A PID control system and method for liquid level of magnetic separator

Abstract

The invention relates to the field of liquid level control of a vertical ring pulsating high gradient magnetic separator, and in particular to a magnetic separator liquid level PID control system and method. The specific method comprises: collecting real-time slurry liquid level data in a magnetic separator by means of an ultrasonic liquid level sensor; extracting a first dynamic target liquid level value set by a host computer of the magnetic separator, and calculating a first liquid level deviation value; establishing a BP neural network model, and training the BP neural network model; outputting liquid level dynamic adjustment parameters by means of the BP neural network model, and adjusting the switching amount of a pulsating bucket valve according to the liquid level dynamic adjustment parameters; and continuously monitoring the slurry liquid level by means of a feedback judgment strategy. The invention solves the problem in the prior art that, when the liquid level is lower than an optimal sorting liquid level, a sorting ring cannot be completely immersed in the slurry, its pulsating stroke mechanism cannot fully function, magnetic particles cannot be fully captured by a magnetic medium in the sorting ring, and materials cannot be effectively sorted.

Description

Technical Field

[0001] The invention relates to the field of liquid level control of a vertical ring pulsating high gradient magnetic separator, and in particular to a liquid level PID control system and method for a magnetic separator. Background Art

[0002] In the separation process of the vertical ring high gradient magnetic separator, the separation liquid level is an important parameter that affects the mineral separation index. When the liquid level is lower than the optimal separation liquid level, the separation ring cannot be completely immersed in the slurry, its pulsation stroke mechanism cannot fully function, and the magnetic particles cannot be fully captured by the magnetic medium in the separation ring, resulting in the inability to effectively separate the materials, which greatly affects the separation index. When the liquid level is higher than the optimal separation liquid level, the magnetic separator will have a phenomenon of ore bubbling, and the original ore slurry will flow into the tailings, resulting in a high tailings grade. At the same time, the separation magnetic medium will be saturated, resulting in the magnetic minerals not being fully separated and flowing out with the tailings, not only reducing the concentrate grade, but also increasing the tailings grade, thereby reducing the mineral separation recovery rate. However, the existing vertical ring high gradient magnetic separator liquid level control mainly relies on manual adjustment, the operator's labor intensity is high, and the operation timeliness is poor. At present, most magnetic separator liquid level controls realize liquid level control in mechanical structure and electrical logic control, but the system has hysteresis and delay, and the control accuracy is relatively low. With the development of intelligent algorithms, intelligent upgrades to magnetic separator liquid level control have become possible.

[0003] In the existing disclosed invention technologies, for example, the patent with publication number CN114515647A discloses a control method and equipment for a magnetic separator system, the method comprising: obtaining a liquid level detection result and a density detection result of a combined medium barrel; obtaining information of an inflow medium flowing into the combined medium barrel and information of an outflow medium flowing out of the combined medium barrel; and adjusting the amount of inflow medium and the density of inflow medium of the combined medium barrel according to the liquid level detection result, the density detection result, the inflow medium information and the outflow medium information, so that the liquid level of the medium in the combined medium barrel reaches a set liquid level value, and the density of the medium in the combined medium barrel reaches a set density value.

[0004] The above patent has a relatively complicated adjustment process when considering the changes in the information of the inflow medium and the outflow medium, and has a poor real-time response capability to the changes in the information of the inflow medium and the outflow medium, and has the problems described in the background technology. Summary of the invention

[0005] The technical problem to be solved by the present invention is that in the prior art, when the liquid level is lower than the optimal sorting liquid level, the sorting ring cannot be completely immersed in the slurry, its pulsating stroke mechanism cannot fully function, the magnetic particles cannot be fully captured by the magnetic medium in the sorting ring, and the material cannot be effectively sorted. A magnetic separator liquid level PID control system and method are proposed.

[0006] In order to achieve the above object, the technical solution of a magnetic separator liquid level PID control method of the present invention comprises the following steps:

[0007] S1: Collect real-time slurry level data in the magnetic separator through ultrasonic level sensors;

[0008] S2: extracting the first dynamic target liquid level value set by the upper computer of the magnetic separator, importing the real-time slurry liquid level data and the first dynamic target liquid level value into the slurry liquid level deviation value calculation strategy to calculate the first liquid level deviation value;

[0009] S3: Establish a BP neural network model, and use the particle swarm algorithm to perform global optimization on the BP neural network to complete the training of the BP neural network model;

[0010] S4: inputting the first liquid level deviation value into the trained BP neural network model, outputting the liquid level dynamic adjustment parameter through the BP neural network model, and adjusting the switching amount of the pulsating bucket valve according to the liquid level dynamic adjustment parameter;

[0011] S5: The slurry level is continuously monitored through the feedback judgment strategy, and the pulsation bucket valve switching value is adjusted according to the real-time feedback information to complete the adjustment and correction of the slurry level.

[0012] Specifically, S2 includes the following specific steps:

[0013] S21: extracting real-time slurry level data in the magnetic separator according to S1;

[0014] S22: extracting the first dynamic target liquid level value set by the upper computer of the magnetic separator;

[0015] S23: input the real-time slurry level data and the first dynamic target level value into the PLC system, and import the real-time slurry level data and the first dynamic target level value into the slurry level deviation value calculation strategy through the PLC system to calculate the level deviation value. The slurry level deviation value calculation strategy is as follows:

[0016] ;

[0017] in, is the first liquid level deviation value;

[0018] They are respectively the real-time slurry level data and the first dynamic target level value of the magnetic separator at the monitoring time point t.

[0019] Specifically, S3 includes the following specific steps:

[0020] S31: Establishing a BP neural network model, assigning an initial position and an initial velocity to a particle swarm through a particle swarm algorithm, and mapping the initial position and initial velocity of the particle swarm to the weights of each node of the BP neural network, wherein the weights of each node include: weight and bias;

[0021] S32: using the training data to perform multiple iterations of training on the BP neural network, and calculating the fitness of each particle, wherein the calculation strategy of the fitness of each particle is: calculating the mean square error between the predicted acceleration and the actual acceleration of the particle;

[0022] S33: updating the individual historical optimal weight and the global optimal weight of each particle, and synchronously updating the speed and position of the particle, wherein the individual historical optimal weight is the optimal weight achieved by the particle in the previous iteration, and the global optimal weight is the optimal weight achieved in the entire particle swarm;

[0023] S34: Calculate and extract the mean fitness data of the particles, and compare the mean fitness data of the particles with a preset error threshold. When the mean fitness data of the particles is less than or equal to the preset error threshold, the iterative training of the BP neural network is completed. Otherwise, return to step S32 to perform the next round of iterative training and recalculate the fitness data of each particle.

[0024] Specifically, in S4, the output of the liquid level dynamic adjustment parameter specifically includes:

[0025] S41: inputting the first liquid level deviation value into the trained BP neural network model;

[0026] S42: Outputting the liquid level dynamic adjustment parameter through the BP neural network model, wherein the liquid level dynamic adjustment parameter The details are as follows:

[0027] ;

[0028] in, are respectively the proportional coefficient, integral coefficient and differential coefficient of the first liquid level deviation value;

[0029] S43: adjusting the switching value of the pulsating bucket valve according to the liquid level dynamic adjustment parameter.

[0030] Specifically, in S5, the continuous monitoring of the slurry level through the feedback judgment strategy specifically includes:

[0031] S51: Return to step S1, and extract the real-time slurry level data in the magnetic separator again through the ultrasonic level sensor;

[0032] S52: synchronously extracting a second dynamic target liquid level value set by the upper computer of the magnetic separator;

[0033] S53: Importing the real-time slurry level data and the second dynamic target level value into the slurry level deviation value calculation strategy to calculate the second level deviation value;

[0034] S54: Continuously monitor the slurry level through feedback judgment strategy;

[0035] When the second liquid level deviation value is greater than or equal to the first liquid level deviation value, it is determined that the closed-loop control of the magnetic separator fails, and a manual liquid level control operation prompt is issued to the vertical ring high gradient magnetic separator operator;

[0036] When the second liquid level deviation value is less than the preset liquid level deviation threshold, it is determined that the closed-loop control of the magnetic separator is operating normally, and steps S3-S4 are continued to be executed to perform circulating liquid level control on the magnetic separator;

[0037] When the second liquid level deviation value is zero, the liquid level control process of the magnetic separator is completed.

[0038] In addition, a magnetic separator liquid level PID control system of the present invention includes the following modules:

[0039] Ultrasonic liquid level sensor module, liquid level deviation value calculation module, neural network module, liquid level controller module, valve electric execution module;

[0040] The ultrasonic level sensor module monitors the real-time slurry level data in the magnetic separator through an ultrasonic level sensor;

[0041] The liquid level deviation value calculation module is used to extract the first dynamic target liquid level value set by the upper computer of the magnetic separator, and import the real-time slurry liquid level data and the first dynamic target liquid level value into the slurry liquid level deviation value calculation strategy to calculate the first liquid level deviation value;

[0042] The neural network module is used to establish a BP neural network model, and perform global optimization on the initial weight, threshold and output weight of the BP neural network through a particle swarm algorithm to complete the training of the BP neural network model;

[0043] The liquid level controller module is used to input the first liquid level deviation value into the trained BP neural network model, output the liquid level dynamic adjustment parameter through the BP neural network model, and adjust the pulsating bucket valve switch amount according to the liquid level dynamic adjustment parameter;

[0044] The valve electric execution module continuously monitors the slurry level through a feedback judgment strategy, and adjusts the pulsation bucket valve switching amount according to real-time feedback information to complete the adjustment and correction of the slurry level.

[0045] Specifically, the BP neural network model in the neural network module specifically includes: an input layer, a hidden layer and an output layer; wherein the output layer is provided with 4 neurons, corresponding to the bias term, the first dynamic target liquid level value, the real-time slurry liquid level data and the first dynamic target liquid level value, the hidden layer is provided with 5 neurons, and the output of the output layer is ,in, is the PID control parameter;

[0046] The activation function of the hidden layer is a non-negative Sigmoid function; the BP neural network model uses the steepest descent method to correct the weight coefficient of the network, and the learning rate is set to 0.20 and the inertia coefficient is set to 0.05.

[0047] Specifically, the detection range of the ultrasonic liquid level sensor module is 65~600mm; the IP level of the ultrasonic liquid level sensor module is 67;

[0048] The valve electric execution module is used for automatic control of the switch valve of the pulsating bucket, outputs a 4-20mA analog signal to the electric actuator, and automatically adjusts the opening and closing degree of the valve according to the change of the liquid level.

[0049] Specifically, the liquid level controller module further includes a host device unit, and the host device unit is used to display real-time slurry level data and a liquid level curve.

[0050] A storage medium stores instructions, and when a computer reads the instructions, the computer executes any one of the above-mentioned methods for controlling the liquid level PID of a magnetic separator.

[0051] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, any one of the above-mentioned methods for controlling the liquid level of a magnetic separator in PID is implemented.

[0052] Compared with the prior art, the technical effects of the present invention are as follows:

[0053] 1. In the present invention, the host computer displays the liquid level detection value and the liquid level curve in real time, and automatically adjusts the switching amount of the pulsating bucket valve according to the liquid level dynamic adjustment parameters output by the BP neural network model, so as to realize accurate control and adjustment of the slurry level.

[0054] 2. The control algorithm used in the present invention enables the system to have fast response speed, high adjustment accuracy and good stability.

[0055] 3. The device used in the present invention has the advantage of convenient liquid level control, which significantly improves the production efficiency, mineral processing index and automation level of the high gradient magnetic separator. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0057] Figure 1 It is a flow chart of a magnetic separator liquid level PID control method of the present invention;

[0058] Figure 2 It is a structural schematic diagram of a magnetic separator liquid level PID control system of the present invention;

[0059] Figure 3 It is a flow chart of a PSO-BP-PID control algorithm of the present invention. DETAILED DESCRIPTION

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Example 1: Figure 1 , Figure 3 As shown, a PID control method for a liquid level of a magnetic separator according to an embodiment of the present invention is as follows: Figure 1 As shown, the specific steps include:

[0062] S1: Collect real-time slurry level data in the magnetic separator through ultrasonic level sensors;

[0063] S2: extracting the first dynamic target liquid level value set by the upper computer of the magnetic separator, importing the real-time slurry liquid level data and the first dynamic target liquid level value into the slurry liquid level deviation value calculation strategy to calculate the first liquid level deviation value;

[0064] S2 includes the following specific steps:

[0065] S21: extracting real-time slurry level data in the magnetic separator according to S1;

[0066] S22: extracting the first dynamic target liquid level value set by the upper computer of the magnetic separator;

[0067] S23: input the real-time slurry level data and the first dynamic target level value into the PLC system, and import the real-time slurry level data and the first dynamic target level value into the slurry level deviation value calculation strategy through the PLC system to calculate the level deviation value. The slurry level deviation value calculation strategy is as follows:

[0068] ;

[0069] in, is the first liquid level deviation value;

[0070] They are respectively the real-time slurry level data and the first dynamic target level value of the magnetic separator at the monitoring time point t.

[0071] S3: Establish a BP neural network model, and use the particle swarm algorithm to perform global optimization on the BP neural network to complete the training of the BP neural network model;

[0072] S3 includes the following specific steps:

[0073] S31: Establishing a BP neural network model, assigning an initial position and an initial velocity to a particle swarm through a particle swarm algorithm, and mapping the initial position and initial velocity of the particle swarm to the weights of each node of the BP neural network, wherein the weights of each node include: weight and bias;

[0074] S32: using the training data to perform multiple iterations of training on the BP neural network, and calculating the fitness of each particle, wherein the calculation strategy of the fitness of each particle is: calculating the mean square error between the predicted acceleration and the actual acceleration of the particle;

[0075] S33: updating the individual historical optimal weight and the global optimal weight of each particle, and synchronously updating the speed and position of the particle, wherein the individual historical optimal weight is the optimal weight achieved by the particle in the previous iteration, and the global optimal weight is the optimal weight achieved in the entire particle swarm;

[0076] S34: Calculate and extract the mean fitness data of the particles, and compare the mean fitness data of the particles with a preset error threshold. When the mean fitness data of the particles is less than or equal to the preset error threshold, the iterative training of the BP neural network is completed. Otherwise, return to step S32 to perform the next round of iterative training and recalculate the fitness data of each particle.

[0077] S4: inputting the first liquid level deviation value into the trained BP neural network model, outputting the liquid level dynamic adjustment parameter through the BP neural network model, and adjusting the switching amount of the pulsating bucket valve according to the liquid level dynamic adjustment parameter;

[0078] In S4, the output of the liquid level dynamic adjustment parameter specifically includes:

[0079] S41: inputting the first liquid level deviation value into the trained BP neural network model;

[0080] S42: Outputting the liquid level dynamic adjustment parameter through the BP neural network model, wherein the liquid level dynamic adjustment parameter The details are as follows:

[0081] ;

[0082] in, are respectively the proportional coefficient, integral coefficient and differential coefficient of the first liquid level deviation value;

[0083] S43: adjusting the switching value of the pulsating bucket valve according to the liquid level dynamic adjustment parameter.

[0084] S5: The slurry level is continuously monitored through the feedback judgment strategy, and the pulsation bucket valve switching value is adjusted according to the real-time feedback information to complete the adjustment and correction of the slurry level.

[0085] In S5, the continuous monitoring of the slurry level through the feedback judgment strategy specifically includes:

[0086] S51: Return to step S1, and extract the real-time slurry level data in the magnetic separator again through the ultrasonic level sensor;

[0087] S52: synchronously extracting a second dynamic target liquid level value set by the upper computer of the magnetic separator;

[0088] S53: Importing the real-time slurry level data and the second dynamic target level value into the slurry level deviation value calculation strategy to calculate the second level deviation value;

[0089] S54: Continuously monitor the slurry level through feedback judgment strategy;

[0090] When the second liquid level deviation value is greater than or equal to the first liquid level deviation value, it is determined that the closed-loop control of the magnetic separator fails, and a manual liquid level control operation prompt is issued to the vertical ring high gradient magnetic separator operator;

[0091] When the second liquid level deviation value is less than the preset liquid level deviation threshold, it is determined that the closed-loop control of the magnetic separator is operating normally, and steps S3-S4 are continued to be executed to perform circulating liquid level control on the magnetic separator;

[0092] When the second liquid level deviation value is zero, the liquid level control process of the magnetic separator is completed.

[0093] Example 2: Figure 2 As shown, a magnetic separator liquid level PID control system according to an embodiment of the present invention includes the following modules:

[0094] Ultrasonic liquid level sensor module, liquid level deviation value calculation module, neural network module, liquid level controller module, valve electric execution module;

[0095] The ultrasonic level sensor module monitors the real-time slurry level data in the magnetic separator through an ultrasonic level sensor;

[0096] The liquid level deviation value calculation module is used to extract the first dynamic target liquid level value set by the upper computer of the magnetic separator, and import the real-time slurry liquid level data and the first dynamic target liquid level value into the slurry liquid level deviation value calculation strategy to calculate the first liquid level deviation value;

[0097] The neural network module is used to establish a BP neural network model, and perform global optimization on the initial weight, threshold and output weight of the BP neural network through a particle swarm algorithm to complete the training of the BP neural network model;

[0098] The liquid level controller module is used to input the first liquid level deviation value into the trained BP neural network model, output the liquid level dynamic adjustment parameter through the BP neural network model, and adjust the pulsating bucket valve switch amount according to the liquid level dynamic adjustment parameter;

[0099] The valve electric execution module continuously monitors the slurry level through the feedback judgment strategy, and adjusts the pulsation bucket valve switch value according to the real-time feedback information to complete the adjustment and correction of the slurry level. The BP neural network model in the neural network module specifically includes: input layer, hidden layer and output layer; wherein, the output layer is provided with 4 neurons, corresponding to the bias term, the first dynamic target level value, the real-time slurry level data and the first dynamic target level value, the hidden layer is set to 5 neurons, and the output of the output layer is ,in, is the PID control parameter;

[0100] The activation function of the hidden layer is a non-negative Sigmoid function; the BP neural network model uses the steepest descent method to correct the weight coefficient of the network, and the learning rate is set to 0.20 and the inertia coefficient is set to 0.05.

[0101] The detection range of the ultrasonic liquid level sensor module is 65~600mm; the IP level of the ultrasonic liquid level sensor module is 67;

[0102] Exemplarily, in this embodiment, the sensor model is Slon-mic+35 / IU / SC / G;

[0103] The valve electric execution module is used for automatic control of the switch valve of the pulsating bucket, outputs a 4-20mA analog signal to the electric actuator, and automatically adjusts the opening and closing degree of the valve according to the change of the liquid level.

[0104] Exemplarily, in this embodiment, the model of the electric actuator is a Z series intelligent non-invasive electric actuator.

[0105] Example 3: Exemplarily, in this example, a method for determining the transfer function of a magnetic separator liquid level control system using a system identification method is provided, which is as follows:

[0106] When the feed rate is stable and the magnetic separator reaches a stable state under specific working conditions, the drainage bucket, supplementary water and flushing water remain constant. In the open-loop mode, the valve opening is given. After reaching a stable state, the valve opening is changed to affect the change of the magnetic separator liquid level, and the data of the liquid level value changing over time are recorded. The transfer function of the liquid level control system is calculated based on the recorded data using the MATLAB system identification toolbox.

[0107] Embodiment 4: Exemplarily, a communication method between MATLAB and a PLC controller is provided in this embodiment, which specifically includes:

[0108] The control algorithm written in MATLAB is implemented through OPC UA communication and applied to the on-site PLC control system.

[0109] The OPC UA server of this system is S7-1200 PLC and the client is MATLAB software.

[0110] Preferably, the server configuration in this embodiment is as follows:

[0111] First, establish a connection with the PLC, modify the network IP address and subnet mask to fixed values ​​in the computer adapter options, set the PG / PC interface, and assign parameters to the communication interface. Create communication data variables. The variables used in this system are the proportional coefficient, integral coefficient, differential coefficient, liquid level detection value, liquid level setting value, and actuator opening control of the control algorithm. After the configuration is completed, you can communicate with MATLAB to realize data transmission and reading and writing.

[0112] The client is MATLAB software, using the integrated OPC toolbox to set up a connection with the server and write programs to read and write variables.

[0113] Embodiment 5: This embodiment provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0114] The processor executes the above-mentioned magnetic separator liquid level PID control method by calling the computer program stored in the memory.

[0115] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement a magnetic separator liquid level PID control method provided in the above method embodiment. The electronic device may also include other components for implementing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0116] Embodiment 6: This embodiment provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0117] When the computer program runs on a computer device, the computer device executes the above-mentioned magnetic separator liquid level PID control method.

[0118] For example, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0119] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or / and a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0120] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0121] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0124] In summary, compared with the prior art, the technical effects of the present invention are as follows:

[0125] 1. In the present invention, the host computer displays the liquid level detection value and the liquid level curve in real time, and automatically adjusts the switching amount of the pulsating bucket valve according to the liquid level dynamic adjustment parameters output by the BP neural network model, so as to realize accurate control and adjustment of the slurry level.

[0126] 2. The control algorithm used in the present invention enables the system to have fast response speed, high adjustment accuracy and good stability.

[0127] 3. The device used in the present invention has the advantage of convenient liquid level control, which significantly improves the production efficiency, mineral processing index and automation level of the high gradient magnetic separator.

[0128] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A magnetic separator liquid level PID control method, characterized in that: The method comprises the following specific steps: S1: Collect real-time slurry level data in the magnetic separator through ultrasonic level sensors; S2: extracting the first dynamic target liquid level value set by the upper computer of the magnetic separator, importing the real-time slurry liquid level data and the first dynamic target liquid level value into the slurry liquid level deviation value calculation strategy to calculate the first liquid level deviation value; S3: Establish a BP neural network model, and use the particle swarm algorithm to perform global optimization on the BP neural network to complete the training of the BP neural network model; S4: inputting the first liquid level deviation value into the trained BP neural network model, outputting the liquid level dynamic adjustment parameter through the BP neural network model, and adjusting the switching amount of the pulsating bucket valve according to the liquid level dynamic adjustment parameter; S5: The slurry level is continuously monitored through the feedback judgment strategy, and the pulsation bucket valve switching value is adjusted according to the real-time feedback information to complete the adjustment and correction of the slurry level.

2. A magnetic separator liquid level PID control method according to claim 1, characterized in that: S2 includes the following specific steps: S21: extracting real-time slurry level data in the magnetic separator according to S1; S22: extracting the first dynamic target liquid level value set by the upper computer of the magnetic separator; S23: input the real-time slurry level data and the first dynamic target level value into the PLC system, and import the real-time slurry level data and the first dynamic target level value into the slurry level deviation value calculation strategy through the PLC system to calculate the level deviation value. The slurry level deviation value calculation strategy is as follows: ； in, is the first liquid level deviation value; They are respectively the real-time slurry level data and the first dynamic target level value of the magnetic separator at the monitoring time point t.

3. A magnetic separator liquid level PID control method according to claim 2, characterized in that: S3 includes the following specific steps: S31: Establishing a BP neural network model, assigning an initial position and an initial velocity to a particle swarm through a particle swarm algorithm, and mapping the initial position and initial velocity of the particle swarm to the weights of each node of the BP neural network, wherein the weights of each node include: weight and bias; S32: using the training data to perform multiple iterations of training on the BP neural network, and calculating the fitness of each particle, wherein the calculation strategy of the fitness of each particle is: calculating the mean square error between the predicted acceleration and the actual acceleration of the particle; S33: updating the individual historical optimal weight and the global optimal weight of each particle, and synchronously updating the speed and position of the particle, wherein the individual historical optimal weight is the optimal weight achieved by the particle in the previous iteration, and the global optimal weight is the optimal weight achieved in the entire particle swarm; S34: Calculate and extract the mean fitness data of the particles, and compare the mean fitness data of the particles with a preset error threshold. When the mean fitness data of the particles is less than or equal to the preset error threshold, the iterative training of the BP neural network is completed. Otherwise, return to step S32 to perform the next round of iterative training and recalculate the fitness data of each particle.

4. A magnetic separator liquid level PID control method according to claim 3, characterized in that: In S4, the output of the liquid level dynamic adjustment parameter specifically includes: S41: inputting the first liquid level deviation value into the trained BP neural network model; S42: Outputting the liquid level dynamic adjustment parameter through the BP neural network model, wherein the liquid level dynamic adjustment parameter The details are as follows: ； in, are respectively the proportional coefficient, integral coefficient and differential coefficient of the first liquid level deviation value; S43: adjusting the switching value of the pulsating bucket valve according to the liquid level dynamic adjustment parameter.

5. A magnetic separator liquid level PID control method according to claim 4, characterized in that: In S5, the continuous monitoring of the slurry level through the feedback judgment strategy specifically includes: S51: Return to step S1, and extract the real-time slurry level data in the magnetic separator again through the ultrasonic level sensor; S52: synchronously extracting a second dynamic target liquid level value set by the upper computer of the magnetic separator; S53: Importing the real-time slurry level data and the second dynamic target level value into the slurry level deviation value calculation strategy to calculate the second level deviation value; S54: Continuously monitor the slurry level through feedback judgment strategy; When the second liquid level deviation value is greater than or equal to the first liquid level deviation value, it is determined that the closed-loop control of the magnetic separator fails, and a manual liquid level control operation prompt is issued to the vertical ring high gradient magnetic separator operator; When the second liquid level deviation value is less than the preset liquid level deviation threshold, it is determined that the closed-loop control of the magnetic separator is operating normally, and steps S3-S4 are continued to be executed to perform circulating liquid level control on the magnetic separator; When the second liquid level deviation value is zero, the liquid level control process of the magnetic separator is completed.

6. A magnetic separator liquid level PID control system, which is implemented based on a magnetic separator liquid level PID control method according to any one of claims 1 to 5, characterized in that: The system includes the following modules: Ultrasonic liquid level sensor module, liquid level deviation value calculation module, neural network module, liquid level controller module, valve electric execution module; The ultrasonic level sensor module monitors the real-time slurry level data in the magnetic separator through an ultrasonic level sensor; The liquid level deviation value calculation module is used to extract the first dynamic target liquid level value set by the upper computer of the magnetic separator, and import the real-time slurry liquid level data and the first dynamic target liquid level value into the slurry liquid level deviation value calculation strategy to calculate the first liquid level deviation value; The neural network module is used to establish a BP neural network model, and perform global optimization on the initial weight, threshold and output weight of the BP neural network through a particle swarm algorithm to complete the training of the BP neural network model; The liquid level controller module is used to input the first liquid level deviation value into the trained BP neural network model, output the liquid level dynamic adjustment parameter through the BP neural network model, and adjust the pulsating bucket valve switch amount according to the liquid level dynamic adjustment parameter; The valve electric execution module continuously monitors the slurry level through a feedback judgment strategy, and adjusts the pulsation bucket valve switching amount according to real-time feedback information to complete the adjustment and correction of the slurry level.

7. A magnetic separator liquid level PID control system according to claim 6, characterized in that: The BP neural network model in the neural network module specifically includes: an input layer, a hidden layer and an output layer; wherein the output layer is provided with 4 neurons, corresponding to the bias term, the first dynamic target liquid level value, the real-time slurry liquid level data and the first dynamic target liquid level value, the hidden layer is provided with 5 neurons, and the output of the output layer is ,in, is the PID control parameter; The activation function of the hidden layer is a non-negative Sigmoid function; the BP neural network model uses the steepest descent method to correct the weight coefficient of the network, and the learning rate is set to 0.20 and the inertia coefficient is set to 0.

05.

8. A magnetic separator liquid level PID control system according to claim 7, characterized in that: The detection range of the ultrasonic liquid level sensor module is 65~600mm; the IP level of the ultrasonic liquid level sensor module is 67; The valve electric execution module is used for automatic control of the switch valve of the pulsating bucket, outputs a 4-20mA analog signal to the electric actuator, and automatically adjusts the opening and closing degree of the valve according to the change of the liquid level.

Images