Sand blender running state real-time monitoring system based on digital twinning

Abstract

The application relates to the field of digital twinning and intelligent monitoring of industrial equipment, in particular to a real-time monitoring system for the running state of a sand mixer based on digital twinning, which comprises a main control module, a variable frequency drive module, a resistance calculation module, a data processing module and a drive recovery module. The main control module monitors the state of a sand discharge door and the sand discharge duration, and outputs a short-time power-off instruction when preset sand discharge conditions are met. The variable frequency drive module blocks the pulse width modulation pulse output of a frequency converter, so that the main drive motor enters a free coasting state, the back electromotive force signal is collected, the rotor angular velocity at the start and end time of the power-off detection period is determined, the stator current and voltage are collected during the load mixing stage, and the real-time total resistance torque is output. The resistance calculation module calculates the mechanical friction resistance torque. The data processing module deducts the baseline to obtain the actual sand resistance torque and calculates the sand viscosity parameter. The drive recovery module recovers the pulse width modulation pulse output and controls the main drive motor to recover to the set rotating speed. The application can separate the mechanical wear resistance and the sand viscosity resistance without additionally setting internal sensors of the sand mixer, and reduces the baseline drift.

Description

Technical Field

[0001] This invention relates to the field of digital twins and intelligent monitoring of industrial equipment, specifically a real-time monitoring system for the operating status of a sand mixer based on digital twins. Background Technology

[0002] With the increasing demands on the stability of molding sand quality and continuous operation of equipment in the casting production process, sand mixers not only need to monitor the mixing state of molding sand in real time during operation, but also need to minimize the interference of factors such as mechanical wear of equipment and changes in bottom sand accumulation on the monitoring results. Therefore, how to accurately obtain operating state parameters that can reflect the true viscosity changes of molding sand under continuous production cycle has become an urgent problem to be solved in the field of intelligent monitoring of sand mixers. Traditional sand mixer operation status monitoring currently mainly relies on the following methods: directly detecting the mixing status by additionally deploying sensors inside the equipment, making overall estimates based on the current or power signal of the main drive motor, and compensating for the resistance signal using a fixed baseline or long-period average baseline. However, methods such as additional sensor deployment, overall estimation based on current or power, and fixed baseline or long-cycle average compensation all have certain drawbacks. For example, additional sensor deployment is difficult to adapt to the internal working conditions of sand mixers with high dust and high humidity, and increases the complexity of equipment deployment and maintenance. When making overall estimations based on current or power signals, mechanical wear resistance and molding sand viscosity resistance are coupled and difficult to separate effectively. Fixed baseline or long-cycle average compensation methods cannot reflect the dynamic changes of scraper wear, bottom sand accumulation, and residual sand additional resistance in a timely manner, which can easily lead to drift of the no-load resistance torque reference, causing deviation in the calculation of molding sand viscosity parameters and affecting the consistency and accuracy of digital twin analysis results between different batches. Summary of the Invention

[0003] The purpose of this invention is to provide a real-time monitoring system for the operating status of a sand mixer based on digital twins, and to solve the following technical problems: It avoids disrupting continuous production due to additional machine downtime required for measurement operations, and can separate mechanical wear resistance from molding sand viscosity resistance without the need for additional sensors inside the sand mixer. This reduces baseline drift caused by scraper wear and changes in bottom sand accumulation, thereby solving the problem of insufficient stability in molding sand quality prediction.

[0004] The objective of this invention can be achieved through the following technical solutions: A real-time monitoring system for the operating status of a sand mixer based on digital twins includes: The main control module is used to monitor the status of the sand discharge gate of the sand mixer and the duration of sand discharge, and outputs a short-term power-off command when the preset sand discharge conditions are met; The variable frequency drive module includes a frequency converter and a main drive motor driven by the frequency converter. It is used to block the pulse width modulation pulse output of the frequency converter after receiving a short-term power failure command, so that the main drive motor enters a free gliding state, collects back electromotive force signals and determines the first rotor angular velocity at the start time and the second rotor angular velocity at the end time of the power failure detection period; it is also used to collect the stator current and voltage of the main drive motor during the mixed load stage and output the real-time total resistance torque. The resistance calculation module is used to calculate the mechanical friction resistance torque of the current batch of sand unloading stage based on the rotor angular velocity at the start and end of the power outage detection period, the duration of the power outage detection period, the pre-calibrated equivalent rotational inertia of the system, the pre-determined residual sand resistance correction parameters, and the wind resistance and iron loss combined resistance torque obtained by referring to the pre-set motor loss mapping table. The data processing module is used to take the mechanical friction resistance torque as the reference for the no-load resistance torque, subtract the reference from the real-time total resistance torque to obtain the actual molding sand resistance torque, and calculate and output the molding sand viscosity parameters based on the actual molding sand resistance torque. The drive recovery module is used to determine the rotor speed and electrical angle based on the back electromotive force signal after the power failure detection period ends, restore the pulse width modulation pulse output, and control the main drive motor to return to the set speed.

[0005] As a further aspect of the present invention, when the sand unloading door is in the open state and the sand unloading duration reaches the preset sand unloading condition, the main control module determines that it has entered the power failure detection period.

[0006] As a further aspect of the present invention, the duration of the power outage detection period is set according to the preset allowable speed drop range and historical operating parameters. The frequency conversion drive module determines the first rotor angular velocity at the start of the power outage and the second rotor angular velocity at the end of the power outage during the power outage detection period, and transmits them to the resistance calculation module.

[0007] As a further aspect of the present invention, the resistance calculation module calculates the mechanical friction resistance torque based on the first rotor angular velocity, the second rotor angular velocity, the duration of the power outage detection period, the system's equivalent moment of inertia, the residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss.

[0008] As a further aspect of the present invention, the residual sand resistance correction parameter is determined based on the sand unloading duration and historical sand unloading test data.

[0009] As a further aspect of the present invention, at the end of the power failure detection period, the main control module stops outputting short-term power failure commands; the drive recovery module resumes pulse width modulation pulse output and controls the main drive motor to resume to the set speed; the resistance calculation module sends the mechanical friction resistance torque to the data processing module as a new no-load resistance torque reference.

[0010] As a further aspect of the present invention, in the loading and mixing stage of the next adjacent batch, the frequency conversion drive module outputs the real-time total resistance torque to the data processing module, and the data processing module calculates the actual molding sand resistance torque based on the real-time total resistance torque and the updated mechanical friction resistance torque of the previous adjacent batch.

[0011] As a further aspect of the present invention, during the load mixing cycle, the data processing module uses the mechanical friction resistance torque updated in the adjacent previous batch for benchmark subtraction.

[0012] The beneficial effects of this invention are: 1) This invention calculates the mechanical friction resistance torque based on the rotor angular velocity at the beginning and end of the power-off detection period, the system's equivalent moment of inertia, the residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss using a resistance calculation module. It can convert the mechanical resistance without electromagnetic driving force into the deceleration process during free sliding, thereby distinguishing mechanical wear resistance from motor body losses and residual sand additional resistance, improving the authenticity and usability of the no-load resistance torque benchmark. The residual sand resistance correction parameters are determined based on the sand unloading duration and historical sand unloading test data, enabling quantitative correction for actual working conditions where the sand unloading end is not absolutely no-load. This reduces the impact of residual sand additional resistance on the no-load resistance torque benchmark under different evacuation degrees, avoiding the drift and continuous accumulation of compensation errors caused by fixed baselines or simple no-load treatment in the prior art. Attached Figure Description

[0013] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of a real-time monitoring system for the operating status of a sand mixer according to an embodiment. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figure 1 A real-time monitoring system for the operating status of a sand mixer based on digital twins includes: The main control module is used to monitor the status of the sand discharge gate of the sand mixer and the duration of sand discharge, and outputs a short-term power-off command when the preset sand discharge conditions are met; The variable frequency drive module includes a frequency converter and a main drive motor driven by the frequency converter. It is used to block the pulse width modulation pulse output of the frequency converter after receiving a short-term power failure command, so that the main drive motor enters a free gliding state, collects back electromotive force signals and determines the first rotor angular velocity at the start time and the second rotor angular velocity at the end time of the power failure detection period; it is also used to collect the stator current and voltage of the main drive motor during the mixed load stage and output the real-time total resistance torque. The resistance calculation module is used to calculate the mechanical friction resistance torque of the current batch of sand unloading stage based on the rotor angular velocity at the start and end of the power outage detection period, the duration of the power outage detection period, the pre-calibrated equivalent rotational inertia of the system, the pre-determined residual sand resistance correction parameters, and the wind resistance and iron loss combined resistance torque obtained by referring to the pre-set motor loss mapping table. The data processing module is used to take the mechanical friction resistance torque as the reference for the no-load resistance torque, subtract the reference from the real-time total resistance torque to obtain the actual molding sand resistance torque, and calculate and output the molding sand viscosity parameters based on the actual molding sand resistance torque. The drive recovery module is used to determine the rotor speed and electrical angle based on the back electromotive force signal after the power failure detection period ends, restore the pulse width modulation pulse output, and control the main drive motor to return to the set speed.

[0016] This embodiment provides a real-time monitoring system for the operating status of a sand mixer based on digital twins. The system takes the sand unloading stage and the subsequent loaded mixing stage within a production cycle of the sand mixer as continuously associated processing objects. During the period when the sand unloading stage is close to emptying, the main control module issues a short-term power-off command to temporarily block the pulse width modulation pulse output of the frequency conversion drive module, and the main drive motor enters a free gliding state. The rotor angular velocity before and after the power-off detection period is determined based on the back electromotive force signal. The resistance calculation module combines the system's equivalent rotational inertia, residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss to calculate the mechanical friction resistance torque of the current batch of sand unloading stage; the data processing module then uses this mechanical friction resistance torque as the reference for the no-load resistance torque, subtracts the real-time total resistance torque of the subsequent loaded mixing stage, obtains the actual molding sand resistance torque, and outputs the molding sand viscosity parameters accordingly. After the power failure detection period ends, the drive recovery module resumes the pulse width modulation pulse output, so that the main drive motor returns to the set speed. In this way, without the need to install additional internal sensors in the sand mixer, mechanical wear resistance and molding sand viscosity resistance can be separated, reducing baseline drift caused by scraper wear and changes in bottom sand accumulation. Furthermore, based on the above implementation method, the status signal of the sand unloading gate and the sand unloading duration signal are obtained, and the current production cycle is triggered and judged according to the preset sand unloading conditions to generate a short-term power-off command. Among them, the sand unloading door status signal refers to the opening and closing status quantity read by the main control module through the equipment side input interface, such as the sand unloading door opening level signal, and the sand unloading duration signal refers to the time criterion obtained by the main control module after periodically accumulating the sand unloading door opening status, such as the cumulative value of continuous opening to a preset proportion threshold. In practice, the main control module can read the status of the sand unloading door through the status acquisition channel of the program control system and write the status to the local instruction buffer within the control cycle. When the continuously recorded open states in the buffer meet the preset sand unloading conditions, the main control module generates a short-time power-off command in the same control link and writes the command into the control register corresponding to the frequency converter, which is then sent to the frequency converter drive module by the high-speed fieldbus. This implementation does not perform global fixed parameter correction on the entire mixed signal, but instead selects a periodic interval where the molding sand has been basically emptied but the mechanical resistance is still observable by triggering the conditions during the sand unloading stage, thus avoiding interference from the load viscosity change signal on the mechanical resistance extraction process. This embodiment can accurately select the power failure detection period based on the sand unloading condition triggering mechanism, avoid erroneous extraction of resistance characteristics under non-no-load approximate conditions, and ultimately solve the technical defect of the existing technology that the no-load resistance torque benchmark is difficult to update in real time due to the lack of a targeted triggering mechanism. Furthermore, based on the above implementation method, the back electromotive force signal of the main drive motor is obtained, and the pulse width modulation pulse output of the frequency converter is blocked according to the short-time power-off command, so as to generate the starting rotor angular velocity and the ending rotor angular velocity corresponding to the power-off detection period. Among them, the back electromotive force signal refers to the three-phase signal fed back by the stator winding when the main drive motor is freely gliding, such as the three-phase voltage change read by the frequency converter sampling channel after power failure. The starting rotor angular velocity and the ending rotor angular velocity refer to the rotor mechanical motion parameters obtained by back electromotive force inversion at the boundary of the power failure detection period, such as the angular velocity value at the beginning boundary of power failure and the angular velocity value at the end boundary of power failure. In practice, after receiving a short-time power-off command written to the control register by the main control module via the fieldbus, the frequency converter drive module blocks the pulse width modulation pulse output, causing the main drive motor to enter a free-slip state. The back electromotive force signal is collected using the internal sampling circuit of the frequency converter, and the sampling result is parsed into angular velocity data corresponding to the boundary of the power failure detection period in the local processing unit; The local computing unit uses the following rules for parsing: the local computing unit specifically uses back EMF zero-crossing detection combined with software phase-locked loop logic to obtain the basic phase discrete sequence of the rotor by detecting the voltage zero-crossing point of the three back EMF signals. The basic phase discrete sequence is input into a locally preset software phase-locked loop structure, and the internal proportional-integral (PI) regulator is used to filter out high-frequency commutation interference. The discrete iterative relationship of the PI regulator is as follows: in, For the input phase deviation, and These are the proportional and integral gains, respectively, derived from the rotor electrical angular frequency estimate. After integration, a smooth and continuous electrical angle sequence is output; the time derivative of the electrical angle sequence is calculated to generate the corresponding rotor angular velocity data, and the angular velocity data is written into the bus mapping register for the resistance calculation module to read; This implementation utilizes the back EMF sampling capability of existing frequency converters to extract deceleration characteristics directly corresponding to mechanical resistance during the short-term free-slip phase of the main drive motor, transforming the mechanical information that was originally easily obscured by the electromagnetic torque under load into a calculable change in angular velocity. Furthermore, based on the above implementation method, the rotor angular velocity at the start and end of the power outage detection period is obtained. Based on the system's equivalent moment of inertia, residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss, dynamic calculations are performed on the current batch of sand unloading stage to generate mechanical friction resistance torque. Among them, the system equivalent moment of inertia refers to the comprehensive inertial parameters exhibited by the main drive motor and the transmission components of the sand mixer during rotation, such as the moment of inertia value solidified after identification in the no-load state during the equipment commissioning stage, which is used to reflect the resistance torque conversion scale corresponding to the change in angular velocity. The residual sand resistance correction parameter refers to the correction amount used to reflect the additional damping effect of the molding sand not being completely emptied during the sand unloading stage. For example, the correction coefficient gradually approaches the no-load state as the sand unloading continues. It is used to reflect the actual working condition that the balance has been basically emptied but is not absolutely no-load. The combined resistance torque of wind resistance and iron loss refers to the additional resistance formed by wind resistance and iron loss of the main drive motor in this speed range. The combined resistance torque of wind resistance and iron loss is a dynamic corresponding value determined by the frequency conversion drive module after calculating the average angular velocity based on the rotor angular velocity at the start and end of the power failure detection period, and then looking up the preset motor loss mapping table with the average angular velocity as an index. It is used to reflect the motor body loss at the corresponding speed from the total deceleration effect. In order to obtain the mechanical frictional resistance torque during the current batch of sand unloading stage It can comprehensively correct parameters for residual sand resistance. Equivalent moment of inertia of the system Rotor angular velocity at the start of the power outage detection period Rotor angular velocity at the end of the power outage detection period Duration of power outage detection period and the combined drag torque of wind resistance and iron loss Using a relational formula, its calculation formula is as follows: The mechanical resistance without electromagnetic driving force can be calculated by using the natural deceleration process during the power outage period through the above relationship; This implementation method uses the coordinated constraints of the system's equivalent moment of inertia, residual sand resistance correction parameters, and motor loss terms to quantitatively evaluate the deceleration process, avoiding the misclassification of bottom residual sand damping and motor body losses as scraper mechanical wear. Furthermore, based on the above implementation method, the stator current and voltage during the load mixing stage are obtained, and the real-time total resistance torque is deducted from the benchmark according to the electromagnetic torque conversion result to generate the actual molding sand resistance torque, and the molding sand viscosity parameters are determined. Among them, the real-time total resistance torque refers to the total resistance torque information output by the main drive motor during the load mixing stage and calculated by the frequency conversion drive module. The frequency conversion drive module collects the stator current and voltage of the main drive motor during the load mixing stage, obtains the quadrature axis current and direct axis current through coordinate transformation, calculates the real-time electromagnetic torque by combining the number of motor pole pairs and flux linkage parameters, and obtains the real-time total resistance torque after deducting the dynamic torque compensation caused by the system inertia; for example, the resistance torque fluctuation during the molding sand mixing process after adding water is used to reflect the changes in molding sand viscosity and mechanical wear resistance at the same time. The unloaded resistance torque benchmark refers to the aforementioned mechanical friction resistance torque, such as the benchmark value measured at the end of the current batch of sand unloading and used for the next loading stage, which is used to reflect the foundation resistance not caused by the rheological properties of molding sand. The actual molding sand resistance moment refers to the effective resistance component obtained by subtracting the no-load resistance moment reference from the real-time total resistance moment. For example, it may only reflect the resistance moment portion that reflects changes in molding sand due to water addition, mixing, and densification. This component serves as the direct input for calculating viscosity parameters in a digital twin model. The actual molding sand resistance moment is used to extract this component. It can measure the real-time total drag torque during the loading and mixing phase. Subtract mechanical friction torque Using a relational formula, its calculation formula is as follows: The above relationship is used to calculate the resistance component directly related to the rheological behavior of molding sand from the total resistance torque. In industrial control and digital twin analysis, when evaluating the quality of extraction of viscosity-related features of molding sand, baseline drift sensitivity and batch-to-batch consistency are usually the focus. When a fixed baseline is used for deduction, as scraper wear intensifies or bottom sand accumulation increases, the non-molding sand component in the total resistance torque will gradually increase, which can easily cause an overall shift in viscosity parameters in continuous batches. When updating the mechanical friction resistance torque by batch using this implementation method, the data processing module subtracts the measured reference of the adjacent cycle in each load mixing cycle, which can separate the molding sand resistance characteristics from the mechanical wear characteristics in terms of source, thereby improving the comparability of viscosity parameters between different batches. Furthermore, based on the above implementation method, the back electromotive force signal analysis result is obtained, and the pulse width modulation pulse output is restored according to the power failure detection period end condition to generate the control result of the main drive motor restoring to the set speed; Among them, the rotor speed and electrical angle refer to the rotor motion state parameters identified by the drive recovery module based on the back electromotive force signal, that is, the speed and phase quantities used to match the current rotation state before the drive is restored. These speed and phase quantities are obtained by the aforementioned extraction based on zero-crossing detection and software phase-locked loop logic, and are used to reflect the adaptation of the pulse width modulation pulse to the current rotation state of the motor during recovery. The set speed refers to the target speed that the sand mixer needs to maintain in the subsequent production cycle, such as the operating speed required before entering the next batch of feeding waiting or before mixing under load, which is used to reflect the maintenance of production cycle continuity; In practical implementation, the drive recovery module reads the back electromotive force analysis result at the end of the power failure detection period from the bus mapping register, determines the rotor speed and electrical angle in the local control unit, and updates the inverter output control parameters accordingly, releases the pulse width modulation pulse blocking state, and restores the main drive motor to the set speed. This implementation uses the rotor state variables that are already available when the power outage ends for recovery control, avoiding the introduction of additional waiting due to re-searching the motor state; This embodiment can realize drive recovery based on the rotor state determined by the back electromotive force, avoiding the significant impact of power failure detection on the subsequent process cycle, and ultimately solving the technical defect of the prior art that is difficult to use for continuous production due to the need for additional machine stoppage for measurement actions.

[0017] In a preferred embodiment of the present invention, when the sand unloading door is in the open state and the sand unloading duration reaches the preset sand unloading condition, the main control module determines that it enters the power failure detection period. The duration of the power outage detection period is set according to the preset allowable speed drop range and historical operating parameters. During the power outage detection period, the frequency converter drive module determines the first rotor angular velocity at the start of the power outage and the second rotor angular velocity at the end of the power outage, and transmits them to the resistance calculation module.

[0018] The system obtains the sand unloading gate opening status and sand unloading duration, and determines the timing of entering the power outage detection period based on preset sand unloading conditions, thereby determining the power outage detection period and its duration. Among them, the preset sand unloading conditions refer to the criteria for determining whether the sand unloading gate is kept open and the sand unloading duration can be approximated as the end of the emptying period. For example, the sand unloading duration reaches a preset proportion of the standard total sand unloading time. The purpose is to make the power failure detection occur in the periodic interval where the molding sand has been basically discharged and the mechanical resistance can still be stably characterized. The allowable speed drop range refers to the acceptable speed drop limit of the main drive motor during a short power outage. For example, the allowable drop ratio corresponding to the rated angular velocity can be used as the judgment boundary to reflect the balance between resistance extraction accuracy and subsequent drive recovery stability. Historical operating parameters refer to the information on angular acceleration changes recorded by the equipment during its historical period, such as the absolute value of the historical maximum angular acceleration, which is used to provide a conservative constraint on the length of the power outage detection period. In practice, the main control module reads the input status of the sand unloading door in each control cycle and accumulates the sand unloading duration during the door opening period. When the current cycle of sand unloading duration Greater than or equal to the effective air displacement coefficient Total sand unloading time preset by the system The product of, that is, satisfies When the power outage detection period begins, the sand unloading duration is determined to be within the specified time. Less than the effective air displacement coefficient Total sand unloading time preset by the system When the product is multiplied, the main control module maintains normal monitoring status and does not output a short-term power-off command; Effective air displacement coefficient The value can be within a preset ratio range, which is used to define the boundary of the working condition that has been basically emptied. Furthermore, the duration of the power outage detection period. It can be preset based on the allowable speed reduction range and historical operating parameters to ensure that it does not exceed the rated angular velocity. Ratio of maximum permissible speed drop The product divided by the absolute value of the historical maximum angular acceleration The obtained quotient satisfies The purpose of this relationship calculation is to ensure that an observable angular velocity difference can be generated during the power outage detection period, without causing the main drive motor speed to drop too much and affect the recovery drive. In a specific implementation, the standard total unloading time of the sand mixer is set. The effective emptying coefficient is 20 seconds. Set to 0.85, when the sand unloading duration of the current cycle... When 17 seconds have elapsed, the main control module determines that the preset sand unloading conditions have been met and enters the power-off detection period. At this point, we further assume the rated angular velocity of the main drive motor. The maximum permissible speed drop ratio is 150 rad / s. Limited to 5%, and the absolute value of the historical maximum angular acceleration recorded by the system. for Based on the aforementioned rules, (150 × 5%) / 15 = 0.5 seconds is calculated, which is the set duration of the power outage detection period. The deceleration time is no more than 0.5s, which allows for effective deceleration characteristics while limiting excessive drops in engine speed. This implementation method limits the power outage window based on the sand unloading duration and historical operating capacity, so that the triggering conditions change in tandem with the actual operating conditions of the equipment. This transforms the simple process of powering off too early or too late into targeted control for near-no-load conditions. This embodiment can determine the power outage detection period based on the joint judgment of sand unloading status and historical operating parameters, avoiding the window selection deviating from the actual emptying stage, and ultimately solving the technical defect of existing technology that causes angular velocity sampling distortion due to lack of working condition constraints. The back electromotive force signal during the power outage detection period is acquired. Based on the boundary of the power outage detection period, the rotor motion state is analyzed to determine the first rotor angular velocity and the second rotor angular velocity, and written into the read register corresponding to the resistance calculation module. The first rotor angular velocity refers to the rotor angular velocity at the power-off start boundary, such as the angular velocity value obtained by inversion from the effective back electromotive force after the pulse width modulation pulse is blocked and the stator freewheeling decays to zero. It is used to reflect the initial motion state characterizing the mechanical resistance calculation. The second rotor angular velocity refers to the rotor angular velocity at the power-off end boundary, such as the angular velocity value obtained by analysis at the end of the power-off detection period, which is used to reflect the end motion state after characterizing the free gliding process. During the power failure detection period, the frequency converter drive module continuously samples the three-phase back electromotive force signals of the stator winding and updates the data in the local sampling buffer according to the control cycle. Based on the sampling results of the start and end boundaries of the time period, the frequency conversion drive module analyzes and obtains the first rotor angular velocity and the second rotor angular velocity respectively, and writes the analyzed angular velocity values ​​into the bus mapping register. The resistance calculation module releases the corresponding buffer immediately after reading the data. This implementation method focuses on sampling the rotor angular velocity at the boundary of the power outage detection period, which can directly correspond to the deceleration process within the free gliding range, thus providing a clear time basis for subsequent inertia conversion and drag inversion. In industrial control drives, to determine whether the resistance inversion is reliable, we usually focus on whether the boundary sampling is aligned with the actual free gliding range. If the boundary sampling deviates, the angular velocity difference will be mixed with the acceleration and deceleration effects of the electric drive stage, resulting in the mechanical resistance estimation being too high or too low. With this implementation method, both the first rotor angular velocity and the second rotor angular velocity are determined within the power-off detection period, and the length of this period is constrained by the allowable speed drop range and historical operating parameters, thus maintaining good boundary consistency. This embodiment can reliably extract the angular velocities at both ends based on boundary analysis during the power outage detection period, avoiding the inclusion of information from the non-free gliding phase in the angular velocity difference, and ultimately solving the technical defect of the prior art that causes the distortion of the basic data for resistance calculation due to unclear sampling boundaries.

[0019] In a preferred embodiment of the present invention, the resistance calculation module calculates the mechanical friction resistance torque based on the first rotor angular velocity, the second rotor angular velocity, the duration of the power outage detection period, the system's equivalent moment of inertia, the residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss. The residual sand resistance correction parameters are determined based on the sand unloading duration and historical sand unloading test data; At the end of the power outage detection period, the main control module stops outputting short-term power outage commands; the drive recovery module resumes pulse width modulation pulse output and controls the main drive motor to return to the set speed; the resistance calculation module sends the mechanical friction resistance torque to the data processing module as a new no-load resistance torque reference. The first rotor angular velocity, the second rotor angular velocity, and the duration of the power outage detection period are obtained. Based on the system's equivalent moment of inertia, residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss, the mechanical resistance during the sand unloading stage is converted to generate the mechanical friction resistance torque. Among them, the equivalent rotational inertia of the system refers to the inertia parameter jointly exhibited by the main drive motor and transmission system of the sand mixer. For example, the inertia value identified and stored in the controller under no-load conditions during the initial commissioning of the equipment is used to reflect the inertial scale corresponding to converting the change in angular velocity into resistance torque. The combined resistance torque of wind resistance and iron loss refers to the resistance term caused by wind resistance and iron loss of the main drive motor in the current speed range. For example, the loss value obtained by looking up the table based on the motor speed-loss relationship is used to reflect the interference of the inherent loss of the motor body on the mechanical friction judgment. The residual sand resistance correction parameter refers to the additional damping correction introduced when a small amount of molding sand remains at the end of the sand unloading process. It is used to reflect the difference in actual working conditions where the load is close to empty but there is still residual sand. The purpose of this formula is to convert the deceleration process during free sliding into the mechanical friction resistance torque during the sand unloading stage; This implementation method uses inertia, loss and residual sand correction terms in the calculation, so that the mechanical friction resistance torque is no longer mixed with motor body loss and residual molding sand damping, thus improving its usability as a reference quantity. Acquire the sand unloading duration and historical sand unloading test data, and correct the residual sand resistance based on the degree of venting during the sand unloading stage, and determine the residual sand resistance correction parameters; Among them, historical sand unloading test data refers to the resistance change record of the equipment under different sand unloading durations, such as the relationship between the angular velocity decay and the degree of venting at the end of multiple batches of sand unloading, which is used to reflect the empirical boundary for the additional resistance of residual sand. Additional resistance from residual sand refers to the additional resistance formed at the end of the sand unloading stage due to a small amount of residual molding sand still in contact with the scraper and bottom plate. For example, the additional frictional resistance caused by the residual sand layer before the sand unloading is completely finished. It is used to reflect that this stage is not an absolutely empty machine state. To make the mechanical friction resistance torque closer to the actual foundation friction, residual sand resistance correction parameters can be constructed based on the duration of sand unloading, and these parameters can be determined. : This relation uses the natural constant. Based on the bottom line, combined with the duration of sand unloading And empirical constants obtained by fitting historical sand unloading test data. and ; When establishing the aforementioned constants, the system collects real residual sand additional resistance data under different sand unloading durations during the offline calibration phase. Using the resistance attenuation rate as the target variable, the least squares method is used for nonlinear curve fitting to determine the basic coefficient representing the initial residual sand influence intensity. and the reciprocal of the decay time constant representing the venting rate Its value is fixed based on the opening of the sand discharge gate and the sand discharge capacity of the mixing scraper, so as to ensure the accuracy of parameter matching under different equipment operating conditions; The purpose of this formula is to make the influence of residual sand weaker as the sand unloading duration increases, and to make the correction parameters closer to the no-load conditions. In a specific implementation, an empirical constant is set based on fitting historical sand unloading test data. and When the sand unloading duration When the time is 10s, the system calculates the residual sand resistance correction parameters at this time. ; When the sand unloading duration When the summation is further accumulated to 15 seconds, the calculation is obtained. This indicates that as the degree of sand unloading and venting increases, the nonlinear effect of residual sand additional damping weakens, and the correction parameter gradually transitions smoothly to coefficient 1 under absolute no-load conditions, thus realizing a transparent conversion of the residual sand damping effect. This implementation method uses the combined constraints of sand unloading duration and historical sand unloading test data to quantitatively correct the additional resistance of residual sand, transforming small resistance terms that are easily ignored into correction parameters that can be used in the calculation. In signal processing and operating condition identification, if the additional resistance of residual sand is ignored, the mechanical friction resistance torque is easily overestimated when the sand unloading is not fully completed, and too much reference is deducted in the next loading stage, resulting in a lower molding sand resistance characteristic. After adopting this implementation method, the residual sand resistance correction parameter will be adjusted according to the degree of sand unloading, which can reduce the impact of different sand unloading fullness on the consistency of the no-load resistance torque benchmark. The signal indicating the end of the power failure detection period is obtained. Based on the drive recovery conditions, the pulse width modulation pulse output is recovered, and the mechanical friction resistance torque is written into the reference register unit corresponding to the data processing module to update the no-load resistance torque reference. Among them, the power failure detection period end signal refers to the power failure termination control condition determined by the main control module based on the preset duration, such as the stop signal after the power failure detection period count reaches the set length, which is used to reflect the switching boundary between the end of mechanical resistance sampling and the start of drive recovery. The unloaded resistance torque benchmark refers to the basic resistance value used for deducting the total resistance torque in the subsequent loaded mixing stage. For example, the mechanical friction resistance torque just calculated in the current sand unloading stage is used to reflect the bottom value of mechanical resistance that is independent of the viscosity of molding sand. In practical implementation, the main control module stops outputting short-term power failure commands when the power failure detection period ends. At this time, the system enters two parallel control links: In the first link, the drive recovery module immediately releases the pulse blockade inside the frequency converter based on the back electromotive force analysis result, restores the pulse width modulation pulse output, and controls the main drive motor to restore to the set speed. In the second link, the resistance calculation module needs to go through a microsecond-level bus read delay to obtain the second rotor angular velocity, and execute the aforementioned dynamic calculation formula containing floating-point operations in the local control unit. After the calculation link is completed, the converted mechanical friction resistance torque is asynchronously written into the reference register unit of the data processing module, thereby eliminating the timing paradox of absolute synchronous response in physical execution. The data processing module will release the old value immediately after reading the baseline value in the next load mixing cycle; This implementation method directly transfers the mechanical friction resistance torque to the data processing module asynchronously during the drive recovery phase, so that the measurement and subsequent use are continuously connected within adjacent production cycles. This embodiment can realize the timely use of mechanical resistance measurement results based on the independent parallel recovery and reference update timing logic after the power outage ends. It avoids weakening the correction effect of subsequent load data due to the lag in reference update or command blocking, and finally solves the technical defect of the prior art that the compensation is not timely because the measurement results cannot be entered into the calculation in a timely manner with the cycle.

[0020] In a preferred embodiment of the present invention, during the load mixing stage of the next adjacent batch, the frequency conversion drive module outputs the real-time total resistance torque to the data processing module, and the data processing module calculates the actual molding sand resistance torque based on the real-time total resistance torque and the updated mechanical friction resistance torque of the previous adjacent batch. During the load mixing cycle, the data processing module uses the mechanical friction resistance torque updated in the adjacent previous batch for benchmark subtraction.

[0021] Obtain the real-time total resistance torque of the next adjacent batch during the loading and mixing stage. Based on the updated mechanical friction resistance torque of the previous adjacent batch, subtract the real-time total resistance torque from the benchmark to generate the actual molding sand resistance torque. Among them, the real-time total resistance torque refers to the total resistance information calculated by the frequency conversion drive module based on the stator current and voltage of the main drive motor during the load mixing cycle. For example, the resistance torque increases when the molding sand becomes more adhesive after water is added. It is used to reflect both the mechanical foundation resistance and the rheological resistance of the molding sand. The mechanical friction resistance torque updated in the previous batch refers to the baseline value obtained through power failure detection in the previous sand unloading stage and written into the data processing module. It is used to reflect the latest no-load resistance reference for the current load cycle. The actual molding sand resistance torque refers to the effective resistance component retained after deducting the mechanical reference. For example, it only corresponds to the resistance torque part that corresponds to the mixing, humidity change and compaction of molding sand. It is used to reflect the true contribution of the current molding sand itself to the load of the sand mixer. In specific implementation, the frequency conversion drive module writes the real-time total resistance torque calculated during the load mixing stage into the bus mapping register according to the control cycle. After reading it, the data processing module subtracts the mechanical friction resistance torque of the previous batch in the reference register unit to obtain the actual molding sand resistance torque corresponding to the current cycle. The real-time data read is released immediately at the end of the control cycle. This implementation method always uses the mechanical friction resistance torque that was just updated in the previous batch for subtraction, so that the baseline change can be adjusted synchronously with the wear of the scraper and the change of the bottom sand accumulation, thereby reducing the interference of slowly changing mechanical factors on the extraction of molding sand features. This embodiment can calculate the actual molding sand resistance torque based on the reference transfer between adjacent batches, avoiding the long-term accumulation of mechanical wear components in the real-time total resistance torque and contamination of the load characteristics, and ultimately solving the technical defect of the prior art that the molding sand resistance is not accurately identified due to the excessively long reference update cycle. Obtain the actual molding sand resistance torque, and calculate the viscosity parameters of the current molding sand based on the resistance characteristics required for digital twin analysis, generating molding sand viscosity parameters that can be used for production monitoring; Among them, molding sand viscosity parameter refers to the result parameter used to characterize the current rheological properties of molding sand, such as the value reflecting the degree of adhesion and flow change of molding sand after mixing, which is used to provide a judgment basis directly related to molding sand quality for production monitoring; The loaded mixing cycle refers to the operating range of the molding sand under the action of feeding, watering and continuous mixing load, such as the entire cycle of the main drive motor continuously outputting the mixing load. It is used to reflect that the total resistance torque is most significantly affected by the characteristics of the molding sand in this range. The data processing module continuously uses the mechanical friction resistance torque updated by the adjacent previous batch to perform benchmark subtraction within the loaded mixing cycle to obtain the actual molding sand resistance torque. The specific calculation process for digital twin analysis performed by the data processing module includes: The data processing module extracts the actual molding sand resistance torque within a specific feature window during the loaded mixing stage. The specific feature window is set as the steady-state operating time period after the water addition step ends and the rate of change of the actual molding sand resistance torque is lower than a preset fluctuation threshold for multiple consecutive sampling cycles during the loaded mixing stage. The module also calculates the time-domain characteristics within this specific feature window, such as the average resistance torque and the fluctuation variance. The time-domain features are used as query keys and input into a multidimensional mapping matrix of resistance features and viscosity parameters that has been pre-established based on a digital twin virtual environment and a large number of offline calibration experiments. To clarify the construction process and internal mapping logic of the digital twin virtual environment, its construction includes the following defined steps and physical quantity boundaries: A virtual entity of the equipment is established based on the three-dimensional geometric feature parameters of the sand mixer and the electromechanical coupling dynamic equations of the main drive motor. The electromechanical coupling dynamic equations are set as follows: in, For electromagnetic torque, This represents the total load resistance torque during the mixed-load phase. The system's viscous damping coefficient is defined as follows: A simulation model of sand particle collision and adhesion contact based on the discrete element method is embedded within this model. The simulation model uses the Hertz-Minderlin JKR contact mechanics calculation framework, and its normal adhesion mechanics equation is set as follows: in, For the surface energy parameters of molding sand, For the equivalent elastic modulus, The radius of the particle contact surface is used to quantitatively characterize the liquid bridge adhesion effect between molding sand particles caused by water mixing. The simulation was driven by using known preset molding sand viscosity parameters with multiple gradients as environmental input variables, and the average value and fluctuation variance of the virtual resistance torque at the corresponding virtual main drive motor were extracted. The virtual calculation results are cross-validated and error-calibrated with actual offline calibration data obtained from material testing of a large number of physical devices. The loss function for error calibration is set as follows: in, The variance weighting coefficients are used to adjust the surface energy parameters of molding sand in the discrete element simulation model using the gradient descent algorithm. The coefficient of sliding friction between particles is calculated until the loss function is lower than the preset convergence threshold. The convergence threshold is calibrated based on the allowable root mean square error of multiple offline tests, which characterizes the limit of the calculation deviation between the model output and the entity data. The relationship between the input viscosity parameter and the output resistance feature at this time is extracted and finally solidified to generate the above multidimensional mapping matrix. The multidimensional mapping matrix uses the average resistance torque during the loaded mixing stage as the first dimension coordinate and the resistance torque fluctuation variance as the second dimension coordinate. The matrix nodes store the molding sand viscosity benchmark value obtained from offline test calibration. Data flow and analysis are performed through spatial interpolation algorithms. For example, using bilinear interpolation, based on the average value and variance of the current input drag torque, four adjacent reference nodes in the multidimensional mapping matrix are located. The smooth viscosity value within the interval enclosed by the four nodes is calculated by the inverse weight of the distance between the two points, and the viscosity parameter of the current molding sand is directly obtained and output. In a specific implementation, the aforementioned multidimensional mapping matrix is ​​set with a mean step size. variance step size Mesh generation is performed, and the average actual molding sand resistance torque extracted from the system within the steady-state window is... And the variance of the fluctuation is At that time, the data processing module locates four adjacent reference grid nodes containing the coordinates in the mapping matrix based on the time domain characteristics: node A (190,10) corresponds to a viscosity coefficient of 1.5, node B (210,10) corresponds to a viscosity coefficient of 1.7, node C (190,15) corresponds to a viscosity coefficient of 1.6, and node D (210,15) corresponds to a viscosity coefficient of 1.85; Dimensionality reduction was performed using a bilinear interpolation algorithm: the first layer of interpolation was performed based on the inverse distance ratio in the mean dimension, resulting in an intermediate viscosity value of 1.6 at the mean of 200 and variance of 10, and an intermediate viscosity value of 1.725 at the mean of 200 and variance of 15. Based on the actual variance value of 12, a second-level interpolation is performed on the variance dimension to derive the final current molding sand viscosity parameter. ; The above multidimensional interpolation operation can reduce the computational bias of a single dimension and reduce the coupling interference of mechanical wear changes on the model input. In digital twin modeling, the evaluation of whether the input features are suitable as the basis for predicting molding sand viscosity usually focuses on the degree of decoupling between the features and mechanical wear changes. If the total resistance torque is used directly, as the wear of the scraper increases, the model input will show an overall rise that is unrelated to the molding sand, resulting in inconsistent predictions between batches. After adopting this implementation method, the molding sand viscosity parameter is established on the basis of the actual molding sand resistance torque after deducting the mechanical friction resistance torque of the previous batch, which can make the input characteristics of the same formula maintain good stability under different wear stages. This embodiment can calculate the molding sand viscosity parameters based on the updated mechanical friction resistance torque of the adjacent previous batch, avoiding the deviation of the digital twin model due to the long-term influence of the slow change of mechanical wear factors, and ultimately solving the technical defect of insufficient stability of molding sand quality prediction caused by the lack of a batch-based deduction mechanism in the existing technology.

[0022] The above system is deployed in the field structure consisting of the program control system, frequency converter and host computer server of the sand mixer. The main control module is deployed in the program control system, the frequency conversion drive module, the resistance calculation module and the drive recovery module are deployed in the frequency converter, and the data processing module is deployed as a digital twin analysis module in the host computer server. The main control module reads the sand unloading gate opening and closing signal through the equipment side status acquisition channel and accumulates the sand unloading duration within the control cycle; When the preset sand unloading conditions are met, the main control module writes a short-time power-off command to the control register corresponding to the frequency converter drive module via the high-speed fieldbus. During the power failure detection period, the frequency conversion drive module blocks the pulse width modulation pulse output and uses its own sampling circuit to collect three-phase back electromotive force signals. The rotor angular velocity obtained at the power failure boundary is written into the bus mapping register for the resistance calculation module to read. The resistance calculation module calculates the mechanical friction resistance torque based on the system's equivalent moment of inertia, residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss. After the power outage detection period ends, the mechanical friction resistance torque is written into the reference register unit of the data processing module. The drive recovery module synchronously restores the pulse width modulation pulse output based on the back electromotive force analysis results, so that the main drive motor returns to the set speed; After entering the next batch of loaded mixing stage, the frequency conversion drive module continues to output the real-time total resistance torque according to the control cycle. The data processing module reads the real-time total resistance torque and calls the updated mechanical friction resistance torque of the previous batch for benchmark subtraction to obtain the actual molding sand resistance torque, and further calculates the viscosity parameters of the current molding sand. This viscosity parameter is output as a production monitoring result to the digital twin analysis module, which is used to maintain the data input required for judging the quality of molding sand in subsequent cycles; This application example can complete the measurement of mechanical resistance during the sand unloading stage, the deduction of the load characteristics of the next batch, and the output of molding sand viscosity parameters within a continuous production cycle, avoiding the impact of additional downtime or manual calibration on production continuity.

[0023] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A real-time monitoring system for the operating status of a sand mixer based on digital twins, characterized in that, include: The main control module is used to monitor the status of the sand discharge gate of the sand mixer and the duration of sand discharge, and outputs a short-term power-off command when the preset sand discharge conditions are met; The variable frequency drive module includes a frequency converter and a main drive motor driven by the frequency converter. It is used to block the pulse width modulation pulse output of the frequency converter after receiving the short-time power failure command, so that the main drive motor enters a free gliding state, collects the back electromotive force signal, and determines the first rotor angular velocity at the start time and the second rotor angular velocity at the end time of the power failure detection period. It is also used to collect the stator current and voltage of the main drive motor during the load mixing phase and output the real-time total resistance torque; The resistance calculation module is used to calculate the mechanical friction resistance torque of the current batch of sand unloading stage based on the rotor angular velocity at the start and end of the power outage detection period, the duration of the power outage detection period, the pre-calibrated equivalent rotational inertia of the system, the pre-determined residual sand resistance correction parameters, and the wind resistance and iron loss combined resistance torque obtained by consulting the pre-set motor loss mapping table. The data processing module is used to take the mechanical friction resistance torque as the no-load resistance torque benchmark, subtract the benchmark from the real-time total resistance torque to obtain the actual molding sand resistance torque, and calculate and output the molding sand viscosity parameters based on the actual molding sand resistance torque. The drive recovery module is used to determine the rotor speed and electrical angle based on the back electromotive force signal after the power failure detection period ends, restore the pulse width modulation pulse output, and control the main drive motor to return to the set speed.

2. The real-time monitoring system for the operating status of a sand mixer based on digital twins as described in claim 1, characterized in that, When the sand unloading door is open and the sand unloading duration reaches the preset sand unloading condition, the main control module determines that the power failure detection period has begun.

3. The real-time monitoring system for the operating status of a sand mixer based on digital twins according to claim 2, characterized in that, The duration of the power outage detection period is set according to the preset allowable speed drop range and historical operating parameters. The variable frequency drive module determines the first rotor angular velocity at the start of the power outage and the second rotor angular velocity at the end of the power outage during the power outage detection period, and transmits them to the resistance calculation module.

4. The real-time monitoring system for the operating status of a sand mixer based on digital twins according to claim 3, characterized in that, The resistance calculation module calculates the mechanical friction resistance torque based on the first rotor angular velocity, the second rotor angular velocity, the duration of the power outage detection period, the system's equivalent moment of inertia, the residual sand resistance correction parameters, and the combined resistance torque of wind resistance and iron loss.

5. The real-time monitoring system for the operating status of a sand mixer based on digital twins according to claim 4, characterized in that, The residual sand resistance correction parameter is determined based on the sand unloading duration and historical sand unloading test data.

6. The real-time monitoring system for the operating status of a sand mixer based on digital twins according to claim 5, characterized in that, When the power outage detection period ends, the main control module stops outputting the short-term power outage command; The drive recovery module restores the pulse width modulation pulse output and controls the main drive motor to return to the set speed; The resistance calculation module sends the mechanical friction resistance torque to the data processing module as a new reference for the no-load resistance torque.

7. The real-time monitoring system for the operating status of a sand mixer based on digital twins according to claim 6, characterized in that, During the loading and mixing phase of the next adjacent batch, the variable frequency drive module outputs the real-time total resistance torque to the data processing module. The data processing module calculates the actual molding sand resistance torque based on the real-time total resistance torque and the mechanical friction resistance torque updated in the previous adjacent batch.

8. The real-time monitoring system for the operating status of a sand mixer based on digital twins according to claim 7, characterized in that, During the load mixing cycle, the data processing module uses the mechanical friction resistance torque updated in the adjacent previous batch for benchmark subtraction.

Images