Intelligent regulation and control method for dynamic pressure control of filter cake forming

Abstract

The invention provides an intelligent regulation and control method for dynamic pressure control of filter cake forming, and belongs to the technical field of intelligent regulation and control based on computer data processing. Firstly, the characteristics of a solid-liquid mixture are extracted by arranging a pressure sensor and a flow sensor, meanwhile, the state characteristics of the solid-liquid mixture are recognized through a camera, a mode recognition model of the filter cake state is built by integrating multi-dimensional information, and misjudgment of the model caused by a single parameter is avoided. After the filter cake forming stage is confirmed, dynamic modeling is conducted on time series data through the long-short-term memory network, nonlinear correlation between pressure fluctuation and the time dimension in the filter cake forming process is established, and the pressure control target value of the filter press is calculated. A pressure sensor is used for monitoring a pressure control true value of the filter press, and the pressure difference between the pressure control true value and a target value is used as the input of the improved segmented PID controller to finally realize the dynamic regulation and control of the pressure of the filter press. The method is wide in application scene, the detection accuracy in the filter cake forming stage is improved, and the working automation degree of the filter press is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent regulation and control based on computer data processing, and particularly relates to a filter cake forming dynamic pressure control intelligent regulation method. BACKGROUND

[0002] A filter press is a mechanical device that uses a special filter medium to apply pressure to the object to allow the liquid to be separated. It is a commonly used solid-liquid separation equipment and is widely used in many fields such as mining, metallurgy, food, pharmaceutical, sewage treatment, etc.

[0003] Taking the mine scene as an example, the filter press is mainly used for concentrate dewatering, tailings treatment and wastewater recovery in mine beneficiation, which directly affects the beneficiation efficiency and resource recovery rate. Concentrate dewatering reduces the water content of concentrate, thereby improving the smelting efficiency and avoiding the high cost of drying and storage. Dry tailings can realize solid-liquid separation and reduce environmental pollution. In the recycling of beneficiation wastewater, the filter press recovers water resources through solid-liquid separation, reducing the pressure of wastewater discharge.

[0004] The pressure of the filter press is not the higher the better, and needs to be adjusted comprehensively according to the material characteristics, process parameters and equipment performance. The pressure is a key parameter for the dewatering of the filter press, but too high or too low pressure will affect the effect. Insufficient pressure will cause the filter cake to have more residual water and incomplete dewatering; too high pressure may damage the pore structure of the filter cake, thereby hindering the discharge of water, and even damaging the equipment.

[0005] The filter cake forming is divided into three stages, and the control methods used for different stages are also different.

[0006] 1. Filter cake formation stage: Under the action of vacuum, water is extracted and solid particles are gradually accumulated to form filter cake. At this time, low pressure control is adopted to avoid filter cloth blockage by quickly feeding. The pressure is low and the feeding rate is fast, which helps to reduce unevenness.

[0007] 2. Filter cake dewatering stage: With the continuous action of vacuum, the water in the filter cake is further reduced to the required moisture content. At this time, the pressure should be gradually increased, the stable feeding rate should be maintained, and the water should be discharged through high pressure to form a dense filter cake. Real-time monitoring of pressure changes is required to avoid overpressure and equipment damage.

[0008] 3. Filter cake stripping stage: After dewatering, the filter cake is stripped from the filter medium and enters the next process. In this stage, high pressure steady state control is adopted, and the pressure sensor adjusts the feeding pump speed or valve opening to accurately control the pressure fluctuation range.

[0009] At present, the filter cake forming stage of the filter press mainly relies on manual observation, and the pressure control strategy is manually adjusted to perform the filter cake forming stage matching. Due to the long pressure filtration process and single operation process, the physiological fatigue and mental fatigue of the operator are the main factors affecting the production efficiency (such as improper operation causing pressure material and material overflow), so the traditional manual pressure control system needs to be optimized. SUMMARY

[0010] In view of the above problems, the present application designs a filter cake forming dynamic pressure control intelligent regulation method, comprising the following steps: S1, a first pressure sensor is arranged on the filtrate outlet pipeline to monitor the filtration resistance, a flow sensor is arranged on the straight pipe section of the feed to monitor the flow of the feed fluid, and a camera is installed above the filtrate water collection tank to identify the color of the filtrate; A second pressure sensor is installed at the position of the pressure gauge of the filter press for detecting the system pressure and transmitting the signal to the controller, and the real value of the pressure control of the filter press is obtained after data preprocessing; S2, the real-time obtained filtration resistance, feed flow and filtrate color are established into a multi-modal array, and after preprocessing, they are input into the stage recognition model to indicate the stage of filter cake forming according to the principle of relative majority, including filter cake forming stage, filter cake dewatering stage and filter cake stripping stage; S3, the filtration resistance, feed flow, filtrate color and real value of the pressure of the filter press after data preprocessing are taken as the input of the long short-term memory network (LSTM), the time series data is dynamically modeled, the nonlinear relationship between the pressure fluctuation and the time dimension in the filter cake forming process is established, and the target value of the pressure control of the filter press is calculated; S4, according to the different characteristics of the filter cake forming stage, the corresponding segmented PID controller is established, the pressure difference between the real value of the pressure control and the target value of the pressure control is taken as the input of the segmented PID controller, and the output of the segmented PID controller controls the working pressure of the filter press, so as to realize the dynamic regulation of the pressure of the filter press.

[0011] Preferably, the preprocessing process of the filtration resistance, feed flow, filtrate color and real value of the pressure of the filter press comprises: S11, an average value array with a length of 10 is established to smooth the sampling data, and the data after average value filtering is And ; The smooth filtering result of the feed fluid flow sensor at time t is represented as The smooth filtering result of the filtrate color camera at time t is represented as The smooth filtering result of the real value of the pressure control of the filter press at time t is represented as The smooth filtering result of the real value of the pressure control of the filter press at time t is represented as The smooth filtering result of the real value of the pressure control of the filter press at time t is represented as The smooth filtering result of the real value of the pressure control of the filter press at time t is represented as The smooth filtering result of the real value of the pressure control of the filter press at time t is represented as Smooth filtering processing result of filtering pressure data (filtrate outlet pipeline first pressure sensor monitoring filtering resistance pressure sensor data) at time; S12, filtering pressure data, flow data and pressure control real value data are distributed to 0-1, normalized to the same coordinate axis; obtain And ; S13, the image data executed by the average value filter is executed HSL transformation, first, R, G, B three components are normalized to 【0, 1】, the hue H, color saturation S and intensity component L are calculated; the range of H component is 【0, 360】, and the range of S and L component is 【0, 1】.

[0012] Preferably, the specific data processing process of the stage recognition model in S2 is:

[0013] By setting the first threshold value and the second threshold value, the relationship between the filtering resistance and the filter cake formation, dehydration and peeling stage is defined, when the filtering resistance input value is less than the first threshold value, it is defined as being in the filter cake formation stage; when the filtering resistance input value is greater than or equal to the first threshold value and less than or equal to the second threshold value, it is defined as being in the filter cake dehydration stage; when the filtering resistance input value is greater than the second threshold value, it is defined as being in the filter cake peeling stage; By setting the third threshold value and the fourth threshold value, the relationship between the feed fluid flow and the filter cake formation, dehydration and peeling stage is defined, when the feed fluid flow input value is less than the third threshold value, it is defined as being in the filter cake formation stage; when the feed fluid flow input value is greater than or equal to the third threshold value and less than or equal to the fourth threshold value, it is defined as being in the filter cake dehydration stage; when the feed fluid flow input value is greater than the fourth threshold value, it is defined as being in the filter cake peeling stage; After the filtrate color image data is executed HSL transformation, the first judgment condition group and the second judgment condition group are set based on the values of S component and L component, so as to define the relationship between the filtrate color and the filter cake formation, dehydration and peeling stage, the constant threshold value in the first judgment condition group and the second judgment condition group is taken from the empirical value; when the filtrate color meets the first judgment condition group, it is defined as being in the filter cake formation stage; when the filtrate color meets the second judgment condition group, it is defined as being in the filter cake peeling stage; when the filtrate color does not meet the first judgment condition group and does not meet the second judgment condition group, it is defined as being in the filter cake dehydration stage.

[0014] Preferably, based on the preprocessed filtering pressure data : If At this time, it is in the filter cake formation stage; If At this time, it is in the filter cake dehydration stage; If at this time is in the filter cake peeling stage.

[0015] Preferably, based on the pretreated flow data : If at this time is in the filter cake forming stage; If at this time is in the filter cake dewatering stage; If at this time is in the filter cake peeling stage.

[0016] Preferably, based on the converted S component and L component: If at least one of the following three conditions is met, it is determined that at this time is in the filter cake forming stage: ; ) ; ) If at least one of the following three conditions is met, it is determined that at this time is in the filter cake peeling stage: ; ) ; ) ; If the above two conditions are not met at the same time, it is determined that at this time is in the filter cake dewatering state.

[0017] Preferably, the S3 is specifically: The data pretreated structured data filtering resistance , feed flow , filtrate color , and are defined as the input parameter matrix of the LSTM model; the expressions and weight parameter matrices of the system forgetting gate, input gate, and output gate are set to meet the preconditions of the LSTM model calculation, establish a nonlinear relationship between pressure fluctuations and time dimension in the filter cake forming process, calculate the pressure filter pressure control target value, and output the results of the LSTM model .

[0018] Preferably, in the S4, the difference between the actual value of the pressure filter pressure control and the pressure control calculation value output by the LSTM is taken as the input of the positional ID control algorithm, and the output of the PID algorithm model is taken as the given value of the pressure filter pressure control; if it is in the filter cake forming stage, a PID controller is used, if it is in the filter cake dewatering stage, an incomplete differential PID controller is selected, and if it is in the filter cake peeling stage, a PI controller is selected; the actual position of the corresponding actuator is directly output by controlling the cumulative error.

[0019] Preferably, the S4 specific data processing process is: The error at time t is denoted as e(t), and the calculation expression is as follows: ; If it is in the filter cake forming stage, a PID controller is adopted, and the expression is as follows: ; In the formula, denotes the proportional coefficient of the PID controller; denotes the integral coefficient of the PID controller, denotes the differential coefficient of the PID controller; If it is in the filter cake dewatering stage, an incomplete differential PID controller is selected: A first-order inertia link is connected in series after the output of the PID controller to form an incomplete differential PID controller, u1(t) denotes the output of the PID controller, and the transfer function of the first-order inertia link D(s) is as follows: ; In the formula, is the step response coefficient of the first-order inertia link, and the expression of the incomplete differential controller is as follows: ; In the formula, , denotes the calculation result, denotes the output result of the PID controller; If it is in the filter cake peeling stage, a PI controller is selected, and the expression is as follows: ; In the formula, denotes the proportional coefficient of the PI controller; denotes the differential coefficient of the PI controller.

[0020] Compared with the prior art, the innovation points of the present application include: (1) Mode recognition of the filter cake stage is realized by multi-modal data fusion analysis: the characteristics of the solid-liquid mixture are extracted in combination with sensor and image data, and a mode recognition model of the filter cake state is constructed by comprehensively constructing multi-dimensional information; (2) A deep model is constructed to predict the pressure control target value: the pressure control target value of the filter press is predicted in a deep modeling manner; the filter resistance, feed flow, filtrate color and real value of the filter press pressure in the structured data set after data preprocessing are taken as the input of the long short-term memory network (LSTM), the time series data is dynamically modeled, the nonlinear correlation between the pressure fluctuation and the time dimension in the filter cake forming process is established, and the pressure control target value of the filter press is calculated; (3) Establishing an adaptive pressure control strategy: according to the different characteristics of the three stages of filter cake forming, dewatering and peeling, an improved segmented PID control segmented control model is established, the pressure difference between the real value and the target value of pressure control is taken as the input of the improved segmented PID controller, and the output of the improved segmented PID controller controls the working pressure of the filter press to realize the dynamic regulation and control of the filter press pressure.

[0021] The beneficial effects brought by the innovation points of the application include: Improve the detection accuracy of the filter cake forming stage: the method can accurately identify the filter cake forming stage by performing multi-dimensional information comprehensive judgment through multi-modal data, avoiding the hysteresis of traditional manual judgment and reducing the misjudgment rate caused by long-time work mental fatigue; Improve the automation degree of the filter press: the adaptive pressure control strategy replaces the adjustment of the pressure control system after traditional manual judgment, and improves the production automation degree.

[0022] Wide application scenarios: the systematic judgment can be widely used in filter presses in mines, metallurgy, food, pharmaceutical and other scenes, and the method is an automatic identification of the physical properties of the input material, not a parameter optimization for a single device. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 The overall technical route flowchart of the application.

[0024] Figure 2 The filter cake forming stage and the controller selection relationship block diagram.

[0025] Figure 3 The block diagram of the incomplete differential controller.

[0026] Figure 4 The response curve diagram of the traditional controller to the step response signal.

[0027] Figure 5 The response curve diagram of the improved segmented PID controller to the step response signal.

[0028] Figure 6 The EMC interference simulation result diagram of the traditional PID controller.

[0029] Figure 7 The EMC interference simulation result diagram of the improved segmented PID controller. DETAILED DESCRIPTION

[0030] The overall process of the filter cake forming dynamic pressure control intelligent regulation and control method based on multi-modal detection and deep modeling is as shown in Figure 1 ​S1, construct a multi-modal multi-dimensional information structured dataset; monitor the filtration resistance by arranging a pressure sensor at the filtrate outlet pipeline, monitor the flow of the feed fluid by arranging a flow sensor at the straight pipe section of the feed, and identify the filtrate color by installing a camera above the filtrate water collection tank. The filtration resistance, feed flow, and filtrate color are used as raw data to establish a multi-modal array, which is pre-processed as a structured dataset to jointly indicate the pattern recognition of the filter cake formation state. A pressure sensor is installed at the position of the pressure gauge of the filter press to detect the system pressure and transmit the signal to the controller, and the real value of the filter press pressure control is obtained after data preprocessing; S2, establish a pattern recognition model according to multi-dimensional information combined with the characteristics of the filter cake formation stage of the filter press; the logical relationship between the filtration resistance, feed flow, and filtrate color in the structured dataset after data preprocessing and the formation, dehydration, and peeling stages of the filter cake formation is explained, and the stage of the filter cake formation is indicated according to the principle of relative majority; S3, predict the target value of the filter press pressure control in a deep modeling manner; the filtration resistance, feed flow, filtrate color, and filter press pressure real value in the structured dataset after data preprocessing are used as the input of the long short-term memory network (LSTM), the time series data is dynamically modeled, the nonlinear relationship between the pressure fluctuation and the time dimension in the filter cake formation process is established, and the target value of the filter press pressure control is calculated; S4, establish an adaptive pressure control strategy; according to the different characteristics of the formation, dehydration, and peeling stages of the filter cake formation, the corresponding improved segmented PID control segmented control model is established, the pressure difference between the real value and the target value of the pressure control is used as the input of the improved segmented PID controller, and the output of the improved segmented PID controller is used to control the working pressure of the filter press to realize the dynamic regulation and control of the filter press pressure.

[0031] The specific implementation process of the present application will be described in detail below in conjunction with specific embodiments.

[0032] I. Data acquisition and preprocessing In order to ensure the integrity of the perception information, the present application extracts multi-modal data to establish a multi-dimensional information structured dataset; the filtration resistance is monitored by a pressure sensor, the flow of the feed fluid is monitored by a flow sensor, the filtrate color is identified by a camera, and the real value of the filter press pressure control is monitored by a pressure sensor. The time series structured array with aligned time axis is established by the average value filtering and normalization method.

[0033] Firstly, the sampling interval of all perception devices (including input parameters pressure sensor-filter pressure, flow sensor, and camera; output parameter pressure sensor-filter press pressure control real value) is set to 500 ms, and the structured array composed of pressure sensor P, flow sensor Q, and camera image F is collected , This represents an array of input parameters. This represents the actual pressure control data for the filter press.

[0034] To balance the volatility of the acquired data, the sampled data undergoes smoothing filtering. Taking the filtering of time-series data acquired by a pressure sensor as an example, the specific process includes: (1) Let The pressure sensor value is constantly filtered. Collect and record In the array This represents the pressure sensor values ​​from the previous 10 moments. This represents the pressure sensor values ​​from the previous 9 time points. (2) After smoothing filtering The filter pressure sensor value at each step of the calculation process is... It satisfies the following formula. ; Similarly, we can obtain Therefore, the data after average filtering is , This represents the array of input parameters after average filtering. express The smoothing and filtering results of the feed fluid flow sensor at all times. express The smoothing filter result of the filtrate color camera at any given time. express The result of smoothing and filtering the actual pressure control value of the filter press at any time.

[0035] Pressure sensor that has undergone average filtering and flow sensor To implement a 0-1 distribution and normalize to the same coordinate axis, the specific process is as follows: (1) Let The pressure sensor value after undergoing average filtering is: ,Record In the array This represents the pressure sensor values ​​from the previous 10 time points. This represents the pressure sensor values ​​from the previous 9 time points. (2) Perform 0-1 distribution calculation to obtain the filter pressure pressure sensor data processing results Proceed to the next step, the formula is as follows. ; In the formula express The minimum value in the array; denotes the maximum value in the array.

[0036] Similarly, we get and . denotes the result of the 0-1 distribution after smoothing filtering processing of the fluid flow sensor at the moment. denotes the result of the 0-1 distribution after smoothing filtering processing of the filter press pressure control real value at the moment.

[0037] Perform HSL transformation on the image data after performing average value filtering. The image material captured by the camera is in RGB color format. Convert the RGB color space to HSL (hue, saturation, intensity) color space which is closer to the recognition effect of human eyes. The specific method is to first normalize R, G, B three components to 【0, 1】. The conversion relationship between hue H and RGB is as follows: ; Here The calculation relationship is as follows: ; The conversion relationship between saturation S and RGB is as follows: ; The conversion relationship between intensity component L and RGB is as follows: ; From the above formula, it is concluded that after normalizing R, G, B to 【0, 1】 interval, the range of H component is 【0, 360】, and the range of S and L components is 【0, 1】.

[0038] Based on the above data acquisition and processing process, the input parameter structured data set and the output parameter are obtained.

[0039] II. Stage recognition model data processing Filter cake forming is divided into three stages of formation, dehydration and stripping. At the initial stage of filter press operation, the filtrate flow is usually large and turbid. As the filter cake gradually forms, the filtration resistance will gradually increase, the filtrate flow will significantly decrease, and the color will gradually become clear, indicating that the filter cake has been preliminarily formed and is entering the pressing stage. In order to accurately identify the pattern, the pressure sensor reflects the filtration resistance; the flow sensor monitors the flow of the feed fluid; the camera identifies the color of the filtrate. Define the relationship between filtration resistance, feed fluid flow, filtrate color and filter cake formation, dehydration and stripping stage. Determine the stage of filter cake formation through the voting mechanism.

[0040] 1. Define the relationship between filtration resistance and the filter cake formation, dewatering, and peeling stages. Based on the principle that filtration resistance gradually increases as the filter cake gradually forms, the following judgment is made: like At this point, the filter cake is forming. like At this point, the filter cake is in the dehydration stage; like At this point, the filter cake is being peeled off.

[0041] 2. Define the relationship between feed fluid flow rate and the filter cake formation, dewatering, and peeling stages. Initially, the filtrate flow rate is typically high; as the filter cake gradually forms, the filtrate flow rate decreases significantly. The following criteria are applied: like At this point, the filter cake is forming. like At this point, the filter cake is in the dehydration stage; like At this point, the filter cake is being peeled off.

[0042] 3. Define the relationship between filtrate color and the stages of filter cake formation, dehydration, and peeling. Initially, the filtrate is turbid; as the filter cake gradually forms, the filtrate color gradually becomes clearer. Although there is no direct relationship between H, S, and L, based on their color distribution characteristics, the following relationship exists when dividing the color regions: When the lightness (L) component is extremely high or extremely low, the hue (H) will have no effect on color recognition. When the saturation S, which represents color saturation, is extremely low or even close to 0, the hue H will fluctuate within a certain range and cannot be stable; When the lightness value (L) is extremely high or extremely low, the color saturation (S) will have no effect on color recognition. The color is expressed by splitting it into three components: HSL. Since the color subdivision in the colored area is more complex, only the threshold range of black (completely cloudy, corresponding to the filter cake formation stage) and white (completely clear, corresponding to the filter cake peeling stage) in the non-colored area is judged. The constant data empirical value used as the judgment threshold is selected in the formula. If at least one of the following three conditions is met, it is determined that the filter cake is currently in the formation stage. ; )&( ; )&( ; If at least one of the following three conditions is met, it is determined that the filter cake is currently in the peeling stage. ; )&( ; )&( ; If the above two conditions are not met at the same time, it is determined that the filter cake is in a dewatering state at this time.

[0043] 4. Determine the stage of filter cake formation by voting mechanism, adopt the principle of relative majority, and more than 2 votes in indicate the stage of filter cake formation.

[0044] III. Predicting the pressure control target value of the filter press in a deep modeling manner The filtration resistance, feed flow rate, filtrate color and real value of the filter press pressure in the structured data set after data preprocessing are taken as the input of the long short-term memory network (LSTM), the nonlinear association between pressure fluctuation and time dimension in the filter cake formation process is established, and the pressure control target value of the filter press is calculated. Define the input parameter matrix of the LSTM model, set the expressions and weight parameter matrices of the system forgetting gate, input gate and output gate, and meet the preconditions for calculating the LSTM model. The specific method is as follows: Let the current time step be t, the input is , the hidden state at the last time step is , and the cell state is , then the calculation process of LSTM is as follows: The expression of the forgetting gate is as follows: ; In the formula, represents the weight matrix of the forgetting gate; represents the bias term of the forgetting gate, represents the Sigmoid function, .

[0045] The input gate is used to control the generation and storage of new information, and the activation value of the input gate is as follows: ; In the formula, represents the output of the input gate; represents the weight matrix of the input gate; represents the bias term of the input gate, The expression of the candidate new information is as follows: ; In the formula, represents the next state predicted by the cell algorithm, represents the weight matrix of the cell update; represents the bias term of the cell update, The cell state is updated by combining the results of the forgetting gate and the input gate, and the expression is as follows: wherein denotes the cell update output; denotes element-wise multiplication (Hadamard product).

[0046] The expression of the output gate is as follows: wherein denotes the output of the output gate; denotes the weight matrix of the output gate; denotes the bias term of the output gate, The expression of the hidden state update is as follows: The calculated filter press pressure control target value is the output result of the LSTM model .

[0047] Four, establish adaptive pressure control strategy The filter press pressure control strategy is related to the filter cake forming stage. The filter cake forming stage requires low pressure fast response control, and a PID controller is selected; the filter cake dewatering stage requires pressure step-up control, and an incomplete differential PID controller is selected; the filter cake stripping stage requires high pressure steady state control, and a PI controller is selected. Accordingly, an improved segmented PID control algorithm is established, as shown in Figure 2 .

[0048] The position type PID controller is selected, which controls through cumulative error and can realize high control precision and stability. The actual position of the corresponding actuator is directly output without manual / automatic switching operation, and the process is simple.

[0049] The difference between the real value of the filter press pressure control and the pressure control calculation value output by the LSTM is taken as the input of the position type PID control algorithm, and the output of the PID algorithm model is taken as the given value of the filter press pressure control.

[0050] The error at time t is denoted as e(t), and the calculation expression is as follows, If it is in the filter cake forming stage, a PID controller is adopted, and the expression is as follows, wherein denotes the proportional coefficient of the PID controller; denotes the integral coefficient of the PID controller, denotes the differential coefficient of the PID controller; ​​​​​If in the filter cake dewatering stage, select the incomplete differential PID controller, as shown in Figure 3

[0051] PID controller output in series with a first-order inertia link, constitute an incomplete differential PID controller, u1(t) represents the output of the PID controller, the transfer function of the first-order inertia link D(s) as follows, In the formula The step response coefficient of the first-order inertia link, the expression of the incomplete differential controller as follows, In the formula , The calculation results, The output results of the PID controller.

[0052] If in the filter cake stripping stage, select the PI controller, expression as follows, In the formula The proportional coefficient of the PI controller; The differential coefficient of the PI controller.

[0053] Five, experimental analysis The improved segmented PID is simulated on Matlab, as shown in Figure 4 and Figure 5 The experimental results show that under the same step signal superimposed with the same random signal interference as input, the improved segmented PID controller and the traditional PID controller have obvious optimization in control effect. The step response of the PID controller is relatively volatile in the first 1-3 seconds, and although it is relatively stable, it still has small fluctuations; while the step response of the improved segmented PID controller is obviously more stable, with fast response speed, superior stability and anti-interference performance.

[0054] As shown in Table 1, the PID and improved segmented PID parameter configuration and simulation results, the experimental results show that the improved segmented PID controller compared with the traditional PID controller, overshoot, adjustment time, response time index performance improvement is obvious: Table 1 Parameter configuration and simulation results

[0055] As Figure 6 and Figure 7 ​​​​As shown in the figure, wherein the purple line represents input, the yellow line represents noise, the green line represents output, the horizontal coordinate represents time, and the vertical coordinate represents EMC noise response; the experimental results show that compared with the traditional PID controller, the improved segmented PID controller can quickly restore the balance state after the controller is disturbed by electromagnetic interference, and has good anti-interference and robustness.

[0056] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Those skilled in the art can make various modifications and changes to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0057] Although the specific embodiments of the present application are described above, they are not intended to limit the scope of protection of the present application. Those skilled in the art should understand that various modifications or changes made on the basis of the technical solutions of the present application without creative labor are still within the protection scope of the present application.

Claims

1. A method for intelligent control of dynamic pressure during filter cake forming, characterized in that, Includes the following steps: S1, by installing a first pressure sensor in the filtrate outlet pipe to monitor the filtration resistance, installing a flow sensor in the feed straight pipe section to monitor the flow rate of the feed fluid, and installing a camera above the filtrate water collection tank to identify the filtrate color; A second pressure sensor is installed at the location of the pressure gauge built into the filter press to detect the system pressure and transmit the signal to the controller. After data preprocessing, the true pressure control value of the filter press is obtained. S2, the real-time obtained filtration resistance, feed flow rate and filtrate color are used to establish a multimodal array, which is then preprocessed and input into the stage identification model to indicate the stage of filter cake formation based on the principle of relative majority, including the filter cake formation stage, filter cake dehydration stage and filter cake peeling stage. S3 uses the pre-processed filtration resistance, feed flow rate, filtrate color, and actual filter press pressure as inputs to a Long Short-Term Memory (LSTM) network to dynamically model the time series data, establish a nonlinear relationship between pressure fluctuations and the time dimension during the filter cake forming process, and calculate the target value for filter press pressure control. S4. Based on the different characteristics of the filter cake forming stage, a corresponding segmented PID controller is established. The pressure difference between the actual pressure control value and the pressure control target value is used as the input of the segmented PID controller. The output of the segmented PID controller controls the working pressure of the filter press, thereby realizing the dynamic regulation of the filter press pressure.

2. The intelligent control method for dynamic pressure control in filter cake forming as described in claim 1, characterized in that, The pretreatment process for filtration resistance, feed flow rate, filtrate color, and actual filter press pressure includes: S11, Create an average value array of length 10 to perform smoothing filtering on the sampled data. The data after average filtering is: and ; express The smoothing and filtering results of the feed fluid flow sensor at all times. express The smoothing filter result of the filtrate color camera at any given time. express The result of smoothing and filtering the actual pressure control value of the filter press at any time; S12, distribute the filter pressure data, flow data, and actual pressure control data into a 0-1 distribution and normalize them to the same coordinate axis; thus obtaining... and ; S13. Perform HSL transformation on the image data that has undergone average filtering. First, normalize the three components R, G, and B to the range of [0, 1]. Calculate the hue H, color saturation S, and intensity component L. The range of the H component is [0, 360], while the ranges of the S and L components are [0, 1].

3. The intelligent control method for dynamic pressure regulation of filter cake forming as described in claim 1, characterized in that: The specific data processing procedure for the stage identification model in S2 is as follows: By setting a first threshold and a second threshold, the relationship between filtration resistance and the filter cake formation, dewatering, and peeling stages is defined. When the input value of filtration resistance is less than the first threshold, it is defined as being in the filter cake formation stage; when the input value of filtration resistance is greater than or equal to the first threshold and less than or equal to the second threshold, it is defined as being in the filter cake dewatering stage; when the input value of filtration resistance is greater than the second threshold, it is defined as being in the filter cake peeling stage. By setting a third and a fourth threshold, the relationship between the feed fluid flow rate and the filter cake formation, dewatering, and peeling stages is defined. When the feed fluid flow rate input value is less than the third threshold, it is defined as being in the filter cake formation stage; when the feed fluid flow rate input value is greater than or equal to the third threshold and less than or equal to the fourth threshold, it is defined as being in the filter cake dewatering stage. When the feed fluid flow rate input value is greater than the fourth threshold, it is defined as being in the filter cake stripping stage; After performing HSL transformation on the filtrate color image data, a first set of judgment conditions and a second set of judgment conditions are set based on the values ​​of the S and L components, thereby defining the relationship between filtrate color and the filter cake formation, dehydration, and peeling stages. The constant thresholds in the first set of judgment conditions and the second set of judgment conditions are all taken from empirical values. When the filtrate color meets the first set of judgment conditions, it is defined as being in the filter cake formation stage; when the filtrate color meets the second set of judgment conditions, it is defined as being in the filter cake peeling stage; when the filtrate color does not meet either the first set of judgment conditions or the second set of judgment conditions, it is defined as being in the filter cake dehydration stage.

4. The intelligent control method for dynamic pressure control of filter cake forming as described in claim 3, characterized in that: Based on preprocessed filtration pressure data : like At this point, the filter cake is forming. like At this point, the filter cake is in the dehydration stage; like At this point, the filter cake is being peeled off.

5. The intelligent control method for dynamic pressure control of filter cake forming as described in claim 3, characterized in that: Based on preprocessed traffic data : like At this point, the filter cake is forming. like At this point, the filter cake is in the dehydration stage; like At this point, the filter cake is being peeled off.

6. The intelligent control method for dynamic pressure control of filter cake forming as described in claim 3, characterized in that: Based on the transformed S and L components: If at least one of the following three conditions is met, it is determined that the filter cake is currently in the forming stage: ; )&( ; )&( If at least one of the following three conditions is met, it is determined that the filter cake is currently in the peeling stage: ； )&( ； )&( ； If neither of the above two judgment conditions is met at the same time, it is determined that the filter cake is in a dehydration state.

7. The intelligent control method for dynamic pressure control of filter cake forming as described in claim 1, characterized in that: Specifically, S3 is: Filtering resistance from preprocessed structured data Feed flow rate Filtrate color as well as , is defined as the input parameter matrix of the LSTM model; The expressions for the system's forget gate, input gate, and output gate, as well as the weight parameter matrices, are defined to satisfy the preconditions for LSTM model calculation. A nonlinear relationship between pressure fluctuations and the time dimension during filter cake forming is established, and the target pressure control value for the filter press is calculated to provide the output results for the LSTM model. .

8. The intelligent control method for dynamic pressure control of filter cake forming as described in claim 1, characterized in that: In step S4, the filter press pressure is controlled to the actual value. Pressure control calculation value output by LSTM The difference is used as the input of the position-type ID control algorithm, and the output of the PID algorithm model is used as the given value for the pressure control of the filter press. If it is in the filter cake forming stage, a PID controller is used; if it is in the filter cake dewatering stage, an incomplete derivative PID controller is selected; if it is in the filter cake peeling stage, a PI controller is selected. Control is achieved by accumulating errors, and the actual position of the corresponding actuator is directly output.

Images