Belt conveying automatic dust suppression system for multi-working-condition adjustment

Abstract

The invention discloses a belt conveying automatic dust suppression system for multi-working-condition adjustment, relates to the technical field of industrial automation centralized control, and is used for solving the problem of poor belt dust suppression scheduling under multiple working conditions. According to the method, a closed-loop regulation and control system of global state atlas and disturbance fragment recognition is constructed, dust abnormity, belt speed abrupt change and wind speed and humidity fluctuation are quickly recognized, and a dust suppression unit set influenced by disturbance is determined based on propagation verification; a programmed scheduling scheme is generated through constraint modeling and feasible solution screening, safety interlocking, resource boundaries and start-stop intervals are considered, spraying, suction and sealing are linked in a minimum range, local degradation control and centralized rolling rescheduling are combined, unification of on-site disposal and global optimization is achieved, and the scheduling efficiency is improved. The self-adaptability and the dust suppression effect of the system can be further improved, ineffective starting of the whole line is avoided, the fault risk is reduced, and stable operation is kept.

Description

Technical Field

[0001] This invention relates to the field of centralized control technology for industrial automation, and more specifically, to an automatic dust suppression system for belt conveyors oriented towards multi-condition adjustment. Background Technology

[0002] Belt conveyors, as the main material transfer method in bulk material industries such as mining, steel, cement, and coal, have brought serious dust pollution problems while improving transportation efficiency and reducing labor costs. Especially at material discharge points, transfer stations, and transshipment points, large-scale dust can easily be generated due to the impact of falling materials, airflow disturbance, and high-speed operation of the conveyor belt. This not only affects the visibility of the working environment and the life of equipment, but also poses a threat to workers' health, becoming a key issue for occupational diseases and environmental supervision. To control dust spread, the industry has widely adopted methods such as spray dust suppression, enclosed enclosures, and dust extraction fans.

[0003] The shortcomings of existing technologies: Belt conveyors easily generate a large amount of dust at the material discharge port and transfer station, affecting the working environment and equipment lifespan. Existing dust suppression methods include spraying, enclosed hoods, and dust removal fans, but they have obvious shortcomings. For example, spraying is mostly designed with fixed parameters, making it difficult to adaptively adjust with changes in transport volume and wind speed, resulting in insufficient dust suppression or excessive water spraying. Enclosed structures are prone to gaps due to belt misalignment or wear during long-term operation, causing dust leakage. Moreover, dust removal fans are limited by fixed air volume control, which can easily lead to excessive energy consumption or insufficient negative pressure. More importantly, these dust suppression units mostly operate independently, lacking modeling and centralized scheduling of the linkage relationship between multiple conveyors and multiple dust suppression devices, making it difficult to achieve global optimization control under complex and variable working conditions. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the following solution is proposed to solve the problem of poor belt dust suppression scheduling under multiple working conditions in the above-mentioned background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An automatic dust suppression system for belt conveyors with multi-condition adjustment includes a conveying data acquisition module, a condition identification module, a scheduling scheme generation module, a scheduling control module, and a feedback update module. The modules are connected by signals.

[0007] The data acquisition module is used to construct a global status map based on dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate data collected from the conveyor link nodes, and to complete synchronization under a unified clock.

[0008] The working condition identification module is used to identify the working conditions of the global state map, divide it into steady-state segments and disturbance segments, and determine the set of dust suppression units affected by disturbances.

[0009] The scheduling scheme generation module is used to perform constraint modeling and solution of the start-up and shutdown sequence, execution timing and action parameters of the dust suppression unit through the centralized controller, and generate a programmed scheduling scheme.

[0010] The scheduling and control module is used to execute spraying, suction or closure operations on each dust suppression unit according to the scheduling plan, and to execute local degradation control when abnormal conditions are triggered.

[0011] The feedback update module is used to verify the effect of the controller based on the execution feedback, and to update the scheduling scheme in a rolling manner when a deviation is detected, so as to realize optimized control under multiple working conditions.

[0012] Furthermore, the methods for establishing the global state graph include:

[0013] Define the nodes of the conveying section, the nodes of the transfer point, and the nodes of the dust suppression unit according to the conveying route and the transfer relationship, and establish the topological connection; calibrate the dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow sensors deployed at each node, collect data and set the time with a unified clock;

[0014] A sliding window is used to align the time of multi-source data, perform anomaly removal and missing measurement interpolation, and complete the unification of units.

[0015] The alignment data is mapped to the corresponding nodes or connecting edges according to their spatial location and device identifier, generating node state vectors and edge state vectors.

[0016] The operating condition derived indicators are calculated based on the node state vector and written into the node attributes to form an attribute set with a time interval.

[0017] The attribute set is incrementally stored and maintained using a graph data structure, and a global state graph is output.

[0018] Furthermore, it is used for operating condition identification of the global state map, including:

[0019] Read the node state vectors and edge state vectors sorted by time from the global state graph;

[0020] Within a unified sliding window, variable point detection is performed on the dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate sequences, and the start and end times are located.

[0021] Consistency checks are performed on adjacent nodes and upstream and downstream connections to eliminate isolated fluctuations and retain continuous changes across nodes.

[0022] Based on the change point location and consistency verification results, the operation process is divided into steady-state segments and disturbance segments;

[0023] Determine the set of dust suppression units affected by disturbance along the conveying topology;

[0024] Generate operating condition labels containing time intervals and impact ranges and write them into the global state map.

[0025] Furthermore, based on the change point location and consistency verification results, the operation process is divided into steady-state segments and disturbance segments, including:

[0026] Based on the variable point localization results, the variable point boundaries are sorted in ascending order by time and duplicate variable point boundaries are removed to form an initial segmentation point sequence; the time axis is divided into continuous candidate segments by adjacent segmentation points.

[0027] For each candidate segment, the dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate are judged within a unified sliding window based on the preset stable band and hysteresis band, and the duration of deviation is recorded.

[0028] If a candidate segment has a threshold deviation that satisfies the minimum duration and changes in the same direction at the upstream or downstream related nodes and passes the consistency check, the segment is marked as a perturbation segment; otherwise, it is marked as a steady-state segment.

[0029] Furthermore, the set of dust suppression units affected by the disturbance was identified, including:

[0030] Starting from the start and end times and occurrence nodes of the disturbance segment, candidate paths are generated along the transport topology and material flow direction;

[0031] Perform time-delay propagation verification on nodes and connecting edges on candidate paths within a unified sliding window, requiring that the order and delay of parameter changes fall within a preset propagation time window;

[0032] Based on wind direction or negative pressure direction, the opening and closing status of the enclosure and the opening and closing status of the valve, propagation gating and blocking judgment are performed to eliminate blocked paths and isolated nodes.

[0033] The remaining nodes are subjected to adjacent consistency and repeatability verification to eliminate occasional fluctuations;

[0034] The set of dust suppression units that pass propagation verification, gating judgment and consistency verification is marked as the set affected by disturbance and written into the global state map.

[0035] Furthermore, it is used to perform constraint modeling and solution of the start-up and shutdown sequence, execution timing, and action parameters of the dust suppression units through a centralized controller, generating a programmed scheduling scheme, including:

[0036] Establish decision variables and define a discrete set of start / stop flags, start and end gaps, and action parameters for each dust suppression unit;

[0037] Set hard constraints, including safety interlocks, mutual exclusion or coexistence rules, resource boundary conditions, equipment cooling and minimum intervals, maximum number of parallel actions, and time windows and their order of execution;

[0038] Set the protection and recovery conditions as state constraints, and map scheduling requirements and coverage windows according to the set of disturbances and the working condition labels;

[0039] Construct a constraint satisfaction model and solve it to obtain a feasible instruction set; select a solution from the feasible solutions according to the sorting rules, in order of priority to cover the disturbed window, the fewest number of actions, the minimum switching frequency and the shortest total execution time;

[0040] Generate a programmed scheduling scheme that includes unit identifiers, timestamps, and action parameters, and send it to the field controller.

[0041] Furthermore, it is used to perform spraying, suction, or sealing operations on each dust suppression unit according to the scheduling plan, and to perform local degradation control when abnormal conditions are triggered, including:

[0042] Load the scheduling scheme and complete the parameter mapping and verification with local actuators and sensors;

[0043] The execution queue is generated by parsing the time slot scheduling scheme and locking the associated lock quantity and resource usage;

[0044] Before execution, conduct pre-inspection of safety interlocks and resource thresholds such as water pressure, flow rate, and negative pressure air volume, and complete pre-charging or pre-draining according to the preset procedures.

[0045] Commands for opening and closing the spray valve and setting the flow rate, starting and stopping the negative pressure fan and setting its speed, and opening and closing the enclosure and setting its position are issued through the local controller.

[0046] During the execution process, water pressure, flow rate, fan current, conveyor belt speed, and dust concentration are periodically collected and compared with the monitoring conditions.

[0047] When abnormal conditions such as water pressure drop, fan overflow, sudden drop in conveyor belt speed or abnormal dust rebound are detected, local degradation control is triggered, which prioritizes shutting down the spray, reducing suction or switching to a safety sub-strategy, and performing an emergency shutdown when necessary.

[0048] When the anomaly is resolved and the recovery conditions are met, the relevant actuators are reset in a controlled sequence and the system returns to the current effective time slot or enters a waiting state.

[0049] Furthermore, it is used to verify the effect through the controller based on execution feedback, and to continuously update the scheduling scheme when deviations are detected, thereby achieving optimized control under multiple operating conditions, including:

[0050] Collect the execution timestamps, status codes, and feedback parameters of water pressure, flow rate, fan current, dust concentration, and conveyor belt speed reported by each dust suppression unit, and align them with the target time slots and coverage windows of the programmed scheduling scheme;

[0051] Construct an effect verification vector to determine whether the perturbation fragment coverage is complete, whether the recovery time exceeds the limit, whether the resource consumption is abnormal, and whether the protection condition is triggered.

[0052] When a deviation is detected, a rolling rescheduling is triggered, and a feasible set of instructions that meets the hard constraints and timing constraints is searched within the prediction time window.

[0053] Based on the sorting rules, the scheme that best covers the disturbed window and has the fewest number of actions and switching frequencies is selected from the feasible solution set to generate an updated programmed scheduling scheme.

[0054] The technical effects and advantages of the automatic dust suppression system for belt conveyors with multi-condition adjustment according to the present invention are as follows:

[0055] This invention achieves rapid identification and response to various disturbances during belt conveyor transport, such as abnormal dust concentration, sudden changes in conveyor belt speed, and fluctuations in wind speed and humidity, by constructing a closed-loop control system based on a global state map and disturbance segment identification. Based on node state vectors and a propagation verification mechanism, the set of dust suppression units affected by disturbances is determined, significantly improving the system's adaptability to dynamic environmental changes. Compared with traditional spray control methods that rely on fixed timing or human experience, this invention can quickly adjust spraying, suction, and sealing actions within a minimal range, avoiding large-scale ineffective startups across the entire line and ensuring the targeted and energy-efficient nature of the dust suppression strategy.

[0056] Furthermore, this invention introduces constraint modeling and feasible solution screening mechanisms, comprehensively considering safety interlocks, resource boundaries, start-stop intervals, and parallel constraints to construct a programmed scheduling scheme under multiple operating conditions. This effectively improves the rationality of action sequence and resource allocation, avoiding the risk of secondary failures caused by insufficient water pressure or fan overload. Through local degradation control and centralized rescheduling coordination, a dual-layer closed-loop control of rapid on-site safety handling and centralized optimized scheduling is achieved, thereby ensuring the continuity and reliability of dust suppression response. Through rolling feedback updates and verification mechanisms, this invention further enhances the dynamic correction capability of the scheduling scheme, enabling the system to maintain stable dust suppression effects, controllable energy consumption, and high operational safety in environments with frequent multi-condition disturbances, such as ports and raw material storage yards. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of the structure of an automatic dust suppression system for belt conveyors with multi-condition adjustment according to the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] In order to achieve the above objectives, Figure 1 A schematic diagram of the structure of an automatic dust suppression system for belt conveyors with multi-condition adjustment according to the present invention is given. Specifically, it includes a conveying data acquisition module, a condition identification module, a scheduling scheme generation module, a scheduling control module, and a feedback update module. The modules are connected to each other through signals.

[0060] The data acquisition module is used to construct a global status map based on dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate data collected from the conveyor link nodes, and to complete synchronization under a unified clock.

[0061] The working condition identification module is used to identify the working conditions of the global state map, divide it into steady-state segments and disturbance segments, and determine the set of dust suppression units affected by disturbances.

[0062] The scheduling scheme generation module is used to perform constraint modeling and solution of the start-up and shutdown sequence, execution timing and action parameters of the dust suppression unit through the centralized controller, and generate a programmed scheduling scheme.

[0063] The scheduling and control module is used to execute spraying, suction or closure operations on each dust suppression unit according to the scheduling plan, and to execute local degradation control when abnormal conditions are triggered.

[0064] The feedback update module is used to verify the effect of the controller based on the execution feedback, and to update the scheduling scheme in a rolling manner when a deviation is detected, so as to realize optimized control under multiple working conditions.

[0065] The specific implementation details of the data acquisition module are as follows:

[0066] On a belt conveyor line covering the raw material stockpile to the crushing station, firstly, based on the actual conveying route and transfer relationship, each straight conveyor belt segment is determined as a conveying segment node, each drop, transfer, or merging point is determined as a transfer point node, and the spray valve group, negative pressure fan, and enclosure directly associated with each transfer point are determined as dust suppression unit nodes. A unique identifier and spatial coordinates are assigned to each of the above three types of nodes. A directed topological connection is established according to the material flow direction between the conveying segment node and the transfer point node, and the spray pipe or suction pipe directly associated with the connection is recorded in the connection attribute.

[0067] Subsequently, dust concentration sensors, conveyor belt speed sensors, ambient wind speed and humidity sensors, and water pressure and flow sensors are deployed at each node. The conveyor belt speed sensors are installed at the corresponding conveyor section nodes. When the transfer point node and the dust suppression unit node obtain the conveyor belt speed, they use the conveyor belt speed sensor readings of the upstream or downstream conveyor section nodes directly associated with them and assign the readings to that node. Zero-point calibration and range calibration are performed on all sensors. Linearization is performed using the calibration curve provided by the manufacturer, and the calibrated sampled values ​​are bound to the unique identifier of the corresponding node. The system uses a unified clock as the time reference, and all sampled values ​​are written with timestamps at the acquisition end to achieve unified clock timing.

[0068] After the data arrives at the centralized controller, a sliding window is used for time alignment. The window length and sliding step size of the sliding window are fixed before the system is put into operation and remain consistent throughout the system. For multiple sampling records falling into the same time window, they are arranged in chronological order and the average value within the window is used as the representative value. When there is only a single sampling record at the end of the window, that record is used as the representative value. Missing measurement processing follows the principle of prioritizing the previous valid window of the node, followed by adjacent points of the same type of physical quantity. When a representative value of a certain type of sensor is missing in the window, the representative value of the previous window of the node is first searched and filled forward. If there is no representative value in the previous window, the representative value of another sensor in the same window that is directly related to the same type of physical quantity is used as a temporary estimate. For example, the temporary estimate of water pressure and flow rate can be the representative value of water pressure and flow rate recorded at the inlet of the pipeline directly connected to the node.

[0069] Anomaly rejection is based on two criteria: first, the range criterion, where a sampled value exceeding the calibrated measurable range is considered an anomaly and rejected; second, the rate of change criterion, where a sudden change in the sampled value between two adjacent windows that does not conform to the equipment's physical limits is considered an anomaly and rejected. A fixed unit system is used for all measurements: dust concentration is standardized to mass concentration, conveyor belt speed to length per unit time, ambient wind speed and humidity to length per unit time and relative humidity percentage, respectively, and water pressure and flow rate to pressure and volume per unit time, respectively. When the original unit of a sampled value differs from the above units, unit conversion is performed item by item according to the conversion relationships provided in the sensor's instruction manual.

[0070] After time alignment and data cleaning, representative values ​​are mapped to corresponding nodes or connecting edges according to spatial location and device identifier: For each conveyor segment node, transfer point node, and dust suppression unit node, node state vectors are formed in a fixed order, namely dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate; For each connecting edge, an edge state vector is formed. In the edge state vector, the dust concentration is obtained by subtracting the representative dust concentration of the upstream node from the representative dust concentration of the same window of the downstream node, the conveyor belt speed is obtained by taking the representative value of the upstream conveyor segment node, the ambient wind speed and humidity are obtained by taking the representative value of the ambient wind speed and humidity sensor deployed along the path of the connecting edge, and the water pressure and flow rate are obtained by taking the representative value of the pipeline directly associated with the connection in the window. The above mapping process is completed window by window. After each mapping is completed, the node state vector and edge state vector of the window and the start and end times of the corresponding time window are written into the temporary cache in memory.

[0071] Based on the node state vector, operating condition derived indices are calculated and written into the node attributes to form an attribute set with a time interval. The operating condition derived indices include the following defined items: First, stability indices, namely, the magnitude of change of four parameters—dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate—relative to the representative values ​​of the previous window within a given window, and the duration of this change within consecutive windows; second, consistency indices, namely, whether the direction of change of the four parameters of the same node is consistent within the same window, and whether it is consistent with the direction of change of adjacent nodes in the material flow direction; third, support indices, namely, whether the correspondence between water pressure and flow rate and the spray execution state remains responsive within consecutive windows. All of the above indices are recorded in text: the magnitude of change is represented by three symbols: rising, falling, and basically unchanged; the duration is represented by the number of consecutive windows; consistency is represented by matching and non-matching; and support is represented by responding and not responding. The resulting node attributes, together with the aforementioned node state vector, constitute the attribute set of that node within that time interval.

[0072] A graph data structure is used to incrementally store and maintain the attribute set to output a global state graph. The graph data structure consists of three parts: a node set, a connection edge set, and an attribute set. The node set stores the unique identifiers and spatial coordinates of the conveying section nodes, transfer point nodes, and dust suppression unit nodes. The connection edge set stores the directed connections of material flow and their correspondence with the spray pipes or suction pipes. The attribute set stores the node state vectors, edge state vectors, and node attributes indexed by time windows. Incremental maintenance writes new attribute sets step by step according to the time window order, while retaining at least two historical windows to support cross-window consistency judgment. When the set limit for the number of historical windows to be retained is reached, the attribute set of the earliest window is removed in a first-in-first-out manner. The output of the global state graph is provided in the form of an interface. Each time, the interface returns the node set, connection edge set, and attribute set of the current window and the preset number of review windows.

[0073] The specific implementation details of the working condition identification module are as follows:

[0074] The data acquisition module defines the transport section nodes, transfer point nodes, and dust suppression unit nodes according to the transport route and transfer relationship, and establishes a topological connection, assigning a unique identifier and spatial location to each node. Dust concentration sensors, conveyor belt speed sensors, ambient wind speed and humidity sensors, and water pressure and flow sensors are deployed at each node and calibrated. The collected data is timed at the acquisition end using a unified clock and then sent to the central controller. The central controller uses a unified sliding window for time alignment.

[0075] Within each time window, the representative value of the window is obtained by taking multiple sampling records of the same physical quantity at the same node in chronological order. When a physical quantity of a certain node is missing in the current window, the representative value of the previous window is used to fill the gap. If it is still missing, the representative value of the physical quantity of the pipeline inlet or upstream node directly connected to the node is used as a temporary estimate. Anomaly removal is performed according to the range criterion and the rate of change criterion. After unifying the dimensions, the representative value is mapped to the corresponding node or connecting edge according to the spatial location and equipment identifier to form the node state vector and the edge state vector. Based on this, the operating condition derived index is calculated and written into the node attribute. The operating condition derived index includes stability index, consistency index and support index. The stability index is used to describe the change range of the four types of parameters and the number of continuous windows when compared with the previous window.

[0076] Consistency indicators are used to describe whether the change direction of the four types of parameters is consistent within the same node and between adjacent nodes in the material flow direction. Support indicators are used to describe the response correspondence between water pressure and flow rate and spray execution status within a continuous window. Finally, the attribute set is incrementally stored and maintained using a graph data structure, and a global state map is output for subsequent use.

[0077] Operating condition identification involves processing the global state map:

[0078] The system reads the node state vectors and edge state vectors sorted by time from the global state graph. Within a unified sliding window, it performs change point detection on the dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate sequences to locate the start and end times. The change point detection is based on the node state vectors and node attributes. The judgment rule is that when the representative value of any physical quantity deviates from the preset stable zone and continues to reach the minimum duration outside the hysteresis zone, the moment is recorded as a candidate change point. The stable zone is determined by the normal fluctuation range of the equipment and the statistics of historical stable operation. The hysteresis zone is a fixed buffer zone outside the boundary of the stable zone to avoid frequent switching due to instantaneous swing.

[0079] Subsequently, consistency verification is performed on the node where the candidate variable point is located and its upstream and downstream connecting edges: if the same type of physical quantity of adjacent nodes or connecting edges changes in the same direction and remains continuous within a unified sliding window in the material flow direction, the candidate variable point is retained; otherwise, it is regarded as an isolated fluctuation and is eliminated. After variable point location and consistency verification, the set of candidate variable points that pass the verification and the corresponding time are recorded as the basis for subsequent division.

[0080] When dividing the operation process into steady-state segments and disturbance segments, the candidate variable point boundaries are first sorted in ascending time based on the variable point location results and deduplicated to form an initial segmentation point sequence. Then, the time axis is divided into continuous candidate segments with adjacent segmentation points as boundaries. For each candidate segment, within a unified sliding window, threshold discrimination is performed on dust concentration, conveyor belt speed, ambient wind speed and humidity, and water pressure and flow rate based on preset stable and hysteresis bands, and the duration of deviation from the stable band is recorded. When a threshold deviation that meets the minimum duration exists within a candidate segment, and the same direction occurs in the upstream or downstream related nodes, the disturbance segment is considered a candidate segment. When a candidate segment changes but still passes the consistency check, it is marked as a perturbation segment. When all four parameters in a candidate segment are within the stable band, or when there is a short-term deviation but the duration does not reach the minimum duration, or when there is no unidirectional continuity with adjacent nodes, the candidate segment is marked as a steady-state segment. To avoid over-segmentation, adjacent segments of the same type are merged without violating the consistency check. Isolated segments with a duration less than the minimum interval are absorbed according to the category of neighboring segments. Finally, a set of segments with start and end times and category labels is output and written into the global state graph.

[0081] The set of dust suppression units affected by disturbances is determined starting from the start and end times of the disturbance segment and the node where it occurs: candidate paths are generated along the conveying topology and material flow direction. The candidate paths cover all reachable nodes and connecting edges downstream from the node where the disturbance occurs, and candidate paths corresponding to the branch are generated simultaneously when there is a return material branch. Time-delay propagation verification is performed on the candidate paths. The time-delay propagation verification requires that the order and time delay of parameter changes of adjacent nodes or connecting edges along the path fall within a preset propagation time window. The preset propagation time window is determined based on the material arrival time range derived from the conveying section length and conveyor belt speed, and on this basis, the upper and lower time limits are formed by considering the airflow propagation delay caused by ambient wind speed and negative pressure suction. Passing the verification indicates that the disturbance may propagate along the path.

[0082] For candidate paths that pass the verification, propagation gating and blocking judgments are performed based on wind direction or negative pressure direction, the opening and closing status of the enclosure, and the opening and closing status of the valve: when the wind direction and negative pressure direction work together to guide the dust flow to a specific path, the path is retained; when the enclosure is closed and the valve is closed and can isolate the influence of disturbances, the path is blocked; for paths that are neither guided nor completely blocked, they are retained for further verification.

[0083] For the nodes on the remaining path, perform adjacency consistency and repeatability verification. Adjacency consistency requires that the parameter change direction of adjacent nodes along the path remains consistent within a unified sliding window. Repeatability verification requires that the consistency repeats within at least two consecutive windows to exclude occasional fluctuations. Finally, the dust suppression unit nodes that pass the time delay propagation verification, propagation gating and blocking judgment, and adjacency consistency and repeatability verification are marked as the set of dust suppression units affected by the disturbance, and this set, along with the start and end times of the disturbance segment, is written into the global state graph.

[0084] In a specific operational application, assuming that a transfer point node experiences a sustained deviation of its dust concentration representative value from the stable zone within a unified sliding window, reaching the minimum duration, and simultaneously, the conveyor belt speed representative value associated with this transfer point node increases, and the dust concentration representative values ​​of its downstream conveyor nodes and the ambient wind speed and humidity representative values ​​change in the same direction and pass consistency verification in subsequent windows, then the change point location and consistency verification are successful. This resulting disturbance segment serves as the starting point, generating candidate paths along the material flow direction, and comparing the order and time delay of changes at each node within a preset propagation time window. When the enclosure opens... When the closed state is displayed as open and the valve opening / closing state is displayed as open, the path is not blocked. At the same time, the wind direction or negative pressure direction indicates that the airflow is heading downstream. Therefore, the path satisfies the propagation gating and blocking judgment. After verifying the adjacent consistency and repeatability of adjacent nodes on the path, the relevant dust suppression unit nodes on the path are written into the dust suppression unit set affected by the disturbance. The set and the start and end times of the disturbance segment are written into the global state map. This completes the entire process from reading the global state map, changing point detection, consistency verification, dividing the steady state segment and the disturbance segment, to determining the dust suppression unit set affected by the disturbance.

[0085] The specific implementation details of the scheduling scheme generation module are as follows:

[0086] In the centralized controller, decision variables are first established: start / stop flags, start and end gaps, and discrete sets of action parameters are set for each dust suppression unit (spray valve assembly, negative pressure fan, and enclosed hood). The start and end gaps are based on a unified clock, with the time axis divided into continuous time slots according to the length of a unified sliding window. Both start and end gaps are represented by time slot numbers to ensure consistency with the time index of the global state graph. The discrete sets of action parameters are enumerated based on the stable operating conditions allowed by the equipment: the discrete set of action parameters for the spray valve assembly includes valve opening / closing status, spray flow rate level, and nozzle angle level; the discrete set of action parameters for the negative pressure fan includes fan start / stop status and speed level; and the discrete set of action parameters for the enclosed hood includes opening / closing status and position level.

[0087] The above enumerated items are all derived from the equipment technical manual and the field commissioning boundary. The centralized controller only selects instruction combinations from these discrete sets and does not generate parameter values ​​outside the sets.

[0088] Hard constraints are set and cannot be violated during the solution process. Safety interlocks are used to ensure the safety of personnel, machines, and equipment. For example, operations requiring the opening of maintenance doors are prohibited when the enclosure is open, and all start / stop flags are forced to stop when an emergency stop signal is valid. Mutual exclusion or coexistence rules are used to constrain process logic. For example, the valve opening / closing state of the spray valve group and the spray flow rate level must simultaneously point to the same execution state, and the opening / closing state of the enclosure and the start / stop state of the negative pressure fan must be coordinated sequentially within a specified order. Resource boundary conditions are used to limit the occupation of common resources. For example, the maximum simultaneous supply of water pressure and flow rate on the same pipeline, and the negative pressure air volume. The maximum simultaneous suction volume allowed in the same suction branch, the maximum number of parallel actions, and the upper limit of the number of simultaneous actions allowed in the same work area; equipment cooling and minimum interval are used to limit frequent start-ups and shutdowns. For example, after a negative pressure fan stops, it must wait for a preset cooling time before it can be restarted. After a spray valve group closes, it must wait for a minimum interval before it can be reopened. Time windows and sequence relationships are used to fix the key sequence. For example, within the coverage time window that a certain set affected by disturbance needs to be covered, the opening and closing state of the enclosure must first reach the set position, then the spray valve group is allowed to be opened, then the negative pressure fan is allowed to be started, and the process is repeated in reverse order at the end.

[0089] The guardian condition and the recovery condition are set as state constraints. When the guardian condition is triggered, the system is forced to fall back to the safe state. When the recovery condition is met, the system is allowed to return from the safe state to the planned state.

[0090] Given the set of disturbance-affected variables and operating condition labels, the centralized controller organizes the above decision variables and hard constraints into a constraint satisfaction model. The solution process adopts a discrete search method from enumeration to pruning to verification.

[0091] First, within each coverage window that needs to be covered, all candidate instruction combinations that satisfy the discrete set of action parameters are generated for the dust suppression units in the set affected by disturbances. Then, based on safety interlocks, mutual exclusion or coexistence rules, resource boundary conditions, equipment cooling and minimum interval, maximum number of parallel actions, and window and sequence relationship, each condition is pruned to eliminate candidate combinations that do not meet any condition. The remaining candidate combinations are then verified for the reachability of the protection and recovery conditions to confirm that they can fall back to the safe state when the protection condition is triggered and can return to the planned state when the recovery condition is met. The set of candidate combinations that passes the verification is the feasible instruction set.

[0092] Within the feasible instruction set, the central controller selects a scheme according to sorting rules: first, it selects the instruction set that can achieve complete coverage within the full coverage time window; if there are multiple instruction sets that meet the coverage requirements, it selects the instruction set with the fewest actions; if there are still multiple parallel candidates, it selects the instruction set with the lowest switching frequency; if there are still parallel candidates, it selects the instruction set with the shortest total execution time; when there are still parallel candidates, it uses a fixed order of unit identifiers for stable selection to ensure that the result is deterministic.

[0093] After the scheme is determined, the central controller generates a programmed scheduling scheme and sends it to the field controllers. The programmed scheduling scheme uses a list structure to list the unit identifier, timestamp, and action parameters of each instruction. The unit identifier corresponds to the unique identifier of the node in the global state graph, the timestamp corresponds to the absolute time of the start and end gap under a unified clock, and the action parameters correspond to the specific enumeration items in the discrete set of action parameters of the dust suppression unit (e.g., the spray flow level is a certain fixed level, the nozzle angle is a certain fixed level, the fan speed is a certain fixed level, and the enclosure position is a certain fixed level). When sending the scheme, the central controller also attaches the current coverage time window, the associated operating condition label, and the identifier of the set affected by disturbance, which are used by the field controllers to perform consistency checks before execution.

[0094] After receiving the programmed scheduling scheme, the field controller verifies the mapping relationship between each instruction and the local actuators and sensors. Once the verification is successful, the instruction is executed according to the timestamp. When the field controller detects a guardian condition being triggered during execution, it first performs a safety rollback locally and sends the event and timing back to the central controller. Based on this, the central controller immediately initiates a new feasible instruction set selection if the original coverage window has not yet been completed. It maintains the same sorting rules as before, generates an update instruction that matches the current remaining time slot, and issues it again, thereby ensuring that the goal of prioritizing the coverage window is not compromised.

[0095] In a specific scheduling scenario, assume the affected set includes spray valve assemblies at two transfer points and a negative pressure fan on an adjacent branch. The coverage window starts from a certain time slot and then continues from the start time slot to the end time slot. The centralized controller first enumerates discrete sets of action parameters for each of the three dust suppression units, forming candidate instruction combinations. It then checks the resource boundary conditions of water pressure and flow rate for these candidate instruction combinations. If the simultaneous supply on the same pipeline exceeds the allowable upper limit, the combination causing the exceedance is eliminated. For the negative pressure fan, it checks the maximum number of parallel actions. If the maximum number of simultaneous actions in the work area has been reached, no new activation is allowed within that time slot. Based on the time window and sequence, the spray valve assembly can only be opened after the enclosure reaches the specified position, and the negative pressure fan can only be activated after the spray valve assembly is opened. The feasible instruction set obtained after pruning is selected in the order of priority for covering the affected time window, minimum number of actions, minimum switching frequency, and shortest total execution time. The resulting programmed scheduling scheme lists the unit identifier, corresponding timestamp, and action parameters of each of the three dust suppression units. If a sudden drop in water pressure occurs during the initial stage of execution and triggers the protection condition, the field controller will first shut down the spray valve group and send back the event. The central controller will then select a feasible instruction set within the remaining time slots, prioritizing the continuity of the coverage window, and issue a new programmed scheduling scheme. This will achieve the comprehensive goal of prioritizing the coverage window, minimizing the number of actions, minimizing the switching frequency, and minimizing the total execution time by using a defined sorting rule.

[0096] The specific implementation details of the scheduling and control module are as follows:

[0097] On the field controller side, the programmed scheduling scheme is first loaded, and the unit identifiers, timestamps, and action parameters in the scheduling scheme are mapped to local actuators and sensors one by one: the valve opening and closing status, spray flow level, and nozzle angle position of the spray valve group are mapped to the electric valve actuator, flow meter, and nozzle angle actuator; the start / stop status and speed level of the negative pressure fan are mapped to the frequency converter and fan current sensor; and the opening and closing status and position position of the enclosure are mapped to the electric actuator and position limit switch. After mapping, parameter verification is performed. The verification includes: whether the timestamps in the scheduling scheme are consistent with the unified clock, whether all action parameters are within the discrete set of action parameters, and whether the unit identifiers correspond one-to-one with the local hardware channels. Missing or inconsistent instructions are marked as unexecutable and immediately reported to the central controller, while executable instructions continue the subsequent process.

[0098] The field controller analyzes the programmed scheduling scheme according to time slots, generates an execution queue arranged strictly in chronological order, and locks relevant safety interlock signals and resource occupancy before each time slot enters execution. The safety interlocks include state signals that do not allow execution, such as emergency stop, maintenance door opening, over-temperature and over-speed. Resource occupancy includes the expected occupancy of water pressure and flow on the corresponding water supply pipeline, the expected occupancy of negative pressure air volume on the corresponding suction branch, and the maximum number of parallel actions allowed in the work area.

[0099] After locking is completed, a pre-check is performed: The safety interlock is checked to be in a non-triggered state; resource thresholds such as water pressure, flow rate, and negative pressure airflow are checked to ensure they meet the minimum requirements for this time slot. If resource thresholds are insufficient, pre-charging or pre-draining is performed according to preset procedures to establish usable operating conditions. For example, before spraying, the upstream water supply valve is opened to establish water pressure; before the negative pressure fan starts, the bypass valve is opened for short-term exhaust to gradually increase the negative pressure airflow; and before spraying, the hood position is adjusted to the predetermined position. If any pre-check is not met, the relevant actions for the current time slot are postponed, the reason is recorded, and the local system enters a waiting state or executes a degradation strategy.

[0100] After entering the execution time slot, the field controller issues instructions according to the queue: for the opening and closing of the spray valve and the spray flow level, the instructions are issued to the target channels of the electric valve and the flow meter to ensure that the valve is opened first and then the flow is increased, and when canceling, the flow is reduced first and then the valve is closed; for the start and stop of the negative pressure fan and the speed level, the instructions are issued to the frequency converter to ensure that the speed increase and decrease are within the allowable acceleration and deceleration time of the equipment; for the opening and closing of the enclosure and the position setting, the instructions are issued to the actuator to allow related linkage actions only after the enclosure is in place. During execution, the field controller periodically collects water pressure, flow rate, fan current, conveyor belt speed, and dust concentration within a uniform sliding window sampling period, and compares them item by item with the protection conditions: when the water pressure is lower than the minimum working water pressure of the sprayer for a consecutive preset sampling period, it is determined as water pressure drop; when the fan current exceeds the rated allowable current limit for a consecutive preset sampling period, it is determined as fan overcurrent; when the conveyor belt speed shows an unplanned significant decrease relative to the previous time window and continues to reach the minimum duration, it is determined as a sudden drop in conveyor belt speed; when the sprayer has been turned on but the dust concentration is higher than the baseline concentration of the steady-state segment for a consecutive preset sampling period and exceeds the hysteresis zone, it is determined as abnormal dust rebound.

[0101] It should be noted that the values ​​of the above-mentioned continuous preset sampling period number, minimum working water pressure, rated allowable current upper limit, baseline concentration, hysteresis band, and minimum duration were all determined based on the equipment manual and on-site commissioning records before the system was put into operation, and were stored as local parameters for consistency judgment.

[0102] When any protective condition is triggered, the field controller immediately executes local degradation control. The degradation order follows the principles of safety priority and resource protection: first, the spray is shut down to prevent water accumulation and slippage due to insufficient water pressure or a sudden drop in conveyor belt speed; second, the negative pressure fan speed is reduced or temporarily stopped to avoid continuous overflow; then, the system switches to a safety sub-strategy, such as keeping the enclosure in its current position to avoid frequent opening and closing, suspending unnecessary spraying and suction operations, and retaining only actions that meet the minimum safety and environmental protection requirements. When the protective condition continues to deteriorate, an interlock is triggered, or personnel safety is threatened, an emergency shutdown is executed and the shutdown state is maintained until the recovery conditions are confirmed to be met. Each step of the degradation or emergency shutdown is recorded with an execution timestamp and status code and reported to the central controller through the communication channel, so that the central controller can perform rolling rescheduling or issue new programmed scheduling schemes accordingly.

[0103] When the anomaly is resolved and the recovery conditions are met (e.g., water pressure recovers to or above the minimum operating water pressure for spraying within a preset number of consecutive sampling cycles, fan current returns to the rated allowable current range within a preset number of consecutive sampling cycles, conveyor belt speed returns to the speed range of the current work plan within a preset number of consecutive sampling cycles, and dust concentration falls back to the baseline concentration and hysteresis zone), the field controller resets the relevant actuators in the controlled sequence: first, the enclosure is adjusted to the target position; then, the negative pressure fan is accelerated to the target speed level within the allowed acceleration time; finally, the spray valve is opened and gradually increased to the target spray flow level. After the reset is completed, if the current time slot is still valid, it returns to the current valid time slot to continue execution; if the current time slot has expired, it enters a waiting state until the next suitable time slot or receives an updated programmed scheduling scheme from the central controller. Throughout the process, the field controller continuously maintains resource occupancy locks and safety interlock locks, releasing the locks promptly after the action is completed or canceled to ensure that subsequent time slots can accurately assess resource availability; at the same time, it archives the execution timestamp, status code, records of guard condition triggering and recovery condition fulfillment.

[0104] The specific implementation details of the feedback update module are as follows:

[0105] The centralized controller receives the execution timestamps, status codes, water pressure, flow rate, fan current, dust concentration, and conveyor belt speed feedback parameters reported by each dust suppression unit in chronological order under a unified clock. It then aligns these parameters with the target time slots and coverage windows recorded in the programmed scheduling scheme. Each target time slot corresponds to a feedback summary record. If multiple sampling records exist for the same parameter within the same time slot, the records are sorted chronologically, and a representative value for that time slot is taken. If a parameter is missing in a time slot, the representative value from the previous time slot is read to fill the gap. If the parameter is still missing in the previous time slot, a representative value of the same type of physical quantity directly related to that node is used as a temporary estimate (e.g., the water pressure or flow rate representative value of the inlet pipe directly connected to that node) without changing the parameter name and meaning. After alignment, the controller generates a feedback alignment table for each coverage window. This table stores the representative values ​​of each parameter, along with the corresponding execution timestamps and status codes, with the target time slot as the row and the parameter name as the column, for subsequent effect verification.

[0106] The construction of the effect verification vector follows four criteria. First, whether the coverage of the disturbance segment is complete: within the coverage window, check each time slot whether the dust suppression units in the set affected by the disturbance maintain the prescribed action parameters (opening and closing status and spray flow rate of spray valves, starting and stopping status and speed of negative pressure fans, opening and closing status and position of enclosures) according to the programmed scheduling scheme. If it is found that any dust suppression unit affected by the disturbance fails to reach or exits the required action parameters in any target time slot, and the gap continues to reach the minimum duration, it is recorded as insufficient coverage.

[0107] Second, whether the recovery time exceeds the limit: Starting from the end of the coverage window, continuously read the representative value of dust concentration and compare it with the baseline concentration and hysteresis band of the steady-state segment. When the dust concentration continuously reaches within the baseline concentration and hysteresis band, it is recorded as recovery completed. If it is not completed within the preset allowable recovery time, it is recorded as recovery too slow.

[0108] Third, whether the resource usage is abnormal: the representative values ​​of the feedback parameters corresponding to the water pressure, flow rate and negative pressure air volume in each target time slot are checked. When they exceed the limit or are significantly lower than the minimum working requirements compared with the resource boundary conditions, they are recorded as resource abnormalities.

[0109] Fourth, whether the protection condition is triggered: When the field controller reports a status code indicating that the protection condition has been triggered during execution (e.g., water pressure drop, fan overcurrent, sudden drop in conveyor belt speed, abnormal dust rebound), the target time slot is marked as protection triggered. The effect verification vector is composed of the above four judgment results in chronological order, which is used to directly drive whether to enter rolling rescheduling;

[0110] When the effect verification vector shows any deviation in coverage, recovery is too slow, resource anomalies, or guardian triggering, the controller immediately triggers rolling rescheduling and determines the prediction time window.

[0111] The prediction time window is determined according to the coverage priority principle: if the coverage time window has not yet ended, the prediction time window extends from the current target time slot to the end of the coverage time window; if the coverage time window has ended but the recovery is too slow, the prediction time window extends from the current target time slot to the end of the material lag effect period derived from the conveyor section length and conveyor belt speed, and adds the airflow propagation buffer time related to ambient wind speed and negative pressure suction. Within the prediction time window, the controller generates candidate instruction combinations for dust suppression units in the set affected by disturbances, based on predetermined decision variables (start / stop flags, start and end time slots, discrete set of action parameters), and prunes them one by one according to hard constraints and timing constraints: hard constraints include safety interlocks, mutual exclusion or coexistence rules, resource boundary conditions, equipment cooling and minimum interval, and maximum number of parallel actions; timing constraints include the sequential relationship within the coverage time window and the necessary start and stop sequence.

[0112] After pruning, the controller verifies the reachability of the guardian and recovery conditions for the candidate combinations. It confirms that the system can fall back to the safe state when the guardian condition is triggered and can return to the planned state in sequence when the recovery condition is met. The candidate combinations that pass the verification form a feasible instruction set.

[0113] Within the feasible instruction set, the controller selects a scheme according to a predetermined sorting rule. First, it selects the instruction set that can achieve the most sufficient coverage of the coverage window or the lag effect period within the prediction time window. The so-called most sufficient coverage means that, without violating hard constraints and timing constraints, the dust suppression units in the affected set should achieve the required action parameters as seamlessly as possible within the target time slot that needs to be maintained.

[0114] If multiple instruction sets exist that can achieve equal coverage adequacy, the instruction set with the fewest actions is selected. If the number of actions is still the same, the instruction set with the lowest switching frequency is selected. The lowest switching frequency means that the sum of the number of times each dust suppression unit switches from one action parameter to another within the prediction time window is the smallest. If there are still ties, the instruction set with the shortest total execution time is selected, and a fixed unit identifier order is used as the final ties to eliminate them, so as to ensure the determinism of the selection process.

[0115] Based on the above selection, the controller obtains an updated programmatic scheduling scheme. The scheme records the unit identifier, timestamp, and action parameters for each item, and explicitly marks the corresponding coverage window or lag effect period range. The updated scheme is sent to the field controller through the existing communication channel. At the same time, the effect verification vector, deviation type, prediction time window range, and updated scheme identifier are written into the global status map and log system for subsequent working condition identification and rolling update calls.

[0116] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0117] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0118] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0119] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0120] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A belt transport automatic dust suppression system oriented to multi-condition adjustment, characterized in that: The system comprises a conveying data acquisition module, a working condition identification module, a scheduling scheme generation module, a scheduling control module, and a feedback updating module, and the modules are connected through signals; The conveying data acquisition module is used for collecting dust concentration, conveying belt speed, environmental wind speed and humidity, and water pressure and flow data from the conveying link nodes, constructing a global state atlas, and completing synchronization under a unified clock; The working condition identification module is used for identifying the working conditions of the global state atlas, dividing stable segments and disturbance segments, and determining a set of dust suppression units affected by the disturbance; The scheduling scheme generation module is used for modeling and solving the start-stop sequence, execution timing, and action parameters of the dust suppression units through the centralized controller, and generating a programmed scheduling scheme; The scheduling control module is used for executing the spraying, suction, or sealing operations of each dust suppression unit according to the scheduling scheme, and executing local degradation control when an abnormal condition is triggered; The feedback updating module is used for verifying the effect according to the execution feedback through the controller, and rolling updating the scheduling scheme when a deviation is detected, to realize optimized control under multiple working conditions.

2. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 1, characterized in that: The global state atlas is established in the following manner: The conveying section nodes, transfer point nodes, and dust suppression unit nodes are defined according to the conveying route and transfer relationship, and a topological connection is established; the dust concentration, conveying belt speed, environmental wind speed and humidity, and water pressure and flow sensors arranged at each node are calibrated, data is collected, and a unified clock is assigned; Multi-source data is time-aligned using a sliding window, abnormal data is removed, missing data is interpolated, and dimension unification is completed; The aligned data is mapped to the corresponding nodes or connection edges according to the spatial position and equipment identification, and node state vectors and edge state vectors are generated; Working condition derived indicators are calculated based on the node state vectors and written into the node attributes, forming an attribute set with a time interval; The attribute set is incrementally stored and maintained in a graph data structure, and the global state atlas is output.

3. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 2, characterized in that: The working condition identification of the global state atlas comprises: Reading the time-ordered node state vectors and edge state vectors from the global state atlas; Performing change point detection and locating the start and end times of the dust concentration, conveying belt speed, environmental wind speed and humidity, and water pressure and flow sequences within a unified sliding window; Performing consistency checking on adjacent nodes and upstream and downstream connection edges to remove isolated fluctuations and retain continuous changes across nodes; Dividing the running process into stable segments and disturbance segments according to the change point positioning and consistency checking results; Determining a set of dust suppression units affected by the disturbance along the conveying topology direction; Generating a working condition label containing a time interval and an impact range and writing it into the global state atlas.

4. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 3, characterized in that: The running process is divided into stable segments and disturbance segments according to the change point positioning and consistency checking results, comprising: Based on the change point positioning results, the change point boundaries are sorted in ascending order of time and de-duplicated to form an initial split point sequence; the time axis is divided into continuous candidate segments with adjacent split points as boundaries; For each candidate segment, the dust concentration, conveying belt speed, environmental wind speed and humidity, and water pressure and flow are threshold discriminated within a unified sliding window according to the preset stable band and hysteresis band, and the deviation duration is recorded. When there is a threshold deviation meeting the minimum duration within a candidate segment and a same-direction change occurs at the upstream or downstream related node and passes the consistency check, the segment is marked as a disturbance segment, otherwise it is marked as a steady-state segment.

5. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 4, characterized in that: And determine the dust suppression unit set affected by the disturbance, including: With the start and end time of the disturbance segment and the occurrence node as the starting point, generate a candidate path along the conveying topology and material flow direction; Perform time delay propagation check on the nodes and connecting edges on the candidate path within a unified sliding window, requiring the sequence of parameter changes and the delay to fall within the preset propagation time window; Perform propagation gating and blocking determination based on wind direction or negative pressure direction, closed cover opening and closing state, and valve opening and closing state, and eliminate blocked paths and isolated nodes; Perform adjacent consistency and repeatability verification on the remaining nodes to exclude accidental fluctuations; Mark the dust suppression unit set that passes the propagation check, gating determination and consistency verification as the disturbance affected set and write it into the global state atlas.

6. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 5, characterized in that: For the centralized controller to constrain modeling and solving of the start-stop sequence, execution timing and action parameters of the dust suppression unit, generate a programmed scheduling scheme, including: Establish decision variables, define start-stop flags, start and end time slots, and action parameter discrete sets for each dust suppression unit; Set hard constraints, including safety interlocking, mutual exclusion or coexistence rules, resource boundary conditions, device cooling and minimum interval, maximum number of actions, and time window and sequence; Set guardian conditions and recovery conditions as state constraints, and map scheduling requirements and coverage time windows according to the disturbance affected set and working condition labels; Build a constraint satisfaction model and solve it to obtain a feasible instruction set; select a scheme from the feasible solution according to the sorting rules, in order to meet the coverage of the disturbance time window first, the minimum number of actions, the minimum switching frequency and the shortest total execution time; Generate a programmed scheduling scheme containing unit identification, timestamp and action parameters and issue it to the local controller.

7. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 6, characterized in that: For each dust suppression unit to perform spraying, suction or sealing operations according to the scheduling scheme, and perform local degradation control when an abnormal condition is triggered, including: Load the scheduling scheme and complete parameter mapping and verification with local actuators and sensors; Parse the scheduling scheme by time slot to generate an execution queue and lock the associated interlocking amount and resource occupation; Before entering the execution, pre-check the safety interlocking and resource thresholds such as water pressure, flow, negative pressure air volume, and complete the pre-charge or pre-discharge according to the preset process; Issue instructions for spraying valve opening and closing, flow setting, negative pressure fan start and stop, speed setting, and closed cover opening and closing through the local controller; Periodically collect water pressure, flow, fan current, conveyor belt speed and dust concentration during execution and compare them with the guardian conditions; Trigger local degradation control when detecting abnormal conditions such as water pressure drop, fan overcurrent, conveyor belt speed drop or dust abnormal rebound, and preferentially close the spray, reduce the suction or switch to a safe sub-strategy and perform emergency shutdown if necessary; When the abnormality is resolved and the recovery condition is met, reset the related actuators in the controlled order and return to the current valid time slot or enter the waiting state.

8. The belt transport automatic dust suppression system oriented to multi-condition adjustment according to claim 7, characterized in that: The application relates to a method for realizing optimal control under multiple working conditions by controlling a controller to perform effect verification according to execution feedback and to roll over an updating scheduling scheme when deviation is detected, and the method comprises the following steps: Collecting execution time stamps, state codes, water pressures, flow rates, fan currents, dust concentration and conveying belt speed feedback parameters reported by each dust suppression unit and aligning the parameters with target time slots and coverage time windows of a programmed scheduling scheme; Constructing an effect verification vector to respectively determine whether a disturbed segment coverage is complete, whether a recovery time is over-limit, whether resource occupation is abnormal and whether a guardian condition is triggered; Triggering rolling rescheduling when deviation is detected, searching for a feasible instruction set meeting hard constraints and time sequence constraints within a prediction time window; Selecting a scheme with the most sufficient coverage of a disturbed window, the least action times and the least switching frequencies in a feasible solution set according to a sorting rule to generate an updated programmed scheduling scheme.

Images