculation input data; Calling a finite element meshing tool according to the stress calculation input data, meshing the three-dimensional solid model, and calculating a unit node stress numerical distribution data set based on Hooke's law; Stress cloud map data is constructed according to the stress numerical distribution data set, the principal stress value is calculated using the vonMises stress criterion, the residual stress cloud map data is obtained by calibrating the coordinates of the stress peak point, and the residual stress concentration area is determined through the stress distribution cloud map.

3. The method according to claim 1, characterized in that The microscopic observation method is used to characterize and analyze the surface and internal microscopic morphology of the rubber shell in the residual stress concentration area, obtain the morphological characteristics of microcrack formation, and combine the material mechanical property parameters, including fracture toughness and fracture strain, to analyze the mechanical properties formed by microcracks, including: The surface morphology image of the rubber shell is collected by a microscopic scanning device, and the microcrack boundary contour curve data is obtained by using a Sobel edge extractor according to the morphology image; Arranging a strain sampling point array within the gauge range for the microcrack boundary contour curve data, recording the corresponding relationship between the fracture strain value and the microcrack boundary position according to the strain sampling point array, and obtaining a strain data field distribution diagram; According to the strain data field distribution diagram, the microcrack morphology features are segmentedly extracted using an edge detection operator, and the crack depth value, crack elongation rate and crack orientation angle are calculated through the microcrack morphology features to obtain a crack morphology feature parameter matrix; The crack tip opening displacement value, stress intensity factor and fracture toughness are calculated based on the crack morphology characteristic parameter matrix and the generalized fracture criterion. The normal stress component and the shear stress component are calibrated in the micro-region fracture unit through the opening displacement value to obtain the crack extension mechanical parameter field.

4. The method according to claim 1, characterized in that: The mechanical properties of the constructed three-dimensional solid model of the new energy battery shell are tested, the residual stress level is determined according to the yield strength of the material, the influence of the residual stress level on the mechanical properties of the shell is analyzed, and the residual stress control threshold is preset. When the residual stress exceeds the residual stress control threshold, it is determined that the mechanical properties of the shell do not meet the requirements, including: Orthogonal loading points are arranged on the surface of the shell according to the three-dimensional solid load loader, and the displacement field data of the shell surface is collected by the digital image correlation optical strain meter, and the uniaxial tensile load is applied to the loaded area to obtain the strain distribution data of the shell; Extract the tensile strain value, compressive strain value and shear strain value of the measuring point from the strain distribution data of the rubber shell, and calibrate the linear segment value of the material using a tensile stress-strain curve calculator to obtain a yield point stress-strain curve; According to the yield point stress-strain curve, the shear modulus and bulk modulus of the rubber shell material are solved using the Lame parameter calculator, and the residual stress value and stress concentration factor are calculated to obtain the residual stress distribution data; According to the residual stress distribution data, a numerical comparator is used to compare the residual stress value with a preset stress threshold, and the residual stress peak point position and stress level are calibrated to obtain the mechanical property determination result.

5. The method according to claim 1, characterized in that The electrical performance testing method of the dielectric constant and volume resistivity of the constructed three-dimensional solid model of the new energy battery gel shell is performed to evaluate the electrical insulation performance of the constructed three-dimensional solid model of the new energy battery gel shell, and determine the microcrack density control range. If the microcrack density exceeds the range, it is judged that the insulation performance of the constructed three-dimensional solid model of the new energy battery gel shell is reduced, including: According to the three-dimensional solid test platform, the test electrodes are arranged on the surface of the plastic shell, the surface dielectric constant value is collected by the capacitance measuring instrument, and the volume resistivity value is recorded by the high-voltage DC resistance meter to obtain the basic data of electrical performance; Extract the dielectric constant distribution value and the volume resistivity distribution value from the basic electrical performance data, scan and measure the insulation strength of the material through a voltage-current characteristic curve instrument, and obtain insulation performance evaluation data; For the insulation performance evaluation data, a region growth segmenter is used to process the microcrack image in blocks, and the number of cracks per unit area and the crack surface area are calculated by a standard convolutional neural network to generate microcrack density feature data; According to the microcrack density characteristic data, the microcrack characteristic values ​​are mapped to the insulation performance parameters through a correlation matrix calculator to determine the insulation strength weakening area calibration result; It also includes: using a dielectric constant tester and a high resistance meter to measure the dielectric constant and volume resistivity of the plastic shell material, judging whether the plastic shell material meets the insulation requirements based on the electrical insulation performance standard, counting the number and distribution of internal microcracks, analyzing the correlation between microcrack density and insulation performance, and obtaining the control range and threshold of the microcrack density of the plastic shell material.

6. The method according to claim 5, characterized in that The dielectric constant tester and high resistance meter are used to measure the dielectric constant and volume resistivity of the plastic shell material, and based on the electrical insulation performance standard, it is judged whether the plastic shell material meets the insulation requirements, the number and distribution of internal microcracks are counted, and the correlation between microcrack density and insulation performance is analyzed to obtain the control range and threshold of the microcrack density of the plastic shell material, including: Use the test instrument to apply AC voltage and DC voltage to the surface of the plastic shell to obtain the capacitance and resistance values, and obtain the comprehensive data of electrical characteristics according to the breakdown voltage scan; Extracting the dielectric constant value and dielectric loss angle according to the electrical characteristic comprehensive data, and obtaining insulation performance characteristic data by comparing with the preset insulation standard value through a numerical comparator; Running a regional growing algorithm on the insulation performance characteristic data, extracting the number of cracks per unit area and the crack extension length, and constructing an insulation strength characteristic matrix through a convolutional neural network; The correlation value between the crack density parameter and the insulation strength parameter is calculated according to the insulation strength characteristic matrix. If the correlation value exceeds a preset standard threshold, the microcrack control interval data is determined.

7. The method according to claim 1, characterized in that If it is determined that the insulation of the constructed three-dimensional solid model of the new energy battery gel shell is reduced, the injection molding process parameters are optimized, including injection temperature, holding time, and cooling rate, and the optimal process parameter combination is obtained through orthogonal experimental design, including: According to the injection molding specification, the process parameter calculator is used to arrange the injection molding melt temperature range, holding time range and cooling rate range, and the material viscosity value and crystallinity value corresponding to each group of parameters are calibrated through the physical property parameter database to obtain the injection molding process parameter matrix; Extracting each group of process configuration parameters from the injection molding process parameter matrix, recording the cavity pressure sensing point, the temperature sensing point and the flow front sensing point in real time in the injection molding controller, and obtaining the injection molding process parameter set; For the injection molding process parameter set, a gradient boosting tree is used to extract features from parameter curve data, the insulation contribution value of a single parameter and the parameter coupling index are calculated, and a set of process parameter influencing factors is generated; According to the process parameter influencing factor set, a parameter response surface is constructed through a central composite designer, and a traversal search is performed on the parameter space under the constraint of optimizing insulation performance to calibrate the optimal values ​​of melt temperature, holding time and cooling rate.

8. The method according to claim 1, characterized in that The method of producing an injection molded plastic shell based on an optimal process parameter combination, annealing the injection molded plastic shell to promote residual stress release, applying low-frequency vibration to the plastic shell to induce residual stress redistribution, reduce stress concentration, inhibit microcrack propagation, and perform annealing and vibration aging treatment to improve the residual stress state of the plastic shell includes: Arrange a temperature sensor array on the surface of the plastic shell according to the process control parameters, use a temperature collector to obtain the temperature field distribution and crystallinity change data, and obtain the annealing temperature change curve; The heating rate and the holding time parameters are extracted through the annealing temperature variation curve, and the internal displacement field of the rubber shell is numerically solved by a stress relaxation calculator to obtain a stress relaxation field distribution diagram; According to the stress relaxation field distribution diagram, a frequency sweep signal is applied to the plastic shell by using an acoustic exciter, and the frequency sweep signal is processed by a neural network fitter to obtain stress redistribution data; A vibration sensor array is arranged in the stress concentration area of ​​the rubber shell according to the stress redistribution data, and the collected signal of the vibration sensor array is processed by a phase delay calculator to obtain the microcrack growth rate and the stress intensity factor.

9. The method according to claim 1, characterized in that: After annealing and vibration aging treatment, the surface of the plastic shell is scanned by infrared thermal imaging method to identify the local heating area caused by residual stress. Combined with the stress distribution cloud map in the three-dimensional solid model of the plastic shell, the influence of microcrack extension in the hot spot area on the insulation performance of the battery tab is predicted, the risk of battery short circuit in the hot spot area is evaluated, and the anti-short circuit design optimization suggestions are proposed for the plastic shell in the high-risk area, including: According to the scanning data obtained by the thermal imaging acquisition device, the surface temperature of the plastic shell is calibrated by an infrared detector, and the infrared detector uses an area integrator to calculate the hot zone coverage to obtain the hot zone distribution characteristic data; Extracting the temperature gradient field and the thermal stress field from the heat zone distribution characteristic data, and performing partition matching on the stress distribution cloud map by using a regional growing algorithm to obtain a thermal stress coupling distribution map; According to the thermal stress coupling distribution diagram, a multi-layer perceptron is used to perform feature recognition on the microcrack growth path, and the distance between the tab area and the crack tip is calculated in a spatial coordinate system to obtain microcrack evolution prediction data; According to the microcrack evolution prediction data, a resistance threshold detection array is arranged near the tab insulation layer, and the short circuit risk area is calibrated by the insulation resistance attenuation curve to obtain a structural optimization layout diagram.

10. The method according to claim 1, characterized in that The construction of the quality reliability evaluation system for plastic shell injection molding is to comprehensively consider residual stress, microcrack density, mechanical properties, and electrical properties, and adopt a fuzzy comprehensive evaluation method to quantitatively evaluate the quality of the plastic shell, and screen and improve the plastic shell with defects, including: The detection data of stress distribution value, crack density value, mechanical strength value and insulation resistance value are obtained on the surface of the rubber shell by a detection data collector, wherein the detection data collector is arranged with stress sampling points, crack density observation points, mechanical performance test points and insulation performance test points; According to the stress distribution value, crack density value, mechanical strength value and insulation resistance value, a range normalizer is used to perform interval mapping on the collected values ​​to obtain a standardized evaluation parameter matrix; For the standardized evaluation parameter matrix, the principal component analyzer is used to calculate the correlation of the characteristic indicators, and the hierarchical structure method is used to quantify the weights of the performance parameters to obtain a comprehensive performance evaluation matrix; For the comprehensive performance evaluation matrix, triangular fuzzy numbers are used to construct a fuzzy relationship matrix, and various parameters are weighted and integrated through an indicator synthesis operator. The quality evaluation result set and the quality improvement parameter set are generated by comparing with the preset quality level threshold.

Images