An automatic mixing intelligent control method for a marine fuel filling synergist

Abstract

This application discloses an automatic mixing and intelligent control method for marine fuel refueling enhancers, belonging to the field of marine fuel refueling technology. Based on the relationship between the main fuel flow fluctuation amplitude and a preset threshold, a PID control algorithm or fuzzy control algorithm is adopted. The atomization pressure threshold is calibrated based on the enhancer viscosity and nozzle diameter. The air pressure of the pneumatic diaphragm pump is adjusted first to meet the threshold, and then the opening of the adjustable proportional pneumatic regulating valve is fine-tuned to match the fuel flow ratio. When the fuel flow fluctuation reaches the compensation threshold, a gradual proportional correction strategy is initiated. When any core parameter deviates from the set range, a linkage control is triggered based on linear interpolation. After accumulating multiple batches of data, a dedicated operating condition parameter library is generated through improved K-means clustering and multiple linear regression analysis, achieving automatic matching of optimal control parameters when starting a new operating condition. This method achieves precise adaptive control of the enhancer ratio, efficient mixing uniformity, comprehensive safety protection, and intelligent closed-loop iteration of data.

Description

Technical Field

[0001] This application discloses an automatic mixing and intelligent control method for marine fuel filling enhancers, belonging to the field of marine fuel filling technology. Background Technology

[0002] Fuel cleaning and enhancement agents, as key additives for improving combustion efficiency and reducing fuel consumption and pollutant emissions, are increasingly widely used in marine fuel refueling. However, the energy-saving and emission-reduction effects of these agents highly depend on their uniform mixing with high-viscosity heavy fuel oil at the molecular level. Existing refueling technologies have significant shortcomings in terms of proportion control precision, mixing uniformity, safety protection, and adaptability to operating conditions, making it difficult to meet the comprehensive demands of modern shipping for high efficiency, environmental protection, and safety.

[0003] Among them, the automatic addition technology for marine fuel cleaning and enhancement agents focuses on achieving precise synchronous injection and efficient fusion of the enhancement agent during fuel transportation. This technology needs to take into account the high viscosity characteristics of heavy fuel oil used in inland waterways and ocean-going vessels, and be adapted to complex operating scenarios such as fixed refueling stations at docks or mobile refueling vessels. At the same time, it must comply with special safety specifications such as marine explosion-proof and backflow prevention.

[0004] Existing technologies generally employ fixed-ratio or open-loop control methods to inject enhancers into the main pipeline, lacking real-time sensing and dynamic response capabilities to fluctuations in the main fuel flow rate. This results in addition ratio errors often exceeding ±10%, severely impacting the enhancement effect. Furthermore, the single injection point and simple mixing structure cannot overcome the poor flowability of high-viscosity fuels, easily leading to oil-enhancer stratification and low mixing uniformity. In addition, safety designs largely rely on one-way valves to prevent backflow, lacking a dual protection mechanism linked to pressure, posing a risk of fuel backflow from the main pipeline contaminating the additive storage tank. Moreover, the absence of filter clogging warning functions makes it susceptible to flow control failure or even equipment shutdown due to increased pressure differential. These shortcomings render existing devices unsuitable for diverse fuel types, dynamic filling conditions, and harsh marine environments, exhibiting poor versatility, low reliability, and untraceable data. There is an urgent need for an automated intelligent mixing and control method and system for marine fuel enhancers that integrates precise regulation, efficient mixing, multiple protections, and intelligent closed-loop control. Summary of the Invention

[0005] The main purpose of this application is to provide a solution to the problems in the prior art.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] The purpose of this invention is to provide an automatic mixing and intelligent control method and system for marine fuel additives, which can effectively solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] An automated intelligent control method for mixing marine fuel additives includes the following steps:

[0010] Construct a main pipeline system, an auxiliary pipeline system, a monitoring and control system, and a hybrid enhancement system. Input operating condition data through a human-machine interface, and automatically call and preset the optimal control parameters matched in the self-learning parameter library of operating conditions, and perform self-testing of sensors and actuators.

[0011] Core parameters are collected synchronously at a predetermined sampling period, the main fuel flow fluctuation amplitude is calculated for multiple consecutive sampling periods, and a PID control algorithm or a fuzzy control algorithm is decided based on the relationship between the main fuel flow fluctuation amplitude and a preset fluctuation threshold.

[0012] The atomization pressure threshold is calibrated based on the viscosity of the enhancer and the nozzle diameter. When the actual atomization pressure in the auxiliary pipeline is lower than the atomization pressure threshold, the intake pressure of the pneumatic diaphragm pump is adjusted first until the target is met. After the target is met, the opening of the adjustable proportional pneumatic regulating valve is finely adjusted to match the fuel flow ratio.

[0013] When the fluctuation range of fuel flow reaches the preset compensation threshold range, the gradual proportional correction strategy is activated, and the enhancer flow is adjusted according to the first-order inertial link formula.

[0014] When any of the core parameters deviates from the set range, a linkage control measure is triggered based on the linear interpolation method.

[0015] The system records all parameter data at predetermined time intervals to form an operation log. After accumulating multiple batches of data, it generates a dedicated operating condition parameter library through improved K-means clustering and multiple linear regression analysis. When a new operating condition is started, it automatically matches the optimal control parameters with a similarity greater than or equal to the preset similarity threshold.

[0016] Furthermore, the main pipeline system is used to transport marine fuel oil, including a main fuel oil flow sensor and a mixing section;

[0017] The auxiliary pipeline system is a delivery channel for the synergist, including quick connectors, precision filters, one-way isolation valves, pneumatic diaphragm pumps, mass flow meters, adjustable proportional pneumatic regulating valves, and check valves. The precision filter adopts a stainless steel sintered mesh structure and is equipped with a differential pressure transmitter. The pneumatic diaphragm pump is a marine explosion-proof model, and the flow rate is coarsely adjusted by adjusting the inlet air pressure. The one-way isolation valve is connected in series with a pressure sensor to form a double backflow protection.

[0018] The monitoring and control system includes a central control unit, a human-machine interface, a closed-loop feedback module, and an alarm module.

[0019] The mixing enhancement system includes a multi-point symmetrical bypass mixing nozzle group evenly arranged along the circumference of the mixing section of the main pipeline, a helical blade static mixer installed downstream of the additive injection point, and a mixing section heat tracing and insulation layer. The nozzle outlet of the multi-point symmetrical bypass mixing nozzle group forms a predetermined angle with the inner wall of the main pipeline and faces the fuel flow direction. The helical blade static mixer has a predetermined number of blades, blade angle, and blade spacing.

[0020] Furthermore, the central control unit is communicatively connected to the main fuel flow sensor, mass flow meter, pressure sensor, temperature sensor, differential pressure transmitter, and actuator;

[0021] The central control unit has a built-in proportional addition algorithm based on PID or fuzzy control. Based on the real-time data transmitted by the fuel flow sensor in the main pipeline and the addition ratio set by the human-machine interface, it automatically calculates the target flow rate of the enhancer. Combined with the enhancer flow rate in the auxiliary pipeline, pipeline pressure, and filter differential pressure data, it outputs control signals to the pneumatic diaphragm pump to adjust the intake pressure and the adjustable proportional pneumatic regulating valve to achieve matching of the enhancer flow rate.

[0022] Furthermore, the step of synchronously collecting multiple core parameters at a predetermined sampling period, calculating the main fuel flow fluctuation amplitude for multiple consecutive sampling periods, and deciding whether to adopt a PID control algorithm or a fuzzy control algorithm based on the relationship between the main fuel flow fluctuation amplitude and a preset fluctuation threshold, further includes:

[0023] The central control unit collects main fuel flow, enhancer flow, main pipeline pressure, auxiliary pipeline atomization pressure, mixing section temperature, and filter pressure differential data in real time.

[0024] Calculate the flow fluctuation amplitude over multiple consecutive sampling periods based on the current flow and average flow.

[0025] When the flow rate fluctuation is less than the preset fluctuation threshold, it is determined to be a steady-state refueling condition and the PID control algorithm is activated. When the flow rate fluctuation is greater than or equal to the preset fluctuation threshold, it is determined to be a dynamic refueling condition and the fuzzy control algorithm is automatically switched.

[0026] Furthermore, the step of calibrating the atomization pressure threshold based on the enhancer viscosity and nozzle diameter, and prioritizing adjusting the pneumatic diaphragm pump intake pressure until the threshold is reached when the actual atomization pressure in the auxiliary pipeline is lower than the threshold, and then finely adjusting the opening of the adjustable proportional pneumatic regulating valve to match the fuel flow ratio after reaching the threshold, also includes:

[0027] The calibrated pressure threshold is determined based on the synergist viscosity and nozzle diameter. calibrate the atomization pressure threshold to ensure the atomization particle size range;

[0028] Detect the pressure in real time, collect the actual value of the atomization pressure of the auxiliary pipeline, and calculate the pressure deviation in combination with the atomization pressure threshold

[0029] Hierarchical regulation. If the pressure deviation or the actual value of the atomization pressure of the auxiliary pipeline is less than the atomization pressure threshold, only adjust the intake pressure of the pneumatic diaphragm pump, and configure the adjustment amount until the actual value of the atomization pressure of the auxiliary pipeline is greater than or equal to the atomization pressure threshold;

[0030] If the pressure deviation or the actual value of the atomization pressure of the auxiliary pipeline is greater than or equal to the atomization pressure threshold, keep the intake pressure of the pump body unchanged, adjust the opening of the adjustable proportional pneumatic control valve, and configure the opening correction amount until the actual flow rate of the synergist matches the set ratio.

[0031] Furthermore, when the fluctuation range of the fuel flow rate reaches within the preset compensation threshold range, start the progressive proportional correction strategy, and adjust the synergist flow rate according to the first-order inertia link formula, which further includes:

[0032] Detect the fluctuation range of the fuel flow rate in real time. When the fluctuation range of the fuel flow rate is greater than or equal to the compensation threshold, trigger the compensation mode;

[0033] Calculate the correction coefficient and time constant according to the progressive adjustment using the first-order inertia link, and determine the progressive adjustment curve;

[0034] Update the correction value of the synergist flow rate at the preset sampling period, and synchronously collect the actual flow rate feedback until the fluctuation range of the fuel flow rate is less than the compensation threshold, and then exit the compensation mode;

[0035] Record the fluctuation range of the fuel flow rate, adjustment curve, and stable time data of this fluctuation to form a dynamic compensation curve library.

[0036] Furthermore, when any of the core parameters deviates from the set range, trigger the linkage control measures according to the linear interpolation method, including:

[0037] When any of the core parameters is abnormal, the control unit automatically triggers the linkage control;

[0038] Based on the principle of first compensation, then warning, and then shutdown, formulate a linkage control plan for the abnormality of the core parameter. The correction amount of all control parameters is calculated by the linear interpolation method, and the formula is:

[0039] ;

[0040] Wherein, is the parameter correction amount, is the actual value, is the set value, This is the alarm threshold;

[0041] The maximum correction amount is limited by the device's hardware parameters.

[0042] Furthermore, the process of recording full parameter data at predetermined time intervals to form an operation log, accumulating multiple batches of injection data, and generating a dedicated operating condition parameter library through improved K-means clustering and multiple linear regression analysis, automatically matching the optimal control parameters with a similarity greater than or equal to a preset similarity threshold when a new operating condition is started, also includes:

[0043] Feature parameters were extracted, and oil type, ambient temperature, fuel viscosity, mixing section temperature, atomization pressure threshold, addition ratio, flow fluctuation compensation threshold, PID / KP parameters, and fuzzy rule base correction coefficient were selected as clustering samples.

[0044] Based on K-means clustering, the accumulated multi-batch refueling data are divided into heavy fuel oil-low temperature condition, heavy fuel oil-normal temperature condition, and diesel-all temperature range condition. The cluster centers are typical features of each type of condition.

[0045] A multiple linear regression model was established using proportionality error, mixing uniformity, and settling time as evaluation indicators to correlate the evaluation indicators with the control parameters.

[0046]

[0047] in As evaluation indicators, To adjust parameters, For regression coefficients, For constant terms;

[0048] To find the optimal parameters, constraints are constructed based on proportional error, mixing uniformity, and settling time. The optimal solution of the regression model is then obtained to determine the optimal control parameters for various operating conditions.

[0049] The parameter library is updated, and after adding new batches of data, the clustering and regression are re-performed for each preset number of batches to update the optimal parameter library and achieve self-learning iteration.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] Precise adaptive proportional control capability

[0052] Breaking through the limitations of traditional fixed ratios: Through the adaptive switching mechanism of PID / fuzzy dual algorithms, combined with the flow fluctuation compensation strategy, the ratio error is stably controlled within a preset narrow range, which is significantly improved compared with existing technologies; the wide range ratio setting supports flexible adjustment and fully adapts to the efficiency improvement needs of different types of oil such as heavy fuel oil and diesel.

[0053] Highly efficient and reliable mixing uniformity guarantee

[0054] The innovative three-stage mixing enhancement system relies on the synergistic effect of multi-point annular atomization injection, spiral blade static mixer, and heat tracing and insulation to achieve a preset high level of mixing uniformity, completely solving the problem of oil-additive stratification of high-viscosity fuel and enhancer; the coordinated control of atomization pressure ensures that the atomized particle size is stable within the preset micron range, creating ideal conditions for molecular-level fusion.

[0055] Comprehensive and multi-layered security protection mechanisms

[0056] Triple safety features are implemented: the pneumatic diaphragm pump is designed to be free of electrical sparks, meeting marine explosion-proof standards; the one-way isolation valve and pressure sensor provide dual backflow protection, eliminating the risk of tank contamination; and the filter differential pressure warning and multi-parameter linkage control prevent equipment downtime and ensure continuous filling operations.

[0057] Intelligent closed-loop data traceability and optimization

[0058] Achieve intelligent iteration throughout the entire process: Automatically record full parameter data to support energy efficiency audits; The self-learning algorithm for operating conditions achieves one-time debugging and lifelong adaptation through clustering and regression analysis, greatly improving the ease of operation; Closed-loop feedback and adaptive correction ensure long-term stable operation of the system, meeting the comprehensive needs of modern shipping for efficiency, environmental protection and safety. Attached Figure Description

[0059] Figure 1 A flowchart illustrating the workflow of an intelligent control method for automatic mixing of marine fuel additives, as claimed in an embodiment of the present invention.

[0060] Figure 2 This is a second workflow diagram of an intelligent control method for automatic mixing of marine fuel additives, as claimed in an embodiment of the present invention.

[0061] Figure 3 The third workflow diagram is shown for an intelligent control method for automatic mixing of marine fuel additives, as claimed in an embodiment of the present invention.

[0062] Figure 4 The fourth workflow diagram is shown for an intelligent control method for automatic mixing of marine fuel additives, as claimed in an embodiment of the present invention.

[0063] Figure 5 The fifth flowchart is shown for an intelligent control method for automatic mixing of marine fuel additives, as claimed in an embodiment of the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0065] The terms "first," "second," and "third" in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of those features. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationships and movements between components in a specific orientation (as shown in the figures). If the specific orientation changes, the directional indications also change accordingly. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0066] References to embodiments herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0067] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0068] The purpose of this invention is to provide an automatic mixing and intelligent control method for marine fuel additives, which can effectively solve the problems mentioned in the background art.

[0069] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0070] According to the first embodiment of the present invention, referring to Figure 1 This invention claims protection for an intelligent control method for automatic mixing of marine fuel additives, comprising the following steps:

[0071] S1 constructs a main pipeline system, an auxiliary pipeline system, a monitoring and control system, and a hybrid enhancement system. By inputting operating condition data through a human-machine interface, it automatically calls and presets the optimal control parameters matched in the self-learning parameter library of operating conditions and performs self-testing of sensors and actuators.

[0072] S2, synchronously collect core parameters at a predetermined sampling period, calculate the main fuel flow fluctuation amplitude for multiple consecutive sampling periods, and decide whether to adopt a PID control algorithm or a fuzzy control algorithm based on the relationship between the main fuel flow fluctuation amplitude and the preset fluctuation threshold.

[0073] S3. The atomization pressure threshold is calibrated according to the viscosity of the enhancer and the nozzle diameter. When the actual atomization pressure of the auxiliary pipeline is lower than the atomization pressure threshold, the intake pressure of the pneumatic diaphragm pump is adjusted first until the target is met. After the target is met, the opening of the adjustable proportional pneumatic regulating valve is finely adjusted to match the fuel flow ratio.

[0074] S4. When the fluctuation range of fuel flow reaches the preset compensation threshold range, the gradual proportional correction strategy is activated to adjust the enhancer flow according to the first-order inertial link formula.

[0075] S5, when any of the core parameters deviates from the set range, a linkage control measure is triggered based on the linear interpolation method;

[0076] S6 records full parameter data at predetermined time intervals to form an operation log. After accumulating multiple batches of data, it generates a dedicated working condition parameter library through improved K-means clustering and multiple linear regression analysis. When a new working condition is started, it automatically matches the optimal control parameters with a similarity greater than or equal to the preset similarity threshold.

[0077] Furthermore, the main pipeline system is used to transport marine fuel oil, including a main fuel oil flow sensor and a mixing section;

[0078] The auxiliary pipeline system is a delivery channel for the synergist, including quick connectors, precision filters, one-way isolation valves, pneumatic diaphragm pumps, mass flow meters, adjustable proportional pneumatic regulating valves, and check valves. The precision filter adopts a stainless steel sintered mesh structure and is equipped with a differential pressure transmitter. The pneumatic diaphragm pump is a marine explosion-proof model, and the flow rate is coarsely adjusted by adjusting the inlet air pressure. The one-way isolation valve is connected in series with a pressure sensor to form a double backflow protection.

[0079] The monitoring and control system includes a central control unit, a human-machine interface, a closed-loop feedback module, and an alarm module.

[0080] The mixing enhancement system includes a multi-point symmetrical bypass mixing nozzle group evenly arranged along the circumference of the mixing section of the main pipeline, a helical blade static mixer installed downstream of the additive injection point, and a mixing section heat tracing and insulation layer. The nozzle outlet of the multi-point symmetrical bypass mixing nozzle group forms a predetermined angle with the inner wall of the main pipeline and faces the fuel flow direction. The helical blade static mixer has a predetermined number of blades, blade angle, and blade spacing.

[0081] Furthermore, the central control unit is communicatively connected to the main fuel flow sensor, mass flow meter, pressure sensor, temperature sensor, differential pressure transmitter, and actuator;

[0082] The central control unit has a built-in proportional addition algorithm based on PID or fuzzy control. Based on the real-time data transmitted by the fuel flow sensor in the main pipeline and the addition ratio set by the human-machine interface, it automatically calculates the target flow rate of the enhancer. Combined with the enhancer flow rate in the auxiliary pipeline, pipeline pressure, and filter differential pressure data, it outputs control signals to the pneumatic diaphragm pump to adjust the intake pressure and the adjustable proportional pneumatic regulating valve to achieve matching of the enhancer flow rate.

[0083] In this embodiment, the main pipeline system is used to transport marine fuel oil (including heavy fuel oil, diesel oil and other marine oil products). The pipe diameter is 150mm (100-200mm specifications can be adapted according to the filling flow requirements). The design working pressure is 0.5MPa (pressure resistance limit ≥1.0MPa). The normal working temperature is controlled at 35±5℃. Both ends adopt GB / T 9119-2010 standard marine universal flange interfaces. The sealing material is fluororubber, which is suitable for the harsh working conditions of high temperature, high humidity and high salt spray on ships. The high-pressure filling hoses are connected to the oil outlet of the filling ship and the oil inlet of the receiving ship respectively.

[0084] Along the fuel flow direction on the main pipeline, a main fuel flow sensor, a mixing section, a pressure gauge, and a shut-off valve are installed sequentially. The main fuel flow sensor is preferably a Coriolis type (optimal for high-viscosity fuels), with an ultrasonic type as an alternative. It has a measurement range of 50-500 m³ / h, an accuracy of ≤±0.5%, and features resistance to media viscosity interference and anti-scaling capabilities. It can accurately collect real-time flow data of high-viscosity fuel and transmit it to the central control unit in real-time via a 4-20mA analog signal. The mixing section is made of 316L stainless steel, with a length of 1.5-2.0m. The inner wall is polished (roughness Ra≤1.6μm), and the outer wall is equipped with an electric heating insulation layer (power 500-800W). It is equipped with a temperature sensor and a temperature controller to achieve constant temperature heating (temperature fluctuation ≤±2℃), effectively reducing fuel viscosity, improving the fusion ability of fuel and additives, and avoiding uneven mixing caused by fuel adhering to the walls in low-temperature environments.

[0085] The auxiliary pipeline system serves as the delivery channel for the synergist. The pipe diameter is one-fifth that of the main pipeline (i.e., 30mm, compatible with a 150mm main pipeline). The material is also 316L stainless steel. Along the flow direction of the synergist, the pipeline is installed in sequence as follows: synergist delivery pipeline quick connector (selected according to ISO 7241-B standard, quick plug-in and reliable sealing), 5-10μm precision filter (the filtration accuracy can be adjusted according to the type of synergist, the filter screen is made of stainless steel sintered mesh, which can be disassembled, washed and reused), one-way isolation valve (adopting a lifting structure, opening pressure ≤0.1MPa, excellent sealing performance), pneumatic diaphragm pump, high-precision mass flow meter for synergist, adjustable proportional pneumatic regulating valve and check valve.

[0086] The pneumatic diaphragm pump adopts the QBY type marine explosion-proof pneumatic diaphragm pump, with a flow range of 0.1-5 m³ / h and a head ≥30 m. It is driven by a pneumatic motor, generating no electrical sparks, and its explosion-proof rating reaches Ex d II BT4 Gb, fully meeting marine explosion-proof requirements. Coarse flow adjustment can be achieved by adjusting the inlet air pressure (0.2-0.6 MPa). A one-way isolation valve and a pressure sensor (measuring range 0-1.0 MPa, accuracy ≤±0.5%) are connected in series to form a double backflow protection. When the main pipeline pressure rises abnormally (exceeding 0.6 MPa), the pressure sensor immediately sends a signal to the central control unit. The control unit quickly shuts down the pneumatic diaphragm pump and the adjustable proportional pneumatic regulating valve, while the one-way isolation valve automatically closes, completely eliminating the risk of backflow of fuel in the main pipeline and preventing the accumulation of performance enhancers. The precision filter is equipped with a differential pressure transmitter. When the filter screen becomes clogged and the inlet-outlet pressure difference is ≥0.15MPa, it automatically sends a differential pressure alarm signal to the human-machine interface to remind the operator to clean or replace the filter screen in time to avoid proportional loss of control due to flow obstruction. The adjustable proportional pneumatic regulating valve adopts a linear stroke structure with an adjustment accuracy of ≤±1% and an adjustment range of 0-100%. It receives a 4-20mA control signal from the central control unit and can realize continuous and precise adjustment of the synergist flow rate. It supports two modes: fixed ratio setting and stepless adjustment to meet the needs of different working conditions.

[0087] An optional additive preheating and viscosity-reducing module can be added to the auxiliary pipeline. This module uses electric heat tracing or a heat exchange jacket (electric heat tracing is preferred for its ease of installation), with a power of 300-500W and equipped with a temperature controller. The temperature is strictly controlled at ≤60℃ (to prevent high-temperature failure of the synergist), with temperature fluctuations ≤±2℃. This effectively reduces the flow resistance of high-viscosity synergists and improves their atomization effect, making it particularly suitable for low-temperature environments or applications using high-viscosity synergists. The auxiliary pipeline is connected to the mixing section of the main pipeline at its end, enabling precise injection of the synergist.

[0088] The monitoring and control system is the core control unit of the device, integrating a central control unit, a human-machine interface, a closed-loop feedback module, and an alarm module. All modules work together to achieve precise control, data recording, and early warning of anomalies.

[0089] The central control unit adopts a Siemens S7-1200 series marine PLC controller (or an equivalent MCU controller), with an IP65 protection rating, suitable for harsh marine environments. It features a built-in proportional addition algorithm based on PID or fuzzy control (automatically switching according to operating conditions; PID control is used in steady state, and fuzzy control is used during dynamic fluctuations), with a calculation cycle of ≤100ms. This algorithm automatically calculates the target enhancer flow rate based on real-time data transmitted from the main fuel flow sensor and the addition ratio set via the human-machine interface. Simultaneously, it combines data such as enhancer flow rate in the auxiliary pipeline, pipeline pressure, and filter differential pressure to output control signals to the pneumatic diaphragm pump (to adjust the intake pressure) and the adjustable proportional pneumatic regulating valve (to adjust the opening), achieving precise matching of the enhancer flow rate. Furthermore, the algorithm has a flow fluctuation compensation mechanism. When the fuel flow rate is unstable during the initial addition phase (fluctuation amplitude ≥10%), the compensation mode is automatically activated to slowly adjust the enhancer flow rate, avoiding sudden changes in the ratio and achieving a smooth transition, ensuring that the ratio error remains stable within ≤±2%.

[0090] The human-machine interface uses a 7-inch industrial-grade touchscreen with an IP65 protection rating. It supports bilingual operation in Chinese and English and features a simple layout, including four modules: parameter setting area, real-time monitoring area, data query area, and alarm area. The parameter setting area supports manual input of fixed addition ratios within the range of 500:1 to 20000:1 and is equipped with a stepless adjustment knob (adjustment accuracy 1:1). Operators can adjust the ratio parameters in real time according to the type of oil and the model of the synergist. Simultaneously, the flow fluctuation compensation threshold (default 5%-15% adjustable), filter differential pressure alarm threshold (default 0.15MPa), and pipeline pressure alarm threshold (default 0.6MPa) can be set. The real-time monitoring area... The system displays key parameters such as main fuel flow, enhancer flow, mixing ratio, pipeline pressure, mixing section temperature, and enhancer preheating temperature in real time, providing a clear view of the unit's operating status. The data query area automatically records data throughout the entire refueling process (storage capacity ≥ 100,000 records, storage period ≥ 1 year), supports queries by date and refueling batch, and allows data export via USB flash drive (Excel format), providing data support for subsequent ship energy efficiency audits and emission reduction effect assessments. The alarm area displays various abnormal alarms in real time, such as excessive differential pressure, abnormal pressure, excessive flow, and abnormal temperature, while simultaneously issuing audible and visual alarms with an alarm volume ≥ 80dB and a solid red alarm light to remind operators to handle the situation promptly.

[0091] The closed-loop feedback module consists of various sensors, including a main fuel flow sensor, an enhancer mass flow meter, a pressure sensor, a temperature sensor, a differential pressure transmitter, and signal transmission lines. It collects real-time operating data from the main pipeline, auxiliary pipelines, and the mixing enhancement system, transmitting this data to the central control unit for comparison with set values. If a deviation occurs (proportional deviation > ±2%, pressure deviation > ±0.05MPa, temperature deviation > ±2℃, etc.), the control unit immediately adjusts the control signal until the parameters return to the set range, forming a closed-loop control system of acquisition-comparison-adjustment-feedback to ensure long-term stable operation of the device. The alarm module is linked to the closed-loop feedback module. When abnormal data occurs, an alarm is immediately triggered, recording the alarm time, alarm type, and abnormal parameters. Simultaneously, it can automatically take corresponding protective measures based on the severity of the abnormality (minor abnormalities only trigger an alarm; severe abnormalities shut down related equipment to prevent the fault from escalating).

[0092] The mixing enhancement system is used to improve the mixing uniformity of high-viscosity fuel and enhancer, solving the problem of uneven mixing in existing technologies. It consists of a multi-point symmetrical bypass mixing nozzle group, a spiral blade static mixer and a heat-tracing insulation layer for the mixing section, achieving a three-stage mixing effect of atomization initial mixing, shear secondary mixing and constant temperature enhanced fusion.

[0093] The connection between the auxiliary pipeline and the main pipeline adopts a multi-point annular injection design. 4-8 (preferably 6) micro-orifice nozzles with a diameter of 1-2 mm (preferably 1.5 mm) are evenly arranged along the circumference of the mixing section of the main pipeline. The nozzles are made of 316L stainless steel and adopt an atomization structure (atomization particle size 50-100μm). The nozzle outlet is at a 30° angle to the inner wall of the main pipeline and faces the fuel flow direction to ensure that the enhancer is evenly injected into the inner wall of the main pipeline in an atomized state. The shear force generated by the fuel flow is used to achieve the initial diffusion and mixing of the enhancer and fuel, avoiding local accumulation caused by single-point injection. A 316L stainless steel spiral blade static mixer is installed 0.5-1.0 meters (preferably 0.8 meters) downstream of the additive injection point. The mixer is 0.5-0.8 meters long, with 8-12 blades at a 45° angle and a 50mm spacing. As the mixed fluid flows through, it is cut, stirred, divided, and merged by the blades, forming strong radial vortices and axial flows. This achieves secondary enhanced mixing, completely breaking down the oil-additive stratification and ensuring a mixing uniformity of ≥95%. Simultaneously, the heating and insulation layer in the mixing section continuously provides a constant temperature environment, reducing fuel viscosity, enhancing the molecular-level fusion of fuel and the additive, further improving the mixing effect, and fully leveraging the energy-saving and emission-reduction effects of the additive.

[0094] Furthermore, referring to Figure 2 S2 further includes:

[0095] S21, the central control unit collects main fuel flow, enhancer flow, main pipeline pressure, auxiliary pipeline atomization pressure, mixing section temperature and filter pressure difference main fuel flow data in real time;

[0096] S22, calculate the flow fluctuation amplitude for multiple consecutive sampling periods based on the current flow and average flow;

[0097] S23, when the flow rate fluctuation amplitude is less than the preset fluctuation threshold, it is determined to be a steady-state refueling condition and the PID control algorithm is activated; when the flow rate fluctuation amplitude is greater than or equal to the preset fluctuation threshold, it is determined to be a dynamic refueling condition and the fuzzy control algorithm is automatically switched.

[0098] In this embodiment, the central control unit collects main fuel flow data in real time, with a sampling period of ≤100ms, and calculates the flow fluctuation amplitude ΔQ over 5 consecutive sampling periods, where ΔQ = |current flow - average flow| / average flow × 100%;

[0099] When ΔQ < 5%, it is determined to be a steady-state refueling condition, and the PID control algorithm is activated; when ΔQ ≥ 5%, it is determined to be a dynamic refueling condition, such as the initial stage of refueling, fuel line pressure fluctuations, sudden changes in fuel viscosity, etc., and it automatically switches to the fuzzy control algorithm without manual intervention.

[0100] The central control unit collects main fuel flow data in real time with a sampling period of 100ms. This fixed sampling period ensures data consistency, and the flow fluctuation amplitude is calculated over five consecutive sampling periods. The calculation formula is:

[0101]

[0102] in , For the first Main fuel flow rate per sampling period, per unit .

[0103] when When the condition is determined to be a steady-state refueling condition, the PID control algorithm is activated; when When the condition is determined to be dynamic filling, the system automatically switches to a fuzzy control algorithm without requiring manual intervention.

[0104] Specific examples of sampling data are shown in Table 1;

[0105] Table 1 Sampling Data Table

[0106]

[0107] Specific criteria for differentiating dynamic refueling conditions:

[0108] Initial refueling phase: Within 0-5 minutes after refueling starts, the flow rate gradually increases from 0 to the set value, for three consecutive 5-sampling cycles. All are ≥30%, and the flow rate shows a continuous upward trend;

[0109] Fuel line pressure fluctuation: The main line pressure deviates from the set value (0.5MPa) by more than ±0.08MPa, and the flow rate fluctuation is synchronized with the pressure fluctuation. Between 5% and 20%;

[0110] Sudden change in fuel viscosity: The temperature in the mixing section remained unchanged, but the flow rate detected by the main fuel flow sensor showed irregular fluctuations. Between 5% and 15%, and lasting for more than two 5-sampling cycles;

[0111] Other dynamic operating conditions: switching of the receiving vessel's oil circuit, adjustment of the injection pump frequency, etc., with a sudden increase / decrease in flow rate ≥10%, and a single... ≥20%.

[0112] Dual-algorithm collaborative logic: The PID control algorithm is used for steady-state conditions. It uses closed-loop adjustment of proportional (P), integral (I), and derivative (D) parameters, with a proportional coefficient P = 2.5-3.5, an integral time I = 0.8-1.2s, and a derivative time D = 0.1-0.3s. This accurately offsets small flow deviations and ensures stable proportional control without fluctuations. The fuzzy control algorithm is used for dynamic conditions. It does not require the establishment of a precise mathematical model. Through a preset fuzzy rule library for flow fluctuation, pressure regulation, and opening correction, it quickly responds to sudden changes in flow. For example, when the flow suddenly increases by 10%, the pump pressure and regulating valve opening are adjusted within 0.5s, avoiding large deviations in the proportional control and solving the pain point of proportional control runaway under dynamic conditions in existing technologies.

[0113] The specific logic and calculation process of the PID control algorithm:

[0114] The PID control algorithm uses incremental PID, which is suitable for continuous regulation under steady-state conditions. The output is the opening increment of the adjustable proportional pneumatic control valve. and the increase in inlet pressure of the pneumatic diaphragm pump The core formula is:

[0115]

[0116]

[0117] in:

[0118] deviation Set the synergist flow rate Actual synergist flow rate ,unit ;

[0119] proportionality coefficient The corresponding proportionality coefficient P = 3.0, and the integral coefficient Integral time Differential coefficients Differential time ;

[0120] For the current sampling period, , For the first two sampling periods;

[0121] The intake pressure increment is 0.2 times the opening increment, achieving synchronous matching between pump body pressure and regulating valve opening.

[0122] The implementation example of the PID control algorithm is shown in Table 2. Under steady-state conditions, the enhancer flow rate is set to 0.2. ;

[0123] Table 2 PID Control Algorithm Data Table

[0124]

[0125] The fuzzy control algorithm employs two-dimensional fuzzy control. The inputs are the flow fluctuation amplitude ΔQ (fuzzy set: small S, medium M, large L, extra-large XL) and the flow change rate dq / dt (fuzzy set: negative large NB, negative small NS, zero Z, positive small PS, positive large PB). The outputs are the regulating valve opening correction ΔU (fuzzy set: negative large NB, negative small NS, zero Z, positive small PS, positive large PB) and the pump inlet pressure correction ΔP (fuzzy set: negative large NB, negative small NS, zero Z, positive small PS, positive large PB). The core steps are:

[0126] Fuzzing will combine ΔQ (0-100%) and dq / dt ( The precise values ​​of ) are mapped to fuzzy subsets using a triangular membership function;

[0127] Fuzzy reasoning is based on a pre-defined rule base.

[0128] Defuzzing uses the centroid method to convert the fuzzy output into a precise correction value, enabling rapid adjustment.

[0129] The core fuzzy rule base contains 4 × 5 = 20 rules, with key rules selected.

[0130] Table 3 Core Fuzzy Rule Table

[0131]

[0132] Example of fuzzy control algorithm, dynamic operating condition - flow rate suddenly increases by 10%;

[0133] Given that the main fuel flow rate is from 200 It surged to 220 , ΔQ=10% (medium M), dq / dt=20 (Zhengda PB), set the addition ratio to 1000:1, and set the synergist flow rate from 0.2... Rise to 0.22 .

[0134] According to the rule base, fuzzy inference gives ΔU = PB (opening correction + 5%) and ΔP = PB (pressure correction + 0.05 MPa).

[0135] The defuzzy logic is resolved to accurately output an opening increment of 5% and a pressure increment of 0.05 MPa.

[0136] The execution result was that the regulating valve opening increased from 30% to 35% within 0.5 seconds, the pump inlet pressure increased from 0.3MPa to 0.35MPa, and the actual flow rate of the synergist was 0.218. The proportional error is only ±0.9%, solving the problem of proportional loss of control under dynamic working conditions.

[0137] Furthermore, referring to Figure 3 S3 further includes:

[0138] S31, Calibrate the pressure threshold. Based on the synergist viscosity and nozzle diameter, calibrate the atomization pressure threshold to ensure the atomization particle size range.

[0139] S32, Real-time pressure detection, acquisition of actual atomization pressure values ​​in the auxiliary pipeline, and calculation of pressure deviation based on the atomization pressure threshold.

[0140] S33, Layered control: If the pressure deviation is greater than 0 or the actual value of the atomization pressure of the auxiliary pipeline is less than the atomization pressure threshold, only adjust the air inlet pressure of the pneumatic diaphragm pump and configure the adjustment amount until the actual value of the atomization pressure of the auxiliary pipeline is greater than or equal to the atomization pressure threshold.

[0141] S34, if the pressure deviation is less than or equal to 0 or the actual value of the atomization pressure of the auxiliary pipeline is greater than or equal to the atomization pressure threshold, keep the pump body air inlet pressure unchanged, adjust the opening of the adjustable proportional pneumatic regulating valve, configure the opening correction amount, until the actual flow rate of the synergist matches the set ratio.

[0142] In this embodiment, the linkage auxiliary pipeline pressure sensor, pneumatic diaphragm pump, and adjustable proportional pneumatic regulating valve are used to bind and control the atomization pressure with the addition ratio. Based on the synergist viscosity and nozzle specifications, the atomization pressure threshold is preset to 0.2-0.6 MPa. The central control unit collects auxiliary pipeline pressure data in real time. When the pressure is lower than the atomization threshold, the intake pressure of the pneumatic diaphragm pump is adjusted first, by 0.01 MPa every 0.1 seconds, to ensure that the atomized particle size is stable at 50-100 μm. When the pressure reaches the atomization threshold, the opening of the regulating valve is finely adjusted to match the fuel flow ratio, thus solving the problem of uneven mixing caused by the existing technology that only controls the ratio and ignores atomization.

[0143] Specific control algorithm and calculation process

[0144] The atomization pressure coordinated control adopts a priority-based hierarchical control algorithm. The core principle is to first maintain the atomization pressure, then match the addition ratio, and set the atomization pressure threshold. The pressure is 0.2-0.6 MPa, determined based on the synergist viscosity and nozzle specifications. Specific steps are as follows:

[0145] Pressure threshold calibration: based on synergist viscosity and nozzle diameter The atomization pressure threshold is calibrated using the following formula: ( unit , (At the time), ensure the atomized particle size is 50-100μm;

[0146] Real-time pressure monitoring: Collects actual values ​​of atomization pressure in auxiliary pipelines. Calculate pressure deviation ;

[0147] Tiered regulation:

[0148] like or Adjust only the inlet pressure of the pneumatic diaphragm pump, and adjust the amount. until ;

[0149] like or Maintain a constant pump inlet pressure, adjust the opening of the adjustable proportional pneumatic control valve, and adjust the opening correction amount. Until the actual flow rate of the synergist matches the set ratio.

[0150] In a specific embodiment, two operating conditions are used: a nozzle diameter of 1.5mm, a set addition ratio of 500:1, and a fuel flow rate of 200. The synergist flow rate is set at 0.4. )

[0151] Example 1: High viscosity synergist ( )

[0152] Calibration atomization pressure threshold ;

[0153] Initial detection , Start the pump body pressure regulation:

[0154] The pressure increases by 0.01 MPa every 0.1 seconds, and after 1.5 seconds... The threshold has been reached;

[0155] At this point, the actual flow rate of the synergist is 0.35. ,deviation Adjust the opening of the regulating valve:

[0156] Opening correction amount The opening degree increased from 30% to 42.5%;

[0157] Results: Atomized particle size 80 μm, actual synergist flow rate 0.402 The proportional error is ±0.5%.

[0158] Example 2: Low viscosity synergist ( )

[0159] Calibration atomization pressure threshold ;

[0160] Initial detection Directly adjust the opening of the regulating valve:

[0161] Initial actual flow rate: 0.38 Opening correction amount The opening degree increased to 35%;

[0162] Results: Atomized particle size 60 μm, actual synergist flow rate 0.399 g / L The proportional error is ±0.25%.

[0163] Example 3: Sudden Pressure Drop Due to Abrupt Change in Operating Conditions

[0164] During normal operation , Sudden pipeline leak caused It dropped to 0.35 MPa;

[0165] Immediately pause the regulating valve adjustment and start the pump body pressure regulation. After 1 second... Restored to 0.45 MPa;

[0166] The regulating valve was restored to its original state, and the flow deviation was corrected. The atomized particle size did not exceed 50-100μm throughout the process, and the proportional error was ≤±2%.

[0167] Furthermore, referring to Figure 4 S4 further includes:

[0168] S41, Real-time detection of fuel flow fluctuation amplitude; When the fuel flow fluctuation amplitude is greater than or equal to the compensation threshold, the compensation mode is triggered.

[0169] S42, based on the calculation of correction coefficients and time constants using a first-order inertial element for gradual adjustment, determine the gradual adjustment curve;

[0170] S43, update the enhancer flow correction value according to the preset sampling cycle, and collect actual flow feedback simultaneously until the fuel flow fluctuation is less than the compensation threshold, and exit the compensation mode.

[0171] S44 records the fluctuation range of fuel flow, adjustment curve, and stabilization time data for this fluctuation, forming a dynamic compensation curve library.

[0172] In this embodiment, addressing industry pain points such as unstable flow rate and insufficient atomization pressure for high-viscosity media during the initial filling stage, an innovative flow fluctuation compensation + atomization pressure linkage control logic is designed to achieve the dual goals of precise proportioning and atomization compliance, as detailed below:

[0173] Flow fluctuation compensation logic: The preset flow fluctuation compensation threshold is adjustable from 5% to 15%. When the flow fluctuation reaches the compensation threshold due to the initial stage of refueling or sudden changes in operating conditions, the central control unit automatically starts the compensation mode and adopts a gradual proportional correction strategy. According to the flow fluctuation trend, the flow rate of the enhancer is slowly adjusted (adjustment rate ≤ 0.5% / s). At the same time, the fluctuation data is recorded to form a dynamic compensation curve, which can be directly called in the same operating conditions in the future to shorten the stabilization time.

[0174] Specific algorithm formula for progressive scaling correction strategy

[0175] When the traffic fluctuation reaches the compensation threshold, set it to... When the adjustment is 5%-15%, activate the compensation mode and adjust the synergist flow rate accordingly. A gradual adjustment is achieved using a first-order inertial element; the core formula is:

[0176] ;

[0177] ;

[0178] in:

[0179] Set the flow rate for the theoretical synergist after flow fluctuations, per unit ;

[0180] To adjust the time, the unit ;

[0181] This is a correction factor, with a value between 0 and 1. This represents the actual fluctuation range of fuel flow. The compensation threshold;

[0182] The time constant is set to 2 seconds to ensure that the adjustment rate is ≤0.5% / s;

[0183] Actual adjustment rate If the calculation rate exceeds the threshold, it will automatically... Increase the speed to meet the requirements.

[0184] Calculation process:

[0185] Real-time monitoring of fuel flow fluctuations ,when When this occurs, compensation mode is triggered;

[0186] Calculate the correction factor according to the formula. and time constant Determine the gradual adjustment curve;

[0187] The adjuvant flow correction value is updated every 100ms sampling period, and the actual flow feedback is collected synchronously until... Exit the compensation mode;

[0188] Record this fluctuation Data such as adjustment curves and stabilization time are used to form a dynamic compensation curve library.

[0189] Specific implementation example, compensation threshold Set the addition ratio to 1000:1 and the initial fuel flow rate to 200. ;

[0190] Operating condition: Initial fuel flow rate from 180 It surged to 220 , This triggers the compensation mode, and the theoretically set flow rate of the synergist is increased from 0.18. Rise to 0.22 .

[0191] Calculate the correction factor time constant Adjust the curve to ;

[0192] Enhancer flow rate correction values ​​at different times:

[0193] : ;

[0194] : , the adjustment rate 0.083% / s < 0.5% / s;

[0195] : ;

[0196] : ;

[0197] : , drop to 8% < 10%, and exit the compensation mode;

[0198] Result: The stabilization time is only 8s. The overall filling stabilization time is shortened from 15 - 20 minutes in the prior art to 5 - 10 minutes, and the proportional error is ≤ ±2% throughout the process.

[0199] Further, the S5 includes:

[0200] When any of the core parameters is abnormal, the control unit automatically triggers linkage regulation;

[0201] Based on the principle of first compensation, then warning, and then shutdown, a linkage regulation plan is formulated for the abnormal core parameters. The correction amount of all regulation parameters is calculated by the linear interpolation method, and the formula is:

[0202] ;

[0203] Where is the parameter correction amount, is the actual value, is the set value, is the alarm threshold;

[0204] is the maximum correction amount, which is limited by the device hardware parameters.

[0205] Among them, in this embodiment, the temperature in the mixing section is abnormal (set value , alarm threshold )

[0206] Condition 1: The temperature is on the low side, , and it does not reach the alarm threshold of 25°C;

[0207] Calculate the temperature deviation rate ;

[0208] Linkage regulation measures:

[0209] The heat tracing and insulation layer: The power is increased from 600W to Maximum power 800W, heating rate 1℃ / min;

[0210] Atomization pressure: increased from 0.45 MPa to , maximum 0.6MPa;

[0211] Synergist injection rate: from 0.4 Down to Reduce mixed load;

[0212] Feedback determination: When the temperature rises above 33℃, gradually restore the original parameters until 35℃.

[0213] Operating Condition 2: Temperature is too high. The alarm threshold of 45℃ has not been reached.

[0214] Temperature deviation rate ;

[0215] Linked control measures: heating power reduced to 500W, atomization pressure reduced by 0.05MPa, injection rate increased by 0.05 MPa. .

[0216] In another embodiment, the filter pressure difference is abnormal (set value). Warning threshold Alarm threshold )

[0217] Operating condition: differential pressure Warning threshold;

[0218] Deviation rate ;

[0219] Coordinated regulatory measures:

[0220] The inlet pressure of the pneumatic diaphragm pump was increased from 0.4 MPa to... To compensate for traffic loss;

[0221] Alarm module: Triggers audible and visual warning, 80dB, flashing red light, and HMI display indicating that the filter screen is about to become clogged; please maintain it promptly.

[0222] Data recording: Real-time recording of differential pressure changes; if the differential pressure rises to 0.15 MPa within 30 minutes, shutdown protection is triggered.

[0223] In another embodiment, the main pipeline pressure is abnormal, and the set value... Warning threshold Alarm threshold ;

[0224] Operating condition: Pressure Warning threshold;

[0225] Deviation rate ;

[0226] Coordinated regulatory measures:

[0227] Adjustable proportional pneumatic control valve: Reduce opening by 50% × 10% = 5%, reducing the amount of synergist injected;

[0228] Main pipeline shut-off valve: electrically adjustable opening to reduce fuel flow and alleviate pressure rise;

[0229] Pressure sensor: Increase sampling frequency to 50ms to monitor pressure changes in real time.

[0230] Example 4: The flow rate of the synergist is abnormal, and the ratio deviation is > ±2%, but the alarm threshold of ±5% is not reached;

[0231] Operating conditions: Proportional deviation +3%, fuel flow rate 200 The synergist was set at 0.2. The actual value is 0.206. ;

[0232] Deviation rate ;

[0233] Linkage control measures: The opening of the regulating valve is reduced by 60% × 5% = 3%, and the air inlet pressure of the pump body is reduced by 60% × 0.02 MPa = 0.012 MPa.

[0234] In another embodiment: if the preheating temperature of the mixing section is abnormal, the preheating temperature of the synergist is set to ≤60℃, and the alarm threshold is 65℃;

[0235] Operating condition: Preheating temperature 62℃

[0236] Deviation rate ;

[0237] Linkage control measures: Reduce the power of the preheating module by 40% × 100W = 40W, stop increasing the synergist flow rate, and maintain the current injection rate.

[0238] Example 6: Abnormal atomization pressure, set value 0.45MPa, warning threshold 0.2 / 0.6MPa, alarm threshold 0.15 / 0.65MPa;

[0239] Operating conditions: atomization pressure 0.2MPa, warning threshold;

[0240] Deviation rate ;

[0241] Linkage control measures: The pump body intake pressure is increased by 0.01MPa / 0.1s until the pressure is restored to 0.45MPa, and the proportional adjustment is paused to prioritize atomization.

[0242] Furthermore, referring to Figure 5 S6 further includes:

[0243] S61, extract feature parameters and select oil type, ambient temperature, fuel viscosity, mixing section temperature, atomization pressure threshold, addition ratio, flow fluctuation compensation threshold, PID / KP parameters, and fuzzy rule base correction coefficient as clustering samples;

[0244] S62, based on K-means clustering, divides the accumulated multi-batch refueling data into heavy fuel oil-low temperature condition, heavy fuel oil-normal temperature condition and diesel-all temperature range condition, with the cluster centers being typical features of each type of condition;

[0245] S63 uses multiple linear regression, with proportional error, mixing uniformity, and settling time as evaluation indicators, to establish a multiple linear regression model between evaluation indicators and control parameters.

[0246] S64, Solve for the optimal parameters. Based on the proportional error, mixing uniformity, and settling time, construct constraints, solve for the optimal solution of the regression model, and determine the optimal control parameters for various working conditions.

[0247] S65, update the parameter library. After adding new batch data, re-cluster and regress for each preset number of batches to update the optimal parameter library and achieve self-learning iteration.

[0248] In this embodiment, a closed-loop iteration of control data tracing, operating condition self-learning, and parameter adaptive correction is innovatively achieved by combining the data storage module. The specific logic is as follows:

[0249] Full-process data traceability: Automatically records all parameter data for each refueling, with a sampling interval of 1 second, including flow rate, pressure, temperature, proportional parameters, algorithm switching status, alarm information, etc. for each time period, with a storage capacity of ≥100,000 records, and supports querying and exporting by batch and date, providing data support for ship energy efficiency auditing and control parameter optimization.

[0250] Operating condition self-learning and parameter correction: The central control unit has a built-in operating condition self-learning algorithm. After accumulating more than 50 batches of refueling data, it automatically analyzes the optimal control parameters (algorithm parameters, atomization pressure, compensation threshold) under different oil types (heavy fuel oil, diesel), different ambient temperatures, and different viscosities, forming a dedicated operating condition parameter library; under the same operating conditions in the future, the optimal parameters are automatically called, without the need for repeated manual adjustments, achieving one-time adjustment and lifelong adaptation, improving the convenience of operation.

[0251] The self-learning algorithm for operating conditions employs an improved K-means clustering algorithm combined with multiple linear regression. Its core principle is to perform cluster analysis on the refueling data to extract characteristic parameters for different operating conditions, and then determine the optimal control parameters through regression analysis. The steps are as follows:

[0252] Feature parameter extraction: Nine types of feature parameters were selected as clustering samples: oil type, ambient temperature, fuel viscosity, mixing section temperature, atomization pressure threshold, addition ratio, flow fluctuation compensation threshold, PID / KP parameters, and fuzzy rule base correction coefficient.

[0253] K-means clustering: The accumulated 50+ batches of refueling data were divided into 3 operating conditions (heavy fuel oil - low temperature, heavy fuel oil - normal temperature, diesel - all temperature range), and the cluster centers were the typical features of each operating condition.

[0254] Multiple linear regression: Using proportion error, mixing uniformity, and settling time as evaluation indicators, a multiple linear regression model is established between the evaluation indicators and the control parameters.

[0255]

[0256] in As evaluation indicators, To adjust parameters, For regression coefficients, For constant terms;

[0257] Optimal parameter solution: With proportional error ≤ ±2%, mixing uniformity ≥ 95%, and stabilization time ≤ 10 min as constraints, solve the optimal solution of the regression model to determine the optimal control parameters for various working conditions;

[0258] Parameter library update: After adding new batches of data, clustering and regression are performed every 10 batches to update the optimal parameter library and achieve self-learning iteration.

[0259] Data preprocessing: The collected raw data is normalized to eliminate the influence of dimensions. The formula is as follows: ;

[0260] Cluster analysis: Set the number of clusters k=3, calculate the Euclidean distance from each sample to the cluster center, and iterate repeatedly until the cluster centers are stable;

[0261] Regression analysis: The least squares method is used to solve for the regression coefficients, minimizing the sum of squared residuals;

[0262] Optimal solution: The optimal combination of control parameters is found under constraints using linear programming.

[0263] Parameter matching: When a new operating condition starts, the current feature parameters are extracted and matched with the cluster centers in the parameter library (similarity ≥ 80%), and the corresponding optimal control parameters are called.

[0264] In a specific implementation, a total of 60 batches of refueling data were clustered into 3 types of operating conditions;

[0265] Step 1: Feature parameter extraction and normalization, selecting key parameters;

[0266] Table 4 Key Parameter Extraction Feature Table

[0267]

[0268] Step 2: K-means clustering

[0269] The cluster centers for the three types of working conditions were calculated, and the sample similarity was ≥85%, indicating good clustering results.

[0270] Step 3: Multiple linear regression, taking operating condition 1 as an example, the evaluation index is proportional error. ;

[0271] Select atomization pressure PID-KP Compensation threshold Establish a regression model with the following variables as independent variables:

[0272]

[0273] The regression coefficients obtained by solving the least squares method are significant, indicating a good model fit. .

[0274] Step 4: Solving for optimal parameters (constraints: , , , )

[0275] The optimal parameters are obtained by solving: , , At this time, the proportional error The mixing uniformity was 98%, and the stabilization time was 6 minutes.

[0276] Step 5: Implementing New Operating Condition Matching

[0277] New operating condition: Heavy fuel oil, ambient temperature 6℃, fuel viscosity 980. The similarity to condition 1 is 96% ≥ 80%;

[0278] Execution: Automatically calls the optimal parameters for operating condition 1 (atomization pressure 0.55MPa, PID-KP=3.5, compensation threshold 8%), without manual adjustment. It reaches stability 5 minutes after starting the filling, with a proportional error of ±0.9% and a mixing uniformity of 97.5%.

[0279] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0280] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0281] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A method for automatic mixing and intelligent control of marine fuel additives, characterized in that, Includes the following steps: Construct a main pipeline system, an auxiliary pipeline system, a monitoring and control system, and a hybrid enhancement system. Input operating condition data through a human-machine interface, and automatically call and preset the optimal control parameters matched in the self-learning parameter library of operating conditions, and perform self-testing of sensors and actuators. Core parameters are collected synchronously at a predetermined sampling period, the main fuel flow fluctuation amplitude is calculated for multiple consecutive sampling periods, and a PID control algorithm or a fuzzy control algorithm is decided based on the relationship between the main fuel flow fluctuation amplitude and a preset fluctuation threshold. The atomization pressure threshold is calibrated based on the viscosity of the enhancer and the nozzle diameter. When the actual atomization pressure in the auxiliary pipeline is lower than the atomization pressure threshold, the intake pressure of the pneumatic diaphragm pump is adjusted first until the target is met. After the target is met, the opening of the adjustable proportional pneumatic regulating valve is finely adjusted to match the fuel flow ratio. When the fluctuation range of fuel flow reaches the preset compensation threshold range, the gradual proportional correction strategy is activated, and the enhancer flow is adjusted according to the first-order inertial link formula. When any of the core parameters deviates from the set range, a linkage control measure is triggered based on the linear interpolation method. The system records all parameter data at predetermined time intervals to form an operation log. After accumulating multiple batches of data, it generates a dedicated operating condition parameter library through improved K-means clustering and multiple linear regression analysis. When a new operating condition is started, it automatically matches the optimal control parameters with a similarity greater than or equal to the preset similarity threshold.

2. The automatic mixing and intelligent control method for marine fuel additives according to claim 1, characterized in that, The main pipeline system is used to transport marine fuel oil, including a main fuel oil flow sensor and a mixing section; The auxiliary pipeline system is a delivery channel for the synergist, including quick connectors, precision filters, one-way isolation valves, pneumatic diaphragm pumps, mass flow meters, adjustable proportional pneumatic regulating valves, and check valves. The precision filter adopts a stainless steel sintered mesh structure and is equipped with a differential pressure transmitter. The pneumatic diaphragm pump is a marine explosion-proof model, and the flow rate is coarsely adjusted by adjusting the inlet air pressure. The one-way isolation valve is connected in series with a pressure sensor to form a double backflow protection. The monitoring and control system includes a central control unit, a human-machine interface, a closed-loop feedback module, and an alarm module. The mixing enhancement system includes a multi-point symmetrical bypass mixing nozzle group evenly arranged along the circumference of the mixing section of the main pipeline, a helical blade static mixer installed downstream of the additive injection point, and a mixing section heat tracing and insulation layer. The nozzle outlet of the multi-point symmetrical bypass mixing nozzle group forms a predetermined angle with the inner wall of the main pipeline and faces the fuel flow direction. The helical blade static mixer has a predetermined number of blades, blade angle, and blade spacing.

3. The automatic mixing and intelligent control method for marine fuel additives according to claim 2, characterized in that, The central control unit is communicatively connected to the main fuel flow sensor, mass flow meter, pressure sensor, temperature sensor, differential pressure transmitter, and actuator. The central control unit has a built-in proportional addition algorithm based on PID or fuzzy control. Based on the real-time data transmitted by the fuel flow sensor in the main pipeline and the addition ratio set by the human-machine interface, it automatically calculates the target flow rate of the enhancer. Combined with the enhancer flow rate in the auxiliary pipeline, pipeline pressure, and filter differential pressure data, it outputs control signals to the pneumatic diaphragm pump to adjust the intake pressure and the adjustable proportional pneumatic regulating valve to achieve matching of the enhancer flow rate.

4. The automatic mixing and intelligent control method for marine fuel additives according to claim 1, characterized in that, Synchronously collect multiple core parameters at a predetermined sampling period, calculate the fluctuation amplitude of the main fuel flow rate for consecutive sampling periods, and make a decision to adopt a PID control algorithm or a fuzzy control algorithm based on the relationship between the fluctuation amplitude of the main fuel flow rate and a preset fluctuation threshold. It further includes: The central control unit real-time collects the main fuel flow rate data including the main fuel flow rate, synergist flow rate, main pipeline pressure, auxiliary pipeline atomization pressure, mixing section temperature, and filter differential pressure; Calculate the flow rate fluctuation amplitude for consecutive sampling periods based on the current flow rate and the average flow rate; When the flow rate fluctuation amplitude is less than the preset fluctuation threshold, it is determined as a steady-state filling condition, and the PID control algorithm is enabled. When the flow rate fluctuation amplitude is greater than or equal to the preset fluctuation threshold, it is determined as a dynamic filling condition, and automatically switch to the fuzzy control algorithm.

5. The automatic mixing and intelligent control method for marine fuel additives according to claim 1, characterized in that, Calibrate the atomization pressure threshold according to the synergist viscosity and nozzle diameter. When the actual atomization pressure of the auxiliary pipeline is lower than the atomization pressure threshold, first increase the intake pressure of the pneumatic diaphragm pump until it reaches the standard, and then finely adjust the opening of the adjustable proportional pneumatic control valve to match the fuel flow rate ratio. It further includes: The calibrated pressure threshold is determined based on the synergist viscosity and nozzle diameter. calibrate the atomization pressure threshold to ensure the atomization particle size range; Real-time detect the pressure, collect the actual value of the auxiliary pipeline atomization pressure, and calculate the pressure deviation in combination with the atomization pressure threshold; Layered control, if the pressure deviation If the actual value of the atomization pressure in the auxiliary pipeline is less than the atomization pressure threshold, only adjust the air inlet pressure of the pneumatic diaphragm pump and configure the adjustment amount until the actual value of the atomization pressure in the auxiliary pipeline is greater than or equal to the atomization pressure threshold. If the pressure deviation If the actual value of the atomization pressure in the auxiliary pipeline is greater than or equal to the atomization pressure threshold, keep the pump body air inlet pressure constant, adjust the opening of the adjustable proportional pneumatic regulating valve, configure the opening correction amount, until the actual flow rate of the synergist matches the set ratio.

6. The automatic mixing and intelligent control method for marine fuel additives according to claim 1, characterized in that, When the fuel flow rate fluctuation amplitude reaches within the preset compensation threshold range, start the progressive proportional correction strategy and adjust the synergist flow rate according to the first-order inertia link formula. It further includes: Real-time detect the fuel flow rate fluctuation amplitude. When the fuel flow rate fluctuation amplitude is greater than or equal to the compensation threshold, trigger the compensation mode; Calculate the correction coefficient and time constant according to the progressive adjustment using the first-order inertia link to determine the progressive adjustment curve; Update the synergist flow rate correction value at a preset sampling period, synchronously collect the actual flow rate feedback until the fuel flow rate fluctuation amplitude is less than the compensation threshold, and then exit the compensation mode; Record the fuel flow rate fluctuation amplitude, adjustment curve, and stable time data of this fluctuation to form a dynamic compensation curve library.

7. The automatic mixing and intelligent control method for marine fuel additives according to claim 1, characterized in that, When any of the core parameters deviates from the set range, trigger the linkage control measures according to the linear interpolation method, including: When any of the core parameters is abnormal, the control unit automatically triggers the linkage control; Based on the principle of first compensation, then warning, and then shutdown, formulate a linkage control plan for the abnormal core parameters. The correction amount of all control parameters is calculated by the linear interpolation method, and the formula is: ； in, For parameter correction amount, This is the actual value. For setting value, This is the alarm threshold; The maximum correction amount is limited by the device's hardware parameters.

8. The automatic mixing and intelligent control method for marine fuel additives according to claim 1, characterized in that, Record the full parameter data at a predetermined time interval to form an operation log. After accumulating multiple batches of filling data, generate a dedicated working condition parameter library through improved K-means clustering and multiple linear regression analysis. When a new working condition is started, automatically match the optimal control parameters with a similarity greater than or equal to the preset similarity threshold. It further includes: Extract characteristic parameters, and select the oil product type, ambient temperature, fuel viscosity, mixing section temperature, atomization pressure threshold, addition ratio, flow rate fluctuation compensation threshold, PID / KP parameters, and fuzzy rule base correction coefficient as clustering samples; Based on K-means clustering, divide the accumulated multiple batches of filling data into heavy fuel oil - low temperature working condition, heavy fuel oil - normal temperature working condition, and diesel - full temperature range working condition. The clustering centers are the typical characteristics of each working condition; A multiple linear regression model was established using proportionality error, mixing uniformity, and settling time as evaluation indicators to correlate the evaluation indicators with the control parameters. ； in As evaluation indicators, To adjust parameters, For regression coefficients, For constant terms; To find the optimal parameters, constraints are constructed based on proportional error, mixing uniformity, and settling time. The optimal solution of the regression model is then obtained to determine the optimal control parameters for various operating conditions. The parameter library is updated, and after adding new batches of data, the clustering and regression are re-performed for each preset number of batches to update the optimal parameter library and achieve self-learning iteration.

Images