A method and system for controlling clean coal ash content in a dense medium separation process

Abstract

This invention proposes a method and system for controlling the ash content of clean coal in a heavy media separation process. The method includes: collecting historical data of the heavy media separation process; data preprocessing; constructing and training a state estimation model and a system dynamic model; and performing real-time control of the clean coal ash content based on the constructed state estimation model and system dynamic model. This invention effectively solves the modeling problem of large time delay characteristics in the heavy media separation process, comprehensively considers the influence of feed pressure on the clean coal ash content, enabling the model to more accurately learn the underlying physical characteristics of the heavy media separation process, and can promptly take appropriate control actions to compensate for the influence of external interference factors. It has a fast response speed, strong anti-interference ability, and is insensitive to noise in ash content measurement. Applying this invention to production can improve the quality of coal products through precise control of clean coal ash content; it can also fully realize automated control, reduce the labor intensity of operators, and improve production efficiency.

Description

Technical Field

[0001] This invention belongs to the field of coal mine production technology, and relates to computer technology, specifically to a method and system for controlling the ash content of clean coal in a heavy media separation process. Background Technology

[0002] Heavy media separation is a coal preparation method that separates coal particles and gangue in a suspension of heavy media with a density greater than water. During heavy media separation, to ensure that the ash content of the clean coal meets production targets, it is necessary to control the ash content of the clean coal by adjusting the density setpoint of the heavy media suspension. Currently, the density control of heavy media suspensions is largely automated in the industry.

[0003] However, due to the slow response of clean coal ash to the density adjustment of heavy medium suspension during heavy medium separation, and the large noise in the online measurement results of clean coal ash, as well as the influence of feed pressure fluctuations, these factors make it difficult for traditional control methods to achieve accurate ash control. Therefore, manual control is still widely used at present. Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes a method and system for controlling the ash content of clean coal in the heavy media separation process, based on a constructed state estimation model and a system dynamic model.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] A method for controlling the ash content of clean coal in a heavy media separation process includes the following steps:

[0007] Step 1: Collect historical data of the heavy medium separation process, including the density of the heavy medium suspension, feed pressure, and ash content of the clean coal.

[0008] Step 2: Perform center-average filtering on the clean coal ash content data for data preprocessing;

[0009] Step 3: Construct the Encoder-Decoder model. Based on the historical data collected in Step 1 and the preprocessed clean coal ash data from Step 2, construct a training dataset. Train the Encoder-Decoder model on this training dataset. Each training sample in the training set contains: the input to the Encoder network, the input to the Decoder network, and the output of the Decoder network. The input to the Encoder network includes: the density of the heavy medium suspension from time t-T1 to time t-1. Feed pressure from time t-T1 to time t-1 And unfiltered clean coal ash content data from time t-T1+1 to time t The output of the encoder network is input to the decoder network. The input to the decoder network also includes the density of the heavy medium suspension from time t to time t+T2-1. The feed pressure from time t to time t+T2-1 The output of the decoder network includes the clean coal ash content from time t+1 to time t+T2 after center-averaged filtering. T1 and T2 are the first preset duration and the second preset duration, respectively;

[0010] Step 4: Construct historical production sequences. Based on the current density of the heavy medium suspension and the feed pressure, generate N alternative input sequences. Based on the constructed state estimation model and system dynamic model, perform real-time control of the clean coal ash content.

[0011] Furthermore, each sample point in the historical data in step 1 is time series data.

[0012] Furthermore, in step 2, the center-average filtering is implemented using the following formula:

[0013]

[0014] Where yk is the clean coal ash content data at time k collected in step 1. The data represents the ash content of clean coal at time t after central filtering, where 2K+1 is the filter width and K is a preset value.

[0015] Furthermore, the Encoder model is a state estimation model, which predicts the current state of the system based on the system's most recent input and output history sequence, in the following form:

[0016]

[0017] Where x t This is an estimate of the current state. and These are the density sequence and feed pressure sequence of the heavy medium suspension from time t-T1 to time t-1, respectively. Let θ be the clean coal ash sequence from time t-T1+1 to time t, where T1 is the first preset time and θ is the model parameter;

[0018] The Decoder model includes a state transition model and an output model.

[0019] The state transition model describes the state transition characteristics of the system, and takes the following form:

[0020] x t+1 =h φ (x t ,ut ,d t )

[0021] Where x t Let u be the system state at time t. t Let d be the density of the heavy medium suspension at time t. t Let φ be the feed pressure at time t, and φ be the model parameter.

[0022] The output model maps the system state output by the state transition model to the controlled output value, and its form is as follows:

[0023]

[0024] Where x t Let y be the system state at time t. t The output of clean coal ash content at time t. These are the model parameters.

[0025] Furthermore, step 4 includes the following sub-steps:

[0026] Step 4-1: Construct a historical production sequence, which includes the density of the heavy medium suspension from time t-T1 to time t-1. and feed pressure And the clean coal ash content sequence from time t-T1+1 to time t.

[0027] Step 4-2: Based on the constructed historical production sequence, use the Encoder to predict the system state x at the current moment. t ;

[0028] Step 4-3, based on the current density u of the heavy medium suspension t and feed pressure d t Generate N candidate input sequences, which are assumed system inputs within a future time range of T2, including the heavy medium suspension density and feed pressure from time t to time t+T2-1; in each candidate input sequence, the feed pressure at any time is equal to the feed pressure d at the current time. t The density of a heavy medium suspension at any given time is equal to a random value, which varies from the density range (u... t -a,u t +a) is obtained by uniform sampling, where a is a preset parameter;

[0029] Step 4-4: Based on the N candidate input sequences, generate N predicted output sequences. For each candidate input sequence, based on the system state x obtained in Step 4-2... tThe Decoder model is used to iteratively obtain a predicted output sequence, which represents the predicted system output under the condition that the system input in the future time range T2 is the currently selected input sequence;

[0030] Steps 4-5: Based on the obtained N predicted output sequences, compare them with the current ash content setpoint and calculate their absolute deviations to obtain N absolute deviation sequences. Based on the absolute deviation sequences, obtain the optimal heavy medium suspension as the current control action.

[0031] Furthermore, in steps 4-5, the step of obtaining the optimal heavy medium suspension as the current control action based on the absolute deviation sequence includes the following process: summing each absolute deviation sequence to obtain N absolute deviation sums; determining the candidate input sequence corresponding to the prediction output sequence with the smallest absolute deviation sum as the optimal input sequence; and using the density value of the heavy medium suspension corresponding to the optimal input sequence as the current control action.

[0032] This invention also provides a control system for the ash content of clean coal in a heavy media separation process, including a historical data collection module, a data preprocessing module, a model building and training module, and an ash content control module. The historical data collection module is used to implement the function of step 1 of the method for controlling the ash content of clean coal in a heavy media separation process; the data preprocessing module is used to implement the function of step 2 of the method for controlling the ash content of clean coal in a heavy media separation process; the model building and training module is used to implement the function of step 3 of the method for controlling the ash content of clean coal in a heavy media separation process; and the ash content control module is used to implement the function of step 4 of the method for controlling the ash content of clean coal in a heavy media separation process.

[0033] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory, characterized in that the processor executes the computer program to implement the steps of the method for controlling the ash content of clean coal in the heavy media separation process provided by the present invention.

[0034] The present invention also provides a computer-readable storage medium storing a computer program / instructions thereon, characterized in that the computer program / instructions, when executed by a processor, implement the steps of the method for controlling the ash content of clean coal in the heavy media separation process provided by the present invention.

[0035] The beneficial effects of this invention are as follows:

[0036] 1. This invention employs an Encoder-Decoder network structure to construct a state estimation model and a system dynamic model. The state estimation model estimates the unknown system state based on short-term historical data of the heavy media separation process, serving as the initial state for the system dynamic model. The system dynamic model effectively addresses the modeling problem of the large time delay characteristics of the heavy media separation process by introducing state variables. Furthermore, during model training, the filtered clean coal ash content is used as the output of the system dynamic model, and the influence of feed pressure on the clean coal ash content is comprehensively considered during modeling, enabling the model to more accurately learn the underlying physical characteristics of the heavy media separation process.

[0037] 2. This invention employs model-based predictive control technology, which, compared to manual control and traditional PID control schemes, can promptly take appropriate control actions to compensate for the influence of external interference factors. It has a fast response speed, strong anti-interference ability, and is not sensitive to noise in ash content measurement.

[0038] 3. Applying the solution of this invention to production can improve the quality of coal products by precisely controlling the ash content of clean coal; it can also fully realize automated control, reduce the labor intensity of operators, and improve production efficiency. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the method for controlling the ash content of clean coal in the heavy media separation process provided by the present invention;

[0040] Figure 2 This is a schematic diagram of the sub-steps of the ash content control process. Detailed Implementation

[0041] The technical solutions provided by the present invention will be described in detail below with reference to specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0042] This invention proposes a method for controlling the ash content of clean coal in a heavy media separation process, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0043] Step 1: Collect historical data of the heavy medium separation process. Each sample point in the dataset is a time series data point, containing the heavy medium suspension density u, feed pressure d, and clean coal ash content y.

[0044] Step 2, Data Preprocessing. A center-average filter is applied to the clean coal ash content data to eliminate measurement noise. The center-average filtering method is as follows:

[0045]

[0046] Where yk is the clean coal ash content data at time k collected in step 1. The data represents the ash content of clean coal at time t after central filtering, where 2K+1 is the filter width and K is a preset value.

[0047] Step 3, Model Construction and Training. The model adopts an Encoder-Decoder structure, where the Encoder model is the state estimation model and the Decoder model is the system dynamics model.

[0048] State estimation models predict the current state of a system based on its most recent input and output history. Their form is as follows:

[0049]

[0050] Where x t This is an estimate of the current state. and These are the density sequence and feed pressure sequence of the heavy medium suspension from time t-T1 to time t-1, respectively. Let θ be the clean coal ash sequence from time t-T1+1 to time t, where T1 is the first preset time and θ is the model parameter.

[0051] The system dynamic model consists of two sub-models: a state transition model and an output model.

[0052] (a) The state transition model describes the state transition characteristics of the system, and its form is as follows:

[0053] x t+1 =h φ (x t u t ,d t )

[0054] Where x t Let u be the system state at time t. t Let d be the density of the heavy medium suspension at time t. t Let φ be the feed pressure at time t, and φ be the model parameter.

[0055] (b) The output model maps the system state output by the state transition model to the controlled output value, and its form is as follows:

[0056]

[0057] Where x t Let y be the system state at time t. t The system output at time t is the clean coal ash content. These are the model parameters.

[0058] The state estimation model is an encoder network that transforms a sequence into a state vector, implemented using network structures such as transformers or LSTM. The system dynamics model is a decoder network that describes the dynamic characteristics of the system, and can be implemented using an LSTM network. During implementation, it is necessary to ensure that the dimensions of the state vectors in the encoder and decoder networks are equal.

[0059] Model Training. Based on the data collected in step 1 and the preprocessed data in step 2, a training dataset is constructed. Each training sample in the training set contains three parts: the input to the Encoder network, the input to the Decoder network, and the output of the Decoder network. The input to the Encoder network includes... and in For unfiltered clean coal ash data, the input to the Decoder network is... and The output of the Decoder network is in The ash content of the clean coal after center-averaged filtering is given, where T1 and T2 are the first and second preset time periods, respectively. An Encoder-Decoder model is trained on the training set.

[0060] Step 4, Ash Content Control. Real-time control of clean coal ash content is performed based on the constructed state estimation model and system dynamic model. At any control time t, the specific steps include:

[0061] Step 4-1: Construct the historical production sequence. The historical production sequence includes the density of the heavy medium suspension from time t-T1 to the current time t-1. and feed pressure And the clean coal ash content sequence from time t-T1+1 to time t. T1 is the first preset duration.

[0062] Step 4-2: Based on the constructed historical production sequence, use the state estimation model to predict the system state x at the current moment. t .

[0063] Step 4-3, based on the current density u of the heavy medium suspension t and feed pressure d t Generate N candidate input sequences. Each candidate input sequence represents the system input within a hypothetical future time range of T2, containing the heavy medium suspension density and feed pressure from time t to time t+T2-1, where T2 is a second preset time duration. In each candidate input sequence, ensure that the feed pressure at any given time is equal to the feed pressure d at the current time. tLet the density of the heavy medium suspension at any given time be equal to a random value, which is selected from the density range (u... t -a,u t The density search space is obtained by uniform sampling in +a), where a is a preset parameter that determines the size of the density search space at the current control time.

[0064] Step 4-4: Based on the N candidate input sequences, generate N predicted output sequences. For each candidate input sequence, based on the system state x obtained in Step 4-2... t A predicted output sequence can be obtained iteratively using the system dynamic model, representing the predicted system output under the condition that the system input within the future time range T2 is the currently selected input sequence.

[0065] Steps 4-5: Based on the obtained N predicted output sequences, compare them with the current ash content setting and calculate their absolute deviations to obtain N absolute deviation sequences. Further, sum the absolute deviation sequences to obtain N absolute deviation sums. The candidate input sequence corresponding to the predicted output sequence with the smallest absolute deviation sum is determined as the optimal input sequence.

[0066] After obtaining the optimal input sequence, the density value of the heavy medium suspension in it is used as the current control action.

[0067] The present invention provides a method for controlling the ash content of clean coal in a heavy media separation process, which should be implemented by computer software. Therefore, the present invention also provides a control system for the ash content of clean coal in a heavy media separation process, including a historical data collection module, a data preprocessing module, a model building and training module, and an ash content control module. The historical data collection module is used to collect historical data of the heavy media separation process, specifically implementing step 1 of the method for controlling the ash content of clean coal in a heavy media separation process. The data preprocessing module is used to perform center-average filtering on the clean coal ash content data, specifically implementing step 2 of the method for controlling the ash content of clean coal in a heavy media separation process. The model building and training module is used to construct an Encoder-Decoder model and train the model based on a training dataset, specifically implementing step 3 of the method for controlling the ash content of clean coal in a heavy media separation process. The ash content control module is used to perform real-time control of the ash content of clean coal based on the constructed state estimation model and the system dynamic model, specifically implementing step 4 of the method for controlling the ash content of clean coal in a heavy media separation process.

[0068] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method for controlling the ash content of clean coal in the heavy media separation process provided by the present invention.

[0069] The present invention also provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the method for controlling the ash content of clean coal in the heavy media separation process provided by the present invention.

[0070] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. A method for controlling the ash content of clean coal in a heavy media separation process, characterized in that, Includes the following steps: Step 1: Collect historical data of the heavy medium separation process, including the density of the heavy medium suspension, feed pressure, and ash content of the clean coal. Step 2: Perform center-average filtering on the clean coal ash content data for data preprocessing; Step 3: Construct the Encoder-Decoder model. Based on the historical data collected in Step 1 and the preprocessed clean coal ash data in Step 2, construct a training dataset and train the Encoder-Decoder model on the training dataset. Each training sample in the training set contains: the input of the Encoder network, the input of the Decoder network, and the output of the Decoder network. The input to the encoder network includes: the density of the heavy medium suspension from time t-T1 to time t-1. Feed pressure from time t-T1 to time t-1 And unfiltered clean coal ash content data from time t-T1+1 to time t The output of the encoder network is input to the decoder network. The input to the decoder network includes the density of the heavy medium suspension from time t to time t+T2-1. The feed pressure from time t to time t+T2-1 The output of the decoder network includes the clean coal ash content from time t+1 to time t+T2 after center-averaged filtering. T1 and T2 are the first preset duration and the second preset duration, respectively; The Encoder model is a state estimation model. The state estimation model predicts the current state of the system based on the system's most recent input and output history sequence, and takes the following form: Where x t This is an estimate of the current state. and These are the density sequence and feed pressure sequence of the heavy medium suspension from time t-T1 to time t-1, respectively. Let θ be the clean coal ash sequence from time t-T1+1 to time t, where T1 is the first preset time and θ is the model parameter; The Decoder model includes a state transition model and an output model. The state transition model describes the state transition characteristics of the system, and takes the following form: x t+1 =h φ (x t ,u t ,d t ) Where x t Let u be the system state at time t. t Let d be the density of the heavy medium suspension at time t. t Let φ be the feed pressure at time t, and φ be the model parameter. The output model maps the system state output by the state transition model to the controlled output value, and its form is as follows: Where x t Let y be the system state at time t. t The output of clean coal ash content at time t. These are model parameters; Step 4: Construct historical production sequences. Based on the current density of the heavy medium suspension and the feed pressure, generate N alternative input sequences. Based on the constructed state estimation model and system dynamic model, perform real-time control of the clean coal ash content.

2. The method for controlling the ash content of clean coal in the heavy media separation process according to claim 1, characterized in that, Each sample point in the historical data in step 1 is time series data.

3. The method for controlling the ash content of clean coal in the heavy media separation process according to claim 1, characterized in that, In step 2, the center-average filtering is implemented using the following formula: Among them, y k The data represents the ash content of clean coal at time k, collected in step 1. The data is the clean coal ash content at time t after center filtering, where 2K+1 is the filtering width and K is a preset value.

4. The method for controlling the ash content of clean coal in the heavy media separation process according to claim 1, characterized in that, Step 4 includes the following sub-steps: Step 4-1: Construct a historical production sequence, which includes the density of the heavy medium suspension from time t-T1 to time t-1. and feed pressure And the clean coal ash content sequence from time t-T1+1 to time t. Step 4-2: Based on the constructed historical production sequence, use the Encoder to predict the system state x at the current moment. t ; Step 4-3, based on the current density u of the heavy medium suspension t and feed pressure d t Generate N candidate input sequences, which are assumed system inputs within a future time range of T2, including the heavy medium suspension density and feed pressure from time t to time t+T2-1; in each candidate input sequence, the feed pressure at any time is equal to the feed pressure d at the current time. t The density of a heavy medium suspension at any given time is equal to a random value, which varies from the density range (u... t -a,u t +a) is obtained by uniform sampling, where a is a preset parameter; Step 4-4: Based on the N candidate input sequences, generate N predicted output sequences; for each candidate input sequence, based on the system state x obtained in step 4-2... t The Decoder model is used to iteratively obtain a predicted output sequence, which represents the predicted system output under the condition that the system input in the future time range T2 is the currently selected input sequence; Steps 4-5: Based on the obtained N predicted output sequences, compare them with the current ash content setpoint and calculate their absolute deviations to obtain N absolute deviation sequences. Based on the absolute deviation sequences, obtain the optimal heavy medium suspension as the current control action.

5. The method for controlling the ash content of clean coal in the heavy media separation process according to claim 4, characterized in that, In steps 4-5, the step of obtaining the optimal heavy medium suspension as the current control action based on the absolute deviation sequence includes the following process: summing each absolute deviation sequence to obtain N absolute deviation sums; determining the candidate input sequence corresponding to the prediction output sequence with the smallest absolute deviation sum as the optimal input sequence; and using the density value of the heavy medium suspension corresponding to the optimal input sequence as the current control action.

6. A control system for clean coal ash content in a heavy media separation process, characterized in that, The method includes a historical data collection module, a data preprocessing module, a model building and training module, and an ash content control module. The historical data collection module is used to implement the function of step 1 of the method for controlling the ash content of clean coal in the heavy media separation process as described in any one of claims 1-5. The data preprocessing module is used to implement the function of step 2 of the method for controlling the ash content of clean coal in the heavy media separation process as described in any one of claims 1-5. The model building and training module is used to implement the function of step 3 of the method for controlling the ash content of clean coal in the heavy media separation process as described in any one of claims 1-5. The ash content control module is used to implement the function of step 4 of the method for controlling the ash content of clean coal in the heavy media separation process as described in any one of claims 1-5.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method for controlling the ash content of clean coal in the heavy media separation process according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When executed by a processor, the computer program / instruction implements the steps of the method for controlling the ash content of clean coal in the heavy media separation process as described in any one of claims 1-5.

Images