Load balancing regulation and control method of liquefied gas carrier liquid cargo operation system simulator

Abstract

The invention provides a load balancing regulation and control method for a liquefied gas carrier liquid cargo operation system simulator, and belongs to the technical field of traffic transportation, and the method comprises the steps: firstly building a liquid tank pressure response model and a ship state monitoring model, and providing support for safe operation; and deploying a real-time load monitoring module to collect performance data, and carrying out preprocessing and load evaluation to comprehensively grasp the working state of the system. On the basis, dynamic task priority adjustment and load trend prediction are adopted, it is ensured that key tasks are processed in time, and the overall efficiency of the system is improved. And finally, making a dynamic load balancing strategy based on load distribution evaluation, realizing optimal configuration of resources, and monitoring and adjusting in real time to cope with load change so as to ensure stable and reliable operation of the system. According to the scheme, the technologies of modeling analysis, real-time monitoring, dynamic scheduling and the like are integrated, and the technical problems that a complex coupling relation exists among subsystems, local overload or resource waste is possibly caused by unbalanced load distribution, and the overall performance is affected are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of transportation, and in particular relates to a load balancing control method for a liquefied gas ship liquid cargo operating system simulator. Background Art

[0002] Liquefied gas carriers are special ships used to transport liquefied flammable gases such as liquefied natural gas (LNG) or liquefied petroleum gas (LPG). During the loading and unloading process of liquefied gas carriers, it is necessary to strictly control the pressure and temperature in the liquid tanks, maintain the physical state of the liquid cargo, and ensure the buoyancy and strength of the ship. At the same time, it is also necessary to monitor and control the operating status of each component of the system, reasonably allocate system resources, and improve the overall operating efficiency. At present, most liquefied gas carrier liquid cargo operating systems are controlled and monitored by human-computer interaction, relying on the experience and judgment of the operator. With the expansion of the system scale and the increase in operational complexity, manual intervention methods are difficult to meet the requirements of real-time, precision, and stability.

[0003] Some research institutions and enterprises have developed computer simulation software for liquid cargo operation systems of liquefied gas tankers to assist in the design and optimization of loading and unloading operations. These simulation software generally adopt a modular structure, including functional units such as liquid cargo tank simulation, pressure and temperature control, and ship stability calculation. However, most of these simulation systems simulate the loading and unloading process statically, lacking real-time monitoring and dynamic regulation of the system's operating status. In actual applications, there are complex coupling relationships between the subsystems, and uneven load distribution may lead to local overload or resource waste, affecting overall performance. In addition, existing simulation software is usually unable to achieve effective integration with real liquid cargo loading and unloading equipment, making it difficult to conduct comprehensive simulation tests on actual systems. Summary of the invention

[0004] In view of this, the present invention provides a load balancing control method for a liquefied gas tanker liquid cargo operating system simulator, which can solve the technical problems that complex coupling relationships between subsystems and uneven load distribution may cause local overload or resource waste, affecting overall performance.

[0005] The present invention is achieved in that:

[0006] The present invention provides a load balancing control method for a liquefied gas ship liquid cargo operating system simulator, comprising the following steps:

[0007] S10, establishing a tank pressure response model: constructing a tank pressure response model according to the cargo loading volume and the vapor discharge volume; determining the parameters of the tank pressure response model by experimentally measuring the pressure change data during the loading process; the tank pressure response model considers the gain coefficient, time constant, loading rate, return air rate and initial pressure, and is used to predict the tank pressure change;

[0008] S20. Establishing a ship status monitoring model: constructing a ship status monitoring model including transverse stability, longitudinal stability, inclination and draft parameters; the ship status monitoring model considers the impact of changes in the weight of liquid cargo or ballast water on various ship parameters, and evaluates the buoyancy, stability and strength of the ship through real-time calculation, so as to provide safety guarantee for liquid cargo operation;

[0009] S30, deploying a real-time load monitoring module: deploying a real-time load monitoring module in the liquefied gas tanker liquid cargo operating system system to collect the central processing unit utilization rate, memory occupancy rate, network bandwidth utilization rate and disk read and write speed performance data of each device or unit; obtaining a comprehensive performance index through weighted calculation to provide basic data support for subsequent load balancing;

[0010] S40, data preprocessing and load evaluation: preprocessing the collected performance data, including data cleaning and normalization; using the load evaluation function to calculate the load value of each device or unit, considering the performance index, load change rate and random disturbance factors, and comprehensively evaluating the working status of each component of the liquefied gas tanker liquid cargo operation system;

[0011] S50, task priority calculation and dynamic adjustment: calculate the priority value of each task according to the urgency, importance, resource requirements and waiting time of the task; dynamically adjust the task priority taking into account time decay factors and random fluctuations to ensure that key tasks are processed in a timely manner and improve the overall efficiency of the liquefied gas tanker liquid cargo operation system;

[0012] S60, Load trend prediction: Use the long short-term memory neural network model to predict future load trends based on historical load data; capture the long-term and short-term dependencies of load changes through the calculation of forget gates, input gates, output gates, and unit states, and provide forward-looking guidance for the formulation of load balancing strategies;

[0013] S70, formulation of load balancing strategy: based on the load balancing degree calculation formula, evaluate the current load distribution of the liquefied gas tanker liquid cargo operating system; consider the load difference of each device or unit and the random disturbance factor, formulate a load balancing strategy, achieve the optimal configuration of the system resources of the liquefied gas tanker liquid cargo operating system, and improve the overall operation efficiency;

[0014] S80, real-time scheduling and load adjustment: using the load change detection function, real-time monitoring of the load changes of the liquefied gas tanker liquid cargo operating system; by calculating the first-order difference and second-order derivative of the load, rapid response to load mutations, dynamic adjustment of task allocation and resource allocation, to ensure that the liquefied gas tanker liquid cargo operating system maintains stable and efficient operation under various working conditions.

[0015] Among them, the liquefied gas ship liquid cargo operating system simulator adopts a three-layer architecture design, including a host computer closed loop, a host closed loop and a front-end closed loop; the host computer closed loop is used to display the human-machine interface of the liquefied gas ship loading and unloading system and simulate data display and response; the host closed loop is used to simulate the liquid cargo loading and unloading functions and internal unit simulation according to the instructions and data issued by the host computer; the front-end closed loop is used to perform data interaction with the real unit equipment of the liquefied gas ship loading and unloading system simulator. The liquefied gas ship liquid cargo operating system simulator also includes an architecture interaction mechanism, which includes: a data communication protocol, which is used to ensure that the data transmission between the host computer closed loop, the host closed loop and the front-end closed loop is accurate; real-time data processing, which is used to ensure that the data processing and response time of each layer meet the system requirements; a synchronization mechanism, which is used to realize data synchronization based on timestamps to ensure data consistency between the host computer, the host and the front end. The liquefied gas ship liquid cargo operating system simulator is composed of a liquefied natural gas ship cargo tank simulation unit, a liquefied petroleum gas ship simulation unit, a reliquefaction system simulation unit, a nitrogen production system simulation unit and a communication module unit; a combined structure is adopted, and the industrial control host, data transmission module, and power supply system are connected together through a bus, supporting the selection of multiple data transmission modules, and communicating with the host computer through an Ethernet interface. Each unit of the liquefied gas ship liquid cargo operating system simulator adopts a physical interface, and the transmitted signals are all real physical signals; the virtualization unit runs in the software system, receives real input signals through data drive or simulation drive, and generates a response signal at the corresponding interface.

[0016] Furthermore, the tank pressure response model is specifically expressed as follows:

[0017] [z p ]=[M1 / (T1s+1)M2 / (T2s+1)][v1;v2]+[c p ];

[0018] In the formula, z p is the tank pressure in KPa; M1, M2 are gain coefficients, dimensionless, and M1+M2=1; T1, T2 are time constants in seconds, and T1+T2=T ε ; v1 is the loading rate, in m 3 / h; v2 is the return air rate, unit is m 3 / h;c p is the initial pressure of the liquid tank, in KPa; s is the Laplace operator.

[0019] Parameter acquisition method: Through experimental measurement, the average pressure of liquid cargo vapor at the beginning of loading is 2KPa, and the average pressure of gas-liquid mixture in the tank is 12KPa after loading is completed. The loading process lasts for 15 hours. τ= 36000s. The experimental steps include: 1) preparing the liquefied gas tanker and the onshore loading and unloading facilities; 2) starting loading and recording the initial pressure; 3) recording the tank pressure every 30 minutes; 4) recording the final pressure after loading is completed.

[0020] The ship status monitoring model is specifically expressed as follows:

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030] Where GM1 is the new transverse stability height, in m; GM is the original transverse stability height of the ship, in m; m is the weight of the change in liquid cargo or ballast water, in t; D is the displacement of the ship, in t; d m is the average draft of the ship, in m; z m is the vertical coordinate of the changing position of liquid cargo or ballast water, in meters; ρ x is the density of each liquid, in t / m 3 ;i x is the moment of inertia of each liquid surface area, in m4; GM L1 is the new longitudinal stability height, in m; GM L is the original longitudinal stability height of the ship, in m; θ is the heel angle, in rad; φ is the longitudinal heel angle, in rad; y m is the transverse coordinate of the changing position of liquid cargo or ballast water, in meters; m is the longitudinal coordinate of the changing position of liquid cargo or ballast water, in meters; F is the longitudinal coordinate of the drift center, in m; d F is the original bow draft of the ship, in m; d A is the original stern draft of the ship, in m; d F1 is the new bow draft of the ship, in m; d A1is the new stern draft of the ship, in m; L is the length of the ship, in m; TPC is the tons of the ship per centimeter of draft, in t / cm; MTC is the initial state ship's longitudinal moment per centimeter, in t·m / cm.

[0031] Parameter acquisition method: ship original parameters (such as GM, GM L , D, d F , d A The relevant parameters of liquid cargo or ballast water (such as m, x m ,y m , z m etc.) are measured in real time by level gauge and weight calculation system.

[0032] The performance indicator calculation formula of the real-time load monitoring module is as follows:

[0033] P i =αC i +βM i +γN i +δD i +ε i ;

[0034] Where P i is the comprehensive performance index of the ith device or unit, dimensionless; C i is the CPU usage, ranging from 0 to 100%; M i N is the memory usage, ranging from 0 to 100%; i is the network bandwidth utilization, the value range is 0-100%; D i is the disk read and write speed, in MB / s; α, β, γ, δ are weight coefficients, dimensionless, and α+β+γ+δ=1; ε i is the error term, which follows the normal distribution N(0,σ 2 ), where σ is the standard deviation, usually taken as 0.1.

[0035] Parameter acquisition method: C i , M i , N i , D i Obtained in real time through system monitoring tools (such as top, iostat, etc.). The weight coefficients α, β, γ, and δ are determined through expert experience or historical data analysis.

[0036] The load evaluation function is specifically expressed as follows:

[0037]

[0038] Where, L iis the load evaluation value of the ith device or unit, dimensionless; P ij is the jth performance index of the ith device or unit, dimensionless; w j is the weight of the jth performance indicator, dimensionless, and n is the number of performance indicators; λ is the weight coefficient of the load change rate, dimensionless, usually ranging from 0.1 to 0.5; is the rate of change of performance indicators, in units of 1 / s; μ i is a random disturbance term, which obeys the uniform distribution U(-a,a), where a is the disturbance amplitude, usually taking the value of 0.05.

[0039] Parameter acquisition method: P ij Obtained through the real-time monitoring module in step S3. j and λ are determined through historical data analysis and expert experience. It is obtained by calculating the difference between the performance indicators of two adjacent time points and dividing it by the time interval.

[0040] The task priority calculation formula is specifically expressed as follows:

[0041]

[0042] In the formula, PR k is the priority value of the kth task, dimensionless; U k is the urgency of the task, ranging from 1 to 10; I k is the task importance, ranging from 1 to 10; R k is the task resource requirement, ranging from 1 to 10; T k is the task waiting time, in seconds; ω1, ω2, ω3, ω4 are dimensionless weight coefficients, and ω1+ω2+ω3+ω4=1; λ is the time attenuation coefficient, in 1 / s, usually in the range of 0.001-0.01; ξ k is a random fluctuation term, which obeys the normal distribution N(0,σ 2 ), where σ is the standard deviation, usually taken as 0.1.

[0043] Parameter acquisition method: U k , I k , R k Specified by the task submitter or set by the system default. k Calculated by the system clock. The weight coefficients ω1, ω2, ω3, ω4 and λ are determined by historical data analysis and expert experience.

[0044] Parameter acquisition method: By collecting historical load data, the LSTM network is trained using the back propagation algorithm to optimize the weight matrix and bias terms. The training steps include: 1) data preprocessing; 2) building the LSTM network structure; 3) setting the loss function and optimizer; 4) iterative training; 5) model verification and tuning.

[0045] The load balancing degree calculation formula is specifically expressed as follows:

[0046]

[0047] Where B is the load balance degree, dimensionless, ranging from 0 to 1. The closer to 1, the more balanced the load. L i is the load of the ith device or unit, dimensionless; is the average load, dimensionless; m is the total number of devices or units; η is the random disturbance term, which follows the normal distribution N(0,σ 2 ), where σ is the standard deviation, usually taken as 0.01.

[0048] Parameter acquisition method: L i It is calculated by the load evaluation function in step S4. It is the arithmetic mean of all equipment or unit loads.

[0049] The load change detection function is specifically expressed as follows:

[0050]

[0051] Where D(t) is the load change detection function, dimensionless; L i (t) is the load of the i-th device or unit at time t, dimensionless; m is the total number of devices or units; κ is the second-order derivative weight coefficient, dimensionless, usually ranging from 0.1 to 0.5; It is the second-order derivative of the load, indicating the acceleration of the load change, in units of 1 / s 2 ρ i It is a random fluctuation term, which obeys the uniform distribution U(-b,b), where b is the fluctuation amplitude, which is usually 0.05.

[0052] Parameter acquisition method: L i (t) is calculated in real time by the load evaluation function in step S4. The load values ​​at three adjacent time points are approximated using the central difference method. κ is determined through historical data analysis and expert experience.

[0053] Compared with the prior art, the load balancing control method of a liquefied gas ship liquid cargo operating system simulator provided by the present invention has the following beneficial effects:

[0054] Firstly, the method constructs a tank pressure response model and a ship status monitoring model, accurately simulating the pressure change law and ship buoyancy during the liquid cargo loading and unloading process. Through the analysis and modeling of actual loading data, the mathematical expression of the tank pressure changing with the cargo volume, return air volume and initial pressure is obtained. At the same time, based on the ship parameters and the changes in the position of liquid cargo, a comprehensive evaluation model of transverse and longitudinal stability, draft change, etc. is established. These models provide basic support for the dynamic simulation and optimization control of the system.

[0055] Secondly, the method deploys a real-time load monitoring module to collect performance indicators such as CPU usage, memory usage, network bandwidth, and disk I / O of each hardware device and functional unit in real time. Comprehensive performance indicators are obtained through weighted calculation to provide a basis for load evaluation and balancing control. On this basis, the load evaluation function is combined with performance change trends and random disturbance factors to comprehensively analyze the working status of each component of the system, providing a basis for dynamic load balancing.

[0056] Thirdly, this method integrates the functions of task priority calculation, load trend prediction and load balancing strategy formulation. By considering factors such as task urgency, importance and resource requirements, the task priority is adjusted dynamically to ensure that key tasks are responded to in a timely manner. At the same time, the LSTM neural network is used to predict future load change trends and provide forward-looking guidance for the formulation of load balancing strategies. Based on the load balancing degree calculation formula, the current load distribution of the system is dynamically evaluated, and an optimization strategy is formulated to achieve a reasonable allocation of system resources.

[0057] Finally, the method also designs a real-time load change detection and dynamic adjustment mechanism. By analyzing the first-order difference and second-order derivative of the load, the load mutation can be quickly identified, and the task allocation and resource allocation can be adjusted in a timely manner to ensure that the system maintains stable and efficient operation under various working conditions.

[0058] In summary, the load balancing control method of the liquefied gas tanker liquid cargo operating system simulator proposed in the present invention, based on the existing simulation technology, integrates the functions of ship status monitoring, load assessment, task priority allocation, trend prediction and dynamic balancing, realizes comprehensive perception and intelligent control of the system operation status, and solves the technical problems in actual applications that there are complex coupling relationships between subsystems, and uneven load distribution may lead to local overload or resource waste, affecting the overall performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flow chart of the method provided by the present invention;

[0060] Figure 2 It is a structural design diagram in the embodiment;

[0061] Figure 3It is a structural diagram of the simulator system in the embodiment;

[0062] Figure 4 This is a further optimization strategy diagram in the embodiment. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0064] like Figure 1 As shown, it is a flow chart of a load balancing control method of a liquefied gas ship liquid cargo operating system simulator provided by the present invention, and the method comprises the following steps:

[0065] S10. Establishing a tank pressure response model: constructing a tank pressure response model based on the cargo loading and vapor discharge; determining the parameters of the tank pressure response model by experimentally measuring the pressure change data during the loading process; the tank pressure response model considers the gain coefficient, time constant, loading rate, return air rate and initial pressure to predict the tank pressure change;

[0066] S20. Establish a ship status monitoring model: Construct a ship status monitoring model that includes transverse stability, longitudinal stability, inclination and draft parameters; the ship status monitoring model considers the impact of changes in the weight of liquid cargo or ballast water on various ship parameters, and evaluates the buoyancy, stability and strength of the ship through real-time calculation, providing safety protection for liquid cargo operations;

[0067] S30. Deploy real-time load monitoring module: deploy a real-time load monitoring module in the liquid cargo operating system of the liquefied gas tanker to collect the CPU usage rate, memory occupancy rate, network bandwidth utilization rate and disk read and write speed performance data of each device or unit; obtain comprehensive performance indicators through weighted calculation to provide basic data support for subsequent load balancing;

[0068] S40, data preprocessing and load evaluation: preprocess the collected performance data, including data cleaning and normalization; use the load evaluation function to calculate the load value of each device or unit, consider the performance index, load change rate and random disturbance factors, and comprehensively evaluate the working status of each component of the liquefied gas tanker liquid cargo operation system;

[0069] S50, Task priority calculation and dynamic adjustment: Calculate the priority value of each task according to the urgency, importance, resource requirements and waiting time of the task; Consider time decay factors and random fluctuations, dynamically adjust task priorities, ensure that key tasks are processed in a timely manner, and improve the overall efficiency of the liquefied gas tanker liquid cargo operation system;

[0070] S60, Load trend prediction: Use the long short-term memory neural network model to predict future load trends based on historical load data; capture the long-term and short-term dependencies of load changes through the calculation of forget gates, input gates, output gates, and unit states, and provide forward-looking guidance for the formulation of load balancing strategies;

[0071] S70, formulation of load balancing strategy: Based on the load balancing degree calculation formula, evaluate the current load distribution of the liquefied gas tanker liquid cargo operating system; consider the load differences of each device or unit and random disturbance factors, formulate a load balancing strategy, achieve the optimal configuration of the system resources of the liquefied gas tanker liquid cargo operating system, and improve the overall operation efficiency;

[0072] S80, real-time scheduling and load adjustment: use the load change detection function to monitor the load changes of the liquefied gas tanker liquid cargo operating system in real time; by calculating the first-order difference and second-order derivative of the load, quickly respond to load mutations, dynamically adjust task allocation and resource allocation, and ensure that the liquefied gas tanker liquid cargo operating system maintains stable and efficient operation under various working conditions.

[0073] The specific implementation methods of the above steps are described in detail below:

[0074] The specific implementation of step S10 is as follows: First, a tank pressure response model is constructed. The model takes into account the influencing factors during the loading process, including the loading volume, the steam exhaust volume, and the initial pressure. By experimentally measuring the pressure change data during the loading process, the model parameters are determined, and the change law of the tank pressure over time is obtained. The specific mathematical expression of the tank pressure response model is as follows:

[0075] [z p ]=[M1 / (T1s+1)M2 / (T2s+1)][v1;v2]+[c p ];

[0076] Among them, z p is the tank pressure in KPa; M1 and M2 are gain coefficients, dimensionless, and M1+M2=1; T1 and T2 are time constants in seconds, and T1+T2=T ε ; v1 is the loading rate, in m 3 / h; v2 is the return air rate, unit is m 3 / h;c p is the initial pressure of the liquid tank, in KPa; s is the Laplace operator.

[0077] Through experimental measurement, the average pressure of liquid cargo vapor is 2KPa at the beginning of loading, and the average pressure of gas-liquid mixture in the tank is 12KPa after loading is completed. The loading process lasts for 15 hours. τ= 36000s. The experimental steps include: 1) preparing the liquefied gas tanker and the onshore loading and unloading facilities; 2) starting loading and recording the initial pressure; 3) recording the tank pressure every 30 minutes; 4) recording the final pressure after loading is completed.

[0078] Such a tank pressure response model can accurately reflect the changing pattern of tank pressure during loading and provide basic support for subsequent system simulation and optimization control.

[0079] The specific implementation of step S20 is: constructing a ship status monitoring model. The model includes key parameters such as the ship's transverse stability, longitudinal stability, inclination and draft, and can evaluate the ship's buoyancy, stability and strength in real time. The specific mathematical expression of the ship status monitoring model is as follows:

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087] Among them, GM1 is the new transverse stability height, in meters; GM is the original transverse stability height of the ship, in meters; m is the weight of the change in liquid cargo or ballast water, in tons; D is the displacement of the ship, in tons; d m is the average draft of the ship, in m; z m is the vertical coordinate of the changing position of liquid cargo or ballast water, in meters; ρ x is the density of each liquid, in t / m 3 ;i x is the moment of inertia of each liquid surface area, in m 4 GM L1 is the new longitudinal stability height, in m; GM L is the original longitudinal stability height of the ship, in m; θ is the heel angle, in rad; φ is the longitudinal heel angle, in rad; y m is the transverse coordinate of the changing position of liquid cargo or ballast water, in meters; m is the longitudinal coordinate of the changing position of liquid cargo or ballast water, in meters; F is the longitudinal coordinate of the drift center, in m; d F is the original bow draft of the ship, in m; dA is the original stern draft of the ship, in m; d F1 is the new bow draft of the ship, in m; d A1 is the new stern draft of the ship, in m; L is the length of the ship, in m; TPC is the tons of the ship per centimeter of draft, in t / cm; MTC is the initial state ship's longitudinal moment per centimeter, in t·m / cm.

[0088] These parameters can be obtained from ship design drawings, stability manuals and real-time measurement systems. By establishing such a ship status monitoring model, the floating state, stability and strength of the ship can be grasped in real time, providing safety protection for liquid cargo operations.

[0089] The specific implementation of step S30 is: deploying a real-time load monitoring module. This module collects performance index data of each device or unit in the liquefied gas tanker liquid cargo operating system, including central processing unit utilization, memory occupancy, network bandwidth utilization, and disk read and write speed, etc. The comprehensive performance index is obtained through weighted calculation to provide basic data support for subsequent load evaluation and balanced regulation.

[0090] Comprehensive performance index P i The calculation formula is as follows:

[0091] P i =αC i +βM i +γN i +δD i +ε i ;

[0092] Among them, C i is the CPU usage, ranging from 0 to 100%; M i N is the memory usage, ranging from 0 to 100%; i is the network bandwidth utilization, the value range is 0-100%; D i is the disk read and write speed, in MB / s; α, β, γ and δ are dimensionless weight coefficients, and α+β+γ+δ=1; ε i is the error term, which follows the normal distribution N(0,σ 2 ), where σ is the standard deviation, usually taken as 0.1.

[0093] By deploying this real-time load monitoring module, the performance indicator data of each component of the system can be obtained in real time, providing an important basis for subsequent load evaluation and balancing strategy formulation.

[0094] The specific implementation of step S40 is: pre-processing and load evaluation of the collected performance index data. First, the raw data is cleaned and normalized to eliminate dimensional differences and outliers. Then, the load evaluation function is used to calculate the load value of each device or unit, taking into account the performance index, load change rate and random disturbance factors, and comprehensively evaluating the working status of each component of the system.

[0095] The specific expression of the load evaluation function is as follows:

[0096]

[0097] Among them, L i is the load evaluation value of the ith device or unit, dimensionless; P ij is the jth performance index of the ith device or unit, dimensionless; w j is the weight of the jth performance indicator, dimensionless, and n is the number of performance indicators; λ is the weight coefficient of the load change rate, dimensionless, usually ranging from 0.1 to 0.5; is the rate of change of performance indicators, in units of 1 / s; μ i is a random disturbance term, which obeys the uniform distribution U(-a,a), where a is the disturbance amplitude, usually taking the value of 0.05.

[0098] Such a load evaluation process can fully reflect the working status of each component of the system and provide a basis for the subsequent formulation of load balancing strategies.

[0099] The specific implementation of step S50 is: calculate the priority value of each task according to the urgency, importance, resource requirements and waiting time of the task, consider the time decay factor and random fluctuation, dynamically adjust the task priority, ensure that the key tasks are processed in time, and improve the overall efficiency of the system.

[0100] The task priority calculation formula is as follows:

[0101]

[0102] Among them, PR k is the priority value of the kth task, dimensionless; U k is the urgency of the task, ranging from 1 to 10; I k is the task importance, ranging from 1 to 10; R k is the task resource requirement, ranging from 1 to 10; T k is the task waiting time, in seconds; ω1, ω2, ω3 and ω4 are dimensionless weight coefficients, and ω1+ω2+ω3+ω4=1; λ is the time attenuation coefficient, in 1 / s, usually in the range of 0.001-0.01; ξk is a random fluctuation term, which obeys the normal distribution N(0,σ 2 ), where σ is the standard deviation, usually taken as 0.1.

[0103] Through such task priority calculation and dynamic adjustment mechanism, it can be ensured that key tasks are processed in a timely manner and the overall operation efficiency of the system is improved.

[0104] The specific implementation of step S60 is: using a long short-term memory (LSTM) neural network model to predict future load trends based on historical load data. The LSTM network can capture the long-term and short-term dependencies of load changes and provide forward-looking guidance for load balancing strategy formulation.

[0105] By collecting historical load data, the back propagation algorithm is used to train the LSTM network and optimize the weight matrix and bias terms. The training steps include: 1) data preprocessing; 2) building the LSTM network structure; 3) setting the loss function and optimizer; 4) iterative training; 5) model verification and tuning.

[0106] The specific implementation of step S70 is: based on the load balancing calculation formula, evaluate the current load distribution of the system, consider the load differences and random disturbance factors of each device or unit, formulate a load balancing strategy, achieve the optimal configuration of system resources, and improve the overall operation efficiency.

[0107] The load balancing calculation formula is as follows:

[0108]

[0109] Among them, B is the load balance degree, dimensionless, and the value range is 0-1. The closer to 1, the more balanced the load; L i is the load of the ith device or unit, dimensionless; is the average load, dimensionless; m is the total number of devices or units; η is the random disturbance term, which follows the normal distribution N(0,σ 2 ), where σ is the standard deviation, usually taken as 0.01.

[0110] Through such load balancing evaluation and optimization strategy formulation, reasonable allocation of system resources can be achieved and overall operating efficiency can be improved.

[0111] The specific implementation of step S80 is: using the load change detection function to monitor the system load change in real time, by calculating the first-order difference and second-order derivative of the load, quickly responding to load mutations, dynamically adjusting task allocation and resource allocation, and ensuring that the system maintains stable and efficient operation under various working conditions.

[0112] The specific expression of the load change detection function is as follows:

[0113]

[0114] Where D(t) is the load change detection function, dimensionless; L i (t) is the load of the i-th device or unit at time t, dimensionless; m is the total number of devices or units; κ is the second-order derivative weight coefficient, dimensionless, usually ranging from 0.1 to 0.5; It is the second-order derivative of the load, indicating the acceleration of the load change, in units of 1 / s 2 ρ i It is a random fluctuation term, which obeys the uniform distribution U(-b,b), where b is the fluctuation amplitude, which is usually 0.05.

[0115] Through such load change detection and dynamic adjustment mechanism, the system can be ensured to maintain stable and efficient operation under various working conditions.

[0116] Specifically, the principle of the present invention is to establish a multi-level monitoring and control mechanism to achieve comprehensive perception and dynamic optimization of the system operation status.

[0117] Firstly, the method constructs the tank pressure response model and the ship status monitoring model to provide basic support for system simulation. The tank pressure response model takes into account factors such as cargo volume, vapor discharge volume and initial pressure. By analyzing and modeling the actual loading data, the mathematical expression of the tank pressure changing with time is obtained. The model can accurately reflect the dynamic change law of tank pressure during loading.

[0118] The ship status monitoring model includes key parameters such as transverse and longitudinal stability, inclination and draft. The model calculates various stability and strength indicators of the ship according to the changes in the weight of liquid cargo or ballast water, providing safety protection for liquid cargo operations. By real-time monitoring of the ship's floating state and stability, abnormal situations can be discovered in time and corresponding control measures can be taken.

[0119] Secondly, this method deploys a real-time load monitoring module in the system to collect performance indicator data of each device or unit in real time, including CPU usage, memory usage, network bandwidth, and disk I / O. The comprehensive performance indicators are obtained through weighted calculation to provide a data basis for subsequent load evaluation. Such a real-time monitoring mechanism can fully perceive the current operating status of the system and provide a basis for dynamic load balancing.

[0120] On this basis, this method uses load evaluation function, combined with performance indicators, load change trends and random disturbances, to comprehensively evaluate the workload of each device or unit. Through such load modeling and evaluation process, the actual operating status of each component of the system can be objectively reflected, providing a basis for the subsequent formulation of load balancing strategies.

[0121] In addition, this method integrates the functions of task priority calculation, load trend prediction and load balancing strategy formulation. By considering factors such as the urgency, importance and resource requirements of the task, the priority of the task is dynamically adjusted to ensure that key tasks are responded to in a timely manner. At the same time, the LSTM neural network is used to predict the future load change trend and provide forward-looking guidance for the formulation of load balancing strategies. Based on the load balancing degree calculation formula, the current load distribution of the system is dynamically evaluated, and an optimization strategy is formulated to achieve the reasonable allocation of system resources.

[0122] Finally, the method designs a real-time load change detection and dynamic adjustment mechanism. By analyzing the first-order difference and second-order derivative of the load, the load mutation can be quickly identified, and the task allocation and resource allocation can be adjusted in time to ensure that the system maintains stable and efficient operation under various working conditions.

[0123] A specific embodiment of the present invention is given below: A large liquefied natural gas transportation enterprise has a large fleet of liquefied gas ships, which undertakes the task of natural gas transportation at home and abroad. In order to improve the safety and efficiency of liquid cargo loading and unloading operations, the enterprise decided to develop a set of liquefied gas ship liquid cargo operation system simulator based on the present invention to fully simulate and optimize the loading and unloading operation process.

[0124] like Figure 2 The simulator shown in the figure adopts a three-layer architecture design, including the upper computer closed loop, the host closed loop and the front-end closed loop. The upper computer closed loop is responsible for the human-machine interaction interface and the display and response of simulated data; the host closed loop simulates the liquid cargo loading and unloading functions and the internal unit simulation according to the upper computer instructions; the front-end closed loop is responsible for data interaction with the real liquefied gas ship loading and unloading system equipment. The three closed loops are orderly linked through data communication protocols, real-time data processing and synchronization mechanisms to ensure accurate data transmission and timely processing at each layer, and achieve data consistency between the upper computer, the host and the front end.

[0125] Figure 3 The overall structure diagram of the liquefied gas tanker liquid cargo operation system simulator is given. As can be seen from the figure, the simulator consists of a liquefied natural gas tanker cargo hold simulation unit, a liquefied petroleum gas tanker simulation unit, a reliquefaction system simulation unit, a nitrogen generation system simulation unit, and a communication module unit. Each unit adopts a combined structure, connecting the industrial control host, data transmission module, and power supply system through a bus, supporting the selection of multiple data transmission modes, and realizing communication with the host computer through an Ethernet interface.

[0126] In the specific implementation process, the liquefied gas tanker liquid cargo operating system simulator mainly performs the following steps:

[0127] The first step is to build a tank pressure response model. Through a 15-hour pressure test experiment on a certain type of LNG ship during the loading process, the data recorded that the average pressure of the liquid cargo vapor was 2kPa at the beginning of loading, and the average pressure of the gas-liquid mixture in the tank was 12kPa after loading was completed. Based on the experimental data, the following tank pressure response model was established:

[0128] [z p ]=[0.6 / (36s+1)0.4 / (12s+1)][v1;v2]+[2];

[0129] Among them, z p is the tank pressure, in kPa; v1 is the loading rate, in m 3 / h; v2 is the return air rate, unit is m 3 / h. The model takes into account factors such as cargo volume, vapor discharge and initial pressure, and can accurately reflect the change pattern of tank pressure over time.

[0130] The second step is to build a ship status monitoring model. The model includes the transverse stability high GM1 and the longitudinal stability high GM L1 , heel angle θ, pitch angle φ, average draft d m According to the design parameters of a certain type of LNG ship, the following ship status monitoring model was established:

[0131]

[0132]

[0133] Where, m is the added liquid cargo weight, which is 300 tons; D is the displacement, which is 30,000 tons; d F For the original bow draft, take 5.6m; d A For the original tail draft, take 5.9m; x F is the longitudinal coordinate of the center of drift, taking 50m; L is the length of the ship, taking 80m; TPC is the tonnage per centimeter of draft, taking 2.5 tons / cm; MTC is the trim moment per centimeter, taking 800 tons·m / cm. By real-time monitoring of these parameters, the buoyancy and strength of the ship can be evaluated, providing safety protection for liquid cargo operations.

[0134] The third step is to deploy a real-time load monitoring module. This module collects the performance indicators of each device or unit in the system in real time, including CPU usage, memory occupancy, network bandwidth utilization, and disk read and write speed. Based on expert experience, the weight coefficient of each indicator is determined:

[0135] α=0.4, β=0.3, γ=0.2, δ=0.1;

[0136] Substitute the collected performance data into the comprehensive performance index calculation formula:

[0137] P i =0.4C i +0.3M i +0.2N i +0.1D i +ε i ;

[0138] Among them, ε i Obey normal distribution N(0,0.01 2 ). Through the deployment of this module, the operating status of each component of the system can be mastered in real time.

[0139] The fourth step is to preprocess and load evaluate the collected performance indicator data. First, the raw data is cleaned and normalized to eliminate dimensional differences and outliers. Then, the load value L of each device or unit is calculated using the load evaluation function. i , taking into account performance indicators, load change rate and random disturbance factors:

[0140]

[0141] Among them, P i1 is the comprehensive performance index of the ith device or unit, is the rate of change of performance index, μ i It obeys the uniform distribution U(-0.05,0.05). Such a load evaluation process can fully reflect the working status of each component of the system.

[0142] Step 5: Calculate the priority value PR of each task based on the urgency, importance, resource requirements and waiting time of the task. k :

[0143]

[0144] Among them, U k is the urgency of the task (1-10), k is the task importance (1-10), R k is the task resource requirement (1-10), T k is the task waiting time (unit: seconds), ξ k Obeys normal distribution N(0,0.1 2 ). By dynamically adjusting task priorities, critical tasks can be ensured to receive timely responses.

[0145] The sixth step is to use LSTM neural network to predict future load change trends. After back-propagation training and optimization, the LSTM model can capture the long-term and short-term dependencies of load changes and provide forward-looking guidance for the subsequent formulation of load balancing strategies.

[0146] Step 7: Evaluate the current load distribution of the system based on the load balancing degree B:

[0147]

[0148] Among them, L i is the load of the ith device or unit, is the average load, and m is the total number of devices or units. Combined with the load prediction of the LSTM model, an optimization strategy is formulated to achieve reasonable allocation of system resources.

[0149] Step 8: Use the load change detection function d(t) to monitor the real-time changes of the system load:

[0150]

[0151] Once a sudden load change is detected, task allocation and resource configuration are adjusted immediately to ensure that the system maintains stable and efficient operation under various working conditions.

[0152] In order to further optimize system performance and improve the response speed and resource utilization of the simulator, the present invention adopts the following load balancing strategy and performance optimization measures. The strategy diagram is as follows: Figure 4 As shown:

[0153] (1) Task Dynamic Allocation Strategy

[0154] Real-time monitoring: The front-end closed-loop system monitors the key performance indicators of the device or unit in real time, including CPU usage, memory usage, and network bandwidth.

[0155] Random forest prediction: Uses random forest algorithm to predict future load trends of equipment or units based on historical data and real-time monitoring information.

[0156] Load balancing algorithm: Use advanced load balancing algorithm (Least Connections) to optimize task distribution and ensure balanced resource usage.

[0157] Dynamic adjustment: Dynamically adjust task allocation based on prediction results and load balancing algorithms, and prioritize new tasks to devices or units with lower loads. This strategy is designed to prevent single-point overload and improve overall system stability and response speed.

[0158] (2) Software Simulation

[0159] Loading and unloading system simulation: simulate the liquid cargo loading and unloading process of liquefied cargo ships, and ensure the safety of ship loading and unloading by controlling the loading and unloading rate, high-load compressor return air rate, and shore station return air rate.

[0160] Full-process stability simulation: It runs through the entire process after the ship is under attack, including the loading and unloading of liquid cargo and the navigation of the ship.

[0161] Three-dimensional simulated vision on the cabin roof: Apply virtual reality technology to the cabin roof of the liquid cargo ship loading and unloading simulator to achieve three-dimensional visual reality and enhance the physical reality of human-computer interaction.

[0162] (3) Resource management optimization

[0163] Resource pool mechanism: Use the resource pool mechanism to manage system resources to avoid memory leaks and unnecessary resource consumption.

[0164] Garbage collection mechanism: Implement an efficient garbage collection mechanism to clean up useless resources and ensure the stability and efficient operation of the system.

[0165] Resource monitoring: Monitor system resource usage in real time, and promptly detect and handle resource waste or abnormal usage.

[0166] (4) Parallel computing and acceleration technology

[0167] Parallel computing: In the closed-loop simulation calculation of the host, parallel computing technology (such as multi-threading or multi-process) is used to increase the computing speed and reduce the task execution time.

[0168] Hardware acceleration: Use the hardware acceleration library (CUDA) to accelerate computing tasks and further optimize system response time. According to computing requirements, select the appropriate acceleration solution for optimization.

[0169] Optimize algorithms: Apply efficient parallel algorithms to improve computing performance and ensure that the system can handle a large number of concurrent tasks without degrading performance.

[0170] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A load balancing control method for a liquefied gas tanker liquid cargo operating system simulator, characterized in that: The following steps are involved: S10. Establishing a tank pressure response model: constructing a tank pressure response model according to the cargo loading amount and the vapor discharge amount; The pressure change data during the loading process is measured experimentally to determine the parameters of the tank pressure response model; the tank pressure response model considers the gain coefficient, time constant, loading rate, return air rate and initial pressure to predict the tank pressure change; S20. Establishing a ship status monitoring model: constructing a ship status monitoring model including transverse stability, longitudinal stability, inclination and draft parameters; the ship status monitoring model considers the impact of changes in the weight of liquid cargo or ballast water on various ship parameters, and evaluates the buoyancy, stability and strength of the ship through real-time calculation, so as to provide safety guarantee for liquid cargo operation; S30, deploying a real-time load monitoring module: deploying a real-time load monitoring module in the liquefied gas tanker liquid cargo operating system system to collect the central processing unit utilization rate, memory occupancy rate, network bandwidth utilization rate and disk read and write speed performance data of each device or unit; obtaining a comprehensive performance index through weighted calculation to provide basic data support for subsequent load balancing; S40, data preprocessing and load evaluation: preprocessing the collected performance data, including data cleaning and normalization; using the load evaluation function to calculate the load value of each device or unit, considering the performance index, load change rate and random disturbance factors, and comprehensively evaluating the working status of each component of the liquefied gas tanker liquid cargo operation system; S50, task priority calculation and dynamic adjustment: calculate the priority value of each task according to the urgency, importance, resource requirements and waiting time of the task; dynamically adjust the task priority taking into account time decay factors and random fluctuations to ensure that key tasks are processed in a timely manner and improve the overall efficiency of the liquefied gas tanker liquid cargo operation system; S60, Load trend prediction: Use the long short-term memory neural network model to predict future load trends based on historical load data; capture the long-term and short-term dependencies of load changes through the calculation of forget gates, input gates, output gates, and unit states, and provide forward-looking guidance for the formulation of load balancing strategies; S70, formulation of load balancing strategy: based on the load balancing degree calculation formula, evaluate the current load distribution of the liquefied gas tanker liquid cargo operating system; consider the load difference of each device or unit and the random disturbance factor, formulate a load balancing strategy, achieve the optimal configuration of the system resources of the liquefied gas tanker liquid cargo operating system, and improve the overall operation efficiency; S80, real-time scheduling and load adjustment: using the load change detection function, real-time monitoring of the load changes of the liquefied gas tanker liquid cargo operating system; by calculating the first-order difference and second-order derivative of the load, rapid response to load mutations, dynamic adjustment of task allocation and resource allocation, to ensure that the liquefied gas tanker liquid cargo operating system maintains stable and efficient operation under various working conditions.

2. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: The liquefied gas ship liquid cargo operating system simulator adopts a three-layer architecture design, including a host computer closed loop, a host closed loop and a front-end closed loop; the host computer closed loop is used to display the human-machine interface of the liquefied gas ship loading and unloading system and simulate data display and response; the host closed loop is used to simulate the liquid cargo loading and unloading functions and internal unit simulation according to the instructions and data issued by the host computer; the front-end closed loop is used to perform data interaction with real unit equipment of the liquefied gas ship loading and unloading system simulator.

3. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 2 is characterized in that: The liquefied gas tanker liquid cargo operating system simulator also includes an architecture interaction mechanism, which includes: a data communication protocol, which is used to ensure that the data transmission between the upper computer closed loop, the host closed loop and the front-end closed loop is accurate; real-time data processing, which is used to ensure that the data processing and response time of each layer meet the system requirements; a synchronization mechanism, which is used to realize timestamp-based data synchronization to ensure data consistency between the upper computer, the host and the front end.

4. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: The liquefied gas ship liquid cargo operating system simulator consists of a liquefied natural gas ship cargo hold simulation unit, a liquefied petroleum gas ship simulation unit, a reliquefaction system simulation unit, a nitrogen production system simulation unit and a communication module unit; it adopts a combined structure, connects the industrial control host, data transmission module, and power supply system together through a bus, supports the selection of multiple data transmission modules, and realizes communication with the host computer through an Ethernet interface.

5. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: Each unit of the liquefied gas tanker liquid cargo operating system simulator adopts a physical interface, and the transmitted signals are all real physical signals; the virtualization unit runs in the software system, receives real input signals through data drive or simulation drive, and generates a response signal at the corresponding interface.

6. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: The tank pressure response model is specifically expressed as follows: <h2 style=";text-align:left;direction:ltr">[z<h2 style=";text-align:left;direction:ltr"> p <h2 style=";text-align:left;direction:ltr"> ]=[M1 / (T1s+1)M2 / (T2s+1)][v1;v2]+[c<h2 style=";text-align:left;direction:ltr"> p <h2 style=";text-align:left;direction:ltr"> ]; In the formula, z p is the tank pressure in KPa; M1, M2 are gain coefficients, dimensionless, and M1+M2=1; T1, T2 are time constants in seconds, and T1+T2=T ε ; v1 is the loading rate, in m 3 / h; v2 is the return air rate, unit is m 3 / h;c p is the initial pressure of the liquid tank, in KPa; s is the Laplace operator.

7. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: The ship status monitoring model is specifically expressed as follows: Wherein, GM1 is the new transverse stability height, in m; GM is the original transverse stability height of the ship, in m; m is the weight of the change in liquid cargo or ballast water, in t; D is the displacement of the ship, in t; d m is the average draft of the ship, in m; z m is the vertical coordinate of the changing position of liquid cargo or ballast water, in meters; ρ x is the density of each liquid, in t / m 3 ; i x is the moment of inertia of each liquid surface area, in m 4 ; GM L1 is the new longitudinal stability height, in m; GM L is the original longitudinal stability height of the ship, in m; θ is the heel angle, in rad; φ is the pitch angle, in rad; y m is the transverse coordinate of the changing position of liquid cargo or ballast water, in meters; m is the longitudinal coordinate of the changing position of liquid cargo or ballast water, in meters; F is the longitudinal coordinate of the drift center, in m; d F is the original bow draft of the ship, in m; d A is the original stern draft of the ship, in m; d F1 is the new bow draft of the ship, in m; d A1 is the new stern draft of the ship, in m; L is the length of the ship, in m; TPC is the tons of ship draft per centimeter, in t / cm; MTC is the longitudinal moment of the ship per centimeter in the initial state, in t·m / cm.

8. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: The performance indicator calculation formula of the real-time load monitoring module is as follows: P i =αC i +βM i +γN i +δD i +e i ; Where P i is the comprehensive performance index of the ith device or unit, dimensionless; C i is the CPU usage, the value range is 0-100%; M i N is the memory usage, ranging from 0 to 100%; i is the network bandwidth utilization, the value range is 0-100%; D i is the disk read and write speed, in MB / s; α, β, γ, δ are weight coefficients, dimensionless, and α+β+γ+δ=1; ε i is the error term; The load evaluation function is specifically expressed as follows: Where, L i is the load evaluation value of the ith device or unit, dimensionless; P ij is the jth performance index of the ith device or unit, dimensionless; w j is the weight of the jth performance indicator, dimensionless, and n is the number of performance indicators; λ is the weight coefficient of the load change rate, dimensionless; is the rate of change of the performance index, in units of 1 / s; μ i is a random disturbance term.

9. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1 is characterized in that: The task priority calculation formula is specifically expressed as follows: In the formula, PR k is the priority value of the kth task, dimensionless; U k is the urgency of the task, ranging from 1 to 10; I k is the task importance, ranging from 1 to 10; R k is the task resource requirement, ranging from 1 to 10; T k is the task waiting time, in seconds; ω1, ω2, ω3, ω4 are dimensionless weight coefficients, and ω1+ω2+ω3+ω4=1; λ is the time attenuation coefficient, in 1 / s; ξ k is a random fluctuation term, which obeys normal distribution.

10. The load balancing control method of the liquefied gas tanker liquid cargo operating system simulator according to claim 1, characterized in that: The load balancing degree calculation formula is specifically expressed as follows: Where B is the load balance degree, dimensionless, ranging from 0 to 1. The closer to 1, the more balanced the load. L i is the load of the ith device or unit, dimensionless; is the average load, dimensionless; m is the total number of devices or units; η is a random disturbance term, which obeys a normal distribution; the load change detection function is specifically expressed as follows: Where D(t) is the load change detection function, dimensionless; L i (t) is the load of the ith device or unit at time t, dimensionless; m is the total number of devices or units; κ is the second-order derivative weight coefficient, dimensionless, usually ranging from 0.1 to 0.5; It is the second-order derivative of the load, indicating the acceleration of the load change, in units of 1 / s 2 ; ρ i is a random fluctuation term, which obeys the uniform distribution U(-b,b), where b is the fluctuation amplitude.

Images