Full-automatic shutdown automatic control method and system based on rice mill

Abstract

The invention relates to a full-automatic shutdown automatic control method and system based on a rice mill, and belongs to the technical field of rice mill control, and the method comprises the following steps: a, monitoring the feeding flow, and transmitting the feeding flow data to a control system; b, the control system dynamically adjusts the working power of a motor of the rice husking machine according to the feeding flow; c, continuously collecting torque data of the main shaft of the rice mill, and filtering the torque data through a Kalman filter; d, a shutdown module of the rice mill is triggered, and the shutdown module is made to operate; e, after the shutdown module is triggered, the shutdown module executes a graded deceleration strategy; step f, after the rice mill is stopped, automatically generating a parameter file through the Internet of Things module, and pushing the parameter file to the management terminal; the method has the beneficial effects that the flow change is automatically responded through multiple feeding flow intervals, the operation error is reduced, and the control performance is improved through continuous iteration of the flow-power mapping model along with production experience accumulation.

Description

Technical Field

[0001] The present invention belongs to the technical field of rice mill control, and particularly relates to a full-automatic shutdown automatic control method and system based on a rice mill. Background Art

[0002] A rice mill is an important device for processing paddy into rice and is widely used in the fields of agricultural production and food processing. In the prior art, the shutdown of a rice mill mostly relies on manual operation. The operator needs to constantly monitor the rice milling process and manually turn off the rice mill when the rice milling is completed, causing the rice mill to stop.

[0003] However, there are some problems with the manual shutdown of the rice mill. On the one hand, manual operation is prone to misjudgment, resulting in too long or too short rice milling time, affecting the quality and yield of rice. On the other hand, manual on-duty increases labor costs, and the operator is prone to fatigue during long-term work, posing a safety hazard. In addition, for some existing rice mills with an automatic shutdown function, the shutdown judgment basis is single and cannot accurately adapt to the rice milling requirements of different types of paddy, resulting in low accuracy and reliability of automatic shutdown. Summary of the Invention

[0004] The present invention provides a full-automatic shutdown automatic control method and system based on a rice mill, which is used to solve the technical problem of low accuracy of automatic shutdown of the rice mill in the prior art. It covers different production scenarios in multiple feed flow ranges, avoids the limitations of a single fixed power, requires no manual intervention, automatically responds to flow changes, reduces operation errors, and continuously iterates the flow-power mapping model trained by historical data with the accumulation of production experience to improve control performance.

[0005] To achieve the above object, the present invention is realized through the following technical solutions:

[0006] A full-automatic shutdown automatic control method based on a rice mill includes the following steps:

[0007] Step a: Monitor the feed flow through a flow sensor installed at the feed inlet of the rice mill and transmit the feed flow data to the control system;

[0008] Step b: The control system dynamically adjusts the working power of the rice mill motor based on a preset flow-power mapping model and according to the feed flow; wherein, the flow-power mapping model is obtained by training historical production data and includes power adjustment coefficients corresponding to at least three different feed flow ranges;

[0009] Step c: Continuously collect the torque data of the main shaft of the rice mill and perform torque data filtering processing through a Kalman filter to eliminate the noise interference caused by the mechanical vibration of the main shaft of the rice mill;

[0010] Step d: When any of the following conditions is detected, trigger the shutdown module of the rice mill to make the shutdown module operate:

[0011] The feed flow rate is continuously lower than 15% of the rated flow rate for 30 seconds;

[0012] The filtered torque value exceeds 120% of the safety threshold for 10 seconds continuously;

[0013] The control system receives an abnormal signal from the quality sensor at the discharge port, and the abnormal signal indicates that the broken rice rate of the finished rice exceeds the preset standard;

[0014] Step e: After triggering the shutdown module, the shutdown module executes a hierarchical deceleration strategy:

[0015] The first stage: Reduce the motor power to 50% of the current value within 0.5 seconds;

[0016] The second stage: Subsequently, linearly reduce it to 20% of the rated power within 2 seconds;

[0017] The third stage: Completely cut off the power supply within the last 1 second and activate the mechanical braking device;

[0018] Step f: After the rice mill stops, automatically generate a parameter file containing the shutdown time, the reason for triggering the shutdown, and the hierarchical deceleration strategy through the Internet of Things module, and push it to the management terminal.

[0019] Optionally, in step a, when the rice mill starts feeding, the flow sensor starts to monitor the feed flow rate, and converts the collected flow data into an electrical signal or a digital signal at a set frequency, and transmits it to the control system through the connection line;

[0020] The control system receives and analyzes the feed flow data transmitted by the flow sensor, and converts the feed flow data into a visual flow value or chart for intuitive viewing of the feed flow situation.

[0021] Optionally, in step b, dynamically adjust the working power of the rice mill motor as follows: Monitor the feed flow rate, determine the feed flow rate interval to which the feed flow rate belongs, and call the corresponding coefficient to calculate the working power. The formula is:

[0022] Working power = Base power × Power adjustment coefficient of the corresponding interval.

[0023] Optionally, in step b, the working process of the control system is as follows:

[0024] Receive data acquisition: The feed flow rate monitored by the flow sensor is transmitted to the control system;

[0025] Model matching and calculation: The control system determines the feed flow rate interval and the power adjustment coefficient to which it belongs according to the current flow rate, and calculates the target power;

[0026] Power regulation execution: adjust the motor output power through the inverter or motor controller so that the motor output power quickly approaches the motor's working power;

[0027] Closed-loop feedback optimization: Continuously collect adjusted production data to regularly update the feed flow range and power adjustment coefficient to improve control accuracy.

[0028] Optionally, the flow-power mapping model is a piecewise linear model, and the steps for constructing the piecewise linear model are:

[0029] Step S1: Data collection and preprocessing: collect measured data of flow rate and corresponding power, remove outliers and normalize;

[0030] Step S2: interval division;

[0031] Step S3: parameter estimation;

[0032] Step S4: Model verification and optimization.

[0033] Or, the piecewise linear model is constructed as:

[0034] Divide the flow range into intervals, each interval corresponds to a linear segment, then the model is expressed as:

[0035] ;

[0036] in, represents the feed flow rate, Indicates the working power of the corresponding motor, Divided flow intervals, satisfying and ; and They are respectively The linear coefficients of the intervals; the dividing points of adjacent intervals The node positions are set according to prior knowledge or data characteristics.

[0037] Optionally, in step c, a torque sensor is selected and installed at a key position of the main shaft of the rice mill to obtain a torque signal; the Kalman filter is based on a state space model, and through two steps of prediction and update, it uses the state equation and the observation equation to optimally estimate the state.

[0038] Optionally, in step d, the shutdown module establishes a shutdown linkage mechanism:

[0039] Level 1 warning: When the triggering conditions are about to be met, an alarm is issued;

[0040] Secondary shutdown: When the trigger condition is reached, the machine shuts down immediately.

[0041] Optionally, in step e, the hierarchical deceleration strategy is as follows:

[0042] First stage: Within 0 - 0.5 seconds, quickly reduce the power and buffer.

[0043] Second stage: Within 0.5 - 2.5 seconds, linearly decelerate and transition.

[0044] Third stage: Within 2.5 - 3.5 seconds, perform safety stop braking.

[0045] Optionally, in step f, after the management terminal receives the file, it supports displaying the shutdown information in the form of a visual chart.

[0046] The fully automatic shutdown control system for a rice milling machine includes:

[0047] A flow sensor for monitoring the feed flow rate and transmitting the feed flow rate data.

[0048] A control system for receiving the feed flow rate data collected by the flow sensor.

[0049] A torque sensor for collecting the torque data of the main shaft of the rice milling machine.

[0050] A Kalman filter for filtering the torque data.

[0051] A shutdown module for performing the shutdown of the rice milling machine.

[0052] An Internet of Things module for automatically generating a parameter file and pushing it to the management terminal.

[0053] The flow sensor is connected to the shutdown module through the control system, the torque sensor is connected to the shutdown module through the Kalman filter, and the shutdown module is connected to the Internet of Things module.

[0054] Advantages of the present invention:

[0055] 1. The present invention utilizes a piecewise linear model of the flow - power mapping model. The piecewise linear model divides the input variable (such as the feed flow rate) into multiple intervals, and within each interval, an independent linear formula is used to describe the variable relationship. By the feed flow rate interval to which the feed flow rate belongs, the corresponding coefficient is called to calculate the working power. Multiple feed flow rate intervals cover different production scenarios (such as raw material batch changes, output fluctuations), avoiding the limitations of a single fixed power, without manual intervention, automatically responding to flow rate changes, reducing operation errors. The flow - power mapping model trained based on historical data is continuously iterated with the accumulation of production experience, improving the control performance.

[0056] 2. The present invention is a Kalman filter based on a state space model. Through two steps of prediction and update, using the state equation and the observation equation, it optimally estimates the system state. In the processing of the main shaft torque data of a rice milling machine, the main shaft torque is used as the state variable, and the data collected by the sensor is used as the observation variable. According to the mechanical characteristics and operating rules of the rice milling machine, accurate state equations and observation equations are established, combined with the process noise covariance matrix and the observation noise covariance matrix, to achieve real-time filtering of the torque data. In the prediction step, based on the state estimate value at the previous moment and the state equation, the state at the current moment is predicted; in the update step, combined with the observation value at the current moment, the observation equation is corrected to obtain a more accurate torque estimate value, effectively eliminating the noise interference generated by mechanical vibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0058] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0059] Figure 2 It is a schematic diagram of the working process of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The following will describe the embodiments of the present application in detail with reference to the drawings.

[0061] Embodiment 1;

[0062] As Figure 1 shown, this embodiment provides a fully automatic shutdown automatic control system based on a rice milling machine, including:

[0063] A flow sensor for monitoring the feed flow rate and transmitting the feed flow rate data;

[0064] A control system for receiving the feed flow rate data collected by the flow sensor;

[0065] A torque sensor for collecting the torque data of the main shaft of the rice milling machine;

[0066] A Kalman filter for filtering the torque data;

[0067] A shutdown module for executing the shutdown of the rice milling machine;

[0068] An Internet of Things module for automatically generating a parameter file and pushing it to a management terminal (not shown in the figure);

[0069] The flow sensor is connected to the shutdown module through the control system, the torque sensor is connected to the shutdown module through the Kalman filter, and the shutdown module is connected to the Internet of Things module;

[0070] The feed flow collected by the flow sensor is transmitted to the control system. The control system receives the feed flow collected by the flow sensor. The torque sensor collects the torque of the main shaft of the rice mill, and the torque data is filtered and processed through the Kalman filter. The shutdown module triggers the shutdown module of the rice mill according to the feed flow and the filtered torque value, so that the shutdown module operates. The shutdown module executes the shutdown of the rice mill. After the shutdown of the rice mill is completed, a parameter file is automatically generated through the Internet of Things module and pushed to the management terminal.

[0071] Embodiment 2;

[0072] As Figure 2 shown, based on Embodiment 1, this embodiment provides a full-automatic shutdown and automatic control method for a rice mill, including the following steps:

[0073] Step a: Monitor the feed flow through a flow sensor installed at the feed inlet of the rice mill and transmit the feed flow data to the control system;

[0074] Step b: The control system dynamically adjusts the working power of the rice mill motor based on a preset flow-power mapping model and according to the feed flow; wherein, the flow-power mapping model is obtained by training historical production data and includes power adjustment coefficients corresponding to at least three different feed flow ranges;

[0075] Step c: Continuously collect the torque data of the main shaft of the rice mill and perform torque data filtering processing through the Kalman filter to eliminate the noise interference caused by the mechanical vibration of the main shaft of the rice mill;

[0076] Step d: When any of the following conditions is detected, trigger the shutdown module of the rice mill to make the shutdown module operate:

[0077] The feed flow is continuously lower than 15% of the rated flow for 30 seconds (if the feed volume is continuously too low, it may cause insufficient materials in the rice milling chamber, resulting in idling wear or a significant decrease in processing efficiency), and the rated flow is set according to the actual situation;

[0078] The filtered torque value continuously exceeds 120% of the safety threshold for 10 seconds (a sudden change or continuous excessive torque may indicate that the pressure between the rice milling roller and the material is too large (such as: foreign object jamming, abnormal raw material humidity), which is likely to cause equipment overload damage), and the safety threshold is set according to the actual situation;

[0079] The control system receives an abnormal signal from the quality sensor at the discharge port. The abnormal signal indicates that the broken rice rate of the finished rice exceeds the preset standard (the excessive broken rice rate (judged according to the actual situation) reflects that the rice milling parameters (such as milling pressure, rotation speed) are unreasonable or the equipment is worn, and it is necessary to stop the machine in time for adjustment to prevent unqualified products from flowing into the next process);

[0080] Step e: After triggering the shutdown module, the shutdown module executes a hierarchical deceleration strategy:

[0081] The first stage: Reduce the motor power to 50% of the current value within 0.5 seconds;

[0082] The second stage: Then linearly reduce it to 20% of the rated power within 2 seconds;

[0083] The third stage: Completely cut off the power supply within the last 1 second and activate the mechanical braking device;

[0084] Step f: After the rice mill stops, the IoT module automatically generates a parameter file containing the shutdown time, the reason for triggering the shutdown, and the hierarchical deceleration strategy, and pushes it to the management terminal.

[0085] Embodiment 3;

[0086] Based on Embodiment 2, in step a, when the rice mill starts feeding, the flow sensor starts to monitor the feeding flow rate, and converts the collected flow rate data into an electrical signal or a digital signal at a set frequency, and transmits it to the control system through the connection line;

[0087] The control system receives and analyzes the feeding flow rate data transmitted by the flow sensor, and converts the feeding flow rate data into a visual flow rate value or chart for intuitive viewing of the feeding flow rate situation.

[0088] Specifically, when the rice mill starts feeding, the flow sensor based on the principle of electromagnetic induction starts quickly. Taking the electromagnetic induction type flow sensor as an example, the excitation coil inside it generates an alternating magnetic field. When the conductive rice grain fluid passes through the flow sensor pipeline, it cuts the magnetic induction line to generate an induced electromotive force, and the magnitude of the induced electromotive force has a linear relationship with the fluid flow rate; the flow sensor samples the collected feeding flow rate data at a high frequency of 10 - 50 times per second, and through the built-in analog-to-digital conversion module (ADC), accurately converts it into an electrical signal of 0 - 5V or a digital signal following the Modbus or Profibus industrial communication protocol, and stably transmits it to the control system through a connection line with strong anti-interference ability such as shielded twisted pair or optical fiber.

[0089] The control system uses an embedded industrial computer as the core and is equipped with a dedicated data acquisition module. After receiving the signal, digital filtering is first performed to eliminate noise data generated by electromagnetic interference or mechanical vibration. Subsequently, through a preset algorithm, the original signal is converted into the actual flow rate value (unit: kg / min). To achieve more intuitive monitoring, the control system uses SCADA (Supervisory Control and Data Acquisition) software to present the real-time flow rate data in the form of dynamic line charts and bar charts on the human-machine interface. At the same time, it supports the storage and query functions of historical data, facilitating the analysis of the change trend of the feed flow rate and enabling the intuitive viewing of the feed flow rate situation.

[0090] Example 4;

[0091] Based on Example 2, in step b, the construction of the flow-power mapping model is based on data obtained through training with historical production data. Specifically:

[0092] The data source is the historical production data collected on the feed flow rate, motor power, and rice milling quality indicators (such as: broken rice rate, rice yield rate) of the rice milling machine under different working conditions;

[0093] Training objective: Through data analysis (such as: regression analysis, machine learning algorithms), establish the mapping relationship between the feed flow rate and the motor power to ensure that the model can accurately reflect the actual production law.

[0094] The feed flow rate interval is divided into three different feed flow rate intervals, including: low flow rate interval, medium flow rate interval, and high flow rate interval, and each interval corresponds to a power adjustment coefficient;

[0095] The low flow rate interval means less feed, less resistance required for rice milling, and the motor power needs to be adjusted downwards according to the power adjustment coefficient. For example, the power adjustment coefficient is 0.8; when the flow rate < 50 kg / h, the working power = basic power × 0.8;

[0096] The medium flow rate interval means moderate feed, and the power operates according to the basic power or the power adjustment coefficient to balance efficiency and energy consumption. For example, the power adjustment coefficient is 1.0; when 50 kg / h ≤ flow rate ≤ 100 kg / h, the working power = basic power × 1.0;

[0097] The high flow rate interval means high feed, large resistance for rice milling, and the power needs to be adjusted upwards according to the power adjustment coefficient. For example, the power adjustment coefficient is 1.2 to ensure the rice milling effect and equipment safety; when the flow rate > 100 kg / h, the working power = basic power × 1.2;

[0098] Dynamically adjust the working power of the rice milling machine motor as follows: Monitor the feed flow rate, determine the feed flow rate interval to which the feed flow rate belongs, and call the corresponding coefficient to calculate the working power. The formula is:

[0099] Working power = Base power × Power adjustment coefficient for the corresponding interval.

[0100] The working process of the control system is as follows:

[0101] Receiving data acquisition: The feed flow rate monitored by sensors (such as flow sensors) is transmitted to the control system;

[0102] Model matching and calculation: The control system determines the feed flow rate interval and power adjustment coefficient to which it belongs based on the current flow rate, and calculates the target power (such as 0.8, 1.0, or 1.2 times the power adjustment coefficient in the power adjustment coefficient);

[0103] Power adjustment execution: The output power of the motor is adjusted through a frequency converter or motor controller to make the output power of the motor quickly approach the working power of the motor;

[0104] Closed-loop feedback optimization: Continuously collect the adjusted production data (such as the output power of the motor, rice milling quality) to be used for regularly updating the feed flow rate interval and power adjustment coefficient to which it belongs, and improve the control accuracy.

[0105] Multiple feed flow rate intervals can cover different production scenarios (such as raw material batch changes, output fluctuations), avoiding the limitations of a single fixed power, without manual intervention, automatically responding to flow rate changes, reducing operation errors. The flow-power mapping model trained based on historical data can be continuously iterated with the accumulation of production experience to improve the performance of the control system.

[0106] The specific flow-power mapping model is a piecewise linear model. A piecewise linear model is a model that divides the input variable (such as flow rate) into multiple intervals, and within each interval, an independent linear formula is used to describe the variable relationship. It solves the problem that a single linear model cannot accurately describe complex non-linear relationships, while retaining the simplicity and interpretability of the linear model.

[0107] The working principle of the piecewise linear model is as follows:

[0108] The value range of the independent variable (such as feed flow rate ) is divided into several continuous intervals, such as low, medium, and high flow rate intervals;

[0109] Within each flow rate interval, the dependent variable (such as motor power ) and the independent variable satisfy a linear relationship, but the linear equation parameters (slope, intercept) in different intervals are different;

[0110] It is applicable to the working power consumption characteristics of the motor showing different under different loads:

[0111] When the feed flow rate is low, the motor mainly overcomes the no-load resistance, and the working power increases slowly (small slope);

[0112] At medium feed flow rate, the working power increases linearly with the flow rate (medium slope);

[0113] At high feed flow rate, the material load increases and the working power needs to be rapidly increased (large slope).

[0114] Specifically, the piecewise linear model is constructed as follows:

[0115] The flow rate range is divided into intervals, and each interval corresponds to a linear segment, then the model is expressed as:

[0116] ;

[0117] where, represents the feed flow rate (input variable), represents the working power of the corresponding motor (output variable), are respectively the divided flow rate intervals, satisfying and (maximum feed flow rate).

[0118] and are respectively the linear coefficients of the th interval, that is, is the intercept, is the slope, and are determined by data fitting.

[0119] The demarcation points between adjacent intervals are nodes, and the node positions are set according to prior knowledge or data characteristics (such as: equally spaced division, division based on inflection points).

[0120] The specific steps for constructing the piecewise linear model are as follows:

[0121] Step S1: Data collection and preprocessing: Collect the measured data of the flow rate and the corresponding power , remove outliers and normalize.

[0122] Step S2: Interval division (determine node positions):

[0123] Equally spaced division: The flow rate range is evenly divided into intervals, such as: .

[0124] Division based on inflection points: Identify the inflection points of the flow rate - power curve through data analysis (such as: derivative method, clustering algorithm) as the interval demarcation points.

[0125] Adaptive partitioning: Use an optimization algorithm (e.g., dynamic programming) to automatically determine the node positions and minimize the model fitting error.

[0126] Step S3: Parameter estimation (linear fitting): Find a set of parameters to obtain the model predicted values , and use the least squares method to minimize the error between the model predicted values and the actual observed values to obtain the minimized sum of squared errors: Minimizing the sum of squared errors is achieved by minimizing the squared errors, avoiding the cancellation of positive and negative errors, and assigning higher weights to larger errors to ensure that the overall fitted curve is close to the data points (mean points);

[0127] The data points in the th interval for minimizing the sum of squared errors are , where is a set of sample data pairs, indicating that when the flow rate is , the power observed value is , is the data set from to , covering all samples in the th interval, is the number of samples in the th interval, which is .

[0128] Minimizing the sum of squared errors in

[0129] , ; where and are the mean of the flow rate and the mean of the power in the th interval and the th interval, respectively.

[0130] For , represents the deviation of the flow rate from the interval mean , represents the deviation of the power from the interval mean , and the product of the two deviations reflects the and co-variation direction and degree:

[0131] If and are both greater than or less than the mean (positively correlated), the product is positive;

[0132] If and one is greater than the mean value and the other is less than the mean value (negatively correlated), the product is negative.

[0133] Therefore, the sum of the numerators of the slope is called covariance, which measures the and degree of linear correlation.

[0134] is the square of the deviation of the flow rate . The sum is called variance, which measures the dispersion degree of the flow rate data. The denominator of the slope is used to normalize the covariance to eliminate the influence of the flow rate's own fluctuation on the slope.

[0135] The slope represents the average change in power when the flow rate increases by one unit within the interval (e.g., when the flow rate increases by 0.5 kg / min, the power increases by 0.5 W). If , it means that the power is independent of the flow rate within this interval (e.g., standby state).

[0136] For , taking the mean value on both sides of the linear function , we get: . After transposing, we obtain the intercept formula, indicating that the fitting line must pass through the mean point of the data points, that is, the line passes through the concentration point (mean point) of the data distribution.

[0137] The intercept is, when , represents the initial power within the interval (e.g., the no-load power consumption of the device). If the interval of does not contain 0 (e.g., ), then

[0138] Step S4: Model verification and optimization: Use the root mean square error (RMSE) or the coefficient of determination ( ) index to evaluate the model fitting effect; if the error is too large, adjust the number of intervals or the node positions and refit the parameters.

[0139] Example 5;

[0140] Based on Embodiment 2, in step c, a torque sensor is selected and installed at the key position of the main shaft of the rice milling machine to ensure that torque signals can be obtained. The analog signal output by the torque sensor is preprocessed by amplification and filtering through a signal conditioning circuit to improve the signal quality, and then the analog signal is converted into a digital signal by an A / D converter for subsequent processing. At the same time, a microcontroller or data acquisition card with high stability is used to set a reasonable sampling frequency to ensure that torque data can be continuously and stably collected. For example, the sampling frequency is set according to the rotational speed and torque change of the rice milling machine, and the sampling frequency is set to 100Hz - 1000Hz to facilitate continuous and stable collection of torque data.

[0141] The Kalman filter is based on the state space model. Through two steps of prediction and update, using the state equation and observation equation of the system, the optimal estimation of the system state is carried out. In the processing of the main shaft torque data of the rice milling machine, the main shaft torque is used as the state variable, and the data collected by the sensor is used as the observation variable. According to the mechanical characteristics and operating laws of the rice milling machine, accurate state equations and observation equations are established, combined with the process noise covariance matrix and the observation noise covariance matrix, to realize real-time filtering of torque data. In the prediction step, based on the state estimation value at the previous moment and the state equation, the state at the current moment is predicted; in the update step, combined with the observation value at the current moment, the observation equation is corrected to obtain a more accurate torque estimation value, effectively eliminating the noise interference caused by mechanical vibration.

[0142] The core of the Kalman filter is the state space model, which abstracts the main shaft torque of the rice milling machine into state equations and observation equations. The state equation describes the change law of torque in the time series. For example, if the change of the main shaft torque at adjacent moments conforms to the first-order autoregressive model, the state transition matrix can reflect the change trend of torque from one moment to the next moment. The observation equation establishes the relationship between the sensor measurement value and the actual torque. Due to the measurement error of the sensor, the observation matrix needs to consider the accuracy and installation position of the sensor.

[0143] Specifically, for the state equation (describing the change law of torque over time), the dynamic change of the main shaft torque is affected by the internal dynamics of the system (such as load fluctuation, mechanical damping) and external disturbances, and is established as a first-order linear equation:

[0144] ;

[0145] where is the state variable the first derivative with respect to time describes the rate of change of the state variable over time, reflecting the dynamic evolution trend of the system state, is the state variable, the actual value of the spindle torque (scalar or vector, considering multi-dimensional dynamics), with the unit of N·m; is the state matrix, describing the self-decay characteristic of the torque (e.g., the natural decay of the torque caused by mechanical damping), with the unit of 1 / s; is the input matrix, describing the influence of the external control input (e.g., the driving force of the motor) on the torque, with the unit of N·m / V (if the input is a voltage signal).

[0146] is the input variable, the external control input (e.g., the motor voltage or current), with the unit of V or A;

[0147] is the process noise, a random disturbance following a Gaussian distribution (e.g., the torque fluctuation caused by the impact of material particles), that is , where is the representation of the multivariate normal distribution (also called Gaussian distribution), used to describe the statistical characteristics of the process noise, is the process noise covariance matrix.

[0148] For the observation equation (establishing the relationship between the sensor measurement value and the actual torque), the measurement value of the sensor (e.g., torque sensor) contains the actual torque and the measurement noise:

[0149] ;

[0150] where is the observation variable, the torque value measured by the sensor, with the unit of N·m.

[0151] is the observation matrix, the measurement sensitivity matrix (the sensor gain when it is a scalar), describing the mapping from the actual torque to the measurement value, usually 1 (if the sensor directly outputs the torque value).

[0152] is the measurement noise, a measurement error following a Gaussian distribution (e.g., sensor noise, signal interference), that is , where is the representation of the multivariate normal distribution (also called Gaussian distribution), used to describe the statistical characteristics of the measurement noise, is the observation noise covariance matrix.

[0153] The process noise covariance matrix is used to quantify the uncertainties within the system. For example, during the operation of a rice milling machine, factors such as uneven grain sizes and minor wear of transmission components can cause random variations in torque, which are reflected in the process noise covariance matrix. The observation noise covariance matrix, on the other hand, reflects the inaccuracy of sensor measurements. The parameters of the observation noise covariance matrix can be determined by conducting multiple calibration experiments on the sensor and statistically analyzing the measurement errors. In practical applications, it is usually necessary to continuously adjust the process noise covariance matrix and the observation noise covariance matrix based on experience and experimental data to optimize the filtering effect.

[0154] Example 6;

[0155] Based on Example 2, in step d, the shutdown module establishes a shutdown linkage mechanism:

[0156] First-level warning: When the trigger conditions are about to be met (the feed flow rate is continuously below 15% of the rated flow rate for 30 seconds; the filtered torque value continuously exceeds 120% of the safety threshold for 10 seconds; the control system receives an abnormal signal from the quality sensor at the discharge port), an alarm is issued;

[0157] Second-level shutdown: When the trigger conditions are reached, the machine is immediately shut down. The specific trigger conditions are shown in Table 1 below:

[0158] Table 1 Trigger conditions for shutdown

[0159]

[0160] Example 7;

[0161] Based on Example 2, in step e, the hierarchical deceleration strategy is as follows:

[0162] First stage: Rapid power reduction and buffering (0 - 0.5 seconds);

[0163] Within 0.5 seconds, the motor output power is rapidly reduced to 50% of the current operating power. In this stage, current closed-loop control technology is adopted to accurately adjust the motor torque and avoid mechanical shocks caused by sudden power drops. At the same time, the system continuously monitors the motor speed and load changes, and dynamically adjusts the power reduction curve to ensure stable operation of the equipment.

[0164] Second stage: Linear deceleration transition (0.5 - 2.5 seconds);

[0165] Within 2 seconds, the motor power is continuously and smoothly reduced from the current value to 20% of the rated power. In this stage, the PID control strategy is used to adjust the output current in real-time according to the preset deceleration curve, so that the motor speed decreases uniformly. During this period, the sensor continuously collects motor temperature and vibration data. Once an abnormality is detected, the emergency handling procedure is immediately triggered to ensure the safety of the equipment.

[0166] Third stage: Safe stop braking (2.5 - 3.5 seconds);

[0167] In the last second, the power supply of the motor is cut off through high-voltage DC braking technology, instantly eliminating the electromagnetic torque. At the same time, the hydraulically driven mechanical braking device responds quickly, and through the close fitting of the friction plate and the brake disc, mechanical locking is achieved. During the braking process, the pressure sensor monitors the braking torque in real time to ensure that the braking force meets the safety standards and avoid sliding or rebounding phenomena.

[0168] The stepwise deceleration strategy effectively reduces the inertial impact during the equipment shutdown process, extends the service life of mechanical components, and ensures the safety of operators and equipment through the coordinated cooperation of electrical braking and mechanical braking.

[0169] Embodiment 8;

[0170] Based on Embodiment 2, in step f, when the rice mill executes the shutdown instruction and completes all operation processes, a signal indicating that the motor speed has returned to zero is captured, and then the IoT module is triggered to start the data acquisition program. The IoT module reads the shutdown timestamp stored in the PLC control system in real time through the industrial data bus, accurate to the millisecond level; synchronously retrieves the fault diagnosis system log to identify the specific reasons for triggering the shutdown, covering more than 20 preset scenarios such as overload protection, material shortage alarm, and abnormal equipment temperature rise; at the same time, it analyzes the stepwise deceleration strategy parameters, including the deceleration stage division, the time threshold for each stage, and the core data of the speed reduction gradient.

[0171] The IoT module formats the shutdown time of the collected data according to the ISO8601 time standard, constructs a parameter file in JSON format, the file structure follows the industrial IoT data exchange specification (IEC62541), and uses the AES-256 encryption algorithm to ensure the security of data transmission. After receiving the file, the management terminal supports displaying the shutdown information in the form of a visual chart.

[0172] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope recorded in the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claimed rights.

Claims

1. An automatic shutdown automatic control method based on a rice milling machine, characterized in that, The steps include: Step a: monitoring the feed flow rate by means of a flow sensor installed at the feed inlet of the rice mill, and transmitting the feed flow rate data to a control system; Step b: The control system dynamically adjusts the working power of the rice milling machine motor according to the feed flow rate based on a preset flow-power mapping model; wherein the flow-power mapping model is obtained by training historical production data and includes power adjustment coefficients corresponding to at least three different feed flow intervals; Step c: continuously collecting torque data of the main shaft of the rice milling machine, and filtering the torque data through a Kalman filter to eliminate noise interference caused by mechanical vibration of the main shaft of the rice milling machine; Step d: When any of the following conditions is detected, the shutdown module of the rice mill is triggered to operate: The feed flow rate is lower than 15% of the rated flow rate for 30 seconds continuously; The filtered torque value exceeds 120% of the safety threshold for 10 seconds; The control system receives an abnormal signal from the discharge port quality sensor, and the abnormal signal indicates that the broken rice rate of the finished rice exceeds a preset standard; Step e: After the shutdown module is triggered, the shutdown module executes a graded deceleration strategy: Phase 1: Reduce the motor power to 50% of the current value within 0.5 seconds; Stage 2: Then the power is linearly reduced to 20% of the rated power within 2 seconds; The third stage: completely cut off the power supply within the last second and activate the mechanical brake device; Step f: After the rice mill is shut down, a parameter file including the downtime, the reason for triggering the shutdown, and the graded deceleration strategy is automatically generated through the Internet of Things module and pushed to the management terminal.

2. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 1, characterized in that, In the step a, when the rice mill starts to feed, the flow sensor starts to start and monitor the feed flow, and converts the collected flow data into an electrical signal or a digital signal according to a set frequency, and transmits it to the control system through a connecting line; The control system receives and analyzes the feed flow data transmitted by the flow sensor, and converts the feed flow data into a visual flow value or chart to facilitate intuitive viewing of the feed flow situation.

3. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 1, characterized in that In the step b, the working power of the rice milling machine motor is dynamically adjusted by monitoring the feed flow rate, determining the feed flow rate interval to which the feed flow rate belongs, and calling the corresponding coefficient to calculate the working power. The formula is: Working power = basic power × power adjustment coefficient of the corresponding range.

4. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 3, wherein, In step b, the working process of the control system is: Receiving data collection: The feed flow rate monitored by the flow sensor is transmitted to the control system; Model matching and calculation: The control system determines the feed flow range and power adjustment coefficient according to the current flow, and calculates the target power; Power regulation execution: adjust the motor output power through the inverter or motor controller so that the motor output power quickly approaches the motor's working power; Closed-loop feedback optimization: Continuously collect adjusted production data to regularly update the feed flow range and power adjustment coefficient to improve control accuracy.

5. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 3, characterized in that, The flow-power mapping model is a piecewise linear model, and the steps of constructing the piecewise linear model are: Step S1: Data collection and preprocessing: collect measured data of flow rate and corresponding power, remove outliers and normalize; Step S2: interval division; Step S3: Parameter estimation; Step S4: Model verification and optimization; Or, the piecewise linear model is constructed as: Divide the flow rate range into intervals, and each interval corresponds to a linear segment. Then the model is expressed as: ； Among them, represents the feed flow rate, represents the operating power of the corresponding motor, are respectively the divided flow rate intervals, satisfying and ; and are respectively the linear coefficients of the th interval; the demarcation points of adjacent intervals are nodes, and the node positions are set according to prior knowledge or data characteristics.

6. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 1, wherein, In step c, a torque sensor is selected and installed at a key position of the main shaft of the rice mill to obtain torque signals; based on the state space model, the Kalman filter performs optimal estimation of the state through two steps of prediction and update, using the state equation and the observation equation.

7. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 1, characterized in that, In step d, the shutdown module establishes a shutdown linkage mechanism: First-level warning: When the trigger condition is about to be met, an alarm is issued; Second-level shutdown: When the trigger condition is reached, the machine is immediately shut down.

8. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 1, wherein In step e, the hierarchical deceleration strategy is: First stage: Within 0 - 0.5 seconds, quickly reduce the power and buffer; Second stage: Within 0.5 - 2.5 seconds, linearly decelerate and transition; Third stage: Within 2.5 - 3.5 seconds, perform a safety stop braking.

9. The fully automatic shutdown automatic control method based on a rice milling machine according to claim 1, characterized in that, In step f, after receiving the file, the management terminal supports displaying the shutdown information in the form of a visual chart.

10. A full-automatic shutdown automatic control system based on a rice milling machine, which is used to execute the full-automatic shutdown automatic control method based on a rice milling machine according to any one of claims 1-9, characterized in that, Including: A flow sensor for monitoring the feed flow rate and transmitting the feed flow rate data; A control system for receiving the feed flow rate data collected by the flow sensor; A torque sensor for collecting the torque data of the main shaft of the rice mill; A Kalman filter for filtering the torque data; A shutdown module for performing the shutdown of the rice mill; An Internet of Things module for automatically generating a parameter file and pushing it to the management terminal; The flow sensor is connected to the shutdown module through the control system, the torque sensor is connected to the shutdown module through the Kalman filter, and the shutdown module is connected to the Internet of Things module.

Images