Deep-sea mining accurate feeding method and device based on intelligent monitoring and regulation

Abstract

The invention provides a deep-sea mining precise feeding method and device based on intelligent monitoring and regulation, and the method comprises the following steps: a plurality of feed ports are formed in the bottom of a storage tank, each feed port is connected with a spiral feeder, the screw pitches of the spiral feeders are different, and the discharge ports of the spiral feeders are connected with the same main material conveying pipeline; setting an Actor network to predict action parameters based on the state parameters at the current moment; and executing the action parameter, obtaining a real reward value and a state parameter at the next moment based on an execution result of the action parameter, and stopping prediction of the action parameter when the ore concentration of the outlet of the main material conveying pipeline in the state parameter at the next moment meets a preset standard. According to the scheme, the spiral feeders with different screw pitches are arranged, the multiple sensors are installed at the outlet and the inlet of each spiral feeder, and the feeding precision is accurately controlled based on data of the sensors.

Description

Technical Field

[0001] The present application relates to the technical field of ore transportation, and particularly to a precise feeding method and device for deep-sea mining based on intelligent monitoring and regulation. Background Art

[0002] In the field of deep-sea mining, precise feeding and pulp mixing technologies play a crucial role in the efficient and stable operation of the entire mining operation. However, there are many deficiencies in this technology at present.

[0003] Firstly, from the aspect of feeding accuracy, precise flow control is an important link to ensure the smooth progress of deep-sea mining. The solid content in the pulp must be strictly controlled within ±3% of the target value. If the feeding accuracy cannot be guaranteed and the solid content fluctuates greatly, it is extremely easy to cause the lifting pipe to be blocked, which will not only reduce the mining efficiency, but also may cause equipment failures, increase maintenance costs and safety risks. However, the existing feeding technologies are difficult to meet such high-precision requirements in flow control and are difficult to meet the growing needs of deep-sea mining.

[0004] In terms of the stability of equipment operation, the deep-sea environment is complex and harsh, and the feeding equipment faces severe tests. The strong corrosiveness of seawater will accelerate the corrosion and damage of the equipment, shorten the service life of the equipment, and increase maintenance costs. At the same time, the seabed geological conditions are complex, and the properties of ores (such as particle size distribution, concentration, etc.) vary greatly. This requires the feeding equipment to be able to adjust the operating parameters in a timely manner according to the changes in ore characteristics to ensure stable feeding. However, traditional feeding equipment lacks an effective regulation mechanism when dealing with these complex changes and cannot ensure the stable operation of the equipment under different working conditions. Summary of the Invention

[0005] The embodiments of the present application provide a precise feeding method and device for deep-sea mining based on intelligent monitoring and regulation. By setting screw feeders with different pitch sizes, and installing multiple sensors at the outlets and inlets of each screw feeder, and precisely controlling the feeding accuracy based on the sensor data.

[0006] In the first aspect, the embodiments of the present application provide a precise feeding method for deep-sea mining based on intelligent monitoring and regulation, and the method includes: A plurality of feeding ports are provided at the bottom of the storage tank, each feeding port is connected to a screw feeder, and the pitch of each screw feeder is different. The discharge ports of each screw feeder are respectively connected to the same main conveying pipeline, and a main and auxiliary conveying valve group is arranged at the discharge port of each screw feeder. A makeup water pump is connected to the main conveying pipeline. The main and auxiliary conveying valve group is composed of a main conveying valve and an auxiliary conveying valve connected in parallel, and the flux of the main conveying valve is greater than that of the auxiliary conveying valve; Set up the Actor network and obtain the state parameters at the current moment. The Actor network predicts action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degrees of the main and auxiliary material conveying valve groups, the flow rate of the makeup water pump, and the ore concentration at the outlet of the total material conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degrees of the main and auxiliary material conveying valve groups; Execute the action parameters, and obtain the true reward value and the state parameters at the next moment based on the execution result of the action parameters. Set up the Critic network. The Critic network evaluates the value of the action parameters based on the state parameters at the current moment to obtain the predicted reward value, updates the parameters of the Actor network based on the predicted reward value, calculates the difference between the predicted reward value and the true reward value to obtain the reward difference, and updates the parameters of the Critic network with the reward difference. Use the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment. Stop predicting the action parameters when the ore concentration at the outlet of the total material conveying pipeline in the state parameters at the next moment meets the preset standard.

[0007] In a second aspect, an embodiment of the present application provides a maintenance and detection method based on intelligent monitoring and control, including: Obtain the corrosion state parameters of the corrosion maintenance components of each screw feeder at the current moment; Construct a current network and a target network with the same initial parameters based on the optimization objective. Input the corrosion state at the current moment into the current network to obtain the Q value of the current network. Select and execute a maintenance action based on the Q value of the current network with an ε-greedy strategy. The Q value is the weight of each maintenance action in the corrosion state at the current moment. The maintenance actions include overall maintenance, partial maintenance, and no maintenance; Input the corrosion state at the current moment into the target network at each moment to obtain the Q value of the target network. Calculate the target value based on the Q value of the current network and the Q value of the target network. Define a loss function according to the target value, and update the parameters of the current network based on the result of the loss function. Synchronize the parameters of the current network to the target network every fixed moment.

[0008] In a third aspect, an embodiment of the present application provides a deep-sea mining precise feeding device based on intelligent monitoring and control, including: A setting module is used to set a plurality of feeding ports at the bottom of a storage tank. Each feeding port is connected to a screw feeder, and the pitch of each screw feeder is different. The discharge ports of each screw feeder are respectively connected to the same total conveying pipeline, and a main and auxiliary conveying valve group is set at the discharge port of each screw feeder. A make-up water pump is connected to the total conveying pipeline. The main and auxiliary conveying valve group is composed of a main conveying valve and an auxiliary conveying valve connected in parallel, and the flux of the main conveying valve is greater than that of the auxiliary conveying valve; A prediction module is used to set an Actor network and obtain state parameters at the current moment. The Actor network predicts action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degree of the main and auxiliary conveying valve group, the flow rate of the make-up water pump, and the ore concentration at the outlet of the total conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degree of the main and auxiliary conveying valve group; An iteration module is used to execute the action parameters, obtain a real reward value and state parameters at the next moment based on the execution result of the action parameters, set a Critic network. The Critic network evaluates the value of the action parameters based on the state parameters at the current moment to obtain a predicted reward value, updates the parameters of the Actor network based on the predicted reward value, calculates the difference between the predicted reward value and the real reward value to obtain a reward difference, updates the parameters of the Critic network with the reward difference, and uses the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment. When the ore concentration at the outlet of the total conveying pipeline in the state parameters at the next moment meets the preset standard, the prediction of the action parameters is stopped.

[0009] In a fourth aspect, an embodiment of the present application provides a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute a precise feeding method for deep-sea mining based on intelligent monitoring and regulation or a maintenance and detection method based on intelligent monitoring and regulation.

[0010] In a fifth aspect, an embodiment of the present application provides a readable storage medium. A computer program is stored in the readable storage medium. The computer program includes program codes for controlling a process to execute the process. The process includes a precise feeding method for deep-sea mining based on intelligent monitoring and regulation or a maintenance and detection method based on intelligent monitoring and regulation.

[0011] The main contributions and innovations of the present invention are as follows: In the embodiments of the present application, multiple screw feeders operate in parallel with different pitch parameters, enabling the transportation of ores with different particle size gradations. Through a unique screw phase design, vibrator configuration, real-time monitoring, and dynamic speed regulation, the uniform transportation of pulp particles is ensured, and the fluctuation of the solid content is controlled within ±3% of the target value, guaranteeing the uniformity of solid feeding in the riser pipe and effectively preventing blockages. In the embodiments of the present application, torque and vibration sensors are arranged on the screw shaft to monitor the transportation state in real time. Once the motor torque suddenly increases or the amplitude of the vibrator increases beyond the threshold, the protection mechanism and reverse rotation mechanism are immediately activated, which can effectively prevent and solve blockage problems, reduce downtime, ensure the continuous and stable feeding process, and improve production efficiency. In the embodiments of the present application, the dual network is trained to optimize the decision-making, achieving the balance of pulp flow rate and the stable operation of the system, enhancing the intelligence and accuracy of feeding control, and formulating an anti-corrosion maintenance strategy for the feeding mechanism, comprehensively considering various factors to define the reward function, reasonably planning maintenance measures, minimizing equipment losses, optimizing maintenance intervals, and reducing unplanned downtime.

[0012] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more concise and understandable. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings: Figure 1 is a flowchart of a precise feeding method for deep-sea mining based on intelligent monitoring and regulation according to an embodiment of the present application; Figure 2 is a structural diagram of a deep-sea mining device according to an embodiment of the present application; Among them, 1 is a storage tank, 2 is a screw feeder, 3 is a total material transportation pipeline, 4 is a main and auxiliary material transportation valve group, 401 is the main material transportation valve group, 402 is the auxiliary material transportation valve group, and 5 is a water replenishing pump.

[0014] Figure 3 is a flowchart of a maintenance and detection method based on intelligent monitoring and regulation according to an embodiment of the present application; Figure 4 is a structural block diagram of a precise feeding device for deep-sea mining based on intelligent monitoring and regulation according to an embodiment of the present application; Figure 5 is a schematic hardware structure diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. On the contrary, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0016] It should be noted that: In other embodiments, the steps of the corresponding methods are not necessarily executed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may also be combined into a single step for description in other embodiments.

[0017] Embodiment 1 An embodiment of the present application provides a precise feeding method for deep-sea mining based on intelligent monitoring and regulation. By setting screw feeders with different pitch sizes, and installing multiple sensors at the outlets and inlets of each screw feeder, and precisely controlling the feeding accuracy based on the sensor data. Specifically, referring to Figure 1 , the method includes: A plurality of feed inlets are provided at the bottom of the storage tank 1, each feed inlet is connected to a screw feeder 2, and the pitch of each screw feeder 2 is different. The discharge ports of each screw feeder are respectively connected to the same main conveying pipeline 3, and a main and auxiliary conveying valve group 4 is provided at the discharge port of each screw feeder 2. A makeup water pump 5 is connected to the main conveying pipeline. The main and auxiliary conveying valve group 4 is composed of a main conveying valve 401 and an auxiliary conveying valve 402 connected in parallel, and the flux of the main conveying valve 401 is greater than that of the auxiliary conveying valve 402; Set up an Actor network and obtain the state parameters at the current moment. The Actor network predicts the action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degree of the main and auxiliary conveying valve group, the flow rate of the makeup water pump, and the ore concentration at the outlet of the main conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degree of the main and auxiliary conveying valve group; Execute the action parameters, obtain the true reward value and the state parameters at the next moment based on the execution result of the action parameters, set up the Critic network, which evaluates the value of the action parameters based on the state parameters at the current moment to obtain the predicted reward value, update the parameters of the Actor network based on the predicted reward value, calculate the difference between the predicted reward value and the true reward value to obtain the reward difference, and update the parameters of the Critic network with the reward difference. Use the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment. Stop predicting the action parameters when the ore concentration at the outlet of the total material conveying pipeline 3 in the state parameters at the next moment meets the preset standard.

[0018] In some embodiments, the structure of the deep-sea mining equipment in this solution is as Figure 2 shown. The storage tank 1 of the deep-sea mining equipment is used to collect the ore collected by the mining equipment, and the storage tank 1 is placed horizontally. The feeding port at the bottom of the storage tank 1 is a conical feeding port, and each feeding port is connected to a screw feeder 2, and the feeding port is used to convey the ore in the storage tank into each screw feeder.

[0019] Specifically, an overflow hole 102 is provided on the storage tank 1 in this solution, and the overflow hole 102 is used to discharge the excess seawater during the ore accumulation process. And an ore detection and stirring device is provided in the storage tank 1, and when the detection and stirring device detects that a certain proportion of the ore in the storage tank 1 is stationary, it stirs the ore to activate this part of the ore.

[0020] In some embodiments, the stagger amount of the spiral blades of two adjacent screw feeders 2 is half a pitch, and the spiral phase difference is 2π / n. The advantage is to solve the periodic fluctuation problem in the screw feeder, where n is the total number of screw feeders.

[0021] Specifically, each screw feeder 2 in this solution is connected to the storage tank through a separate feeding port, so as to ensure that the slurry in the screw feeder will not flow into another one.

[0022] Specifically, the inside of the screw feeder in this solution is a single-spiral structure, and the surface of the blade of the screw feeder 2 is coated with laser micro-texture or polytetrafluoroethylene (PTFE) to reduce the adhesion force and prevent the fine particles from sticking and causing jamming problems. A jet system is also installed in the variable pitch section of the main spiral, equipped with high-pressure nozzles (0.5~1MPa), and water is sprayed regularly to remove the particles accumulated at the root of the blade to ensure the smooth operation of the equipment. Through this design, the efficient and stable operation of the screw feeder 2 under different pulp concentrations and particle distribution changes is ensured.

[0023] In some embodiments, the makeup water pump 5 is used to supply water to the total material conveying pipeline 3 to ensure the ore concentration at the outlet of the total material conveying pipeline 3.

[0024] Specifically, in order to more precisely adjust the ore concentration at the outlet of the total material conveying pipeline 3, the water output of the makeup water pump 5 in this solution is set to a fixed value, so as to ensure that the ore concentration at the outlet of the total material conveying pipeline 3 can be adjusted only by adjusting the rotation speed of the screw feeder 2 and the opening degree of the main and auxiliary material conveying valve group 4.

[0025] In some embodiments, the rotation speed of each screw feeder 2, the opening degree of the main and auxiliary material conveying valve group 4, and the flow rate of the makeup water pump 5 are obtained by capturing the parameters of the background control system. The ore concentration at the outlet of each screw feeder 2 is obtained by installing a flow sensor and a concentration and particle size counter at the outlet of each screw feeder 2, and then the ore concentration at the outlet of the total material conveying pipeline 3 is calculated from the ore concentration at the outlet of each screw feeder 2 and the flow rate of the makeup water pump 5.

[0026] In some specific embodiments, ultrasonic signals are sent to the outlet of each screw feeder 2, and the returned ultrasonic signals are received by the flow sensor and the concentration and particle size sensors. For the measurement of sediment particle size and concentration distribution, considering the attenuation effect of different ore particle sizes on ultrasonic waves, the inversion formula for ore concentration and particle size is obtained by combining the optimal regularization algorithm. The formula is as follows:

[0027] Where A is the coefficient matrix, F is the discretized ore size frequency distribution, matrix H is the smoothing matrix, G is the vector composed of the ultrasonic attenuation coefficients at different frequencies obtained from actual measurements, and the parameter γ can be obtained by minimizing the following γ function:

[0028] Where I is the identity matrix, m is the number of frequencies of the ultrasonic waves used and is also the order of the identity matrix.

[0029] In some specific embodiments, since the flow rate of the makeup water pump 5 in this solution is a fixed value, to ensure that the ore concentration at the outlet of the total material conveying pipeline 3 meets the preset standard in this solution, it is only necessary to ensure that the conveying capacity of each screw feeder 2 meets the standard. The formula for the conveying capacity of each screw feeder 2 is expressed as follows:

[0030] Where Q0 is the conveying capacity, ω is the rotation speed, c is the ore concentration, θ is the intelligent valve opening degree, Q is the flow rate of the seawater pump, c 0 is the ore concentration at the outlet of the total material conveying pipeline 3, where the conveying capacity is the weight of the material conveyed per unit time.

[0031] In some embodiments, an Actor network is set up and the weight coefficients of the Actor network are initialized. The state parameters at the current moment are input into the Actor network. The formula for the Actor network to predict the action parameters based on the state parameters at the current moment is as follows:

[0032] Wherein, d t is the action parameter predicted at the current moment t, f() is the Actor network, and δ f is the weight coefficient of the Actor network. γ t is t the random noise at time

[0033] In this solution, the state parameters at the current moment are first converted into a data format that can be processed by the algorithm and then input into the Actor network. The formula is as follows:

[0034] Where ω is the rotational speed, c is the ore concentration, π is the opening of the intelligent valve, Q is the flow rate of the seawater pump, c and 0 is the ore concentration at the outlet of the total conveying pipeline 3.

[0035] Specifically, the Actor network predicts the probability distribution of various actions based on the state parameters at the current moment, and selects the action with the highest probability as the action parameter based on the probability distribution of various action parameters.

[0036] In some embodiments, the action parameters are transmitted to the control system of the corresponding screw feeder 2 in the form of electrical signals, and the control system adjusts the rotational speed of the screw conveyor and the opening of the main and auxiliary feeding valve group 4.

[0037] Furthermore, when adjusting the opening of the main and auxiliary feeding valve group 4 according to the action parameters, if the adjustment value of the opening of the main and auxiliary feeding valve group 4 is greater than or equal to the set threshold, the opening of the main feeding valve 401 is adjusted and the opening of the auxiliary feeding valve 402 remains unchanged. If the adjustment value of the opening of the main and auxiliary feeding valve group 4 is less than the set threshold, the opening of the auxiliary feeding valve 402 is adjusted and the opening of the main feeding valve 401 remains unchanged.

[0038] That is to say, since the throughput of the main feeding valve 401 in this solution is greater than that of the auxiliary feeding valve 402, when the ore concentration at the outlet of the total conveying pipeline 3 changes greatly, this solution will adjust the opening of the main feeding valve 401. If the ore concentration at the outlet of the total conveying pipeline 3 changes slightly, this solution will adjust the opening of the auxiliary feeding valve 402, so as to ensure that under the coordinated operation of the main and auxiliary feeding valves, the main and auxiliary feeding valve group 4 can adapt to more strongly fluctuating working conditions.

[0039] In some embodiments, the true reward value obtained based on the execution result of the action parameter is determined by a reward function, and the formula of the reward function is expressed as follows:

[0040] Wherein, p 1 is the concentration difference reward function, p 2 is the sparse reward function, p 3 is the formalized reward function.

[0041] The concentration difference reward function calculates the reward based on the difference between the actual concentration and the target concentration. The sparse reward function gives a reward only when the target is completed, otherwise the reward is 0. The formalized reward function refines and decomposes the task objective and designs a reward method with multiple intermediate rewards.

[0042] In some embodiments, multiple sets of experience data are obtained and stored in an experience database, and the Actor network and the Critic network are trained with the experience data. Each set of experience data includes the state parameter, action parameter, true reward value at the current moment, and the state parameter at the next moment.

[0043] In some embodiments, the parameters of the Actor network are updated by the gradient descent method, and the parameters of the Critic network are updated by minimizing the reward difference.

[0044] Specifically, in the training of the Actor network, the policy gradient theorem is used to calculate the policy gradient, and the weight parameters of the policy network are updated by the gradient descent method. Its goal is to optimally approximate the expected policy function so that the optimal action parameter can be selected under the given state parameter.

[0045] Specifically, in the training of the Critic network, the Critic network is trained with multiple sets of experience data in the experience database, and the parameters of the Critic network are continuously optimized by minimizing the reward difference during the training process.

[0046] In some specific embodiments, in order to stabilize the training process and improve the convergence speed, a target Actor network and a target Critic network are used for training, so as to avoid the tight coupling between the Actor network and the Critic network, thereby reducing the influence of gradient noise. The formula is expressed as follows:

[0047] Wherein, is the target Actor network, is the target Critic network, is the weight coefficient of the target Actor network, is the weight coefficient of the target Critic network, and λ is the update coefficient.

[0048] In some embodiments, a torque sensor and a vibration sensor are arranged on the spiral shaft of each spiral feeder 2. When the torque value detected by the torque sensor is greater than the torque threshold or the amplitude value detected by the vibration sensor is greater than the amplitude threshold, the rotation speed of the corresponding spiral feeder 2 is reduced to the safe rotation speed. If the torque value is still greater than the torque threshold or the amplitude value is still greater than the amplitude threshold at the safe rotation speed, the spiral shaft is reversed for a first period of time continuously.

[0049] Specifically, due to the complex working environment of the spiral feeder 2, internal blockage is likely to occur. In this solution, the torque sensor and the vibration sensor are used to detect the blockage condition of the spiral feeder 2. When it is detected that the torque value is greater than the torque threshold or the amplitude value is greater than the amplitude threshold, it indicates that the spiral feeder 2 is blocked. At this time, the blockage is first removed by reducing the rotation speed. If the blockage is not removed after reducing the rotation speed, the spiral shaft is controlled to reverse to loosen the blocked ore.

[0050] Furthermore, when the torque value of a spiral feeder 2 is greater than the torque threshold or the amplitude value is greater than the amplitude threshold, it is considered that the spiral feeder 2 is a blocked spiral feeder, and the execution of the stop action parameter in the corresponding spiral feeder 2 is stopped.

[0051] Embodiment 2 A maintenance and detection method for a spiral feeder based on intelligent monitoring and regulation, referring to Figure 3 , includes: Obtain the corrosion state parameters of the corrosion maintenance components of each spiral feeder 2 at the current moment; Construct a current network and a target network with the same initial parameters based on the optimization goal. Input the corrosion state at the current moment into the current network to obtain the Q value of the current network. Select and execute the maintenance action based on the Q value of the current network with the ε-greedy strategy, where the Q value is the weight of each maintenance action under the corrosion state at the current moment, and the maintenance actions include overall maintenance, partial maintenance, and no maintenance; Input the corrosion state at the current moment into the target network at each moment to obtain the Q value of the target network. Calculate the target value based on the Q value of the current network and the Q value of the target network. Define the loss function according to the target value, and update the parameters of the current network based on the result of the loss function. Synchronize the parameters of the current network to the target network every fixed moment.

[0052] In some specific embodiments, the corrosion maintenance components of the screw feeder 2 include a barrel, a screw blade, a valve core, and the inner wall of the pipeline. The corrosion state parameter at the current moment is the remaining wall thickness of the corrosion maintenance component at the current moment divided by the initial wall thickness, which is expressed by the formula:

[0053] Wherein, is the corrosion parameter at the current moment t, is the remaining wall thickness at the current moment t, and D is the initial wall thickness.

[0054] In some embodiments, the Savitzky-Golay wavelet smoothing method is used to perform noise reduction processing on the corrosion pipeline degradation data obtained by online monitoring, so as to obtain the wall thickness of each corrosion maintenance component. A preventive maintenance threshold and a failure threshold are defined for each corrosion maintenance component. When the corrosion state parameter of the corrosion maintenance component exceeds the preventive maintenance threshold, it indicates that the component should be repaired. When the corrosion state parameter of the corrosion maintenance component exceeds the failure threshold, it indicates that the corrosion maintenance component should be replaced.

[0055] In some specific embodiments, the optimization objective is to minimize one of equipment loss, minimum maintenance interval, minimum unplanned downtime, minimum maintenance cost, and minimum energy consumption. The optimization objective is set according to the requirements in the implementation application.

[0056] Specifically, taking minimizing equipment loss as the optimization objective is used to keep the corrosion rate of the device material within the set range. Taking the minimum maintenance interval as the optimization objective is used to avoid the additional costs caused by over-maintenance and under-maintenance. Taking the minimum unplanned downtime as the optimization objective is used to keep the screw feeder 2 running efficiently. Taking the minimum maintenance cost as the optimization objective is used to reduce the costs of anti-corrosion coatings, replacement parts, and maintenance operations. Taking the minimum energy consumption as the optimization objective is used to reduce the increase in flow resistance and energy consumption caused by corrosion.

[0057] Furthermore, based on the optimization objective, the reward function of the current network is defined, and the formula of the reward function is expressed as:

[0058] Wherein, is the corrosion control reward, is the unplanned downtime penalty, is the maintenance cost penalty, is the energy consumption penalty, is the remaining life reward, 、 、 、 、 are the corresponding weight parameters.

[0059] In some specific embodiments, in the ε-greedy strategy, by setting the initial exploration probability and the initial exploration probability decays with the Q-value training process. The formula is expressed as:

[0060] where the initial exploration probability takes a value of 0.5, λ is the decay rate of the exploration probability, and at each moment, with a probability of 1 - the highest value of the current network Q-value is selected, and with a probability of a maintenance action is randomly selected.

[0061] In some embodiments, the actions that can be selected by the current network and the target network in this solution at each moment include overall maintenance, partial maintenance, and no maintenance. The formula is expressed as follows:

[0062] where represents partial maintenance, represents overall maintenance, represents no maintenance.

[0063] In some embodiments, by defining a state transition function to represent the state transition relationship of a certain corrosion maintenance component from time t to time t + 1 after a maintenance action, the formula is expressed as follows:

[0064] where is the remaining wall thickness after maintenance at the previous maintenance moment, is the maintenance action, is the corrosion parameter at the current moment t.

[0065] In some specific embodiments, the Q-value output by the current network is Q( S t , a, θ), and the Q-value output by the target network is ( S t , a, θ'), where is the corrosion state at the current moment, a is the corresponding maintenance action, θ is the weight of each maintenance action in the current network, and θ' is the weight of each maintenance action in the target network. Then, the calculation formula for the target value at the current moment is as follows:

[0066] where is the target value at the current moment t, is the reward value at time t + 1.

[0067] The formula of the loss function defined based on the target value is expressed as follows:

[0068] Wherein, is the loss function, and N is the number of samples.

[0069] That is to say, after obtaining the loss function, the parameters of the current network are updated according to the value of the loss function, so that the maintenance actions selected by the current network are getting better and better. The formula is expressed as follows:

[0070] Wherein, α is the learning rate.

[0071] In some specific embodiments, the formula for synchronizing the parameters of the current network to the target network at fixed intervals is expressed as follows:

[0072] That is to say, in this solution, the ε-greedy strategy is used to utilize the Q value of the target network and the Q value of the current network to improve the prediction accuracy of the current network for maintenance actions.

[0073] Specifically, when the life of the corroded maintenance component is exhausted or the number of unplanned shutdowns exceeds the set threshold, the iteration of the current network ends.

[0074] In some specific embodiments, the Q value of the current network at each moment and the actually obtained corrosion state at the next moment are combined into a state-action pair and input into the experience pool, so as to train the current network and the target network through the data in the experience pool.

[0075] Embodiment III Based on the same concept, referring to Figure 4 , the present application also proposes an intelligent monitoring and control-based precise feeding device for deep-sea mining, including: A setting module for providing a plurality of feeding ports at the bottom of the storage tank 1, each feeding port is connected to a screw feeder 2, and the pitch of each screw feeder 2 is different. The discharge ports of each screw feeder are respectively connected to the same main conveying pipeline 3, and a main and auxiliary conveying valve group 4 is provided at the discharge port of each screw feeder 2. A make-up water pump 5 is connected to the main conveying pipeline. The main and auxiliary conveying valve group 4 is composed of a main conveying valve 401 and an auxiliary conveying valve 402 connected in parallel, and the flux of the main conveying valve 401 is greater than that of the auxiliary conveying valve 402; A prediction module, configured to set up an Actor network and obtain state parameters at the current moment. The Actor network predicts action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degrees of the main and auxiliary material conveying valve groups, the flow rate of the makeup water pump, and the ore concentration at the outlet of the total material conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degrees of the main and auxiliary material conveying valve groups. An iteration module, configured to execute the action parameters, obtain a true reward value and state parameters at the next moment based on the execution result of the action parameters, set up a Critic network. The Critic network evaluates the value of the action parameters based on the state parameters at the current moment to obtain a predicted reward value, updates the parameters of the Actor network based on the predicted reward value, calculates the difference between the predicted reward value and the true reward value to obtain a reward difference value, updates the parameters of the Critic network with the reward difference value, uses the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment, and stops predicting the action parameters when the ore concentration at the outlet of the total material conveying pipeline 3 in the state parameters at the next moment meets a preset standard.

[0076] Embodiment 4 This embodiment also provides an electronic device, referring to Figure 5 , including a memory 404 and a processor 402. A computer program is stored in the memory 404, and the processor 402 is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0077] Specifically, the above processor 402 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured as one or more integrated circuits implementing the embodiments of the present application.

[0078] Among them, the memory 404 may include a mass storage 404 for data or instructions. By way of example and not limitation, the memory 404 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In a suitable case, the memory 404 may include a removable or non-removable (or fixed) medium. In a suitable case, the memory 404 may be inside or outside the data processing device. In a specific embodiment, the memory 404 is a non-volatile memory. In a specific embodiment, the memory 404 includes a read-only memory (ROM) and a random access memory (RAM). In a suitable case, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or a flash memory, or a combination of two or more of these. In a suitable case, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM may be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0079] The memory 404 can be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 402.

[0080] The processor 402 reads and executes the computer program instructions stored in the memory 404 to implement any one of the deep-sea mining precise feeding methods based on intelligent monitoring and regulation in the above embodiments.

[0081] Optionally, the above electronic device may further include a transmission device 406 and an input / output device 408. Among them, the transmission device 406 is connected to the above processor 402, and the input / output device 408 is connected to the above processor 402.

[0082] The transmission device 406 can be used to receive or send data via a network. Specific examples of the above network may include wired or wireless networks provided by the communication provider of the electronic device. In one example, the transmission device includes a network adapter (abbreviated as NIC), which can be connected to other network devices through a base station and thus communicate with the Internet. In one example, the transmission device 406 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0083] The input / output device 408 is used to input or output information. In this embodiment, the input information can be the state parameters at the current moment, etc., and the output information can be the reward value, action parameters, etc.

[0084] Optionally, in this embodiment, the above processor 402 can be set to execute the following steps through a computer program: A plurality of feed inlets are provided at the bottom of the storage tank 1, each feed inlet is connected to a screw feeder 2, and the pitch of each screw feeder 2 is different. The discharge ports of each screw feeder 2 are respectively connected to the same main material conveying pipeline 3, and a main and auxiliary material conveying valve group 4 is provided at the discharge port of each screw feeder 2. A water replenishing pump 5 is connected to the main material conveying pipeline. The main and auxiliary material conveying valve group 4 is composed of a main material conveying valve 401 and an auxiliary material conveying valve 402 connected in parallel, and the flux of the main material conveying valve 401 is greater than that of the auxiliary material conveying valve 402; Set up an Actor network and obtain the state parameters at the current moment. The Actor network predicts the action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degree of the main and auxiliary material conveying valve group, the flow rate of the water replenishing pump, and the ore concentration at the outlet of the main material conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degree of the main and auxiliary material conveying valve group; Execute the action parameters, obtain the true reward value and the state parameters at the next moment based on the execution result of the action parameters, set up the Critic network, the Critic network evaluates the value of the action parameters based on the state parameters at the current moment to obtain the predicted reward value, update the parameters of the Actor network based on the predicted reward value, calculate the difference between the predicted reward value and the true reward value to obtain the reward difference, update the parameters of the Critic network with the reward difference, use the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment, and stop predicting the action parameters when the ore concentration at the outlet of the total feeding pipeline 3 in the state parameters at the next moment meets the preset standard.

[0085] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation manners, and will not be elaborated here.

[0086] Generally, various embodiments can be implemented in hardware or dedicated circuits, software, logic, or any combination thereof. Some aspects of the present invention can be implemented in hardware, while other aspects can be implemented by firmware or software executed by a controller, a microprocessor, or other computing devices, but the present invention is not limited thereto. Although the various aspects of the present invention can be shown and described as block diagrams, flowcharts, or using some other graphical representation, it should be understood that, as a non-limiting example, the blocks, devices, systems, techniques, or methods described herein can be implemented in hardware, software, firmware, dedicated circuits or logic, general hardware or a controller, or other computing devices, or some combination thereof.

[0087] Embodiments of the present invention can be implemented by computer software, which can be executed by a data processor of a mobile device, such as in a processor entity, or implemented by hardware, or implemented by a combination of software and hardware. A computer software or program (also referred to as a program product), including software routines, applets, and / or macros, can be stored in any device-readable data storage medium, and they include program instructions for performing specific tasks. The computer program product can include one or more computer-executable components configured to execute the embodiments when the program runs. One or more computer-executable components can be at least one software code or a part thereof. Additionally, in this regard, it should be noted that any block in the logical flow, as Figure 5 shown, can represent a program step, or an interconnected logical circuit, block, and function, or a combination of a program step and a logical circuit, block, and function. The software can be stored on physical media such as memory chips or storage blocks implemented within a processor, magnetic media such as hard disks or floppy disks, and optical media such as, for example, DVDs and their data variants, CDs. The physical media is a non-transitory medium.

[0088] Those skilled in the art should understand that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered that the scope described in this specification is covered.

[0089] The above embodiments only express several implementation manners of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A precise feeding method for deep - sea mining based on intelligent monitoring and regulation, characterized in that, It includes the following steps: There are multiple feed inlets at the bottom of the storage tank (1). Each feed inlet is connected to a screw feeder (2), and the pitch of each screw feeder (2) is different. The discharge ports of each screw feeder are respectively connected to the same total material conveying pipeline (3), and a main and auxiliary material conveying valve group (4) is arranged at the discharge port of each screw feeder (2). A water replenishing pump (5) is connected to the total material conveying pipeline. The main and auxiliary material conveying valve group (4) is composed of a main material conveying valve (401) and an auxiliary material conveying valve (402) connected in parallel, and the flux of the main material conveying valve (401) is greater than that of the auxiliary material conveying valve (402); Set up an Actor network and obtain the state parameters at the current moment. The Actor network predicts the action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degree of the main and auxiliary material conveying valve group, the flow rate of the water replenishing pump, and the ore concentration at the outlet of the total material conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degree of the main and auxiliary material conveying valve group; Execute the action parameters, and obtain the real reward value and the state parameters at the next moment based on the execution result of the action parameters. Set up a Critic network. The Critic network evaluates the value of the action parameters based on the state parameters at the current moment to obtain the predicted reward value, updates the parameters of the Actor network based on the predicted reward value, calculates the difference between the predicted reward value and the real reward value to obtain the reward difference, updates the parameters of the Critic network with the reward difference, and uses the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment. When the ore concentration at the outlet of the total material conveying pipeline (3) in the state parameters at the next moment meets the preset standard, stop predicting the action parameters.

2. The precise feeding method for deep - sea mining based on intelligent monitoring and regulation according to claim 1, wherein, The dislocation amount of the spiral blades of two adjacent screw feeders (2) is half a pitch, and the spiral phase difference is 2π / n, where n is the total number of screw feeders.

3. The precise feeding method for deep - sea mining based on intelligent monitoring and regulation according to claim 1, wherein The rotation speed of each screw feeder (2), the opening degree of the main and auxiliary material conveying valve group (4), and the flow rate of the water replenishing pump (5) are obtained by capturing the parameters of the background control system. The ore concentration at the discharge port of each screw feeder (2) is obtained by installing a flow sensor and a concentration and particle size counter at the discharge port of each screw feeder (2), and the ore concentration at the outlet of the total material conveying pipeline (3) is calculated from the ore concentration at the discharge port of each screw feeder (2) and the flow rate of the water replenishing pump (5).

4. A precise feeding method for deep - sea mining based on intelligent monitoring and regulation according to claim 1, characterized in that, Set up an Actor network and initialize the weight coefficients of the Actor network. The formula for the Actor network to predict the action parameters based on the state parameters at the current moment is as follows: ； Among them, d t is the action parameter predicted at the current moment t, γ t is t the random noise at the moment.

5. A precise feeding method for deep - sea mining based on intelligent monitoring and regulation according to claim 1, characterized in that, When adjusting the opening degree of the main and auxiliary material feeding valve group (4) according to the action parameters, if the adjusted value of the opening degree of the main and auxiliary material feeding valve group (4) is greater than or equal to the set threshold, the opening degree of the main material feeding valve (401) is adjusted, and the opening degree of the auxiliary material feeding valve (402) remains unchanged. If the adjusted value of the opening degree of the main and auxiliary material feeding valve group (4) is less than the set threshold, the opening degree of the auxiliary material feeding valve (402) is adjusted, and the opening degree of the main material feeding valve (401) remains unchanged.

6. The precise feeding method for deep - sea mining based on intelligent monitoring and regulation according to claim 1, wherein, A torque sensor and a vibration sensor are arranged on the screw shaft of each screw feeder (2). When the torque value detected by the torque sensor is greater than the torque threshold or the amplitude value detected by the vibration sensor is greater than the amplitude threshold, the rotation speed of the corresponding screw feeder (2) is reduced to the safe rotation speed. If the torque value is still greater than the torque threshold or the amplitude value is still greater than the amplitude threshold at the safe rotation speed, the screw shaft is reversed continuously for the first period of time.

7. A maintenance detection method based on intelligent monitoring and regulation, characterized in that, Including: Obtaining the corrosion state parameters of the corrosion repair components of each screw feeder (2) at the current moment; Based on the optimization objective, constructing a current network and a target network with the same initial parameters. Inputting the corrosion state at the current moment into the current network to obtain the Q value of the current network. Selecting and executing a maintenance action based on the Q value of the current network with an ε-greedy strategy, where the Q value is the weight of each maintenance action under the corrosion state at the current moment, and the maintenance actions include overall maintenance, partial maintenance, and no maintenance; At each moment, inputting the corrosion state at the current moment into the target network to obtain the Q value of the target network. Calculating the target value based on the Q value of the current network and the Q value of the target network. Defining a loss function according to the target value, and updating the parameters of the current network based on the result of the loss function. Synchronizing the parameters of the current network to the target network every fixed moment.

8. An accurate feeding device for deep-sea mining based on intelligent monitoring and regulation, characterized in that, Including: A setting module for arranging a plurality of feeding ports at the bottom of the storage tank (1). Each feeding port is connected to a screw feeder (2), and the pitch of each screw feeder (2) is different. The discharge ports of each screw feeder are respectively connected to the same total material conveying pipeline (3), and a main and auxiliary material feeding valve group (4) is arranged at the discharge port of each screw feeder (2). A water replenishing pump (5) is connected to the total material conveying pipeline. The main and auxiliary material feeding valve group (4) is composed of a main material feeding valve (401) and an auxiliary material feeding valve (402) connected in parallel, and the throughput of the main material feeding valve (401) is greater than that of the auxiliary material feeding valve (402); A prediction module for setting an Actor network and obtaining the state parameters at the current moment. The Actor network predicts the action parameters based on the state parameters at the current moment. The state parameters include the rotation speed of each screw feeder, the ore concentration at the discharge port, the opening degree of the main and auxiliary material feeding valve group, the flow rate of the water replenishing pump, and the ore concentration at the outlet of the total material conveying pipeline. The action parameters include the rotation speed of each screw conveyor and the opening degree of the main and auxiliary material feeding valve group; An iterative module is used to execute action parameters, obtain a true reward value and state parameters at the next moment based on the execution result of the action parameters, set a Critic network, and the Critic network evaluates the value of the action parameters based on the state parameters at the current moment to obtain a predicted reward value. Update the parameters of the Actor network based on the predicted reward value, calculate the difference between the predicted reward value and the true reward value to obtain a reward difference, and update the parameters of the Critic network with the reward difference. Use the Actor network and the Critic network with updated parameters to predict the action parameters at the next moment. When the ore concentration at the outlet of the total feeding pipeline (3) in the state parameters at the next moment meets the preset standard, stop predicting the action parameters.

9. An electronic device, comprising a memory and a processor, characterized in that, A computer program is stored in the memory, and the processor is configured to run the computer program to execute a precise feeding method for deep-sea mining based on intelligent monitoring and regulation according to any one of claims 1-6 or a maintenance detection method based on intelligent monitoring and regulation according to claim 7.

10. A readable storage medium, characterized in that, A computer program is stored in the readable storage medium, and the computer program includes program codes for controlling a process to execute the process, and the process includes a precise feeding method for deep-sea mining based on intelligent monitoring and regulation according to any one of claims 1-6 or a maintenance detection method based on intelligent monitoring and regulation according to claim 7.

Images