Lithium battery wrench monitoring system comprising intelligent sensor

Abstract

The present application relates to the technical field of intelligent electric tools and industrial automation assembly detection, specifically to a lithium battery wrench monitoring system containing intelligent sensors; containing multi-dimensional data acquisition, energy flow modeling, dynamic residual observation, state decoupling and closed-loop feedback modules; the system solves the theoretical benchmark through the electromechanical energy transmission model, and converts the comparison between the actual output and the theoretical value into the dissipation rate residual; the core is to decouple the fluctuation root according to the residual characteristics: the residual is determined as power disturbance within the tolerance range, and as mechanical failure beyond the range; the present application solves the confusion problem of power limited torque attenuation and mechanical impedance type abnormality, and realizes the accurate distinction of tool performance decline and process abnormality.

Description

Technical Field

[0001] This invention relates to the field of intelligent power tools and industrial automation assembly and inspection technology, specifically to a lithium-ion battery wrench monitoring system incorporating intelligent sensors. Background Technology

[0002] The evolution of industrial automation assembly technology has led to the increasingly widespread application of lithium-ion battery tightening tools in high-cycle scenarios such as automotive assembly lines, which in turn increases the complexity of their operational data. Currently, monitoring tightening quality mainly relies on traditional torque-angle curves or static amplitude determination methods. However, in resource-constrained embedded working conditions, this method, which relies on a single physical quantity or static threshold, has significant limitations. Because fluctuations in lithium-ion battery voltage can have a nonlinear coupling effect on motor output characteristics, existing technologies struggle to effectively distinguish between soft performance degradation caused by insufficient battery power and process-side mechanical anomalies caused by bolt stripping or misthreading. This confusion can easily lead to false alarms in the monitoring system, resulting in unnecessary re-tightening, scrapping, and production downtime, severely impacting production efficiency and cost control. Therefore, how to eliminate the interference of power fluctuations on monitoring accuracy and achieve precise decoupling between tool-side disturbances and process-side anomalies has become an urgent problem to be solved in this field. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a lithium-ion battery wrench monitoring system incorporating intelligent sensors. Specifically, the technical solution of this invention includes: The system includes a multi-dimensional data acquisition module, which synchronously acquires the input electrical parameters and output mechanical parameters of the system during the tightening operation of the lithium-ion battery wrench, and generates a multi-dimensional time-series dataset; an energy flow modeling module, which pre-sets an electromechanical energy transfer model, calculates the theoretical mechanical output value based on the input electrical parameters, and uses the theoretical mechanical output value as a virtual reference benchmark; a dynamic residual observation module, which compares the actual output mechanical parameters acquired by the multi-dimensional data acquisition module with the theoretical mechanical output value to generate a dissipation rate residual in the energy transfer process; a state decoupling module, which, based on the numerical characteristics of the dissipation rate residual, separates the root causes of fluctuations in the output mechanical parameters into tool-side power supply disturbances or process-side mechanical anomalies, and generates a classification result; and a closed-loop feedback module, which, in response to the classification result, adaptively compensates the control parameters of the lithium-ion battery wrench or generates an operation quality alarm signal.

[0004] Preferably, the multidimensional data acquisition module includes: a signal synchronization unit, used to synchronously lock four signals—battery terminal voltage, bus current, motor speed, and output shaft torque—at a preset high-frequency sampling rate; and a density calculation unit, used to calculate the input power flow rate based on the battery terminal voltage and the bus current, and to calculate the output work density based on the output shaft torque and the motor speed, using the input power flow rate and the output work flow rate as the core features of the multidimensional time series dataset.

[0005] Preferably, the energy flow modeling module includes: a transfer function storage unit for storing a standard energy transfer function, which defines the relationship between the efficiency of converting motor current into output torque and the rotational speed under ideal friction coefficient and standard battery internal resistance conditions; and a benchmark extrapolation unit for inputting the currently acquired battery terminal voltage and bus current into the standard energy transfer function to predict the theoretical torque value that the system should achieve under the current power supply state without relying on actual torque sensor readings.

[0006] Preferably, the dynamic residual observation module includes: a difference calculation unit, used to calculate in real time the difference between the actual output shaft torque obtained by the multi-dimensional data acquisition module and the theoretical torque value generated by the benchmark extrapolation unit; and a residual quantization unit, used to normalize the difference to generate the dissipation rate residual, which characterizes the degree to which the actual energy transfer efficiency deviates from the theoretical model.

[0007] Preferably, the state decoupling module includes: a power disturbance identification unit, used to monitor the actual output shaft torque and the dissipation rate residual; when the actual output shaft torque is lower than a preset minimum torque threshold and the dissipation rate residual remains within a preset zero tolerance range, the current state is determined to be a soft performance degradation caused by insufficient battery power, and a tool-side power disturbance signal is generated; the zero tolerance range is used to define the allowable error range between the theoretical model and the actual system under normal operating conditions.

[0008] Preferably, the state decoupling module further includes: a mechanical anomaly identification unit, used to monitor the dissipation rate residual; when the dissipation rate residual exceeds the preset zero tolerance range, it determines that there is a nonlinear mechanical resistance change in the current energy transmission path and generates a process-side mechanical anomaly signal; the process-side mechanical anomaly signal corresponds to physical conditions such as bolt stripping, misthreading, or abnormal friction coefficient, and the mechanical anomaly identification unit marks such conditions as unqualified tightening quality.

[0009] Preferably, the closed-loop feedback module includes: an energy injection compensation unit, used to calculate the incremental duty cycle of the pulse width modulation signal within the thermal tolerance range of the motor in response to the power disturbance signal on the tool side; and a drive execution unit, used to adjust the drive voltage of the motor based on the incremental duty cycle, thereby increasing the input energy to offset the effect of the battery voltage drop, so that the actual output shaft torque approaches the preset target torque threshold, and maintains the judgment of qualified tightening quality after successful compensation.

[0010] Preferably, it further includes: a health prediction module, used to extract historical trend data of the dissipation rate residual, identify the baseline drift of the dissipation rate residual over time, and when the baseline drift exceeds a preset aging threshold, reversely infer the increase in battery internal resistance and output a battery health warning signal.

[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. This system constructs an electromechanical energy transfer model and generates a virtual reference benchmark, converting the comparison between the actual output and the theoretical value into a dissipation rate residual. The system can perform binary classification based on the numerical characteristics of the residual: when the actual torque decreases but the residual remains within the zero tolerance range, it is determined to be a soft performance degradation caused by insufficient battery power; when the residual exceeds the tolerance range, it is determined to be a mechanical fault such as bolt stripping or abnormal friction. This mechanism effectively solves the technical problem that traditional monitoring methods cannot distinguish between power-limited torque attenuation, commonly known as lack of strength to tighten, and mechanical impedance anomalies, commonly known as inability to tighten. It eliminates false alarms caused by lithium battery voltage fluctuations and avoids unnecessary re-tightening or scrapping operations. 2. Through the closed-loop feedback module, this system transforms the lithium-ion battery wrench from a soft characteristic source affected by battery status into a hard characteristic source with voltage stabilization capability. When a drop in battery voltage is detected, the system automatically calculates and increases the duty cycle of the pulse width modulation signal within the motor's thermal tolerance range, injecting incremental energy to offset the impact of insufficient voltage. This compensation mechanism enables the tool to drive the actual output torque close to the preset target even when the battery is half-charged or the voltage fluctuates, thereby extending the effective number of operations per charge and ensuring the stability of the tightening process under different power levels. 3. Unlike traditional monitoring methods that rely on static amplitude, this system uses a multi-dimensional data acquisition module to synchronously lock onto high-frequency signals and calculates the input power density and output work density as core features. This dynamic analysis method based on energy flow rate enables the model to keenly capture millisecond-level energy impacts, such as those generated at the moment a bolt is seated. Even in complex industrial environments, the system can accurately identify sudden increases in frictional heat generation or sudden disappearance of load, such as abrupt changes in nonlinear mechanical resistance like stripping, based on the normalized dissipation rate residual. This provides a high-precision quality control method for precision assembly. 4. This system taps into the long-term value of monitoring data. By extracting historical trend data of dissipation rate residuals, it identifies the drift of the residual baseline. Using this drift, it inversely extrapolates the increase in battery internal resistance, thereby quantifying the battery's health status. This unexpected synergistic effect allows the system to diagnose battery aging without adding extra battery testing circuitry while performing tightening tasks, and to issue early warnings before battery performance deteriorates to the point of affecting operations, effectively reducing the risk of production line downtime due to sudden battery failure. Attached Figure Description

[0012] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation

[0013] 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.

[0014] Example 1: Please see Figure 1 A lithium-ion battery wrench monitoring system incorporating intelligent sensors includes: a multi-dimensional data acquisition module for simultaneously acquiring the system's input electrical parameters and output mechanical parameters during the tightening operation of the lithium-ion battery wrench, and generating a multi-dimensional time-series dataset; an energy flow modeling module for pre-setting an electromechanical energy transfer model, calculating the theoretical mechanical output value based on the input electrical parameters, and using the theoretical mechanical output value as a virtual reference benchmark; a dynamic residual observation module for comparing the actual output mechanical parameters acquired by the multi-dimensional data acquisition module with the theoretical mechanical output value, generating a dissipation rate residual for the energy transfer process; a state decoupling module for separating the root causes of output mechanical parameter fluctuations into tool-side power supply disturbances or process-side mechanical anomalies based on the numerical characteristics of the dissipation rate residual, and generating a classification result; and a closed-loop feedback module for adaptively compensating the control parameters of the lithium-ion battery wrench or generating an operation quality alarm signal in response to the classification result.

[0015] This embodiment details the core architecture of a lithium-ion battery wrench monitoring system incorporating intelligent sensors, aiming to solve the challenge of high-precision monitoring of tightening quality in resource-constrained embedded working conditions. The system connects to the brushless motor controller and output shaft torque sensor of the lithium-ion battery wrench via a multi-dimensional data acquisition module, simultaneously acquiring the system's input electrical parameters and output mechanical parameters at a high sampling rate of 1kHz-10kHz. This data is then encapsulated into a multi-dimensional time-series dataset. This serves as the basis for subsequent energy flow analysis; The energy flow modeling module constructs an electromechanical energy transfer model based on first principles of physics, reads the input electrical parameters in real time, calculates the theoretical mechanical output value that should be generated under the current power supply capacity, and establishes this value as the virtual reference benchmark of the monitoring system. On this basis, the dynamic residual observation module performs real-time differential calculation between the actual collected output mechanical parameters and the above-mentioned theoretical mechanical output value to generate the dissipation rate residual of the energy transfer process. The state decoupling module uses the numerical characteristics of the dissipation rate residual to perform binary classification of the root causes of output fluctuations using a logic decision tree, clearly distinguishing between power supply disturbances on the tool side and mechanical anomalies on the process side, and generating classification judgment results. The closed-loop feedback module responds to the classification judgment results. If it is a power supply disturbance, it adaptively compensates the control parameters to maintain constant output. If it is a mechanical anomaly, it directly generates a work quality alarm signal. This embodiment breaks through the limitations of traditional monitoring that relies solely on torque-angle curves by introducing a virtual reference benchmark. In high-cycle production scenarios such as automobile assembly lines, the system can effectively distinguish between insufficient tightening power due to low battery charge and inability to tighten due to stripped bolt threads. It eliminates false alarms caused by lithium battery voltage fluctuations, significantly improves the accuracy of yield determination for tightening operations, and avoids process misjudgments caused by fluctuations in tool performance.

[0016] Example 2: The multidimensional data acquisition module includes: a signal synchronization unit, used to synchronously lock four signals: battery terminal voltage, bus current, motor speed, and output shaft torque at a preset high-frequency sampling rate; and a density calculation unit, used to calculate the input power flow rate based on the battery terminal voltage and bus current, and to calculate the output work density based on the output shaft torque and motor speed, using the input power flow rate and output work flow rate as the core features of the multidimensional time series dataset.

[0017] This embodiment further refines the data acquisition and feature extraction process of the multi-dimensional data acquisition module; the signal synchronization unit adopts an FPGA-based hardware triggering mechanism to send a unified synchronization lock signal, which serves as a high-precision clock pulse and simultaneously triggers the ADC to read the battery terminal voltage. and bus current And latch the motor speed of the encoder. and the output shaft torque of the strain gauge This ensures that the alignment error of the four signals on the time axis is less than 10 microseconds; the density calculation unit performs physical characterization calculations of energy density, incorporating input power density. With output work density The calculation formula is as follows: ; ; in, The source is a synchronously acquired voltage sensor, and its physical meaning is the instantaneous voltage at the battery terminal, with the unit being V; The source is a Hall current sensor, and its physical meaning is the instantaneous current of the busbar, with the unit being A; The source is the actual torque sensor, and its physical meaning is the actual output shaft torque, with the unit being N·m; The source is a Hall encoder, and its physical meaning is the real-time speed of the motor, measured in RPM. The source is the density calculation unit, and its physical meaning is the input power density. In this embodiment, it is characterized by instantaneous electric power for tools with a fixed volume, and the unit is W. The source is the density calculation unit, and its physical meaning is the output work density. In this embodiment, it is characterized by instantaneous mechanical power for tools of fixed volume, and the unit is W. The system uses the calculated input power density and output work density as the core features of the multidimensional time series dataset for subsequent modules to call; This embodiment calculates power density and uses it as a core feature, so that the system no longer relies solely on static amplitude, but performs dynamic analysis based on energy flow rate. This design enables the monitoring model to keenly capture millisecond-level energy impacts such as those generated at the moment a bolt is seated, laying a solid data foundation for high-precision residual calculation. It is particularly suitable for precision assembly processes with extremely high requirements for transient response.

[0018] Example 3: The energy flow modeling module includes: a transfer function storage unit, which stores the standard energy transfer function, which defines the relationship between the efficiency of converting motor current into output torque and the speed under ideal friction coefficient and standard battery internal resistance conditions; and a benchmark extrapolation unit, which inputs the currently collected battery terminal voltage and bus current into the standard energy transfer function to predict the theoretical torque value that the system should achieve under the current power supply state without relying on actual torque sensor readings.

[0019] This embodiment details the specific logic of the energy flow modeling module in constructing a virtual reference benchmark; the transfer function storage unit stores pre-calibrated standard energy transfer functions. This function is a motor efficiency map measured under standard laboratory conditions (full charge, standard temperature, and standard load), defining a baseline efficiency coefficient for the conversion of electrical energy into mechanical energy at specific speeds and currents. It should be noted that this efficiency coefficient... It represents the mechanical torque transmission efficiency, i.e., output torque / electromagnetic torque, rather than simply the power conversion efficiency; It is important to note that the standard energy transfer function Based on the experimental calibration of the whole system, its numerical mapping relationship implicitly includes the modulation and conversion efficiency of the controller from DC bus current to motor phase current under different duty cycles, as well as the mechanical transmission loss of the reduction mechanism, thus supporting the direct utilization of bus current. Map the final output shaft torque; This definition ensures that when the motor speed... The value is 0, meaning during the tightening and holding phase. The static transmission coefficient remains non-zero, thus avoiding the physical paradox of the power efficiency model calculating zero at zero speed. Considering the nonlinear effects of battery voltage drops on motor magnetic field saturation and driver dead zone, this embodiment introduces a voltage coupling coefficient into the standard energy transfer function. The benchmark simulation unit uses the currently acquired voltage and current, combined with the motor constants, to execute the theoretical torque prediction formula: ; in, Motor torque constant, in units of ; Efficiency lookup factor under standard operating conditions; in, Voltage coupling coefficient, its calculation formula is: ; To ensure the validity of the physical meaning and prevent negative values ​​from being calculated under extremely low-pressure conditions, the formula needs to be subject to non-negativity constraints: ; in, The preset minimum coupling coefficient, such as 0.1, ensures that the theoretical torque prediction value always remains in the correct direction before the battery is undervoltage cutoff. This is a preset voltage sensitivity coefficient used to characterize the voltage deviation from the nominal value. It has a slight suppressive effect on torque output capability; This unit outputs This serves as a virtual reference benchmark for the system, independent of the actual torque sensor readings; This embodiment constructs a digital twin torque signal that fully reflects the torque state that the mechanical transmission system should have when it is perfect and the input electrical energy is converted normally. When the battery voltage drops and causes the input current to change, this theoretical value is dynamically adjusted accordingly, thus providing a mathematically meaningful benchmark for distinguishing between performance degradation caused by lack of power and mechanical failure caused by faulty parts. It should be noted that the electromechanical energy transfer model is based on steady-state energy balance; when mechanical abnormalities such as bolt stripping occur, the inertial acceleration torque generated by the motor rotor... This will cause dynamic decoupling between the electromagnetic torque generated by the actual current and the steady-state load torque of the output shaft. At this time, the steady-state model cannot predict this inertial component, which leads to a sharp increase in the dissipation rate residual. The system uses this model failure feature to capture transient mechanical faults.

[0020] Example 4: The dynamic residual observation module includes: a difference calculation unit, used to calculate in real time the difference between the actual output shaft torque obtained by the multi-dimensional data acquisition module and the theoretical torque value generated by the benchmark extrapolation unit; and a residual quantization unit, used to normalize the difference and generate dissipation rate residuals, which characterize the degree to which the actual energy transfer efficiency deviates from the theoretical model.

[0021] This embodiment illustrates the specific steps of the dynamic residual observation module in generating diagnostic indicators; the difference calculation unit acquires the actual output shaft torque in real time. Compared with theoretical torque value The residual quantization unit performs normalization calculations to eliminate the influence of the system operating conditions on the error amplitude, generating normalized dissipation rate residuals. The calculation formula is as follows: ; in, The source is a torque sensor, and its physical meaning is the actual output shaft torque, with the unit being N·m; The source is the energy flow modeling module, and its physical meaning is the theoretical torque value, with the unit being N·m; The source is the configuration of the operating parameters, and its physical meaning is the target torque set in the current tightening process, which is a constant. The source is a system preset constant, the physical meaning is to prevent the removal of the zero minimum value, and the unit is N·m; The source is the calculation result, and its physical meaning is the dissipation rate residual, which is dimensionless. This embodiment transforms complex physical signals into an intuitive health index by calculating the normalized dissipation rate residual. When the system is working normally, the residual always approaches 0 regardless of the fluctuation of the battery power. This design greatly simplifies the complexity of subsequent logical judgments, enabling the system to maintain consistent discrimination sensitivity in different torque levels of work tasks.

[0022] Example 5: The state decoupling module includes a power disturbance identification unit, which monitors the actual output shaft torque and dissipation rate residual. When the actual output shaft torque is lower than the preset minimum torque threshold and the dissipation rate residual remains within the preset zero tolerance range, the current state is determined to be a soft performance degradation caused by insufficient battery power, and a tool-side power disturbance signal is generated. The zero tolerance range is used to define the allowable error range between the theoretical model and the actual system under normal operating conditions.

[0023] This means that when the residual is within this range, it indicates that the actual output follows the decay trend predicted by the theoretical model. Follow The decrease was consistent with the physical expectation and no abnormal dissipation outside the model occurred; This embodiment describes the judgment logic of the power disturbance identification unit in the state decoupling module; this unit monitors in real time. and and enforce the joint conditional judgment: Condition one: ; Condition two: ; in, The source is the process setting, and its physical meaning is the minimum torque threshold for a qualified tightening, with the unit being N·m; The source is the upper limit of the 3σ value of the residual statistical distribution collected under standard load during the tool's factory calibration phase. The physical meaning of is the zero-value tolerance range, which is used to define the inherent modeling noise floor of the theoretical model and the actual system, and is dimensionless. The source is a system preset, and its physical meaning is the length of the sliding time window used for calculating the residual mean, i.e., the number of sampling points, in units of points; If the above conditions are met simultaneously, that is, the actual torque does not meet the standard but the residual is extremely small and within the noise floor range, it indicates that the actual value matches the theoretical value. The motor has tried its best to convert the input energy but is limited by the input source. The system determines that the current state is a power supply disturbance on the tool side and generates a corresponding signal. This embodiment accurately identifies the degradation of soft performance caused by battery depletion or excessive internal resistance. In actual production lines, this logic avoids misreporting insufficient torque caused by low battery as bolt tightening failure, thereby preventing operators from performing unnecessary re-tightening or scrapping operations, effectively reducing the misjudgment rate and rework costs on the production line.

[0024] Example 6: The state decoupling module also includes a mechanical anomaly identification unit, which monitors the dissipation rate residual. When the dissipation rate residual exceeds the preset zero tolerance range, it determines that there is a nonlinear mechanical resistance change in the current energy transmission path and generates a process-side mechanical anomaly signal. The process-side mechanical anomaly signal corresponds to physical conditions such as bolt stripping, misthreading, or abnormal friction coefficient. The mechanical anomaly identification unit marks such conditions as unqualified tightening quality.

[0025] This embodiment describes the execution logic of the mechanical anomaly identification unit in the state decoupling module; this unit continuously monitors the dissipation rate residual. ; in response If the residual integral within the time window exceeds the preset energy threshold, the system determines that the energy has been dissipated or leaked unexpectedly during transmission. This physical phenomenon corresponds to the actual torque deviating significantly from the theoretical prediction value, such as a sharp increase in frictional heat generation or the sudden disappearance of load due to thread stripping. Subsequently, the system generates a mechanical abnormality signal on the process side, clearly indicating that the current working condition is bolt stripping, misthreading, or abnormal friction coefficient, and marks this operation as unqualified tightening quality. This embodiment enables the system to detect minute mechanical faults by monitoring abrupt changes in residuals. Even when the battery is fully charged and the system is in a strong power state, if slippage occurs, the actual torque decreases while the theoretical torque remains high. The residual will increase instantly and trigger an alarm, ensuring strict control over the quality of mechanical connections under various battery charge conditions.

[0026] Example 7: The closed-loop feedback module includes: an energy injection compensation unit, which is used to calculate the incremental duty cycle of the pulse width modulation signal within the thermal tolerance range of the motor in response to the power disturbance signal on the tool side; and a drive execution unit, which is used to adjust the drive voltage of the motor based on the incremental duty cycle, thereby offsetting the impact of battery voltage drop by increasing input energy, making the actual output shaft torque approach the preset target torque threshold, and maintaining the judgment of qualified tightening quality after successful compensation.

[0027] This embodiment describes the automatic compensation mechanism of the closed-loop feedback module for power supply disturbances; the energy injection compensation unit responds to the received tool-side power supply disturbance signal and enables the automatic compensation mechanism based on... A virtual thermal model is used to quantify the heat tolerance range of the motor, and the calculation formula is as follows: ; Among them, the system initialization is set Only when the real-time heat accumulation Less than the preset heat capacity limit Only then is the system allowed to calculate the incremental duty cycle of the pulse width modulation signal. ;like The system then triggers thermal protection logic, forcing... Suspend compensation to prevent motor overheating; among them, The system sampling time interval is expressed in seconds (s). This formula represents the virtual heat accumulation based on the square integral of the current. It employs the equivalent thermal energy characterization method, assuming the equivalent resistance of the loop. The temperature rise trend of the motor windings can be directly characterized by the integral of the square of the current, which has already been normalized. Its physical unit is defined as... ; The source is the motor heat dissipation characteristic calibration, and its physical meaning is the dimensionless heat dissipation coefficient, which characterizes the natural heat decay rate within a single sampling period; to ensure that it can be implemented by those skilled in the art, its calculation formula is defined as follows: ;in, The thermal time constant of the motor is expressed in seconds (s). In this embodiment, it is preferably a constant between 30s and 60s, representing the time required for the motor temperature rise to reach 63.2% of its steady-state value. This calculation formula is based on a discretized approximation of a first-order thermal hysteresis element, and its effectiveness is predicated on the system sampling time interval. Much smaller than the thermal time constant of the motor ,Right now To ensure real-time compensation, the control loop frequency of the drive execution unit is set to be no lower than the sampling frequency of the multi-dimensional data acquisition module, for example... To ensure energy injection is completed within a millisecond-level time window of a single tightening operation; the drive unit adjusts the motor's drive voltage based on the following control law: ; ; Final output duty cycle It needs to be processed by upper and lower limits, and the formula is as follows: ; This saturation limit ensures that the system will not output an illegal duty cycle when an abnormal surge in bus voltage results in a calculated negative increment. in, The source is the original control loop calculation, and its physical meaning is the basic duty cycle, which is dimensionless. The source is the calculation result, and its physical meaning is the incremental duty cycle, which is dimensionless. The source is the controller hardware parameters, and its physical meaning is the maximum allowable duty cycle of PWM, such as 0.95 or 1.0, to prevent calculation overflow; The source is a system setting; its physical meaning is the nominal battery voltage, and the unit is V. The data source is real-time acquisition, and its physical meaning is the current measured voltage drop, with the unit being V. The source is the motor's thermal withstand capability test, and its physical meaning is the compensation gain coefficient, with the unit being 1 / V; The source is the motor's thermal characteristic specifications or experimental calibration; its physical meaning is the maximum allowable thermal accumulation threshold of the motor windings, and the unit is... When this value is exceeded, it indicates that the motor is at risk of overheating; After output saturation limiting Adjusting the motor voltage increases the input energy to offset the impact of battery voltage drops, driving the actuator to force the actual output shaft torque to approach the preset target torque threshold, and maintaining the tightening quality qualified judgment after successful compensation; This embodiment transforms the lithium-ion wrench from a soft characteristic source affected by the battery state into a hard characteristic source with voltage stabilization characteristics. When the battery voltage drops, it automatically increases the throttle to inject more energy, ensuring that the torque output to the bolt meets the standard even when the battery is half-charged. This compensation mechanism significantly extends the effective number of operations per charge, improving the tool's endurance and operational consistency.

[0028] Example 8: It also includes a health prediction module, which is used to extract historical trend data of dissipation rate residuals, identify the baseline drift of dissipation rate residuals over time, and when the baseline drift exceeds the preset aging threshold, reverse the increase in battery internal resistance and output a battery health warning signal.

[0029] This embodiment describes the logic of the health prediction module in mining the long-term value of data; the module extracts historical trend data of the dissipation rate residuals after each tightening; based on this, the system identifies the baseline drift of the residuals over time. The physical cause of baseline drift is battery aging, which increases its internal resistance. The significant increase resulted in an actual voltage drop under high current conditions exceeding that of the voltage coupling coefficient in Example 3. The linear compensation range; At this point, the actual output torque is affected. The decay rate is faster than the theoretical model. The prediction speed causes the residual baseline to drift to one side; to ensure data robustness, The calculation selects data from the stable segment where the torque is between 50% and 100% of the target value during the tightening process, and calculates the torque within this range. The arithmetic mean of the values ​​is used to eliminate non-stationary noise introduced by startup shock and shutdown oscillation; the system performs an internal resistance reverse derivation step to quantify the increase in battery internal resistance. The calculation formula is: ; Among them, mapping coefficients Based on battery electrochemical characteristics, such as the slope of the internal resistance-voltage curve, calibrated empirical coefficients are used to map dimensionless residual drift into resistance increments with physical units. ; The system calculates the battery health status (SOH) using the above deduction results or their equivalent drift characteristics. The calculation formula is as follows: ; in, The source is statistical calculation, and its physical meaning is the average residual value of the current period's stable segment, which is dimensionless; The source is system storage, and its physical meaning is the residual baseline value in the initial state of the battery, which is dimensionless; The source is a preset value, and its physical meaning is the aging threshold when the battery needs to be scrapped. It is dimensionless. The source is experimental calibration; its physical meaning is the sensitivity adjustment coefficient, which is dimensionless. Introduced in the formula This is to eliminate noise caused by measurement. The abnormal situation of SOH>100% that occurs at that time is introduced at the same time. The function is designed to prevent the calculation of a negative SOH value under extreme aging conditions, ensuring a closed-loop logic. In response to the baseline drift exceeding the preset aging threshold, i.e., the SOH is lower than the preset health line, the system outputs a battery health warning signal. This embodiment produces an unexpected synergistic effect: while monitoring the quality of bolt tightening, no additional battery testing circuit is needed. The health status of the battery can be diagnosed by analyzing the energy deviation during the tightening process, enabling predictive management of tool maintenance and reducing the risk of production line downtime due to sudden battery failure.

[0030] 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 lithium-ion battery wrench monitoring system incorporating intelligent sensors, characterized in that: include: The multi-dimensional data acquisition module is used to simultaneously acquire the input electrical parameters and output mechanical parameters of the system during the tightening operation of the lithium battery wrench, and generate a multi-dimensional time series dataset. The energy flow modeling module is used to pre-set the electromechanical energy transmission model, calculate the theoretical mechanical output value based on the input electrical parameters, and use the theoretical mechanical output value as a virtual reference benchmark. The dynamic residual observation module is used to compare the actual output mechanical parameters collected by the multidimensional data acquisition module with the theoretical mechanical output values ​​to generate the dissipation rate residual of the energy transmission process. The state decoupling module is used to separate the root cause of the output mechanical parameter fluctuation into tool-side power supply disturbance or process-side mechanical abnormality based on the numerical characteristics of the dissipation rate residual, and generate a classification judgment result; the closed-loop feedback module is used to adaptively compensate the control parameters of the lithium battery wrench or generate a work quality alarm signal in response to the classification judgment result. The dynamic residual observation module includes: a difference calculation unit, used to calculate in real time the difference between the actual output shaft torque obtained by the multi-dimensional data acquisition module and the theoretical torque value generated by the benchmark extrapolation unit; and a residual quantization unit, used to normalize the difference to generate the dissipation rate residual, which characterizes the degree to which the actual energy transfer efficiency deviates from the theoretical model. The state decoupling module includes a power disturbance identification unit, used to monitor the actual output shaft torque and the dissipation rate residual. When the actual output shaft torque is lower than a preset minimum torque threshold and the dissipation rate residual remains within a preset zero tolerance range, the current state is determined to be a soft performance degradation caused by insufficient battery power, and a tool-side power disturbance signal is generated. The zero tolerance range is used to define the allowable error range between the theoretical model and the actual system under normal operating conditions. The state decoupling module further includes a mechanical anomaly identification unit, used to monitor the dissipation rate residual. When the dissipation rate residual exceeds the preset zero tolerance range, it determines that there is a nonlinear mechanical resistance change in the current energy transmission path and generates a process-side mechanical anomaly signal. The process-side mechanical anomaly signal corresponds to physical conditions such as bolt stripping, misthreading, or abnormal friction coefficient. The mechanical anomaly identification unit marks such conditions as unqualified tightening quality.

2. The lithium-ion battery wrench monitoring system incorporating an intelligent sensor according to claim 1, characterized in that: The multidimensional data acquisition module includes: a signal synchronization unit, used to synchronously lock four signals—battery terminal voltage, bus current, motor speed, and output shaft torque—at a preset high-frequency sampling rate; and a density calculation unit, used to calculate the input power flow rate based on the battery terminal voltage and the bus current, and to calculate the output work density based on the output shaft torque and the motor speed, using the input power flow rate and the output work flow rate as the core features of the multidimensional time series dataset.

3. The lithium-ion battery wrench monitoring system incorporating an intelligent sensor according to claim 1, characterized in that: The energy flow modeling module includes: a transfer function storage unit for storing a standard energy transfer function, which defines the relationship between the efficiency of converting motor current into output torque and the rotational speed under ideal friction coefficient and standard battery internal resistance conditions; and a benchmark extrapolation unit for inputting the currently collected battery terminal voltage and bus current into the standard energy transfer function to predict the theoretical torque value that the system should achieve under the current power supply state without relying on actual torque sensor readings.

4. The lithium-ion battery wrench monitoring system incorporating an intelligent sensor according to claim 1, characterized in that: The closed-loop feedback module includes: an energy injection compensation unit, used to calculate the incremental duty cycle of the pulse width modulation signal within the thermal tolerance range of the motor in response to the power disturbance signal on the tool side; and a drive execution unit, used to adjust the drive voltage of the motor based on the incremental duty cycle, thereby increasing the input energy to offset the effect of the battery voltage drop, making the actual output shaft torque approach the preset target torque threshold, and maintaining the judgment of qualified tightening quality after successful compensation.

5. The lithium-ion battery wrench monitoring system incorporating an intelligent sensor according to claim 1, characterized in that: Also includes: The health prediction module is used to extract historical trend data of the dissipation rate residual, identify the baseline drift of the dissipation rate residual over time, and when the baseline drift exceeds a preset aging threshold, reverse the increase in battery internal resistance and output a battery health warning signal.

Images