A method and system for intelligent scheduling of trailing suction hopper dredging projects

By dynamically updating the departure time and operation sequence based on the workload and tidal conditions, the construction section and navigation energy consumption of the trailing suction hopper dredging project are optimized, solving the problem of low construction organization efficiency in the existing technology and achieving more efficient construction management.

CN123117736APending Publication Date: 2026-11-17CCCC GUANGHANG DREDGING CO
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611618446.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The current dredging schedule for trailing suction hopper dredgers relies on fixed construction period constraints and human experience, resulting in a rough matching of construction sections with vessel capacity. The impact of tidal changes is difficult to take into account, leading to vessel waiting times, increased energy consumption, and low construction organization efficiency.

Method used

By combining the spatial distribution of the project workload, the vessel's operational capacity, the construction water depth and tidal conditions, the departure time and operation sequence are dynamically updated to form an intelligent scheduling mechanism, optimize the allocation of construction sections and navigation energy consumption of multiple trailing suction hopper dredgers, and realize dynamic adjustment of the construction process.

Benefits of technology

Reduce ship waiting time and wasted navigation, improve section connections and energy utilization, and enhance the stability of construction organization and schedule control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117736A_ABST
    Figure CN123117736A_ABST
Patent Text Reader

Abstract

This invention relates to the field of process control technology, specifically to an intelligent scheduling method and system for trailing suction hopper dredging projects. The method includes the following steps: dividing construction sections based on the distribution of work volume and the vessel's capacity and draft, establishing matching relationships; reading candidate departure times; predicting and prioritizing energy consumption for full-load and empty-load navigation based on the direction of the tidal current and speed and draft; generating multi-vessel schedules based on node schedules, navigation windows, and operational cycles; and updating the priority and schedule of remaining work volume based on water depth, shallow siltation, and voyage status. In this invention, by incorporating the energy consumption of vessel capacity sections, task voyages, and schedule windows into a unified sorting constraint, the departure times and operational sequences take into account both energy utilization and construction continuity. Furthermore, the method adjusts section allocation according to changes in work volume, reducing vessel waiting and ineffective navigation, alleviating the problem of delayed response to changes in tidal currents and siltation in fixed schedules, and improving the efficiency of multi-vessel collaboration.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of process control technology, and in particular to an intelligent scheduling method and system for trailing suction hopper dredging projects. Background Technology

[0002] The field of process control technology involves using computer information systems to record, analyze, and coordinate the task sequence, status changes, and execution conditions in continuous or phased work processes. Specifically, intelligent scheduling for trailing suction hopper dredging projects involves dividing the dredging project into several tasks, recording the dependencies, planned times, vessel occupancy status, and construction progress of each task in a database, and having the server sequentially read the task status data according to the task flow, updating the status of unexecuted, in-process, and completed tasks, and adjusting the execution order and planned time of subsequent tasks based on task dependencies and resource occupancy records.

[0003] Existing dredging projects using trailing suction hopper dredgers rely heavily on fixed schedules and manual experience to allocate vessels. The matching of construction sections with vessel capacity is rather crude. Departure times are usually determined according to the planned sequence, making it difficult to incorporate the impact of tidal changes on the energy consumption of round-trip voyages into a unified comparison. Conflicts easily arise between navigation windows and work cycles. When siltation occurs in shallow water areas and the status of voyages changes continuously during construction, adjustments to the original plan lag behind, easily causing vessel waiting, uneven connection between sections, increased energy consumption, and decreased stability in meeting milestones, thus restricting the overall efficiency of construction organization. Summary of the Invention

[0004] This invention aims to provide an intelligent scheduling method and system for trailing suction hopper dredging projects. By associating the spatial distribution of the workload, the vessel's operational capacity, the construction water depth, the key construction periods, and the navigation window, and incorporating the vessel's navigation energy consumption corresponding to different candidate departure times under tidal conditions into the scheduling basis, the remaining workload and construction priority are dynamically updated based on the water depth, shallow points, siltation, and voyage completion status during construction. This allows the allocation of construction sections, the order of operations, and the departure times of multiple trailing suction hopper dredging vessels to be uniformly determined and adjusted according to project constraints, vessel suitability conditions, navigation energy consumption, and on-site construction status, thereby forming an intelligent scheduling mechanism suitable for trailing suction hopper dredging projects.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent scheduling method for trailing suction hopper dredging projects, comprising the following steps: The process involves acquiring data on the project workload distribution, key milestones, navigation windows, and the capacity, draft, speed, and operational cycles of each trailing suction hopper dredger. Based on these data, construction sections are divided, and vessel section matching relationships are determined. Candidate departure times are read, and based on the vessel section matching relationships, the round-trip route from the trailing suction hopper dredger to the dumping area is used. Downstream and crossstream currents are determined according to tidal current forecasts and vessel heading. Speed, draft, downstream and crossstream currents are input into a navigation energy consumption prediction model to calculate propulsion power for fully loaded departure and empty return, and the total energy consumption for each voyage is accumulated. Candidate departure times are sorted in ascending order according to total voyage energy consumption to construct a candidate departure sequence. Combining key milestones, navigation windows, and vessel operational cycles, departure times and operational sequences are determined, generating a multi-vehicle scheduling plan. Real-time acquisition of water depth measurement results, shallow point distribution, siltation changes, and voyage completion status; determination of remaining work volume and construction priority in construction sections; and updating of section allocation, work sequence, and departure time in multi-vessel scheduling plan based on remaining work volume, construction priority, and candidate departure sequences.

[0006] For the division of construction sections and the matching relationship of vessel sections, the planar position, current bottom elevation and design bottom elevation corresponding to the distribution of engineering volume are further recorded. Calculation units are divided according to the continuous construction boundary. Based on the planar adjacency relationship and construction depth difference of adjacent calculation units, the engineering volume is merged and accumulated to generate candidate sections. The target engineering volume and minimum construction water depth of the candidate sections are read. The hopper capacity is determined as the single-voyage loading capacity. The draft is compared with the minimum construction water depth, and trailing suction hopper dredgers that do not meet the safe margin water depth condition are eliminated to obtain a set of candidate vessels. The single-voyage loading capacity of the trailing suction hopper dredgers in the candidate vessel set is compared with the target engineering volume, and adjacent candidate sections with the same candidate vessel set are merged to generate construction sections and establish vessel section matching relationships.

[0007] The safe margin water depth condition can be determined by the construction water depth corresponding to each calculation unit within the candidate section. The minimum construction water depth is determined as the minimum construction water depth. The full-load draft of the trailing suction hopper dredger corresponding to the candidate section is obtained, and the difference between the minimum construction water depth and the full-load draft is calculated as the water depth margin. The water depth margin is compared with the preset safe margin water depth threshold. If the water depth margin is not less than the safe margin water depth threshold, the trailing suction hopper dredger is determined to meet the safe margin water depth condition; otherwise, the corresponding trailing suction hopper dredger is excluded from the candidate vessel set.

[0008] To determine the total energy consumption of the voyage, the trailing suction hopper dredger, construction section, and dumping area corresponding to the candidate departure time are determined based on the matching relationship between vessel sections. Tidal current forecasts are matched with the route positions and expected arrival times from the construction section to the dumping area and from the dumping area to the construction section to obtain tidal current data for the round trip routes. Based on the tidal current velocity, tidal current direction, and the vessel's heading at each route position, the flow direction is decomposed along the vessel's heading to obtain downstream flow, and the flow component perpendicular to the vessel's heading is determined as transverse flow, forming the route flow field input data. The speed, draft, and route flow field input data corresponding to fully loaded departure and empty return are input into the navigation energy consumption prediction model to obtain the propulsion power of each segment of the round trip route. The energy consumption is accumulated according to the segment's navigation time, and the energy consumption of fully loaded departure and empty return is combined to obtain the total energy consumption of the voyage.

[0009] The navigation energy consumption prediction model can be constructed based on historical voyage data. It obtains synchronous speed, draft, ship heading, tidal current speed, tidal current direction, and measured propulsion power from historical voyages. The tidal current is decomposed according to the ship heading to obtain historical downstream current and historical cross current. Samples are labeled according to full load or empty load conditions to form a training sample set. Using speed, draft, historical downstream current, and historical cross current as inputs, and measured propulsion power as the prediction target, the model is trained in groups according to ship type and operating condition to obtain sub-models for different operating conditions. The model is validated using historical voyages that did not participate in the training. When the propulsion power prediction deviation meets the preset error threshold, the model parameters are fixed, and an index relationship between ship type, operating condition, and model parameters is established.

[0010] When obtaining the energy consumption for full-load departure and empty-load return, the vessel identifier of the corresponding trailing suction hopper dredger is read according to the vessel segment matching relationship. The navigation energy consumption model parameters that match the vessel type and main propulsion device are called according to the vessel identifier, and the operating condition identifier is determined according to the full-load departure or empty-load return status. The speed, draft, downstream current, lateral current and navigation operating condition parameters are formed into a continuous input sequence according to the expected elapsed time of each segment, and input into the navigation energy consumption prediction model to obtain the segment propulsion power sequence of each segment. Then, the segment navigation time is determined according to the distance and corresponding speed of each segment, and the segment propulsion power sequence is associated with the segment navigation time segment by segment and accumulated.

[0011] For multi-vessel scheduling plans, a record of the corresponding departure time and energy consumption is established for the same trailing suction hopper vessel and the same construction section. Candidate departure sequences are obtained by arranging them from low to high based on the total energy consumption of the voyage. Candidate departure times are read item by item according to the candidate departure sequences. The cycle end time and the expected passage time of the restricted section are calculated based on the vessel's operation cycle. The expected passage time of the restricted section is compared with the navigation window, and the cycle end time is compared with the node schedule to form an executable candidate sequence. Then, the departure time of each trailing suction hopper vessel is determined in sequence, and the operation order is determined according to the time sequence of the corresponding operation cycle. The correspondence between the vessel, construction section, departure time and operation order is established.

[0012] During the dynamic updating of the multi-vessel scheduling plan, the water depth measurement results, shallow point distribution, siltation changes, and voyage completion status of the same measurement cycle are associated with the spatial boundaries of the construction sections. The measured bottom elevation is compared with the design bottom elevation to identify substandard construction units. The dredging volume of these substandard units is adjusted based on siltation changes, and the remaining workload for each construction section is accumulated. Construction priorities are determined according to a hierarchical relationship: urgency takes precedence, followed by shallow point treatment, and then siltation treatment. When the first three priorities are the same, the order is determined by the remaining workload from largest to smallest. Construction sections to be constructed are read from highest to lowest priority. Trailing hopper dredgers that can be reassigned are determined based on the voyage completion status. From the candidate vessels that meet the vessel-section matching relationship, a departure candidate is selected according to the candidate departure sequence. Based on this, the construction sections, vessels, work sequence, and departure times are updated.

[0013] Water depth measurement results, shallow point distribution, siltation changes, and voyage completion status can be obtained through the depth sounding equipment of the construction survey vessel or trailing hopper dredger and the onboard operation records. Water depth data with location markers and collection times are collected, aggregated according to the planar grid of the construction section, and compared grid-by-grid with the design bottom elevation. Grid positions where the measured bottom elevation is higher than the design bottom elevation and their corresponding elevation differences are spatially correlated to generate water depth measurement results and shallow point distribution. The measured bottom elevation of two adjacent measurement cycles within the same grid is read, and siltation changes are determined by comparing bottom elevation changes based on the sequence of measurement times. Records of loading, departure, dredging, and return corresponding to vessel and voyage markers are retrieved from the onboard operation records. The corresponding voyage is considered complete when all preset operation stages are recorded.

[0014] The present invention also provides an intelligent scheduling system for trailing suction hopper dredging projects to implement the above method, including a construction zone matching module, an energy consumption prediction module, a multi-vessel scheduling module, and a scheduling module. The construction zoning matching module is used to acquire the distribution of work volume, node schedules, navigation windows, and the capacity, draft, speed, and operational cycles of each trailing suction hopper dredger, dividing the construction zones and determining the matching relationships between vessel sections. The energy consumption prediction module is connected to the construction zoning matching module and is used to calculate the propulsion power and total energy consumption of the voyage for fully loaded departure and empty return based on candidate departure times, vessel section matching relationships, tidal current forecasts, and vessel headings. The multi-vessel scheduling module is connected to the energy consumption prediction module and is used to construct candidate departure sequences, and determine departure times and operational sequences based on node schedules, navigation windows, and vessel operational cycles, generating a multi-vessel scheduling plan. The scheduling module is connected to the multi-vessel scheduling module and is used to determine the remaining work volume and construction priorities based on real-time acquired water depth measurement results, shallow point distribution, siltation changes, and voyage completion status, and to update the section allocation, operational sequence, and departure times of the multi-vessel scheduling plan.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a continuous scheduling logic is formed by adapting to different sections, prioritizing voyage energy consumption, and dynamically rearranging the schedule. Construction sections are divided based on the distribution of work volume and cabin capacity and draft, and a matching relationship between the vessels and the sections is established. This coordinates the vessel's capacity with the tasks of each section. The tidal current direction and speed and draft of the round-trip route are used to calculate the energy consumption for fully loaded departure and empty return. The departure priority is formed according to the total energy consumption, and the work sequence is constrained by the node schedule, navigation window, and work cycle. At the same time, the remaining work volume and priority are updated based on the water depth, shallow siltation, and voyage status, so that the section allocation and departure time are adjusted synchronously with changes on site. This reduces vessel waiting and ineffective voyages, improves section connection and energy utilization, and enhances the stability of schedule control and construction organization. Attached Figure Description

[0016] Figure 1 A schematic diagram of the overall data flow for the scheduling and dispatching method of trailing suction hopper dredging projects; Figure 2 This is a schematic diagram illustrating the formation of the construction section and the matching relationship between the vessel sections; Figure 3 A schematic diagram showing the input-output relationship and parameter index of a navigation energy consumption prediction model; Figure 4 A schematic diagram illustrating the relationship between candidate departure sequences and multi-ship scheduling plans; Figure 5 This is a schematic diagram illustrating the multi-vessel scheduling update relationship based on construction status feedback; Figure 6 This is a schematic diagram of the module interaction of the scheduling system for trailing suction hopper dredging projects. Detailed Implementation

[0017] The following describes the embodiments of the present invention. In the present invention, the engineering quantity distribution refers to the existing bottom elevation, the design bottom elevation, and the spatial distribution of the dredging quantity corresponding to the planar position within the construction area. The node construction period is used to define the time boundaries that the corresponding construction task or vessel operation cycle should meet. The navigation window is used to define the allowable time interval for the trailing suction hopper dredger to pass through restricted sections. The vessel operation cycle includes the loading, departure, dumping, and return processes corresponding to one dredging voyage, and the scheduling sequence constraints are formed by the time states corresponding to each process.

[0018] A construction section is a spatial construction object formed based on continuous construction boundaries, differences in construction depth, workload, and the adaptability of the trailing suction hopper dredger to the construction water depth. The vessel section matching relationship indicates whether a particular trailing suction hopper dredger is capable of performing dredging tasks in the corresponding construction section. Its formation process considers both the single-voyage loading capacity of the trailing suction hopper dredger and the safety constraints between its full-load draft and the construction water depth. Candidate departure times are the departure times participating in the scheduling comparison. Each candidate departure time is associated with the corresponding trailing suction hopper dredger, construction section, and dumping area during the energy consumption calculation process.

[0019] Downstream current is the component of tidal current velocity along the ship's course, while transverse current is the component of tidal current velocity perpendicular to the ship's course. The navigation energy consumption prediction model is used to establish predictive relationships between ship speed, draft, downstream current, transverse current, and propulsion power, and calls appropriate model parameters based on ship type and whether the ship is fully loaded or unloaded. Total voyage energy consumption includes energy consumption during the fully loaded departure process and energy consumption during the unloaded return process, which is used as the basis for ranking candidate departure times based on energy consumption.

[0020] A multi-vessel scheduling plan must at least include the correspondence between vessels, construction sections, departure times, and work sequences. Water depth measurement results reflect the actual water depth status of the construction section; shallow point distribution indicates the locations where the measured bottom elevation is higher than the design bottom elevation and the corresponding elevation difference; siltation changes indicate the bottom elevation changes at the same spatial location between adjacent measurement periods; and voyage completion status indicates whether the loading, departure, dumping, and return phases corresponding to vessel and voyage identifiers have been completed. Remaining work volume and construction priority are updated based on the above construction status data and continue to influence the subsequent allocation of sections, work sequences, and determination of departure times.

[0021] Example 1 Please see Figure 1 This embodiment provides a scheduling method for trailing suction hopper dredging projects. In this embodiment, an executable relationship between the vessel and the construction section is established based on the spatial distribution of the project volume and the vessel's construction capacity. Then, tidal current forecasts are matched with candidate departure times, and the total energy consumption of each voyage (fully loaded departure and empty return) is calculated. A candidate departure sequence is formed based on the total energy consumption of each voyage. A multi-vessel scheduling plan is generated under the constraints of node schedules, navigation windows, and vessel operation cycles. The scheduling results are updated during construction using water depth, shallow points, siltation, and voyage status. The main process of this method includes four main steps, S1 to S4.

[0022] S1 obtains the project quantity distribution, node schedule, navigation window, and the hopper capacity, draft, speed, and vessel operation cycle of each trailing suction hopper vessel. Based on the project quantity distribution, hopper capacity, and draft, the construction sections are divided and the vessel section matching relationship is determined.

[0023] Please see Figure 2Each spatial data object in the quantity distribution corresponds to at least a planar location, existing bottom elevation, and design bottom elevation. The planar location determines the spatial affiliation of the data object within the construction area; the existing bottom elevation characterizes the current bottom surface location; and the design bottom elevation characterizes the target bottom surface location. Continuous construction boundaries define the spatial range within which dredging operations can be continuously organized. The construction depth difference determines whether adjacent calculation units are suitable for merging into the same candidate section.

[0024] S101. The distribution of the workload is spatially divided according to the continuous construction boundary to form calculation units. The planar adjacency relationship and construction depth difference of adjacent calculation units are read. Units belonging to the same continuous construction boundary and satisfying the adjacency merging relationship are merged, and the corresponding workload is accumulated during the merging process to form candidate sections. Each candidate section corresponds to at least a planar spatial boundary, a target workload, and a minimum construction water depth. The target workload is used for subsequent comparison with the single-voyage loading capacity of the trailing suction hopper dredger, and the minimum construction water depth is used to determine whether a fully loaded trailing suction hopper dredger has the water depth conditions for entry and construction.

[0025] The minimum construction water depth is determined by the construction water depth corresponding to each calculation unit within the candidate section. The construction water depths belonging to the same candidate section are read one by one, and the minimum construction water depth is taken as the minimum construction water depth for that candidate section. This selection of the minimum value ensures that the vessel's water depth adaptability judgment uses the location with the most stringent constraints within the candidate section as the comparison object, avoiding the use of only the section average water depth while ignoring locally shallow water locations.

[0026] S102, read the target engineering quantity and minimum construction water depth of the candidate section, and read the hopper capacity and full-load draft of each trailing suction hopper vessel. The hopper capacity is the single-voyage loading capacity of the corresponding trailing suction hopper vessel. For each candidate section, calculate the water depth difference between its minimum construction water depth and the full-load draft of the trailing suction hopper vessel to be judged, and use this water depth difference as the water depth margin.

[0027] The safe margin water depth threshold is a comparison parameter using the same length dimension as the minimum construction water depth and the draft under full load conditions. It is used to reserve a safe margin of water depth beyond the draft of a fully loaded vessel. The water depth margin is compared with the safe margin water depth threshold. When the water depth margin is not less than the safe margin water depth threshold, the corresponding trailing hopper dredger is retained in the candidate vessel set for that candidate section; when the water depth margin is less than the safe margin water depth threshold, the corresponding trailing hopper dredger is excluded from the candidate vessel set for that candidate section.

[0028] The comparison process uses the lowest construction water depth as the section-side input, the full-load draft as the vessel-side input, the water depth margin as the intermediate result, and the retention or exclusion status of the candidate vessel set as the output. The resulting candidate vessel set is then passed to the candidate section merging process and the scheduling process. Vessels not included in the candidate vessel set are not considered as scheduling candidates for the corresponding construction section.

[0029] S103, for candidate sections, compare the single-voyage loading capacity of each trailing suction hopper dredger within the candidate vessel set with the target engineering volume of the candidate section, and simultaneously compare the candidate vessel sets corresponding to spatially adjacent candidate sections. Adjacent candidate sections with the same candidate vessel set are merged to form construction sections. The merged construction sections retain the correspondence with their candidate vessel sets, thus forming a vessel section matching relationship.

[0030] The output of the vessel segment matching relationship should at least indicate the corresponding status between the construction segment and the trailing suction hopper dredger allowed to perform the construction task in that segment. Subsequent energy consumption calculations, when reading candidate departure times, will only establish voyage calculation objects for combinations of vessels and construction segments that satisfy this matching relationship, thus preventing vessels that do not meet the construction water depth constraints from entering the candidate departure sequence for the corresponding segment.

[0031] S2, read the candidate departure time, based on the ship section matching relationship, along the round-trip route from the trailing suction hopper to the dumping area, determine the downstream and lateral currents according to the tidal current forecast and the ship's heading, input the speed, draft, downstream and lateral currents into the navigation energy consumption prediction model, calculate the propulsion power for fully loaded departure and empty return, and accumulate the total energy consumption of the voyage.

[0032] Candidate departure times exist as time values ​​that can be corresponding to the scheduling timeline. For each candidate departure time, the corresponding trailing suction hopper dredger and construction section are read according to the vessel section matching relationship, and the dumping area corresponding to the construction task is determined, thus forming a candidate voyage object. The candidate voyage object has a fully loaded departure route from the construction section to the dumping area and an empty return route from the dumping area to the construction section.

[0033] S201 processes round-trip routes based on their location and estimated transit times. For fully loaded outbound routes, each segment is determined from the construction section based on its location, and the estimated transit time for each segment is calculated based on candidate departure times and the estimated travel time for each segment. For empty return routes, the estimated transit time for each segment is determined by the return sequence after dredging is completed. By correlating the route location and estimated transit time with tidal current forecasts, the tidal current velocity and direction for each segment are obtained.

[0034] The spatial location of the tidal current forecast corresponds to the location of the flight segment, and the temporal information corresponds to the estimated transit time of the flight segment. For the same candidate departure time, fully loaded departures and empty returns pass through the route positions at different times, and the tidal current data corresponding to their estimated transit times are read separately. Therefore, the round-trip routes in a single voyage are not considered to have the same tidal current conditions.

[0035] S202: For each segment, the ship's heading of the trailing hopper dredger is read, and the corresponding tidal current velocity and direction are decomposed according to the ship's heading. The flow component along the ship's heading is denoted as downstream flow, and the flow component perpendicular to the ship's heading is denoted as transverse flow, thus forming the route flow field input data corresponding to the segment position and the expected elapsed time.

[0036] Downstream current describes the longitudinal effect of tidal current along the direction of the ship's movement or in the opposite direction, while transverse current describes the lateral effect of tidal current relative to the ship's course. Both, along with ship speed and draft, serve as inputs for propulsion power prediction. The route flow field input data are arranged according to the estimated elapsed time of each segment, ensuring that the propulsion power prediction for each segment uses the tidal current state at the corresponding time for that segment.

[0037] S203, please refer to Figure 3 The navigation energy consumption prediction model is constructed using synchronously recorded data from historical voyages, including speed, draft, ship heading, tidal current speed, tidal current direction, and measured propulsion power. First, the tidal current is decomposed according to the ship's heading corresponding to the historical voyage, resulting in historical downstream and cross currents. Samples are then labeled as either fully loaded or unloaded based on the loading status of the historical voyage. Each training sample uses speed, draft, historical downstream and cross currents as input fields, and the measured propulsion power corresponding to the same data collection state as the prediction target field.

[0038] Training samples are grouped according to ship type and operating condition, ensuring that samples within the same group correspond to the same ship type and the same full-load or empty-load condition. A predictive relationship between input fields and measured propulsion power is established within each group. The model parameters corresponding to that group are adjusted during training to form sub-models for different operating conditions. Specifically, each sub-model employs a multi-layer feedforward neural network regression model. The input layer has four input nodes corresponding to speed, draft, downstream current, and cross current, while the output layer has one output node to output propulsion power. At least one fully connected hidden layer is placed between the input and output layers, and a nonlinear activation function is used to fit the nonlinear coupling relationship between the input fields. Before training, each input field within the same group is normalized or standardized. The mean square error between the predicted propulsion power output by the network and the measured propulsion power of the corresponding sample is used as the loss function. The gradient of the loss function relative to each network weight and bias is calculated through backpropagation, and the network weights and biases are iteratively updated using a gradient descent optimization algorithm until the loss function converges or the preset training termination condition is met. The connection weights and biases obtained after training constitute the model parameters of the ship type and the corresponding sub-models for prediction under full load or no load conditions, thereby determining the nonlinear mapping relationship between speed, draft, downstream current and lateral current to propulsion power.

[0039] The processing boundaries of the sub-model for different operating conditions are fixed as reading speed, draft, downstream current, and lateral current, and outputting the propulsion power required to maintain the corresponding navigation state.

[0040] After the model parameters are trained, historical voyages not involved in the corresponding training process are used as verification objects. The speed, draft, historical downstream current, and historical lateral current from the verification voyages are input into the corresponding sub-model for each working condition to obtain the predicted propulsion power. The predicted propulsion power is then compared with the synchronously measured propulsion power to form the propulsion power prediction deviation. A preset error threshold and the adopted prediction deviation are used with the same evaluation metric. When the propulsion power prediction deviation is less than the preset error threshold, the model parameters of the sub-model for that working condition are fixed.

[0041] After the model parameters are solidified, an index relationship is established based on the ship type and operating condition. When calling up actual voyages, the corresponding ship type and main propulsion system are first determined by the ship identifier, and then the operating condition identifier is determined based on whether the ship is departing fully loaded or returning empty. Based on this, the navigation operating condition parameters that match the current ship and navigation operating conditions are obtained.

[0042] S204. For the fully loaded departure process, the ship's current speed and draft at full load are used as the ship's state input, and the downstream and lateral currents of each fully loaded departure segment are used as the environmental input. These are combined with the aforementioned navigation parameters to form a continuous input sequence arranged according to the expected elapsed time. The navigation energy consumption prediction model reads this continuous input sequence segment by segment and outputs the propulsion power required to maintain the corresponding speed in each segment under full load departure conditions, forming a propulsion power sequence for each segment under full load departure conditions.

[0043] For the empty return journey, the ship's speed and draft, corresponding to the current empty state of the trailing hopper dredger, are used as the ship's state inputs, and the downstream and lateral currents of each return segment are used as the environmental inputs. The navigation condition parameters corresponding to the empty condition are then called to form a continuous input sequence for the return journey. The navigation energy consumption prediction model outputs the corresponding propulsion power sequence for each return segment in sequence.

[0044] S205: For each flight segment, the distance and corresponding speed are read, and the flight time for that segment is determined accordingly. The propulsion power of a segment is correlated with the flight time of the same segment segment, and the energy consumption of the propulsion power within the corresponding flight time is used to form the energy consumption of that segment. The energy consumption of each fully loaded outward flight segment is accumulated sequentially according to the route, resulting in the fully loaded outward energy consumption; the energy consumption of each empty return flight segment is accumulated sequentially according to the return flight, resulting in the empty return energy consumption. When propulsion power is expressed in kilowatts and flight time in hours, the corresponding energy consumption is represented in kilowatt-hours.

[0045] The energy consumption for a fully loaded departure and an empty return is combined to obtain the total energy consumption for the current trailing suction hopper dredger, the current construction section, and the current candidate departure time. For different candidate departure times, the above-mentioned processes of tidal current matching, flow field decomposition, propulsion power prediction, and energy consumption accumulation are repeated to form a one-to-one correspondence record between candidate departure times and total departure energy consumption.

[0046] S3 sorts the candidate departure times in ascending order according to the total energy consumption of the voyage, constructs a candidate departure sequence, and determines the departure time and operation sequence by combining the node schedule, navigation window and ship operation cycle, and generates a multi-ship scheduling plan.

[0047] Please see Figure 4 For the same trailing hopper dredger and the same construction section, different candidate departure times and their corresponding total energy consumption for each voyage are recorded as departure time-energy consumption correspondence records. These correspondence records are arranged from lowest to highest total energy consumption for each voyage, and the sorted candidate departure times constitute the candidate departure sequence corresponding to the combination of the trailing hopper dredger and the construction section. The candidate departure sequence retains the candidate departure time, the corresponding total energy consumption for each voyage, and the association between the vessel and the construction section.

[0048] S301: Read the candidate departure times according to the order of the candidate departure sequence. For each candidate departure time, calculate the end time of the corresponding voyage cycle based on the vessel's operational cycle, and calculate the estimated passage time of the vessel through the restricted section based on the round-trip route and operational cycle sequence. Compare the cycle end time with the node schedule, and compare the estimated passage time of the restricted section with the navigation window.

[0049] The air traffic window is constrained by the time interval that allows passage through restricted air routes. Only if the expected passage time of the restricted air route falls within the allowed time range of the corresponding air traffic window, and the cycle end time meets the corresponding node schedule, will the current candidate departure time be retained as an executable candidate. Candidates that do not simultaneously meet the above time constraints will not enter the executable candidate sequence.

[0050] S302, candidates that meet the node schedule and air traffic window constraints are grouped into an executable candidate sequence according to their original candidate departure sequence order. Thus, the total energy consumption of the voyage determines the priority order of reading the candidates, while the node schedule and air traffic window determine whether a candidate can enter the actual schedule. The two types of processing respectively play the roles of sorting and time constraint.

[0051] S303: Determine the departure time of each trailing suction hopper dredger sequentially from the executable candidate sequence. After the departure time is determined, the operation sequence is determined according to the time order of the corresponding vessel operation cycles on the scheduling timeline, and a correspondence is established between vessels, construction sections, departure times, and operation sequences to form a multi-vessel scheduling plan. The multi-vessel scheduling plan serves as the scheduling status output during the construction execution phase and also as the basis for subsequent dynamic adjustments to the current schedule.

[0052] When a voyage is executed according to a multi-vessel scheduling plan, its vessel identification, voyage identification, construction section, and departure time can be correlated with subsequent shipboard operation records. After the completion of each stage, such as loading, departure, dumping mud, and return, the voyage completion status is used to determine whether the corresponding vessel can enter the next reallocation, thus linking the initial scheduling results with the status updates of the construction stages.

[0053] S4 acquires real-time water depth measurement results, shallow point distribution, siltation changes, and voyage completion status to determine the remaining workload and construction priority of the construction section. Based on the remaining workload, construction priority, and candidate departure sequences, it updates the section allocation, work sequence, and departure time of the multi-vehicle scheduling plan.

[0054] Please see Figure 5 The construction survey vessel or trailing hopper dredger uses depth sounding equipment to collect water depth data with location markers and acquisition times. Location markers are used to assign water depth data to the corresponding construction section and the planar grid within the construction section, while acquisition times are used to distinguish measurement cycles and establish the time correspondence of the same grid in different measurement cycles.

[0055] S401: Water depth data for the same measurement period are collected according to the spatial boundaries and planar grids of the construction section. For each grid, the measured bottom elevation corresponding to the water depth measurement result is compared with the design bottom elevation grid by grid. When the measured bottom elevation is higher than the design bottom elevation, the grid is identified as a substandard grid, and its grid location and the elevation difference between the measured bottom elevation and the design bottom elevation are recorded. The locations of all substandard grids and their corresponding elevation differences form a shallow point distribution.

[0056] For the same grid, the measured bottom elevation is read from two adjacent measurement cycles. The changes in bottom elevation are compared according to the order of the two measurement cycles, and the siltation change of the corresponding grid is determined by this change. Thus, the water depth measurement results reflect the current measured state, the shallow point distribution reflects the spatial location where the design bottom elevation has not yet been reached, and the siltation change reflects the changing trend of the bottom surface state between adjacent measurement cycles.

[0057] Simultaneously, the loading completion record, departure completion record, mud dumping completion record, and return completion record corresponding to the vessel identifier and voyage identifier are retrieved from the ship's operation log. If a completion record exists for all four preset operation stages, the voyage status is determined to be completed. If no completion record has been generated for any preset operation stage, the corresponding voyage is not considered a completed voyage for vessel reallocation determination.

[0058] S402, based on the spatial boundary of the construction section, associate the water depth measurement results, shallow point distribution, and siltation changes within the same measurement period. For substandard construction units where the measured bottom elevation is higher than the design bottom elevation, determine the remaining dredging volume based on their failure to reach the design bottom elevation, and correct this dredging volume according to the corresponding siltation changes. Accumulate the dredging volumes of all substandard construction units within the same construction section after correction for siltation changes to obtain the remaining workload for that construction section.

[0059] The remaining workload serves as a data object representing the scale of unfinished construction in a section during scheduling updates, and does not directly replace construction priority. Construction priority is also subject to hierarchical constraints based on the urgency of milestone deadlines, the status of shallow water treatment, and the status of siltation treatment.

[0060] S403 stipulates that construction sections to be constructed are arranged according to a hierarchical relationship: priority is given to sections with urgency, followed by shallow spot treatment, and then siltation treatment. First, the urgency of the construction sections is compared; if the urgency is the same, the shallow spot treatment priority is compared; if the first two priorities are the same, the siltation treatment priority is compared. When multiple construction sections have the same priority in the first three levels, the remaining work volume of each construction section is read, and the priority is determined according to the remaining work volume from largest to smallest. The results of the above hierarchical comparison form the construction priority.

[0061] The above priority generation process uses a layer-by-layer comparison method. If no identical priority is found in the previous layer, the conditions of the next layer are not used to change the already determined priority relationship. Therefore, the remaining workload is only used to determine the priority of sections when the priorities of the first three layers are consistent, and is not used as a scoring item to be added together with the priority of urgency, shallow spot treatment, or siltation treatment.

[0062] S404: Read the sections to be constructed according to construction priority from high to low. For the current section to be constructed, determine the trailing suction hopper dredgers that can be reassigned based on the completion status of the voyage. Vessels that have completed the voyage are added to the list of vessels that can be reassigned, while vessels with pre-set work phases that have not yet been completed remain in their current voyage status.

[0063] For trailing suction hopper dredgers that can be reassigned, the vessel segment matching relationship formed in S1 is invoked, retaining only candidate vessels that match the current construction segment. From these candidate vessels, the candidate departure sequences already formed in S2 and S3 are read, and departure candidates that meet the current scheduling conditions are obtained according to these sequences. Subsequently, the corresponding construction segment, vessel, and candidate departure time are written into the updated scheduling position in the original multi-vessel scheduling plan, and the operation sequence is updated based on the new vessel operation cycle time relationship.

[0064] After the update, the section allocation, work sequence, and departure time in the multi-vessel scheduling plan correspond to the current remaining work volume, construction priority, and the status of vessels that can be reassigned. Once a new measurement cycle generates new depth measurement results, shallow point distribution, and siltation changes, or after a new voyage completion record is formed, the corresponding update is performed according to S4. This ensures that the scheduling status continues to evolve with the on-site construction status without altering the vessel safety matching constraints formed in S1 or the ranking basis determined by the total energy consumption of the voyage in the candidate departure sequences.

[0065] Through S1 to S4, the spatial state of the project quantity is first converted into a matching relationship between construction sections and vessel sections; the vessel, route, and time-varying tidal current states are then converted into the total energy consumption of the voyage corresponding to each candidate departure time; the total energy consumption of the voyage is sorted in ascending order and constrained by node schedules, navigation windows, and vessel operation cycles to form an executable multi-vehicle scheduling plan; the water depth, shallow points, siltation, and voyage completion status generated during construction then have a reverse effect on the remaining project quantity, construction priority, and scheduling position updates of the construction section. The output of each stage becomes the explicit input of the next stage, thus forming a continuous processing chain from construction zoning, navigation energy consumption calculation, scheduling generation to construction status feedback updates.

[0066] Example 2 Please see Figure 6 This embodiment provides a scheduling system for trailing suction hopper dredging projects, including a construction zone matching module, an energy consumption prediction module, a multi-vessel scheduling module, and a scheduling module. The same process of zone matching, voyage energy consumption calculation, schedule generation, and feedback update is achieved through data transmission between modules.

[0067] The construction zoning matching module receives the project quantity distribution, milestone schedules, navigation windows, and the hopper capacity, draft, speed, and operational cycle of each trailing suction hopper vessel. The project quantity distribution is used to perform calculation unit division, candidate section merging, and target project quantity accumulation. The hopper capacity is used to determine the loading capacity for a single voyage, and the draft is compared with the minimum construction water depth and the safe margin water depth threshold. The construction zoning matching module outputs the matching relationship between construction sections and vessel sections, and then transmits the vessel section matching relationship to the energy consumption prediction module.

[0068] The energy consumption prediction module receives candidate departure times and the vessel segment matching relationships output by the construction zone matching module, and determines the trailing suction hopper dredger, construction section, and dumping area corresponding to the candidate departure times. The module then reads the route position, estimated transit time, and tidal current forecast for the round trip, decomposes the tidal current into downstream and transverse currents based on the vessel's heading, and calls upon model parameters matching the vessel type, main propulsion system, and full-load or empty-load operating conditions. Speed, draft, downstream current, and transverse current are processed by the navigation energy consumption prediction model to form a segment propulsion power sequence, which, combined with the segment's navigation time, yields the energy consumption for full-load departure, empty-load return, and total voyage energy consumption. The energy consumption prediction module then transmits the candidate departure times, corresponding total voyage energy consumption, and the correlation data between the vessel and the construction section to the multi-vessel scheduling module.

[0069] The multi-vessel scheduling module receives the aforementioned associated data and arranges candidate departure times from low to high based on the total energy consumption of the voyage, according to the same trailing hopper dredger and the same construction section, forming a candidate departure sequence. The module reads the node duration, navigation window, and vessel operation cycle, compares the cycle end time and the estimated passage time of the restricted section corresponding to the candidate departure times, retains candidates that meet the corresponding time constraints, and determines the departure time and operation sequence of each vessel accordingly. The output data object is the multi-vessel scheduling plan, which establishes the correspondence between vessels, construction sections, departure times, and operation sequences. The multi-vessel scheduling module transmits the current multi-vessel scheduling plan and the candidate departure sequences used in the scheduling to the scheduling module.

[0070] The scheduling module receives the current multi-vessel scheduling plan and candidate departure sequences output by the multi-vessel scheduling module. It also receives water depth measurement results, shallow point distribution, and siltation changes from the sounding equipment of the construction survey vessel or trailing hopper dredger, and receives the voyage completion status from the shipboard operation records. The scheduling module correlates the measurement data with the spatial boundaries of the construction section, identifies the substandard construction units, and accumulates the remaining workload. Then, it determines the construction priority based on the urgency of the schedule, the priority of shallow point treatment, the priority of siltation treatment, and the size of the remaining workload when the first three priorities are consistent.

[0071] The scheduling module determines the construction section to be updated based on construction priority, identifies the trailing suction hopper dredgers that can be reassigned based on the voyage completion status, then uses the vessel section matching relationship to exclude vessels that do not match the current construction section, and determines new departure candidates based on the candidate departure sequence. The new construction section, vessels, and candidate departure times are written into the corresponding scheduling positions of the current multi-vessel scheduling plan, and the operation sequence and departure times are updated to form the updated multi-vessel scheduling plan. When subsequent depth measurement results, shallow point distribution, siltation changes, or voyage completion status change, the updated multi-vessel scheduling plan continues to serve as the current scheduling status for the next round of scheduling processing.

[0072] The methods for forming construction sections, determining minimum construction water depth, comparing water depth margins, decomposing tidal currents, determining the input-output relationship of the navigation energy consumption prediction model, sorting direction of candidate departure sequences, comparing navigation windows, comparing node durations, determining construction priority levels, and updating schedules in the above embodiments are only used to illustrate the possible implementations of the present invention. Any equivalent substitutions, modifications, or combinations made by those skilled in the art to the corresponding processing steps, module acceptance methods, data field names, or equivalent processing subjects or carrier forms without departing from the technical solutions described in this invention should fall within the protection scope of this technical solution, but the established input, processing, output, and feedback relationships between the various technical objects should not be changed due to changes in names or carrier forms.

Claims

1. A method for intelligent scheduling and dispatching of trailing suction hopper dredging projects, characterized in that, Includes the following steps: The project quantity distribution, node construction period, navigation window and the hopper capacity of each trailing suction hopper vessel, the draft corresponding to the vessel's operating conditions, including the draft under full load and empty load conditions, as well as the speed and vessel operation cycle, are obtained. Based on the project quantity distribution, the hopper capacity and the draft, the construction sections are divided and the vessel section matching relationship is determined. Read the candidate departure time, and based on the matching relationship of the ship segments, along the round-trip route from the trailing suction hopper to the dumping area, determine the downstream current and the lateral current according to the tidal current forecast and the ship's heading. Input the ship speed, the draft, the downstream current and the lateral current into the navigation energy consumption prediction model, calculate the propulsion power for fully loaded departure and empty return, multiply the propulsion power of each segment by the corresponding segment's navigation time or integrate it according to time to obtain the navigation energy consumption of each segment, and accumulate the energy consumption for fully loaded departure and empty return to obtain the total energy consumption of the voyage. The candidate departure times are sorted in ascending order according to the total energy consumption of the voyage to construct a candidate departure sequence. The departure times and operation sequences are determined by combining the node construction period, the navigation window and the ship operation cycle to generate a multi-ship scheduling plan. Real-time acquisition of water depth measurement results, shallow point distribution, siltation changes, and voyage completion status; determination of the remaining workload and construction priority of the construction section; and updating of the section allocation, operation sequence, and departure time of the multi-vessel scheduling plan based on the remaining workload, construction priority, and candidate departure sequence.

2. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 1, characterized in that, The steps of dividing the construction section and determining the matching relationship of the vessel section include: Record the plane position, current bottom elevation and design bottom elevation corresponding to the distribution of the project volume, divide the calculation unit according to the continuous construction boundary, merge and accumulate the project volume according to the plane adjacency relationship and construction depth difference of the adjacent calculation unit, and generate candidate sections; Read the target engineering volume and minimum construction water depth of the candidate section, determine the hopper capacity as the single voyage loading capacity, compare the draft with the minimum construction water depth, eliminate trailing suction hopper dredgers that do not meet the safe margin water depth conditions, and obtain the candidate vessel set. The single-voyage loading capacity of the trailing suction hopper dredgers in the candidate vessel set is compared with the target engineering volume. Adjacent candidate segments that match the candidate vessel set are merged to generate the construction segment and establish the vessel segment matching relationship.

3. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 2, characterized in that, The specific safe margin water depth conditions are as follows: Obtain the construction water depth corresponding to each calculation unit within the candidate section, determine the minimum construction water depth as the minimum construction water depth, obtain the full-load draft of the trailing suction hopper corresponding to the candidate section, calculate the water depth difference between the minimum construction water depth and the full-load draft, and obtain the water depth margin. A preset safe margin water depth threshold is read, and the margin water depth is compared with the safe margin water depth threshold. When the margin water depth is not less than the safe margin water depth threshold, it is determined that the trailing suction hopper vessel meets the safe margin water depth condition; when the margin water depth is less than the safe margin water depth threshold, the corresponding trailing suction hopper vessel is excluded from the candidate vessel set.

4. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 1, characterized in that, The steps for calculating the total energy consumption of the voyage include: Read the candidate departure times, determine the corresponding trailing suction hopper dredger, construction section and dumping area according to the vessel section matching relationship, match the tidal current forecast according to the route position and expected arrival time from the construction section to the dumping area and from the dumping area to the construction section, and obtain the round-trip route tidal current data; The flow velocity and direction corresponding to the tidal current data of the round-trip route, as well as the ship's heading at each route position of the trailing suction hopper, are obtained. The flow direction is decomposed along the ship's heading to obtain the downstream flow, and the flow component perpendicular to the ship's heading is determined as the transverse flow, thereby generating the route flow field input data. The speed, draft, and route flow field input data corresponding to the fully loaded departure and empty return are respectively input into the navigation energy consumption prediction model. The propulsion power of each segment of the round trip route is obtained in sequence, and the energy consumption is accumulated according to the segment travel time. The energy consumption of the fully loaded departure and the energy consumption of the empty return are combined to obtain the total energy consumption of the voyage.

5. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 4, characterized in that, The process of constructing the navigation energy consumption prediction model includes: The ship speed, draft, ship heading, tidal current speed, tidal current direction and measured propulsion power are obtained synchronously from the historical voyage. The tidal current is decomposed according to the ship heading to obtain the historical downstream current and historical cross current. The samples are labeled according to the full load or empty load conditions to generate a training sample set. Using the speed, draft, historical downstream current and historical lateral current as inputs, and the measured propulsion power as the prediction target, the model is trained in groups according to ship type and operating condition to obtain sub-models for different operating conditions. The sub-model for different operating conditions is validated using historical voyages that were not used in training. When the propulsion power prediction deviation meets the preset error threshold, the model parameters are fixed, and an index relationship between ship type, operating condition and the model parameters is established to obtain the navigation energy consumption prediction model.

6. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 4, characterized in that, The process of obtaining the energy consumption for fully loaded departure and empty return specifically includes: The vessel identifier of the corresponding trailing suction hopper vessel is read according to the vessel segment matching relationship. The navigation energy consumption model parameters that match the vessel type and main propulsion device are called according to the vessel identifier. The corresponding operating condition identifier is determined according to the state of sailing out fully loaded or returning empty, and the navigation operating condition parameters are obtained. The speed, draft, downstream current, lateral current, and navigation parameters are formed into a continuous input sequence according to the expected elapsed time of each segment. This sequence is then input into the navigation energy consumption prediction model to predict the propulsion power required to maintain the corresponding speed for each segment under full-load departure or empty-load return conditions, thus obtaining the segment propulsion power sequence. The sailing time of each segment is determined according to the range and corresponding speed of each segment. The propulsion power sequence of each segment is associated with the sailing time of the corresponding segment and accumulated to obtain the energy consumption for full-load departure or empty-load return.

7. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 1, characterized in that, The steps for generating the multi-ship scheduling plan include: Obtain the total energy consumption of the voyage corresponding to each candidate departure time, establish a record of departure time and energy consumption corresponding to the same trailing suction hopper vessel and the same construction section, and arrange the record of departure time and energy consumption corresponding to the voyage in order of low to high according to the total energy consumption of the voyage to obtain the candidate departure sequence. Read the candidate departure times one by one according to the candidate departure sequence, calculate the cycle end time and the expected passage time of the restricted section based on the ship operation cycle, compare the expected passage time of the restricted section with the navigation window, and compare the cycle end time with the node schedule to filter out the executable candidate sequence; The departure time of each trailing suction hopper vessel is determined sequentially from the executable candidate sequence, the operation sequence is determined according to the time sequence of the corresponding operation cycle, the correspondence between the vessel, construction section, departure time and operation sequence is established, and the multi-vessel scheduling plan is generated.

8. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 1, characterized in that, The steps of determining the remaining workload and the construction priority and updating the multi-vessel scheduling plan include: The water depth measurement results, shallow point distribution, siltation changes, and voyage completion status are obtained. Data from the same measurement cycle are associated with the spatial boundary of the construction section. The measured bottom elevation is compared with the design bottom elevation to determine the substandard construction units. The dredging volume of the substandard construction units is corrected based on the siltation changes. The remaining engineering volume of each construction section is accumulated. The construction sections are arranged according to a hierarchical relationship of priority based on urgency, followed by shallow spot treatment, and then siltation treatment. When the first three levels of priority are the same for multiple construction sections, the order of priority is determined according to the remaining amount of work from largest to smallest, thus generating the construction priority. The construction sections to be constructed are read from high to low according to the construction priority. The trailing suction hopper vessels that can be reassigned are determined according to the completion status of the voyage. Among the candidate vessels that meet the vessel section matching relationship, the departure candidates are selected according to the candidate departure sequence. The corresponding construction section, vessel and candidate departure time are written into the original schedule position. The section allocation, the operation sequence and the departure time are updated to generate an updated multi-vessel schedule plan.

9. The intelligent scheduling method for trailing suction hopper dredging projects according to claim 8, characterized in that, The process of acquiring the water depth measurement results, the shallow point distribution, the siltation changes, and the cruise completion status in real time specifically includes: Collect water depth data with location markers and collection time output from the depth sounding equipment of the construction survey vessel or trailing suction hopper vessel. Collect the water depth data according to the planar grid of the construction section and compare it grid by grid with the design bottom elevation to obtain the water level elevation corresponding to the collection time. Correct the depth sounding data for water level, depth sounding transducer installation deviation and necessary attitude. Convert the water depth data into the measured bottom elevation under a unified elevation benchmark. Spatially correlate the grid positions where the measured bottom elevation is higher than the design bottom elevation and the corresponding elevation difference to generate the water depth measurement results and the shallow point distribution. Read the measured bottom elevation of two adjacent measurement cycles of the same grid, compare the changes in bottom elevation according to the measurement time to determine the siltation change, and read the loading, departure, dumping and return completion records corresponding to the ship identification and voyage identification from the ship operation record. When all preset operation stages are recorded as completed, the corresponding voyage is determined to be completed, and the voyage completion status is obtained.

10. An intelligent scheduling system for trailing suction hopper dredging projects, characterized in that, The system is used to implement the method according to any one of claims 1-9, comprising: The construction zone matching module is used to obtain the project quantity distribution, node schedule, navigation window, and the hopper capacity, draft, speed and vessel operation cycle of each trailing suction hopper vessel. Based on the project quantity distribution, hopper capacity and draft, the construction zone is divided and the vessel zone matching relationship is determined. The energy consumption prediction module, connected to the construction zone matching module, is used to read candidate departure times, and based on the vessel section matching relationship, along the round-trip route from the trailing suction hopper to the dumping area, determine the downstream current and the lateral current according to the tidal current forecast and the vessel's heading. The speed, the draft, the downstream current and the lateral current are input into the navigation energy consumption prediction model to calculate the propulsion power for fully loaded departure and empty return and accumulate the total energy consumption of the voyage. The multi-ship scheduling module, connected to the energy consumption prediction module, is used to sort the candidate departure times in ascending order according to the total energy consumption of the voyage, construct a candidate departure sequence, and determine the departure time and operation sequence by combining the node construction period, the navigation window and the ship operation cycle, and generate a multi-ship scheduling plan. The scheduling module, connected to the multi-vessel scheduling module, is used to acquire water depth measurement results, shallow point distribution, siltation changes and voyage completion status in real time, determine the remaining workload and construction priority of the construction section, and update the section allocation, operation sequence and departure time of the multi-vessel scheduling plan based on the remaining workload, the construction priority and the candidate departure sequence.