A plum kernel and pulp separation chute conveying device and method

Abstract

The present application relates to the technical field of agricultural product processing, and discloses a plum kernel-meat separation chute conveying device and method, which obtains real-time rotation state of double excitation vibrators and extracts dynamic phase error, obtains instantaneous angular velocity deviation through filtering processing, calculates parasitic lateral disturbance force induced by error, introduces an experience forgetting factor to perform state-decay discrete quadratic integration on the disturbance force to obtain lateral slip accumulation displacement, performs nonlinear coupling on accumulated shrinkage of the chute opening degree in a limited time window and the lateral displacement to calculate dynamic jam risk margin, and finally extracts one-way deterioration gain as an adaptive weight according to the risk margin to generate an instantaneous frequency compensation instruction for feedforward control of a lagging side excitation vibrator. The device comprises a memory and a processor, and the above method program is executed by the processor. The present application realizes dynamic prediction and active feedforward inhibition of jam risk, and greatly improves the stability and production efficiency of the conveying system.

Description

Technical Field

[0001] This invention belongs to the field of agricultural product processing technology, and in particular relates to a plum pit and flesh separation chute conveying device and method. Background Technology

[0002] In the industrial processing of dried plums and stone fruits, the chute conveyor system is a core component, undertaking the crucial tasks of material orientation, grading, and separation of pits and flesh. These systems typically employ a dual-vibrator drive structure, using the inertial force generated by vibration to achieve continuous throwing and sliding motion of materials within the chute. However, in actual industrial operation, factors such as manufacturing deviations in the vibrator, random fluctuations in electrical system load, and uneven material distribution can easily disrupt the phase synchronization of the dual vibrators. This phase asynchrony directly induces complex parasitic lateral interference forces, leading to unexpected lateral displacement and localized accumulation of materials within the chute. When the lateral accumulation reaches a critical threshold and couples with the gradually contracting geometry of the chute along the conveying direction, severe material jamming accidents are inevitable. Traditional control strategies primarily rely on fixed frequency adjustment mechanisms or hysteresis-based feedback control. These methods cannot accurately quantify the probability and evolution of jamming risks in real time within a dynamically changing production environment. They typically only trigger passive alarm responses after a jamming incident occurs, completely lacking the ability to proactively predict and suppress unsteady synchronization errors. This technological limitation not only causes frequent production interruptions and significantly reduces processing efficiency but also greatly increases the complexity of equipment maintenance and operating costs. Therefore, there is an urgent need to develop an intelligent conveying solution capable of real-time assessment of the dynamic characteristics of phase errors, accurate calculation of the impact of parasitic lateral disturbances, dynamic quantification of jamming risk margins, and implementation of adaptive feedforward control. This is crucial to overcome the current technological bottlenecks in the operational reliability and intelligence level of agricultural product processing equipment. Summary of the Invention

[0003] The purpose of this invention is to provide a plum pit and flesh separation chute conveying device and method to solve the above-mentioned problems.

[0004] This invention is implemented as follows: a method for conveying dried plum kernels and meat in a chute, comprising the following steps: Step S10: acquiring the real-time rotational state of the dual exciters and extracting the dynamic phase error, performing differential and first-order low-pass filtering on it to obtain the instantaneous angular velocity deviation; Step S20: calculating the parasitic lateral disturbance force induced by the unsteady synchronization of the dual exciters based on the dynamic phase error; Step S25: introducing an empirical forgetting factor reflecting the characteristics of disturbance dissipation and accumulation release, performing a discrete quadratic integral operation with state decay on the parasitic lateral disturbance force to obtain the lateral sliding accumulation displacement of the material; Step S30: acquiring the dimensionless cumulative shrinkage of the chute opening along the conveying direction within a limited time window, performing a nonlinear multiplication coupling operation with the lateral sliding accumulation displacement to calculate and obtain the dynamic jamming risk margin; Step S40: extracting the unidirectional risk deterioration gain based on the dynamic jamming risk margin, introducing it as an adaptive weight into the feedforward control equation to obtain the instantaneous frequency compensation command, and performing feedforward control on the lag-side exciter.

[0005] A further technical solution, in step S10, the process of obtaining the instantaneous angular velocity deviation includes: calculating the rate of change of the dynamic phase error at the current moment relative to the previous sampling moment, and using it as the current error increment; using a first-order low-pass filtering algorithm, weighting and fusing the current error increment with the historical instantaneous angular velocity deviation at the previous sampling moment according to a preset filtering weight, to obtain the instantaneous angular velocity deviation at the current moment, wherein the filtering weight is used to balance the dynamic response of the sensor and the suppression of high-frequency noise.

[0006] In a further technical solution, in step S20, the parasitic lateral disturbance force is determined by the centrifugal force vector synthesis of the dual exciter eccentric blocks when dynamic phase error exists; the amplitude of the parasitic lateral disturbance force is positively correlated with the physical properties of the exciter eccentric blocks and the positive function of the reference operating frequency, and exhibits a nonlinear trigonometric function mapping relationship with the change of dynamic phase error.

[0007] A further technical solution, in step S25, the discrete quadratic integral operation adopts a state recursive model with a double forgetting factor, specifically including: converting the parasitic lateral disturbance force at the current moment into an instantaneous lateral acceleration based on the equivalent total mass, and superimposing it with the lateral slip velocity at the previous moment after being attenuated by the first forgetting factor to obtain the lateral slip velocity at the current moment, where the first forgetting factor is used to characterize the dissipation characteristics of the lateral disturbance; converting the integral of the lateral slip velocity at the current moment into an instantaneous lateral displacement increment, and superimposing it with the lateral slip accumulation displacement at the previous moment after being attenuated by the second forgetting factor to obtain the lateral slip accumulation displacement at the current moment, where the second forgetting factor is used to characterize the system's release and reset characteristics for material accumulation.

[0008] A further technical solution, in step S30, includes the following nonlinear multiplicative coupling operation logic for calculating and obtaining the dynamic jamming risk margin: Based on the geometric gradient of the chute opening descent and the instantaneous longitudinal conveying speed within a limited time window, a first basic evaluation term characterizing the degree of longitudinal spatial contraction is constructed; based on the lateral sliding accumulation displacement and the geometric projection characteristics of the sidewall, a lateral penalty term characterizing the degree of lateral accumulation blockage is constructed; the first basic evaluation term and the lateral penalty term are multiplicatively coupled, and combined with the dynamic disturbance term obtained after normalization of the parasitic lateral disturbance force, the initial nominal safety margin of the system is subtracted, and the dynamic jamming risk margin is obtained through comprehensive evaluation.

[0009] A further technical solution, in step S40, includes the following steps: determining the changing trend of the dynamic jamming risk margin within adjacent sampling periods; when the dynamic jamming risk margin shows a deteriorating trend, generating an exponentially increasing amplification weight based on the deterioration rate; when the dynamic jamming risk margin shows a mitigating trend, forcibly shielding the influence of the deterioration rate to suppress oscillations; the feedforward control equation generates the final instantaneous frequency compensation command by proportionally calculating the deviation between the target safe jamming risk margin and the current true dynamic jamming risk margin, and multiplying it by the amplification weight in a feedforward manner.

[0010] A plum pit and meat separation chute conveying device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned plum pit and meat separation chute conveying method.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention overcomes the limitations of traditional methods that rely solely on single-dimensional monitoring or post-event alarms. It calculates the actual lateral slip-stacking displacement using a discrete quadratic integral and a state decay model, and nonlinearly couples this displacement with the longitudinal geometric shrinkage of the chute to construct a multi-dimensional dynamic jamming risk margin model. This model can quantify the probability of jamming under complex dynamic conditions in advance and with high precision.

[0012] 2. An innovative approach is adopted, incorporating unidirectional risk deterioration gain and exponential amplification weights to upgrade static risk assessment into a feedforward control system with trend-aware capabilities. When the system detects a risk deterioration trend, it can generate frequency compensation commands in real time at the initial stage, actively adjusting the lag-side exciter to suppress parasitic lateral interference caused by phase asynchrony at the source, effectively solving the problem of severe lag in response of traditional feedback control.

[0013] 3. By introducing a differential and first-order low-pass filtering mechanism to fuse historical data at the data acquisition layer, high-frequency noise glitches from industrial sensors are effectively filtered out, ensuring the stability and accuracy of core input data (instantaneous angular velocity deviation). Combined with a deeply integrated hardware and software architecture, this enables the efficient implementation of complex anti-jamming feedforward algorithms, effectively preventing frequent downtime due to misjudgments, extending equipment lifespan, and reducing maintenance costs. Attached Figure Description

[0014] Figure 1 A flowchart of a plum pit and flesh separation chute conveying method provided by the present invention; Figure 2 The block diagram of the logic and data flow of the anti-jamming control system provided by the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0017] like Figure 1 and Figure 2 As shown, a plum pit and flesh separation chute conveying method provided in one embodiment of the present invention includes the following steps: Step S10: Obtain the real-time rotational state of the dual exciters and extract the dynamic phase error. Perform differential and first-order low-pass filtering on the error to obtain the instantaneous angular velocity deviation. Dual exciters refer to two excitation devices, typically configured in pairs in vibratory conveying equipment, to generate vibration force. They generate periodic excitation force through the rotation of eccentric blocks, driving the chute to vibrate, thereby achieving material conveying. Real-time rotational state refers to the instantaneous information of dynamic parameters such as rotational speed and phase of the dual exciters during operation. This information is usually continuously monitored and collected by sensors. Dynamic phase error refers to the instantaneous difference between the rotational phases of the dual exciters during synchronous operation due to various disturbance factors. Ideally, the dual exciters should maintain a constant phase difference (usually 0 or 180 degrees) to generate a stable resultant force. Instantaneous angular velocity deviation refers to the instantaneous difference between the actual angular velocity of the dual exciter at a certain moment and its ideal or average angular velocity; it reflects the non-steady-state characteristics of the exciter's operation.

[0018] Step S20: Based on the dynamic phase error, calculate the parasitic lateral disturbance force induced by the unsteady synchronization of the dual exciters; the parasitic lateral disturbance force refers to the component perpendicular to the conveying direction generated when the dual exciters are out of phase. This component is the parasitic lateral disturbance force, which causes the material to move laterally within the chute.

[0019] Step S25: Introduce an empirical forgetting factor reflecting the characteristics of disturbance dissipation and accumulation release, and perform a discrete quadratic integral operation with state decay on the parasitic lateral disturbance force to obtain the lateral sliding accumulation displacement of the material. The empirical forgetting factor is a decay coefficient used in dynamic system modeling to characterize the degree of influence of historical states on the current state. Introducing this factor can simulate the characteristics of the gradual dissipation or release of the lateral accumulation state of the material over time after being disturbed. Discrete quadratic integral operation refers to performing two integration operations on a physical quantity in a discrete time series. In the scenario of lateral material movement, it is usually used to convert force into velocity, and then velocity into displacement to simulate the dynamic response of the material. Lateral sliding accumulation displacement refers to the cumulative displacement of the material perpendicular to the conveying direction during the chute conveying process due to the parasitic lateral disturbance force. This displacement directly reflects the degree of material accumulation on the sidewall of the chute.

[0020] Step S30: Obtain the dimensionless cumulative shrinkage of the chute opening along the conveying direction within a limited time window, and perform a nonlinear multiplication coupling operation with the lateral sliding accumulation displacement to calculate the dynamic jamming risk margin. The dimensionless cumulative shrinkage refers to the cumulative decrease in the geometric opening of the chute along the conveying direction within a specific time window. After dimensionless processing, it is used to characterize the limitation of the chute structure on the material throughput capacity. The nonlinear multiplication coupling operation refers to the correlation calculation of two or more physical quantities through a nonlinear multiplication relationship. This coupling method can more accurately reflect the complex interactions between different factors, such as the combined impact of material accumulation and chute shrinkage on jamming risk. The dynamic jamming risk margin refers to the system's resistance or safety margin to material jamming accidents during the plum kernel and meat separation chute conveying process. This margin changes in real time, reflecting the probability of jamming occurring under the current operating conditions.

[0021] Step S40: Extract the unidirectional risk deterioration gain based on the dynamic jamming risk margin, and introduce it as an adaptive weight into the feedforward control equation to obtain the instantaneous frequency compensation command, thereby performing feedforward control on the lag-side exciter. The unidirectional risk deterioration gain is a positive gain value extracted from the rate of change of the dynamic jamming risk margin when it shows a deteriorating trend. This gain is used to quantify the degree of risk deterioration and serves as the input to the control system. Adaptive weights are parameters dynamically adjusted in the control system based on changes in system state or environment. Introducing adaptive weights makes the control strategy more flexible to cope with complex and changing operating conditions. The feedforward control equation is a control algorithm that generates control commands in advance based on predictions or measurements of system disturbances to counteract the impact of disturbances on the system. The instantaneous frequency compensation command is a frequency adjustment command generated in real time by the feedforward control system based on the calculated risk margin and sent to the exciter, aiming to eliminate or reduce the phase error of the exciter. The lag-side exciter is an exciter in a dual-exciter system whose rotational phase lags behind that of the other exciter. Typically, feedforward control commands are applied to the lagging exciter to bring its phase up and restore synchronization.

[0022] This embodiment provides a method for separating dried plum kernels and flesh using a chute for conveying. Its main feature is that it achieves real-time assessment and active suppression of the risk of blockage during the conveying process through a series of steps.

[0023] First, in step S10, the real-time rotational state of the dual exciters needs to be acquired and the dynamic phase error extracted. This is then differentially processed and subjected to a first-order low-pass filter to obtain the instantaneous angular velocity deviation. Several methods can be used to acquire the real-time rotational state of the dual exciters. For example, an optical encoder can be installed on the rotational shaft of each exciter, and its rotational angle and speed can be monitored in real time using the pulse signal output by the encoder. Another method is to use a Hall sensor or proximity sensor to detect marker points on the exciter's eccentric block, and to calculate the rotational state by the time interval between the marker points passing through the sensor. These methods can all provide the raw rotational data of the exciters, and then the dynamic phase error can be extracted by comparing the rotational angles or time series of the two exciters. After obtaining the dynamic phase error, it is differentially processed, for example, by calculating the difference in phase error between adjacent sampling times to approximate its rate of change. Subsequently, a first-order low-pass filter can be used to process the differential signal, for example, by using a simple moving average algorithm or an exponentially weighted average algorithm to filter out high-frequency noise, thereby obtaining a smoother and more accurate instantaneous angular velocity deviation.

[0024] Secondly, in step S20, the parasitic lateral disturbance force induced by the unsteady synchronization of the dual exciters is calculated based on the aforementioned dynamic phase error. This step aims to quantify the impact of phase asynchrony on the lateral movement of the material. One approach is to directly map different magnitudes of dynamic phase errors to corresponding parasitic lateral disturbance forces using pre-established experimental data or simulation models. For example, a lookup table can be constructed, storing the lateral force values ​​measured under different phase errors. When the system detects a certain dynamic phase error, it retrieves the corresponding lateral disturbance force from the lookup table. Another approach is to use simplified linear or nonlinear functions to approximate the relationship between phase error and lateral disturbance force. For example, when the phase error is small, the lateral force and phase error have an approximately linear relationship; when the phase error is large, the lateral force may exhibit a saturation trend.

[0025] Further, in step S25, an empirical forgetting factor reflecting the characteristics of disturbance dissipation and accumulation release is introduced. A discrete quadratic integral operation with state decay is performed on the parasitic lateral disturbance force to obtain the lateral sliding accumulation displacement of the material. To simulate the dynamic response of the material, the calculated parasitic lateral disturbance force can first be discretely integrated to obtain the lateral sliding velocity of the material. For example, the Euler method can be used, multiplying the force at the current moment by the sampling time interval and accumulating it to the velocity at the previous moment. Subsequently, the lateral sliding velocity is discretely integrated again to obtain the lateral sliding accumulation displacement of the material. In this process, an empirical forgetting factor can be introduced. For example, during each integration accumulation, the velocity or displacement at the previous moment is multiplied by a fixed decay coefficient less than 1 to simulate the energy dissipation and gradual release of accumulation caused by factors such as friction and collision within the chute.

[0026] Next, in step S30, the dimensionless cumulative shrinkage of the chute opening along the conveying direction within a defined time window is obtained. This shrinkage is then coupled nonlinearly with the lateral sliding accumulation displacement to calculate the dynamic jamming risk margin. To obtain the dimensionless cumulative shrinkage of the chute opening, the chute geometry can be measured in advance, and the total reduction in chute opening within that time window can be calculated based on the material conveying speed and the defined time window. For example, it can be assumed that the chute shrinkage is linear, estimated by measuring the openings at the inlet and outlet ends and combining this with the conveying distance. Subsequently, this shrinkage is coupled nonlinearly with the lateral sliding accumulation displacement. For example, the two can be simply multiplied, or an empirical coefficient can be introduced to adjust their relative weights to reflect the combined effect of material accumulation and chute shrinkage on jamming risk.

[0027] Finally, in step S40, a unidirectional risk deterioration gain is extracted based on the aforementioned dynamic jamming risk margin. This gain is then used as an adaptive weight in the feedforward control equation to obtain an instantaneous frequency compensation command, and feedforward control is applied to the lag-side exciter. To extract the unidirectional risk deterioration gain, the changing trend of the dynamic jamming risk margin at consecutive sampling times can be monitored. For example, if the risk margin at the current time is lower than that at the previous time, it indicates that the risk is deteriorating, and a positive gain value can be set. If the risk margin remains unchanged or improves, the gain is zero or negative. Subsequently, this gain is used as an adaptive weight and introduced into the feedforward control equation. For example, a simple proportional control (P control) structure can be used, multiplying the deviation between the risk margin and the target safety margin by a proportional gain, and further multiplying by the adaptive weight to generate an instantaneous frequency compensation command. This command is then sent to the driver of the lag-side exciter to adjust its operating frequency so that its phase catches up, thereby actively suppressing phase error.

[0028] The following example will provide a more detailed explanation of the above technical solution: Suppose that on a plum pit and flesh separation production line, the chute conveyor system is operating stably. However, due to load fluctuations caused by changes in material batches, the dual vibrators begin to exhibit slight phase asynchrony.

[0029] At this point, the system initiates step S10. Sensors (such as photoelectric encoders) mounted on the dual exciters monitor their rotation angles and speeds in real time and transmit this data to the processor. The processor calculates the dynamic phase error by comparing the real-time angles of the two exciters. For example, if the phase of exciter A leads that of exciter B by 5 degrees, the dynamic phase error is 5 degrees. Subsequently, the processor performs differential processing on this dynamic phase error, calculates its rate of change, and applies a first-order low-pass filter to filter out measurement noise, thereby obtaining a smooth instantaneous angular velocity deviation. This deviation value accurately reflects the degree of deviation in the synchronization state of the dual exciters.

[0030] Next, in step S20, the processor calculates the parasitic lateral disturbance force induced by the phase asynchrony based on the dynamic phase error corresponding to this instantaneous angular velocity deviation, using a preset mapping relationship (such as a lookup table or empirical formula stored in the system). For example, if the phase error is 5 degrees, the system may calculate that there is a lateral disturbance force of 10 Newtons to the left of the slide.

[0031] The system then proceeds to step S25. The processor takes the 10-Newton parasitic lateral disturbance force as input and introduces two empirical forgetting factors. The first forgetting factor simulates the decay characteristics of the material's lateral sliding velocity after being subjected to a lateral force; the second forgetting factor simulates the release characteristics of the material's lateral accumulation displacement. The processor performs discrete quadratic integration on the lateral disturbance force. Specifically, it first converts the force into lateral sliding velocity, and then converts the velocity into lateral sliding accumulation displacement. During each integration accumulation, the effect of the forgetting factors is considered, allowing the historical accumulation effect to gradually decay, thus more realistically reflecting the actual accumulation of material on the chute sidewall. For example, after a period of accumulation, the system may calculate that the material has generated a 2-millimeter lateral sliding accumulation displacement on the left side of the chute.

[0032] Simultaneously, in step S30, the system acquires the dimensionless cumulative shrinkage of the chute's opening along the conveying direction within a defined time window. For example, in the next 5 seconds, the geometric opening of the chute will cumulatively shrink by 0.5 mm. The processor performs a nonlinear multiplication coupling operation on this 0.5 mm cumulative shrinkage and the previously calculated 2 mm lateral sliding accumulation displacement. This coupling operation comprehensively considers the actual situation of material accumulation and the limitation of chute geometric shrinkage, thereby calculating a dynamic jamming risk margin. For example, if the calculated risk margin is 0.3, it indicates that the current system has a certain jamming risk, and the margin is low.

[0033] Finally, in step S40, the processor extracts a unidirectional risk deterioration gain based on this dynamic jamming risk margin. For example, if the current risk margin of 0.3 is lower than the previous time step of 0.4, indicating that the risk is deteriorating, the system extracts a positive deterioration gain. This deterioration gain is then introduced into the feedforward control equation as an adaptive weight. The feedforward control equation calculates an instantaneous frequency compensation command based on the deviation between the current risk margin and the target safety risk margin, combined with this adaptive weight. For example, the command might be to increase the frequency of the lagging exciter by 0.5 Hz. This command is immediately sent to the driver of the lagging exciter, increasing its frequency and accelerating its rotation, allowing its phase to catch up, reducing dynamic phase error, thereby reducing parasitic lateral disturbance force, and ultimately suppressing the lateral slippage and accumulation of material, thus preventing jamming accidents. Through this series of closely coordinated steps, the system achieves real-time prediction and proactive intervention of potential jamming risks.

[0034] This application further proposes that in step S10, the process of obtaining the instantaneous angular velocity deviation includes: calculating the rate of change of the dynamic phase error at the current moment relative to the previous sampling moment, and using it as the current error increment; using a first-order low-pass filtering algorithm, weighting and fusing the current error increment with the historical instantaneous angular velocity deviation at the previous sampling moment according to a preset filtering weight to obtain the instantaneous angular velocity deviation at the current moment, wherein the filtering weight is used to balance the dynamic response of the sensor and high-frequency noise suppression; its calculation formula is: in, This represents the instantaneous angular velocity deviation at the current time t; This represents the dynamic phase error at the current time t; This represents the dynamic phase error at the previous sampling time. Indicates the discrete sampling interval; Denotes the first-order low-pass filter coefficients and ; This represents the instantaneous angular velocity deviation at the previous sampling moment.

[0035] Specifically, discrete sampling interval In a digital control system, phase error refers to the time interval for periodic measurement and data acquisition of continuously changing physical quantities (such as phase error). It is the foundation for converting continuous signals into discrete digital signals, determining the system's response speed to dynamic changes and the granularity of data processing. This interval can be determined by the system clock frequency and sampling control logic; for example, by triggering an analog-to-digital converter (ADC) for data acquisition via a timer interrupt, or by reading sensor data at a fixed frequency through software loops. In some cases, different sensors or actuators may have different sampling periods, but a synchronization mechanism can ensure that data is processed at a uniform discrete point in time. (Synchronization phase error) Perform differential operations, where the synchronization phase error This refers to the instantaneous difference between the actual rotation angles of the two exciters during operation. Differential calculation refers to calculating the change of this error between adjacent sampling times. Through differential calculation, the trend and rate of change of the phase error can be preliminarily reflected, providing basic data for evaluating the relative motion state of the exciters. One implementation method is to directly input the phase error at the current moment. Subtract the phase error from the previous sampling time. Another approach, when considering more complex system dynamics, is to use higher-order differencing or to fit the phase error data over a period of time using the least squares method and then take the derivative to obtain a smoother differencing result. This involves fusing the historical state from the previous moment. Obtaining instantaneous angular velocity deviation This refers to the fact that when calculating the instantaneous angular velocity deviation at the current moment, not only is the phase error change at the current moment considered, but the instantaneous angular velocity deviation calculated at the previous sampling moment is also taken into account. This fusion mechanism is key to implementing first-order low-pass filtering. It makes the current calculation result influenced by historical data, thereby smoothing instantaneous fluctuations and improving the stability of the calculation. One implementation method is to use a weighted average, as shown in the formula, combining the current difference result with the instantaneous angular velocity deviation at the previous moment according to a certain weight. Another method is to use other forms of recursive filtering algorithms such as exponentially weighted moving average (EWMA), and control the degree of influence of historical data by adjusting the attenuation factor. First-order low-pass filter coefficients This is a weighting factor between 0 and 1, used to control the degree of influence of the current input on the output and the smoothing effect of historical data on the output. It determines the filter's response speed and noise suppression capability. A larger value... A higher value means the filter responds more quickly to the current input, but the smoothing effect is weaker; a smaller value... A value of [value missing] indicates a stronger smoothing effect but a slower response time. This coefficient can be optimized through experimental testing and system identification methods, and adjusted according to the requirements of response speed and noise immunity in actual application scenarios. For example, in scenarios with high real-time requirements, it can be appropriately increased. Value; however, in scenarios with severe noise interference and higher requirements for smoothness, it can be reduced. value.

[0036] This scheme acquires the real-time rotational state of the dual exciters and extracts the dynamic phase error. Then, it performs differential calculations on this dynamic phase error at discrete sampling intervals to initially capture the trend of phase change. To overcome the interference of high-frequency noise during digital sampling on the calculation of instantaneous angular velocity deviation, this scheme cleverly introduces a first-order low-pass filtering mechanism. This mechanism weights and fuses the phase error differential result at the current moment with the instantaneous angular velocity deviation calculated at the previous sampling moment, utilizing the first-order low-pass filter coefficients. The two values ​​are weighted and averaged to generate the instantaneous angular velocity deviation at the current moment. This recursive calculation method ensures that the output of the instantaneous angular velocity deviation no longer depends solely on the sampling data at a single moment, but exhibits a continuous and smooth evolution characteristic, effectively suppressing drastic fluctuations caused by instantaneous noise. This is achieved by adjusting the filter coefficients. It can flexibly balance the system's sensitivity to phase change response and anti-interference capability, ensuring that the calculation results of instantaneous angular velocity deviation are stable and accurate, providing reliable input data for subsequent accurate evaluation of the unsteady synchronization state of the dual exciter, thereby improving the robustness of the entire control system from the source.

[0037] As one specific implementation, this method can be implemented in a control system based on a digital signal processor (DSP) or a high-performance microcontroller (such as an ARM Cortex-M series processor). The real-time rotational state of the dual exciters can be acquired by a rotary encoder mounted on each exciter shaft, and the pulse signals output by these encoders are fed into the timer / counter module of the microcontroller. Dynamic phase error The phase error can be calculated by comparing the angular positions of the two exciters at each sampling time. For example, if the two encoders output pos1(t) and pos2(t) respectively, the phase error can be expressed as pos1(t) - pos2(t). Discrete sampling interval This can be set via a system timer interrupt, for example, to 1 millisecond. In each timer interrupt service routine, the current dynamic phase error is read. The dynamic phase error at the previous moment. Deviation from the instantaneous angular velocity at the previous moment These values ​​are then read from memory; they were calculated and stored in the previous interrupt cycle. Subsequently, based on the aforementioned formula, floating-point arithmetic is used to calculate the current instantaneous angular velocity deviation. Calculation of first-order low-pass filter coefficients. This can be pre-configured in the system firmware, for example, set to 0.2, which typically provides a good balance between noise suppression and response in typical mechanical vibration systems. After the calculation is complete, the newly obtained... and This value will be stored as the "previous moment" value for the next sampling period. This iterative process ensures a continuous and smooth estimation of the instantaneous angular velocity deviation.

[0038] Through the above technical solution, this application effectively solves the problem that direct differential calculation of phase error in discrete digital control systems is easily affected by high-frequency noise, leading to inaccurate calculation of instantaneous angular velocity deviation. By introducing a first-order low-pass filter mechanism and incorporating the historical state from the previous moment, this solution can smooth the differential result of dynamic phase error, significantly suppressing instantaneous noise generated during digital sampling. This allows the calculated instantaneous angular velocity deviation to more stably and accurately reflect the true synchronization state of the dual exciters, avoiding misjudgments caused by data fluctuations. Therefore, it provides reliable basic data for subsequent accurate calculation of parasitic lateral disturbance force, fundamentally improving the robustness and control accuracy of the entire plum kernel and meat separation chute conveying method, and ensuring the stability and reliability of the system under dynamic operating conditions.

[0039] This application further proposes that in step S20, the parasitic lateral disturbance force is determined by the vector synthesis of the centrifugal force of the dual exciter eccentric blocks when dynamic phase error exists; the amplitude of the parasitic lateral disturbance force is positively correlated with the physical properties of the exciter eccentric blocks and the positive function of the reference operating frequency, and exhibits a nonlinear trigonometric function mapping relationship with the change of dynamic phase error; its calculation formula is:

[0040] in, This represents the parasitic lateral disturbance force generated at the current time t; This indicates the mass of the eccentric block of the vibrator; This indicates the centroid eccentricity of the eccentric block; This represents the reference circular frequency under ideal steady-state conditions of the dual exciter. This represents the dynamic phase error at the current time t.

[0041] Specifically, dynamic phase error This refers to the instantaneous difference in rotation angle between two vibrators during operation. This error is not constant but changes dynamically over time, reflecting the real-time deviation of the vibrator synchronization state. It can be obtained by using angle sensors (e.g., rotary encoders, Hall effect sensors) mounted on the vibrators to monitor the rotational position of each vibrator in real time, and then calculating the difference between the position signals of the two vibrators; alternatively, it can be indirectly calculated by analyzing the phase information of the vibrator drive current or voltage. Parasitic lateral disturbance force. This refers to the unexpected lateral force acting on the chute conveyor system caused by the phase asynchrony of the two vibrators. This force can cause lateral movement or accumulation of material within the chute, and is one of the main causes of jamming risk. This disturbance force can be understood as the resultant force of the centrifugal force in the lateral component when the vibrator eccentric block rotates asynchronously. The "mapping" in "mapped to peak envelope" means converting one physical quantity (dynamic phase error) into another physical quantity (parasitic lateral disturbance force) through a specific mathematical relationship. "Peak envelope" emphasizes that this mapping relationship captures the maximum possible value of the disturbance force or the boundary of its changing trend, rather than every detail of instantaneous fluctuations. This means that through this mapping, the maximum intensity of the lateral disturbance force that can be achieved under a given phase error can be obtained, thus providing a conservative and effective upper limit for risk assessment. In the calculation formula, m (the mass of the vibrator eccentric block) refers to the mass of the eccentric block used to generate vibration in the vibrator, which is an inherent physical quantity determined during the vibrator design. r (eccentricity of the eccentric block's center of mass) refers to the distance from the center of mass of the eccentric block to the rotation axis of the vibrator, which is also an inherent physical quantity in the design of the vibrator. (Reference circular frequency under ideal steady state of dual exciter) refers to the rated or design circular frequency of dual exciter under ideal synchronous and stable operating conditions, representing the vibration frequency reference when the system is working normally. This term is the core of the formula; it measures the dynamic phase error. The equation is introduced in the form of a sine function, with half of its angle taken. This nonlinear processing method can accurately reflect the influence of phase error on the peak value of lateral disturbance force. When the phase error is small, the lateral disturbance force is also small; when the phase error increases, the lateral disturbance force increases nonlinearly until it reaches its maximum value. This sine function relationship is derived based on the synchronization theory of dual exciters and can accurately capture the physical relationship between phase error and the lateral component of the resultant force.

[0042] This application's solution establishes a mapping relationship between dynamic phase error and parasitic lateral disturbance force, thereby achieving a quantitative characterization of the unsteady-state synchronous-induced disturbance of the excitation system. In the conveying process of the plum pit and meat separation chute, the real-time deviation of the synchronization state of the dual exciters is a key factor leading to lateral material deviation and jamming. Traditional methods struggle to accurately capture the impact of this nonlinear fluctuation on material movement. This solution addresses this by using the real-time acquired dynamic phase error... As input, the physical parameters of the exciter itself are combined, namely the mass m of the eccentric block, the eccentricity r of the eccentric block's center of mass, and the reference circular frequency under ideal steady-state conditions of the dual exciter. Through a specific nonlinear functional relationship, Calculate the parasitic lateral disturbance force generated at the current time t. This calculation method cleverly utilizes the characteristics of the sine function to effectively capture the nonlinear influence of phase error on lateral disturbance force at different levels, especially when the phase error is small or large, it can more accurately reflect the actual mechanical effect. In this way, the abstract problem of phase asynchrony is transformed into specific, quantifiable mechanical parameters, providing an accurate input basis for subsequent assessment of the lateral sliding and stacking displacement of materials. This allows the system to dynamically adjust the control strategy based on the real-time changing phase error, thereby effectively improving the ability to predict jamming risks. Compared to using only phase error as a linear input, this scheme, by introducing physical parameters and nonlinear mapping, can more accurately characterize the parasitic lateral disturbance force induced by the unsteady synchronization of the exciter, thus providing a more reliable basis for subsequent jamming risk assessment and feedforward control, significantly improving the accuracy and robustness of the entire conveying method.

[0043] As a specific implementation method, in the plum pit and meat separation chute conveying method, when it is necessary to calculate the parasitic lateral disturbance force induced by the unsteady synchronization of the dual exciters, the real-time rotation state of the dual exciters is first obtained through step S10 and the dynamic phase error is extracted. For example, a high-precision rotary encoder mounted on each exciter shaft can be used to acquire its angular position signal in real time, and the dynamic phase error can be obtained by calculating the difference between the two encoder signals. This dynamic phase error is then... The data is input into a preset calculation module. This module stores the mass m of the exciter eccentric block, the eccentricity r of the eccentric block's center of mass, and the reference circular frequency under ideal steady-state conditions of the dual exciter. Parameters such as these. For example, for a specific type of exciter, the mass m of its eccentric block can be 0.5 kg, the eccentricity r of the center of mass can be 0.02 m, and the reference circular frequency... It can be 100 radians per second. The calculation module uses the formula... Perform calculations. For example, if at a certain time t, the dynamic phase error is measured... If the value is 0.1 radians, then the calculation module will set m=0.5, r=0.02. , Substituting into the formula, the parasitic lateral disturbance force at the current moment can be calculated. This calculation result This can then be used as input for calculating the lateral sliding and stacking displacement of the material in subsequent step S25.

[0044] Through the above technical solution, this application overcomes the shortcomings of traditional methods in quantifying the driving effect of nonlinear phase error on the lateral deviation of materials. By nonlinearly mapping the dynamic phase error into the peak envelope of the parasitic lateral disturbance force, and combining it with the physical parameters of the exciter, the precise quantification of the disturbance force induced by the unsteady synchronization of the exciter is achieved. This precise mechanical characterization enables the system to more realistically reflect the dynamic characteristics of the dual exciters in the asynchronous state, thus providing a more accurate input for subsequent calculation of the lateral slippage and accumulation displacement of materials, and significantly improving the accuracy and reliability of jamming risk prediction.

[0045] This application further proposes that in step S25, the discrete quadratic integral operation adopts a state recursive model with a double forgetting factor, specifically including: converting the parasitic lateral disturbance force at the current moment into an instantaneous lateral acceleration based on the equivalent total mass, and superimposing it with the lateral slip velocity at the previous moment after attenuation processing by the first forgetting factor to obtain the lateral slip velocity at the current moment, where the first forgetting factor is used to characterize the dissipation characteristics of the lateral disturbance; converting the integral of the lateral slip velocity at the current moment into an instantaneous lateral displacement increment, and superimposing it with the lateral slip accumulation displacement at the previous moment after attenuation processing by the second forgetting factor to obtain the lateral slip accumulation displacement at the current moment, where the second forgetting factor is used to characterize the system's release and reset characteristics for material accumulation; the specific calculation formula is as follows:

[0046]

[0047] in, and Let represent the lateral sliding velocities at the current time t and the previous time, respectively. and These represent the lateral slip stacking displacement at the current time t and the previous time, respectively. Indicates parasitic lateral disturbance force. Indicates the equivalent total mass. Indicates the discrete sampling interval; and These are the experiential forgetting factors, among which Characterizes the dissipation properties of the disturbance. Characterizes the accumulation and release properties, and and The values ​​are all between 0 and 1.

[0048] Specifically, the empirical forgetting factor is a parameter used to weight the influence of historical data on the current state. It allows old information to gradually decay during the calculation process, thus better reflecting the dynamic changes of the system. For example, the empirical forgetting factor can be determined through experimental calibration or numerical simulation based on the physical properties of the material (such as the coefficient of friction and viscosity), or it can be dynamically adjusted based on real-time feedback from the system operation using an adaptive algorithm to optimize its value. The accumulated lateral slip-accumulation displacement refers to the lateral offset gradually accumulated by the material within the chute due to lateral disturbance forces. This displacement is not an instantaneous measurement but rather considers the cumulative effect of factors such as material inertia and friction over time. It can be achieved by integrating a mathematical model of the material's lateral movement within the chute, or by installing high-precision displacement sensors or a visual recognition system on the chute sidewall to monitor the lateral distribution of the material in real time and perform cumulative calculations. The lateral slip velocity is the instantaneous speed of the material in the lateral direction of the chute. This velocity can be obtained based on the differential calculation of the lateral slip-accumulation displacement, or it can be directly measured by a non-contact velocity sensor (such as a laser Doppler velocimeter) installed on the chute sidewall. The equivalent total mass refers to the effective mass of the material and the chute components involved in the lateral movement, which needs to be considered when calculating the lateral motion of the material. This mass can be estimated through dynamic analysis of the average filling amount of material in the chute and the chute structure, or it can be accurately determined through vibration testing and mass calibration during the system design phase. The specific calculation formula can be: ;in, The inherent static total mass of the chute body. The average bulk density of the dried plum material was measured in advance. The material filling volume is calculated in real time by a material level height sensor installed at the chute feed inlet. In this embodiment, the dynamic effective mass coefficient for bulk materials participating in lateral sliding motion is defined as follows: The value range is limited to 0.65 to 0.80. During control system initialization, the static mass... With density As a constant, it is pre-stored in memory, and the processor only needs to update it according to the material level during operation. This can achieve an equivalent total mass. Accurate, real-time dynamic acquisition is achieved. The discrete sampling interval is the time step for the system to acquire and calculate data. The setting of this interval needs to take into account the real-time requirements of the control system, the data refresh rate of the sensor, and the dynamic response speed of the material movement. For example, it can be set between 10 milliseconds and 100 milliseconds to ensure that key dynamic changes can be captured.

[0049] This application's solution introduces an empirical forgetting factor to construct a cumulative displacement calculation model that reflects the dynamic evolution of materials, thus achieving a leap from instantaneous mechanical analysis to time-series dynamic behavior analysis. Specifically, the solution first utilizes the lateral slip velocity from the previous moment. Parasitic lateral disturbance force induced by the unsteady synchronization of the dual exciter at the current moment Weighted fusion is performed, in which the experiential forgetting factor is included. The disturbance dissipation characteristics were characterized, effectively simulating the energy dissipation and frictional resistance of materials under vibration, thereby calculating the lateral slip velocity at the current moment. This approach avoids drastic fluctuations in calculation results caused by instantaneous disturbance force fluctuations, resulting in smoother velocity calculations that conform to physical laws. Building upon this, the scheme further incorporates the lateral slip-stacking displacement at the current moment. Displacement state compared to the previous moment Weighted accumulation is performed, including the experiential forgetting factor. This model characterizes the accumulation and release properties, accurately depicting the accumulation effect of materials within the chute over time. By preserving and attenuating historical states, the model dynamically captures the complete process from initial material displacement to the formation of accumulation, thus providing accurate displacement data support for subsequent jamming risk assessment. The core contribution of this scheme lies in transforming discrete mechanical disturbances into a continuous displacement evolution process. Through a parameterized forgetting factor, it achieves accurate modeling of material behavior under complex conveying environments, significantly improving the system's ability to perceive risks induced by unsteady synchronicity. In conjunction with steps S10 and S20, this scheme can accurately calculate the parasitic lateral disturbance force based on real-time acquired dynamic phase errors. Furthermore, it can accurately quantify the cumulative lateral sliding and accumulation displacement of the material, providing a more refined and dynamic data foundation for subsequent jamming risk assessment.

[0050] As a specific implementation method, an industrial-grade programmable logic controller (PLC) or an embedded system can be used as the computing platform. This platform uses a preset discrete sampling interval (e.g., The system performs computational tasks periodically. Within each sampling period, the system first obtains the parasitic lateral disturbance force at the current moment from step S20 above. Then, using the previous moment's lateral sliding velocity stored in system memory... and preset experience forgetting factor (For example, ), combined with equivalent total mass (For example, ), calculate the lateral slip velocity at the current moment. Next, the system uses the newly calculated... The lateral sliding stacking displacement stored at the previous moment and preset experience forgetting factor (For example, ), calculate the lateral slip stacking displacement at the current moment. Among them, the experiential forgetting factor and The values ​​can be calibrated based on factors such as the actual frictional characteristics, adhesion, and material of the chute, through preliminary experimental testing and data analysis. For example, for dried plums, It may be set lower to reflect less dissipation, while for moist dried plums, The value may be set too high. Equivalent total mass It can be estimated based on the geometry of the chute, the average bulk density of the material, and the vibration characteristics of the vibrator.

[0051] Through the above technical solution, this application overcomes the limitations of traditional methods that rely solely on instantaneous mechanical responses, achieving dynamic and cumulative modeling of the lateral sliding and accumulation process of materials. This modeling method can more accurately reflect the energy dissipation and accumulation release characteristics of materials during actual conveying, thereby significantly improving the prediction accuracy of lateral material displacement and potential jamming risks. By providing more refined and dynamic lateral sliding and accumulation displacement data, this solution provides a solid foundation for subsequent jamming risk margin calculations (such as step S30 above), enabling the entire anti-jamming control system to detect risks earlier and more accurately, thereby achieving more proactive and effective preventive control, effectively reducing the probability of jamming accidents, and improving the stability and production efficiency of the plum pit and meat separation chute conveying process.

[0052] This application further proposes that the nonlinear multiplicative coupling operation logic for calculating and obtaining the dynamic jamming risk margin in step S30 includes: constructing a first basic evaluation term characterizing the degree of longitudinal spatial contraction based on the geometric gradient of the chute opening descent and the instantaneous longitudinal conveying speed within a limited time window; constructing a lateral penalty term characterizing the degree of lateral stacking blockage based on the lateral sliding stacking displacement and the sidewall geometric projection characteristics; multiplying and coupling the first basic evaluation term and the lateral penalty term, and combining them with the dynamic disturbance term obtained after normalization of the parasitic lateral disturbance force, subtracting the initial nominal safety margin of the system, and comprehensively evaluating to obtain the dynamic jamming risk margin; specifically, determining the instantaneous longitudinal conveying speed of the chute based on the real-time driving frequency of the dual exciter and the chute inclination angle. Furthermore, by coupling the dimensionless cumulative opening shrinkage with the lateral slip stacking displacement through a sliding window mechanism, the dynamic jamming risk margin is calculated using the following formula:

[0053] in, This represents the dynamic jamming risk margin at the current time t; This represents the geometric gradient of the chute opening decreasing along the conveying direction; Indicates a time window; This represents the effective longitudinal displacement within the sliding window; Indicates the instantaneous longitudinal conveying speed; Indicates the initial opening of the feed end gap; Indicates the geometric projection coefficient of the sidewall; Indicates lateral sliding accumulation displacement; Indicates parasitic lateral disturbance force; Indicates the equivalent total mass; Indicates the reference angular frequency; This represents the discrete sampling interval.

[0054] Specifically, the real-time drive frequency of the dual exciter refers to the instantaneous output frequency of the motor or frequency converter driving the dual exciter. This frequency can be measured in real-time using an encoder or frequency sensor mounted on the motor shaft, or directly obtained from the control output signal of the frequency converter. The chute inclination angle refers to the angle of inclination of the conveying chute relative to the horizontal plane. This angle can be preset and fixed, or it can be dynamically adjusted using a hydraulic or electric actuator and monitored in real-time using an inclination sensor or encoder. Instantaneous longitudinal conveying speed. This refers to the instantaneous velocity of the material moving along the conveying direction within the chute. It can be calculated using a pre-established kinematic model or a lookup table method, based on the real-time driving frequency of the dual vibrators, the chute inclination angle, the chute's vibration characteristics, and the material's friction coefficient. Alternatively, it can be calculated in real-time by tracking and calculating the material's trajectory using a visual recognition system. Specifically, the pre-established kinematic model employs a semi-empirical analytical equation based on the throwing state of bulk materials, and its mathematical expression is: In the formula, This is a correction factor for the efficiency of material sliding conveying, with a value ranging from 0.85 to 0.95. This is the nominal amplitude reference value generated longitudinally by the dual exciter; This is the real-time drive frequency output from the frequency converter to the dual exciter; The current tilt angle of the chute is obtained by the MEMS tilt sensor; The structural vibration direction angle is fixed for the chute. In another optional implementation, to reduce the real-time computing power overhead of the processor, the kinematic model can also be implemented using a two-dimensional mesh lookup table method generated by offline calibration: the control system has a pre-stored velocity mapping table, which uses the real-time drive frequency of the frequency converter. The horizontal axis is the inclination angle of the groove (1Hz step). With the vertical axis as the reference (step size 0.5°), the output value corresponding to the grid intersection point is the empirical value of the longitudinal conveying speed calibrated in advance using a laser Doppler velocimeter. The processor can realize the instantaneous longitudinal conveying speed using bilinear interpolation. High-frequency retrieval. The sliding window mechanism is a data processing method that defines a fixed-length time window. Within this window, continuously collected data is processed. As time progresses, the window slides forward, constantly incorporating new data and discarding older data, thus achieving dynamic accumulation and analysis of historical data. The dimensionless cumulative opening shrinkage is an indicator reflecting the cumulative change in the chute's geometric constraints with longitudinal conveying. It is expressed as the geometric gradient of the chute's opening decreases along the conveying direction. The product of the effective longitudinal displacement within the sliding window and divided by the initial opening of the feed end gap. Dimensionless transformation is used to quantify the degree of longitudinal spatial contraction encountered by materials during transportation. Lateral sliding and stacking displacement. It is the parasitic lateral disturbance force on the material in the chute The resulting lateral offset and accumulation, calculated in the preceding steps, reflects the unintended lateral aggregation of material. The nonlinear multiplicative coupling operation combines the dimensionless cumulative opening shrinkage with the lateral slip accumulation displacement in a multiplicative form, and then applies this value through the sidewall geometric projection coefficient. The influence of lateral displacement is weighted to reflect the nonlinear synergistic effect of longitudinal geometric constraints and lateral packing on jamming risk. Dynamic jamming risk margin. It is a real-time changing quantitative indicator used to assess the risk of material jamming in the chute at the current moment; the smaller the value, the higher the risk of jamming. The geometric gradient of the chute opening decreases along the conveying direction. These are the design parameters for the chute structure, representing the shrinkage rate of the chute width along the conveying direction. They can be obtained from design drawings or actual measurements. (Limited time window) This refers to the time length used to accumulate longitudinal displacement in the sliding window mechanism; its setting should comprehensively consider the system response speed and the influence range of historical data. Initial opening of the feed end gap. This is the initial width at the chute inlet, a baseline parameter for the chute's geometry. Sidewall geometric projection coefficient. This is an empirically mapped weight based on the cross-sectional profile characteristics of the chute, used to adjust the nonlinear contribution of lateral sliding and accumulation displacement to the final jamming risk. This coefficient is strongly correlated with the inclination angle of the chute sidewall and its surface friction characteristics: when the chute sidewall inclination angle is steep (e.g., a V-shaped chute) or the surface friction coefficient is large, lateral sliding of material easily transforms into upward climbing and accumulation. A larger value needs to be set to amplify the penalty effect; conversely, if the bottom of the chute is relatively flat (such as a U-shaped wide-bottom chute), then... The corresponding reduction. In a conventional embodiment of the present invention, The dimensionless value range is recommended to be limited to between 0.5 and 2.0, for example, for nominal dried plum materials and standard inclined chutes. It can be preset to 1.2. Parasitic lateral interference force Equivalent total mass Reference Circular Frequency and discrete sampling interval These are all parameters that have been determined in the aforementioned steps. They serve as inputs in the calculation of dynamic constrained risk margin and collectively affect the accuracy of risk assessment.

[0055] This application achieves a quantitative assessment of the risk of jamming during the conveying process by constructing a dynamic jamming risk margin model. First, the instantaneous longitudinal conveying speed is determined based on the real-time drive frequency of the dual vibrators and the chute inclination angle. This feature ensures that risk assessment can respond in real time to changes in excitation parameters, providing fundamental dynamic data for subsequent calculations. Secondly, a sliding window mechanism is introduced to accumulate the dimensionless cumulative shrinkage of the opening. This feature effectively captures the changing trend of the chute geometry within a specific time period, avoiding misjudgments caused by instantaneous fluctuations. This cumulative shrinkage is then compared with the lateral sliding accumulation displacement. By employing nonlinear multiplicative coupling, this application successfully correlates the geometric constraints of longitudinal conveying with the physical offset of lateral stacking, reflecting the comprehensive stress and stacking state of materials within the confined space of the chute. Finally, by introducing the sidewall geometric projection coefficient... The lateral displacement is corrected, and the parasitic lateral disturbance force is taken into account. The final risk margin calculation is performed. This model can comprehensively reflect the jamming risk of materials during the conveying process due to the combined effects of lateral deviation and longitudinal space contraction, thus providing a precise decision-making basis for subsequent adaptive control. This scheme, combined with the calculations of dynamic phase error, parasitic lateral disturbance force, and lateral slip accumulation displacement in the previous steps, forms a complete chain from the vibrator's operating state to the material jamming risk, enabling the system to conduct risk assessment and control based on more comprehensive information. As a specific implementation method, the above technical solution can be implemented in the following way. In the plum pit and meat separation chute conveying device, the processor can acquire the driving frequency of the dual vibrators in real time, for example, by reading the frequency signal output by the frequency converter. The chute inclination angle can be measured by a MEMS inclination sensor installed at the bottom of the chute. Based on these real-time data, combined with the pre-stored semi-empirical analytical equations or two-dimensional grid velocity mapping tables of the material throwing state, the processor calculates and retrieves the instantaneous longitudinal conveying speed in the current sampling period. To calculate the dimensionless cumulative shrinkage of the opening, the processor maintains a length of... A sliding array is used to store the instantaneous longitudinal conveying speed over a recent period of time. At each sampling moment, the processor will generate new... Add to the array, remove the oldest data, and then process all data in the array. Summing yields the effective longitudinal displacement within the sliding window. This effective longitudinal displacement is related to the preset geometric gradient of the chute opening along the conveying direction. Multiply by the product, then divide by the initial opening of the feed end gap. This yields the dimensionless cumulative contraction of the opening. Simultaneously, the processor obtains the lateral slip stacking displacement at the current moment from the preceding steps. Parasitic lateral disturbance force Finally, the processor substitutes these calculation results into the dynamic jamming risk margin. In the formula, the sidewall geometric projection coefficient is... Equivalent total mass Reference Circular Frequency and discrete sampling interval All parameters are preset or known system parameters, thus enabling real-time calculation of dynamic jamming risk margin. .

[0056] Through the above technical solution, this application can effectively couple the dynamic characteristics of longitudinal material conveying with the risk of lateral stacking, overcoming the limitations of traditional methods that only focus on a single dimension of risk. This solution, by introducing the nonlinear coupling of chute geometric shrinkage and lateral stacking displacement, makes the assessment of jamming risk more comprehensive and accurate, enabling earlier identification of potential jamming risks. This provides a reliable basis for subsequent proactive intervention and control, significantly improving the stability and reliability of the plum pit and meat separation chute conveying process.

[0057] This application further proposes that in step S40, the extraction logic of the unidirectional risk deterioration gain and the acquisition process of the instantaneous frequency compensation command include: judging the changing trend of the dynamic jamming risk margin in adjacent sampling periods; when the dynamic jamming risk margin shows a deteriorating trend, generating an exponentially increasing amplification weight based on the deterioration rate; when the dynamic jamming risk margin shows a mitigating trend, forcibly shielding the influence of the deterioration rate to suppress oscillation; the feedforward control equation generates the final instantaneous frequency compensation command by proportionally calculating the deviation between the target safe jamming risk margin and the current true dynamic jamming risk margin, and multiplying it by the amplification weight in a feedforward manner; the formula for calculating the instantaneous frequency compensation command is:

[0058] in, This indicates the instantaneous frequency compensation command output at the current time t; This represents the function that maximizes the gain from unidirectional risk deterioration. The preset degradation acceleration factor is expressed in seconds. and These represent the dynamic congestion risk margins at the current and previous moments, respectively. Indicates the discrete sampling interval; This represents the proportional mapping gain of the feedforward control. This indicates the target safety margin for jamming risk.

[0059] Instantaneous frequency compensation command This is a signal output by the control system used to adjust the operating frequency of the vibrator in real time. Its function is to dynamically correct the drive frequency of the vibrator based on the current system state and risk assessment results, in order to offset or reduce parasitic lateral interference caused by phase errors, thereby maintaining stable material transport within the chute. This command can be an analog voltage signal, output to the frequency converter via a digital-to-analog converter (DAC), which adjusts the power supply frequency of the vibrating motor; or, this command can be a digital quantity, sent to the intelligent frequency converter or motor controller via a communication interface (such as Modbus or CAN bus), where its internal algorithm implements frequency adjustment. The maximum value function max(0,...) for extracting the unidirectional risk deterioration gain is used to specifically identify and quantify the trend of risk deterioration from the rate of change of risk margin. It calculates the rate of change of risk margin by comparing the risk margin at the current moment with that at the previous moment, retaining only positive values ​​(i.e., a decrease in risk margin, indicating risk deterioration), and truncating negative values ​​(an increase in risk margin, indicating risk improvement) to zero. This function can be implemented in a digital signal processor (DSP) or microcontroller (MCU) using conditional statements; alternatively, it can be implemented in hardware logic such as an FPGA using comparator and selector circuits. The preset degradation acceleration factor is in seconds. This is a pre-set parameter used to adjust the degree of nonlinear influence of the risk degradation gain on the frequency compensation command. Its dimension is seconds, meaning that multiplying it by the risk degradation rate yields a dimensionless exponential term, thus controlling the amplification effect of the exponential function on the compensation command. This factor can be stored as a system configuration parameter in the controller's non-volatile memory and loaded during system initialization; alternatively, it can be adjusted and optimized online via a human-machine interface (HMI) or host computer software. Dynamic jamming risk margin and This is an indicator that measures the risk of congestion in the current conveying system; the smaller the value, the higher the risk of congestion. These risk margin values ​​can be calculated in real time by the risk assessment module in the system and stored in memory. The calculation process can be based on a preset mathematical model and real-time sensor data. Discrete sampling interval This refers to the time period during which the system collects and processes data on key parameters such as exciter status and risk margin. It determines the real-time performance and response speed of the control system. It can be precisely controlled by the system clock or timer interrupt, for example, set to 10 milliseconds, 20 milliseconds, or 50 milliseconds. The feedforward control's proportional-mapped gain... It is a scaling factor used to linearly map the deviation between the risk margin and the target margin, as well as the effect of the risk degradation gain, to the output amplitude of the frequency compensation command. It can be stored as a controller parameter in the configuration table, and its value can be optimized through offline simulation, system identification, or online adaptive algorithms. Target safety jamming risk margin. It is a preset ideal or expected level of congestion risk. The settings can be configured based on the requirements of the plum pit and meat separation process, material characteristics, and equipment operating experience, and stored in the parameter configuration area of ​​the controller.

[0060] This application introduces a unidirectional risk deterioration gain to achieve quantitative perception of the changing trend of jamming risk, thereby upgrading simple risk margin feedback into adaptive feedforward control with trend prediction capabilities. This scheme continuously monitors the dynamic jamming risk margin. And calculate its time interval in discrete sampling intervals. The rate of change within the risk margin. By maximizing the function max(0,...), the system can accurately extract the unidirectional gain of risk margin decline (i.e., risk deterioration), effectively filtering out fluctuations in risk improvement and ensuring the targeted nature of control commands. This unidirectional risk deterioration gain is then nonlinearly amplified through an exponential function, where a preset degradation acceleration factor is used. The amplification intensity was adjusted to enable the system to generate a nonlinear compensation gain when the risk deteriorates rapidly, thus achieving a rapid response to sudden congestion risks. Finally, this amplified deterioration gain is compared with the current risk margin. and target safety jam risk margin The deviations between them are multiplicatively coupled and the gain is mapped by a scaling factor. Adjustments are made to generate instantaneous frequency compensation commands. This instruction is used for feedforward control of the hysteresis exciter. This design not only considers the degree to which the current risk deviates from the target, but also dynamically corrects the compensation intensity based on the risk deterioration trend. This allows the feedforward control to adjust the excitation frequency in real time according to the risk evolution, thereby actively suppressing parasitic lateral disturbances caused by phase errors and effectively improving the stability of the conveying process.

[0061] As a specific implementation method, the control system of the plum pit and meat separation chute conveying method can be deployed on an industrial programmable logic controller (PLC), such as a Siemens S7-1500 series PLC, which integrates motion control functions. The real-time rotational state of the dual exciters can be obtained through rotary encoders (e.g., incremental encoders with 1024 pulses per revolution) mounted on the exciter shafts. The pulse sequences generated by these encoders are fed into the PLC's high-speed counting module. Dynamic phase error. Calculated by comparing the phase difference between two encoder signals. Discrete sampling interval. It can be set to 20 milliseconds and synchronized by the PLC's cyclic interrupt organization block (OB). Instantaneous frequency compensation instruction. The calculation is performed in the PLC control program, where the max(0,...) function can be implemented using standard PLC ladder diagrams or structured text programming. A preset degradation acceleration factor is also included. It can be configured as a floating-point parameter in a PLC data block, for example Seconds. Scale-mapped gain It can also be a configurable parameter, for example Hertz / risk unit. Target safety jamming risk margin. The value can be set to 0.8 based on experimental data from specific dried plum varieties. The calculated... The signal is then output as an analog signal (e.g., 0-10V) from the PLC's analog-to-digital converter to the frequency converter (VFD) that controls the lag-side exciter motor. The VFD interprets this analog signal as a frequency adjustment command, dynamically changing the motor speed and thus adjusting the exciter's frequency. This allows the system to adaptively adjust the lag-side exciter's frequency in real time to counteract detected risk deterioration.

[0062] Through the above technical solution, this application can shift from traditional static risk assessment to adaptive control based on dynamic trend perception, effectively solving the problem that relying solely on static risk margin is insufficient to capture the dynamic trend of risk deterioration. This method achieves rapid and sensitive response to sudden jamming risks by accurately extracting the unidirectional risk deterioration gain and using an exponential function for nonlinear amplification, significantly reducing the lag in the control system's response to jamming risks. Therefore, the system can intervene promptly through frequency compensation in the early stages of risk deterioration, thereby enhancing the active suppression capability against non-steady-state synchronization errors, effectively preventing lateral accumulation of materials in the chute and the occurrence of jamming accidents, and improving the stability and reliability of the plum pit and meat separation chute conveying process.

[0063] In some of the solutions mentioned above in this application, a plum pit and meat separation chute conveying method is proposed to achieve directional grading and pit and meat separation of materials. However, in this process, due to the lack of deep integration between hardware carrier and software program, the above-mentioned complex dynamic phase error assessment, parasitic lateral disturbance force calculation and adaptive feedforward control logic are difficult to be automatically and in real time implemented in actual industrial processing equipment. This results in a disconnect between the control algorithm and the physical conveying device, and it is unable to effectively cope with real-time disturbances in the production environment.

[0064] In response, a plum pit and meat separation chute conveying device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned plum pit and meat separation chute conveying method.

[0065] Memory is a device used to store data and instructions. It can include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD), or hard disk drive (HDD). Memory plays a crucial role in the device, carrying all the computer program code, algorithm parameters, real-time acquired data, and intermediate calculation results required to implement the plum pit and meat separation chute conveying method. The processor is the core computing unit of the device, responsible for executing the computer program stored in the memory. The processor can be one or more central processing units (CPUs), microcontrollers (MCUs), digital signal processors (DSPs), field-programmable gate arrays (FPGAs), or application-specific integrated circuits (ASICs). The processor processes the real-time status data acquired from the dual exciters by executing instructions and generates corresponding control instructions according to preset control logic. A computer program is a collection of instructions designed to guide the processor to complete a specific task. This program can be embedded in the microcontroller as firmware or run as an application program on an operating system. The computer program serves as the software carrier for implementing the plum pit and meat separation chute conveying method. It defines the complete logical flow from data acquisition, phase error analysis, disturbance force calculation, risk margin assessment to feedforward control command generation. When the processor executes the program, it implements any of the aforementioned plum pit and meat separation chute conveying methods. This means that by running the computer program stored in memory, the processor can automatically complete all steps of the plum pit and meat separation chute conveying method. This includes real-time acquisition of the rotational state of the dual exciters, extraction of dynamic phase error, calculation of instantaneous angular velocity deviation, assessment of parasitic lateral disturbance force, calculation of lateral slippage and accumulation displacement, determination of dynamic jamming risk margin, and finally generation of instantaneous frequency compensation commands for feedforward control of the lag-side exciter.

[0066] The dried plum pit and meat separation chute conveying device of this application provides a physical execution platform for the aforementioned intelligent anti-jamming control logic by constructing an integrated hardware and software system. Specifically, the memory stores a complete set of algorithm programs including phase error processing, disturbance force modeling, risk margin calculation, and feedforward control command generation. The processor, as the core computing unit, reads the operating status data of the dual vibrators in real time and executes dynamic closed-loop control according to the stored program. By integrating the processor and memory into the conveying device, digital management of the chute conveying process is realized. When the processor executes the program, it can obtain the rotational state of the dual vibrators in real time and calculate the instantaneous angular velocity deviation according to the preset differential and filtering algorithms in the memory, thereby providing a data basis for subsequent disturbance force assessment. The processor further calculates the parasitic lateral disturbance force based on the dynamic phase error and uses the empirical forgetting factor to perform discrete quadratic integration on the disturbance force, thereby accurately obtaining the lateral sliding and accumulation displacement of the material. Building upon this, the processor uses a sliding window mechanism to nonlinearly couple the chute opening and contraction with the lateral displacement, calculating the dynamic jamming risk margin. Based on the risk deterioration gain, it generates frequency compensation commands in real time to provide feedforward control for the lag-side exciter. This hardware-software integrated architecture ensures the control algorithm can respond to unsteady synchronization errors in the production process with extremely high real-time performance, thereby achieving proactive suppression of jamming risk at the physical level and improving the automation level and operational reliability of agricultural product processing equipment. The introduction of this device enables the complex calculation and control logic in the aforementioned plum pit and meat separation chute conveying method to operate efficiently and accurately in a real industrial environment, transforming theoretical algorithms into operable practical productivity and effectively solving the problem of disconnect between the control algorithm and the physical conveying device.

[0067] In one specific implementation, the memory can be a combination of embedded flash memory and dynamic random access memory (DRAM). The embedded flash memory stores the computer program code and fixed algorithm parameters for the plum pit and meat separation conveyor method, while the DRAM stores real-time acquired dual exciter status data, intermediate calculation results, and dynamically updated control variables. The processor can be a high-performance industrial-grade microcontroller (such as an ARM Cortex-M series or DSP series), which integrates the necessary analog-to-digital converter (ADC) interface for acquiring sensor signals from the exciter, and a pulse width modulation (PWM) output interface for controlling the exciter's drive frequency. The computer program can be written in C / C++ and compiled and burned into the microcontroller's embedded flash memory. When the device starts, the microcontroller (processor) loads the program from the flash memory and begins cyclic execution. It reads the speed sensor signals of the dual exciter in real time through the ADC interface, converts these signals into digital quantities, and stores them in the DRAM. Subsequently, based on the program instructions stored in flash memory, the processor performs a series of mathematical processes on the data in DRAM, including differential processing, filtering, trigonometric function operations, and integration, to calculate the instantaneous angular velocity deviation, parasitic lateral disturbance force, lateral slip stacking displacement, and dynamic jamming risk margin. Finally, based on the calculated risk margin, the processor generates instantaneous frequency compensation instructions through program logic and outputs them to the inverter of the lag-side exciter via the PWM interface, thereby adjusting its drive frequency in real time and actively suppressing jamming risk.

[0068] Through the above technical solution, this application provides a chute conveying device for separating dried plum pits and meat. By organically combining a memory, processor, and computer program, it provides a robust hardware carrier and software execution environment for the aforementioned chute conveying method for separating dried plum pits and meat. This enables the automated and real-time operation of complex dynamic phase error assessment, parasitic lateral disturbance force calculation, and adaptive feedforward control logic in actual industrial processing equipment, effectively bridging the gap between control algorithms and physical conveying devices in traditional solutions. This device can respond to dynamic disturbances in the production environment in real time, accurately quantify jamming risks, and actively perform feedforward control, thereby significantly reducing the probability of jamming accidents during the separation of dried plum pits and meat, improving production efficiency and equipment reliability, and avoiding maintenance costs and production losses caused by accidents.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for conveying dried plum kernels and flesh using a chute, characterized in that, Includes the following steps: Step S10: Obtain the real-time rotational state of the dual exciter and extract the dynamic phase error, perform differential and first-order low-pass filtering on it to obtain the instantaneous angular velocity deviation; Step S20: Calculate the parasitic lateral disturbance force induced by the unsteady synchronization of the dual exciters based on the dynamic phase error; Step S25: Introduce an empirical forgetting factor that reflects the characteristics of disturbance dissipation and accumulation release, and perform a discrete quadratic integral operation with state decay on the parasitic lateral disturbance force to obtain the lateral sliding accumulation displacement of the material. Step S30: Obtain the dimensionless cumulative shrinkage of the chute opening along the conveying direction within a limited time window, and perform a nonlinear multiplication coupling operation with the lateral sliding accumulation displacement to calculate and obtain the dynamic jamming risk margin. Step S40: Extract the unidirectional risk deterioration gain based on the dynamic jamming risk margin, and introduce it as an adaptive weight into the feedforward control equation to obtain the instantaneous frequency compensation command, and perform feedforward control on the hysteresis exciter.

2. The plum pit and flesh separation chute conveying method according to claim 1, characterized in that, In step S10, the process of obtaining the instantaneous angular velocity deviation includes: Calculate the rate of change of the dynamic phase error at the current moment relative to the previous sampling moment, and use it as the current error increment; Using a first-order low-pass filtering algorithm, the current error increment and the historical instantaneous angular velocity deviation at the previous sampling moment are weighted and fused according to a preset filtering weight to obtain the instantaneous angular velocity deviation at the current moment. The filtering weight is used to balance the dynamic response of the sensor and the suppression of high-frequency noise.

3. The plum pit and flesh separation chute conveying method according to claim 2, characterized in that, In step S20, the parasitic lateral disturbance force is determined by the vector synthesis of the centrifugal force of the dual exciter eccentric block when there is dynamic phase error; The amplitude of the parasitic lateral disturbance force is positively correlated with the physical properties of the exciter eccentric block and the positive function of the reference operating frequency, and exhibits a nonlinear trigonometric function mapping relationship with the change of dynamic phase error.

4. The plum pit and flesh separation chute conveying method according to claim 3, characterized in that, In step S25, the discrete quadratic integral operation employs a state recursive model with a double forgetting factor, specifically including: The parasitic lateral disturbance force at the current moment is converted into instantaneous lateral acceleration based on the equivalent total mass. This is then superimposed with the lateral slip velocity at the previous moment after being attenuated by the first forgetting factor to obtain the lateral slip velocity at the current moment. The first forgetting factor is used to characterize the dissipation characteristics of the lateral disturbance. The integral of the lateral slip velocity at the current moment is converted into the instantaneous lateral displacement increment, which is then superimposed with the lateral slip accumulation displacement at the previous moment after being attenuated by the second forgetting factor to obtain the lateral slip accumulation displacement at the current moment. The second forgetting factor is used to characterize the system's release and reset characteristics for material accumulation.

5. The plum pit and flesh separation chute conveying method according to claim 4, characterized in that, In step S30, the nonlinear multiplication coupling operation logic for calculating and obtaining the dynamic jamming risk margin includes: Based on the geometric gradient of the chute opening descent and the instantaneous longitudinal conveying velocity within a limited time window, a first basic evaluation term characterizing the degree of longitudinal spatial contraction is constructed. Based on the lateral slip stacking displacement and sidewall geometric projection characteristics, a lateral penalty term is constructed to characterize the degree of lateral stacking blockage. The first basic evaluation term is multiplicatively coupled with the lateral penalty term, and combined with the dynamic interference term obtained after normalization of the parasitic lateral interference force, the initial nominal safety margin of the system is deducted, and the dynamic jamming risk margin is obtained by comprehensive evaluation.

6. The plum pit and flesh separation chute conveying method according to claim 5, characterized in that, In step S40, the extraction logic of unidirectional risk degradation gain and the acquisition process of instantaneous frequency compensation command include: The dynamic jamming risk margin is judged to change in adjacent sampling periods. When the dynamic jamming risk margin shows a deteriorating trend, an amplified weight with exponential growth is generated based on the deterioration rate. When the dynamic jamming risk margin shows a mitigating trend, the influence of the deterioration rate is forcibly shielded to suppress oscillation. The feedforward control equation generates the final instantaneous frequency compensation command by proportionally calculating the deviation between the target safety jam risk margin and the current real dynamic jam risk margin, and multiplying it by an amplification weight in a feedforward manner.

7. A plum pit and flesh separation chute conveying device, characterized in that, include: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the plum pit and flesh separation chute conveying method as described in any one of claims 1 to 6.

Images