A Dynamic Weighing Method and Device for Tobacco Stems Based on Multi-Degree-of-Freedom Intelligent Damping
By using multi-degree-of-freedom intelligent damping technology, a coupled model of flow and vibration is constructed to correct the weighing signal in real time, which solves the metering accuracy and stability problems of the GF2 tobacco stem separation device in cigarette machines and improves weighing accuracy and stability.
Patent Information
- Application Number
- CN202611255169.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-18
- Publication Date
- 2026-11-17
AI Technical Summary
The existing GF2 tobacco stem separation device has problems with poor metering accuracy and unstable operation in cigarette making machines. In particular, it is difficult to achieve accurate weighing under multi-directional coupled vibration and flow disturbance, and it lacks multi-dimensional vibration acquisition, real-time flow prediction and adaptive damping control.
By employing multi-degree-of-freedom intelligent damping technology, a flow prediction model, a vibration prediction model, and a vibration-flow coupling error model are constructed. Combined with a magnetorheological damper and an adaptive control law, the weighing signal data is corrected in real time to obtain the actual weight of the tobacco stem.
It improves weighing accuracy and stability, reduces errors caused by vibration and flow disturbance, adapts to actual industrial scenarios, and meets the production needs of cigarette equipment.
Smart Images

Figure CN123111441A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of metering supporting cigarette making equipment, and specifically relates to a tobacco stem dynamic weighing method and device based on multi-degree-of-freedom intelligent damping. Background Art
[0002] At present, in the cigarette making, assembling and packaging links of the tobacco industry, the GF2 type tobacco stem separation device is key equipment used in supporting with high-speed cigarette makers, which is mainly responsible for the pneumatic separation, separation and conveying of tobacco stems during the cigarette making process. Its operation stability and metering accuracy directly affect cigarette quality control, tobacco stem reuse rate and production cost control. The supporting dynamic weighing link of the existing GF2 device has metering accuracy and stability that are difficult to meet the requirements of current refined production management. For example, Patent CN115077670A only adopts a unidirectional passive vibration damping structure, which is not designed for a three-axis coupled vibration environment; Patent CN113819985A can only separately handle weighing errors caused by a single interference, and does not realize coupled compensation of vibration and flow errors; while the general industry solutions generally have the problems of a single vibration damping mode and lack of compensation mechanisms.
[0003] During the operation of the GF2 device, its own mechanical vibration is superimposed with the excitations from the cigarette maker and upstream and downstream equipment, forming complex multi-directional coupled vibration. On the one hand, the high-speed operation of the internal fan and conveying mechanism of the GF2 device will generate inherent vibration, whose vibration frequency is close to the natural frequency of the weighing platform, which easily causes resonance; on the other hand, the periodic impact generated by the high-speed rotating components and cigarette making and assembling actions of the high-speed cigarette maker, as well as the vibration of the downstream tobacco stem conveying equipment, will be transmitted to the GF2 device through the frame and the ground, further intensifying the multi-directional vibration of the weighing platform. The existing single-axis modeling and vibration damping methods only perform passive vibration isolation for vertical vibration, cannot cover the coupled interference in horizontal and vertical directions, and cannot effectively suppress the superposition of multi-source vibrations of different frequencies, resulting in the vibration acceleration of the weighing platform remaining at a relatively high level for a long time. Actual measured data shows that its peak vibration acceleration can reach more than 0.5g, and the weighing error caused by vibration can be as high as 5.2%, which seriously affects the accuracy of metering results.
[0004] Meanwhile, fluctuations in the physical properties of the incoming tobacco stems and dynamic changes during the conveying process lead to frequent variations in the feed flow rate. During cigarette machine operation, the moisture content, density, and particle size of the incoming tobacco stems naturally differ. Combined with the instability of stem separation during air classification, this causes instantaneous fluctuations in the feed flow rate, with fluctuations exceeding ±10%. Furthermore, the conveyor belt speed of the GF2 unit is affected by factors such as load and equipment wear, resulting in slight fluctuations that further exacerbate flow disturbances. Existing devices do not integrate flow disturbance and vibration control in their design; they can only reduce errors through simple post-hoc averaging or fixed compensation coefficients, failing to achieve real-time prediction and advance compensation for flow disturbances. The weighing deviation caused by these combined disturbances is difficult to eliminate at its root. In actual production, the weighing error caused by flow fluctuations can reach 4.5%. When combined with vibration errors, the overall weighing error is typically around ±5.0%, far exceeding the accuracy control requirements of cigarette equipment.
[0005] Furthermore, existing systems lack an integrated structure for multi-dimensional vibration acquisition, real-time flow prediction, and adaptive damping control. Regarding vibration acquisition, existing solutions mostly use a single vertical acceleration sensor, which cannot comprehensively capture three-axis coupled vibration signals, leading to incomplete vibration error assessment. For flow prediction, simple moving average filtering is often used, which cannot effectively separate effective fluctuation signals from noise, resulting in low prediction accuracy and long lag times. In terms of control, there is a lack of active damping control mechanisms, and passive vibration isolation methods cannot adjust the vibration reduction effect according to real-time vibration conditions, nor can they coordinate with flow compensation. Simultaneously, the parameter calibration process of existing devices is incomplete, relying heavily on manual experience for debugging, lacking standardized multi-condition calibration methods, resulting in poor parameter adaptability and long and difficult on-site debugging cycles. On the hardware side, some solutions use non-standardized custom components, resulting in poor compatibility with the mechanical and electrical interfaces of the GF2 device. Modification and deployment require significant changes to the equipment, affecting the original structure and operational stability, increasing modification risks and costs.
[0006] Therefore, there is an urgent need for a dynamic weighing method and device for tobacco stems based on multi-degree-of-freedom intelligent damping to solve the problems of poor measurement accuracy and long-term instability in the dynamic weighing process of existing technologies, improve measurement accuracy and stability, and meet the actual production needs of cigarette equipment. Summary of the Invention
[0007] The purpose of this invention is to provide a dynamic weighing method and device for tobacco stems based on multi-degree-of-freedom intelligent damping, so as to solve the problems of poor measurement accuracy and long-term unstable operation in the dynamic weighing process in the prior art.
[0008] To achieve the above objectives, embodiments of the present invention provide a dynamic weighing method for tobacco stems based on multi-degree-of-freedom intelligent damping, comprising: Collect multidimensional data from weighing platforms and equipment, construct a flow prediction model, and optimize the weighing platform; The optimized weighing platform and equipment are re-collected and multi-dimensional data, including weighing signal data, vibration signal data and flow signal data. Using the aforementioned traffic prediction model, and based on the traffic signal data, the traffic disturbance prediction error is obtained; Construct a vibration prediction model and obtain the vibration error based on the vibration signal data; A vibration-flow coupling error model is constructed, and the coupling error is obtained based on the flow disturbance prediction error and the vibration error. Based on the vibration error, the flow disturbance prediction error, and the coupling error, the weighing signal data is corrected in real time to obtain the actual weight of the tobacco stem.
[0009] Optionally, multi-dimensional data from the weighing platform and equipment can be collected to construct a flow prediction model and optimize the weighing platform, including: Collect multidimensional data from the weighing platform and equipment, including flow time-series signal data and system state vector; A traffic prediction model is constructed, and the traffic disturbance prediction error value is obtained based on the traffic time series signal data; The active damping control force is obtained based on the predicted flow disturbance error value and the system state vector, and the weighing platform is optimized.
[0010] Optionally, constructing a traffic prediction model and obtaining the traffic disturbance prediction error value based on the traffic time-series signal data includes: Based on the aforementioned traffic flow time-series signal data, a discrete wavelet packet decomposition algorithm is used to obtain the traffic flow disturbance component sequence; An autoregressive prediction model is constructed, and the prediction error value of the flow disturbance component is obtained based on the flow disturbance component sequence.
[0011] Optionally, an autoregressive prediction model is constructed, and based on the flow disturbance component sequence, the flow disturbance prediction error value is obtained, including: Based on the flow disturbance component sequence, a training dataset for an autoregressive prediction model is constructed. The least squares method is used to iteratively fit the training dataset to obtain the regression coefficients for the autoregressive prediction model parameter calibration. Based on the regression coefficients, an autoregressive prediction model is constructed; The current flow disturbance sequence is input into the autoregressive prediction model to obtain the flow disturbance prediction error value.
[0012] Optionally, constructing an autoregressive prediction model based on the regression coefficients includes: Construct a regression function based on formula (1). (1) in, This is the predicted value of traffic flow disturbance. It is a continuous historical flow disturbance sequence. The regression coefficients are the parameters calibrated. This is the adaptive residual correction term.
[0013] Optionally, the active damping control force is obtained based on the flow disturbance prediction error value and the system state vector, and the weighing platform is optimized by including: Construct a multi-degree-of-freedom vibration model; Based on the multi-degree-of-freedom vibration model and the system state vector, a state-space equation is constructed. Stability analysis and control parameter calibration are performed based on the state-space equations to obtain displacement feedback gain, vibration suppression gain and flow compensation gain. An adaptive coupled control law is constructed based on the displacement feedback gain, vibration suppression gain, and flow compensation gain. The active damping control force is calculated based on the flow disturbance prediction error value. The balancing platform is optimized based on the active damping control force.
[0014] Optionally, an adaptive coupled control law is constructed based on the displacement feedback gain, vibration suppression gain, and flow compensation gain. The active damping control force is calculated based on the flow disturbance prediction error value, including: An adaptive coupling control law is constructed based on formula (2). (2) in, For active damping control force, For vibration displacement, For displacement feedback gain, For vibration velocity, To suppress vibration velocity gain, Gain to compensate for flow disturbance prediction errors. This represents the prediction error for flow disturbances.
[0015] Optionally, a vibration prediction model is constructed, and the vibration error is obtained based on the vibration signal data, including: A vibration prediction model is constructed based on formula (3). (3) in, This is the output value of the weighing sensor. For the actual quality of tobacco stems, It is the acceleration due to gravity. This is the weighing error coefficient. The vertical vibration acceleration, For measuring noise; Based on the vibration signal data, the vibration prediction model is used to obtain the vibration error.
[0016] Optionally, a vibration-flow coupling error model is constructed, and the coupling error is obtained based on the flow disturbance prediction error and the vibration error, including: The vibration-flow coupling error model is constructed using formula (4). (4) in, This is the coupling error. For dynamic weight coupling coefficients, For the looseness of the tobacco stems, For real-time traffic change rate, The amplitude of the vibration. For instantaneous vibration error, This represents the instantaneous flow rate error.
[0017] On the other hand, the present invention also provides a dynamic weighing device for tobacco stems based on multi-degree-of-freedom intelligent damping, the device comprising: The weighing platform optimization module is used to collect multi-dimensional data from the weighing platform and equipment, build a flow prediction model, and optimize the weighing platform. The multidimensional signal acquisition module includes a weighing execution unit and a data acquisition unit. The weighing execution unit uses sensors symmetrically installed at the four corners. The data acquisition unit is used to acquire multidimensional data of the weighing platform and equipment. The multidimensional data includes weighing signal data, vibration signal data and flow signal data. The flow disturbance error prediction module is used to obtain the flow disturbance prediction error based on the flow signal data using the flow prediction model. The vibration prediction module is used to construct a vibration prediction model and obtain the vibration error based on the vibration signal data. The coupling error acquisition module is used to construct a vibration-flow coupling error model and acquire the coupling error based on the flow disturbance prediction error and the vibration error. The weighing compensation module is used to correct the weighing signal data in real time based on the vibration error, the flow disturbance prediction error and the coupling error, so as to obtain the actual weight of the tobacco stem.
[0018] The beneficial effects of this invention are: This invention improves weighing accuracy by pre-optimizing the weighing platform before collecting actual data, reducing cross-coupling errors caused by flow disturbances. By constructing a flow prediction model, a vibration prediction model, and a vibration-flow coupling model, it overcomes the limitations of traditional single vibration modeling, accurately capturing the composite interference errors of dynamic weighing. This effectively solves the problems of strong vibration interference, large flow fluctuations, and low weighing accuracy in traditional tobacco stem dynamic weighing devices, making it suitable for real-world industrial scenarios.
[0019] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description
[0020] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a dynamic weighing method for tobacco stems based on multi-degree-of-freedom intelligent damping according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a specific method for collecting multidimensional data from a weighing platform and equipment, constructing a flow prediction model, and optimizing the weighing platform according to an embodiment of the present invention. Figure 3 This is a system block diagram of a dynamic weighing device for tobacco stems based on multi-degree-of-freedom intelligent damping according to an embodiment of the present invention. Detailed Implementation
[0021] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.
[0022] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.
[0023] like Figure 1 The diagram shows a flowchart of a dynamic weighing method for tobacco stems based on multi-degree-of-freedom intelligent damping according to an embodiment of the present invention. Figure 1 In this process, the weighing method may include the following steps: In step S10, multidimensional data of the weighing platform and equipment are collected, a flow prediction model is constructed, and the weighing platform is optimized. In step S11, multidimensional data of the optimized weighing platform and equipment are collected; In step S12, a flow prediction model is used to obtain the flow disturbance prediction error based on the flow signal data collected at the current moment; In step S13, a vibration prediction model is constructed, and the vibration error is obtained based on the vibration signal data; In step S14, a vibration-flow coupling error model is constructed, and the coupling error is obtained based on the flow disturbance prediction error and the vibration error. In step S15, the weighing signal data is corrected in real time based on vibration error, flow disturbance prediction error and coupling error to obtain the actual weight of the tobacco stem.
[0024] In such Figure 1 In the illustrated dynamic weighing method for tobacco stems based on multi-degree-of-freedom intelligent damping, step S10 is used to collect multi-dimensional data of the weighing platform and equipment, construct a flow prediction model, and optimize the weighing platform. In this embodiment, the specific methods for collecting multi-dimensional data of the weighing platform and equipment, constructing a flow prediction model, and optimizing the weighing platform in step S10 can be of various forms known to those skilled in the art. In this embodiment, step S10 can be as follows: Figure 2 The steps shown are described in this. Figure 2 In this context, step S10 may include: In step S20, multidimensional data of the weighing platform and equipment are collected; In step S21, a traffic prediction model is constructed, and the traffic disturbance prediction error value is obtained based on traffic time series signal data; In step S22, the active damping control force is obtained based on the flow disturbance prediction error value and the system state vector to optimize the weighing platform.
[0025] In such Figure 2 In the method shown, step S20 is used to collect multidimensional data of the weighing platform and equipment. In this embodiment, the specific method for collecting multidimensional data of the weighing platform and equipment in step S20 can be of various forms known to those skilled in the art. In this embodiment, firstly, displacement and velocity time-series data of the weighing platform are collected, and a system state vector is constructed based on the displacement and velocity time-series data. Subsequently, real-time raw load signals from the weighing sensor are collected at a fixed sampling frequency. In this embodiment, the sampling frequency is set to 200Hz, and a set of 200 discrete sampling points is buffered every second to construct a raw discrete signal sequence.
[0026] Step S21 is used to construct a flow prediction model and obtain the flow disturbance prediction error value based on the flow time series signal data. In this embodiment, step S21 includes: First, using a discrete wavelet packet decomposition algorithm to obtain the flow disturbance component sequence. Specifically, the commonly used db4 wavelet basis in the industrial metering field is selected, and a 3-level wavelet packet decomposition is set. After the 3-level decomposition, the signal is divided into 8 non-overlapping frequency band components. Among them, the frequency band division corresponds to the on-site interference, that is, the 0th frequency band corresponds to the slow disturbance of the tobacco stem flow and the change of the bulkiness of the stack, the 1st-5th frequency bands correspond to the airflow disturbance of the conveyor belt and the fan, and the 6th-7th frequency bands correspond to the mechanical vibration of the frame and the high-frequency noise of the motor.
[0027] The divided frequency band components are subjected to layer-by-layer discrete wavelet packet decomposition using a low-pass and high-pass conjugate filter bank, completing discrete convolution and downsampling operations. Specifically, the original signal is decomposed into low-frequency approximate components. and high-frequency detail components ; respectively , Decompose again to obtain , , , ;right , , , Continue with the frequency division by two to obtain 8 independent frequency domain subsequences.
[0028] Based on 8 independent frequency domain subsequences, an adaptive soft threshold noise reduction rule is set to remove frequency bands 1-7 corresponding to vibration and airflow, retaining only the low-frequency components of frequency band 0. This component only includes the time series curve of pure flow disturbance caused by changes in tobacco stem feed rate and packing porosity, thus avoiding interference from vibration signals in the judgment of flow disturbance.
[0029] Secondly, an autoregressive prediction model is constructed, and the prediction error value of the flow disturbance component sequence is obtained. Specifically, based on the triple rigid physical constraints of time-varying tobacco stem material, fixed parameters of the weighing hardware, and the upper limit of the rate of change of the material's physical state, a fourth-order autoregressive prediction model is selected. First, based on the high-precision sampling hardware module, the low-frequency flow disturbance time series sequence of tobacco stems after three-level decomposition and purification of discrete wavelet packets is analyzed. A fixed-window sliding sampling method was used to construct a training dataset for a fourth-order autoregressive prediction model, focusing only on the low-frequency drift characteristics of tobacco stem flocculent fiber materials.
[0030] Subsequently, the least squares method is used to iteratively fit the training dataset of the model, with the minimization of the sum of squared residuals of the true error in dynamic weighing of tobacco stems as the convergence objective, to solve for the fourth-order regression coefficients that adapt to the physical characteristics of tobacco stems. In this embodiment, a training convergence threshold of 10 is set. -4The maximum number of iterations is 200, ensuring that the final converged regression coefficients are only adapted to the hardware parameters and material characteristics of this tobacco stem dynamic weighing device, thus avoiding overfitting problems in higher-order models.
[0031] Furthermore, the regression function of the fourth-order autoregressive prediction model is constructed using formula (1): (1) in, This is the predicted value of traffic flow disturbance. It is a continuous historical flow disturbance sequence. The regression coefficients are the parameters calibrated. This is the adaptive residual correction term. In this embodiment, the adaptive residual correction term is driven by the hardware detection data of the real-time physical state of the tobacco stems for adaptive updates: when the weighing system sensor detects sudden changes in the humidity of the tobacco stems, large fluctuations in looseness, and the deviation between the model prediction value and the actual sampled value exceeds a preset threshold, the system automatically fine-tunes the regression coefficient, where the preset threshold can be set to ±0.3%.
[0032] Furthermore, the collected flow disturbance sequence at the current moment is input into a fourth-order autoregressive prediction model to obtain the predicted flow disturbance value. The flow disturbance prediction error value is then calculated using the following formula: ; in, Let be the prediction error value of the flow disturbance at time t. This refers to the specific functional relationship obtained through experimental fitting. The residence time of the tobacco stems on the conveyor belt, in this embodiment, is taken from a value ranging from [value missing]. .
[0033] Step S22 is used to obtain the active damping control force based on the flow disturbance prediction error value and the system state vector, thereby optimizing the weighing platform. In this embodiment, in order to realistically reflect the actual vibration state and improve the adaptability of the model to industrial scenarios, step S22 includes: First, constructing a multi-degree-of-freedom vibration model. Specifically, the weighing platform and vibration isolation system are equivalent to a three-degree-of-freedom mass-spring-damping system, and its dynamic equations are as follows: ; in, The equivalent mass matrix, in this embodiment, can be defined as: , These are the equivalent masses in the three axial directions, respectively; Let be the acceleration vector of the weighing platform in the three axes; The damping matrix, in this embodiment, can be defined as: , These are the damping coefficients in the three axial directions, respectively; Let be the acceleration vector of the weighing platform in the three axes; The stiffness matrix, in this embodiment, can be defined as: , These are the stiffness coefficients in the three axial directions, respectively; Let be the displacement vector of the platform in the three axes, which in this embodiment can be defined as , This represents the platform's displacement in the three axes; The external vibration excitation, in this embodiment, is caused by the operation of the tobacco stem conveying equipment; This is the active damping control force.
[0034] Then, the system state vector is defined: The system is transformed into a state-space form, and the state-space equations are constructed according to the following formula: ; in, Let be the system state vector. For the system matrix, To control the input matrix, To control the input, i.e., the active damping control force , This is the external excitation matrix.
[0035] The system matrix, control input matrix, and external excitation matrix can all be derived from the multi-degree-of-freedom vibration model. Specifically, based on the constructed multi-degree-of-freedom vibration model... , deformed By combining the system state vector, the system matrix can be obtained. Control input matrix External stimulus matrix .
[0036] Furthermore, stability analysis and control parameter calibration are performed based on the state-space equation to obtain displacement feedback gain, vibration suppression gain, and flow compensation gain.
[0037] Furthermore, based on the displacement feedback gain, vibration suppression gain, and flow compensation gain, an adaptive coupled control law is constructed according to formula (2), and the active damping control force is calculated: (2) in, For active damping control force, For vibration displacement, For displacement feedback gain, For vibration velocity, To suppress vibration velocity gain, Gain to compensate for flow disturbance prediction errors. This represents the error value for traffic disturbance prediction.
[0038] Furthermore, based on the calculated active damping control force, the damper is driven to suppress platform vibration in real time, thereby optimizing the weighing platform.
[0039] Step S10 integrates flow disturbance prediction into active damping control, outputting compensating damping force in advance to proactively counteract additional vibrations caused by flow changes before coupling errors occur, thus reducing cross-coupling errors at the hardware source. By dynamically adjusting the damping force based on real-time vibration status and predicted flow disturbances, the vibration of the weighing platform is suppressed, making it adaptable to various tobacco stem looseness and flow conditions.
[0040] Step S11 is used to collect multi-dimensional data of the optimized weighing platform and equipment. In this embodiment, the multi-dimensional data collected in step S11 may include weighing signal data, vibration signal data, and flow signal data. Subsequently, in step S12, a flow prediction model is used to predict flow disturbances in advance based on the real-time collected tobacco stem feed flow signal, and the flow disturbance prediction error is obtained.
[0041] Step S13 is used to construct a vibration prediction model and obtain the vibration error based on the real-time acquired vibration signal data. In this embodiment, step S13 constructs the vibration prediction model according to formula (3): (3) in, This is the output value of the weighing sensor. For the actual quality of tobacco stems, It is the acceleration due to gravity. The weighing error coefficient, in this embodiment, can be obtained through experimental calibration. The specific calibration method is as follows: Under conditions without external vibration interference, load tobacco stems of different known masses, collect the difference between the sensor output value and the actual mass, and calculate the weight error coefficient through linear fitting. The optimal value; It represents the vertical vibration acceleration; Noise, which is mainly caused by sensor error and environmental electromagnetic interference, can be effectively suppressed by the Kalman filter algorithm. The filter parameters can be calibrated according to the actual noise characteristics.
[0042] Subsequently, the weighing error term introduced by vibration is separated using the following formula: ; in, This represents vibration error.
[0043] Step S14 is used to construct a vibration-flow coupling error model, and obtain the coupling error based on the flow disturbance prediction error and vibration error. In this embodiment, step S14 includes: First, setting up a three-factor orthogonal experiment and collecting coupling error benchmark data. Specifically, the control variables of the three-factor orthogonal experiment are the tobacco stem flow rate, material looseness, and frame vibration amplitude, while other factors such as ambient temperature and humidity, conveyor belt speed, etc. remain unchanged.
[0044] With a fixed tobacco stem flow rate and varying material bulkness, multiple sets of pure flow rate errors were collected; with a fixed material bulkness and varying frame vibration amplitude, multiple sets of pure vibration errors were collected; with simultaneous changes in tobacco stem flow rate, material bulkness, and frame vibration amplitude, while keeping other parameters constant, the total measured error under the combined effects of flow rate and vibration was collected.
[0045] The difference between the superposition of pure flow error and pure vibration error and the total measured error is calculated and used as the original sample of coupling error. Through full-condition orthogonal experiments, cross-coupling error datasets corresponding to different flow rates, looseness, and vibration amplitudes are obtained.
[0046] Subsequently, using tobacco stem looseness, real-time flow rate change rate, and vibration amplitude as three-dimensional input variables, and cross-coupling error as the dependent variable, a multivariate linear fitting was employed to obtain the dynamic coupling weight coefficient, with a value range of [missing information]. .
[0047] Furthermore, the vibration-flow coupling error model is constructed using formula (4) to obtain the coupling error: (4) in, This is the coupling error. For dynamic weight coupling coefficients, For the looseness of the tobacco stems, For real-time traffic change rate, The amplitude of the vibration. For instantaneous vibration error, This represents the instantaneous flow disturbance error.
[0048] Step S15 is used to perform real-time correction on the weighing signal data based on vibration error, flow disturbance prediction error, and coupling error to obtain the actual weight of the tobacco stem. In this embodiment, step S15 performs real-time correction on the weighing signal data according to the following formula: ; in, This refers to the actual weight of the tobacco stems at the final output. This is the output value of the weighing sensor.
[0049] On the other hand, embodiments of the present invention also provide a dynamic weighing device for tobacco stems based on multi-degree-of-freedom intelligent damping. The structural block diagram of this device is shown below. Figure 3 As shown. In this Figure 3 The device includes a weighing platform optimization module, a multi-dimensional signal acquisition module, a flow disturbance error prediction module, a vibration prediction module, a coupling error acquisition module, and a weighing compensation module. The weighing platform optimization module collects multi-dimensional data from the weighing platform and equipment, constructs a flow prediction model, and optimizes the weighing platform. The multi-dimensional signal acquisition module includes a weighing execution unit and a data acquisition unit. The weighing execution unit uses sensors symmetrically installed at the four corners, and the data acquisition unit collects multi-dimensional data from the weighing platform and equipment, including weighing signal data, vibration signal data, and flow signal data. The flow disturbance error prediction module uses the flow prediction model to obtain the flow disturbance prediction error based on the flow signal data. The vibration prediction module constructs a vibration prediction model and obtains the vibration error based on the vibration signal data. The coupling error acquisition module constructs a vibration-flow coupling error model and obtains the coupling error based on the flow disturbance prediction error and the vibration error. The weighing compensation module performs real-time correction on the weighing signal data based on the vibration error, flow disturbance prediction error, and coupling error to obtain the actual weight of the tobacco stems.
[0050] In one embodiment of the present invention, the weighing platform adopts a rectangular frame structure. The frame material is lightweight, high-strength aluminum alloy, and the upper surface of the frame is covered with anti-slip and wear-resistant rubber pads to prevent weighing errors caused by the slippage of tobacco stems. The weighing sensor adopts four tension-type weighing sensors, which are respectively installed at the four corners of the bottom of the weighing platform and connected to the support base through buffer connectors. The buffer connectors are made of composite rubber material with an embedded metal skeleton. Their axial stiffness is 5000 N / m and their radial stiffness is 3000 N / m, which not only plays a buffering role but also prevents the vibration of the support base from being directly transmitted to the weighing platform. The support base adopts a welded steel structure with leveling bolts at the bottom, which can be used to adjust the level according to the on-site working conditions to ensure the stability of the weighing platform.
[0051] A triaxial piezoelectric accelerometer, signal conditioner, and data acquisition card are used to collect multi-directional vibration signals from the weighing platform and conveyor equipment in real time. Two triaxial accelerometers are installed at the center of the bottom of the weighing platform and on the frame of the conveyor equipment, respectively. The sensor installed at the bottom of the weighing platform collects vibration acceleration signals in the x, y, and z axes, while the sensor installed on the conveyor frame collects external excitation vibration signals, achieving bidirectional acquisition of vibration signals. The signal conditioner uses a differential amplifier to amplify and filter the weak vibration signals collected by the sensors, suppressing electromagnetic interference. The processed signal amplitude range is 0-5V. The data acquisition card uses a USB interface and can be set to an adjustable sampling frequency of 50-200Hz, with a default setting of 100Hz, converting analog vibration signals into digital signals. A belt-type flow sensor and signal converter are used to collect tobacco stem feed flow signals in real time. The belt-type flow sensor is installed below the tobacco stem inlet, in close contact with the conveyor belt, to collect tobacco stem feed flow signals in real time. The signal converter converts the pulse signals output by the flow sensor into digital signals.
[0052] A magnetorheological damper, drive module, and mounting bracket are used to suppress multi-directional vibrations of the weighing platform. Four magnetorheological dampers are installed between the weighing platform and the support base, corresponding to the installation positions of the load cells, in a diagonally symmetrical arrangement. They are connected to the weighing platform and support base via ball joints, allowing for multi-directional damping force adjustment. The magnetorheological damper adopts a dual-excitation coil design with internal dual-annular throttling channels, corresponding to vibration suppression in the x and z directions respectively. The damping adjustment range is 500-5000N, and the response time is ≤5ms. The drive module uses a constant current drive circuit to convert the control signal output by the controller into a drive current, adjusting the excitation coil current of the magnetorheological damper, thereby adjusting the damping force. The mounting bracket has an adjustable angle design, allowing the installation angle of the damper to be adjusted according to the vibration direction on site, ensuring that the damping force can be accurately applied to the vibration direction.
[0053] It adopts an industrial-grade touch screen, with the Weintek MT8102iE model as an option, for parameter setting, data display, and fault alarm. It can display parameters such as vibration acceleration, flow rate, and weighing value in real time, and supports manual adjustment of parameters such as control gain and sampling frequency. It uses the Modbus communication protocol to realize signal transmission between the controller and various sensors and actuators, and also supports interface with the factory's industrial control system to realize data upload and remote monitoring.
[0054] The beneficial effects of this invention are: This invention employs three-degree-of-freedom coupled modeling, equating the weighing platform and vibration isolation system to a three-degree-of-freedom mass-spring-damping system. Compared to traditional single-degree-of-freedom modeling that only considers vertical vibration, this more realistically reflects the three-axis vibration coupling characteristics of industrial environments, improving the model's adaptability to industrial scenarios. By directly integrating flow disturbance prediction into active damping control, combined control of flow compensation and vibration suppression is achieved, fundamentally reducing weighing errors caused by composite disturbances. Compared to existing active damping control technologies that only consider vibration suppression while ignoring the coupling effect of flow disturbances, this improves control performance. Furthermore, by constructing an adaptive control law, the gain parameters can be automatically adjusted based on vibration intensity and flow fluctuation amplitude, adapting to different industrial conditions.
[0055] This invention separates the effective and interference signals in the flow disturbance signal through discrete wavelet packet decomposition, eliminating invalid components such as environmental interference and sensor noise, thereby improving prediction accuracy. Compared with traditional conventional Fourier transform, which cannot distinguish between instantaneous flow changes and high-frequency noise from frame vibration, discrete wavelet packet decomposition can break down the original weighing time-series signal into multiple layers and multiple frequency bands of independent components, eliminating vibration clutter interference. By employing a fourth-order autoregressive prediction model, adapted to the physical change law of tobacco stem material and matched with the hardware parameters of the device, flow disturbances are predicted in advance, unlike traditional direct acquisition of instantaneous flow with lag compensation, achieving predictive dynamic compensation. Furthermore, the experimental data for the autoregressive prediction model uses actual tobacco stem flow data collected in industrial settings, covering different production periods and different tobacco stem types, improving the versatility of the prediction model.
[0056] This invention achieves joint compensation by coupling vibration error and flow disturbance error, thus fundamentally improving weighing accuracy compared to existing technologies that only consider a single error source. It discovers that fluctuations in feed flow rate alter the tobacco stem accumulation state, thereby changing the load characteristics of the weighing platform and amplifying the offset effect caused by vibration. The vibration error and flow disturbance error exhibit cross-coupling responses, leading to the introduction of a dynamically adjustable coupled error term, significantly improving dynamic weighing accuracy.
[0057] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0058] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0059] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0060] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0061] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0062] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.
[0063] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.
[0064] It should also be noted that 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 process, method, article, or apparatus. Unless otherwise specified, 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 that element.
[0065] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.
Claims
1. A dynamic weighing method for tobacco stems based on multi-degree-of-freedom intelligent damping, characterized in that, The method includes: Collect multidimensional data from weighing platforms and equipment, construct a flow prediction model, and optimize the weighing platform; Collect multidimensional data of the optimized weighing platform and equipment, including weighing signal data, vibration signal data and flow signal data; Using the aforementioned traffic prediction model, and based on the traffic signal data, the traffic disturbance prediction error is obtained; Construct a vibration prediction model and obtain the vibration error based on the vibration signal data; A vibration-flow coupling error model is constructed, and the coupling error is obtained based on the flow disturbance prediction error and the vibration error. Based on the vibration error, the flow disturbance prediction error, and the coupling error, the weighing signal data is corrected in real time to obtain the actual weight of the tobacco stem.
2. The method according to claim 1, characterized in that, Collect multidimensional data from weighing platforms and equipment, construct a flow prediction model, and optimize the weighing platform, including: Collect multidimensional data from the weighing platform and equipment, including flow time-series signal data and system state vector; A traffic prediction model is constructed, and the traffic disturbance prediction error value is obtained based on the traffic time series signal data; The active damping control force is obtained based on the predicted flow disturbance error value and the system state vector, and the weighing platform is optimized.
3. The method according to claim 2, characterized in that, Constructing a traffic prediction model, and obtaining traffic disturbance prediction error values based on the traffic time-series signal data, includes: Based on the aforementioned traffic flow time-series signal data, a discrete wavelet packet decomposition algorithm is used to obtain the traffic flow disturbance component sequence; An autoregressive prediction model is constructed, and the prediction error value of the flow disturbance component is obtained based on the flow disturbance component sequence.
4. The method according to claim 3, characterized in that, Constructing an autoregressive prediction model, based on the flow disturbance component sequence, and obtaining the flow disturbance prediction error value includes: Based on the flow disturbance component sequence, a training dataset for an autoregressive prediction model is constructed. The least squares method is used to iteratively fit the training dataset to obtain the regression coefficients for the autoregressive prediction model parameter calibration. Based on the regression coefficients, an autoregressive prediction model is constructed; The current flow disturbance sequence is input into the autoregressive prediction model to obtain the flow disturbance prediction error value.
5. The method according to claim 4, characterized in that, Based on the regression coefficients, constructing an autoregressive prediction model includes: Construct a regression function based on formula (1). ,(1) in, This is the predicted value of traffic flow disturbance. It is a continuous historical flow disturbance sequence. The regression coefficients are the parameters calibrated. This is the adaptive residual correction term.
6. The method according to claim 2, characterized in that, Based on the predicted flow disturbance error value and the system state vector, the active damping control force is obtained, and the weighing platform is optimized by: Construct a multi-degree-of-freedom vibration model; Based on the multi-degree-of-freedom vibration model and the system state vector, a state-space equation is constructed. Stability analysis and control parameter calibration are performed based on the state-space equations to obtain displacement feedback gain, vibration suppression gain and flow compensation gain. An adaptive coupled control law is constructed based on the displacement feedback gain, vibration suppression gain, and flow compensation gain. The active damping control force is calculated based on the flow disturbance prediction error value. The balancing platform is optimized based on the active damping control force.
7. The method according to claim 6, characterized in that, An adaptive coupled control law is constructed based on the displacement feedback gain, vibration suppression gain, and flow compensation gain. The active damping control force is calculated based on the flow disturbance prediction error value, including: An adaptive coupling control law is constructed based on formula (2). ,(2) in, For active damping control force, For vibration displacement, For displacement feedback gain, For vibration velocity, To suppress vibration velocity gain, Gain to compensate for flow disturbance prediction errors. This represents the prediction error for flow disturbances.
8. The method according to claim 1, characterized in that, Constructing a vibration prediction model and obtaining vibration errors based on the vibration signal data includes: A vibration prediction model is constructed based on formula (3). ,(3) in, This is the output value of the weighing sensor. For the actual quality of tobacco stems, It is the acceleration due to gravity. This is the weighing error coefficient. The vertical vibration acceleration, For measuring noise; Based on the vibration signal data, the vibration prediction model is used to obtain the vibration error.
9. The method according to claim 1, characterized in that, Constructing a vibration-flow coupling error model, and obtaining the coupling error based on the flow disturbance prediction error and the vibration error, includes: The vibration-flow coupling error model is constructed using formula (4). ,(4) in, This is the coupling error. For dynamic weight coupling coefficients, For the looseness of the tobacco stems, For real-time traffic change rate, The amplitude of the vibration. For instantaneous vibration error, This represents the instantaneous flow disturbance error.
10. A dynamic weighing device for tobacco stems based on multi-degree-of-freedom intelligent damping, characterized in that, The device includes: The weighing platform optimization module is used to collect multi-dimensional data from the weighing platform and equipment, build a flow prediction model, and optimize the weighing platform. The multidimensional signal acquisition module includes a weighing execution unit and a data acquisition unit. The weighing execution unit uses sensors symmetrically installed at the four corners. The data acquisition unit is used to acquire multidimensional data of the weighing platform and equipment. The multidimensional data includes weighing signal data, vibration signal data and flow signal data. The flow disturbance error prediction module is used to obtain the flow disturbance prediction error based on the flow signal data using the flow prediction model. The vibration prediction module is used to construct a vibration prediction model and obtain the vibration error based on the vibration signal data. The coupling error acquisition module is used to construct a vibration-flow coupling error model and acquire the coupling error based on the flow disturbance prediction error and the vibration error. The weighing compensation module is used to correct the weighing signal data in real time based on the vibration error, the flow disturbance prediction error and the coupling error, so as to obtain the actual weight of the tobacco stem.
Citation Information
Patent Citations
Wafer anti-interference weighing device and application thereof
CN113819985A
