A full-process intelligent control method applied to heavy medium coal separation

Abstract

This invention discloses a full-process intelligent control method for heavy medium coal preparation, relating to the field of industrial control system technology. It addresses the problem of improving the dynamic response accuracy and separation interface stability to medium density disturbances during heavy medium coal preparation. The method includes: acquiring process monitoring data of the heavy medium coal preparation process within a preset observation period; determining the dynamic disturbance index of the separation zone based on upstream heavy medium density data, separation zone heavy medium density data, separation zone coal slurry flow rate data, and internal flow field data of the separation zone; determining the separation interface drift coefficient based on the coal quality data of the separation products; wherein the separation interface drift coefficient characterizes the degree of displacement of the separation interface between clean coal and gangue caused by changes in the separation flow field state; determining the dynamic response coupling factor based on the separation zone dynamic disturbance index and the separation interface drift coefficient; and adjusting the current heavy medium density based on the dynamic response coupling factor.

Description

Technical Field

[0001] This invention relates to the field of industrial control system technology, and specifically to a full-process intelligent control method for heavy media coal preparation. Background Technology

[0002] Heavy media coal preparation is a core separation process in the field of clean coal processing. It achieves efficient separation of coal and gangue by adjusting the density of the heavy media suspension. The control precision of this process directly affects the quality and yield of the final clean coal product. In actual industrial processes, heavy media coal preparation systems involve multiple interconnected steps, including media preparation, circulation, separation, and product recovery, forming a complex dynamic process. Existing technologies for controlling heavy media coal preparation processes mainly focus on the independent monitoring and closed-loop adjustment of single process parameters (such as separation zone density and feed flow rate).

[0003] However, due to the objective existence of fluctuations in the properties of raw coal and disturbances in equipment operation, disturbances from upstream processes are transmitted to downstream sorting stages in a nonlinear and time-delayed manner through the media circulation system. Existing control strategies often lack a systematic analysis and compensation mechanism for the propagation characteristics of such cross-process and time-delayed disturbances and their impact on the final sorting effect. This makes it difficult to achieve accurate and stable intelligent control at the whole process scale, which restricts the further improvement of sorting efficiency and product quality. Summary of the Invention

[0004] To address the current technical challenges of improving the dynamic response accuracy and separation interface stability to medium density disturbances in heavy medium coal preparation, this invention aims to provide a full-process intelligent control method for heavy medium coal preparation. The specific technical solution adopted is as follows: In a first aspect, the present invention provides a full-process intelligent control method for heavy medium coal preparation, comprising: acquiring process monitoring data of the heavy medium coal preparation process within a preset observation period; the process monitoring data including upstream heavy medium density data, separation zone heavy medium density data, separation zone coal slurry flow rate data, separation zone internal flow field data, and separation product coal quality data; determining a dynamic disturbance index of the separation zone based on the upstream heavy medium density data, separation zone heavy medium density data, separation zone coal slurry flow rate data, and separation zone internal flow field data; wherein, the dynamic disturbance index of the separation zone is used to characterize the comprehensive influence of upstream heavy medium density disturbance on the separation flow field state within the separation zone after being transmitted to the separation zone through medium circulation; determining a separation interface drift coefficient based on the separation product coal quality data; wherein, the separation interface drift coefficient is used to characterize the degree of shift in the separation interface position between clean coal and gangue caused by changes in the separation flow field state; determining a dynamic response coupling factor based on the separation zone dynamic disturbance index and the separation interface drift coefficient; and adjusting the current heavy medium density based on the dynamic response coupling factor.

[0005] Secondly, the present invention provides a full-process intelligent control system for heavy medium coal preparation, comprising: a data acquisition module, a dynamic disturbance assessment module, an interface drift assessment module, a coupling factor calculation module, and a heavy medium density adjustment module; the data acquisition module is used to acquire process monitoring data of the heavy medium coal preparation process within a preset observation period; the process monitoring data includes upstream heavy medium density data, separation zone heavy medium density data, separation zone coal slurry flow data, separation zone internal flow field data, and separation product coal quality data; the dynamic disturbance assessment module is used to determine the dynamic disturbance based on the upstream heavy medium density data, separation zone heavy medium density data, separation zone coal slurry flow data, and separation zone internal flow field data. The system comprises the following modules: a sorting zone dynamic disturbance index (characterizing the comprehensive impact of upstream heavy medium density disturbances transmitted through the medium circulation to the sorting zone on the state of the sorting flow field); an interface drift assessment module (determining the sorting interface drift coefficient based on the coal quality data of the sorted products, characterizing the degree of shift in the position of the clean coal and gangue sorting interface caused by changes in the sorting flow field state); a coupling factor calculation module (determining the dynamic response coupling factor based on the sorting zone dynamic disturbance index and the sorting interface drift coefficient); and a heavy medium density adjustment module (adjusting the current heavy medium density based on the dynamic response coupling factor).

[0006] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer-executable instructions, and when the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform the intelligent control method for the entire process of heavy media coal preparation as described in the first aspect and any possible implementation thereof.

[0007] This invention offers the following advantages: By constructing a closed-loop intelligent control system encompassing data acquisition, dynamic disturbance assessment, interface drift assessment, coupling factor calculation, and density adjustment, it first systematically collects and integrates multi-source time-series data reflecting medium circulation, flow field structure, and sorting results. Secondly, by quantitatively analyzing the dynamic mismatch of upstream density disturbances during transmission and the resulting flow field shift, and combining this with changes in sorting product quality to infer interface stability, it achieves a precise characterization of the dynamic coupling relationships within the process. Finally, based on the dynamic response coupling factor obtained through comprehensive evaluation, it intelligently adjusts the heavy medium density in a directional and quantitative manner. This method effectively overcomes the shortcomings of existing technologies that independently control process parameters and ignore cross-process disturbance transmission and time-series coupling, significantly improving the control accuracy, adaptability, and final product quality stability of the heavy medium coal preparation process when facing fluctuations in raw coal properties and equipment operating conditions. Attached Figure Description

[0008] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the architecture of a fully intelligent control system for heavy media coal preparation, provided in one embodiment of the present invention. Figure 2 This is a schematic flowchart of a fully intelligent control method for heavy media coal preparation, provided as an embodiment of the present invention. Detailed Implementation

[0010] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0011] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0012] The following description, in conjunction with the accompanying drawings, details a specific scheme for an intelligent control method for the entire process of heavy media coal preparation provided by the present invention.

[0013] For example, such as Figure 1 The diagram shown is a schematic representation of the architecture of an intelligent control system (hereinafter referred to as intelligent control system 10) applied to the entire process of heavy medium coal preparation according to an embodiment of the present invention. The intelligent control system 10 includes: a data acquisition module 11, a dynamic disturbance evaluation module 12, an interface drift evaluation module 13, a coupling factor calculation module 14, and a heavy medium density adjustment module 15. The modules are described in detail below: (1) Data acquisition module 11.

[0014] The data acquisition module 11 is responsible for collecting and aggregating all the raw data required for intelligent control within the preset observation period from the sensor network deployed at key nodes throughout the heavy media coal preparation process, providing basic data support for subsequent evaluation and calculation.

[0015] Optionally, the data acquisition module 11 is used to acquire process monitoring data. Specifically, the process monitoring data includes: upstream heavy medium density data collected by density sensors installed at key upstream nodes of the heavy medium circulation system (such as downstream of the supplementary medium device or the main circulation pipeline); heavy medium density data and coal slurry flow rate data of the sorting zone collected by sensors installed at the inlet of the sorting equipment or inside the sorting zone; flow field data inside the sorting zone collected by sensors (such as flow velocity sensors) installed at multiple monitoring points inside the sorting zone; and coal quality data of the sorted products collected by online coal quality testing devices (such as ash analyzers and near-infrared analyzers) installed at the clean coal outlet and gangue outlet.

[0016] The data acquisition module 11 synchronizes and marks all collected multi-source data according to a unified time base to form a structured process monitoring data set, and transmits it to the dynamic disturbance evaluation module 12 and the interface drift evaluation module 13 as the data starting point of the entire intelligent control process.

[0017] (2) Dynamic disturbance assessment module 12.

[0018] The dynamic disturbance assessment module 12 is responsible for receiving upstream heavy medium density data, sorting zone heavy medium density data, sorting zone coal slurry flow data, and sorting zone internal flow field data from the data acquisition module 11. By comprehensively analyzing the disturbance propagation and flow field response in the medium circulation, it determines a quantitative sorting zone dynamic disturbance index, which characterizes the comprehensive impact of upstream density disturbance on the sorting flow field state within the sorting zone.

[0019] Optionally, the dynamic disturbance assessment module 12 is used to determine the dynamic disturbance index of the sorting zone based on upstream heavy medium density data, sorting zone heavy medium density data, sorting zone coal slurry flow data, and sorting zone internal flow field data.

[0020] For example, the dynamic disturbance assessment module 12 can be divided into a propagation mismatch assessment submodule 121 and a response offset assessment submodule 122 to assess the propagation distortion and flow field reconstruction of the disturbance, respectively, which are described below: (2.1) Propagation mismatch assessment submodule 121.

[0021] Optionally, the propagation mismatch assessment submodule 121 is used to determine the propagation dynamic mismatch degree based on the upstream heavy medium density data and the sorting zone heavy medium density data.

[0022] Specifically, the propagation mismatch assessment submodule 121 first identifies a first feature dataset reflecting significant density changes from the upstream heavy medium density data, and then identifies a corresponding second feature dataset from the heavy medium density data in the sorting area. Next, the propagation mismatch assessment submodule 121 analyzes the temporal correspondence between the two feature datasets and calculates the heavy medium density perturbation propagation delay exponent. Finally, the propagation mismatch assessment submodule 121 combines this delay exponent with the amplitude differences of the corresponding changes in the two datasets to calculate the final propagation dynamic mismatch degree. This parameter comprehensively characterizes the temporal delay and amplitude distortion degree during the propagation of the upstream heavy medium density perturbation to the sorting area.

[0023] The propagation mismatch assessment submodule 121 outputs the calculated propagation dynamic mismatch degree to the response offset assessment submodule 122 for subsequent comprehensive calculation.

[0024] (2.2) Response offset evaluation submodule 122.

[0025] Optionally, the response offset evaluation submodule 122 is used to determine the response offset index of the sorting zone based on the coal slurry flow data and the internal flow field data of the sorting zone.

[0026] Specifically, the response offset assessment submodule 122 first calculates the average fluctuation intensity of the coal slurry flow rate data in the sorting zone during the observation period. Then, it acquires the flow velocity data of the internal flow field data in the sorting zone at multiple spatial locations (such as the inlet area and the core sorting area) and calculates the average spatial difference level of the flow velocity between these locations. Finally, based on the average fluctuation intensity and the average spatial difference level, the response offset assessment submodule 122 determines the sorting zone response offset index, which characterizes the degree of change in the internal flow field structure of the sorting zone caused by fluctuations in coal slurry flow rate.

[0027] After obtaining the propagation dynamic mismatch degree and the sorting zone response offset index, the response offset evaluation submodule 122 combines the two to calculate the sorting zone dynamic disturbance index, which characterizes the comprehensive impact of upstream disturbance on the flow field of the sorting zone, and outputs it to the coupling factor calculation module 14.

[0028] (3) Interface drift evaluation module 13.

[0029] The interface drift assessment module 13 is responsible for receiving coal quality data of the sorted products from the data acquisition module 11. By analyzing the asynchronous nature of the changes in the coal quality of light and heavy products, it determines a quantitative sorting interface drift coefficient, which characterizes the degree of displacement of the sorting interface between clean coal and gangue caused by the change in the sorting flow field state.

[0030] Optionally, the interface drift evaluation module 13 is used to determine the interface drift coefficient based on the coal quality data of the sorted products.

[0031] Specifically, the interface drift assessment module 13 first extracts a first variation feature sequence based on the time-varying trend of the light product coal quality data; simultaneously, it extracts a second variation feature sequence based on the time-varying trend of the heavy product coal quality data. Then, the interface drift assessment module 13 calculates the difference in characteristic values ​​between the first and second variation feature sequences at the same time point. Finally, based on the difference in characteristic values ​​throughout the entire observation period, the interface drift assessment module 13 determines the sorting interface drift coefficient.

[0032] The interface drift evaluation module 13 outputs the calculated sorting interface drift coefficient to the coupling factor calculation module 14 for subsequent comprehensive calculation.

[0033] (4) Coupling factor calculation module 14.

[0034] The coupling factor calculation module 14 is responsible for receiving the sorting zone dynamic disturbance index from the dynamic disturbance evaluation module 12 and the sorting interface drift coefficient from the interface drift evaluation module 13. By combining these two indicators, which respectively reflect the intensity of process disturbance and the degree of result deviation, the final dynamic response coupling factor is calculated. This factor is the core decision parameter guiding the adjustment of heavy medium density.

[0035] Optionally, the coupling factor calculation module 14 is used to determine the dynamic response coupling factor based on the dynamic disturbance index of the sorting zone and the drift coefficient of the sorting interface.

[0036] Specifically, the coupling factor calculation module 14 performs a fusion calculation (e.g., multiplication followed by normalization) on the dynamic disturbance index of the sorting zone and the drift coefficient of the sorting interface to generate a dynamic response coupling factor. The magnitude of this factor directly reflects the urgency and intensity of intervention by the control system under the current operating conditions; the larger the factor value, the greater the required adjustment of the heavy medium density. The coupling factor calculation module 14 outputs the calculated dynamic response coupling factor to the heavy medium density adjustment module 15.

[0037] (5) Heavy medium density adjustment module 15.

[0038] The heavy medium density adjustment module 15 is responsible for receiving the dynamic response coupling factor from the coupling factor calculation module 14 and the current heavy medium density and coal quality data of the sorted products from the data acquisition module 11. Based on this, it generates accurate heavy medium density adjustment commands and drives the field actuator to complete closed-loop control, thereby stabilizing the sorting interface and improving product quality.

[0039] Optionally, the heavy medium density adjustment module 15 is used to adjust the current heavy medium density according to the dynamic response coupling factor.

[0040] Specifically, the heavy medium density adjustment module 15 first analyzes the changing trend of the coal quality data of the sorted products to determine the direction of adjustment (increase or decrease) of the heavy medium density. Then, based on the preset maximum allowable adjustment range of the system and the received dynamic response coupling factor, the heavy medium density adjustment module 15 calculates the specific adjustment amount for this operation. Finally, according to the determined adjustment direction and amount, the heavy medium density adjustment module 15 generates a specific target heavy medium density value and converts it into executable process commands (such as adjusting the opening of the replenishing valve or the flow rate) to control the heavy medium circulation system to perform the adjustment action.

[0041] Optionally, the heavy medium density adjustment module 15 is also used to trigger the system to enter the next observation and control cycle after completing one adjustment. The data acquisition module 11 begins to collect new process monitoring data, and each evaluation and calculation module runs in sequence, thereby forming a continuously optimized, end-to-end intelligent control closed loop.

[0042] The intelligent control system 10 and its included modules have been described above.

[0043] For example, such as Figure 2 The diagram shown is a flowchart illustrating a fully intelligent control method for heavy media coal preparation according to an embodiment of the present invention, comprising the following steps: S201. Obtain process monitoring data for the heavy medium coal preparation process within the preset observation period. The process monitoring data includes upstream heavy medium density data, heavy medium density data in the separation zone, coal slurry flow rate data in the separation zone, internal flow field data in the separation zone, and coal quality data of the separation products.

[0044] For example, this step can be performed by the data acquisition module 11 in the intelligent control system 10 described above, and specifically includes the following steps: (1) Acquire upstream heavy medium density data by using sensors installed at key upstream nodes of the heavy medium circulation system.

[0045] Specifically, the data acquisition module 11 continuously collects the density values ​​of the heavy medium suspension within a preset observation period from online density sensors deployed at key upstream nodes of the heavy medium circulation system (such as downstream of the replenishment device, at the outlet of the medium mixing tank, or before the main circulation pipeline enters the sorting equipment), forming an upstream heavy medium density data sequence that changes over time.

[0046] (2) Data on the density of the heavy medium in the sorting zone and the flow rate of the coal slurry in the sorting zone are obtained by sensors installed at the inlet of the sorting equipment or inside the sorting zone.

[0047] Specifically, the data acquisition module 11 collects density data of the heavy medium suspension participating in the separation process from density sensors installed at the inlet of key separation equipment such as heavy medium cyclones or separation tanks or inside the separation zone, thus obtaining heavy medium density data for the separation zone. Simultaneously, it collects the instantaneous flow rate of coal slurry entering the separation zone within the same time period using online flow meters (such as electromagnetic flow meters) installed at the inlet of the separation zone or on key pipelines, thus obtaining coal slurry flow rate data for the separation zone.

[0048] (3) Obtain flow field data inside the sorting zone by using sensors at multiple monitoring points inside the sorting zone.

[0049] Specifically, the data acquisition module 11 directly collects flow velocity information at various spatial locations using flow velocity sensors (such as multi-point flow velocity detection devices) deployed in multiple key areas within the sorting zone (such as near the inlet, the core sorting area, and the area adjacent to the sorting interface), thus forming the flow field data within the sorting zone. In another implementation, this data can also be estimated in real time based on pressure and flow sensor data at the inlet and outlet of the sorting zone, combined with a pre-set fluid dynamics model (e.g., a simplified simulation model based on computational fluid dynamics (CFD), or a neural network prediction model trained based on historical data).

[0050] (4) Obtain coal quality data of sorted products by using online coal quality detection devices installed at the clean coal outlet and gangue outlet.

[0051] Specifically, the data acquisition module 11 continuously collects coal quality indicators (such as ash content or density value) of light and heavy products from the online ash content detection device, density recognition device or image recognition unit set at the diversion outlet of clean coal (light product) and gangue (heavy product) to obtain coal quality data of the sorted products.

[0052] (5) The upstream heavy medium density data, the heavy medium density data of the sorting zone, the coal slurry flow data of the sorting zone, the internal flow field data of the sorting zone, and the coal quality data of the sorting products are synchronized and marked according to a unified time reference to form process monitoring data.

[0053] Furthermore, the data acquisition module 11 synchronizes various types of data from different acquisition nodes with different sampling frequencies using timestamp alignment technology and integrates them into a process monitoring dataset with a unified time series label.

[0054] Thus, the data acquisition module 11 systematically collects and synchronizes multi-source data reflecting the state of the medium, the flow field structure, and the sorting results through the sensor network deployed at key nodes throughout the process, laying a reliable data foundation for subsequent intelligent evaluation and control.

[0055] S202. Based on the upstream heavy medium density data, the heavy medium density data of the sorting zone, the coal slurry flow rate data of the sorting zone, and the internal flow field data of the sorting zone, determine the dynamic disturbance index of the sorting zone. The dynamic disturbance index of the sorting zone characterizes the comprehensive impact of upstream heavy medium density disturbances, after being transmitted to the sorting zone through medium circulation, on the state of the partially sorted flow field within the sorting zone.

[0056] For example, this step can be performed by the dynamic disturbance evaluation module 12 in the intelligent control system 10 described above. Specifically, the dynamic disturbance evaluation module 12 first evaluates the timing delay and amplitude distortion of the disturbance propagation in the medium circulation based on the upstream heavy medium density data and the heavy medium density data of the sorting zone, and determines the propagation dynamic mismatch. Then, the dynamic disturbance evaluation module 12 evaluates the changes in the spatial structure of the flow field caused by the flow fluctuation based on the coal slurry flow data and the internal flow field data of the sorting zone, and determines the sorting zone response offset index. Finally, the dynamic disturbance evaluation module 12 calculates the sorting zone dynamic disturbance index by combining the propagation dynamic mismatch and the sorting zone response offset index. It should be noted that the specific process of the aforementioned sub-steps can be found in S301-S303 below, and will not be repeated here.

[0057] In another possible implementation, when determining the dynamic disturbance index of the sorting area, the dynamic disturbance assessment module 12 can also assign empirical weights trained based on historical working data to the propagation dynamic mismatch degree and the sorting area response offset index, respectively. The dynamic disturbance index of the sorting area can be directly obtained by weighted summation, which is suitable for scenarios with extremely high real-time requirements and limited computing resources.

[0058] Therefore, the dynamic disturbance assessment module 12 quantitatively analyzes the dynamic characteristics of upstream disturbances in two dimensions: medium transmission and flow field response, and integrates them into a comprehensive sorting area dynamic disturbance index, which accurately characterizes the overall influence intensity of the disturbance on the flow field state of the core sorting area.

[0059] S203. Determine the separation interface drift coefficient based on the coal quality data of the separation products. The separation interface drift coefficient is used to characterize the degree of displacement of the separation interface between clean coal and gangue caused by changes in the separation flow field state.

[0060] For example, this step can be performed by the interface drift evaluation module 13 in the intelligent control system 10 described above. Specifically, the interface drift evaluation module 13 first extracts a first change feature sequence based on the time-varying trend of the light product coal quality data; simultaneously, it extracts a second change feature sequence based on the time-varying trend of the heavy product coal quality data. Then, the interface drift evaluation module 13 calculates the difference in characteristic values ​​between the first and second change feature sequences at the same time. Finally, the interface drift evaluation module 13 determines the sorting interface drift coefficient based on the statistics of the characteristic value differences over the entire observation period. It should be noted that the specific procedures for the aforementioned sub-steps can be found in S401-S403 below, and will not be repeated here.

[0061] In another possible implementation, when determining the sorting interface drift coefficient, the interface drift assessment module 13 can also indirectly characterize the synchronicity of the light and heavy product coal quality data change curves by calculating the correlation coefficient or dynamic time regularization distance within the observation period. The lower the correlation coefficient or the larger the distance, the more significant the interface drift is determined, thereby deriving the sorting interface drift coefficient.

[0062] Therefore, by capturing the asynchronous nature of the changes in coal quality between light and heavy products, the interface drift assessment module 13 transforms the difficult-to-observe sorting interface position drift into a quantifiable sorting interface drift coefficient, thereby realizing the indirect online assessment of key indicators of sorting stability.

[0063] S204. Determine the dynamic response coupling factor based on the dynamic disturbance index of the sorting zone and the drift coefficient of the sorting interface.

[0064] For example, this step can be performed by the coupling factor calculation module 14 in the intelligent control system 10 described above. Specifically, the coupling factor calculation module 14 receives the sorting zone dynamic disturbance index from the dynamic disturbance evaluation module 12 and the sorting interface drift coefficient from the interface drift evaluation module 13. The coupling factor calculation module 14 performs a fusion calculation on these two indicators, and an exemplary calculation formula is as follows: in, Represents the dynamic response coupling factor. This represents the dynamic disturbance index of the sorting area. This represents the drift coefficient of the sorting interface. This represents the maximum-minimum normalization function, used to... The value is normalized to the interval [0,1]. The normalization process here is based on the dataset: the product of the sorting zone dynamic disturbance index and the sorting interface drift coefficient, calculated from a large amount of data collected over a historical period. The maximum and minimum values ​​are statistically selected based on this dataset (or set based on historical experience). The normalization result is then set to 1 when the normalized object is greater than the maximum value, and set to 0 when the normalized object is less than the minimum value.

[0065] It is understandable that the dynamic disturbance index of the sorting area... It reflects the propagation intensity and flow field disturbance amplitude of heavy medium density disturbance within the sorting zone, and is a dynamic disturbance index at the process level; the sorting interface drift coefficient This reflects the degree to which the disturbance ultimately causes the product output sorting interface position to deviate from the actual state, and is an output response index at the result level. The larger the value obtained by multiplying the two, the more significant the disturbance has been at both the process propagation and final result levels, the greater the degree to which the system deviates from the steady state, and the more urgent the need for control intervention. The purpose of normalization is to eliminate the influence of dimensions, making the dynamic response coupling factor a dimensionless proportionality coefficient that can be used to directly guide the adjustment quantity.

[0066] Therefore, the coupling factor calculation module 14 generates a dynamic response coupling factor that can comprehensively and quantitatively characterize the current abnormal state and control requirements of the system by coupling the dynamic disturbance index of the process layer with the output response index of the result layer, thus providing the core decision basis for the final precise adjustment.

[0067] S205. Adjust the current heavy medium density according to the dynamic response coupling factor.

[0068] For example, this step can be executed by the heavy medium density adjustment module 15 in the intelligent control system 10 described above. Specifically, the heavy medium density adjustment module 15 first determines whether the sorting interface is moving upward or downward based on the changing trend of the coal quality data of the sorted products (especially the ash content of the light products), thereby determining the adjustment direction of increasing or decreasing the heavy medium density. Then, the heavy medium density adjustment module 15 calculates the specific adjustment amount by multiplying the received dynamic response coupling factor according to the preset maximum adjustment range of the heavy medium density allowed by the system. Finally, the heavy medium density adjustment module 15 generates a target heavy medium density setpoint according to the determined adjustment direction and adjustment amount, combined with the current heavy medium density value, and converts this setpoint into a control command that can drive the actuator (such as the replenishing medium regulating valve, the diverter motor), controlling the heavy medium circulation system to perform the adjustment action in a step-like or proportional manner. It should be noted that the specific process of the aforementioned sub-steps can be found in S501-S503 below, and will not be repeated here.

[0069] In another possible implementation, when the heavy medium density adjustment module 15 adjusts the current heavy medium density according to the dynamic response coupling factor, it can also use historical multi-round adjustment data to train a feedforward prediction model. When calculating the current adjustment amount, it not only relies on the current dynamic response coupling factor, but also uses the model to predict the disturbance trend that may occur in the short term in the future, thereby providing advance compensation for the adjustment command to further improve the system's response speed and stability.

[0070] Therefore, the heavy medium density adjustment module 15 transforms the abstract coupling factor into a specific, executable heavy medium density adjustment command, and realizes real-time intervention in the sorting process through closed-loop execution, thereby correcting the coupling relationship between the flow field response and the interface drift.

[0071] Based on the above technical solution, this invention constructs a closed-loop intelligent control system encompassing data acquisition, dynamic disturbance assessment, interface drift assessment, coupling factor calculation, and density adjustment. First, it systematically collects and integrates multi-source time-series data reflecting medium circulation, flow field structure, and sorting results. Second, by quantitatively analyzing the dynamic mismatch of upstream density disturbances during transmission and the resulting flow field shift, and combining this with changes in the quality of sorted products to infer interface stability, it achieves a precise characterization of the dynamic coupling relationships within the process. Finally, based on the dynamic response coupling factor obtained from the comprehensive evaluation, it performs directional and quantitative intelligent adjustment of the heavy medium density. This method effectively overcomes the shortcomings of existing technologies that independently control process parameters and ignore cross-process disturbance transmission and time-series coupling, significantly improving the control accuracy, adaptability, and final product quality stability of the heavy medium coal preparation process when facing fluctuations in raw coal properties and equipment operating conditions.

[0072] For example, in another embodiment of the present invention, a fully intelligent control method for heavy medium coal preparation is provided, which determines the dynamic disturbance index of the separation zone based on upstream heavy medium density data, separation zone heavy medium density data, separation zone coal slurry flow data, and separation zone internal flow field data. Specifically, this includes the following steps: S301. Determine the propagation dynamic mismatch degree based on the upstream heavy medium density data and the heavy medium density data of the sorting area. The propagation dynamic mismatch degree is used to comprehensively characterize the time delay and amplitude distortion of the propagation of the upstream heavy medium density disturbance to the sorting area.

[0073] For example, this step can be performed by the propagation mismatch assessment submodule 121 in the dynamic disturbance assessment module 12, specifically including the following steps: (1) Identify the first feature dataset from the upstream heavy medium density data and the second feature dataset from the sorting area heavy medium density data.

[0074] Specifically, the propagation mismatch assessment submodule 121 first processes the upstream heavy medium density data sequence and calculates its rate of change sequence between multiple consecutive sampling times.

[0075] Subsequently, the propagation mismatch assessment submodule 121 employs a mutation point detection algorithm based on the peak value of the sorting difference: the change rate sequence is sorted in ascending order according to the absolute value to obtain an ordered sequence; the difference between adjacent values ​​in the ordered sequence is calculated; all differences are traversed, and the position where the largest difference occurs is taken as the split point. The right side of the ordered sequence (i.e., the part with a larger absolute value of the change rate) is identified as a subsequence reflecting a significant change in density, and this subsequence is labeled as the first feature dataset U.

[0076] Similarly, using the exact same algorithm, the heavy medium density data sequence of the sorting area is processed synchronously to identify and obtain the second feature dataset P.

[0077] (2) Determine the propagation delay index of heavy medium density perturbation based on the temporal correspondence between the first feature dataset and the second feature dataset.

[0078] Furthermore, after completing the identification of the first feature dataset U and the second feature dataset P, the propagation mismatch assessment submodule 121 needs to establish the correspondence between the two. The specific pairing rule is as follows: Based on the physical fact that there is a causal relationship between the upstream density perturbation and the sorting area, the first feature dataset U, identified in chronological order, is paired with the first feature dataset U. i Significant change points (corresponding to a rate of change) ), and the second feature dataset P that immediately follows in time and has a consistent trend (increase / decrease) i Significant change points (corresponding to a rate of change) Pair them up to form the first i Group corresponding data. If no corresponding point with a consistent trend is found in group P within a reasonable time window, the data in group U is considered unresponsive and is not included in subsequent calculations.

[0079] Subsequently, the propagation mismatch assessment submodule 121 calculates the time interval on the time axis for each set of data that was successfully paired as described above. Specifically, taking the earlier sampling point in each set of data as the reference, it calculates the time difference between the upstream density change point and the sorting zone density response point. Then, for all corresponding Normalization is performed to eliminate the influence of absolute time scales. Finally, the propagation mismatch assessment submodule 121 calculates the heavy medium density perturbation propagation delay exponent using the following formula. f : Where m represents the number of data groups contained in the first feature dataset U; Indicates the first i The difference between the time when the upstream feature change occurs and the time when the corresponding feature response occurs in the sorting area in the corresponding data of the successfully paired groups, and the result after normalization. This represents a pre-defined, small, non-zero constant used to guarantee performance in extreme cases. f The value is not zero to prevent subsequent... f The multiplication calculation involved results in a product of zero, for example... The value can be 10 to the power of negative cube. It should be noted that the above formula averages the time delay of upstream disturbance propagation to the sorting area. The larger the value of the summation term, the more significant the effect. i The slower the propagation of the disturbance, the more unstable it is. (Average value) f The larger the value, the slower and less efficient the process of upstream heavy medium density perturbation being transmitted to the sorting area throughout the entire observation period, and the more delayed the system's response to the perturbation.

[0080] (3) Determine the propagation dynamic mismatch degree based on the difference in the change magnitude of the corresponding data in the first feature dataset and the second feature dataset, as well as the propagation delay index of the heavy medium density perturbation.

[0081] For example, the propagation mismatch assessment submodule 121 calculates the propagation dynamic mismatch degree using the following formula. : Where f represents the propagation delay exponent of the heavy medium density perturbation calculated above; m represents the number of data groups in the first feature dataset U; In the first feature dataset U, the first feature is represented by the first feature. i The rate of change (i.e., slope) of the data set. This indicates that in the second feature dataset P, the data is related to the above... The rate of change of the characteristic response point of the successfully paired sorting zone; This represents a pre-defined, small, non-zero constant used to guarantee performance in extreme cases. The value is not zero to prevent subsequent... The multiplication calculation involved results in a product of zero, for example... It can take the value of 10 to the power of negative cube.

[0082] It should be noted that the above formula first calculates the average absolute deviation of the corresponding upstream and downstream disturbances in terms of their magnitude of change. This value reflects the degree of amplitude distortion during the propagation of the disturbance; the smaller the difference, the more consistent the response trend. Then, the average level of this amplitude deviation is compared with the time delay exponent. f Multiplication means assessing the physical effects of amplitude mismatch cumulatively over the dimension of time delay. Even with the same amplitude mismatch, if the propagation time is very long (f A larger value indicates a greater cumulative error and potential control risk to the system, thus resulting in a higher calculated propagation dynamic mismatch. The larger it gets, the bigger it becomes.

[0083] S302. Based on the coal slurry flow rate data and the internal flow field data of the sorting zone, determine the sorting zone response offset index. The sorting zone response offset index is used to characterize the degree of change in the internal flow field structure of the sorting zone caused by fluctuations in coal slurry flow rate.

[0084] For example, this step can be performed by the response offset evaluation submodule 122 in the dynamic disturbance evaluation module 12, and specifically includes the following steps: (1) Determine the average fluctuation intensity of coal slurry flow data in the sorting area within the preset observation period.

[0085] Specifically, the response offset assessment submodule 122 calculates the absolute value sequence of the difference between the flow rates of coal slurry at adjacent sampling times within the observation period. Then, it calculates the arithmetic mean of this absolute value sequence as the average fluctuation intensity, used to quantify the severity of changes in coal slurry flow rate throughout the entire observation period.

[0086] (2) Obtain flow velocity data at multiple spatial locations within the sorting zone.

[0087] Furthermore, the response offset evaluation submodule 122 extracts flow velocity data at at least two different spatial locations from the flow field data within the sorting zone. In one specific implementation, the flow field data within the sorting zone includes flow velocity data from multiple key areas within the sorting zone directly collected by flow velocity sensors. These key areas include at least two of the following: the sorting zone entrance area, the core sorting area, and the area adjacent to the sorting interface. The response offset evaluation submodule 122 directly reads real-time flow velocity information from a physical sensor network (such as ultrasonic flowmeters or electromagnetic flowmeters) deployed in one or more of the aforementioned key areas, using this information as the spatial location flow velocity data for subsequent calculations.

[0088] (3) Determine the sorting zone response offset index based on the average fluctuation intensity and the flow velocity data of the internal flow field data of the sorting zone at multiple spatial locations.

[0089] Furthermore, the response migration assessment submodule 122 first calculates the average flow velocity at each spatial location over the entire observation period. Then, it calculates the absolute differences between each pair of these average flow velocities, and finally averages all these differences to obtain the average spatial difference level, which characterizes the non-uniformity or migration degree of the flow field in the spatial structure. Finally, the sorting zone response migration index is calculated using the following formula. h : Where M represents the total number of spatial locations; and represents the average flow velocity (normalized) at the e-th and o-th spatial locations respectively; n represents the number of sampling points for the coal slurry flow data; and They represent the first j The sampling time, the first j +1 sampling time of coal slurry flow rate (normalized); This represents a pre-defined, small, non-zero constant used to guarantee performance in extreme cases. The value is not zero to prevent subsequent... The multiplication calculation involved results in a product of zero, for example... It can take the value of 10 to the power of negative cube.

[0090] It should be noted that the flow rate data mentioned above... , and traffic data , All data have been dimensionless during the data preprocessing stage. Therefore, all operational terms in the formula are dimensionless values, fundamentally ensuring dimensional consistency and avoiding computational logic problems caused by different physical units. The term in the first bracket of the formula aims to comprehensively evaluate the spatial non-uniformity of the flow field, which is divided by... M The design focuses on establishing a proportional relationship between the indicator and the number of monitoring points, facilitating relative comparisons between systems with different configurations; the second bracketed term characterizes the overall intensity of flow fluctuations. The product of the two yields... h , is a dimensionless index that comprehensively characterizes the degree of shift in the flow field structure caused by flow fluctuations.

[0091] Understandably, the first term in the above formula calculates the average velocity difference between all pairs of spatial locations, i.e., the average spatial difference level. A larger value indicates a more significant deviation of the flow field's spatial structure from a uniform distribution. The second term in parentheses represents the previously calculated average fluctuation intensity. Multiplying the two terms physically assesses the degree to which the flow field structure within the sorting zone is reconstructed or shifted under a given flow fluctuation context. h The larger the value, the more drastic the flow fluctuations, and the higher the spatial non-uniformity of the flow field. The combined effect of these two factors leads to a significant shift in the flow field, thus the larger the sorting zone response shift index, the greater the shift.

[0092] S303. Determine the dynamic disturbance index of the sorting zone based on the propagation dynamic mismatch degree and the sorting zone response offset index.

[0093] In this step, the dynamic disturbance assessment module 12 will use the propagation dynamic mismatch degree calculated by the propagation mismatch assessment submodule 121. The sorting zone response offset index h calculated by the response offset evaluation submodule 122 is directly multiplied to obtain the sorting zone dynamic disturbance index. .

[0094] Understandably, this relates to the dynamic mismatch in propagation. The sorting zone response offset index describes the degree of "distortion" (including time delay and amplitude) of upstream density disturbances during propagation. h This describes the degree of "reconstruction" of the flow field within the current sorting zone due to flow fluctuations. The product of the two... G It comprehensively reflects the total dynamic impact of upstream disturbances on the current sorting flow field state within the sorting zone after transmission. G The larger the value, the higher the degree to which the impact of upstream disturbances is transmitted and amplified, and the greater the potential threat to the stability of the sorting area.

[0095] Based on the above technical solution, this invention quantifies the temporal and amplitude distortion (propagation dynamic mismatch) of upstream density disturbances during the medium circulation process and the spatial reconstruction of the sorting zone flow field caused by coal slurry flow fluctuations (sorting zone response offset index). Ultimately, these two are coupled to form a sorting zone dynamic disturbance index that comprehensively characterizes the dual effects of "propagation distortion" and "on-site response." This index no longer reflects a single parameter but profoundly depicts the coupling relationship between cross-process disturbance propagation and the fluid dynamic response within the sorting zone, providing crucial process state input for subsequent accurate assessment of sorting interface stability and implementation of precise density control.

[0096] For example, in another embodiment of the present invention, a fully intelligent control method for heavy media coal preparation is provided, which determines the separation interface drift coefficient based on the coal quality data of the separation products. This specifically includes the following steps: S401. Based on the changing trend of the light product coal quality data in the sorted product coal quality data over time, determine the first change characteristic sequence.

[0097] Specifically, the interface drift assessment module 13 receives light product coal quality data from the data acquisition module 11. In the heavy media coal preparation process, light product coal quality data typically uses ash content as the core indicator, which is continuously acquired by an online ash analyzer located at the clean coal (light product) outlet. The interface drift assessment module 13 first preprocesses the light product ash content time series within a preset observation period. Then, the interface drift assessment module 13 calculates the rate of change of this series between each pair of adjacent sampling times, i.e., the slope value. Specifically, for the x-th point in the time series, its slope value... The slope is calculated by dividing the difference in gray values ​​between two adjacent sampling times by the sampling time interval. All calculated slope values ​​are arranged in chronological order to form the first change characteristic sequence.

[0098] S402. Based on the trend of the change of heavy product coal quality data in the sorted product coal quality data over time, determine the second change characteristic sequence.

[0099] Similarly, the interface drift assessment module 13 synchronously processes the heavy product coal quality data (mainly ash content) from the heavy product outlet. Using the same algorithm as S401, it calculates the slope value of the heavy product ash content time series between each pair of adjacent sampling times. .all Arranged chronologically, they form the second sequence of change characteristics. It should be noted that, to ensure that the two sequences correspond strictly on the time axis, the interface drift evaluation module 13 uses a unified sampling time reference when calculating the slope.

[0100] S403. Determine the sorting interface drift coefficient based on the difference in characteristic values ​​between the first change characteristic sequence and the second change characteristic sequence at the same time.

[0101] Furthermore, the interface drift evaluation module 13 compares the feature values ​​(i.e., slope values) of the first change feature sequence and the second change feature sequence at the same time position (i.e., the same slope value index x). The interface drift evaluation module 13 calculates the sorting interface drift coefficient using the following formula. d : Where N represents the total number of slope values ​​that can be calculated from the light and heavy coal quality data sequences during the entire prediction observation period (i.e., the length of the first or second change characteristic sequence). This represents the slope value at the x-th position (i.e., the x-th group of adjacent sampling times) in the first change feature sequence; This represents the slope value of x at the same position in the second change feature sequence; This represents a pre-defined, small, non-zero constant used to guarantee performance in extreme cases. The value is not zero to prevent subsequent... The multiplication calculation involved results in a product of zero, for example... It can take the value of 10 to the power of negative cube.

[0102] It should be noted that the above formula first calculates the instantaneous difference in the ash content change trends of the light and heavy products at corresponding times for each group. This difference directly reflects the synchronicity of changes in the quality of clean coal and gangue at a specific moment. If the separation interface is stable, the light and heavy products should respond synchronously to changes in raw coal or operating conditions, and this difference should be close to zero; if the difference is large, it indicates that the two products are not changing synchronously, and interface drift exists. Then, the instantaneous differences at all times are averaged to obtain the average drift intensity over the entire observation period, i.e., the separation interface drift coefficient. d . dThe larger the value, the more persistent and significant the asynchronous phenomenon of changes in the quality of light and heavy products is throughout the entire observation process, that is, the more severe the average degree of drift at the sorting interface.

[0103] Based on the above technical solution, this invention extracts the instantaneous trend characteristics of the changes in the coal quality of light and heavy products, respectively, and quantifies the trend difference between the two at the same time, ultimately calculating the sorting interface drift coefficient. This solution cleverly utilizes product quality data that is easily monitored online to indirectly but effectively characterize the stability of the sorting interface position, which is difficult to measure directly. This transforms the key but implicit process state of "interface drift" into a clear and quantifiable indicator. This provides crucial result-level feedback signals for the entire process intelligent control system, enabling the control system to perform closed-loop adjustments based on the direct characterization of the sorting effect, significantly enhancing the perception and control capabilities regarding the stability of the sorting process.

[0104] For example, in another embodiment of the present invention, a fully intelligent control method for heavy medium coal preparation is provided, which adjusts the current heavy medium density according to the dynamic response coupling factor, specifically including the following steps: S501. Based on the changing trend of the coal quality data of the sorted products, determine the direction of adjustment of the heavy medium density.

[0105] In this step, the heavy medium density adjustment module 15 receives the coal quality data of the sorted products from the data acquisition module 11, and focuses on analyzing the changing trend of the light product coal quality data (such as ash content) over time. Specifically, the heavy medium density adjustment module 15 calculates the overall average slope of the light product coal quality data sequence within a preset observation period, denoted as... The determination of the adjustment direction follows these rules: when When the value is positive, it indicates that the ash content of the light product is increasing. This usually means that the separation interface has shifted downwards within the separation zone, and more low-density clean coal has entered the heavy product. In this case, it is necessary to increase the density of the heavy medium (i.e., increase the amount of supplementary medium) to make the separation interface shift upwards back to the target position. Therefore, the adjustment direction is to increase. Conversely, when the value is negative... When the value is negative, the adjustment direction is decrease. This judgment logic can be formally expressed using a sign function. This indicates that the output value of the function (+1 or -1) directly corresponds to the direction of adjustment.

[0106] S502. Determine the adjustment amount based on the preset maximum allowable adjustment range of heavy medium density and the dynamic response coupling factor.

[0107] Specifically, the heavy medium density adjustment module 15 first reads a preset maximum allowable adjustment range for the heavy medium density from the system configuration or control strategy, denoted as . This parameter defines the maximum absolute value that the density of the heavy medium can be changed during a single adjustment, serving as a safety constraint to ensure system stability and prevent over-adjustment. An exemplary rule for its value is: it is determined comprehensively based on the process equipment characteristics, media system stability, and product quality control requirements of the specific coal preparation plant. For example, it can be set to ±5% of the density setpoint (i.e., if the set density is 1.45 g / cm³, then...). (Approximately 0.0725 g / cm³), to ensure that the adjustment is both effective and stable.

[0108] Subsequently, the heavy medium density adjustment module 15 receives the dynamic response coupling factor from the coupling factor calculation module 14. w (Its value has been normalized, for example, between 0 and 1). Adjustment amount Calculated using the following formula: in, w This represents the dynamic response coupling factor, calculated from S204; This indicates the maximum allowable adjustment range for the preset heavy medium density. It can be understood that this represents the normalized coupling factor. w As a proportional coefficient for adjusting intensity. When the system state is close to ideal ( w When the disturbance is close to 0, the required adjustment is very small; when the disturbance is significant and the interface drift is severe ( w When the value approaches 1), an adjustment amount close to the maximum allowable value is enabled. This achieves an adaptive match between the adjustment strength and the degree of system anomaly.

[0109] S503. Generate a target heavy medium density value according to the adjustment direction and the adjustment amount, and control the heavy medium circulation system to perform adjustment.

[0110] Furthermore, the heavy medium density adjustment module 15 combines the adjustment direction and adjustment amount to calculate the final target heavy medium density value. This is specifically achieved through the following formula: in, This represents the calculated target heavy medium density value; This indicates the current heavy medium density value before adjustment was performed; This determines the direction of adjustment (+1 for increase, -1 for decrease). The adjustment amount is determined in S502.

[0111] It should be noted that the above formula is based on the current density (from...). Based on the representation, a directional adjustment value is superimposed to obtain a new setpoint. Subsequently, the heavy medium density adjustment module 15 converts it into process commands (such as valve opening signals and motor speed settings) that can be received by the field actuators (such as replenishment valves and diversion box controllers). To avoid instantaneous impact on the system, this module controls the actuators to smoothly approach the target value in a step-like or proportional manner. After the adjustment command is issued, the system enters the next observation and control cycle, thus forming a closed loop.

[0112] Based on the above technical solution, this invention achieves directional and quantitative intelligent adjustment of heavy medium density by organically combining trend analysis of sorting effect, system state coupling factor, and safety adjustment constraints. This method ensures that each density adjustment has a clear purpose (correcting the direction of interface drift) and precise dosage (adjusting proportionally according to the degree of anomaly), fundamentally avoiding frequent oscillations or overcompensation based on instantaneous deviations in traditional control, and significantly improving the stability, accuracy, and final product quality stability of the control process.

[0113] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0114] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A full-process intelligent control method applied to heavy medium coal separation, characterized in that, include: Acquire process monitoring data of the heavy media coal preparation process within a preset observation period; The process monitoring data includes upstream heavy medium density data, heavy medium density data of the sorting zone, coal slurry flow data of the sorting zone, internal flow field data of the sorting zone, and coal quality data of the sorting products; Based on the upstream heavy medium density data, the heavy medium density data of the sorting zone, the coal slurry flow rate data of the sorting zone, and the internal flow field data of the sorting zone, the dynamic disturbance index of the sorting zone is determined; wherein, the dynamic disturbance index of the sorting zone is used to characterize the comprehensive impact of the upstream heavy medium density disturbance on the state of the partial sorting flow field in the sorting zone after it is transmitted to the sorting zone through medium circulation. Based on the coal quality data of the sorted products, the sorting interface drift coefficient is determined; wherein, the sorting interface drift coefficient is used to characterize the degree of displacement of the sorting interface between clean coal and gangue caused by the change of the sorting flow field state. The dynamic response coupling factor is determined based on the dynamic disturbance index of the sorting zone and the drift coefficient of the sorting interface. The current heavy medium density is adjusted according to the dynamic response coupling factor.

2. The full-process intelligent control method for dense medium coal separation according to claim 1, characterized in that, Based on the upstream heavy medium density data, the heavy medium density data of the sorting zone, the coal slurry flow rate data of the sorting zone, and the internal flow field data of the sorting zone, the dynamic disturbance index of the sorting zone is determined, specifically including: Based on the upstream heavy medium density data and the sorting area heavy medium density data, the propagation dynamic mismatch is determined; wherein, the propagation dynamic mismatch is used to comprehensively characterize the timing delay and amplitude distortion of the propagation of the upstream heavy medium density disturbance to the sorting area; Based on the coal slurry flow rate data and the internal flow field data of the sorting zone, the sorting zone response offset index is determined; the sorting zone response offset index is used to characterize the degree of change in the internal flow field structure of the sorting zone caused by fluctuations in coal slurry flow rate. The sorting zone dynamic disturbance index is determined based on the propagation dynamic mismatch degree and the sorting zone response offset index.

3. The full-process intelligent control method for dense medium coal separation according to claim 2, characterized in that, Based on the upstream heavy medium density data and the sorting zone heavy medium density data, the propagation dynamic mismatch is determined, specifically including: Identify a first feature dataset from the upstream heavy medium density data and identify a second feature dataset from the sorting zone heavy medium density data; The propagation delay index of heavy medium density perturbation is determined based on the temporal correspondence between the first feature dataset and the second feature dataset. The propagation dynamic mismatch is determined based on the difference in the magnitude of change of corresponding data in the first feature dataset and the second feature dataset, as well as the propagation delay index of the heavy medium density perturbation.

4. The intelligent control method for the entire process of heavy media coal preparation according to claim 2, characterized in that, Based on the coal slurry flow rate data and the internal flow field data of the sorting zone, the sorting zone response offset index is determined, specifically including: Determine the average fluctuation intensity of the coal slurry flow data in the sorting area during the preset observation period; Obtain flow velocity data at multiple spatial locations within the sorting zone; The sorting zone response offset index is determined based on the average fluctuation intensity and the flow velocity data of the internal flow field data of the sorting zone at multiple spatial locations.

5. The intelligent control method for the entire process of heavy media coal preparation according to claim 4, characterized in that, The flow field data inside the sorting zone includes flow velocity data of multiple key areas inside the sorting zone directly collected by a flow velocity sensor; the key areas include at least two of the sorting zone entrance area, core sorting area, and adjacent area of ​​the sorting interface.

6. The intelligent control method for the entire process of heavy media coal preparation according to claim 1, characterized in that, Based on the coal quality data of the separated products, the separation interface drift coefficient is determined, specifically including: Based on the changing trend of the light product coal quality data in the sorted product coal quality data over time, a first change feature sequence is determined; Based on the changing trend of heavy product coal quality data in the sorted product coal quality data over time, a second change feature sequence is determined; The sorting interface drift coefficient is determined based on the difference in feature values ​​between the first change feature sequence and the second change feature sequence at the same time.

7. The intelligent control method for the entire process of heavy media coal preparation according to claim 1, characterized in that, Adjusting the current heavy medium density according to the dynamic response coupling factor specifically includes: Based on the changing trend of the coal quality data of the sorted products, the direction of adjustment of the heavy medium density is determined; The adjustment amount is determined based on the preset maximum allowable adjustment range of the heavy medium density and the dynamic response coupling factor; According to the adjustment direction and the adjustment amount, a target heavy medium density value is generated and the heavy medium circulation system is controlled to perform adjustment.

8. The intelligent control method for the entire process of heavy media coal preparation according to claim 7, characterized in that, The method further includes: The process monitoring data is periodically acquired during multiple consecutive preset observation periods following the preset observation period. Continuous closed-loop regulation is implemented based on periodically acquired process monitoring data.

9. The intelligent control method for the entire process of heavy media coal preparation according to claim 3, characterized in that, Identifying the first feature dataset from the upstream heavy medium density data specifically includes: Calculate the rate of change sequence of the upstream heavy medium density data sequence at multiple consecutive time points; Sort the rate of change sequence by absolute value; Based on the distribution of differences between adjacent data in the sorted rate of change sequence, the subsequences with significant changes are identified as the first feature dataset.

10. The intelligent control method for the entire process of heavy media coal preparation according to any one of claims 1-9, characterized in that, Obtain process monitoring data for the heavy media coal preparation process within a preset observation period, specifically including: The upstream heavy medium density data is obtained by sensors installed at key upstream nodes of the heavy medium circulation system. The heavy medium density data and the coal slurry flow rate data of the sorting zone are obtained by sensors installed at the inlet of the sorting equipment or inside the sorting zone. Sensors installed at multiple monitoring points within the sorting zone are used to acquire flow field data within the sorting zone. The coal quality data of the sorted products are obtained by using online coal quality testing devices installed at the clean coal outlet and gangue outlet. The acquired upstream heavy medium density data, the heavy medium density data of the sorting zone, the coal slurry flow rate data of the sorting zone, the internal flow field data of the sorting zone, and the coal quality data of the sorted products are synchronized and marked according to a unified time base to form the process monitoring data.

Images