Ship main propulsion intelligent control system based on multi-source data fusion

Abstract

The application discloses a kind of ship main propulsion intelligent control systems based on multi-source data fusion, it is related to ship main propulsion intelligent control technical field;Solve the problems that the prior art exists dynamic optimization deficiency, data fusion ability is weak, energy efficiency is low, fault-tolerant mechanism is passive, control strategy is rigid, fault recovery is slow and intelligent level is low, cannot cope with complex environmental changes;The present application introduces multi-source data fusion intelligent control algorithm, dynamically optimizes the power distribution and battery scheduling of ship main propulsion system, combines environmental data and equipment data for deep fusion, improves propulsion efficiency and energy management;Through multidimensional data fusion, improve system adaptability and fault-tolerant capability, real-time adjustment and intelligent prediction, reduce fuel consumption, reduce battery aging, improve sailing stability;The system also has intelligent fault detection and prediction mechanism, reduces manual intervention, improves fault-tolerant capability and system intelligent level.

Description

Technical Field

[0001] The present invention relates to the technical field of ship main propulsion intelligent control, and in particular to a ship main propulsion intelligent control system based on multi-source data fusion. Background Art

[0002] Traditional ship propulsion control systems primarily utilize threshold control based on a single sensor (such as a speed sensor) and adjust diesel engine power output via a PID algorithm. Power distribution is fixed (e.g., the diesel engine carries 80% of the load, with the battery providing the remainder), but lack dynamic optimization. Data silos are severe, and environmental data (wind speed, waves) is processed separately from equipment status data. Typical drawbacks include low control accuracy (e.g., response delays exceeding 2 seconds when load changes due to force 6 winds and waves), poor energy efficiency (diesel engines often operate in uneconomical power zones, increasing fuel consumption by 15% to 20%), and short battery life (without a dynamic SOC protection strategy, frequent deep discharges of the battery result in a cycle life of less than 1,000 cycles). Early data fusion technologies primarily relied on simple weighted averaging or Kalman filtering to fuse similar sensor data (e.g., mean filtering of multiple temperature sensors). Environmental data and equipment data were statically linked only through rule bases (e.g., if-then statements). This resulted in weak heterogeneous data fusion capabilities (e.g., modeling errors >20% for the dynamic impact of wind speed on propeller thrust) and poor interference immunity (e.g., high-frequency interference in current signals causes large fluctuations in fusion results, with errors of ±5%). Traditional fault-tolerance mechanisms employed hardware redundancy (e.g., dual generator backup) or simple threshold alarms (e.g., power cutoff when the temperature exceeds 90°C). However, fault recovery relied on manual intervention, resulting in response times >10 seconds. Typical drawbacks included high fault-tolerance costs (hardware redundancy increases system weight and cost, with backup generators accounting for 8%-12% of the total ship cost) and insufficient intelligence (e.g., inability to perform predictive maintenance, and a downtime rate >5% for sudden failures). The key defects of existing technologies include a single data fusion dimension (only integrating data from similar sensors, lacking in-depth correlation between environment + equipment + navigation, such as failing to dynamically couple wave height with propeller speed, resulting in a propulsion efficiency drop of more than 30% when sea conditions deteriorate), rigid control strategy (the fixed power distribution ratio cannot adapt to load fluctuations, resulting in frequent start and stop of the diesel engine, >10 times per hour, and a 12% to 18% reduction in fuel efficiency; charging and discharging scheduling relies on manual experience, SOC protection lags, and accelerates battery aging), poor adaptability to dynamic environments (traditional PID control adjusts overshoot by >20% in nonlinear and strong disturbance scenarios, lacks stability, and lacks multi-objective optimization capabilities, making it difficult to balance fuel economy, equipment life and navigation safety, such as forcing diesel engines to overload to ensure power supply), and passive and inefficient fault tolerance mechanisms (fault diagnosis relies on threshold alarms, with a false alarm rate of up to 15%, and the inability to accurately locate the root cause of the fault. The backup system switching has a long delay, resulting in the risk of power outages for key equipment (such as navigation radar) during the switching period). Summary of the Invention

[0003] In response to the problems mentioned in the above background technology, the present invention proposes a ship main propulsion intelligent control system based on multi-source data fusion.

[0004] The object of the present invention can be achieved through the following technical solutions: A ship main propulsion intelligent control system based on multi-source data fusion, comprising: a data processing module, a control decision module and an execution control module;

[0005] The data processing module is used to perform denoising and data integration on multi-source data, specifically:

[0006] Step 1: Use Mann filtering to eliminate noise from sensor data;

[0007] Step 2: Integrate data from different sensors; use a multi-layer perceptron to obtain features; input multi-source sensor data into the MLP, calculate the hidden layer neurons through the hidden layer to obtain the final output; then perform data fusion based on the output results to obtain the fused data;

[0008] The control decision module generates control instructions based on the system status and navigation requirements, specifically: performing power allocation algorithm optimization analysis:

[0009] S011: Determine optimization objectives, including minimizing fuel consumption, optimizing battery charging and discharging, and ensuring propulsion system power requirements are met;

[0010] S012: Input parameters: Load demand P load , Maximum power of diesel generator Current generator power P gen (t), battery state of charge SOC(t) and charge and discharge efficiency η bat ;

[0011] S013: Calculation process:

[0012] S131: Optimizing the objective function;

[0013] S132: perform constraints;

[0014] S014: through the dynamic power allocation strategy and the charge and discharge scheduling strategy; if the SOC is higher than the set threshold, generate the charge and discharge scheduling instruction 1; if the SOC is lower than the set threshold, generate the charge and discharge scheduling instruction 2;

[0015] When the load is less than the minimum value within the preset load demand range, a battery charging instruction is generated; when the load is greater than the maximum value within the preset load demand range, a battery discharging instruction is generated;

[0016] The execution control module converts the instructions generated by the control decision module into physical actions.

[0017] As a preferred embodiment of the present invention, it also includes a multi-source data acquisition module and a human-computer interaction module;

[0018] The multi-source data acquisition module collects real-time status data of diesel generators, batteries, and drive motors through current sensors, voltage sensors, speed sensors, and temperature sensors. It also collects environmental data and navigation status data, converts the analog signals output by the sensors into digital signals, and transmits the data to the main controller via the CAN bus and Ethernet.

[0019] The human-computer interaction module provides a real-time status visualization interface, receives operation instructions, triggers alarm prompts, supports historical data query and fault diagnosis interaction, and allows emergency intervention.

[0020] As a preferred embodiment of the present invention, the specific process of calculating the hidden layer neurons through the hidden layer to obtain the final output is:

[0021] Multi-layer perceptron is used for feature acquisition; multi-source sensor data is input into MLP, and the output is: X = (x1, x2, ..., x n ); x1,x2,...,x n Represents n sensor data; then normalizes it and calculates the hidden layer neurons through the hidden layer: the formula is: w ij is the input data x i To the hidden layer neuron h j The weight of ; bj is the bias term; f(·) is the activation function; if there are L hidden layers, the calculation formula becomes:

[0022] H (l) =f(W (l) H (l-1) +b (l) ), H (l) is the output of the lth hidden layer, W (l) and b (l) are the weights and biases of this layer, and the activation function f(x) performs a nonlinear transformation; this is then calculated by the neurons in the output layer:

[0023] y is the final output, and σ(·) is the activation function.

[0024] As a preferred embodiment of the present invention, data fusion is performed based on the output results to obtain the fused data in the following specific process:

[0025] Rule-based fusion: Battery SOC calculation:

[0026] Calculate the current SOC through the battery charge and discharge current and capacity;

[0027] Weighted average fusion, calculates the weighted average;

[0028] Kalman fusion, GPS+IMU data fusion; wind speed+ocean current data fusion.

[0029] As a preferred embodiment of the present invention, the specific process of optimizing the objective function is as follows:

[0030] Define the objective function to minimize fuel consumption and battery usage cost: minF = C fuel ×P gen (t)+C battery ×P bat (t), C fuel is the fuel cost coefficient, C battery Cost of battery use.

[0031] As a preferred embodiment of the present invention, the specific constraints are:

[0032] Power balance constraints, diesel generator output power constraints, and battery SOC constraints;

[0033] Power balance constraint: P load =P gen (t)+P bat (t); P bat (t) is positive or negative;

[0034] Diesel generator output power constraints: Battery SOC constraint: SOC min ≤SOC(t)≤SOC max .

[0035] As a preferred embodiment of the present invention, the specific process of the execution control module converting the instructions generated by the control decision module into physical actions is as follows:

[0036] When receiving the charge and discharge scheduling instruction 1, the system increases the diesel generator power, adjusts the fuel injection rate, increases the speed and output power, disconnects the battery discharge circuit, reduces the discharge current, and triggers the trickle charge protection. The system calculates the power distribution factor α1 to distribute the generator and battery power. The inverter adjusts the IGBT switching frequency to match the power supply capacity. The system monitors the temperature and triggers the cooling system when it exceeds the limit. The HMI displays the power supply status and SOC information, and performs fault processing in the event of an abnormality.

[0037] When receiving the second charge and discharge scheduling instruction, the generator power is increased to meet the load and charging requirements, while charging the battery at the same time. The BMS disconnects the discharge circuit, allowing only low-current discharge, and dynamically adjusts the charging mode. The power allocation factor α2 is calculated to optimize the generator and battery charging strategies. When the load exceeds the limit, the propulsion power is reduced to ensure the power supply of key equipment. The charging temperature rise and charging current anomalies are monitored, triggering the cooling fan or disconnecting the charging circuit. The backup power supply is activated, and the voltage estimation mode is switched to in the event of a SOC sensor failure. The HMI updates the power allocation status in real time and provides abnormal alarm prompts.

[0038] When receiving a battery charging command, the BMS self-checks the SOC and temperature, and adjusts cooling or heating if the temperature is abnormal; determines the charging source; disconnects the discharge relay; switches between constant current, constant voltage, and trickle charging modes according to the SOC during the charging process; monitors charging parameters and triggers overvoltage, overcurrent, and overtemperature protection; and performs a soft stop after charging is complete, entering floating charge or standby mode.

[0039] When receiving a battery discharge command, the BMS self-checks the SOC and temperature, and suspends discharge when the SOC is low or the temperature is abnormal; adjusts the discharge power according to the load demand; closes the charging relay and opens the discharge relay to match the load type; uses constant power or pulse discharge mode to optimize energy output; monitors discharge parameters and triggers abnormal protection mechanisms.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention uses Kalman filtering, MLP neural network and weighted fusion strategy to perform spatiotemporal alignment and feature extraction on heterogeneous data such as generator operating conditions (current / temperature), environmental parameters (wind speed / waves) and navigation status (load / speed). For example, the Kalman fusion algorithm using GPS+IMU can reduce the heading estimation error to within 0.5°, while the nonlinear feature mapping based on multi-layer perceptron can effectively identify the SOC estimation deviation caused by battery aging (such as improving the temperature compensation accuracy of the voltage-current curve by 15%). This deep fusion mechanism enables the system to maintain a propeller thrust fluctuation rate of less than 2% under level 6 sea conditions, which is significantly better than the 5% to 8% fluctuation performance of traditional PID control.

[0042] 2. The present invention incorporates the fuel cost coefficient (0.3L / kWh), battery cycle life cost (0.05$ / kWh) and real-time SOC constraint (SOCmin=20%) into the objective function by constructing a mixed integer programming model, and dynamically calculates the power allocation factor α in combination with the rolling horizon optimization algorithm. Actual measured data show that under typical operating conditions (load demand fluctuations of 80-120kW), the system increases the proportion of time the diesel engine operates in the economic power zone (70% to 85% rated power) to 92% through second-level adjustment of the α value (such as the response time <500ms when α jumps from 0.7 to 0.95), which is 17.3% more fuel-efficient than the traditional solution. At the same time, through the threshold trigger mechanism of the charge and discharge scheduling instruction (ΔSOC=5%), the cycle life of the lithium battery can be extended to more than 4,000 times, which is 2.5 times higher than that of the unprotected strategy.

[0043] 3. This invention utilizes a three-level fault-tolerance design: execution layer, decision layer, and hardware layer. At the execution layer, when a generator overload (greater than 110% of rated current) is detected, the PMS can switch to a backup power source within 50ms (e.g., supercapacitors to fill a 20kW shortfall). At the decision layer, if the SOC sensor fails, the system automatically switches to a Kalman prediction model based on open-circuit voltage (OCV), keeping the SOC estimation error within ±3%. At the hardware layer, a dual-CAN bus redundant architecture is employed, ensuring a delay of less than 10ms for slave takeover in the event of a primary bus failure. Actual tests have shown that, when simulating battery thermal runaway (temperature rise rate >5°C / min), the BMS can shut down the charging circuit and restart the liquid cooling system at full power within 200ms, stabilizing the battery temperature below the safety threshold of 45°C. This shortens the fault response time by 60% compared to traditional solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0045] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0046] The technical solutions of the present invention will be described clearly and completely below in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. It should be understood that the terms "include" and "comprising" used in the specification and claims of this disclosure indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0047] It should also be understood that the terminology used in this disclosure is for the purpose of describing specific embodiments only and is not intended to limit the disclosure. As used in this disclosure and the claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should be further understood that the term "and / or" as used in this disclosure and the claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0048] See also Figure 1 As shown, a ship main propulsion intelligent control system based on multi-source data fusion includes: a multi-source data acquisition module, a data processing module, a control decision module, an execution control module and a human-computer interaction module;

[0049] The multi-source data acquisition module collects real-time status data (such as voltage, current, speed, and temperature) from equipment such as diesel generators, batteries, and drive motors through current sensors, voltage sensors, speed sensors, and temperature sensors. It is also used to collect environmental data (such as wind speed and wave height) and navigation status data (such as speed and load). It converts the analog signals output by the sensors into digital signals and transmits the data to the main controller via the CAN bus and Ethernet.

[0050] The data processing module retrieves multi-source data from the main controller for denoising:

[0051] Step 1: Filter and denoise the collected multi-source data: Use the Mann filter to eliminate the noise of the sensor data;

[0052] Step 2: Integrate data from different sensors; use multi-layer perceptron (MLP) to obtain features; input multi-source sensor data into MLP, and set the output to be: X = (x1, x2, ..., x n ); where x1, x2, ..., x n Represents n sensor data, such as battery current, voltage, temperature, motor torque, etc.; then performs normalization processing, such as normalization (0-1) or Z-score normalization, to ensure the consistent distribution of all input data and improve training stability; then, calculates the hidden layer neurons through the hidden layer: the formula is: Where: w ij is the input data x i To the hidden layer neuron h j The weight of ; bj is the bias term; f(·) is the activation function, usually ReLU: f(x) = max(0,x); if there are L hidden layers, the calculation formula becomes: H (l) =f(W(l) H (l-1) +b (l) ), where H (l) is the output of the lth hidden layer, W (l) and b (l) are the weights and biases of this layer, and the activation function f(x) performs a nonlinear transformation; this is then calculated by the neurons in the output layer: Where y is the final output and σ(·) is the activation function. Data fusion is then performed based on the output results. Data fusion includes:

[0053] Rule-based fusion is suitable for situations where there is a clear mathematical relationship between the data, such as battery SOC calculation:

[0054] Calculate the current SOC through the battery charge and discharge current and capacity;

[0055] Weighted average fusion is suitable for calculating the weighted average of multiple sensors providing the same type of data (such as readings of multiple current sensors) to improve accuracy;

[0056] Kalman fusion is suitable for dynamically changing data fusion, such as: GPS + IMU data fusion: improving the accuracy of position, speed and heading estimation; wind speed + ocean current data fusion: predicting the actual navigation status of the ship;

[0057] The control decision module generates control instructions based on the system status and navigation requirements, specifically: performing power allocation algorithm optimization analysis:

[0058] S011: Determine optimization objectives, including minimizing fuel consumption, optimizing battery charging and discharging, and ensuring propulsion system power requirements are met;

[0059] S012: Input parameters, including load demand P load (Total power demand of current ship propulsion system), maximum power of diesel generator (rated maximum power output capacity of the diesel generator), current generator power P gen (t) (actual power output of the diesel generator at time t), battery state of charge (SOC) SOC(t) (the current state of charge of the battery, usually between SOC min and SOC max between) and the charge and discharge efficiency η bat ;

[0060] S013: Calculation process:

[0061] S131: Optimize the objective function, define the objective function, minimize fuel consumption and battery usage cost: minF = C fuel ×Pgen (t)+C battery ×P bat (t), where C fuel is the fuel cost coefficient, which represents the fuel consumption rate of the diesel generator, C battery Battery usage cost, taking into account battery life attenuation and battery cycle cost;

[0062] S132: Setting constraints include: power balance constraint, diesel generator output power constraint and battery SOC constraint; power balance constraint: P load =P gen (t)+P bat (t); where P bat (t) may be positive (discharging) or negative (charging);

[0063] Diesel generator output power constraints: Battery SOC constraint: SOC min ≤SOC(t)≤SOC max ;

[0064] S014: Dynamic power allocation strategy: In order to optimize power allocation, set the power allocation factor α and substitute it into the formula to obtain: P gen (t) = αP load ,P bat (t) = (1-α)P load ; Among them, α is calculated through an optimization algorithm to ensure that the power distribution achieves the lowest fuel consumption while satisfying the constraints;

[0065] Charge and discharge scheduling strategy: If the SOC is higher than the set threshold, that is, SOC>SOCmax-ΔSOC, the first charge and discharge scheduling instruction is generated; when reducing battery charging, the generator is used for power supply first, limiting battery discharge to prevent deep discharge from damaging battery life; if the SOC is lower than the set threshold, that is, SOC<SOCmin+ΔSOC, the second charge and discharge scheduling instruction is generated; limiting battery discharge to prevent the SOC from being too low and causing power failure, increasing the generator power, and giving priority to battery charging;

[0066] When the load is less than the minimum value within the preset load demand range, a battery charging instruction is generated; the diesel generator maintains economical power output, and the excess power is used to charge the battery; when the load is greater than the maximum value within the preset load demand range, a battery discharging instruction is generated; the battery discharge assists the generator in powering the power, reducing the high-load operation of the diesel engine;

[0067] The execution control module converts the instructions generated by the control decision module into physical actions, specifically:

[0068] When receiving the charge and discharge scheduling instruction 1, the diesel generator power is controlled to increase. The main controller sends a power increase instruction to the diesel generator ECU. The ECU adjusts the fuel injection amount (for example, increases the opening of the injection pump solenoid valve), increases the diesel engine speed and output power; the generator output power is gradually increased from the current value (for example, 50kW) to the target value (for example, 90kW);

[0069] Discharge circuit cutoff: The main controller sends a "discharge limit" command to the battery BMS. The BMS controls the discharge contactor to disconnect (or reduces the duty cycle of the DC-DC converter), so that the discharge current approaches zero. The battery output power drops from the current value (such as 30kW) to a preset safety value (such as below 5kW). The BMS continuously monitors the SOC changes. If the SOC continues to rise due to self-discharge or light load, the trickle charge protection is triggered (such as limiting the charging current to ≤1A).

[0070] The main controller calculates the power allocation factor α1 (for example, α1 = 0.95) according to the optimization algorithm and allocates power according to the formula: P gen (t) = α1 × P load ,P bat (t)=(1-α1)×P load If the load demand is 100kW, the generator will provide 95kW and the battery will only provide 5kW (or reverse charging). The PID controller is used to fine-tune the α1 value to ensure a smooth transition of the generator power (e.g., adjusting α1 ± 0.01 per second).

[0071] Propulsion system cooling linkage: If load demand changes, the inverter adjusts the IGBT switching frequency to match the generator's power supply capacity (e.g., increasing the motor speed from 1000 rpm to 1200 rpm). This stabilizes propeller thrust to prevent ship vibration caused by sudden power increases. When the temperature sensor detects a generator temperature greater than 80°C, it triggers the cooling water pump and fan to start (by closing the circuit via a relay) (e.g., increasing the coolant flow rate from 50 L / min to 80 L / min to ensure heat dissipation in the generator).

[0072] Finally, the HMI displays the current status on the operation interface, such as "Generator power supply: 95%" and "Battery discharge limit, SOC = 91%". When the alarm indicator (yellow) prompts "high SOC state", if a generator overload is detected (such as current > 110% of the rated value), the fault diagnosis module immediately executes: cutting off the fuel supply (ECU triggers an emergency shutdown) and switching to a backup power source (such as another generator).

[0073] When receiving the second charge and discharge scheduling instruction, the main controller sends a "power increase + charge" instruction to the diesel generator ECU. The ECU increases the fuel injection rate, raising the output power to meet the load demand while charging the battery. If the generator is originally operating at economic power (such as 60% of rated power), it is now increased to 85% to 95% of rated power (for example, from 80kW to 120kW). The engine charges the battery through a rectifier or bidirectional DC-DC converter, and the charging current is set by the BMS (for example, the current limit is 0.2C, that is, 20A for a 100Ah battery).

[0074] After receiving the command, the BMS immediately cuts off the discharge circuit contactor (physical disconnection) to prohibit battery discharge. If the system requires emergency discharge (such as a sudden load increase), the BMS only allows a very small current discharge (such as ≤5% of the rated current). The charging circuit contactor closes, and the charger switches the charging mode according to the SOC status: when the SOC is extremely low (such as SOC <20%), it is fast charged to the safety threshold at a constant current (such as 0.3C); when the SOC rises above 25%, it switches to constant voltage mode to prevent overcharging.

[0075] Similarly, the power allocation factor α2 is dynamically adjusted according to the optimization algorithm. The objective function is adjusted to minimize battery discharge and meet charging requirements. The power allocation formula is adjusted to: P gen (t) = α2 × P load +P charge ,P bat (t) = 0 (discharge prohibited), where P charge The PMS sends instructions to the generator and BMS via the CAN bus: generator target power: P_gen = 120kW (including 90kW load + 30kW charging), battery charging power upper limit: P_charge_max = 30kW (dynamically adjusted by the BMS based on SOC and temperature);

[0076] If the load demand exceeds the maximum capacity of the generator (e.g., a sudden load increase), the execution module temporarily reduces the propulsion motor power (e.g., by limiting the speed through the inverter) to prioritize power supply to key equipment (e.g., the navigation system);

[0077] If the battery temperature sensor detects a rise in charging temperature (e.g. >40°C), the cooling fan is forced to start (via the relay control circuit) to maintain the battery temperature ≤35°C;

[0078] If the BMS detects an abnormal charging current (e.g., fluctuations >±10%), it immediately disconnects the charging circuit and issues an alarm via the CAN bus. If the generator is overloaded, the fault-tolerance module activates a backup power source (such as another generator or supercapacitor) to temporarily fill the power gap. If the BMS detects a SOC sensor failure, it switches to voltage estimation mode (based on the open-circuit voltage curve) and triggers an HMI alarm indicating "SOC estimation degradation."

[0079] Interface display: HMI is updated to **"Low SOC Protection Mode**, displaying: generator power: 120kW (load 90kW + charging 30kW), battery status: SOC = 23%, charging current = 30A, and corresponding text; if the alarm indicator (red) flashes, it prompts "Battery power is low, charging"; manual override instructions are allowed (such as forcing battery discharge in an emergency), but the security password must be entered and the risk confirmation must be confirmed.

[0080] When receiving a battery charge command, the battery management system (BMS) first performs a self-check and reads the state of charge (SOC). If the SOC is ≥100%, the charging process is terminated. If the SOC falls below this threshold, the system further reads the battery temperature. If the temperature exceeds the safe range (too high or too low), the BMS suspends charging and activates temperature regulation mechanisms (such as activating the cooling fan or heater), while continuously monitoring the battery voltage to prevent overvoltage or short-circuit risks. The system then determines the available charging source (diesel generator, shore power, photovoltaic system, etc.). If a diesel generator is selected, its load status is first checked to avoid overload. If shore power or photovoltaic power is available, external power is prioritized. After confirming the power source, the BMS double-checks the relay status: opening the discharge relay to prevent the risk of a short circuit caused by parallel charging and discharging, and ensuring that the charge relay is open to prevent shock from the direct connection of the high-voltage power source. The charging execution phase involves the following process: first, closing the charge relay to connect the battery to the charging power source, then adjusting the power parameters based on the power source type. If the power source is a DC power source (such as photovoltaic power or DC shore power), the DC / DC converter will stabilize the voltage (for example, stepping down 400VDC to 48VDC). If the power source is an AC power source (such as a diesel generator or AC shore power), the power is converted to DC by an AC-DC rectifier (for example, 380VAC to 400VDC), and then adjusted to the appropriate voltage (such as 48VDC) by the DC / DC converter. The charging mode dynamically switches based on the SOC: constant current mode (SOC ≤ 80%) uses a set constant current to achieve rapid charging; constant voltage mode (SOC 80%-100%) gradually reduces the current to prevent overcharging; and trickle mode (SOC near 100%) uses a low current to maintain floating charge and compensate for self-discharge losses. Throughout the charging process, the BMS monitors temperature, voltage, and current parameters in real time, ensuring safety through three mechanisms: overvoltage protection (charging is interrupted if the limit is exceeded), overcurrent protection (automatic current reduction or disconnection if abnormal current is present), and overtemperature protection (triggering cooling or emergency shutdown). When charging is complete, the system will execute a soft stop process after the SOC reaches 100% or the current drops to the threshold: gradually reduce the current to adapt the battery to the zero-charge state → turn off the charging relay to cut off the circuit → switch to floating charge mode or standby state according to the grid connection status.

[0081] When receiving a battery discharge command, the BMS initiates a self-check and reads the SOC value. If the SOC is ≤ 20%, discharge is prohibited to prevent over-discharge damage. The system also monitors battery temperature, pausing discharge and activating temperature control if abnormalities occur. The discharge strategy dynamically adjusts based on load demand: during low power demand, discharge power is reduced to optimize battery life; during high power demand, power is provided in conjunction with the diesel generator. During the discharge phase, the charging relay is closed before the discharge relay is opened. The power conversion device is adapted to the load type: for DC loads (such as propulsion motors), a DC / DC converter boosts the battery voltage (e.g., 48VDC to 400VDC); for AC loads (such as marine equipment), a DC-AC inverter converts DC power to AC (e.g., 48VDC to 380VAC). The discharge process utilizes multi-mode dynamic regulation: constant power output is maintained under normal operating conditions; short bursts of high power demand (e.g., ship acceleration) trigger pulsed discharge mode to prevent diesel engine overload; and when the SOC falls below 30%, the system automatically switches to low power mode to extend endurance. During the entire discharge cycle, the BMS continuously monitors key parameters and implements over-discharge / over-temperature protection to ensure that the system operates within safety thresholds, and eventually returns to standby mode based on instructions or changes in conditions, waiting for the next operation cycle.

[0082] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A ship main propulsion intelligent control system based on multi-source data fusion, comprising: Data processing module, control decision module and execution control module; characterized by: The data processing module is used to perform denoising and data integration on multi-source data, specifically: Step 1: Use Mann filtering to eliminate noise from sensor data; Step 2: Integrate data from different sensors; use a multi-layer perceptron to obtain features; input multi-source sensor data into MLP, and calculate the hidden layer neurons through the hidden layer to obtain the final output: use a multi-layer perceptron to obtain features; input multi-source sensor data into MLP, and set the output to be: X=(x1,x2,...,x n ); x1,x2,...,x n Represents n sensor data; then normalizes it and calculates the hidden layer neurons through the hidden layer: the formula is: , w ij is the input data x i To the hidden layer neuron h j The weight of ; bj is the bias term; is the activation function; if there are L hidden layers, the calculation formula becomes: , It is The output of the hidden layer, and are the weights and biases of this layer, and the activation function f(x) performs a nonlinear transformation; this is then calculated by the neurons in the output layer: , y is the final output, It is an activation function; then data fusion is performed based on the output results, and the process of obtaining the fused data is as follows: rule-based fusion: battery SOC calculation: , calculate the current SOC through the battery charge and discharge current and capacity; weighted average fusion, calculate the weighted average value; Kalman fusion, GPS + IMU data fusion; wind speed + ocean current data fusion; The control decision module generates control instructions based on the system status and navigation requirements, specifically: performing power allocation algorithm optimization analysis: S011: Determine optimization objectives, including minimizing fuel consumption, optimizing battery charging and discharging, and ensuring propulsion system power requirements are met; S012: Input parameters: Load demand P load , Maximum power of diesel generator , Current generator power , battery state of charge SOC(t) and charge and discharge efficiency ; S013: Calculate the input parameters. The specific process is as follows: S131: Optimizing the objective function; S132: perform constraints; S014: Using a dynamic power allocation strategy and a charge-discharge scheduling strategy, if the SOC is higher than a set threshold, a charge-discharge scheduling instruction 1 is generated; if the SOC is lower than the set threshold, a charge-discharge scheduling instruction 2 is generated; when the load is less than the minimum value within a preset load demand range, a battery charge instruction is generated; when the load is greater than the maximum value within the preset load demand range, a battery discharge instruction is generated; The execution control module converts the instructions generated by the control decision module into physical actions.

2. The ship main propulsion intelligent control system based on multi-source data fusion according to claim 1 is characterized in that: It also includes multi-source data acquisition module and human-computer interaction module; The multi-source data acquisition module collects status data of diesel generators, batteries, and drive motors in real time through current sensors, voltage sensors, speed sensors, and temperature sensors; collects environmental data and navigation status data, converts the analog signals output by the sensors into digital signals, and transmits the data to the main controller through the CAN bus and Ethernet; The human-computer interaction module provides a real-time status visualization interface, receives operation instructions, triggers alarm prompts, supports historical data query and fault diagnosis interaction, and allows emergency intervention.

3. The ship main propulsion intelligent control system based on multi-source data fusion according to claim 1 is characterized in that: The specific process of optimizing the objective function is as follows: Define the objective function to minimize fuel consumption and battery usage cost: minF=C fuel ×P gen (t)+C battery ×P bat (t), C fuel is the fuel cost coefficient, C battery The cost of battery use.

4. The ship main propulsion intelligent control system based on multi-source data fusion according to claim 1 is characterized in that: The specific constraints are as follows: Power balance constraints, diesel generator output power constraints, and battery SOC constraints; Power balance constraint: P load =P gen (t)+P bat (t); P bat (t) is positive or negative; Diesel generator output power constraints: ≤ ; Battery SOC constraint: SOC min ≤SOC(t)≤SOC max .

5. The ship main propulsion intelligent control system based on multi-source data fusion according to claim 1 is characterized in that: The specific process of the execution control module converting the instructions generated by the control decision module into physical actions is as follows: When receiving the charge and discharge scheduling instruction 1, the system increases the diesel generator power, adjusts the fuel injection rate, increases the speed and output power, disconnects the battery discharge circuit, reduces the discharge current, and triggers the trickle charge protection. The system calculates the power distribution factor α1 to distribute the generator and battery power. The inverter adjusts the IGBT switching frequency to match the power supply capacity. The system monitors the temperature and triggers the cooling system when it exceeds the limit. The HMI displays the power supply status and SOC information, and performs fault processing in the event of an abnormality. When receiving the second charge and discharge scheduling instruction, the generator power is increased to meet the load and charging requirements, while charging the battery at the same time. The BMS disconnects the discharge circuit, allowing only low-current discharge, and dynamically adjusts the charging mode. The power allocation factor α2 is calculated to optimize the generator and battery charging strategies. When the load exceeds the limit, the propulsion power is reduced to ensure the power supply of key equipment. The charging temperature rise and charging current anomalies are monitored, triggering the cooling fan or disconnecting the charging circuit. The backup power supply is activated, and the voltage estimation mode is switched to in the event of a SOC sensor failure. The HMI updates the power allocation status in real time and provides abnormal alarm prompts. When receiving a battery charging command, the BMS self-checks the SOC and temperature, and adjusts cooling or heating when the temperature is abnormal; determines the charging source; and disconnects the discharge relay; The charging process switches between constant current, constant voltage, and trickle mode according to SOC; monitors charging parameters and triggers overvoltage, overcurrent, and overtemperature protection; After charging is completed, soft stop is executed and the battery enters the floating charge or standby state; When receiving a battery discharge command, the BMS self-checks the SOC and temperature, and suspends discharge when the SOC is low or the temperature is abnormal; adjusts the discharge power according to the load demand; closes the charging relay and opens the discharge relay to match the load type; uses constant power or pulse discharge mode to optimize energy output; monitors discharge parameters and triggers abnormal protection mechanisms.

Images