High-efficiency operation and anti-blocking control method for high-frequency screen

Abstract

The invention discloses a high-efficiency operation and anti-blocking control method for a high-frequency screen. The method comprises the steps that the current of a preceding-stage belt and the current and frequency of the high-frequency screen are collected through an intelligent control unit; establishing a high-frequency screen current prediction model by using a KNN regression algorithm; calculating and controlling the operation frequency of the high-frequency screen through an interpolation method according to the preceding-stage belt current; the actual current of the high-frequency screen is compared with the model prediction current in real time, when the deviation exceeds a threshold value, the frequency is automatically increased for dredging and timing, if the current falls back, the frequency is recovered, and if the current does not fall back overtime, an alarm is triggered; according to the method, the KNN regression model based on the preceding-stage belt current and the high-frequency screen operation frequency is established, self-adaptive adjustment of the high-frequency screen operation frequency is achieved, the problem of electric energy waste caused by fixed-frequency operation of a traditional high-frequency screen is solved, operation parameters can be dynamically adjusted according to the actual feeding amount, equipment energy consumption is reduced, and the service life of the high-frequency screen is prolonged. And the economical efficiency of the production line is improved.

Description

Technical Field

[0001] This invention relates to the field of high-efficiency operation control technology for high-frequency screens, and in particular to a method for high-efficiency operation and anti-clogging control of high-frequency screens. Background Technology

[0002] A high-frequency screen is a commonly used solid separation device. Its function is to screen and classify materials. When the material enters the screen box, it will be driven by a vibrating motor to generate vibration. The screen will also vibrate due to the vibration. Materials of different particle sizes will move towards the end of the screen due to different inertial forces. The material that passes through the screen will be collected at one end, while the material smaller than the screen aperture will fall to the lower layer, forming a material layer of different particle sizes.

[0003] High-frequency screens typically operate at a fixed frequency. This method cannot adjust the operating frequency according to the amount of incoming material, resulting in wasted electricity. Furthermore, when material blockage occurs, the existing system cannot automatically detect it, severely affecting the screening effect and quality.

[0004] Therefore, in response to the problems mentioned above, this invention proposes a method for high-efficiency operation and anti-clogging control of high-frequency screens. Summary of the Invention

[0005] To overcome the current problem that high-frequency screens lack online monitoring and control equipment, thus failing to achieve high-efficiency operation and automatically detect and clear material blockages, this invention proposes a high-efficiency operation and anti-blockage control method for high-frequency screens. This method uses an intelligent control unit and frequency converter to form a system that enables high-efficiency operation of the high-frequency screen, automatically detects and clears material blockages, and then automatically resumes normal operation.

[0006] The technical solution of this invention is: a method for high-efficiency operation and anti-clogging control of a high-frequency screen, comprising the following steps: S1 collects the current value of the upstream belt motor, the current value of the high-frequency screen inverter, and the operating frequency value in real time through the intelligent control unit; S2. By changing the feed rate of the upstream belt, multiple sets of upstream belt current values, corresponding high-frequency screen current values ​​and operating frequency values ​​are obtained to form a training dataset. The upstream belt current values ​​and operating frequency values ​​in the training dataset are normalized and trained using the KNN regression algorithm to obtain the high-frequency screen current prediction model. S3, During normal system operation, based on the real-time collected front-end belt current value, the operating control frequency of the high-frequency screen is determined from the training dataset using interpolation, and the high-frequency screen is controlled to operate at that frequency. S4 compares the actual current value of the high-frequency screen with the predicted current value of the high-frequency screen current prediction model in real time. When the deviation exceeds the preset threshold, the operating frequency of the high-frequency screen is increased to clear the blockage, and a timer is started. If the actual current value of the high-frequency screen falls back to the normal range within the timer period, the high-frequency screen is restored to the normal operating frequency; otherwise, an alarm device is triggered.

[0007] It is worth noting that this method achieves adaptive adjustment of the operating frequency of the high-frequency screen by establishing a dynamic mapping relationship between current and frequency, thus solving the problem of energy waste caused by fixed frequency operation. The normalization process adopts the linear normalization method to map the data to the [0,1] interval, thereby improving the training efficiency and prediction accuracy of the KNN algorithm.

[0008] Preferably, the intelligent control unit is a PLC, a microcontroller, or an embedded control system.

[0009] Preferably, in step S2, the K value of the KNN regression algorithm is an odd number between 3 and 10.

[0010] Preferably, the value of K is dynamically adjusted based on the size of the training data. When the amount of data is large, a larger value of K is selected, thereby improving the stability and generalization ability of the model.

[0011] Preferably, the interpolation method is a linear interpolation method, specifically including: finding two adjacent preceding belt current values ​​in the training dataset based on the current preceding belt current value. and and the corresponding operating frequency value and Through the formula: ; Calculate the operating control frequency of the high-frequency screen, where I is the current value of the upstream belt current.

[0012] It is worth noting that the interpolation method can smoothly transition between training data points, thereby avoiding the adverse effects of frequency abrupt changes on screening effect and equipment life.

[0013] Preferably, in step S3, the high-frequency screen is controlled to operate at the operating control frequency by an increment.

[0014] Preferably, the increment is 0.5-2Hz, which can improve screening efficiency without causing equipment overload or increased energy consumption due to excessive frequency.

[0015] Preferably, in step S4, when the deviation exceeds a preset threshold, the operating frequency of the high-frequency screen is increased to a fixed value or increased by a certain proportion.

[0016] Preferably, the frequency adjustment method is a phased increase, with each increase being 3-5Hz, until the current drops back or the frequency upper limit is reached, in order to avoid equipment impact.

[0017] Preferably, the alarm device includes an audible and visual alarm, driven by an intelligent control unit.

[0018] Preferably, the method displays the processing data in real time on a display screen, including the current value of the upstream belt, the current value of the high-frequency screen, the operating frequency value, and the alarm status.

[0019] Preferably, in step S2, the feed rate of the front belt is changed by controlling the running speed of the front belt or the opening of the feed valve.

[0020] Preferably, the update cycle of the high-frequency screen current prediction model is periodic or adjusted according to operating data.

[0021] The beneficial effects of this invention are: 1. This invention establishes a KNN regression model based on the current of the preceding belt and the operating frequency of the high-frequency screen, thereby achieving adaptive adjustment of the operating frequency of the high-frequency screen. This overcomes the problem of energy waste caused by the fixed frequency operation of traditional high-frequency screens. The operating parameters can be dynamically adjusted according to the actual feed rate, reducing equipment energy consumption and improving the economic efficiency of the production line.

[0022] 2. This invention uses real-time current monitoring and model prediction comparison, which enables the system to automatically and accurately identify the blockage status without relying on manual judgment. This avoids the decline in screening efficiency and equipment damage caused by blockage, and improves the system's intelligence level and operational reliability.

[0023] 3. Upon detecting material blockage, the system automatically increases the operating frequency of the high-frequency screen to clear the blockage. It also features a timing judgment mechanism, which allows the system to autonomously clear blockages and restore normal operation in most cases. This significantly reduces the frequency of manual intervention and processing time, ensuring the continuity and stability of production.

[0024] 4. This invention uses interpolation combined with training dataset for frequency control, enabling the high-frequency screen to smoothly follow changes in the feed rate. This optimizes the screening effect and avoids the impact of sudden frequency changes on the mechanical structure, thus extending the service life of the equipment. Attached Figure Description

[0025] Figure 1 The diagram shown is a schematic representation of the system framework of the present invention. Figure 2 The diagram shown illustrates the working principle of this invention. Figure 3 The diagram shown is a schematic representation of the model acquisition process of this invention. Figure 4 The diagram shown is a schematic representation of the system operation flow of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1 This invention provides an embodiment of a method for high-efficiency operation and anti-clogging control of a high-frequency screen: This method is achieved through an intelligent control system for high-frequency screens. This system senses the feed rate of the upstream belt, establishes an intelligent model between the feed rate and the optimal operating parameters of the high-frequency screen, and realizes adaptive adjustment of the operating frequency of the high-frequency screen and automatic identification and handling of material blockage based on this model. The system includes an intelligent control unit, a high-frequency screen inverter, a upstream belt current detection unit, a display unit, and an alarm.

[0028] The intelligent control unit can be a PLC, microcontroller, or embedded control system, etc., which has a CPU and industrial control equipment. It needs to have analog input channels (for acquiring current signals), analog output channels (for controlling the frequency converter), digital output channels (for controlling the alarm), and communication interfaces (for connecting to the display unit).

[0029] The high-frequency screen inverter receives a frequency command signal from the intelligent control unit and drives the vibration motor of the high-frequency screen to run at a specified frequency. At the same time, its internal current detection module can feed back the motor operating current to the intelligent control unit.

[0030] The front belt current detection unit uses a current transmitter or ammeter to detect and transmit the operating current of the front belt drive motor in real time. This current value is proportional to the feed rate of the belt and is a key parameter reflecting the feeding situation.

[0031] The display unit is a touch screen or industrial computer connected to the intelligent control unit. It is used to display the system status, key parameters (current of the upstream belt, current and frequency of the high-frequency screen, model prediction value, alarm information) and historical data curves in real time, and can provide a simple human-machine configuration interface.

[0032] The alarm is an audible and visual alarm, driven by the digital output point of the intelligent control unit, and issues an alarm when the system determines that manual intervention is required.

[0033] Please see Figure 2In this system, the material to be processed is conveyed to the high-frequency screen via a front-end belt, where it is screened. The front-end belt is controlled by a three-phase motor, and the high-frequency screen's operating frequency is controlled by a frequency converter. During normal operation, the motor operates at a set frequency, and the current fluctuates within a normal range, with a relatively constant average value. When material blockage occurs, the amount of material on the high-frequency screen increases, causing excessive load on the motor and an increase in the average motor current.

[0034] The system operates in two main phases: the model building phase and the online operation phase.

[0035] During the model building phase, the system conducts a series of tests to obtain current and frequency data corresponding to the stable and efficient operation of the high-frequency screen under different feed rates. It then uses the KNN regression algorithm to train a prediction model that can predict the normal current value of the high-frequency screen based on the current current of the upstream belt (representing the feed rate) and the operating frequency of the high-frequency screen.

[0036] During online operation, the system collects the current of the upstream belt and the actual current and frequency of the high-frequency screen in real time. First, based on the upstream belt current, a basic operating frequency is determined by interpolation, and the high-frequency screen is controlled to operate at this frequency to achieve efficient screening. Then, the current operating conditions (upstream belt current and actual operating frequency) are input into the pre-trained KNN model to obtain a predicted current value. This predicted value is compared with the actual current value of the high-frequency screen in real time. When the actual current is significantly higher than the predicted current, it indicates that material has accumulated on the screen and the load has increased, i.e., material blockage has occurred. At this time, the system will automatically take action to increase the vibration intensity by significantly increasing the operating frequency of the high-frequency screen, in an attempt to shake and discharge the blocked material. If the actual current returns to the normal range within the set time, it means that the blockage has been successfully cleared, and the system automatically returns to the normal operating frequency. If it does not return within the time limit, it is judged as severe material blockage requiring manual cleaning, and an alarm is triggered.

[0037] Furthermore, the workflow of this invention will be described as follows: Please see Figure 3 In this embodiment, the first step is to obtain the training dataset and generate a high-frequency screen current prediction model. Specifically: Ensure the high-frequency screen and upstream belt system are in normal mechanical and electrical condition. Pre-program a data acquisition sequence in the intelligent control unit, starting from zero and gradually and smoothly increasing the feed rate of the upstream belt until the system's maximum design throughput is reached. During this process, control the feed rate to form several stable steps. After each feed rate step stabilizes, the intelligent control unit synchronously acquires and records a set of data. This set of data includes: The stable current value of the preceding belt motor is denoted as: , where n is the total number of data points; The operating frequency of the high-frequency screen inverter that ensures good screening effect at the current feed rate is denoted as: ; The corresponding current of the high-frequency screen is denoted as .

[0038] The collected data is organized into a standard machine learning format. The combination of [previous belt current, high-frequency screen frequency] is used as the input feature X, and the corresponding high-frequency screen current is used as the prediction target Y. The training dataset contains X data. The Y data in the training set are .

[0039] Since the current of the preceding belt and the frequency of the high-frequency screen may differ significantly in both numerical value and dimension, it is necessary to normalize the feature X in order to eliminate the influence of dimensions and improve the convergence speed and prediction accuracy of the KNN algorithm. This invention employs a normalization method to map the data to the interval [0, 1]. For each dimension of the feature X data, the normalization formula is: ; in, and These are the minimum and maximum values ​​of the front-end belt current in the training set, respectively. Similarly, the frequency-dimensional features are normalized using the following formula: ; The normalized feature dataset is denoted as , Similarly.

[0040] Then, the KNN model is trained. This algorithm, for a given sample to be predicted, finds the K closest known samples in the feature space, and then uses the average of the target values ​​of these K samples as the predicted value for the sample to be predicted. The training process essentially involves training the normalized dataset... and These data are stored to form a model database. During training, a key parameter K also needs to be determined. The choice of K value has a significant impact on the model's smoothness and prediction accuracy.

[0041] If the K value is too small, the model is sensitive to noise and prone to overfitting; if the K value is too large, the model is too smooth and may ignore local features of the data. This invention preferably uses a K value between 3 and 10, and preferably an odd number, to avoid ties. In practice, the optimal K value can be selected through cross-validation based on the dataset size. For example, when the number of data points n is small (e.g., <20), 3 or 5 can be selected; when the number of data points is large, 7 or 9 can be selected. After training, a high-frequency screening current prediction model is generated, which is essentially the stored normalized dataset. , With the normalization parameter and K value used.

[0042] Please see Figure 4 In this embodiment, the second step involves using the established model to achieve efficient operation and intelligent anti-clogging of the high-frequency screen. Specifically: After the system starts up, the intelligent control unit collects the current value of the upstream belt motor in real time. Output current value of high frequency screen inverter and current operating frequency .

[0043] To achieve energy-efficient operation by automatically adjusting the frequency based on the incoming material volume, the system uses interpolation to dynamically obtain the recommended operating frequency from the model training data. First, in the raw (unnormalized) training data, the current of the upstream belt is found. In sequence The interval in which the given information is located, i.e., finding the index m, such that... Obtain the optimal operating frequency corresponding to the two endpoints of the interval. and Then, the current feed rate is calculated using a linear interpolation formula. Recommended high-frequency screen operating control frequency The formula is: ; This formula ensures that the frequency changes continuously and smoothly with the feed rate. To ensure a margin in the screening effect, the final frequency setpoint output by the intelligent control unit to the frequency converter can be set within a certain range. Based on this, a small increment Δf is added, where Δf ranges from 0.5Hz to 2Hz. That is, the high-frequency screen is controlled to... = It operates at a frequency of +Δf.

[0044] The system then collects the current value of the front-end belt in real time. With the current operating frequency of the high-frequency screen Combined into feature vectors [ , The vector is then normalized using the normalization parameters determined during the model training phase to obtain the query sample. The sample is then input into a pre-trained KNN regression model. The model calculates the Euclidean distance in the normalized feature space, finds the K nearest neighbors, and uses the average of the normalized current values ​​of the high-frequency screen corresponding to these neighbors as the prediction output. After inverse normalization, the predicted current value of the high-frequency screen under the current operating conditions is obtained. The system continuously calculates the actual operating current value of the high-frequency screen. With prediction absolute deviation between | - | When the deviation continues to exceed the preset threshold If a set delay (e.g., 2-5 seconds) is maintained, it is determined that the high-frequency screen is blocked, thereby triggering the subsequent automatic unblocking process.

[0045] When the system detects a material blockage, the intelligent control unit immediately sends a command to the high-frequency screen inverter to increase the operating frequency from the current value, for example, initially at the current operating frequency. Increasing the frequency by 5Hz significantly enhances the vibration intensity of the screen body, helping to disperse accumulated materials. Simultaneously, a timer with a preset duration of 10 to 30 seconds is activated. During this time, the system continuously monitors the actual current of the high-frequency screen. If the actual current It showed a significant decline and dropped to near the predicted value. If the frequency is within the normal range, it indicates that the blockage has been cleared. At this point, immediately restore the operating frequency of the high-frequency screen to its normal value. And exit the blockage clearing process; if the actual current after the timer times out If the blockage does not return to the normal range, the automatic unblocking function is deemed to have failed. The intelligent control unit then activates the audible and visual alarm and displays information indicating blockage requiring manual intervention.

[0046] This invention provides Embodiment 1: In this embodiment, following the first step of the above embodiment, during the production line commissioning phase, the opening of the raw material silo discharge valve is slowly adjusted in the central control room to simulate eight different feeding steps from 10% to 100% load. The PLC controller automatically records the upstream belt current, high-frequency screen frequency, and current under each step, obtaining a total of eight sets of data. The K value is selected as 5, and through normalization and KNN training, a current prediction model under this specific material and screen configuration is generated.

[0047] Then, in the second step, when the material intake is at 30% load, the system automatically adjusts the high-frequency screen frequency to 38.5Hz. When the material intake increases to 80% load, the frequency automatically rises to 46.2Hz. Compared to the original operating frequency of 50Hz, the energy-saving effect is more significant at low loads, with an average energy consumption reduction of about 18%. On one occasion, due to increased material moisture, the screen became slightly clogged, and the actual current of the high-frequency screen rose from the normal 21A to 24A, while the model predicted a current of 20.8A, exceeding the threshold deviation. The system immediately triggered automatic unclogging, and the frequency jumped from 42Hz to 47Hz. After about 8 seconds, the actual current dropped back to 21.5A, and the system immediately returned to 42Hz operation. The entire process required no human intervention, avoiding a potential downtime accident.

[0048] This invention provides Embodiment 2: This embodiment improves upon the fixed-frequency high-frequency screen in a traditional quartz sand production line, replacing it with this invention. While retaining the original high and low voltage electrical circuits, it adds a PLC, frequency converter, current transmitter, and touchscreen. The modified system then executes a model-building process. During model building, six sets of valid data were acquired, with a K value of 3 selected. After the modification, during low-production periods at night, the frequency automatically operates at around 40Hz, saving energy and significantly reducing equipment noise and mechanical wear.

[0049] Occasionally, fibrous impurities would mix into the materials on this production line, easily causing blockages. Previously, manual inspection and cleaning were required every two hours. After the system was put into operation, it successfully detected impurities in the early stages of accumulation (abnormal current increase of 2.5A) three times and cleared them successfully through automatic frequency increase (4Hz each time), effectively extending the manual cleaning cycle and improving production continuity. On one occasion, a large foreign object caused the automatic unblocking to fail, and the system accurately alarmed, allowing maintenance personnel to quickly arrive and handle the situation, preventing the equipment from running idle for an extended period or being damaged.

[0050] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A high-frequency screen high-efficiency operation and anti-blocking control method, characterized in that, It includes the following steps: S1 collects the current value of the upstream belt motor, the current value of the high-frequency screen inverter, and the operating frequency value in real time through the intelligent control unit; S2. By changing the feed rate of the upstream belt, multiple sets of upstream belt current values, corresponding high-frequency screen current values ​​and operating frequency values ​​are obtained to form a training dataset. The upstream belt current values ​​and operating frequency values ​​in the training dataset are normalized and trained using the KNN regression algorithm to obtain the high-frequency screen current prediction model. S3, During normal system operation, based on the real-time collected front-end belt current value, the operating control frequency of the high-frequency screen is determined from the training dataset using interpolation, and the high-frequency screen is controlled to operate at that frequency. S4 compares the actual current value of the high-frequency screen with the predicted current value of the high-frequency screen current prediction model in real time. When the deviation exceeds the preset threshold, the operating frequency of the high-frequency screen is increased to clear the blockage, and a timer is started. If the actual current value of the high-frequency screen falls back to the normal range within the timer period, the high-frequency screen is restored to the normal operating frequency; otherwise, an alarm device is triggered.

2. The high-frequency screen high-efficiency operation and anti-blocking control method according to claim 1, characterized in that: The intelligent control unit is a PLC, a microcontroller, or an embedded control system.

3. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: In step S2, the K value of the KNN regression algorithm is an odd number between 3 and 10.

4. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that, The interpolation method is a linear interpolation method, specifically including: finding two adjacent preceding belt current values ​​in the training dataset based on the current preceding belt current value. and and the corresponding operating frequency value and Through the formula: ； Calculate the operating control frequency of the high-frequency screen, where I is the current value of the upstream belt current.

5. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: In step S3, the high-frequency screen is controlled to operate at an incremental frequency that increases by one increment.

6. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: In step S4, when the deviation exceeds the preset threshold, the operating frequency of the high-frequency screen is increased to a fixed value or increased by a certain proportion.

7. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: The alarm device includes an audible and visual alarm, which is driven by an intelligent control unit.

8. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: The method displays processing data in real time on a screen, including the current value of the upstream belt, the current value of the high-frequency screen, the operating frequency value, and the alarm status.

9. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: In step S2, the feed rate of the front belt is changed by controlling the running speed of the front belt or the opening of the feed valve.

10. The method for high-efficiency operation and anti-clogging control of a high-frequency screen according to claim 1, characterized in that: The update cycle of the high-frequency screen current prediction model is periodic or adjusted based on operational data.

Images