Cooperative control method for filtration and ultraviolet sterilization links in pure physical ballast water management system

Abstract

The invention discloses a cooperative control method for filtering and ultraviolet sterilization links in a pure physical ballast water management system. According to the method, based on real-time monitoring of water flow characteristics, filter screen states, water quality parameters and ultraviolet sterilization effects, efficient operation of a ballast water treatment system is achieved through intelligent feedback regulation and control, machine learning optimization, self-adaptive flow velocity adjustment and dynamic energy consumption optimization. According to the method, the ballast water treatment efficiency can be improved by 20%, the energy consumption can be reduced by 15%, meanwhile, the stability and adaptability of the system are enhanced, and the method is suitable for ballast water management of various ships and meets the international environmental protection standard.

Description

Technical Field

[0001] The present invention relates to the technical field of ballast water treatment technology, and in particular to a coordinated control method of filtration and ultraviolet sterilization links in a pure physical ballast water management system. Background Art

[0002] At present, ballast water treatment technologies are mainly divided into physical methods, chemical methods and biological methods. Among them, physical methods have become the focus of research and application due to their characteristics of no chemical residues, environmental protection and safety. Physical methods mainly include mechanical filtration, ultraviolet sterilization, high-pressure sterilization and ultrasonic treatment, among which the combination of mechanical filtration and ultraviolet sterilization is the most widely used. Mechanical filtration can remove larger particles of suspended matter and organisms, while ultraviolet sterilization destroys microbial DNA through high-intensity ultraviolet rays to achieve an inactivation effect. Although these two methods have relatively mature technologies in practical applications, their independent control modes often lead to reduced overall system operating efficiency. For example, the degree of blockage of the filtration system directly affects the treatment efficiency of the ultraviolet sterilization unit, and the working state of the ultraviolet sterilization unit also affects the maintenance requirements of the filtration system. However, the existing technology has little research on the coordinated control of the two, and the following problems still exist in the operation of the system:

[0003] 1. The filtration system and the UV sterilization unit fail to work together effectively. Traditional systems often operate the filtration and UV sterilization units as two independent modules, and fail to dynamically adjust the power and flow rate of UV sterilization according to changes in filtration effects, resulting in increased energy consumption or incomplete sterilization.

[0004] 2. Blockage of the filtration system affects the sterilization effect. After the filtration unit has been running for a long time, the filter element may become blocked, resulting in a decrease in the water flow rate. However, the UV sterilization unit is usually designed according to a fixed flow rate. Changes in flow rate may lead to insufficient sterilization time or excessive sterilization, affecting the overall stability of the system.

[0005] 3. High energy consumption and maintenance costs. Due to the lack of effective coordinated control, the system often requires high-intensity UV irradiation to compensate for the instability of the filtration system, which shortens the life of the UV lamp and increases maintenance and replacement costs. Summary of the invention

[0006] In view of this, the purpose of the present invention is to provide a coordinated control method for the filtration and ultraviolet sterilization links in a purely physical ballast water management system. Through an intelligent monitoring and feedback mechanism, adaptive adjustment of the filtration and ultraviolet sterilization units is achieved, thereby improving sterilization efficiency, reducing energy consumption, and reducing equipment maintenance costs, ensuring that the system can operate stably and efficiently under different working conditions.

[0007] To achieve the above object, the present invention adopts the following technical solutions: A collaborative control method for the filtration and ultraviolet sterilization links in a pure physical ballast water management system, comprising the following steps:

[0008] Step 1: Real-time monitor the suspended particle concentration, turbidity, ultraviolet transmittance and water flow velocity data of the ballast water through sensors, and establish a water quality analysis model;

[0009] Step 2: Adopt an intelligent feedback control mechanism to dynamically adjust the filtration accuracy, backwashing frequency and ultraviolet dose requirements according to the real-time water quality data;

[0010] Step 3: Adopt an optimization model based on machine learning, use support vector regression and long short-term memory network to predict the optimal filtration accuracy and ultraviolet sterilization parameters, and optimize the processing strategy according to historical data;

[0011] Step 4: Adopt an adaptive flow rate adjustment mechanism, combine quantitative control PID control, fuzzy control and genetic algorithm to optimize the flow control valve and variable frequency pump, so that the water flow rate is maintained within the best working range of ultraviolet sterilization;

[0012] Step 5: Adopt a dynamic energy consumption optimization strategy, combine the water quality monitoring data, adjust the ultraviolet lamp power and filtration energy consumption, and reduce the overall energy consumption while ensuring the treatment effect.

[0013] In a preferred embodiment, the specific steps of Step 2 include:

[0014] Step 21: Real-time water quality monitoring and data collection:

[0015] Use a turbidity sensor to real-time monitor the turbidity of the ballast water, and use a particulate matter concentration sensor to real-time monitor the suspended particle concentration of the ballast water;

[0016] Use an ultraviolet transmittance sensor to measure the ultraviolet light transmittance and calculate the ultraviolet light dose: Wherein, is the ultraviolet light dose, is the ultraviolet light intensity, is the irradiation time;

[0017] Use a flow velocity sensor to detect the water flow velocity , and ensure that the residence time of the water flow in the ultraviolet sterilization unit meets the following conditions: Wherein, is the effective length of the ultraviolet sterilization unit;

[0018] Step 22: Dynamically adjust the filtration parameters:

[0019] When the suspended particle concentration is higher than the threshold , then automatically increase the filtration accuracy Level or increase backwash frequency ;

[0020] When the filtered turbidity is still higher than the set value, reduce the water flow rate to improve the filtration effect;

[0021] Step 23: UV sterilization power adjustment

[0022] Calculate the UV dose requirement where, is an empirical coefficient, adjusted according to different water qualities;

[0023] According to the UV dose requirement Adjust the UV lamp power where, is the maximum UV power, is the target UV dose;

[0024] Step 24: Flow rate adaptive control

[0025] Calculate the optimal flow rate range where, is the optimal UV sterilization residence time;

[0026] Automatically adjust the water flow rate through a variable frequency pump and a flow control valve to match the optimal UV sterilization residence time; Step 25: Abnormal handling mechanism When the filtered turbidity continues to be higher than the turbidity set value , then alarm and adjust the filtration mode;

[0027] When the biological survival rate of the UV sterilization unit exceeds the survival rate set value , then increase the UV lamp power or reduce the flow rate.

[0028] In a preferred embodiment, the specific steps of step 3 include:

[0029] Step 31: Data collection and feature engineering

[0030] Collect water quality data before and after filtration, including the concentration of suspended particles before filtration , the concentration of suspended particles after filtration , the turbidity NTU before filtration, the turbidity after filtration , the UV transmittance (UVT) after filtration, the water flow rate , the UV lamp power and the sterilization rate after filtration ;

[0031] Dimensionality reduction is performed using principal component analysis (PCA) to extract the key influencing factors : , where is the dimensionality reduction transformation matrix, is the original data feature matrix;

[0032] Step 32: Model training and optimization

[0033] Support vector regression (SVR) is used to predict the optimal filtration accuracy and the ultraviolet dose requirement ;

[0034] Training objective: , where is the sterilization rate predicted by the model, is the sterilization after filtration; N is the summary of historical data samples, and i is the sample index;

[0035] Cross-validation is used to improve the generalization ability of the model;

[0036] Step 33: Real-time prediction and dynamic adjustment:

[0037] Calculate the current water quality state vector :

[0038] The long short-term memory (LSTM) network is used to predict the optimal filtration accuracy at the next moment and the optimal ultraviolet dose requirement at the next moment : , where f is the mapping function of the LSTM network, are the model parameters; the optimal filtration accuracy at the next moment is applied to the filtration module to adjust the accuracy; the optimal ultraviolet dose requirement at the next moment is used to adjust the ultraviolet lamp power ;

[0039] Step 34: Optimization of the control strategy using reinforcement learning:

[0040] Deep Q-network (DQN) is used to optimize the filtration and ultraviolet sterilization strategies;

[0041] Design of the reward function : , where, is the total energy consumption of the ballast water management system, is the weight coefficient; during the training process, the agent learns the optimal control strategy through trial and error to maximize the sterilization rate and minimize the energy consumption;

[0042] Step 35: Deployment and adaptive optimization of the ballast water management system:

[0043] Continuously update the machine learning model through online learning to adapt to different water quality environments;

[0044] Combine the historical ballast water data of the ship to establish a personalized optimization model and improve the adaptability of different shipping routes.

[0045] In a preferred embodiment, step 4 specifically includes:

[0046] Step 41: Flow velocity monitoring and data collection;

[0047] Step 42: Adaptive flow velocity calculation:

[0048] Optimal flow velocity Calculate: ; Represents the optimal UV sterilization residence time;

[0049] When the residence time in the UV sterilization unit Increase the water flow velocity To avoid energy consumption waste caused by over-irradiation;

[0050] When the residence time in the UV sterilization unit Reduce the water flow velocity To ensure the sterilization effect;

[0051] Step 43: Flow velocity adjustment control strategy:

[0052] Adopt Fuzzy Control to achieve adaptive adjustment;

[0053] Set the fuzzy control input variables: the current flow velocity , the UV transmittance (UVT) after filtration and the turbidity (NTU) before filtration; set the control rules:

[0054] If the UV transmittance (UVT) after filtration is low and the turbidity (NTU) before filtration is high, reduce the water flow velocity ;

[0055] If the UV transmittance (UVT) after filtration is high and the turbidity (NTU) before filtration is low, increase the water flow velocity ;

[0056] Adopt fuzzy inference to calculate the adjustment amplitude Among them, Is the target flow velocity;

[0057] Step 44: Closed-loop control and optimization

[0058] Set the error feedback mechanism: Among them, Represents the flow velocity at the current moment, are PID control parameters; if the error is too large, the controller will adaptively and quickly adjust the flow rate; if the error is small, it will be adjusted slowly to ensure the stability of the flow rate;

[0059] Step 45: Optimization and dynamic adjustment of the ballast water management system.

[0060] In a preferred embodiment, the specific steps of step 5 include:

[0061] Step 51: Energy consumption modeling and objective function:

[0062] The total energy consumption of the ballast water management system mainly consists of the energy consumption of the filtration unit and the energy consumption of the ultraviolet disinfection unit : ;

[0063] The energy consumption of the filtration unit is affected by the water flow rate and the backwashing frequency : where, is the flow rate correlation coefficient, is the backwashing energy consumption coefficient;

[0064] The energy consumption of the ultraviolet disinfection unit is affected by the ultraviolet lamp power and the ultraviolet transmittance UVT: ;

[0065] The goal is to minimize the total energy consumption of the ballast water management system while meeting the sterilization rate after filtration constraint: s.t. where, represents the minimum sterilization rate;

[0066] Step 52: Energy consumption optimization strategy based on reinforcement learning:

[0067] Use deep reinforcement learning DQN to optimize the collaborative energy consumption control strategy of filtration and ultraviolet disinfection;

[0068] State variables : water quality parameters (NTU, UVT, C), current flow rate and ultraviolet lamp power ;

[0069] Action space : adjust V, F and ;

[0070] Reward function : , where is the penalty factor;

[0071] During the training process, the agent learns the optimal policy based on historical data and continuously updates it online;

[0072] Step 53: Dynamically adjust the UV lamp power:

[0073] Calculate the optimal UV lamp power ;

[0074] When the UV transmittance UVT is high, reduce the UV lamp power to reduce energy consumption;

[0075] When the sterilization rate after filtration decreases, increase the UV lamp power to ensure the sterilization effect;

[0076] Step 54: Cooperative optimization strategy:

[0077] Optimize the parameter combination through the genetic algorithm GA;

[0078] Fitness function: ;

[0079] Adopt selection, crossover, and mutation operations to search for the optimal energy consumption solution;

[0080] Step 55: Online adaptive optimization of the system:

[0081] Combine reinforcement learning and genetic algorithm to dynamically update the energy consumption optimization strategy;

[0082] Adopt predictive maintenance to prevent the performance degradation caused by the long-term operation of the system.

[0083] Compared with the prior art, the present invention has the following beneficial effects:

[0084] Through the cooperative control of the filtration and UV sterilization links, the present invention realizes the intelligentization, energy saving, and high efficiency of the ballast water treatment system, can meet the emission requirements of international conventions, and improve the overall level of ship ballast water management. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 is a schematic flowchart of a preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0086] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0087] It should be noted that the following detailed description is illustrative and aims to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs.

[0088] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0089] A collaborative control method for the filtration and ultraviolet sterilization links in a pure physical ballast water management system, refer to Figure 1 , including the following steps:

[0090] Step 1. System initialization and data acquisition

[0091] (1) Start the ballast water management system and perform self-check to ensure that the filtration unit, ultraviolet sterilization unit, sensor network, and control system are all in normal working condition.

[0092] (2) The water quality sensor starts to collect real-time raw water quality data, including suspended particle concentration , turbidity (NTU), ultraviolet transmittance (UVT), and water flow rate .

[0093] (3) The system sets the target water quality standard ( target value, minimum sterilization rate requirement), and calculates the initial filtration accuracy , ultraviolet dose requirement and flow rate .

[0094] Step 2. Intelligent feedback regulation

[0095] (1) Monitor the water quality parameters before and after filtration and calculate the filtration efficiency: .

[0096] (2) If is lower than the set threshold, automatically adjust the filtration accuracy or increase the backwashing frequency .

[0097] (3) Through the feedback of ultraviolet transmittance (UVT) and biological survival rate , dynamically adjust the ultraviolet dose:

[0098] Step 3. Optimization Model Based on Machine Learning

[0099] (1) Support Vector Regression (SVR) is used to predict the optimal filtration accuracy and UV dose.

[0100] (2) Calculate the error and optimize the adjustment parameters to achieve the target sterilization rate:

[0101] .

[0102] (3) Long Short-Term Memory (LSTM) network is used for water quality prediction to adjust the filtration and UV sterilization strategies in advance.

[0103] Step 4. Adaptive Flow Rate Regulation

[0104] (1) Calculate the current residence time: .

[0105] (2) Ensure that is near the optimal sterilization time, and adjust the flow rate

[0106] (3) Introduce a fuzzy control strategy to adjust the flow rate through fuzzy logic decision-making to improve the regulation accuracy.

[0107] (4) Use genetic algorithm to optimize the PID control parameters to make the flow rate regulation more accurate and efficient.

[0108] Step 5. Dynamic Energy Consumption Optimization

[0109] (1) Use reinforcement learning to optimize the collaborative energy consumption regulation of filtration and UV sterilization.

[0110] (2) Calculate the total system energy consumption:

[0111] (3) Objective function: s.t.

[0112] (4) Optimize combination through genetic algorithm (GA) to improve the system energy efficiency.

[0113] Step 6. System Fault Warning and Maintenance

[0114] (1) Set the abnormal detection threshold and alarm when the water quality parameters exceed the range.

[0115] (2) Adopt remote monitoring and data analysis to predictively maintain key components and reduce unplanned downtime.

[0116] ​Through the collaborative control of the filtration and ultraviolet sterilization processes, the present invention achieves the intelligentization, energy conservation, and high efficiency of the ballast water treatment system, can meet the discharge requirements of international conventions, and improves the overall level of ship ballast water management.

Claims

1. A collaborative control method for the filtration and ultraviolet sterilization links in a pure physical ballast water management system, characterized in that, It includes the following steps: Step 1: Monitor the suspended particle concentration, turbidity, ultraviolet transmittance and water flow velocity data of the ballast water in real time through sensors, and establish a water quality analysis model; Step 2: Adopt an intelligent feedback control mechanism to dynamically adjust the filtration accuracy, backwashing frequency and ultraviolet dose requirements according to the real-time water quality data; Step 3: Adopt an optimization model based on machine learning, use support vector regression and long short-term memory network to predict the optimal filtration accuracy and ultraviolet sterilization parameters, and optimize the treatment strategy according to historical data; Step 4: Adopt an adaptive flow rate adjustment mechanism, combine quantitative control PID control, fuzzy control and genetic algorithm to optimize the flow control valve and variable frequency pump, so that the water flow rate is maintained within the optimal working range of ultraviolet sterilization; Step 5: Adopt a dynamic energy consumption optimization strategy, combine water quality monitoring data, adjust the power of ultraviolet lamps and filtration energy consumption, and reduce the overall energy consumption while ensuring the treatment effect.

2. The collaborative control method for the filtration and ultraviolet sterilization links in a pure physical ballast water management system according to claim 1, characterized in that The specific content of Step 2 includes: Step 21: Real-time water quality monitoring and data collection: Adopt a turbidity sensor to monitor the turbidity of the ballast water in real time, and adopt a particulate matter concentration sensor to monitor the suspended particle concentration of the ballast water in real time; The ultraviolet transmittance sensor is used to measure the ultraviolet light transmittance and calculate the ultraviolet light dose: wherein, is the ultraviolet light dose, is the ultraviolet light intensity, is the irradiation time; The flow velocity sensor is used to detect the water flow velocity , ensuring the residence time of water flow in the ultraviolet sterilization unit to meet the following conditions: wherein, is the effective length of the ultraviolet sterilization unit; Step 22: Dynamically adjust the filtration parameters: When the concentration of suspended particles is higher than the threshold , the filtration accuracy level will be automatically increased or the backwash frequency will be increased ; When the filtered turbidity is still higher than the set value, the water flow velocity is reduced to improve the filtration effect; Step 23: Adjust the ultraviolet sterilization power Calculating the Ultraviolet Dose Requirement where is an empirical coefficient and is adjusted according to different water qualities; According to the ultraviolet dose requirement Adjust the ultraviolet lamp power wherein is the maximum ultraviolet power, is the target ultraviolet dose; Step 24: Adaptive flow rate control Calculate the optimal flow velocity range wherein, is the optimal ultraviolet sterilization residence time; Automatically adjust the water flow rate through a variable frequency pump and a flow control valve To match the optimal UV sterilization residence time; Step 25: Abnormal handling mechanism When the filtered turbidity continues to be higher than the turbidity set value , an alarm is given and the filtration mode is adjusted; When the biological survival rate of the ultraviolet sterilization unit exceeds the set survival rate , the power of the ultraviolet lamp is increased or the flow rate is decreased.

3. The collaborative control method for the filtration and ultraviolet sterilization links in a pure physical ballast water management system according to claim 1, characterized in that, The specific content of Step 3 includes: Step 31: Data collection and feature engineering Collect water quality data before and after filtration, including the suspended particle concentration before filtration , the suspended particle concentration after filtration , the turbidity NTU before filtration, the turbidity after filtration , the UV transmittance UVT after filtration, the water flow rate , the UV lamp power and the sterilization rate after filtration ; Principal component analysis (PCA) is used for dimensionality reduction to extract key influencing factors : , where is the dimensionality reduction transformation matrix, is the original data feature matrix; Step 32: Model training and optimization Using Support Vector Regression (SVR) to predict the optimal filtration accuracy and the ultraviolet dose requirement ; Training objective: , where is the sterilization rate predicted by the model, is the sterilization after filtration; N is the summary of historical data samples, and i is the sample index; Adopt Cross Validation to improve the generalization ability of the model; Step 33: Real-time prediction and dynamic adjustment: Calculate the current water quality state vector : Using a Long Short-Term Memory (LSTM) network to predict the optimal filtration accuracy at the next moment and the optimal ultraviolet dose requirement at the next moment : , where f is the mapping function of the LSTM network, and are model parameters; applying the optimal filtration accuracy at the next moment to the filtration module to adjust the accuracy; adjusting the ultraviolet lamp power according to the optimal ultraviolet dose requirement at the next moment ;​ Step 34: Optimize the control strategy by reinforcement learning: Adopt Deep Q-Network (DQN) to optimize the filtration and ultraviolet sterilization strategies; Reward Function Design :[[]] , where is the total energy consumption of the ballast water management system, is the weight coefficient; during the training process, the agent learns the optimal control strategy through trial and error to maximize the sterilization rate and minimize the energy consumption; Step 35: Deployment and adaptive optimization of the ballast water management system: Continuously update the machine learning model through online learning to adapt to different water quality environments; Combine the historical ballast water data of the ship to establish a personalized optimization model to improve the adaptability of different shipping routes.

4. The collaborative control method for the filtering and ultraviolet sterilization links in a pure physical ballast water management system according to claim 1, wherein The specific content of Step 4 includes: Step 41: Flow rate monitoring and data collection; Step 42: Calculate the adaptive flow rate: Optimal flow rate Calculation: ; represents the optimal UV sterilization residence time; When the residence time in the ultraviolet sterilization unit is reached, the water flow rate needs to be increased to avoid waste of energy consumption caused by excessive irradiation; When the residence time in the ultraviolet sterilization unit reduce the water flow rate to ensure the sterilization effect; Step 43: Flow rate adjustment control strategy: Adopt Fuzzy Control to achieve adaptive adjustment; Set the input variables of fuzzy control: the current flow rate , the UV transmittance (UVT) after filtration, and the turbidity (NTU) before filtration; Set the control rules: If the ultraviolet transmittance (UVT) after filtration is low and the turbidity (NTU) before filtration is high, then reduce the water flow rate ; If the ultraviolet transmittance (UVT) is high after filtration and the turbidity (NTU) is low before filtration, increase the water flow rate ; Adjustment range is calculated using fuzzy inference wherein is the target flow rate; Step 44: Closed-loop control and optimization Set the error feedback mechanism: Among them, represents the flow rate at the current moment, are the PID control parameters; if the error is too large, the controller will adaptively and quickly adjust the flow rate; if the error is small, it will be adjusted slowly to ensure the stability of the flow rate; Step 45: Optimization and dynamic adjustment of the ballast water management system.

5. The collaborative control method for the filtration and ultraviolet sterilization links in a pure physical ballast water management system according to claim 1, characterized in that The specific content of Step 5 includes: Step 51: Energy consumption modeling and objective function: The total energy consumption of the ballast water management system is mainly composed of the energy consumption of the filtration unit and the energy consumption of the ultraviolet disinfection unit as follows: ; Filter unit energy consumption affected by water flow velocity and backwash frequency as follows: , where is the flow velocity correlation coefficient, is the backwash energy consumption coefficient; Energy consumption of the ultraviolet sterilization unit affected by the power of the ultraviolet lamp and the ultraviolet transmittance UVT: ; The goal is to minimize the total energy consumption of the ballast water management system , while meeting the post-filtration sterilization rate Constraints: s.t. , represents the minimum sterilization rate; Step 52: Energy consumption optimization strategy based on reinforcement learning: Adopt Deep Q-Network (DQN) to optimize the collaborative energy consumption regulation strategy of filtration and ultraviolet sterilization; State variables : Water quality parameters (NTU, UVT, C), current flow rate , UV lamp power ; Action space : Adjust V, F and ; Reward function : , where is the penalty factor; During the training process, the agent learns the optimal strategy based on historical data and continuously updates online; Step 53: Dynamically adjust the power of ultraviolet lamps: Calculate the optimal ultraviolet lamp power ; When the ultraviolet transmittance UVT is high, reduce the power of the ultraviolet lamp to reduce energy consumption; When the sterilization rate after filtration decreases, increase the power of the ultraviolet lamp to ensure the sterilization effect; Step 54: Collaborative optimization strategy: Optimized by Genetic Algorithm GA Parameter combination; Fitness function: ; Adopt selection, crossover and mutation operations to search for the optimal energy consumption solution; Step 55: Online adaptive optimization of the system: Combine reinforcement learning and genetic algorithm to dynamically update the energy consumption optimization strategy; Adopt predictive maintenance to prevent the performance degradation caused by the long-term operation of the system.

Images