A high-precision shock-resistant loss-in-weight scale and its control method

By setting up independent weighing sensors and calibration components in the loss-in-weight scale, and combining multiple algorithms and coordinated actions of mechanisms, the metering accuracy and stability problems of existing loss-in-weight scales during the feeding impact stage are solved, and high-precision closed-loop control under all working conditions is achieved.

CN123111428APending Publication Date: 2026-11-17SHENZHEN JIANHANG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611300894.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-26
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing loss-in-weight scales suffer from low open-loop control accuracy, poor structural linkage, and significant vibration interference during the feeding impact phase, failing to meet the demands of high-end precision batching.

Method used

A high-precision, shock-resistant loss-in-weight scale is adopted. By setting an independent weighing sensor and calibration component in the feeding hopper, combined with weight acquisition and calibration, Kalman filtering, hysteresis compensation and AI dynamic optimization algorithm, closed-loop control under all working conditions is achieved. The double helix mechanism and stirring mechanism work together to eliminate material impact and environmental vibration interference.

Benefits of technology

The metering error of the feeding impact section is controlled within ±0.1% to ±0.2%, ensuring stable discharge and meeting the needs of high-end precision batching.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123111428A_ABST
    Figure CN123111428A_ABST
Patent Text Reader

Abstract

This invention discloses a high-precision, impact-resistant loss-in-weight scale and its control method, relating to the field of loss-in-weight scale measurement and control technology. The loss-in-weight scale includes a main body assembly, a calibration assembly, and a control assembly. The main body assembly includes a feeding hopper, a scale body, a weighing sensor, and a servo motor. The feeding hopper and the scale body are sealed together via a flexible connector. A first stirring mechanism is located on the upper part of the scale body, and a second stirring mechanism and a double-helix mechanism are located below the scale body. The input end of the double-helix mechanism is connected to a transfer box. The calibration assembly includes a data acquisition module with a built-in edge computing unit. The control assembly controls the coordinated operation of the double-helix mechanism, the first stirring mechanism, and the second stirring mechanism based on the compensated flow signal output by the calibration assembly. This invention achieves full closed-loop control of the feeding impact section through high-precision weighing compensation in the upper hopper, three-servo coordinated flexible control, and AI adaptive adjustment, effectively improving the measurement accuracy and operational stability of the loss-in-weight scale.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of loss-in-weight scale measurement and control technology, specifically to a high-precision impact-resistant loss-in-weight scale and its control method. Background Technology

[0002] Loss-in-weight weighing (LIW) systems, as high-precision metering devices with intermittent feeding and continuous discharging, are widely used in quantitative batching processes for various powders and granules. However, existing LWI systems suffer from a core technical challenge during actual operation: during the upper feeding stage, the falling material impacts the lower LWI system, causing signal fluctuations and distortion in the lower weighing sensors. This prevents effective closed-loop control, forcing the system to switch to open-loop volumetric operation. This open-loop control significantly reduces metering accuracy during the feeding impact phase, typically with errors ranging from ±0.3% to 1.5%, severely impacting overall batching accuracy.

[0003] In existing technologies, some solutions attempt to improve accuracy by optimizing PID parameters and adding filtering algorithms, but they do not fundamentally solve the signal distortion problem caused by feeding impact. Other solutions employ dual-weighing-body linkage, but these often use ordinary 4-wire sensors and low-cost AD acquisition modules, resulting in drawbacks such as wire resistance loss, large temperature drift, and weak anti-interference capabilities. Furthermore, they lack multi-servo collaborative control, failing to balance discharge stability and metering accuracy. Meanwhile, existing loss-in-weight scales, if employing a double-helix, split-box, and double-stirring structure, are prone to problems such as double-helix discharge pulsation, sudden flow changes caused by split-box return gaps, and stirring vibrations interfering with the weighing signal, further exacerbating metering errors. Moreover, servo control is mostly synchronous control, failing to achieve coordinated flexible control, making it unsuitable for the complex operating conditions during feeding impact periods and unable to meet the demands of high-end precision batching. Summary of the Invention

[0004] This application provides a high-precision anti-impact loss-in-weight scale and its control method, which can solve the technical problems of low open-loop control accuracy, poor structural linkage, and large vibration interference in the feeding impact section of the loss-in-weight scale in the prior art.

[0005] In a first aspect, embodiments of this application provide a high-precision shock-resistant loss-in-weight scale, comprising: The main body assembly includes a feeding bin, a scale body flexibly connected below the feeding bin, and a weighing sensor installed at the bottom of the feeding bin. The upper part of the scale body is provided with a first stirring mechanism connected to the feeding bin, and the lower part of the scale body is provided with a second stirring mechanism and a double helix mechanism located at the discharge end of the second stirring mechanism. The input end of the double helix mechanism is connected to a transfer box. A calibration component, which is associated with the scale body, includes a data acquisition module. The signal input terminal of the data acquisition module is electrically connected to the signal output terminal of the weighing sensor to acquire the upper weight signal of the feeding hopper. The data acquisition module has a built-in edge computing unit, which includes a weight acquisition calibration unit for calibrating zero drift and system errors, a flow calculation unit for calculating instantaneous feeding flow rate, a filtering unit for eliminating signal interference, a hysteresis compensation unit for correcting material falling hysteresis error, and an AI optimization unit for dynamically optimizing compensation parameters. A control component, which is communicatively connected to the calibration component, is used to receive the compensated flow signal output by the calibration component and control the double helix mechanism, the first stirring mechanism and the second stirring mechanism to work together according to the compensated flow signal.

[0006] In one embodiment, the acquisition module further includes a high-precision analog-to-digital converter (ADC), the weighing sensor signal output terminal of the feeding hopper is connected to the high-precision ADC, and the high-precision ADC is connected to the edge computing unit via an SPI interface.

[0007] In one embodiment, the weighing sensor is a six-wire sensor, with its positive power supply terminal connected to the positive power supply terminal of the high-precision analog-to-digital converter, its negative power supply terminal connected to the negative power supply terminal of the high-precision analog-to-digital converter, its positive signal terminal connected to the first signal input terminal of the high-precision analog-to-digital converter, its negative signal terminal connected to the second signal input terminal of the high-precision analog-to-digital converter, its positive sensing terminal short-circuited to the positive power supply terminal, and its negative sensing terminal short-circuited to the negative power supply terminal.

[0008] In one embodiment, the weight acquisition and calibration unit is configured with a calibration formula: ; in, For the real-time weight of the feeding hopper (1), These are the raw values ​​obtained by the acquisition module. This is the collected value corresponding to the sensor's zero point. The weighing range of the load cell. This is the collected value corresponding to the full scale of the sensor.

[0009] In one embodiment, the flow calculation unit is configured with an instantaneous flow calculation formula: ; in, For the instantaneous discharge flow rate of the feeding hopper, This represents the weight of the feeding hopper in the previous data collection cycle. This represents the weight of the feeding hopper during the current data collection cycle. The sampling period is specified; the filtering unit uses the Kalman filter algorithm to filter the instantaneous flow rate.

[0010] In one embodiment, the hysteresis compensation unit is configured with a hysteresis compensation formula: ; in, This is the amount of delayed compensation. For the instantaneous discharge flow rate of the feeding hopper, The lag time for materials to fall from the feeding hopper to the scale body. This refers to the data collection cycle.

[0011] In one implementation, the AI ​​optimization unit dynamically optimizes the compensation coefficient and adaptively corrects the lag time by real-time collection of instantaneous flow rate, double helix mechanism speed, first stirring mechanism speed, second stirring mechanism speed, and scale body weight fluctuation value, combined with historical operating data, and outputs the optimized true net flow rate. The calculation formula is as follows: ; in, For true net flow, For the instantaneous discharge flow rate of the feeding hopper, The dynamic compensation coefficient optimized for AI. This is the lag compensation amount after AI correction.

[0012] This application also includes a control system. The edge computing unit transmits the calculated real net flow rate Q_real to the control system. The control component switches the control mode according to the current working conditions. When the feeding signal is triggered, it switches to the impact compensation mode and uses Q_real as the main feedback. When the feeding ends, it switches to the stable mode and uses the weighing signal of the scale body 2 as the feedback. The control component outputs control signals to each servo motor to drive the double helix mechanism 5, the first stirring mechanism 3 and the second stirring mechanism 4 to perform coordinated actions. The weighing sensor of the scale body 2 feeds back the weight signal to the control system, forming a closed-loop control.

[0013] Secondly, embodiments of this application provide a control method for a high-precision shock-resistant loss-of-weight scale, employing the aforementioned high-precision shock-resistant loss-of-weight scale, comprising the following steps: The acquisition module obtains the weight signal of the upper chamber of the feeding hopper, and the edge computing unit performs zero-point drift and system error calibration on the upper chamber weight signal; The edge computing unit calculates the instantaneous feed rate based on the calibrated weight signal and eliminates signal interference through a filtering algorithm; The edge computing unit uses a hysteresis compensation algorithm to correct the hysteresis error of material falling from the feeding hopper to the scale body, and dynamically optimizes the compensation parameters through the AI ​​optimization unit, and outputs the compensated real net flow signal to the control component. When the feeding signal is triggered, the control component switches to the impact compensation mode, and uses the compensated real net flow signal as the main feedback to control the double helix mechanism, the first stirring mechanism and the second stirring mechanism to adjust in coordination. In the impact compensation mode, the double helix mechanism is driven by a servo motor through the transfer box, the two screws maintain a set speed difference, the first stirring mechanism maintains a constant low speed, and the second stirring mechanism speeds up to break the arch. When feeding is finished, the control component switches back to stable mode, using the weight signal of the scale body as feedback to achieve closed-loop control.

[0014] In one embodiment, the formula for zero-point drift and system error calibration is: ; in, This represents the real-time weight of the feeding hopper. These are the raw values ​​acquired in real time by the data acquisition module. This is the collected value corresponding to the zero point of the weighing sensor. The range of the weighing sensor. This represents the collected value corresponding to the full scale of the weighing sensor; The formula for calculating the instantaneous feed rate is: ; in, The instantaneous discharge flow rate of the feeding hopper, This represents the weight of the feeding hopper in the previous data collection cycle. This represents the weight of the feeding hopper during the current data collection cycle. For the collection period, The filtering time is 0.01s to 0.05s. The filtering algorithm is the Kalman filtering algorithm, and its filtering parameters are adaptively adjusted by the AI ​​optimization unit according to the real-time signal interference intensity.

[0015] In one implementation, the AI ​​optimization unit collects real-time data on instantaneous flow rate, double helix mechanism speed, first stirring mechanism speed, second stirring mechanism speed, and scale body weight fluctuation, and combines this data with historical operating data to perform the following optimization operations: The compensation coefficient K is dynamically optimized to replace the fixed initial value. Adaptive adjustment of filter parameters; The lag time τ is adaptively adjusted based on the material's falling speed. Based on historical operating data, predict the intensity and duration of the feeding impact, and adjust the compensation and control parameters in advance before the feeding signal is triggered; The final formula for calculating the true net flow is: ; in, This represents the actual net flow rate of the feeding hopper. For the instantaneous discharge flow rate of the feeding hopper, The dynamic compensation coefficients after optimization for the AI ​​optimization unit. The amount of hysteresis compensation corrected for the AI ​​optimization unit.

[0016] The beneficial effects of the technical solutions provided in this application include: 1. By setting up independent weighing sensors and calibration components in the feeding hopper, the system uses the actual net flow rate of the upper hopper as the main feedback for closed-loop control when feeding impact occurs. After feeding is completed, it switches back to the main weighing body feedback, realizing closed-loop control under all working conditions. This fundamentally eliminates the problem of accuracy reduction caused by the traditional loss-in-weight scale having to switch to open-loop volume mode during the feeding impact phase.

[0017] 2. Through the combined effect of multiple algorithms such as weight acquisition calibration, Kalman filtering, hysteresis compensation and AI dynamic optimization, the influence of interference factors such as material impact, environmental vibration and temperature drift on the weight signal can be effectively eliminated, and the measurement error of the feeding impact section can be controlled within ±0.1%~±0.2%.

[0018] 3. Based on the compensation flow signal output by the calibration component, the control component controls the double helix mechanism to maintain the set speed difference, the first stirring mechanism to maintain a constant low to medium speed, and the second stirring mechanism to accelerate and break the arch in the impact compensation mode. The three work together to ensure that the discharge is stable, without blockage or pulsation during the feeding impact.

[0019] 4. The AI ​​optimization unit dynamically optimizes the compensation coefficient, adaptively adjusts the filtering parameters, and corrects the lag time by collecting multi-dimensional data in real time and combining it with historical operating experience. It can also predict the feeding impact based on historical operating data and adjust the parameters in advance to achieve intelligent control that is adaptive to the working condition and suppresses interference. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of a high-precision shock-resistant weight loss scale provided in an embodiment of this application; Figure 2 A step diagram illustrating the control method of a high-precision shock-resistant loss-of-weight scale provided in this application embodiment; Figure 3 Detailed diagram of a control method for a high-precision shock-resistant loss-of-weight scale provided in an embodiment of this application.

[0022] In the diagram: 1. Feeding bin; 2. Scale body; 3. First stirring mechanism; 4. Second stirring mechanism; 5. Double helix mechanism; 6. Transfer box. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0024] In the first aspect, the embodiments of this application provide a high-precision shock-resistant weightless scale, which mainly includes three parts: a body component, a calibration component, and a control component.

[0025] Figure 1 This application provides a schematic diagram of a high-precision, impact-resistant loss-of-weight scale structure, as shown in the embodiments. Figure 1 As shown, the main body component includes a feeding bin 1 and a scale body 2 disposed below the feeding bin 1.

[0026] The feeding hopper 1 is used to store the material to be measured. It has an inlet at the top and an outlet at the bottom. A load cell is installed at the bottom of the feeding hopper 1 to sense the real-time weight of the material in the feeding hopper 1 and convert the weight into a differential voltage signal output. The scale body 2 is the main weighing container of the loss-in-weight scale, used to receive the material from the feeding hopper 1 and perform weighing.

[0027] The upper part of the scale body 2 is provided with a first stirring mechanism 3 connected to the feeding bin 1. In one embodiment, the first stirring mechanism 3 is a vertical stirring mechanism. The stirring shaft of the first stirring mechanism 3 extends from the upper part of the scale body 2 into the bin, and the stirring range covers the upper area of ​​the scale body 2. It is used to stir the material in the upper part of the scale body 2 and prevent the material from bridging in the upper part of the bin.

[0028] Below the scale body 2, there is a second stirring mechanism 4 and a double helix mechanism 5 located at the discharge end of the second stirring mechanism 4. In one embodiment, the second stirring mechanism 4 is a spherical tank stirring mechanism, whose stirring shaft extends laterally from the side wall of the scale body 2 into it, and is used to break up and stir the material near the discharge port to ensure that the material smoothly enters the double helix mechanism 5.

[0029] The double helix mechanism 5 includes two parallel helical shafts, a main helix and a secondary helix, which rotate in opposite directions to force the material inside the scale body 2 to be conveyed and discharged. The input end of the double helix mechanism 5 is connected to a distribution box 6, and the input end of the distribution box 6 is connected to a servo motor, through which the power of the servo motor is synchronously distributed to the two helical shafts.

[0030] In one embodiment, the double helix mechanism 5, the first stirring mechanism 3, and the second stirring mechanism 4 each correspond to a servo motor. The first servo motor drives the double helix mechanism 5 to rotate synchronously through a transfer case. The second servo motor drives the first stirring mechanism 3 after amplifying the torque through a reducer. The third servo motor drives the second stirring mechanism 4 through a flexible coupling. A shock-absorbing pad is added to the servo motor base, and the servo speed output uses a first-order inertial filter to prevent vibrations caused by sudden speed changes from being transmitted to the scale body 2.

[0031] In one embodiment, the feeding hopper 1 and the scale body 2 are sealed together by a flexible dustproof connecting sleeve (soft connection) to ensure that the vibration of the feeding hopper 1 (such as material impact) will not be transmitted to the scale body 2 and its associated weighing sensor. The discharge port of the scale body 2 is connected to the feed box of the double helix mechanism 5 by a flexible rubber compensating joint to further isolate the vibration interference at the discharge end.

[0032] The calibration component is associated with the scale body 2 and includes a data acquisition module for obtaining the weight signal of the upper compartment of the feeding hopper 1. The signal input terminal of the data acquisition module is electrically connected to the signal output terminal of the weighing sensor to obtain the upper compartment weight signal of the feeding hopper 1. The data acquisition module has a built-in edge computing unit, which includes a weight acquisition and calibration unit, a flow calculation unit, a filtering unit, a hysteresis compensation unit, and an AI optimization unit.

[0033] The acquisition module also includes a high-precision analog-to-digital converter (ADC). The weighing sensor signal output terminal of the feeding hopper 1 is connected to the high-precision ADC, which is connected to the edge computing unit via an SPI interface. The edge computing unit has the aforementioned algorithm units built into it.

[0034] In one implementation, the load cell is a six-wire sensor, with its positive power supply terminal ( EXC+ ) connected to the positive power supply terminal of the high-precision analog-to-digital converter unit ( VDD ), negative power supply terminal ( EXC- ) connected to the negative power supply terminal of the high-precision analog-to-digital converter unit ( GND ), positive signal ( SIG+ ) connected to the first signal input terminal of the high-precision analog-to-digital converter unit ( AIN0 ), negative end of signal ( SIG- ) connected to the second signal input terminal of the high-precision analog-to-digital converter unit ( AIN1 ), sensing positive extreme ( SEN+ ) and the positive terminal of the power supply ( EXC+ Short-circuit, inducting the negative terminal ( SEN- ) and the negative terminal of the power supply ( EXC- Short circuit.

[0035] In standard six-wire advanced usage SEN+ and SEN- The induction wire should be connected separately to the reference voltage input terminal of the analog-to-digital converter unit for real-time correction of wire loss and remote compensation. In this embodiment, however, the induction wire is... SEN+ and EXC+ Short circuit, SEN- and EXC- Short-circuiting is equivalent to turning the induction line into a parallel branch of the power supply line, thus halving the total resistance of the power supply line. When changes in the weight of the feeding hopper cause current fluctuations, halving the line resistance means that the voltage drop fluctuation is also halved, thereby indirectly improving the stability of the excitation voltage and ensuring the accuracy of the ratiometric measurement.

[0036] In one implementation, the edge computing unit is a Raspberry Pi, which is fixedly mounted on a mounting bracket on the side of the feeding hopper 1, and transmits signals to the high-precision analog-to-digital converter unit via a ribbon cable. Mounting the edge computing unit close to the sensor results in an extremely short transmission path for analog small signals (mV level), greatly reducing induced noise introduced by long-line transmission and significantly decreasing sensitivity to external electromagnetic interference.

[0037] The control component is communicatively connected to the calibration component, used to receive the compensated flow signal output by the calibration component, and to control the double helix mechanism 5, the first stirring mechanism 3, and the second stirring mechanism 4 to operate collaboratively based on the compensated flow signal. The control component is installed in the mounting cavity on the side of the scale body 2. In one embodiment, it is controlled by... Modbus RTU The communication and edge computing unit is connected, and the three output control signals are connected to the drivers of three servo motors respectively.

[0038] Furthermore, high-precision analog-to-digital converters (such as the ADS1256) operate at high sampling rates (e.g., 50 times per second, i.e., sampling period). The differential voltage signal output by the weighing sensor is acquired at 0.02s and converted into a digital signal. Due to zero-point drift and system errors in the weighing sensor, the weight acquisition and calibration unit built into the edge computing unit calibrates the acquired values.

[0039] The calibration formula is: ; in, This is the real-time weight of feeding hopper 1 (unit: kg). The raw values ​​collected by the high-precision analog-to-digital conversion unit. This is the collected value corresponding to the zero point of the sensor (i.e., the AD value collected when the feeding hopper is empty). The range of the weighing sensor (unit: kg). This is the collected value corresponding to the full scale of the sensor (i.e., the AD value collected when the feeding hopper is fully loaded).

[0040] The system automatically completes zero-point calibration (data acquisition) after power-on. Automatically completes full-scale calibration every 24 hours (data acquisition). This ensures data acquisition accuracy. If the ambient temperature changes significantly between two full-scale calibrations, the system can adjust the calibration based on the built-in temperature compensation coefficient. AD The value is corrected to further suppress the impact of temperature drift on accuracy.

[0041] The flow calculation unit calculates the instantaneous material discharge flow rate based on the real-time changes in the weight of the feeding hopper 1. The calculation formula is:

[0042] ; in, The instantaneous discharge flow rate of feed hopper 1 (unit: kg / s). This represents the weight of feed hopper 1 from the previous data collection cycle. Δt represents the weight of feed hopper 1 in the current data collection cycle, and Δt represents the data collection cycle.

[0043] In this embodiment, the acquisition period The sampling frequency is set to 0.02s, or 50 times per second. This sampling frequency can fully capture the dynamic changes of material falling and is also adapted to the computing power level of the edge computing unit, ensuring real-time performance.

[0044] Since the raw weight signal inevitably contains interference components such as material impact and environmental vibration, the filtering unit uses the Kalman filter algorithm to filter the instantaneous flow rate. The core formula of the Kalman filter algorithm is:

[0045] ; ; in, This is the current filtered weight status value. This is the weight state value after filtering in the previous cycle. This is the state transition matrix (set to 1). For the control matrix (set to 0). To control the input amount, For process noise, This is the currently collected raw weight value. The observation matrix (set to 1). To mitigate noise, this algorithm effectively filters signal fluctuations, ensuring the stability of flow calculations.

[0046] Because there is a certain vertical distance between the feeding bin 1 and the scale body 2, it takes a certain amount of time (i.e., lag time τ) for the material to fall from the feeding bin 1 to the scale body 2. During this time, the weight of the material in the feeding bin 1 has decreased, but the corresponding material has not yet reached the scale body 2. If the instantaneous flow rate Q_in of the feeding bin 1 is directly used as the control feedback, it will lead to control lead and produce errors.

[0047] The lag compensation unit uses a lag compensation algorithm to correct this error. The compensation formula is as follows: ; in, This is the amount of delayed compensation. The instantaneous discharge flow rate of the feeding hopper is 1. The lag time for material to fall from the feeding bin 1 to the scale body 2. Data acquisition cycle. Lag time. The setting is based on the vertical distance between the feeding bin 1 and the scale body 2, with a typical value of 0.1~0.3s, which can be adaptively adjusted according to the actual installation dimensions.

[0048] The AI ​​optimization unit is integrated into the edge computing unit and employs a lightweight AI model (deployed based on TensorFlow Lite, adapted to the computing power of the edge computing unit). The AI ​​optimization unit collects instantaneous traffic in real time. Based on multi-dimensional data such as the rotational speed of the double helix mechanism 5, the rotational speed of the first stirring mechanism 3, the rotational speed of the second stirring mechanism 4, and the weight fluctuation value of the scale body 2, combined with historical operating data (stored locally on the edge computing unit for cumulative learning and optimization), the following optimization operations are performed:

[0049] a) Dynamically optimized compensation coefficient K: The AI ​​model outputs the optimal compensation coefficient K in real time (range 0.95~1.05), replacing the fixed initial value, solving the compensation deviation problem caused by changes in material bulk density and material level, and ensuring compensation under different working conditions. Q_real The accuracy.

[0050] b) Adaptive adjustment of filtering parameters: The AI ​​model identifies the intensity of signal interference (such as the impact intensity of feeding and the amplitude of stirring vibration) in real time and automatically adjusts the process noise of the Kalman filter algorithm. and observation noise Parameters. When the impact interference is large, increase the filter strength to avoid distortion of flow calculation; when the operating conditions are stable, decrease the filter strength to improve the response speed.

[0051] c) Adaptive correction of lag time τ: The AI ​​model automatically corrects the lag time τ based on real-time data of material falling (feeding amount from the hopper, material density) to solve the lag compensation deviation caused by different falling speeds of different materials (powder, granules).

[0052] d) Prediction and early correction of accuracy: The AI ​​model predicts the intensity and duration of the feeding impact based on historical operating data. Before the feeding signal is triggered, the compensation parameters and control parameters are adjusted in advance to suppress the impact interference and avoid fluctuations in metering accuracy.

[0053] The final formula for calculating the true net flow is: ; in, For true net flow, The dynamic compensation coefficient optimized for AI. This is the hysteresis compensation amount after AI correction. Through AI optimization, the metering error of the feeding impact section can be further reduced, ensuring that the overall accuracy is stable at ±0.1%~±0.2%, achieving high-precision closed-loop control.

[0054] Secondly, embodiments of this application provide a control method for a high-precision shock-resistant loss-of-weight scale, employing the aforementioned high-precision shock-resistant loss-of-weight scale, specifically as follows: S1: The high-precision analog-to-digital conversion unit of the acquisition module acquires the upper chamber weight signal from the six-wire weighing sensor of the feeding hopper 1 and converts it into a digital quantity. The built-in computing module of the edge computing unit performs zero-point drift and system error calibration on the raw digital quantity. The calibration formula is described above and will not be repeated here. The system automatically completes zero-point calibration after power-on and automatically completes full-scale calibration according to the set cycle (such as every 24 hours).

[0055] S2: The calculation module calculates the instantaneous discharge flow rate based on the calibrated weight signal and eliminates signal interference through a filtering algorithm. The calculation module calculates the instantaneous discharge flow rate of feeding hopper 1 based on the calibrated weight signal W. . The calculation formula has been described above and will not be repeated here. Δt The acquisition period is set to 0.01s~0.05s (0.02s is preferred in this embodiment).

[0056] Subsequently, the calculation module employs a Kalman filter algorithm to filter the instantaneous flow rate, eliminating signal interference caused by material impact, environmental vibration, and other factors. The Kalman filter parameters are adaptively adjusted by the AI ​​optimization unit based on the real-time signal interference intensity.

[0057] S3: The calculation module uses a hysteresis compensation algorithm to correct the hysteresis error of material falling from the feeding hopper 1 to the scale body 2, and dynamically optimizes the compensation parameters through the AI ​​optimization unit, outputting the compensated true net flow rate signal to the control component. The hysteresis compensation formula is as described above and will not be repeated here. Subsequently, the AI ​​optimization unit dynamically optimizes the compensation coefficient K and adaptively adjusts the filtering parameters by real-time acquisition of instantaneous flow rate, rotation speed of the double helix mechanism 5, rotation speed of the first stirring mechanism 3, rotation speed of the second stirring mechanism 4, and weight fluctuation value of the scale body 2, combined with historical operating data.

[0058] Final True Net Flow The calculation formula is as described above and will not be repeated here. The calculation module will use the compensated true net flow signal. The data is transmitted to the control component via the communication interface.

[0059] S4: When the feeding signal is triggered (i.e., feeding hopper 1 begins feeding material into scale body 2), the control component immediately switches to impact compensation mode. In this mode, the control component uses the compensated true net flow rate signal output by the calibration component. The main feedback mechanism controls the coordinated adjustment of the double helix mechanism 5, the first stirring mechanism 3, and the second stirring mechanism 4.

[0060] The two screws of the double helix mechanism 5 maintain a set speed difference (the main and auxiliary screws maintain a speed difference of 15% to 30%) to ensure stable and pulsation-free material discharge; The first stirring mechanism 3 maintains a constant low to medium speed to ensure that the material on the upper part of the scale body 2 does not bridge. The second mixing mechanism 4 accelerates to break up arches and prevents the material near the discharge port from being compacted.

[0061] Meanwhile, the transfer case further suppresses flow rate fluctuations by compensating for backlash through servo torque feedforward.

[0062] S5: When feeding is complete, the control component switches back to stable mode, using the weight signal of the scale body 2 as feedback to achieve closed-loop control. In stable mode, the control component uses the weight signal of the scale body 2 as feedback to achieve standard... PID Closed-loop control. The AI ​​optimization unit continues to run in real time, assisting in correcting the flow deviation of the main scale and ensuring high-precision metering under all operating conditions. Through the above steps, this invention uses the compensated real net flow signal as the main feedback for closed-loop control throughout the feeding impact process, realizing the transformation from open-loop to closed-loop in the feeding impact section, with no open-loop operation throughout the entire process.

[0063] In the description of this application, it should be noted that the terms "upper," "lower," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two elements. For those skilled in the art, the specific meaning of the above terms in this application can be understood according to the specific circumstances.

[0064] It should be noted that in this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0065] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A high-precision, shock-resistant loss-in-weight scale, characterized in that, include: The main body assembly includes a feeding bin (1), a scale body (2) flexibly connected to the bottom of the feeding bin (1), and a weighing sensor installed at the bottom of the feeding bin (1). The upper part of the scale body (2) is provided with a first stirring mechanism (3) connected to the feeding bin (1). The lower part of the scale body (2) is provided with a second stirring mechanism (4) and a double helix mechanism (5) located at the discharge end of the second stirring mechanism (4). The input end of the double helix mechanism (5) is connected to a transfer box (6). The calibration component is associated with the scale body (2) and includes a data acquisition module. The signal input terminal of the data acquisition module is electrically connected to the signal output terminal of the weighing sensor to acquire the upper weight signal of the feeding hopper. The data acquisition module has a built-in edge computing unit. The edge computing unit includes a weight acquisition calibration unit for calibrating zero drift and system error, a flow calculation unit for calculating instantaneous feeding flow, a filtering unit for eliminating signal interference, a hysteresis compensation unit for correcting material falling hysteresis error, and an AI optimization unit for dynamically optimizing compensation parameters. A control component, which is communicatively connected to the calibration component, is used to receive the compensated flow signal output by the calibration component and control the double helix mechanism (5), the first stirring mechanism (3) and the second stirring mechanism (4) to work together according to the compensated flow signal.

2. The high-precision shock-resistant loss-in-weight scale as described in claim 1, characterized in that, The acquisition module also includes a high-precision analog-to-digital conversion unit. The weighing sensor signal output terminal of the feeding bin (1) is connected to the high-precision analog-to-digital conversion unit. The high-precision analog-to-digital conversion unit is connected to the edge computing unit through the SPI interface.

3. A high-precision shock-resistant loss-in-weight scale as described in claim 2, characterized in that, The weighing sensor is a six-wire sensor. Its positive power supply terminal is connected to the positive power supply terminal of the high-precision analog-to-digital converter, its negative power supply terminal is connected to the negative power supply terminal of the high-precision analog-to-digital converter, its positive signal terminal is connected to the first signal input terminal of the high-precision analog-to-digital converter, its negative signal terminal is connected to the second signal input terminal of the high-precision analog-to-digital converter, its positive sensing terminal is short-circuited with the positive power supply terminal, and its negative sensing terminal is short-circuited with the negative power supply terminal.

4. A high-precision shock-resistant loss-in-weight scale as described in claim 1, characterized in that, The weight acquisition and calibration unit is equipped with a calibration formula: ; in, For the real-time weight of the feeding hopper (1), These are the raw values ​​obtained by the acquisition module. This is the collected value corresponding to the sensor's zero point. The weighing range of the load cell. This is the value collected at the full scale of the sensor.

5. A high-precision shock-resistant loss-in-weight scale as described in claim 1, characterized in that, The flow calculation unit is configured with an instantaneous flow calculation formula: ; in, For the instantaneous discharge flow rate of the feeding hopper (1), The weight of the feeding bin (1) in the previous collection cycle, The weight of the feeding bin (1) in the current collection cycle, The sampling period is specified; the filtering unit uses the Kalman filter algorithm to filter the instantaneous flow rate.

6. A high-precision shock-resistant loss-in-weight scale as described in claim 5, characterized in that, The lag compensation unit is configured with a lag compensation formula: ; in, This is the amount of delayed compensation. For the instantaneous discharge flow rate of the feeding hopper (1), The lag time for material to fall from the feeding bin (1) to the scale body (2) This refers to the data collection cycle.

7. A high-precision shock-resistant loss-in-weight scale as described in claim 1, characterized in that, The AI ​​optimization unit collects real-time data on instantaneous flow rate, rotation speed of the double helix mechanism (5), rotation speed of the first stirring mechanism (3), rotation speed of the second stirring mechanism (4), and weight fluctuation of the scale body (2). Combined with historical operating data, it dynamically optimizes the compensation coefficient and adaptively corrects the lag time, outputting the optimized true net flow rate. The calculation formula is as follows: ; in, For true net flow, For the instantaneous discharge flow rate of the feeding hopper (1), The dynamic compensation coefficient optimized for AI. This is the lag compensation amount after AI correction.

8. A control method for a high-precision shock-resistant loss-of-weight scale, employing the high-precision shock-resistant loss-of-weight scale as described in any one of claims 1 to 7, characterized in that, Includes the following steps: The acquisition module obtains the upper weight signal of the feeding hopper (1), and the edge computing unit performs zero-point drift and system error calibration on the upper weight signal; The edge computing unit calculates the instantaneous feed rate based on the calibrated weight signal and eliminates signal interference through a filtering algorithm; The edge computing unit uses a hysteresis compensation algorithm to correct the hysteresis error of the material falling from the feeding bin (1) to the scale body (2), and dynamically optimizes the compensation parameters through the AI ​​optimization unit, and outputs the compensated real net flow signal to the control component; When the feeding signal is triggered, the control component switches to the impact compensation mode, and uses the real net flow signal after compensation as the main feedback to control the double helix mechanism (5), the first stirring mechanism (3) and the second stirring mechanism (4) to adjust in coordination. In the impact compensation mode, the double helix mechanism (5) is driven by the servo motor through the transfer box (6), the two screws maintain the set speed difference, the first stirring mechanism (3) maintains a constant medium and low speed, and the second stirring mechanism (4) speeds up to break the arch. When feeding is finished, the control component switches back to stable mode and uses the weight signal of the scale body (2) as feedback to achieve closed-loop control.

9. The control method for a high-precision anti-impact loss-in-weight scale as described in claim 8, characterized in that: The formulas for zero-point drift and system error calibration are as follows: ; in, The real-time weight of the feeding hopper (1) These are the raw values ​​acquired in real time by the data acquisition module. This is the collected value corresponding to the zero point of the weighing sensor. The range of the weighing sensor. This represents the collected value corresponding to the full scale range of the weighing sensor; The formula for calculating the instantaneous feed rate is: ; in, The instantaneous discharge flow rate of the feeding hopper (1) The weight of the feeding bin (1) in the previous collection cycle, The weight of the feeding bin (1) in the current collection cycle, For the collection period, The filtering time is 0.01s to 0.05s. The filtering algorithm is the Kalman filtering algorithm, and its filtering parameters are adaptively adjusted by the AI ​​optimization unit according to the real-time signal interference intensity.

10. The control method for a high-precision anti-impact loss-in-weight scale as described in claim 8, characterized in that: The AI ​​optimization unit collects real-time data on instantaneous flow rate, rotation speed of the double helix mechanism (5), rotation speed of the first stirring mechanism (3), rotation speed of the second stirring mechanism (4), and weight fluctuation value of the scale body (2), and performs the following optimization operations in conjunction with historical operating data: The compensation coefficient K is dynamically optimized to replace the fixed initial value. Adaptive adjustment of filter parameters; The lag time τ is adaptively adjusted based on the material's falling speed. Based on historical operating data, predict the intensity and duration of the feeding impact, and adjust the compensation and control parameters in advance before the feeding signal is triggered; The final formula for calculating the true net flow is: ; in, For the actual net flow rate of the feeding silo (1), The instantaneous discharge flow rate of the feeding hopper (1) The dynamic compensation coefficients after optimization for the AI ​​optimization unit. The amount of hysteresis compensation corrected for the AI ​​optimization unit.