Dredging operation parameter pre-dredging intelligent recommendation method based on deep learning network model

Abstract

The invention discloses a dredging operation parameter pre-dredging intelligent recommendation method based on a deep learning network model. The method comprises the following steps: in response to a dredger control request, screening original ship data and original management data according to the type of a dredging working condition to obtain construction data and power data, and respectively determining label information corresponding to the construction data and the power data; training according to the construction data and the corresponding label information to obtain a target construction parameter recommendation model, and training according to the power data and the corresponding label information to obtain a target power pre-distribution model; and according to the optimization mode of the recommended parameters, based on the target construction parameter recommendation model and the target power pre-distribution model, input parameter combination optimization is carried out, an optimal parameter combination is obtained, and according to the optimal parameter combination and the dredging working condition, the target dredger is controlled to work. According to the method, the data can be preprocessed in combination with the dredging working condition, and the optimal recommendation parameters are obtained based on the prediction model and the optimization mode.

Description

Technical Field

[0001] The present invention relates to the technical field of dredging, and particularly to an intelligent pre-dredging operation parameter recommendation method for dredging based on a deep learning network model. Background Art

[0002] A trailing suction hopper dredger faces complex environmental conditions, variable soil properties, and different designed dredging depths during dredging operations. These factors directly affect construction efficiency and ship energy efficiency. Traditionally, construction workers would control the trailing suction hopper dredger with different initial parameters based on manual experience. However, such operations have a high degree of variation, resulting in different yields and energy consumption for each voyage, and it is difficult to ensure stable overall construction efficiency and cost.

[0003] Therefore, how to preprocess data in combination with dredging working conditions and obtain optimal recommended parameters based on a prediction model and an optimization mode to achieve precise and efficient control of the dredger is an urgent problem to be solved at present. Summary of the Invention

[0004] The present invention provides an intelligent pre-dredging operation parameter recommendation method for dredging based on a deep learning network model, which preprocesses data in combination with dredging working conditions and obtains optimal recommended parameters based on a prediction model and an optimization mode of deep learning to achieve precise and efficient control of the dredger.

[0005] According to one aspect of the present invention, there is provided an intelligent pre-dredging operation parameter recommendation method for dredging based on a deep learning network model, including:

[0006] In response to a dredger control request, screening the original ship data and the original management data according to the category of the dredging working conditions to obtain construction data and power data, and respectively determining the label information corresponding to the construction data and the power data;

[0007] Training a target construction parameter recommendation model according to the construction data and its corresponding label information, and training a target power pre-allocation model according to the power data and its corresponding label information;

[0008] Based on the optimization mode of the recommended parameters, performing input parameter combination optimization based on the target construction parameter recommendation model and the target power pre-allocation model to obtain an optimal parameter combination, and controlling the operation of the target dredger according to the optimal parameter combination and the dredging working conditions.

[0009] According to another aspect of the present invention, there is provided an intelligent pre-dredging operation parameter recommendation device for dredging based on a deep learning network model, including:

[0010] A determination module, configured to screen the original ship data and the original management data according to the category of the dredging working conditions in response to a dredging ship control request, so as to obtain construction data and power data, and respectively determine the label information corresponding to the construction data and the power data;

[0011] A training module, configured to train a target construction parameter recommendation model according to the construction data and its corresponding label information, and train a target power pre-allocation model according to the power data and its corresponding label information;

[0012] A control module, configured to perform input parameter combination optimization based on the target construction parameter recommendation model and the target power pre-allocation model according to the optimization mode of the recommended parameters, obtain an optimal parameter combination, and control the operation of the target dredging ship according to the optimal parameter combination and the dredging working conditions.

[0013] According to another aspect of the present invention, there is provided an electronic device, the electronic device includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the intelligent pre-dredging operation parameter recommendation method based on the deep learning network model according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the intelligent pre-dredging operation parameter recommendation method based on the deep learning network model according to any embodiment of the present invention when executed by a processor.

[0018] According to another aspect of the present invention, there is also provided a computer program product, the computer program product includes a computer program, and the computer program implements the intelligent pre-dredging operation parameter recommendation method based on the deep learning network model according to any embodiment of the present invention when executed by a processor.

[0019] In the technical solution of the embodiment of the present invention, in response to a dredger control request, the original ship data and the original management data are screened according to the category of the dredging working condition to obtain construction data and power data, and the label information corresponding to the construction data and the power data is determined respectively; a target construction parameter recommendation model is trained according to the construction data and its corresponding label information, and a target power pre-allocation model is trained according to the power data and its corresponding label information; according to the optimization mode of the recommended parameters, input parameter combination optimization is performed based on the target construction parameter recommendation model and the target power pre-allocation model to obtain an optimal parameter combination, and the target dredger is controlled to operate according to the optimal parameter combination and the dredging working condition. By preprocessing the data in combination with the dredging working condition and obtaining the optimal recommended parameters based on the prediction model and the optimization mode, the present invention can achieve precise and efficient control of the dredger.

[0020] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Brief Description of the Drawings

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0022] Figure 1 It is a flowchart of a pre-dredging intelligent recommendation method for dredging operation parameters based on a deep learning network model provided in Embodiment 1 of the present invention;

[0023] Figure 2 It is a flowchart of a pre-dredging intelligent recommendation method for dredging operation parameters based on a deep learning network model provided in Embodiment 2 of the present invention;

[0024] Figure 3 It is a structural block diagram of a pre-dredging intelligent recommendation device for dredging operation parameters based on a deep learning network model provided in Embodiment 3 of the present invention;

[0025] Figure 4 It is a schematic structural diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Embodiments

[0026] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] It should be noted that the terms "first", "second", "target", "candidate", "alternative", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data used can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. The acquisition, storage, use, processing, etc. of data in the technical solutions of this application all comply with the relevant regulations of national laws and regulations.

[0028] Embodiment 1

[0029] Figure 1 is a flowchart of a method for intelligent pre-dredging operation parameter recommendation based on a deep learning network model provided by Embodiment 1 of the present invention; this embodiment is applicable to the situation where an intelligent integration platform processes original ship data and original management data and trains a model to optimize and obtain an optimal parameter combination for precise and efficient control of a dredger. This method can be executed by a device for intelligent pre-dredging operation parameter recommendation based on a deep learning network model. The device for intelligent pre-dredging operation parameter recommendation based on a deep learning network model can be implemented in the form of hardware and / or software. The device for intelligent pre-dredging operation parameter recommendation based on a deep learning network model can be configured in an electronic device and executed by the intelligent integration platform, as Figure 1 shown, the method for intelligent pre-dredging operation parameter recommendation based on a deep learning network model includes:

[0030] S101. In response to a dredger control request, screen the original ship data and the original management data according to the category of the dredging working condition to obtain construction data and power data, and respectively determine the tag information corresponding to the construction data and the power data.

[0031] Among them, the dredger control request refers to a request for the intelligent integration platform to process the original ship data of the intelligent dredging control system and the original management data of the power management system and perform model training to determine recommended parameters to achieve the control of the target dredger. The categories of dredging working conditions can be dredging conditions and mud suction conditions. The original ship data refers to the ship operation data collected by the intelligent dredging control system, specifically including at least one of the following: mud flow rate, mud concentration, dredging speed, high-pressure water jet pump pressure, overflow cylinder height, and mud pump speed. The original management data refers to the management equipment power data collected by the power management system, specifically including at least one of the following: underwater pump power, in-cabin pump power, and high-pressure water jet pump power. The construction data refers to the data in the original ship data corresponding to the category of the dredging working condition, and the power data refers to the data in the original management data corresponding to the category of the dredging working condition. The tag information corresponding to the construction data refers to the instantaneous output and fuel consumption per 10,000 cubic meters of soil of the target dredger at the collection time corresponding to each group of construction data. The tag information corresponding to the power data refers to the instantaneous output and fuel consumption per 10,000 cubic meters of soil of the target dredger at the collection time corresponding to each group of power data.

[0032] Optionally, according to the dredger control request, if the configuration information of the dredging soil quality and the designed dredging depth is detected, the category of the dredging working condition is determined as the dredging condition; if the configuration information of the discharge distance is detected, the category of the dredging working condition is determined as the mud blowing condition.

[0033] Optionally, filter the original ship data and the original management data to obtain the construction data and the power data, and respectively determine the tag information corresponding to the construction data and the power data, including: according to the category of the dredging working condition, filter the original ship data collected by the intelligent dredging control system within a preset historical time period to determine the construction data, and determine the instantaneous output and fuel consumption per 10,000 cubic meters of soil corresponding to the construction data as the tag information corresponding to the construction data; according to the category of the dredging working condition, filter the original management data collected by the power management system within a preset historical time period to determine the power data, and determine the instantaneous output and fuel consumption per 10,000 cubic meters of soil corresponding to the power data as the tag information corresponding to the power data.

[0034] Optionally, according to the collection time corresponding to each group of construction data, match in the database of the intelligent acquisition platform that locally records the instantaneous output and fuel consumption per 10,000 cubic meters of soil to determine the instantaneous output and fuel consumption per 10,000 cubic meters of soil corresponding to the construction data, that is, determine the tag information corresponding to the construction data; similarly, according to the collection time corresponding to each group of power data, match in the database of the intelligent acquisition platform that locally records the instantaneous output and fuel consumption per 10,000 cubic meters of soil to determine the instantaneous output and fuel consumption per 10,000 cubic meters of soil corresponding to the power data, that is, determine the tag information corresponding to the power data.

[0035] Optionally, according to the category of the dredging working conditions, the intelligent dredging control system, the power management system, the original ship data and the original management data collected within a preset historical time period are respectively screened to determine the construction data and the power data, including: if the category of the dredging working condition is the dredging condition, the mud flow rate, the mud concentration, the dredging speed, the pressure of the high-pressure water injection pump and the height of the overflow cylinder within the preset historical time period are determined as the construction data, and the underwater pump power, the in-cabin pump power and the high-pressure water injection pump power are determined as the power data; if the category of the dredging working condition is the mud blowing condition, the mud flow rate, the mud concentration, the pressure of the high-pressure water injection pump and the mud pump speed within the preset historical time period are determined as the construction data, and the in-cabin pump power and the high-pressure water injection pump power are determined as the power data.

[0036] S102. Train a target construction parameter recommendation model according to the construction data and its corresponding tag information, and train a target power pre-allocation model according to the power data and its corresponding tag information.

[0037] Among them, the deep learning network model refers to the target construction parameter recommendation model and the target power pre-allocation model. The target construction parameter recommendation model and the target power pre-allocation model can be LSTM (Long Short-Term Memory) prediction models. The input data of the target construction parameter recommendation model is the construction data, and the output data is the predicted instantaneous power and the predicted fuel consumption per ten thousand cubic meters of soil, that is, the tag information corresponding to the construction data. The input data of the target power pre-allocation model is the power data, and the output data is the predicted instantaneous power and the predicted fuel consumption per ten thousand cubic meters of soil, that is, the tag information corresponding to the power data.

[0038] Optionally, training a target construction parameter recommendation model according to the construction data and its corresponding tag information includes: inputting the construction data into an initial prediction model (such as an initialized LSTM prediction model) to obtain the predicted instantaneous output and the predicted fuel consumption per ten thousand cubic meters of soil; based on the predicted instantaneous output and the predicted fuel consumption per ten thousand cubic meters of soil, combined with the instantaneous output and the fuel consumption per ten thousand cubic meters of soil corresponding to the construction data, iteratively train the initial prediction model based on the mean square error loss function to obtain the target construction parameter recommendation model.

[0039] It should be noted that the process of iteratively training the initial prediction model according to the power data and the instantaneous output and the fuel consumption per ten thousand cubic meters of soil corresponding to the power data to obtain the target power pre-allocation model is similar to the process of obtaining the target construction parameter recommendation model according to the construction data, and will not be elaborated here.

[0040] Exemplarily, for the target construction parameter recommendation model, the concentration of the input layer can be set as C t and the flow rate as V t and the pressure of the high-pressure water injection pump as P t and the height of the overflow cylinder as Ht , the dredging speed is S t , and the input layer is set to have 5 neurons, and the input vector is represented as X t = [C t , V t , P t , H t , S t . The hidden layer adopts a multi-layer LSTM structure. Let the hidden layer have L layers, and the number of hidden units in the l-th layer be n l (for example, l = 1, 2, 3, n1 = 64, n2 = 64, n3 = 64). The output layer is set to have 2 neurons, representing the instantaneous production Y 1,t and the fuel consumption per 10,000 cubic meters of soil Y 2,t . The calculation of the output layer is: Y j,t = W j,o h t,L + b j,o (j = 1, 2), where W j,o and b j,o are the weight matrix and bias vector of the output layer. The weight parameters of the model are randomly initialized, with values ranging from [-0.1, 0.1], and the bias parameters are initialized to 0.

[0041] Exemplarily, for the target power pre-allocation model, the underwater pump power of the input layer can be set to P u,t , the in-cabin pump power to P i,t , and the high-pressure flushing pump power to P h,t , then the input vector is X′ t = [P u,t , P i,t , P h,t . The hidden layer adopts an L-layer LSTM structure (for example, n1 = 32, n2 = 32, n3 = 32). The output layer is the instantaneous production Y 1,t and the fuel consumption per 10,000 cubic meters of soil Y 2,t . The calculation method of the output layer, the initialization method of the weights and bias parameters are the same as those of the target construction parameter recommendation model.

[0042] Exemplarily, the construction data can be divided into a training set and a validation set according to a ratio (70% for training, 30% for validation) for model training. Let the number of training set samples be N, and the mini-batch gradient descent algorithm (batchsize is set to 32) is used for model training. The loss function selects the mean square error (MSE) function, and the corresponding loss function can be expressed by the following formula:

[0043]

[0044] where, and They are the predicted instantaneous production and the predicted fuel consumption per 10,000 cubic meters of soil predicted by the model. During the training process, the early stopping method can be used to prevent overfitting. When the loss function value on the validation set no longer decreases for 10 consecutive epochs (number of traversals), the training is stopped, and the optimal weight parameters of the model are saved to obtain the target construction parameter recommendation model. It should be noted that the training process of the target power pre-allocation model is similar to that of the target construction parameter recommendation model, so it will not be elaborated here.

[0045] S103. According to the optimization mode of the recommended parameters, based on the target construction parameter recommendation model and the target power pre-allocation model, perform input parameter combination optimization to obtain the optimal parameter combination, and control the operation of the target dredger according to the optimal parameter combination and the dredging working conditions.

[0046] Among them, the optimization mode refers to the strategy mode followed for finding the optimal parameter combination. The optimization mode can be the maximum production mode or the economic production mode. The maximum production mode is the optimization mode aiming at the maximum instantaneous production, and the economic production mode is the optimization mode aiming at the lowest fuel consumption per 10,000 cubic meters of soil. The optimal parameter combination includes an optimal set of construction data and an optimal set of power data.

[0047] Optionally, according to the optimization mode of the recommended parameters, based on the target construction parameter recommendation model and the target power pre-allocation model, perform input parameter combination optimization to obtain the optimal parameter combination, including: constructing an objective function corresponding to the optimization mode according to the optimization mode of the recommended parameters, the weights of the construction model, and the weights of the power model; according to the objective function and the preset optimization algorithm, optimize the input parameter combination of the target construction parameter recommendation model and the target power pre-allocation model to obtain the optimal parameter combination that meets the optimization mode.

[0048] Exemplarily, if the optimization mode is the maximum production mode, the following objective function can be constructed according to the weights of the construction model and the power model:

[0049] J max (θ) = α·Y 1,c (C, V, P, H, S) + β·Y 1,p (P u , P i , P h )

[0050] Among them, Y 1,c is the instantaneous production predicted by the target construction parameter recommendation model (based on the concentration C, flow velocity V, high-pressure water pump pressure P, overflow cylinder height H, and dredging speed S). Y 1,p is the instantaneous production predicted by the target power pre-allocation model (based on the underwater pump power P u , the in-cabin pump power P i, the power P of the high-pressure flush water pump h ). α and β are the weights of the construction model and the power model, satisfying α + β = 1.

[0051] Exemplarily, in the maximum production mode, a hybrid optimization algorithm combining genetic algorithm and gradient ascent method can be adopted to perform input parameter combination optimization to obtain the optimal parameter combination. For example, first, encode the construction parameters and equipment power parameters using the genetic algorithm (adopting binary encoding), generate the initial population, and set the population size to M = 100. Let the i-th individual be θ i , and its fitness value F(θ i ) = Y1(θ i ). The crossover operation adopts single-point crossover, and the crossover probability is set to p c = 0.7. The mutation operation adopts bit mutation, and the mutation probability is set to p m = 0.01. In the process of population evolution in each generation, use the gradient ascent method to perform local optimization on each individual. Let the learning rate be α, and the gradient calculation is as follows: The individual update formula is: After multiple generations of evolution (G > 100 generations), the parameter combination that maximizes the instantaneous production is obtained.

[0052] Exemplarily, if the optimization mode is the economic production mode, then according to the weights of the construction model and the power model, the following objective function can be constructed:

[0053] J econ = min(γ·Y 2,c (C, V, P, H, S) + θY 2,p (P u , P i , P h ))

[0054] where Y 2,c is the fuel consumption per 10,000 cubic meters of soil predicted by the target construction parameter recommendation model (based on the concentration C, flow velocity V, pressure P of the high-pressure flush water pump, height H of the overflow cylinder, and dredging speed S. Y 2,p is the fuel consumption per 10,000 cubic meters of soil predicted by the target power pre-distribution model (based on the power P of the underwater pump u , the power P of the in-cabin pump i , and the power P of the high-pressure flush water pump h ). γ and δ are the weights of the construction model and the power model, satisfying γ + δ = 1.

[0055] Exemplarily, in the economic production mode, the NSGAII (Non-dominated Sorting Genetic Algorithm-II, multi-objective optimization genetic algorithm) algorithm can be adopted to perform input parameter combination optimization to obtain the optimal parameter combination.

[0056] Optionally, according to the optimal parameter combination and the dredging working condition, control the operation of the target dredger, including: if the category of the dredging working condition is the dredging condition, during the navigation of the target dredger, monitor the dredging depth of the target dredger; if the monitored dredging depth is less than the preset depth threshold, control the submersible pump and the toothed draghead according to the optimal parameter combination to perform dredging operations, so as to suck the sediment at the bottom of the water into the target mud tank inside the ship, and realize the dredging operation of the target dredger; if the monitored dredging depth is greater than or equal to the preset depth threshold, control the submersible pump, the in-tank pump and the toothed draghead according to the optimal parameter combination to perform dredging operations, so as to suck the sediment at the bottom of the water into the target mud tank inside the ship, and realize the dredging operation of the target dredger.

[0057] Among them, the target dredger includes a submersible pump, an in-tank pump, a high-pressure water jet pump, an overflow cylinder, a draghead cover, and a propeller propulsion device.

[0058] Optionally, when controlling the submersible pump, the in-tank pump and the toothed draghead to perform dredging operations according to the optimal parameter combination, the submersible pump and the in-tank pump can work in series.

[0059] Optionally, the submersible pump and the toothed draghead can be controlled to perform dredging operations according to the optimal submersible pump power in the optimal parameter combination, and the submersible pump, the in-tank pump and the toothed draghead can be controlled to perform dredging operations according to the optimal submersible pump power and the optimal in-tank pump power in the optimal parameter combination.

[0060] Optionally, according to the optimal parameter combination and the dredging working condition, control the operation of the target dredger, including: if the category of the dredging working condition is the mud blowing condition, control the high-pressure water jet pump to blow the soil in the target mud tank into a slurry mixture according to the optimal parameter combination; when it is detected that the target dredger travels to the target location, control the in-tank pump to discharge the slurry mixture according to the optimal parameter combination, and realize the mud blowing operation of the target dredger.

[0061] Optionally, the propeller speed corresponding to the optimal speed can be determined according to the optimal speed in the optimal parameter combination, and the propeller of the target dredger can be controlled according to the propeller speed to adjust the speed of the target dredger.

[0062] Optionally, after controlling the dredger, if it is detected that the current time reaches the preset cycle time (such as 50 hours), or the number of voyages of the dredger leaving the ship reaches the preset voyage threshold (such as 30 voyages), it is determined that the model update conditions of the target construction parameter recommendation model and the target power pre-allocation model are met. At this time, execute the above-mentioned intelligent pre-recommendation method for dredging operation parameters based on the deep learning network model of the present invention, and retrain the model for parameter recommendation and dredger control.

[0063] In the technical solution of the embodiment of the present invention, in response to a dredger control request, according to the category of the dredging working condition, the original ship data and the original management data are screened to obtain construction data and power data, and the label information corresponding to the construction data and the power data is determined respectively; a target construction parameter recommendation model is trained according to the construction data and its corresponding label information, and a target power pre-allocation model is trained according to the power data and its corresponding label information; according to the optimization mode of the recommended parameters, input parameter combination optimization is performed based on the target construction parameter recommendation model and the target power pre-allocation model to obtain the optimal parameter combination, and the target dredger is controlled according to the optimal parameter combination and the dredging working condition. The present invention can realize precise and efficient control of the dredger by preprocessing data in combination with the dredging working condition and obtaining the optimal recommended parameters based on the prediction model and the optimization mode.

[0064] Embodiment 2

[0065] Figure 2 is a flowchart of an intelligent pre-dredging recommendation method for dredging operation parameters based on a deep learning network model provided by Embodiment 2 of the present invention; on the basis of the above embodiment, this embodiment provides an optimal example of an intelligent integration platform for processing the original ship data and the original management data, training a model to optimize and obtain the optimal parameter combination to realize pre-dredging parameter recommendation. Specifically, as Figure 2 shown, the method includes the following processes:

[0066] Optionally, in response to a dredger control request, the original ship data and the original management data are obtained and preprocessed, and historical data is further screened according to the category of the dredging working condition, that is, the original ship data and the original management data are screened. Specifically, the data under the dredging working condition involves the dredging soil quality and the designed dredging depth, and the data under the mud blowing working condition involves the dredging soil quality and the discharge distance.

[0067] Optionally, according to the obtained construction data and power data, an LSTM prediction model is constructed and the model is trained. Further, in combination with the optimization mode, construction parameter recommendation and power pre-allocation parameter recommendation are respectively performed in the case of the maximum output mode and the economic output mode, that is, the optimal parameter combination is determined. Among them, the construction parameter recommendation can be the mud flow rate, the mud concentration, the dredging speed, the pressure of the high-pressure water jet pump, and the height of the overflow cylinder, and the power pre-allocation parameter recommendation can be the power of the underwater pump, the power of the in-cabin pump, and the power of the high-pressure water jet pump.

[0068] Optionally, the process of training the model according to the historical data and determining the optimal parameter combination can be performed every 50 hours or 30 ship voyages to perform pre-dredging parameter recommendation and realize the control of the target dredger.

[0069] The technical solution of the present invention provides an intelligent pre-dredging recommendation solution for dredging operation parameters based on historical data. It can provide pre-dredging recommendations for dredging parameters in two intelligent dredging modes of maximum output and economic output for trailing suction hopper dredgers. The implementation process is based on intelligent optimization analysis of historical data. By means of a prediction model, the optimal combination of construction parameters under two working conditions of the dredging operation and the mud blowing operation of the trailing suction hopper dredger is determined, and the power pre-allocation of the equipment is carried out, which can provide an initial value for the intelligent one-key dredging of the whole ship and improve the operation efficiency of the dredger.

[0070] Embodiment III

[0071] Figure 3 It is a structural block diagram of an intelligent pre-dredging recommendation device for dredging operation parameters based on a deep learning network model provided in Embodiment III of the present invention; this embodiment is applicable to the situation where an intelligent integration platform processes original ship data and original management data, trains a model to optimize and obtain an optimal parameter combination for precise and efficient control of a dredger. The intelligent pre-dredging recommendation device for dredging operation parameters based on a deep learning network model provided in the embodiments of the present invention can execute the intelligent pre-dredging recommendation method for dredging operation parameters based on a deep learning network model provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method; the intelligent pre-dredging recommendation device for dredging operation parameters based on a deep learning network model can be implemented in the form of hardware and / or software, and is configured in an electronic device with a dredger control function and executed by an intelligent integration platform, such as Figure 3 As shown, the intelligent pre-dredging recommendation device for dredging operation parameters based on a deep learning network model specifically may include:

[0072] A determination module 301, configured to screen the original ship data and the original management data according to the category of the dredging working condition in response to a dredger control request, so as to obtain construction data and power data, and respectively determine the label information corresponding to the construction data and the power data;

[0073] A training module 302, configured to train a target construction parameter recommendation model according to the construction data and its corresponding label information, and train a target power pre-allocation model according to the power data and its corresponding label information;

[0074] A control module 303, configured to perform input parameter combination optimization based on the target construction parameter recommendation model and the target power pre-allocation model according to the optimization mode of the recommended parameters, obtain an optimal parameter combination, and control the operation of the target dredger according to the optimal parameter combination and the dredging working condition.

[0075] In the technical solution of the embodiment of the present invention, in response to a dredger control request, according to the category of the dredging working condition, the original ship data and the original management data are screened to obtain construction data and power data, and the label information corresponding to the construction data and the power data is determined respectively; a target construction parameter recommendation model is trained according to the construction data and its corresponding label information, and a target power pre-allocation model is trained according to the power data and its corresponding label information; according to the optimization mode of the recommended parameters, based on the target construction parameter recommendation model and the target power pre-allocation model, the input parameter combination is optimized to obtain the optimal parameter combination, and the target dredger is controlled to operate according to the optimal parameter combination and the dredging working condition. The present invention can realize precise and efficient control of the dredger by preprocessing data in combination with the dredging working condition and obtaining the optimal recommended parameters based on the prediction model and the optimization mode.

[0076] Further, the determining module 301 is specifically configured to:

[0077] According to the category of the dredging working condition, the intelligent dredging control system and the original ship data collected in a preset historical time period are screened to determine the construction data, and the instantaneous output and fuel consumption per ten thousand cubic meters of soil corresponding to the construction data are determined as the label information corresponding to the construction data;

[0078] According to the category of the dredging working condition, the original management data collected by the power management system in a preset historical time period are screened to determine the power data, and the instantaneous output and fuel consumption per ten thousand cubic meters of soil corresponding to the power data are determined as the label information corresponding to the power data.

[0079] Further, the control module 303 is specifically configured to:

[0080] According to the optimization mode of the recommended parameters, the construction model weight, and the power model weight, a target function corresponding to the optimization mode is constructed;

[0081] According to the target function and a preset optimization algorithm, the input parameter combination of the target construction parameter recommendation model and the target power pre-allocation model is optimized to obtain the optimal parameter combination that meets the optimization mode.

[0082] Further, the determining module 301 is further configured to:

[0083] If the category of the dredging working condition is the dredging operation condition, the mud flow rate, mud concentration, dredging speed, high-pressure water jet pump pressure, and overflow cylinder height within a preset historical time period are determined as the construction data, and the underwater pump power, in-cabin pump power, and high-pressure water jet pump power are determined as the power data;

[0084] If the type of the dredging operation mode is the mud blowing operation mode, the mud flow rate, mud concentration, high-pressure water jet pump pressure, and mud pump speed within a preset historical time period are determined as construction data, and the power of the in-cabin pump and the power of the high-pressure water jet pump are determined as power data.

[0085] Further, the control module 303 is specifically configured to:

[0086] If the type of the dredging operation mode is the dredging operation mode, during the navigation of the target dredger, monitor the dredging depth of the target dredger;

[0087] If it is monitored that the dredging depth is less than the preset depth threshold, control the underwater pump and the toothed cutter head to perform dredging operations according to the optimal parameter combination, so as to suck the sediment at the bottom of the water into the target mud cabin inside the ship, and realize the dredging operation of the target dredger;

[0088] If it is monitored that the dredging depth is greater than or equal to the preset depth threshold, control the underwater pump, the in-cabin pump, and the toothed cutter head to perform dredging operations according to the optimal parameter combination, so as to suck the sediment at the bottom of the water into the target mud cabin inside the ship, and realize the dredging operation of the target dredger.

[0089] Further, the control module 303 is further configured to:

[0090] If the type of the dredging operation mode is the mud blowing operation mode, control the high-pressure water jet pump to blow the soil in the target mud cabin into a mud mixture according to the optimal parameter combination;

[0091] When it is monitored that the target dredger travels to the target location, control the in-cabin pump to discharge the mud mixture according to the optimal parameter combination, and realize the mud blowing operation of the target dredger.

[0092] Embodiment IV

[0093] Figure 4 It is a schematic structural diagram of the electronic device provided in Embodiment IV of the present invention. Figure 4 A schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0094] As Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory communicatively connected to the at least one processor 11, such as read-only memory (ROM) 12, random access memory (RAM) 13, etc. The memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0095] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0096] The processor 11 can be various general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the intelligent pre-dredging operation parameter recommendation method based on a deep learning network model.

[0097] In some embodiments, the intelligent pre-dredging operation parameter recommendation method based on a deep learning network model can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the intelligent pre-dredging operation parameter recommendation method based on a deep learning network model described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the intelligent pre-dredging operation parameter recommendation method based on a deep learning network model in any other appropriate manner (e.g., by means of firmware).

[0098] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), system on a chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0099] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine, or entirely on the remote machine or server.

[0100] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronics, magnetic, optical, electromagnetic, infrared, semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0101] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0102] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0103] The computing system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The relationship between the client and the server is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to address the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0104] In one embodiment, the embodiment of the present invention further includes a computer program product, the computer program product includes a computer program, and when the computer program is executed by a processor, it implements the intelligent pre-dredging operation parameter recommendation method based on a deep learning network model in any embodiment of the present invention.

[0105] In the process of implementing the computer program product, computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages and also conventional procedural programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0106] It should be understood that the various forms of the flow shown above may be used, steps may be reordered, added or deleted. For example, the steps recited in the present invention may be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0107] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent pre-dredging operation parameter recommendation method based on a deep learning network model, characterized in that Including: In response to a dredger control request, filter the original ship data and original management data according to the category of the dredging working condition to obtain construction data and power data, and respectively determine the label information corresponding to the construction data and the power data; Train a target construction parameter recommendation model based on the construction data and its corresponding label information, and train a target power pre-allocation model based on the power data and its corresponding label information; According to the optimization mode of the recommended parameters, perform input parameter combination optimization based on the target construction parameter recommendation model and the target power pre-allocation model to obtain the optimal parameter combination, and control the operation of the target dredger according to the optimal parameter combination and the dredging working condition.

2. The method according to claim 1, wherein Filter the original ship data and original management data to obtain construction data and power data, and respectively determine the label information corresponding to the construction data and the power data, including: Filter the intelligent dredging control system and the original ship data collected in a preset historical time period according to the category of the dredging working condition to determine the construction data, and determine the instantaneous output and fuel consumption per 10,000 cubic meters of soil corresponding to the construction data as the label information corresponding to the construction data; Filter the original management data collected by the power management system in a preset historical time period according to the category of the dredging working condition to determine the power data, and determine the instantaneous output and fuel consumption per 10,000 cubic meters of soil corresponding to the power data as the label information corresponding to the power data.

3. The method according to claim 1, wherein According to the optimization mode of the recommended parameters, perform input parameter combination optimization based on the target construction parameter recommendation model and the target power pre-allocation model to obtain the optimal parameter combination, including: Construct an objective function corresponding to the optimization mode according to the optimization mode of the recommended parameters, the construction model weight, and the power model weight; Optimize the input parameter combination of the target construction parameter recommendation model and the target power pre-allocation model according to the objective function and the preset optimization algorithm to obtain the optimal parameter combination that meets the optimization mode.

4. The method according to claim 2, characterized in that Determine the construction data and the power data, including: If the category of the dredging working condition is the dredging condition, determine the mud flow rate, mud concentration, dredging speed, high-pressure water jet pump pressure, and overflow cylinder height in a preset historical time period as the construction data, and determine the underwater pump power, in-cabin pump power, and high-pressure water jet pump power as the power data; If the category of the dredging working condition is the mud blowing condition, determine the mud flow rate, mud concentration, high-pressure water jet pump pressure, and mud pump speed in a preset historical time period as the construction data, and determine the in-cabin pump power and high-pressure water jet pump power as the power data.

5. The method according to claim 1, characterized in that Control the operation of the target dredger according to the optimal parameter combination and the dredging working condition, including: If the category of the dredging working condition is the dredging condition, monitor the dredging depth of the target dredger during the navigation of the target dredger; If it is detected that the dredging depth is less than the preset depth threshold, control the underwater pump and the toothed rake head to perform dredging operations according to the optimal parameter combination to suck the sediment at the bottom into the target mud tank inside the ship, and realize the dredging operation of the target dredger. If it is detected that the dredging depth is greater than or equal to the preset depth threshold, the underwater pump, the in-cabin pump, and the toothed drag head are controlled according to the optimal parameter combination to perform dredging operations, so as to suck the sediment at the bottom of the water into the target mud tank inside the ship, and realize the dredging operation of the target dredger.

6. The method according to claim 1, wherein Controlling the operation of the target dredger according to the optimal parameter combination and the dredging working condition, including: If the category of the dredging working condition is the mud blowing working condition, the high-pressure water flushing pump is controlled according to the optimal parameter combination to blow the soil in the target mud tank into a slurry mixture; When it is detected that the target dredger travels to the target location, the in-cabin pump is controlled according to the optimal parameter combination to discharge the slurry mixture, and the mud blowing operation of the target dredger is realized.

7. An intelligent pre-dredging operation parameter recommendation device based on a deep learning network model, characterized in that, Including: A determination module, configured to respond to a dredger control request, screen the original ship data and the original management data according to the category of the dredging working condition, so as to obtain construction data and power data, and respectively determine the label information corresponding to the construction data and the power data; A training module, configured to train a target construction parameter recommendation model according to the construction data and its corresponding label information, and train a target power pre-allocation model according to the power data and its corresponding label information; A control module, configured to perform input parameter combination optimization based on the target construction parameter recommendation model and the target power pre-allocation model according to the optimization mode of the recommended parameters, obtain the optimal parameter combination, and control the operation of the target dredger according to the optimal parameter combination and the dredging working condition.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute the intelligent pre-dredging operation parameter recommendation method based on the deep learning network model according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the intelligent pre-dredging operation parameter recommendation method based on the deep learning network model according to any one of claims 1-6 when executed by a processor.

10. A computer program product, characterized in that, The computer program product includes a computer program, and the computer program implements the intelligent pre-dredging operation parameter recommendation method based on the deep learning network model according to any one of claims 1-6 when executed by a processor.

Images