A battery pack production process regulation system and method based on a PLC

Abstract

The application relates to the technical field of new energy battery intelligent manufacturing and industrial automation control, in particular to a battery pack production process regulation system and method based on PLC, which comprises the following steps: mapping physical field data to a unified space-time coordinate system to construct a physical state characteristic vector; generating a comprehensive coupling drift index; based on the comprehensive coupling drift index, generating a dynamic energy output instruction by using a nonlinear adaptive compensation model; writing the dynamic energy output instruction into an executing mechanism to execute closed-loop energy compensation; based on the physical state characteristic vector, calculating a dynamic risk index; in response to the dynamic risk index being greater than a preset risk threshold, triggering an emergency stop instruction and marking a defective product; and in response to the dynamic risk index being less than or equal to the preset risk threshold, maintaining execution of the dynamic energy output instruction; the application solves the problem that a single-dimensional parameter is difficult to represent complex coupling drift, and significantly improves the perception dimension and evaluation accuracy of the system to environmental changes.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing and industrial automation control technology for new energy batteries, specifically a PLC-based battery pack production process control system and method. Background Technology

[0002] In the automated assembly and production of battery packs, key processes such as welding or gluing are significantly affected by multi-physical field coupling factors such as geometric gaps, thermal field distribution and contact pressure. The real-time status of these physical parameters is directly related to the processing quality and structural stability of the product.

[0003] Existing production control solutions typically rely on single-dimensional parameter monitoring or open-loop control based on fixed formulas. When faced with discrete data collected from heterogeneous devices such as laser profilometry, infrared thermography, and pressure sensors, they lack an effective mechanism for unified spatiotemporal coordinate alignment. Due to the lack of a multi-dimensional data fusion evaluation system, traditional methods struggle to accurately quantify the comprehensive deviation of complex processing environments from ideal process benchmarks. Furthermore, fixed parameter control cannot handle dynamic fluctuations caused by accumulated tolerances or thermal buildup effects, resulting in a lack of adaptability in energy output. This makes the system highly susceptible to overload or insufficient compensation, leading to issues such as poor soldering, burn-through, and structural equipment failure. Therefore, achieving a unified spatiotemporal mapping of heterogeneous physical field data, constructing a quantitative index capable of characterizing comprehensive coupling drift, and implementing millisecond-level nonlinear adaptive energy compensation and dynamic risk mitigation based on this index to improve process stability and reduce defect rates have become urgent technical challenges. Summary of the Invention

[0004] The purpose of this invention is to provide a PLC-based battery pack production process control system and method, which can achieve unified spatiotemporal mapping and fusion calculation of heterogeneous physical field data, solve the problem of difficult precise quantification of coupling drift in complex processing environments, and perform closed-loop energy compensation and dynamic risk blocking through a nonlinear adaptive model, thereby avoiding poor soldering, burn-through and equipment structural failure caused by accumulated tolerances or thermal effects, significantly improving process stability and product yield. Specifically, the technical solution of this invention is as follows:

[0005] A PLC-based method for controlling the battery pack production process includes:

[0006] Physical field data are collected using a laser contour sensor, an infrared thermal imager, and a servo pressure sensor.

[0007] Step 1: Map the physical field data to a unified spatiotemporal coordinate system to construct a physical state feature vector;

[0008] Step 2: Generate a comprehensive coupling drift index based on the physical state feature vector and the preset standard process benchmark;

[0009] Step 3: Based on the comprehensive coupling drift index, a dynamic energy output command is generated using a nonlinear adaptive compensation model;

[0010] Step 4: Write the dynamic energy output command into the actuator to perform closed-loop energy compensation;

[0011] Step 5: Calculate the dynamic risk index based on the physical state feature vector;

[0012] Step 6: In response to the dynamic risk index exceeding the preset risk threshold, an emergency stop command is triggered and defective products are marked.

[0013] Step 7: In response to the dynamic risk index being less than or equal to the preset risk threshold, maintain the execution of the dynamic energy output command.

[0014] Optionally, construct a physical state feature vector, including:

[0015] Obtain the vertical clearance between the busbar and the terminal block;

[0016] Obtain the heat accumulation temperature rise of the current area relative to the ambient temperature;

[0017] Obtain the normal clamping force;

[0018] The vertical gap, thermal accumulation temperature rise, and normal clamping force are encapsulated as a physical state feature vector.

[0019] Optionally, preset standard process benchmarks include standard gaps, standard pressures, upper limits of equipment safe temperature, and standard operating cycles for a single process.

[0020] Generate the comprehensive coupling drift index, including:

[0021] Calculate the ratio of the geometric deviation of the vertical clearance relative to the standard clearance;

[0022] Based on the thermal accumulation temperature rise, the upper limit of the equipment's safe temperature, and the standard operating cycle of a single process, the deviation ratio of the thermodynamic dimension is calculated.

[0023] Calculate the ratio of the mechanical deviation of the normal clamping force relative to the standard pressure;

[0024] The geometric deviation ratio, thermodynamic dimension deviation ratio, and mechanical deviation ratio are algebraically weighted and summed to generate a comprehensive coupling drift index.

[0025] Optionally, a nonlinear adaptive compensation model is used to generate dynamic energy output commands, including:

[0026] Obtain the preset baseline energy value, maximum allowable compensation ratio factor, and adjustment sensitivity coefficient of the process formulation;

[0027] Calculate the product of the overall coupling drift index and the regulation sensitivity coefficient;

[0028] Applying the hyperbolic tangent function to the product yields the intermediate value of the saturation gain.

[0029] The gain coefficient is calculated based on the median value of the saturated gain and the maximum allowable compensation ratio factor.

[0030] The reference energy value is corrected using a gain coefficient to generate a dynamic energy output command.

[0031] Optionally, calculate the dynamic risk index, including:

[0032] Obtain the physical state feature vector of the current scan cycle and the physical state feature vector of the previous scan cycle;

[0033] The normalized physical state fluctuation rate is calculated based on the physical state feature vector of the current scan cycle and the physical state feature vector of the previous scan cycle.

[0034] Obtain static drift weights and dynamic fluctuation weights;

[0035] The static risk component is obtained by squared the product of the comprehensive coupling drift index and the static drift weight.

[0036] The dynamic risk component is obtained by calculating the square of the product of the normalized physical state fluctuation rate and the dynamic fluctuation weight.

[0037] The dynamic risk index is generated by summing the static and dynamic risk components and taking the square root.

[0038] Optionally, the normalized physical state fluctuation rate is calculated, including:

[0039] Calculate the changes in vertical clearance, thermal accumulation temperature rise, and normal clamping force in adjacent PLC scanning cycles, respectively.

[0040] Divide the change by the corresponding preset standard process reference to obtain the dimensionless change ratio;

[0041] Calculate the sum of squares of dimensionless rates of change;

[0042] The square root of the sum of squares is taken to obtain the modulus of change.

[0043] Dividing the change modulus by the PLC scan cycle yields the normalized physical state fluctuation rate.

[0044] A PLC-based battery pack production process control system, characterized in that it includes:

[0045] The data acquisition unit is used to acquire physical field data using a laser profile sensor, an infrared thermal imager, and a servo pressure sensor.

[0046] The state mapping unit is used to map physical field data to a unified spatiotemporal coordinate system and construct physical state feature vectors.

[0047] The drift quantization unit is used to generate a comprehensive coupled drift index based on the physical state feature vector and a preset standard process benchmark.

[0048] An adaptive control unit is used to generate dynamic energy output commands based on a comprehensive coupling drift index and a nonlinear adaptive compensation model.

[0049] The closed-loop execution unit is used to write dynamic energy output commands into the execution mechanism and perform closed-loop energy compensation.

[0050] The risk assessment unit is used to calculate the dynamic risk index based on the physical state feature vector.

[0051] The blocking control unit is used to trigger an emergency stop command and mark defective products in response to a dynamic risk index greater than a preset risk threshold; and to maintain the execution of dynamic energy output commands in response to a dynamic risk index less than or equal to a preset risk threshold.

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

[0053] 1. This invention solves the technical problem of asynchronous sampling frequencies and inconsistent spatial references of multi-source data by mapping discrete data from heterogeneous sensors such as laser contouring, infrared thermal imaging, and servo pressure to a unified spatiotemporal coordinate system. This spatiotemporal alignment mechanism eliminates the time delay and bias of data acquisition and can accurately construct physical state feature vectors that reflect the real-time state of the processing interface, laying a high-precision data foundation for subsequent multi-physics coupling analysis.

[0054] 2. This invention proposes a comprehensive coupling drift index calculation method. By normalizing and weighting the vertical gap, thermal accumulation temperature rise, and normal clamping force, the method achieves the unification of multi-dimensional physical dimensions. This method effectively quantifies the comprehensive deviation of the current processing environment from the ideal process benchmark, solves the problem that single-dimensional parameters are difficult to characterize complex coupling drift, and significantly improves the system's perception dimension and evaluation accuracy of environmental changes.

[0055] 3. This invention utilizes a nonlinear adaptive compensation model to generate dynamic energy output commands, achieving millisecond-level closed-loop energy control. This model possesses small-deviation linear response and large-deviation soft-limiting characteristics, which can not only automatically correct process fluctuations caused by accumulated tolerances or thermal effects, but also prevent energy overload caused by sensor noise or extreme operating conditions from the physical level, thereby effectively avoiding quality defects such as poor soldering and burn-through of battery packs and improving process stability.

[0056] 4. This invention constructs a dual-weight risk assessment mechanism based on static drift and dynamic fluctuation, which can accurately calculate the dynamic risk index. This mechanism can not only monitor conventional process deviations, but also keenly detect sudden structural failures such as fixture loosening. Through a hierarchical control strategy, an emergency stop is immediately triggered and defective products are marked when an uncompensable systemic risk is identified, which greatly improves the fault tolerance rate and equipment safety of automated production lines. Attached Figure Description

[0057] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0058] Figure 1 This is a flowchart of the method of the present invention;

[0059] Figure 2 This is a structural diagram of the system of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0061] Example 1:

[0062] Please see Figure 1 A PLC-based method for controlling the production process of battery packs, comprising:

[0063] Physical field data are collected using a laser contour sensor, an infrared thermal imager, and a servo pressure sensor.

[0064] Step 1: Map the physical field data to a unified spatiotemporal coordinate system to construct a physical state feature vector;

[0065] Step 2: Generate a comprehensive coupling drift index based on the physical state feature vector and the preset standard process benchmark;

[0066] Step 3: Based on the comprehensive coupling drift index, a dynamic energy output command is generated using a nonlinear adaptive compensation model;

[0067] Step 4: Write the dynamic energy output command into the actuator to perform closed-loop energy compensation;

[0068] Step 5: Calculate the dynamic risk index based on the physical state feature vector;

[0069] Step 6: In response to the dynamic risk index exceeding the preset risk threshold, an emergency stop command is triggered and defective products are marked.

[0070] Step 7: In response to the dynamic risk index being less than or equal to the preset risk threshold, maintain the execution of the dynamic energy output command;

[0071] Please see Figure 2 A PLC-based battery pack production process control system includes:

[0072] The data acquisition unit is used to acquire physical field data using a laser profile sensor, an infrared thermal imager, and a servo pressure sensor.

[0073] The state mapping unit is used to map physical field data to a unified spatiotemporal coordinate system and construct physical state feature vectors.

[0074] The drift quantization unit is used to generate a comprehensive coupled drift index based on the physical state feature vector and a preset standard process benchmark.

[0075] An adaptive control unit is used to generate dynamic energy output commands based on a comprehensive coupling drift index and a nonlinear adaptive compensation model.

[0076] The closed-loop execution unit is used to write dynamic energy output commands into the execution mechanism and perform closed-loop energy compensation.

[0077] The risk assessment unit is used to calculate the dynamic risk index based on the physical state feature vector.

[0078] The blocking control unit is used to trigger an emergency stop command and mark defective products in response to a dynamic risk index greater than a preset risk threshold; and to maintain the execution of dynamic energy output commands in response to a dynamic risk index less than or equal to a preset risk threshold.

[0079] This embodiment discloses a PLC-based battery pack production process control system and method. The system is configured to solve the process instability problem caused by multi-physical field coupling during the automated assembly of battery packs.

[0080] At the hardware architecture level, the system includes a data acquisition unit, a PLC controller, and actuators. The data acquisition unit is deployed at the welding or gluing station of the production line and specifically includes a laser profile sensor for acquiring the microscopic geometric gap between the busbar and the pole, an infrared thermal imager for monitoring the background thermal field distribution in the welding area, and a servo pressure sensor for acquiring real-time contact pressure. The above sensors communicate with the PLC controller through industrial real-time Ethernet buses such as EtherCAT or Profinet to ensure microsecond-level synchronization of data acquisition.

[0081] At the methodological level, the control method in this embodiment first acquires the aforementioned physical field data through a data acquisition unit; the state mapping unit executes step 1, mapping the discrete data acquired by heterogeneous sensors to a unified spatiotemporal coordinate system; since the sampling frequency and response time of different sensors differ, the system uses the PLC's scanning cycle as a reference, aligns the timestamps of each sensor, and uniformly aligns the spatial coordinates to the CAD reference point of the workpiece, thereby constructing a physical state feature vector describing the current state of the machining interface. ;

[0082] Next, the drift quantization unit executes step 2; this step calls the preset standard process reference stored in the PLC recipe and converts the current physical state feature vector. A multi-dimensional comparison with this benchmark is performed to generate a dimensionless comprehensive coupling drift index. This index aims to quantify the overall deviation of the current processing environment from the ideal process state, solving the technical problem that a single-dimensional parameter cannot characterize complex coupled drift.

[0083] Based on the calculated comprehensive coupling drift index In step 3, the adaptive control unit invokes the nonlinear adaptive compensation model; this model intelligently calculates the energy correction value required for the current process based on the magnitude of the drift, and then generates a dynamic energy output command. In step 4, the closed-loop execution unit writes the instruction into the register of the actuator via a high-speed bus, such as the power setting address of the laser welding controller or the servo drive unit of the glue dispensing pump, to replace the fixed value in the original process formula, thereby completing millisecond-level closed-loop energy compensation.

[0084] While performing compensation, the risk assessment unit performs step 5 in parallel, calculating the dynamic risk index based on the historical trend of the physical state feature vector. This index is used to characterize the stability of a system and can identify sudden failures that cannot be corrected by energy compensation alone.

[0085] The blocking control unit is based on the dynamic risk index. Implement a tiered control strategy; if step 6 determines the dynamic risk index... If the risk exceeds a preset risk threshold, the system identifies it as a sudden anomaly, such as a loose fixture or foreign object intrusion. In this case, an emergency stop command is immediately triggered, and a signal is sent to the MES system to mark the current workpiece as defective. The preset risk threshold is determined based on the 99th percentile of the statistical distribution of risk indices from historical production data when structural failures occur. If step 7 determines the dynamic risk index... If the risk threshold is less than or equal to the preset risk threshold, the system determines that the current process fluctuation is within a controllable range, maintains the execution of dynamic energy output commands, and ensures continuous production.

[0086] Through the closed-loop control logic described above, this embodiment achieves full-process automation from environmental perception, analysis and decision-making to execution feedback, significantly reducing the defect rate caused by accumulated tolerances and thermal effects.

[0087] Example 2:

[0088] Constructing physical state feature vectors includes:

[0089] Obtain the vertical clearance between the busbar and the terminal block;

[0090] Obtain the heat accumulation temperature rise of the current area relative to the ambient temperature;

[0091] Obtain the normal clamping force;

[0092] The vertical gap, thermal accumulation temperature rise, and normal clamping force are encapsulated as a physical state feature vector;

[0093] The preset standard process benchmarks include standard clearance, standard pressure, upper limit of equipment safe temperature, and standard operation cycle for a single process.

[0094] Generate the comprehensive coupling drift index, including:

[0095] Calculate the ratio of the geometric deviation of the vertical clearance relative to the standard clearance;

[0096] Based on the thermal accumulation temperature rise, the upper limit of the equipment's safe temperature, and the standard operating cycle of a single process, the deviation ratio of the thermodynamic dimension is calculated.

[0097] Calculate the ratio of the mechanical deviation of the normal clamping force relative to the standard pressure;

[0098] The geometric deviation ratio, thermodynamic dimension deviation ratio, and mechanical deviation ratio are algebraically weighted and summed to generate a comprehensive coupling drift index.

[0099] This embodiment provides a detailed explanation of the construction logic of the physical state feature vector and the calculation model of the comprehensive coupling drift index;

[0100] Regarding the construction of the physical state feature vector, the system extracts the height difference between the busbar and the electrode contact surface from the point cloud data of the laser profile sensor, and defines it as the vertical gap. The unit is millimeters; simultaneously, based on the aligned workpiece CAD reference points, the system generates a dynamic region of interest (ROI) in the infrared thermal imaging field of view from the infrared thermal imaging data. This ROI covers a 2mm radius on both sides of the weld centerline to eliminate interference from fixture reflections or background heat sources; the highest temperature of the weld area is extracted within the ROI. and subtract the ambient temperature The calculated temperature rise due to thermal accumulation was obtained. The unit is degrees Celsius; in addition, the system reads the feedback current of the servo pressure head and converts it into a force value, defined as the normal clamping force. The unit is Newton; the three physical quantities with different dimensions mentioned above are encapsulated as column vectors. This serves as a unified input benchmark for subsequent calculations;

[0101] Considering the different requirements of physical laws for energy compensation direction—namely, increasing the gap requires increasing energy, while excessive heat accumulation requires decreasing energy—this embodiment constructs a directional vectorized drift model; the specific calculation formula is as follows:

[0102] ;

[0103] in: The preset process weighting coefficients, and satisfying The first term is the geometric compensation component, after removing the absolute value sign, when the actual gap... Larger than standard gap When the first term is positive, it drives an increase in energy to fill the gap; the second term is the thermodynamic compensation component, which introduces a negative coefficient; where... The preset standard process heating curve at the current moment The theoretical temperature rise value; when the actual temperature rise When the temperature rise exceeds the theoretical value, this term is negative, driving energy reduction to prevent burn-through; the third term is the pressure compensation component, preserving the unidirectional or bidirectional effect of pressure deviation on energy. The symbol is a function; through this calculation model, this embodiment successfully solves the problem of dimensional unification and fusion calculation of heterogeneous multiphysics data.

[0104] Example 3:

[0105] A nonlinear adaptive compensation model is used to generate dynamic energy output commands, including:

[0106] Obtain the preset baseline energy value, maximum allowable compensation ratio factor, and adjustment sensitivity coefficient of the process formulation;

[0107] Calculate the product of the overall coupling drift index and the regulation sensitivity coefficient;

[0108] Applying the hyperbolic tangent function to the product yields the intermediate value of the saturation gain.

[0109] The gain coefficient is calculated based on the median value of the saturated gain and the maximum allowable compensation ratio factor.

[0110] The reference energy value is corrected using a gain coefficient to generate a dynamic energy output command.

[0111] This embodiment details the specific process of generating dynamic energy output commands using a nonlinear adaptive compensation model;

[0112] To ensure the compensation effect while preventing over-regulation, the system pre-obtains the baseline energy value in the process formulation. Maximum allowable compensation ratio factor and adjust sensitivity coefficient Among them, the maximum allowable compensation ratio factor Determined by the process window, for example, the range of energy fluctuations allowed by the process. Then set Adjusting the sensitivity coefficient These are system tuning parameters used to determine the control system's response speed to drift.

[0113] Based on the above parameters, the system performs nonlinear gain calculation, and the specific mathematical model is as follows:

[0114] ;

[0115] This calculation process calculates the comprehensive coupling drift index. With adjustment sensitivity coefficient The product; apply the hyperbolic tangent function to the product. The calculation yields the intermediate value of the saturated gain. Due to the hyperbolic tangent function's characteristics of small-deviation linear response and large-deviation soft-limiting, the system provides approximately linear and accurate compensation when the drift is small. It is worth noting that, due to the combined coupling drift exponent... Constructed as a signed numerical value, when When the value is negative, such as when thermal accumulation is dominant, the hyperbolic tangent function outputs a negative gain, resulting in a decrease in the final output energy. Less than the benchmark value To achieve automatic power reduction protection; when When the value is positive, for example, when a large gap dominates, the output energy increases;

[0116] When the drift is too large, the compensation amount will be smoothly limited to... Within the range; based on the median saturation gain and the maximum allowable compensation scaling factor. The final gain coefficient is calculated, and this coefficient is used to compare the reference energy value. Make corrections and generate dynamic energy output commands. This design physically eliminates energy overload caused by sensor noise or extreme operating conditions from the algorithm's underlying layer, avoiding the risk of battery pack burn-out or equipment damage.

[0117] Example 4:

[0118] Calculating the dynamic risk index includes:

[0119] Obtain the physical state feature vector of the current scan cycle and the physical state feature vector of the previous scan cycle;

[0120] The normalized physical state fluctuation rate is calculated based on the physical state feature vector of the current scan cycle and the physical state feature vector of the previous scan cycle.

[0121] Obtain static drift weights and dynamic fluctuation weights;

[0122] The static risk component is obtained by squared the product of the comprehensive coupling drift index and the static drift weight.

[0123] The dynamic risk component is obtained by calculating the square of the product of the normalized physical state fluctuation rate and the dynamic fluctuation weight.

[0124] The dynamic risk index is generated by summing the static and dynamic risk components and taking the square root.

[0125] Calculating the normalized physical state fluctuation rate includes:

[0126] Calculate the changes in vertical clearance, thermal accumulation temperature rise, and normal clamping force in adjacent PLC scanning cycles, respectively.

[0127] Divide the change by the corresponding preset standard process reference to obtain the dimensionless change ratio;

[0128] Calculate the sum of squares of dimensionless rates of change;

[0129] The square root of the sum of squares is taken to obtain the modulus of change.

[0130] Dividing the change modulus by the PLC scan cycle yields the normalized physical state fluctuation rate.

[0131] This embodiment further discloses the calculation logic of the dynamic risk index, which aims to accurately distinguish between compensable process fluctuations and system instability risks;

[0132] The system obtains the current scan cycle. physical state feature vector Compared to the previous scan cycle physical state feature vector Based on these two vectors, the system calculates the normalized physical state fluctuation rate. To completely eliminate the dimensional differences of the various physical parameters and accurately characterize the combined severity of changes in the state of multiple physics fields, a mathematical model based on the Euclidean norm rate of change is constructed, with the specific formula as follows:

[0133] ;

[0134] in, The PLC controller's scan cycle is expressed in seconds. This formula calculates the differences in vertical clearance, thermal accumulation temperature rise, and normal clamping force between adjacent PLC scan cycles, and then divides these differences by the corresponding preset standard process reference. , , This yields three dimensionless instantaneous change ratios; the squares of these three ratios are summed and the square root is taken to calculate the displacement modulus in state space; this modulus is then divided by the time interval. Thus, a quantity with frequency dimensions is obtained ( The overall volatility rate ;

[0135] Based on this, the system obtains the static drift weights from the recipe storage area. and dynamic fluctuation weight Among them, static drift weight The dimensionless constant is used to characterize the degree of attention paid to cumulative deviation; while the dynamic fluctuation weight... Specifically defined as a coefficient with the dimension of time, measured in seconds, used to balance dimensions and calibrate the system's sensitivity to mutation rates; both are obtained through regression analysis of historical fault samples, such as cold solder joints and overvoltage alarms; Dynamic Risk Index The calculation formula is as follows:

[0136] ;

[0137] In this model, due to It is a dimensionless exponent, and The unit is seconds and The unit is The product of seconds is also converted into a dimensionless value, therefore the formula satisfies the strict principle of dimensional consistency; although In Example 2, the values ​​were defined as signed numerical values ​​to distinguish the control direction, but in this risk assessment model, they are calculated using square operations. The system can correctly convert positive drift, such as large gaps, and negative drift, such as overheating, into positive risk accumulation, ensuring that an alarm is triggered as long as the magnitude exceeds the limit, regardless of the direction of the deviation.

[0138] Among them, static risk component This reflects the degree to which the current state deviates from the ideal baseline, while the dynamic risk component... It can keenly detect the drastic changes in system state; this dual-weighting mechanism combining static and dynamic factors endows the system with the dynamic perception capability of sudden mechanical failures. For example, when a clamp suddenly loosens, although static drift... It may not have accumulated to the point of exceeding the standard, but It will surge instantaneously due to drastic changes in physical quantities, thus making The rapid breach of the threshold triggers emergency stop protection, greatly improving the safety and fault tolerance of the production process.

[0139] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended 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.

Claims

1. A PLC-based method for controlling the production process of battery packs, characterized in that, include: Physical field data are collected using a laser contour sensor, an infrared thermal imager, and a servo pressure sensor. Step 1: Map the physical field data to a unified spatiotemporal coordinate system to construct a physical state feature vector; Step 2: Generate a comprehensive coupling drift index based on the physical state feature vector and the preset standard process benchmark; Step 3: Based on the comprehensive coupling drift index, a dynamic energy output command is generated using a nonlinear adaptive compensation model; Step 4: Write the dynamic energy output command into the actuator to perform closed-loop energy compensation; Step 5: Calculate the dynamic risk index based on the physical state feature vector; Step 6: In response to the dynamic risk index exceeding the preset risk threshold, an emergency stop command is triggered and defective products are marked. Step 7: In response to the dynamic risk index being less than or equal to the preset risk threshold, maintain the execution of the dynamic energy output command; Constructing physical state feature vectors includes: Obtain the vertical clearance between the busbar and the terminal block; Obtain the heat accumulation temperature rise of the current region relative to the ambient temperature; Obtain the normal clamping force; The vertical gap, thermal accumulation temperature rise, and normal clamping force are encapsulated as a physical state feature vector; The preset standard process benchmarks include standard clearance, standard pressure, upper limit of equipment safe temperature, and standard operation cycle for a single process. Generate the comprehensive coupling drift index, including: Calculate the ratio of the geometric deviation of the vertical clearance relative to the standard clearance; Based on the thermal accumulation temperature rise, the upper limit of the equipment's safe temperature, and the standard operating cycle of a single process, the deviation ratio of the thermodynamic dimension is calculated. Calculate the ratio of the mechanical deviation of the normal clamping force relative to the standard pressure; The geometric deviation ratio, thermodynamic dimension deviation ratio, and mechanical deviation ratio are algebraically weighted and summed to generate the comprehensive coupling drift index. ; Composite Coupling Drift Index The specific calculation formula is as follows: ； in: The preset process weighting coefficients, and satisfying ; The preset standard process heating curve at the current moment The theoretical temperature rise value; Scan cycle number; Vertical gap; Standard clearance; Heat accumulation and temperature rise; The upper limit of the safe temperature for the equipment; Normal clamping force; Standard pressure; Using a nonlinear adaptive compensation model, dynamic energy output commands are generated, including: Obtain the preset baseline energy value, maximum allowable compensation ratio factor, and adjustment sensitivity coefficient of the process formulation; Calculate the product of the overall coupling drift index and the regulation sensitivity coefficient; Applying the hyperbolic tangent function to the product yields the intermediate value of the saturation gain. The gain coefficient is calculated based on the median value of the saturated gain and the maximum allowable compensation ratio factor. The reference energy value is corrected using a gain factor to generate a dynamic energy output command. ; Dynamic energy output command The specific mathematical model is as follows: ； in, Dynamic energy output command; Baseline energy value; : Maximum allowable compensation ratio factor; Adjust the sensitivity coefficient; Calculating the normalized physical state fluctuation rate includes: Calculate the changes in vertical clearance, thermal accumulation temperature rise, and normal clamping force in adjacent PLC scanning cycles, respectively. Divide the change by the corresponding preset standard process reference to obtain the dimensionless change ratio; Calculate the sum of squares of dimensionless rates of change; The square root of the sum of squares is taken to obtain the modulus of change. Divide the change modulus by the PLC scan cycle to obtain the normalized physical state fluctuation rate. Normalized physical state fluctuation rate The specific formula is as follows: ； in, The scan cycle of the PLC controller, in seconds; : Vertical gap of the previous scan cycle; : Thermal accumulation and temperature rise in the previous scan cycle; Normal clamping force in the previous scan cycle; Calculating the dynamic risk index includes: Obtain the physical state feature vector of the current scan cycle and the physical state feature vector of the previous scan cycle; The normalized physical state fluctuation rate is calculated based on the physical state feature vector of the current scan cycle and the physical state feature vector of the previous scan cycle. Obtain static drift weights and dynamic fluctuation weights; The static risk component is obtained by squared the product of the comprehensive coupling drift index and the static drift weight. The dynamic risk component is obtained by calculating the square of the product of the normalized physical state fluctuation rate and the dynamic fluctuation weight. The dynamic risk index is generated by summing the static and dynamic risk components and taking the square root. Dynamic Risk Index The calculation formula is as follows: ； In this model, Dynamic risk index; Static drift weights; Dynamic fluctuation weight; Normalized physical state fluctuation rate.

2. A PLC-based battery pack production process control system, applied to the PLC-based battery pack production process control method described in claim 1, characterized in that, include: The data acquisition unit is used to acquire physical field data using a laser profile sensor, an infrared thermal imager, and a servo pressure sensor. The state mapping unit is used to map physical field data to a unified spatiotemporal coordinate system and construct physical state feature vectors. The drift quantization unit is used to generate a comprehensive coupled drift index based on the physical state feature vector and a preset standard process benchmark. An adaptive control unit is used to generate dynamic energy output commands based on a comprehensive coupling drift index and a nonlinear adaptive compensation model. The closed-loop execution unit is used to write dynamic energy output commands into the execution mechanism and perform closed-loop energy compensation. The risk assessment unit is used to calculate the dynamic risk index based on the physical state feature vector. The blocking control unit is used to trigger an emergency stop command and mark defective products in response to a dynamic risk index exceeding a preset risk threshold. And in response to the dynamic risk index being less than or equal to a preset risk threshold, the execution of the dynamic energy output command is maintained.

Images